Document 37eMko8YVobaLNBJvJnzK9pZa

14TH JUDICIAL COURT, PARISH OF CALCASIEU, STATE OF LOUISIANA Case No. 94-2695 DEPOSITION OF RICHARD D. IRONS, Ph.D. RAYMOND McFARLAND, et ux, Plaintiffs, VS. HARTFORD ACCIDENT and INDEMNITY COMPANY, et al., Defendants. PURSUANT TO NOTICE and the Louisiana Rules of Civil Procedure, the above-entitled deposition was taken on behalf of Plaintiffs at 4150 East Mississippi, Denver, Colorado, on Friday, December 15, 1995, at 10:30 a.m., before Dyann Patterson-Labo, Registered Professional Reporter and Notary Public within Colorado. 1 2 1 APPEARANCES: 2 For the Plaintiffs: HERSCHEL HOBSON, ESQ. 3 Baggett, McCall & Burgess 3006 Country Club Road 70605 4 Lake Charles, Louisiana 70707-7820 5 For the Defendants: GREGORY W. BELFOUR, ESQ. 6 Jones, Tete, Nolen, Hanchey, Swift & Spears, L.L.P. 7 Post Office Box 910 Lake Charles, Louisiana 70602-0910 8 Also present: Jim Tillson 9 10 11 I N D E X 12 EXAMINATION PAGE 13 December 15, 1995 14 By Mr. Hobson 3 15 By Mr. Belfour 16 17 EXHIBITS 18 (No exhibits were marked.) 19 20 21 22 23 24 25 PATTERSON REPORTING & VIDEO 3 1PR0CEEDINGS 2 RICHARD D. IRONS, 3 having been first duly sworn, was examined and testified 4 as follows: 5 EXAMINATION 6 BY MR. HOBSON: 7 Q. Dr. Irons, you're still at the University of 8 Colorado, are you? 9 A. Yes. 10 Q. And I take it your capacity there hasn't 11 changed since your arrival, or has it? 12 A. I don't remember exactly the last time we met. 13 The only change that has happened since I have been 14 there is I have taken on responsibility as director of a 15 cancer-causation program in the Cancer Center. 16 Q. Tell me about that project. 17 A. I am program director and responsible for our 18 -- the coordinating and promoting of research in cancer 19 causation and carcinogenesis within the Cancer Center. 20 It's an NCI-funded Cancer Center. 21 Q. The Cancer Center is funded by NCI? 22 A. It's an NCI-funded Cancer Center. It's the 23 University of Colorado Health Sciences Center's Cancer 24 Center. 25 Q. And as the director of this project, is it PATTERSON REPORTING & VIDEO 4 1 your job to do research for the NCI, or is this a sort 2 of an administrative post to make sure money keeps 3 coming in? 4 A. No. It's to coordinate research, to encourage 5 collaborative activities amongst faculty, to look for 6 potential opportunities for faculty members to obtain 7 grants, and to encourage researching mechanisms of 8 carcinogenesis. 9 Q. But your job with the NCI cancer-causation 10 project is not to do research, correct? 11 A. I wouldn't say that's a fair characterization, 12 but it does not support directly my research activities. 13 Q. One of the things I really want to cover today 14 with you, and I would like to kind of start there, I 15 guess, is your research activities. What current 16 research activities do you have funded? 17 A. I have a grant from the National Institute of 18 Health. I have a grant from the American Petroleum 19 Institute. I have a grant that I am coming off of in 20 1996 from the National Aeronautics and Space 21 Administration. 22 Q. You say you're "coming off of"? 23 A. It's ending -- it's ending this year. I don't 24 know whether the effective date is December or April. 25 I have -- I have had a -- this last year a PATTERSON REPORTING & VIDEO 5 1 grant from the National Cancer Institute. I finished a 2 grant from the Chemical Manufacturers Association. I 3 have had a number of grants from pharmaceutical 4 companies, the most recent one I believe was 5 Hoffmann-La Roche. I have a number of grants pending. 6 Q. Any others that are currently underway? 7 A. I think that -- I think that covers it. 8 Q. When you use the term "grant" in context with 9 this -- these research activities, is it correct to say 10 that that terminology means that these different 11 entities contract with either you or your administration 12 to do certain kinds of work for which they pay? 13 A. No. That would -- you could say that for 14 contracts and grants. All of those, with the exception 15 of the NCI -- actually, calling the NCI project a grant 16 is probably a misnomer. That's a contract. Everything 17 else has been a grant. 18 Q. What is the distinction that you make between 19 a grant and a contract? 20 A. A contract is a specific agreement -- an 21 agreement to do a specific, characterized piece of work 22 to measure A, B, or C, to specifically do an individual 23 test or project. A grant is to study a specific area. 24 It may be very general, it may be relatively specific, 25 but the experimental design, the things I measure, the PATTERSON REPORTING & VIDEO 6 1 approach I take is at my discretion. 2 Q. And to get a grant, you provide a proposal of 3 what you intend to do with some, as you say, more broad 4 parameters describing the kinds of efforts that you 5 intend to make with the moneys that are going to be 6 provided to you? 7 A. To varying degrees, and varying institutions 8 have different criteria, and depending upon the nature 9 of the proposed work, it may be more specific or less 10 specific as the case may be. 11 Q. Now, the grants are made with agreements that 12 surround them, are they not? 13 A. Sometimes yes, sometimes no. 14 Q. I mean, when you take grant money from the NIH 15 or the API or NASA, it's just given to you with no 16 strings attached? 17 A. It depends. You have lumped those three, and 18 they actually all three are different. In the case of 19 NIH, there is a process which is called extended 20 budgetary authority, which I am subject to, which 21 basically gives me the freedom to manage that grant and 22 the funds made available to it in any manner that I see 23 fit within the context of the work product that I am 24 expected to perform. 25 So whatever budgetary proposals I put forward PATTERSON REPORTING & VIDEO 7 1 in the original grant application I can change if in 2 fact that's a -- if I find it necessary in order to 3 perform the work as I get into it. In the case of NASA, 4 it's fairly stringent with respect to the specific use 5 of funds and what you propose to do. 6 In the case of API, it's extremely -- it's 7 like NIH in the sense that there are no specific 8 provisions with respect to how I reallocate or use funds 9 once the grant is underway. 10 Q. But I think you said even with the NIH grant, 11 which I heard you say is called the loosest arrangement 12 of all, is there still a work product that is involved 13 in the project? 14 A. The work product is what I am expected to do 15 based upon what I proposed to do, and that's evaluated 16 in the context of interim reports and final either 17 renewal of the application or in prospective 18 applications for other funding. 19 Q. Let's start with the NIH grant. When did that 20 grant funding begin, approximately? 21 A. I am in the third year of a five-year cycle. 22 Q. So this is 23 A. So it would be September -- this is '96 24 1994. 25 Q. And you expect that to continue through when? PATTERSON REPORTING & VIDEO 8 1 A. 1998. 2 Q. What is the nature of the work product for the 3 NIH grant? 4 A. It's a study of mechanisms of leukemogenesis 5 associated with exposure to chemotherapeutic agents and 6 benzene. 7 Q. The API grant, when did that start and when do 8 you expect it to end in the current arrangement? 9 A. I believe it started in 1990, and it's been 10 renewable by year, and I have just been informed it will 11 be renewed for '96. 12 Q. What is the nature of the API grant work 13 product? 14 A. Improving the biological basis for evaluation 15 of risk assessment of leukemogenesis and mechanisms of 16 leukemogenesis. 17 Q. The NASA grant, when did it begin, 18 approximately? 19 A. I believe it was 1992. 20 Q. And you say that's in the process of ending 21 now? 22 A. Yes. 23 Q. What was the nature of that grant work? 24 A. Risk characterization associated with extended 25 space-flight environment. PATTERSON REPORTING & VIDEO 9 1 Q. The NCI contract, can you tell me 2 approximately when it began? 3 A. 199 -- I think it was middle to end of '94. 4 Q. And that project ends in '96? 5 A. It's ending now. It's basically done. There 6 is still some paper-writing to do. 7 Q. And what was the endpoint for that NCI 8 contract? 9 A. Characterization of alterations in cytokines 10 associated with benzene poisoning in Chinese workers. 11 Q. The CMA grant that you finished, when did that 12 begin, approximately? 13 A. I think it was 1993, but it could have been 14 '92. 15 Q. Approximately how long did that project last? 16 A. Three to four years. 17 Q. What was the nature of that work? 18 A. Characterization of mechanisms of butadiene 19 leukemogenesis in the mouse. 20 Q. Now, you say that you either have had or have 21 grants from some pharmaceutical companies? 22 A. Yes. 23 Q. Would those be grants or contracts? 24 A. Grants. 25 Q. What was the nature of those? PATTERSON REPORTING & VIDEO 10 1 A. Either characterization of mechanisms of 2 murine leukemogenesis or lymphomagenesis associated with 3 exposure to specific drugs, or potential toxic effects 4 of specific or prototype drugs on human bone marrow 5 function. 6 Q. Are these primarily chemotherapeutic agents 7 for cancer treatment? 8 A. No. 9 Q. What kinds of drugs were they? 10 A. Some are proprietary at this stage, and I am 11 bound by confidentiality not to discuss their nature. 12 Some are agents that are used for a variety of purposes 13 that have either some -- there is some question with 14 respect to whether they're toxic to bone marrow or not. 15 Some have been chemotherapeutic agents. 16 Q. The NIH grant that is currently underway, can 17 you give me the approximate annual funding level for 18 that? 19 MR. BELFOUR: Excuse me, Herschel. I don't 20 know whether it's appropriate for those kinds of 21 information -- that kind of information to be revealed. 22 I'll leave that up to Dr. Irons, because I don't know if 23 there are any confidentiality provisions in there. 24 MR. HOBSON: Well, I would hope that the 25 United States Government grant would not have any PATTERSON REPORTING & VIDEO 11 1 secrets in it about how much the United States is paying 2 Dr. Irons, but if there is something like that, speak 3 up. 4 A. Well, there are issues with respect to 5 overhead and indirect cost that make parity between 6 grants very difficult to compare. What I am trying to 7 do is determine what it is I am getting versus what the 8 institution is paying. 9 Q. (BY MR. HOBSON) Well, I truly was curious to 10 know what the overhead factor for your research is and 11 how it varies from grant to grant. 12 A. Some of that information I am not at liberty 13 to discuss. The overhead -- certainly the negotiated 14 overhead rate for the University of Colorado Health 15 Sciences Center with NIH is, I believe, 51 percent. I 16 believe, including consortium arrangements, that grant 17 is approximately at the level of 200,000 -- I see about 18 200,000 a year in direct cost. 19 Q. You say "in direct" cost as two words? 20 A. In -- those are direct costs, that's the 21 budget that I have to work with. 22 Q. Now, this NIH grant, is it more fund work than 23 just work that you are doing? 24 A. No. 25 Q. Are they activities of the university? PATTERSON REPORTING & VIDEO 12 1 A. No. 2 Q. The API grant, can you tell me approximately 3 what the funding level for that has been and is? 4 A. It's probably on the order of 170- to 180,000. 5 Q. Per year? 6 A. That I see. 7 Q. Per year? 8 A. Yes. 9 Q. And is there a different overhead rate for the 10 API grant? 11 A. Yes. 12 Q. The NASA grant, can you give me approximately 13 how much funding there is for that grant, or was? 14 A. That was approximately $100,000 a year, but it 15 had some very strange indirect cost issues, so I can't 16 -- I can't give you an accurate assessment of what my 17 annual budget was for that. It varied from year to 18 year. 19 Q. The NCI contract that ended, or is ending, 20 approximately what funding level did it reach? 21 A. That I -- I don't remember the specifics on 22 that. It was basically a supplies-based contract, so 23 the costs were relatively -- or the funding level was 24 relatively small. 25 Q. When you say that something is supplies-based PATTERSON REPORTING & VIDEO 13 1 for funding purposes, are you saying that there is no 2 money in that NCI grant for salaries? 3 A. There are no salaries -- I don't think there 4 is any salary money in the NCI grant. 5 Q. Would that be the only one of these grants for 6 which there is no salary funding? 7 A. Yes. 8 Q. Now, for you to take a contract that is salary 9 based -- I'm sorry, that is supply-based only, do you 10 have to get some special dispensation from your 11 authorities there at the university? 12 A. No. I pay overhead on it. 13 Q. I'm sorry. 14 A. I pay overhead on it. 15 Q. What does that mean? 16 A. The university is paid an overhead rate that 17 is within the guidelines for the university for indirect 18 cost recovery, so whether the funds are spent for 19 personnel or not is of no concern to the university. 20 Q. The CMA grant that you finished, can you give 21 me approximate funding level for that grant, please. 22 A. I think my annual budget was between 50- and 23 $70,000. 24 Q. Various pharmaceutical grants, I take it that 25 they vary depending on the work that you were asked to PATTERSON REPORTING & VIDEO 14 1 do? 2 A. Yes. On average, I would say they're about 3 $100,000 a year. 4 Q. That's for all of them or is that each? 5 A. That's on average, so at any given time. 6 Q. Now, when it comes to how you're paid at the 7 university, does the amount of research that you are 8 participating in and get funded for affect how much 9 you're paid? 10 A. No. 11 Q. It just depends -- affects whether or not you 12 get retained, I guess, or are you not expected as part 13 of your duties to entice research to the university so 14 that your department has funds to do research? 15 MR. BELFOUR: Objection, vague and compound. 16 I heard at least two questions, Herschel. I'm not sure 17 which one you want to ask him. 18 Q. (BY MR. HOBSON) Well, let me ask it again. 19 Is part of your job at the University of 20 Colorado to attract research moneys? 21 A. I have three major generic functions or 22 missions as a faculty member at the University of 23 Colorado: teaching, research, and service. And I am 24 expected to provide a balanced response to that mission 25 in all three categories. PATTERSON REPORTING & VIDEO 15 1 Q. And someone in your position who is unable to 2 attract research is less likely to be retained at the 3 university; is that fair? 4 A. Well, I have tenure, so that the actual amount 5 of research dollars that I bring in does not have any 6 bearing on my security or my salary. 7 The fact of the matter is that in an academic 8 environment, if you are not actively publishing in the 9 peer-reviewed literature, you basically don't exist, you 10 don't have -- that is a major role in demonstrating that 11 a faculty member is current, productive, and competent 12 in his or her area. 13 Q. Old adage of "publish or perish"? 14 A. Yes. And you can't publish without research 15 funds. 16 Q. What was the percentage of your time, say, in 17 the last year, that you spent actually teaching at the 18 university? 19 A. Probably on the order of 30 to 40 percent. 20 Q. And the percentage of your time you spent 21 doing research? 22 A. Probably on the order of 40 -- probably on the 23 order of 40 percent. 24 Q. And then your service time would be what? 25 A. 15 percent. Probably spent less this year, PATTERSON REPORTING & VIDEO 16 1 but that would be 2 Q. That, in my notes here, doesn't add up to 100 3 percent. Is there something else you do? 4 A. Yes, administration. 5 Q. What time do you spend doing administration? 6 A. Whatever the remainder is. That would be 7 committees and administrative responsibility. 8 Q. What would qualify in your duties as service 9 work? 10 A. Government advisory work, consulting either 11 with government or industry. 12 Q. Is there some limit that the university puts 13 on you for percentage of time for service work? 14 A. I have a contractual limitation of 15 percent. 15 Q. Maximum or must-do? 16 A. Maximum. 17 Q. Is that based on time spent or money? 18 A. Time. 19 Q. For doing work like the giving of the 20 deposition here today, is that university-related, then? 21 A. As a service, it is considered as fulfilling a 22 service component of my performance of activities as a 23 faculty member. It's independent of my contract as a 24 university employee. 25 Q. The moneys that you will be paid, then, for PATTERSON REPORTING & VIDEO 17 1 today's deposition go to whom? 2 A. To me. 3 Q. Not the university? 4 A. No. 5 Q. And your preparation work and other 6 consultation concerning this lawsuit goes then directly 7 to you and not through the university? 8 A. That's correct. 9 Q. Who sets the rates for your time? You or the 10 university? 11 A. I do. 12 Q. Do you bill that as an individual or do you 13 have a business entity set up to deal with that? 14 A. As an individual. 15 Q. Do you pay any parts of the moneys that you 16 get for this service work, doing depositions, and 17 working on lawsuits, to the university? 18 A. I compensate the university for services used, 19 such as library facilities, that type of activity. 20 Q. No percentage of what you bring in goes to the 21 university as fixed overhead similar to what the 22 university would charge in a grant or a contract? 23 A. No, sir. 24 Q. Let's talk a little bit about your research 25 activities in general. I know that -- at least I think PATTERSON REPORTING & VIDEO 18 1 I know part of your work has been in looking at myeloid 2 stem cells and how those myeloid stem cells and their 3 offspring react to benzene metabolites and drugs. Have 4 I said that correctly? 5 A. That's certainly part of what my work is 6 about. 7 Q. What I would like to do is understand where 8 you are in that research, as far as whether or not you 9 have been able to isolate the myeloid stem cells any 10 better than when I took your deposition before, which 11 was, I don't know, four or five years ago, just try to 12 understand how far you have come in that research of 13 knowing about myeloid stem cells and how the -- how they 14 function at the university-research level that you have 15 done. Does that make sense to you? 16 MR. BELFOUR: Excuse me, Herschel. Four or 17 five years ago in connection with which case, so perhaps 18 we could refer back to that? 19 MR. HOBSON: I don't even remember what it 20 was. 21 A. I don't remember. 22 MR. BELFOUR: If we don't have a reference 23 point, it's hard to say how it has changed since the 24 reference point. 25 MR. HOBSON: He can tell me how it is now. PATTERSON REPORTING & VIDEO 19 1 I'll understand the differences, I think. If not, I'll 2 ask him how it has changed. 3 A. It has changed considerably in the last four 4 or five years. It's changed dramatically in the last 5 couple of years. 6 Q. (BY MR. HOBSON) Where are you in isolating 7 myeloid stem cells and determining how the myeloid stem 8 cells develop into mature cells? 9 A. We can isolate -- we now have developed 10 technologies that allow us to isolate a defined set 11 populations of stem progenitor cells to virtually 12 99 percent purity based upon the criteria that we use. 13 We can distinguish within the stem cell compartment 14 subpopulations of early primitive stem cells. We can 15 distinguish committed myeloid progenitors and lymphoid 16 progenitors from uncommitted cells, and we are in the 17 process of looking at specifically the early events that 18 are associated with exposure to drugs and metabolites on 19 those cells. 20 Some of this work is -- much of this work has 21 recently been published; some of it is not published; 22 but certainly the purification schemes and cell-specific 23 metabolism and toxicity associated with myeloid 24 differentiation is published. 25 Q. And when you say the early stem cell PATTERSON REPORTING & VIDEO 20 1 progenitors, are you talking about, then, the level of 2 the kinds of cells that precede the myeloid stem cells? 3 A. Yes. Myeloid progenitor cell is a relatively 4 late cell in that process. 5 Q. How early, in getting back to the earliest of 6 the cell development, have you been able to go in 7 isolating these stem cells? 8 A. Well, we're in the process now of validating 9 some of our recent developments and results, but I would 10 say that arguably we're at a stage where we can 11 distinguish what is referred to as the long-term 12 repopulating cell from cells that have either 13 intermediate or short-term repopulating ability. So 14 we're beginning to actually subdivide what has 15 historically been referred to as the pluripotential stem 16 cell. It's not really one cell. It's a hierarchy of 17 cells. It's a heterogenous population of cells that 18 have varying propensities and potentials within that 19 compartment. 20 Q. As I read the literature that I see published, 21 some by you and some by others, I see different 22 terminology used which to me describes the same thing, 23 and so I have trouble sometimes following from one study 24 to the next, or even one conversation to the next with 25 the same person; that I am using the same terminology PATTERSON REPORTING & VIDEO 21 1 for the same item, and I don't want to make that 2 mistake. So I would like to take a minute, if we could, 3 and just agree on some terminology so I can try to 4 follow what you are telling me. 5 When you say "pluripotent stem cells," is that 6 a proper term? 7 A. The problem you're discussing is not one that 8 is unique to you. It's a product of a field in the 9 literature that has progressed explosively in the last 10 20 years, so terminology shifts regularly based upon 11 increased knowledge. 12 What you will see in the literature referred 13 to as a "stem cell," "pluripotent stem cell," 14 "totipotent stem cell," varies from person to person, 15 depending upon how current they are with the literature 16 and what particular point in time or history you're 17 reading whatever they're saying. 18 "Pluripotent stem cell" is a useful term. 19 Some people say "totipotent" is more current for 20 referring to the ultimate cell. I still use 21 "pluripotent" in that context, but as I have just said, 22 pluripotent stem cell really refers to a compartment, 23 depending upon what criteria you assign to that cell and 24 what the characteristics are. 25 I tend to also use "stem cell" to refer to PATTERSON REPORTING & VIDEO 22 1 that compartment, except that in certain environments, 2 such as talking to a high school class or a lay 3 population, I may use "stem cell" to mean a variety of 4 different cells; I would not necessarily use it for in a 5 technical setting. 6 Q. "Compartments," that's another term that I see 7 from time to time that I am not sure that I either 8 understand or can follow from one conversation to the 9 next or one article to the next. How do you use the 10 term "compartment"? 11 A. It varies, but basically "a compartment" 12 defines a set or category of characteristics. In the 13 case of stem cell or pluripotent stem cell compartment, 14 what I mean is that all the cells within that 15 compartment have nominally certain characteristics that 16 historically have been used to describe what was thought 17 to be a cell. They may differ with respect to certain 18 potentialities, but they all have certain 19 characteristics within that compartment. 20 You can use "compartment" to define 21 differentiation state or status. For instance, the 22 committed progenitor cell compartment could mean a 23 myeloid progenitor cell; it could mean lymphoid 24 progenitor cells. They're different cells, but they all 25 are at comparable stages in differentiation. PATTERSON REPORTING & VIDEO 23 1 One can say lymphoid compartment or myeloid 2 compartment, and there you're talking about a 3 lineage-specific compartment. And so basically you use 4 compartments to divide up the landscape in such a way as 5 to make it easier to discuss the characteristics of 6 cells that are within these compartments. There is no 7 universal -- for some there is some universal criteria, 8 but there is no absolute criteria for how you want to 9 define a given compartment. It depends upon how you are 10 using it. 11 Q. If you will bear with me and try to work 12 through this so I can try to come to some understanding 13 of how you use the terms, then hopefully we will have 14 some level of communication. It won't be probably your 15 fault, it will be more mine that there won't be any 16 communication, but if you will bear with me, I would 17 appreciate it. 18 If we talk about the pluripotent stem cell 19 compartment -- first of all, is that a proper 20 terminology? 21 A. I would say so, yes. 22 Q. Then that would be a set or a collection of 23 cells, as you use that term? 24 A. Yes. 25 Q. Would all of those cells in the pluripotent PATTERSON REPORTING & VIDEO 24 1 stem cell compartment be noncommitted? 2 A. Yes. 3 Q. I take it there comes a time, then, when the 4 pluripotent stem cells, as they develop, they do commit? 5 A. Yes, and I use -- commitment is consonant with 6 movement into progenitor cell compartment. 7 Q. How many progenitor cell compartments would it 8 be appropriate to talk about, then? 9 A. In the simplest -- in the simplest -- simple 10 is not always the most accurate; but in the simplest 11 context, lymphoid and myeloid; but within the myeloid 12 compartment, you have multiple levels of differentiation 13 and commitment. The myeloid compartment contains cells 14 that are variously referred to as multilineage cells. 15 The myeloid progenitor cell is a multilineage cell. At 16 one point in history many people thought it was the stem 17 cell. It's capable of giving rise to erythrocytes, 18 ultimately erythrocytes, granulocytes, macrophages, 19 platelets. 20 The lymphocyte progenitor cell compartment is 21 understood at the mature end, if you will, probably 22 better than it is at the immature end, in some degree 23 to some degree. 24 There are some subtle differences between the 25 apparent differentiation of lymphoid cells among PATTERSON REPORTING & VIDEO 25 1 species, so the structure of the lymphocyte progenitor 2 cell compartments differs from one species to another. 3 It certainly appears to be. 4 Q. Now, you said that in simple terms there are 5 these two committed progenitor cell compartments: the 6 lymphoid and the myeloid? 7 A. Yes. 8 Q. Now, is there a possible third progenitor cell 9 compartment or have we pretty much isolated the two cell 10 compartments? 11 A. There is a potential -- well, it depends on 12 how far back you want to go on ontogeny. There is some 13 evidence that there is a third compartment early on with 14 respect to differentiation of stroma and fibroblast 15 during embryogenesis, that is a more primitive cell than 16 the hematopoietic pluripotent stem cell. There is, 17 depending on the species, evidence to suggest that the 18 initial committed progenitor cells for lymphoid 19 differentiation between B and T cells is different, and 20 that those are not necessarily analogous compartments. 21 So if we're talking about committed progenitor 22 cell compartments, you can either think of it as being 23 two or three, depending upon how you view lymphoid 24 differentiation. 25 Q. If we're talking about for humans? PATTERSON REPORTING & VIDEO 26 1 A. Yes. 2 Q. Have you described for me this pluripotent 3 stem cell compartment for the noncommitted cells and 4 then the lymphoid and myeloid progenitor cell 5 compartments for humans, or is it different? 6 A. In humans it appears that -- well, the myeloid 7 -- commitment to myeloid differentiation in humans is 8 as I have described. 9 In the case of lymphoid differentiation, there 10 are some -- there is indirect evidence, although it's 11 fairly compelling evidence, to suggest that B cells 12 undergo a defined commitment step to a progenitor cell 13 -- a B lymphoid progenitor cell. The T cells may not; 14 and, in fact, T cells may directly -- or may be derived 15 directly from a pluripotent stem cell or that that 16 differentiation process takes place in the thymus and 17 not the bone marrow. 18 Q. Are you saying there are two pathways then to 19 this T cell? 20 A. It's either/or. We have evidence in mouse of 21 a commitment step at the level of the bone marrow. In 22 humans, so far no one has been able to demonstrate 23 evidence of a commitment step at the level of the bone 24 marrow. 25 Q. You talked from time to time about murine PATTERSON REPORTING & VIDEO 27 1 systems? 2 A. Murine. 3 Q. There you mean rodents? 4 A. Mouse. 5 Q. I must say that as I have gone back and looked 6 at other depositions you have given, sometimes I have 7 trouble understanding in your answer if you are talking 8 about the human or if you are talking about the mouse, 9 and I don't know sometimes if the questioner realizes 10 which one you're answering. So that's what I would like 11 to try to work through that with you too if I could. 12 For humans have you identified any of the 13 components of the pluripotent stem cell compartment? 14 A. In terms of functional characteristics and 15 differentiation, yes. That's been a major focus of 16 ours, has been the reproducible isolation of 17 subpopulations of human stem cells for characterization 18 of function and metabolism and toxicity. 19 Q. But they're pluripotent stem cells, not 20 myeloid or lymphoid? 21 A. That's correct, that's correct. 22 Q. And can you now in your laboratory isolate the 23 pluripotent stem cells to the same, like this 99 percent 24 purity that you were talking about earlier? 25 A. Yes. PATTERSON REPORTING & VIDEO 28 1 Q. And how far have you been able to describe the 2 development of the pluripotent stem cells and isolate 3 them? 4 A. We're in the process of taking the earliest 5 cell population we can define by differentiation 6 characteristics and to validate its function as what is 7 referred to as a long-term repopulating cell. 8 In the current parlance of experimental 9 molecular hematology, there is less focus on pluripotent 10 stem cell and much more focus on function; and the 11 ultimate function that is of concern with respect to 12 functionality today is what is referred to as long-, 13 intermediate- or short-term repopulating ability, 14 because these are the characteristics that are important 15 in bone marrow transplantation. 16 Q. People that have diseases like leukemia, where 17 you can essentially eliminate their blood system, and 18 then the hope is to put in this cell and generate a 19 whole new bloodlike condition that does not have 20 leukemogenic properties? 21 A. That's correct. 22 Q. Are you at that stage? 23 A. Yes, yes. As I said, we're in the process of 24 validating our purification schemes at the level of 25 long-term repopulating cells now. PATTERSON REPORTING & VIDEO 29 1 Q. What do you mean by the difference -- that's a 2 bad question. 3 What is the difference between long-term 4 repopulating and short-term repopulating, as you use it? 5 A. It's probably best discussed in the history of 6 the field of transplantation. Early on it was 7 recognized that there was a subpopulation of bone marrow 8 cells that obviously contained the stem cell population. 9 Bone marrow transplantation originally involved simply 10 the -- and still, in many cases today, does involve the 11 transplantation of whole bone marrow, which obviously 12 contains some stem cells, and it's an effective 13 approach. But in terms of trying to enhance and improve 14 the ability to successfully transplant bone marrow, 15 there has been a major effort to characterize the 16 function of those cells and to purify them. 17 Both in terms of animal studies, and certainly 18 indirectly in human studies, it's been possible to 19 demonstrate that cells within the so-called stem cell 20 compartment have different characteristics with respect 21 to their ability to repopulate a host. 22 If you lethally irradiate a mouse and/or use 23 radiation and chemotherapy to essentially kill off his 24 native bone marrow, it was discovered relatively early 25 on in the -- well, the late '80s, that there was PATTERSON REPORTING & VIDEO 30 1 there were cells that had to be there if that animal was 2 going to survive for six months or a year, but if those 3 cells were the only ones that were given to the mouse, 4 he wouldn't survive for two weeks or a month or three 5 months. 6 And it's been found that there, in fact, are 7 populations of cells that are different with respect to 8 their ability to provide survival following lethal 9 irradiation for different periods of time. So the most 10 primitive cell is one that is capable -- is called a 11 long-term repopulating cell, and that's basically, by 12 definition, a cell that will enable the animal to 13 eventually completely recapitulate normal bone marrow 14 function and survive indefinitely. 15 There are shorter-term repopulating cells that 16 are necessary for immediate survival, but they're not 17 capable of indefinite or perpetual maintenance of bone 18 marrow function. So you really need a heterogeneous 19 population of cells to provide that function if you are 20 going to effectively reconstitute bone marrow. 21 Those are the characteristics that we're using 22 now functionally in terms of providing validation for 23 the subpopulations that we're looking at in the 24 laboratory. 25 Now, what I defined for you is what has been PATTERSON REPORTING & VIDEO 31 1 seen in mouse. We have now assays that are surrogates 2 for the same type of function in humans, so that we can 3 look at the ability -- to a certain extent, we can look 4 at the ability of stem cells -- human stem cells to 5 provide what is called long-term repopulating ability in 6 culture that appears to be analogous to what we see in 7 terms of the isolation of cells that are capable of 8 repopulating, for instance, a mouse. 9 Q. When you say "culture," you're talking about 10 outside the human body? 11 A. Long-term culture, yes. 12 Q. Now, the long-term repopulating cells, are 13 they still in one of the committed progenitor 14 compartments? 15 A. Yes, they're way back. They're at the very, 16 very beginning of the pluripotent stem cell compartment. 17 Committed progenitor cells are not long-term repopulated 18 cells. They're virtually -- they play some role in the 19 intermediate repopulation, but they're not even what I 20 think you could refer to as intermediate repopulating 21 cells. They have a fixed short-term life-span on the 22 order of a few weeks at the most. 23 Q. Say that again about the life-span. 24 A. On the order of a few weeks. 25 Q. The long-term -PATTERSON REPORTING & VIDEO 32 1 A. Short-term. Short-term are certainly no 2 longer than a few weeks. 3 Q. I'm sorry, that's what I missed. 4 Now, the short-term repopulating cells, do 5 they fall in one of these compartments? 6 A. They -- well, again, here there is some 7 fuzziness with respect to the difference between 8 committed and functionality. Committed progenitor cells 9 have some short-term repopulating ability. If you give 10 committed myeloid progenitor cells to an individual or a 11 mouse, they're going to immediately make cells of the 12 granulocytes. So they participate very early on in that 13 process, but they fall within -- certainly in the mouse, 14 within the 14-day window, and in humans probably 15 comparable to that. 16 Q. Can you tell me if there is a difference in 17 how thoroughly you have been able to, I guess, 18 characterize or identify the progenitor cells in these 19 committed compartments, compare the lymphocyte or the 20 lymphoid progenitor cells to the myeloid? In other 21 words, can you do one better than the other, or have you 22 got them both pretty well lined out, in your view? 23 A. Well, we have asked different questions, 24 depending upon the results that we have gotten. We have 25 certainly characterized the myeloid progenitor cell PATTERSON REPORTING & VIDEO 33 1 quite well in terms of functionality, ability to clonate 2 -- the ability to form clones of cells in response to 3 growth factors of cytokines. We certainly know their 4 response to exposure to a number of agents, certainly 5 the metabolites of benzene. We have followed and are 6 currently tracking the response of cells in the 7 pluripotent stem cell compartment as they track into the 8 myeloid compartment and the effects of these agents on 9 those. 10 In the case of myeloid differentiation, we've 11 been able to characterize cell-specific metabolism and 12 the enzymes that are present as a function of 13 differentiation in those cells which confer upon the 14 myeloid progenitor susceptibility to an agent such as 15 benzene. 16 We've been able to characterize and contrast 17 some of the enzymes that are present in myeloid cells 18 from lymphoid cells, and we've been able to characterize 19 and contrast the functional clonogenic responses of 20 lymphoid cells versus the myeloid progenitors in terms 21 of exposure to many of these agents. 22 We still have further studies to do in terms 23 of looking at cell-specific metabolism for a variety of 24 other types of processes in lymphoid cells. 25 Q. You mentioned that for the myeloid line, that PATTERSON REPORTING & VIDEO 34 1 you have done quite a bit of work with the -- or at 2 least some work with the metabolites of benzene? 3 A. Yes. 4 Q. Have you also done the same work with the 5 lymphoid or lymphocyte compartment? 6 A. Yes and no. We have focused primarily on the 7 myeloid progenitor cell differentiation process because 8 that's where we have seen effects, but it's not fair to 9 say that we haven't looked at early events in lymphoid 10 differentiation, because we have, to the extent that 11 these agents have not produced demonstrable alterations 12 in those cells. 13 We have looked at cell-specific metabolism in 14 the myeloid progenitors, because there are very specific 15 enzymatic changes that occur consonant with commitment 16 to myeloid differentiation that are important for the 17 secondary or tertiary metabolism of benzene. We don't 18 see those enzymes present in lymphoid progenitors or 19 lymphoid cells. We have also seen -- we see a different 20 pattern with respect to the enzymology of those cells. 21 We have also seen direct changes in 22 differentiation that are functionally associated with 23 myeloid differentiation that don't occur on the other 24 side with respect to lymphoid differentiation. 25 Subsequent to those studies, we have devoted PATTERSON REPORTING & VIDEO 35 1 most of our time to looking at myeloid differentiation 2 because that's where the action is and that's where we 3 see effects. 4 As I said, there are still other studies that 5 we have in mind to do to look further at lymphoid 6 differentiation, although we're looking at, at this 7 point, a lack of effect rather than positive responses 8 in these systems. 9 Q. Have you also looked at metabolites of benzene 10 and their interactions with cells from the pluripotent 11 stem cell compartment? 12 A. Yes, and we're in the process of doing that 13 now. 14 Q. No results yet? 15 A. I have some results. They're not published at 16 the present time, but yes, we do have some results. 17 Q. Do you see any effects? 18 A. Yes, we see some fairly profound effects 19 associated with shift in differentiation toward myeloid 20 cells. The first effect we see is a change in the 21 pattern of differentiation that occurs in these cells, 22 and it's consistent with a push into the granulocytic 23 committed compartment. 24 Q. When we talk about the effects of the 25 metabolites of benzene on either the myeloid, the PATTERSON REPORTING & VIDEO 36 1 lymphoid, or the progenitor cell compartments, what 2 effects are you looking for? If you could just kind of 3 tell me what those are. 4 A. Well, a pattern has been observed and 5 characterized that is very demonstrable in the evolution 6 of secondary acute myelogenous leukemia. So there are a 7 pattern of changes that have been described, both 8 functional and genetic, that occur in the progression 9 and development of AML, secondary to exposure to 10 chemotherapy or to benzene, that define stages or events 11 that appear to be very important within the process. 12 One of those is a shift in response to growth factors, 13 so the cells respond differently to specific growth 14 factors. In particular, one appears to be an early 15 event and characteristic of agents that are associated 16 with the production of AML. 17 There are also changes that occur with respect 18 to differentiation or the maturation process that are 19 consistent with functional changes in cytokine response 20 that we're looking at, and these are events that precede 21 -- that apparently precede the types of changes that 22 are associated with fixed or permanent changes, such as 23 cytogenetic alterations, where again, in secondary AML, 24 there is a fixed pattern that has been demonstrated that 25 appears to be very strongly expressed in secondary PATTERSON REPORTING & VIDEO 37 1 leukemias. 2 Q. You mentioned that there are growth factors. 3 How do you use that term in this context? 4 A. Technically, we're talking about cytokines. 5 These are low-molecular-weight molecules that function 6 as, if you will, intercellular hormones, or these are 7 molecules that allow developing cells to communicate 8 with each other and that regulate the differentiation 9 process. 10 Q. That regulate the differentiation process? 11 A. They regulate differentiation and growth. 12 There are cytokines that act very -- there are cytokines 13 that appear to be responsible for survival of stem 14 cells. 15 There are cytokines that are called early 16 cytokines in that they're responsible for maintaining 17 early response -- the response of early cells in the 18 commitment process. 19 There are cytokines that are of intermediate 20 level of differentiation that support the growth of 21 maturation of cells that are committed -- that have just 22 been committed. 23 And then there are late-acting cytokines that 24 are primarily associated with late -- with end-stage 25 maturation of committed cells, single-lineage cells. PATTERSON REPORTING & VIDEO 38 1 And then there are a variety of cytokines that 2 cannot do any of those but can influence the behavior of 3 those I just talked about. 4 So it's a fairly complicated paradigm, and if 5 you're going to look at the functionality of these 6 things, you have to look at them -- basically, it's an 7 elaborate dissection and reconstitution process, where 8 you look at how they behave individually and how they 9 behave together. 10 Q. My mind was conjuring up all kinds of 11 questions to ask you as you went along, and I probably 12 will tend to jump around a bit. 13 What you're telling me that you've been able 14 to find, is this now for rodents or is this for humans? 15 A. There are some subtle differences between 16 rodents and humans, and this certainly is not all our 17 work. It's been known for some time that survival 18 there are subtle differences in cytokines that appear to 19 be effective at promoting survival of stem cells. In 20 the mouse you're looking at a cytokine called 21 interleukin-3, and one called stem cell factor, which 22 appears to play a fairly important role as far as the 23 mouse is concerned. 24 In humans it's interleukin-3 and GM-CSF, or 25 granulocyte/macrophage colony-stimulating factor. Stem PATTERSON REPORTING & VIDEO 39 1 cell factor doesn't appear to be a major player in 2 survival of early stem cells in human. 3 As one gets further downstream, the 4 selectivity and the responses of cytokines appear to be 5 fairly comparable between the species. 6 Q. One of the thoughts that came to my mind that 7 I am sure you can answer for me is that in, I guess, a 8 healthy human being or healthy mouse, this system of the 9 blood and its regeneration has a constant ongoing 10 effect, right? 11 A. Yes. 12 Q. And so you have got pluripotent stem cells 13 that are differentiating and developing into those that 14 go down the lymphoid line and those that go down to the 15 myeloid line, and all this is going on simultaneously 16 within the body, right? 17 A. Yes. 18 Q. Did you say that the metabolites of benzene 19 tend to shift the production of the pluripotent stem 20 cells from or more toward the myeloid compartment then? 21 A. More toward myeloid and away from erythroid, 22 so that there is what is an apparent stimulation of 23 granulopoiesis, or the production of granulocytes. This 24 has been reported in animals before. It's been 25 interpreted variously by different authors, but there is PATTERSON REPORTING & VIDEO 40 1 this -- there is evidence out there of this quizzical 2 increase in granulopoietic activity in hematopoietic 3 tissue from mice exposed to high levels of benzene, 4 which has not been entirely understood. 5 And what we're seeing is, there is a shift 6 away from erythroid and a shift toward granulocytic 7 granulation. 8 Q. You say that for humans, as well? 9 A. Well, yes. 10 Q. When you say that for humans, as well, is this 11 in a test system or in human beings? 12 A. It's in a test system. 13 Q. Now, it occurs to me that if you have got a 14 living entity that has a need for myeloid and lymphoid 15 cells being produced on a day-to-day basis to survive, 16 and you create a shift to the myeloid compartment and, 17 as you say, more to the granulocytes, does that not 18 stress then the lymphoid side or the lymphoid 19 compartment? 20 A. Not necessarily. A priori, there would be no 21 reason to reach that conclusion. It depends on the 22 nature of -- it depends on the nature of the insult and 23 the exposure. It also depends upon the duration of 24 exposure. 25 And what I think is ultimately where you may, PATTERSON REPORTING & VIDEO 41 1 in effect, have that situation develop -- and this is 2 this is at this point speculative; we certainly don't 3 have sufficient data to make a definitive statement 4 but there is some evidence that at the earliest stages 5 of stem cell development there is a promotion of 6 differentiation. That may very well play a role in the 7 eventual failure of the bone marrow -- that is to say, 8 in the case of the evolution of a refractory anemia or 9 aplastic anemia, if you will. It may play a role there, 10 because you may, in fact, be shifting the propensity of 11 the earliest cells to stay undifferentiated and capable 12 of recapitulating. Whether or not that has -- we 13 certainly at this point don't have any evidence that 14 there is downstream any specific stress on the lymphoid 15 system. 16 As a matter of fact, we know from animal 17 studies, as well as human exposures to benzene, that, 18 for example, erythrocytes are a relatively late effect 19 in the pantheon of toxicities that are associated with 20 bone marrow toxicity from benzene; certainly not an 21 early effect. And yet we have evidence that there is a 22 primary shift away from erythropoiesis that occurs 23 functionally at the level of those cells, but it is not 24 enough to produce an anemia in any short period of time; 25 and in fact, the animals and humans can survive PATTERSON REPORTING & VIDEO 42 1 following fairly egregious exposures to benzene for long 2 periods of time without demonstrating any clinically 3 significant toxicity. 4 Q. Tell me, for your research that you have done 5 involving the metabolites of benzene, how you carried 6 out these various studies of evaluating the effects of 7 the metabolites on the different pluripotential stem 8 cells and the myeloid and the lymphoid. In other words, 9 do you isolate these particular cells out and conduct an 10 experiment? Do you test more than one at one time? I 11 mean, how does that work? 12 A. It depends upon the nature of the specific 13 question we're asking or the study that we conducted. 14 Early on in murine studies the first thing we did was 15 look at clonogenic activity in response to defining 16 cytokines. So we were looking at their functional 17 ability to form colonies of cells that differentiated 18 and formed the specific types of lineages we're talking 19 about. 20 We started out with whole bone marrow, without 21 any purification scheme, and we used those functional 22 clonogenic activities as the definition of what the cell 23 populations were which we were looking at. So we were 24 indirectly looking at cells by looking at their function 25 and by looking at the effects of various metabolites or PATTERSON REPORTING & VIDEO 43 1 combinations of metabolites on individuals to find 2 cytokine-stimulated clonogenic activity so we could say 3 something about the functions that were being affected. 4 After seeing very selective effects on myeloid 5 differentiation in that system, we went to -- we began 6 to employ purification techniques to hone in on the cell 7 populations and to allow us to be more focused on the 8 physical cells, as well as their individual function. 9 At that point, we began moving into humans and 10 invested a great deal of effort developing the 11 purification schemes we were talking about, because in 12 humans it's much more important in looking at 13 cell-specific function to be able to isolate these cells 14 for a number of reasons. 15 First and foremost, when we do a study in a 16 mouse, we have the ability to completely evacuate the 17 bone marrow, so we don't have to look at a sample of 18 mouse cells. We can look at the universe of -- totality 19 of bone marrow cells from a mouse. We can take a femur 20 and look at every cell that is in it. 21 In humans, we are taking aspirates of femoral 22 bone marrow. They are representative of the bone 23 marrow. And in order to provide for standardization 24 from one sample to another and one person to another, we 25 had to develop purification procedures that allowed us PATTERSON REPORTING & VIDEO 44 1 to be looking at the general population of stem 2 progenitor cells and to know how many we had there from 3 one sample to another. We also need to subfractionate 4 those into defined populations if we're going to look at 5 metabolism and look at individual function. 6 So we began our human studies with the 7 background we had from mice, looking at individual 8 the effects of individual metabolites and compounds on 9 cytokine response, then on semipurified cells, and then 10 into humans, looking at semipurified and now highly 11 purified cells, cytokine response, their ability to form 12 colonies, differentiation, antigen expression, and 13 expression molecules on their surface, as well as the 14 functional characteristics. 15 I hope that was responsive. 16 Q. Well, I think it was. I just wanted to make 17 sure you finished. 18 Now, obviously from the description of the 19 grants and contracts that you told me about, one of the 20 interests that you have had is the endpoint of 21 leukemia. 22 A. Yes. 23 Q. How do you go from these observations of 24 functionality, and any other parameters that you have 25 told me about, to knowing when you have got a cell that PATTERSON REPORTING & VIDEO 1 is going to be leukemogenic, or have you gotten that 2 far? 3 A. We know from -- I mentioned briefly before 4 there is a very strong pattern associated with the 5 development and progression of dysplasias and leukemias 6 that are secondary to exposure to chemotherapeutic 7 agents or benzene. That, superimposed on our general 8 knowledge of cancer biology today, and of the types of 9 events that occur in the progression and development of 10 cancer, gives us -- allows us to set forth hypotheses as 11 to what we would expect to see at a given stage in the 12 leukemogenic process, in the development of leukemia. 13 We are in the process now of looking at the 14 individual steps in that which occur early on. For 15 instance, one would predict from in vivo animal data, 16 from human data associated with exposure to 17 chemotherapeutic agents, the early events associated 18 with altered cytokine response as an early effect. One 19 can see changes associated with functionality of cells 20 that predispose towards the types of lesions that are 21 seen in the progression of AML, and we have then used 22 those changes to generate hypotheses as to how the cells 23 should behave in culture. 24 And what we found so far is that there is very 25 good concordance between the very earliest events that PATTERSON REPORTING & VIDEO 46 1 are seen in culture and events that are predicted by 2 in vivo experiments in mice, as well as in vivo/in vitro 3 studies of patients that have developed dysplasias 4 secondary to chemotherapy. 5 Q. Let's see if I can distill that in my mind a 6 little bit. 7 If we take a human being who has leukemia, and 8 let's say acute myelogenous leukemia, you can find in 9 that person's circulating blood a leukemogenic cell; is 10 that right? 11 A. I think the way you have used the terminology 12 may be misleading. 13 Q. Help me with terminology so that I would use 14 or could use terminology you would use. 15 A. A patient who has developed leukemia is going 16 to have circulating leukemia cells. There are going to 17 be, for that given place in time in the natural history 18 of that patient's disease, abnormal cells that are what 19 we refer to as leukemia cells. They are -- they 20 originate. They may not, however, represent precisely 21 the cells that the leukemia originated from. So in 22 other words, the bone marrow cells that are the -- the 23 original cell -- and I should say cell, because 24 virtually all human leukemias are monoclonal. The 25 original cell, certainly for the vast majority of PATTERSON REPORTING & VIDEO 47 1 secondary AMLs, can be found to originate at the level 2 of the committed myeloid progenitor, so there is a 3 committed myeloid progenitor cell at the apex of this 4 amplification scheme that is associated with the clonal 5 development of an AML. 6 In the case of CML, or chronic myelogenous 7 leukemia, it's a cell within the pluripotent stem cell 8 compartment. Now, some individuals will argue with that 9 and say that even in CML, it's not a pluripotent stem 10 cell; however, in terms of clonal origin, it is a cell 11 with bifunctional characteristics that can give rise to 12 either lymphoid or myeloid cells. So whether or not it 13 is in fact, a long-term repopulating cell or an 14 intermediate-term repopulating cell is not of any 15 significance with respect to defining its 16 bifunctionality. 17 In the case of myeloid cells, there is 18 certainly, for the vast majority of cases, evidence that 19 we're dealing with a multi -- a cell of multilineage 20 potential that is restricted to the myeloid pathway. 21 Those cells are the cells that are the neoplastic 22 transformed cells. They give rise to abnormal cells. 23 They can also replicate. They may have longer life 24 spans than the normal cells and what we normally think 25 of as leukemia cells, but the lesion originates at the PATTERSON REPORTING & VIDEO 48 1 level of the progenitor cells that I am talking about. 2 So when you talk about a leukemogenic cell in 3 the peripheral blood, you're dealing with a leukemia 4 cell; but it is not, by definition, the originating 5 the site of clone origination of the tumor. 6 Q. I have heard a lot of things that I need to 7 sort out with you, if I could. 8 If we could stay, for instance, for a moment 9 here with AML. 10 A. Okay. 11 Q. Because we both know there are lots of kinds 12 of leukemia, it might help me if we could stick with one 13 for a minute. 14 Are you saying that -- let me back up. 15 To have an acute myelogenous leukemia, there 16 has to be a cell which is defective that is capable of 17 reproducing? 18 A. Yes. 19 Q. And I think you said that since leukemia 20 almost all human leukemias are monoclonal, that means 21 that there had to have been, at least in the beginning, 22 one cell that was defective, i.e., had a lesion, that 23 led to the leukemia in the individual. Would that be 24 accurate? 25 A. It has to have a number of lesions. PATTERSON REPORTING & VIDEO 49 1 Q. And all of those lesions have to be in one 2 cell? 3 A. They have to occur not necessarily all in 4 precise order, but there are a number of independent 5 events that must occur in the clonal progression of that 6 lesion that all intersect with one cell. One cell has 7 to basically have all those events occur before you 8 arrive at the stage where it is either a dysplastic 9 cell, which is on the same line, or a leukemia cell. 10 And those changes are associated with functional changes 11 in the survival of that cell, its response to growth 12 factors, and its differentiating potential. 13 Q. So help me again. If we're talking about AML, 14 then we have -- at some point in time there is a cell 15 that is defective with this proper series of lesions, 16 and this cell will replicate itself to produce a bunch 17 of defective cells that become AML in a human being? 18 A. Yes. 19 Q. Now, that defective cell that is at the 20 beginning point of AML, its precursor lineage that got 21 at all these lesions at just the right point in time of 22 its development to become AML, where do those lesions 23 occur in the generation of that cell from its earlier 24 lineage? 25 A. Certainly in terms of the clonal evolution of PATTERSON REPORTING & VIDEO 50 1 the tumor, which means you are assuming that the cell 2 the point at which the cell becomes malignant and the 3 clonal origin of the tumor, okay. That cell, for the 4 majority of cases that have been studied, certainly for 5 secondary leukemias, is at the level of a multilineage 6 myeloid progenitor cell. 7 Q. Say the last sentence again. 8 A. Certainly for the clonal evolution of the 9 tumor, that cell is at the level of a multilineage 10 myeloid progenitor cell. 11 Q. Multilineage myeloid progenitor cell? 12 A. It can be as early as a cell that can go 13 erythroid, myeloid, monocytic, thrombocytic. It may be 14 a cell that is committed to granulocytic/macrophage-type 15 cells, but is basically within that span or spectrum for 16 the vast majority of cases. 17 Q. And so that cell may be even in the 18 pluripotential stem cell compartment? 19 A. A small number of -- in a small number of 20 cases there is some indirect evidence to suggest that is 21 possible. In the majority of cases that have been 22 studied, certainly for secondary leukemias, there is no 23 evidence that that cell has pluripotent characteristics. 24 It appears -- I would say roughly 85 percent of the 25 weight of evidence associated with characterization of PATTERSON REPORTING & VIDEO 51 1 those suggest they are committed to myeloid -- they are 2 myeloid progenitors. One cannot rule out in a 3 subpopulation of AMLs, per se, that there could be 4 involvement of a pluripotent cell, but there is not 5 substantial evidence to support that notion. There is 6 much more evidence to support its committed progenitor. 7 Now, all these lesions we were talking about, 8 or events, have already occurred, because we're talking 9 at that point the clonal evolution of a neoplastic line 10 or clone. So we're talking about events that have 11 preceded that. 12 Q. Is what you're saying now is that some of 13 these lesions that appear in the committed myeloid 14 progenitor cell that becomes the source of the leukemia, 15 some of those lesions may have actually -- or a portion 16 of the lesions may have actually occurred while this 17 cell was still a progenitor stem cell or in the 18 progenitor stem cell compartment? 19 A. That was certainly one of the questions that 20 we set out to ask early on. That's why we were 21 interested, and still are, in early events, because 22 quite frankly, there are a number of independent steps 23 that have to occur in the evolution of a leukemia, but 24 the earliest ones appear to be the most important with 25 respect to the potential to evolve into leukemia in the PATTERSON REPORTING & VIDEO 52 1 first place. It's kind of a numbers game. 2 If independent steps have to occur, the 3 probability that those steps are going to occur changes 4 with respect to what has already happened, and it can 5 influence what is going to happen downstream. So the 6 earliest events are .most important. 7 I would say that the sum total of the work 8 that we have done over the last three years does not 9 support the concept you have just described in terms of 10 a random -- or implied with respect to random events 11 that may occur at the level of pluripotent stem cells. 12 The earliest event we see is a -- which is not a 13 cytotoxic event at all. This is a regulatory response, 14 and the presence of normal cytokines is a shift in 15 differentiation or promotion in differentiation toward 16 myeloid. 17 So if I was going to put it in a -- if I was 18 going to try to put it into some layman's context, it 19 would be more like Marshall McLuhan's, "The Medium is 20 the Message." If the first response that a primitive 21 stem cell has is to go myeloid, the first thing we see 22 is a shift in differentiation. So at that point in time 23 it may not be a normal myeloid progenitor. I can't tell 24 you that, but it certainly thinks it's a myeloid 25 progenitor, and it is acting like a myeloid progenitor. PATTERSON REPORTING & VIDEO 53 1 Q. Are you at the point, then, that you can say 2 that it is biologically implausible to have any kind of 3 secondary leukemia other than an AML? 4 A. I would say that -- based on our current 5 understanding of the processes that are involved in the 6 evolution of leukemia, that the evidence would support 7 that, that it is biologic -- we see no plausibility for 8 other major types of leukemia. That does not mean at 9 this point that we can say, absolutely never, it can't 10 be, but certainly all the evidence points to the -- to 11 that being the case. 12 I think that's also consonant with the 13 epidemiologic literature in that it explains -- merely 14 explains what we see, by far and away, the case in terms 15 of secondary hematopoietic neoplasms following either 16 exposure to chemotherapeutic agents or benzene. 17 If you take all -- if you don't critique the 18 validity of any given study, but you simply add up all 19 the cases that have either been hypothesized as case 20 reports or as quantitative retrospective epidemiology 21 studies or case-control studies for secondary leukemias, 22 you end up with virtually -- I did this exercise once 23 recently -- I think you end up with about 96 percent 24 AMLs, or myelodysplastic syndromes; and everything else 25 that has ever been hypothesized falls into that 4 PATTERSON REPORTING & VIDEO 54 1 percent range. 2 So you have got an overwhelming body of 3 evidence suggesting a predisposition towards AML as the 4 principal, if not the only, secondary leukemia type 5 associated with chemotherapy or occupational exposure. 6 MR. BELFOUR: Excuse me, Herschel. Could we 7 assume that your question which you last asked was 8 directed toward benzene and chemotherapeutic agents? 9 As I recall, you asked if it was biologically 10 implausible to have a secondary leukemia other than AML, 11 and I didn't hear a qualification on that, but Dr. Irons 12 answered, I think, with regard to benzene and 13 chemotherapeutic agents. I don't want there to be a 14 misunderstanding about what you were asking. 15 MR. HOBSON: I think you stated the question I 16 asked. I don't know how that affects the answer you 17 gave, but I think you restated the question I asked. 18 A. I may have missed -- I may have mistaken the 19 generic nature of your question. I was speaking 20 specifically to exposure to alkylating chemotherapeutic 21 agents and to occupational exposure to solvents, of 22 which benzene is the only known leukemogen. 23 If we invoke radiation, we're talking about a 24 totally different story, and radiation -- and there are 25 explanations that I think provide some understanding as PATTERSON REPORTING & VIDEO 55 1 to why -- and that radiation is capable of and, in fact, 2 tends to find resting cells more susceptible than 3 dividing cells, whereas the types of lesions that are 4 seen in the progression of AML associated with exposure 5 to chemicals or drugs are division-dependent processes. 6 Without division, they don't happen. 7 Q. (BY MR. HOBSON) In regard to radiation, are 8 you saying, then, that a cell that is at rest is more 9 susceptible to radiation, if cancer is the endpoint, 10 than one that is dividing? 11 A. It actually can be, yes. Certainly, in terms 12 of the stem cell population, that's the case. That's 13 one of the reasons why, for instance, you see a very 14 respectable incidence of CML following radiation. CML 15 occurs in a cell that, for the most part, is a resting 16 cell, and radiation is effectively doing that. 17 Q. Is it a function of the resting cell being 18 more susceptible or is it a function of just the 19 statistics of how many of the stem cells are actually 20 replicating as opposed to being at rest? 21 A. No, it appears to be more a function of 22 resting cells being more susceptible to 23 radiation-induced damage. 24 Q. What research do you have that points to 25 that? How can you say that? PATTERSON REPORTING & VIDEO 56 1 A. This goes way back. This is work that began 2 in the '50s and has been the subject of a number of 3 studies over the years that have looked at the 4 relationship between radiation-induced damage and 5 chemotherapeutic or chemically induced damage. The 6 chemotherapeutic agents target cycling cells. That's 7 the whole paradigm; that's the whole goal in the 8 selection and design of chemotherapeutic agents: It's 9 to spare stem cells and to target cycling cells. 10 Radiation does not show that kind of selectivity. 11 Radiation obviously will produce damage to replicating 12 cells, but there is evidence to indicate that in certain 13 replicating-cell populations, they're more resistant to 14 radiation-induced damage than the resting ones. 15 Q. Doesn't that depend on where the cell is in 16 this progenitor line? 17 A. To a certain extent, yes, but if you want to 18 target a stem cell, you don't use a chemotherapeutic 19 agent. You use, by far and away, a combination of 20 radiation/ chemotherapy. 21 If you want to leth -- I mean, the classic 22 model in the mouse is lethal irradiation. That's the 23 model for either bone marrow -- for looking at 24 reconstitution or bone marrow transplantation. 25 Now, effectively, you can use chemotherapeutic PATTERSON REPORTING & VIDEO 57 1 agents to obtain bone marrow oblation, but often 2 radiation is used in conditioning; and when chemotherapy 3 is used, you're talking real high doses of agents, very 4 high doses of agents. 5 Q. Getting back to the question I was trying to 6 ask, though, if we talk about the cells that reside in 7 the pluripotent stem cell compartment, can you show me 8 any research that says that those cells that reside in 9 that compartment are either more or less susceptible 10 when they're at rest or when they're replicating, when 11 radiation is the agent? 12 A. Yes. I would have to go back and search my 13 files, because I didn't come prepared to discuss that 14 literature; but, yes, I can find data to support that. 15 Q. I take it you're going to read and sign your 16 deposition? 17 A. Yes. 18 Q. Would you insert on whatever page this is in 19 your errata sheet a note as to what paper you would cite 20 me to, or papers, to that proposition? 21 A. Can we define a question so that I can write 22 it down? 23 Q. Yes, sir. I am trying to find out -- if we 24 talk about the cells that reside at the pluripotent stem 25 cell compartment, as you define that terminology, I am PATTERSON REPORTING & VIDEO 58 1 looking for a paper that you can cite me to that says 2 that cells that are at rest have a different 3 susceptibility to cells that are at replicating if 4 radiation is the agent. Is that a logical question for 5 you to answer? 6 A. Yes. 7 MR. HOBSON: Let's take a break. 8 (Break was taken.) 9 Q. (BY MR. HOBSON) Have you been able to locate 10 one of these defective cells that has the lesions in 11 place, that is capable of replicating, that would later 12 give rise to an individual being diagnosed with AML? 13 MR. BELFOUR: Herschel, could you ask that one 14 again? It may be that it's not vague, I just didn't 15 understand the question. 16 Q. (BY MR. HOBSON) Sure. I am trying to find 17 out if Dr. Irons has been able to locate -- or maybe I 18 said isolate -- one of these cells that is defective 19 that has lesions in place, that is capable of 20 replication, that later would give rise to a person 21 being diagnosed with AML. 22 A. If you are asking me, have we been able to 23 create AML in the test tube, the answer is no. That is 24 certainly an ultimate goal. We're in the process now of 25 defining steps in the process and attempting to produce PATTERSON REPORTING & VIDEO 1 those events and characterize those events. 2 If we are capable ultimately of bringing it 3 all together and integrating it into a single model and 4 producing AML in a test tube, that would be fabulous. 5 We certainly haven't done it yet. Q. Where are you in that process? 7 A. We are in the process -- we have characterized 8 early events in altered differentiation and response and 9 we're linking now with studies looking at early 10 cytogenetic abnormalities associated with exposure to 11 benzene metabolites. 12 So I would say we're -- we are looking at the 13 transition from early transitory changes that are 14 temporal -- temporarily transitive to the first steps 15 that are associated with the first known clonogenic step 16 in the development of AML. 17 Q. What do you mean by "the first known 18 clonogenic step"? 19 A. Certainly, the earliest event -- the earliest 20 permanent event that has been associated with the 21 evolution of AML secondary to occupation or 22 chemotherapeutic exposure is cytogenetic aberrations 23 involving chromosomes 5 and/or 7. Those are certainly 24 the predominant pattern that is seen. 25 Chromosomes 8 and 21 have been implicated as PATTERSON REPORTING & VIDEO 60 1 well, but they're also seen in primary leukemias as 2 frequently as they are in secondary. 3 But the earliest events that are seen are 4 associated with aneuploidy and the potential involvement 5 of chromosomes 5 and/or 7 in the -- certainly the 6 dominant model for secondary AML. 7 We're in the process now of looking at those 8 early events. We certainly have seen aneuploidy. We're 9 in the process now of trying to translate into a system 10 where we can look at the evolution of those types of 11 changes in liquid culture, looking at myeloid stem 12 myeloid progenitor cells and pluripotent stem cells, to 13 see where these events occur. 14 Q. Tell me how you use the term "aneuploidy." 15 A. "Aneuploidy" basically means the addition or 16 loss of a chromosome. One also sees interstitial 17 deletions. These are the types of lesions that are seen 18 in the evolution of secondary AML, and they are 19 precisely the types of abnormalities that are consonant 20 with the molecular effects of the quinone metabolites of 21 benzene, and they appear to be comparable to the effects 22 that these metabolites have on cells in culture, and 23 they're comparable to what has been seen in 24 benzene-poisoned workers in China. 25 Q. When you say the quinone metabolites of PATTERSON REPORTING & VIDEO 61 1 benzene gives you -- I have forgotten the term that you 2 used. You didn't say it gave you the same as, but you 3 used a different modifier, and I have forgotten what you 4 said now. Let me start over. I have lost the question 5 myself. 6 Do you find that you can have these 7 chromosomal changes in chromosomes 5 and 7 when you 8 expose cells to the quinone metabolites of benzene? 9 A. Previous studies by other investigators have 10 suggested involvement of 7. The Chinese-worker studies 11 have looked at chromosomes 7, 8 and 9. We have just 12 recently produced effects associated with 7. We're 13 beginning to look at 5. 7 appears to be certainly a 14 susceptible endpoint associated with exposure to either 15 benzene in vivo in humans or hydroquinone in vitro in 16 human cells. 17 Q. Each time I think you said -- in your answer 18 there you used a term like "suggestive" or "suggested," 19 and then you said "associated with." And what I am 20 trying to find out is, Do you actually find this 21 chromosomal change for the 7 chromosome; or you say, 22 it's just suggested? 23 A. No, we see a loss of 7. We are looking to see 24 whether it's suggested with the whole chromosome or 25 whether it's an interstitial-type deletion. We need to PATTERSON REPORTING & VIDEO 62 1 look at 5 in detail, as well. We also want to compare 2 it to other types of chromosomes to see if there is a 3 selective effect with respect to the chromosomes that 4 are associated with AML, or whether it is a nonspecific 5 effect on virtually any chromosome that just happens to 6 be consistent with cell survival and risk of future 7 events if it happens to involve chromosomes 5 or 7 or 8 8 or 21. 9 Q. At which level in cell development have you 10 done work with the quinone metabolites of benzene that 11 leads you to the finding that the 7 chromosome is 12 damaged? 13 A. A number of cell lines have been looked at by 14 other investigators. We have looked at lymphoid cell 15 lines. We are looking in bone marrow now. All of these 16 studies suggest the predisposition to involvement of 17 chromosome 7 in that. In other words, 7 is certainly a 18 target in replicating cells associated with exposure to 19 these agents. 20 Q. You're looking at the lymphoid compartment? 21 A. We're looking at a lymphoid line basically as 22 a surrogate for bone marrow early on to develop the 23 technique, and we have virtually completed those 24 studies, and we're moving into actual bone marrow cells 25 now. PATTERSON REPORTING & VIDEO 63 1 Q. So at this point are you able to take human 2 cells -- expose human cells to the quinone metabolite of 3 benzene and find chromosome damage to the No. 7 4 chromosome? 5 A. We have not done that. We are in the process 6 of doing that as I speak. 7 Q. And the 7 chromosome is merely one of the 8 chromosomes that has to be affected to actually produce 9 an AML? 10 A. The earliest aberrations that have been seen 11 involve 7 or 5, interstitial deletion of 5 or loss of 5, 12 one copy of 5 and/or loss of 7. 7 appears -- 7 appears 13 to be found more often, but I think it's been looked for 14 more often as well, so that may be an artifact. That's 15 the earliest event that is seen. 16 Other chromosomes have also been implicated, 17 but they are not necessarily a -- they do not allow you 18 to distinguish between primary and secondary leukemias. 19 It doesn't mean they're not involved in certain cases of 20 secondary, but they're also involved just as often in 21 primaries. 5 and 7 early involvement appears to be a 22 hallmark of secondary leukemia. 23 Q. Is the 5 and 7 -- are the 5 and 7 chromosomes 24 only involved in AML or are they involved in other 25 leukemias? PATTERSON REPORTING & VIDEO 64 1 A. For the most part, we're looking at AML, 2 certainly in the case of 5, although 5 is involved as 3 a step in other types of cancers, such as the 4 development of colon carcinoma, although we're looking 5 at different genes that appear to be involved in 5 6 associated with colon carcinoma, and they are in the 7 case of AML. 8 Q. But for the other leukemias, 5 and 7 would not 9 be involved? 10 A. They are not made -- they certainly do not 11 there are not nonrandom chromosomal aberrations in the 12 lymphoid leukemias that involve consistently 5 or 7. 13 That's not to say that you may not find another leukemia 14 type that may have involvement of 5 or 7, but there is 15 certainly no evidence that it is the earliest change or 16 that it is a consistent nonrandom finding in those 17 tumors. 18 I mean, the chance -- the importance of a 19 given chromosomal abnormality really relates to the 20 genes that are affected or involved associated with that 21 abnormality. And those genes -- the significance of an 22 alteration in gene function is dependent upon the stage 23 of differentiation or gene expression in that cell at a 24 given point in time. Let me see if I can give you an 25 example, because I know that is fairly esoteric. PATTERSON REPORTING & VIDEO 65 1 In CML, the primary gene abnormality -2 genetic abnormality that occurs in CML involves the 3 production of the Philadelphia chromosome, and that 4 reciprocal translocation is an abnormal fusion product 5 between two genes: the cluster gene or region and the 6 abl oncogenes, the bcr/abl. That, in turn, indirectly 7 leads to abnormal regulation of the ras oncogene 8 product. The ras gene isn't mutated, but the BCR/ABL 9 fusion product deregulates the function of ras in a 10 pluripotent cell. 11 If that cell was to undergo a mutation in ras, 12 it would probably be of no biological significance 13 whatsoever, because you already have deregulation of 14 ras; and if you mutated the gene, it would have no 15 bearing on the life history of that cell. 16 A ras oncogene mutation in another tumor in 17 another stage and differentiation might be very, very 18 important. So where a gene is affected in the natural 19 history of the cell can make a big difference. 20 In the case of genes on 5, we know that the 21 area 5 that's always affected in secondary AMLs 22 corresponds to a relatively small number of genes that 23 are very important in early steps in myeloid 24 differentiation. If that were to occur in a skin cell, 25 it may have no consequence whatsoever. PATTERSON REPORTING & VIDEO 66 1 Does that make sense? 2 Q. Yes, for the moment. 3 The question that comes to mind next is you 4 say "secondary" versus "primary" leukemias. How do ~.~1 5 know which is primary and which is secondary? 6 A. In the occupational literature that's a very 7 -- that can be a very practical, relevant question, and 8 in fact, the occupational literature is a little fuzzy 9 with respect to the types of findings that are seen 10 because of the problems of assigning exposure and 11 causation, what have you. 12 Having said that, you still see a very 13 pronounced pattern with respect to the characteristics 14 of AMLs in that literature between exposed and 15 controlled populations. 16 But all that aside, where we have a much 17 better handle on AML than many diseases is the 18 relationship between AML secondary to chemotherapy 19 and/or radiation therapy. There you know precisely what 20 the exposures are, you know what the agents are, you 21 have a captive population, you know what the primary 22 diseases are, you know what the secondary leukemias 23 are. And that's a very, very strong, rich, and deep 24 literature base that superimposes a great deal of 25 structure on our interpretation of the occupational PATTERSON REPORTING & VIDEO 67 1 literature. They're very comparable. 2 So we can learn a lot from looking at that 3 literature where we do have a very clear picture of the 4 differences that are seen in primary versus secondary 5 AMLs. 6 Q. Talking about the occupational exposure, I can 7 see your rationale for the chemotherapeutic and the 8 radiation therapies, because as you say, there you have 9 probably extremely well-regulated exposures; at least 10 one would hope that most of the time they are. And so 11 you have your exposure in hand, I guess, or at least 12 it's somewhat understood better. But if you go -- you 13 say that in occupational that's not so. 14 Jumping back for a minute to the 15 chemotherapeutic- and the radiation-secondary leukemias, 16 is the definition of a secondary leukemia for 17 chemotherapeutic and radiation the fact that you have 18 got the leukemia in the exposed population, or is there 19 something about the individual who has received the 20 exposure that you can say, I can look at this individual 21 and say that this is secondary leukemia versus what just 22 turned out to be a primary leukemia that happened? Does 23 that make sense? 24 A. The question makes sense, and the answer is 25 both. PATTERSON REPORTING & VIDEO 68 1 Early on in that literature -- obviously, 2 you're dealing with epidemiologic studies and 3 associations, and early on, the primary definition of a 4 secondary leukemia was one that occurred in a patient 5 who received chemotherapy for some other neoplasm, blit 6 it became very clear very quickly that there was a 7 prescribed pattern, both with respect to the progression 8 and development of that disease, as well as the 9 characteristics of leukemias that were seen that differ 10 dramatically from what you see in the spectrum of 11 leukemias that aren't associated with secondary -- as a 12 secondary. 13 You now -- I think it is arguable that there 14 are specific criteria that can be used to help 15 distinguish between a primary and a secondary that are 16 very strong. There are some changes that aren't 17 particularly useful, but there are changes or 18 characteristics of those that are very useful. 19 For example, we've been talking about 5 20 chromosome 5 and/or 7. If you add all those up, one 21 finds -- well, first of all, just take general 22 cytogenetic nonrandom -- demonstrable nonrandom 23 cytogenetic abnormalities, clonal aberrations, 24 cytogenetic aberrations occurring in a leukemia. 25 If you use banded chromosome analysis, and you PATTERSON REPORTING & VIDEO 69 1 look at primary leukemias, there is no evidence of any 2 previous exposure to a potential or putative cause of 3 aberrations, slightly more than half present with a 4 demonstrable clonal chromosomal aberration, a 50 to 60 5 percent. If you look at patients developing AML 6 secondary to chemotherapy, it's arguably 95 percent. 7 Given the experimental area, you could be looking at 8 virtually 100 percent. It's a huge preponderance. 9 So the presence of a nonrandom clonal 10 cytogenetic aberration provides you with some means of 11 discrimination a priori. Basically, you can't, if you 12 have it, distinguish a priori between a primary or 13 secondary. But if you have no demonstrable chromosomal 14 aberration, it's more probable than not that that is not 15 a secondary leukemia, because there is a much greater 16 predisposition toward clonal chromosomal aberrations in 17 the secondaries. 18 If you take it a step further now and look at 19 the specific involvement of chromosomal aberrations in 20 the various types of leukemias, 8 is found in both, 21 21 is found in both to about the same degree, on the order 22 of, I think -- maybe 5 is low -- probably somewhere 23 between 5 and 10 percent. 24 If you look at 5 and/or 7 together, you're 25 looking at, depending upon the study -- certainly in the PATTERSON REPORTING & VIDEO 70 1 chemotherapeutic literature -- 80 to 90 percent of 2 secondary AMLs involve 5 and/or 7, and only about 15 3 percent, or something like that, of primary, that have a 4 cytogenetic aberration, would involve 5 or 7. So there 5 is a clear preponderance of that type of chromosomal 6 aberration in secondary leukemia. 7 Q. If I asked my question in a little bit 8 different way, would the answer be the same? And the 9 question asked a different way would be -- I guess the 10 preliminary question is, Can you have a primary leukemia 11 in a person who has received chemotherapeutic agents? 12 A. I don't know whether that's a semantic 13 question or a philosophical one. 14 Q. That's why I went down this road, and I wasn't 15 sure we were communicating, because that's what I am 16 trying to find out really, the answer to that question. 17 I mean, if you are going to define a secondary 18 leukemia as any leukemia that appears in someone who has 19 had chemotherapeutic agents administered, then it seems 20 to me to be difficult, then, to separate out primary 21 leukemias based on what you find in those people. 22 A. Let's take the two to whatever percent of 23 individuals who develop AML secondary or AML following 24 intensive chemotherapy who have no clonal chromosomal 25 aberrations. PATTERSON REPORTING & VIDEO 71 1 Q. Say that again. 2 A. If you were to take the relatively small 3 number of individuals who present with AML, one of the 4 subtypes, that does not have a demonstrable nonrandom 5 clonal chromosomal abnormality, we're talking very, very 6 small number of people -- very, very small fraction. 7 If you took it one step further and went in 8 and looked at specific regions of the chromosomes on 9 those individuals that are involved in secondaries and 10 they were intact -- in other words, they're using 11 molecular probes now, not just using visual cytogenetics 12 -- and found that, in fact, they were intact, I think 13 philosophically you would have an interesting question. 14 And that is, Do those represent primary totally 15 independent events that have occurred independent of the 16 toxic exposure those patients have received? 17 1 think that's certainly not -- it's a 18 philosophical -- at this stage of the game, that's a 19 philosophical question, but I think you could certainly 20 ask it in that context. 21 I think the pattern that is seen in the 22 secondaries is so striking, with respect to the 23 predisposition, relative to what is seen in the general 24 population, that for most of those the answer to your 25 question is no, they're not primaries, they're PATTERSON REPORTING & VIDEO 72 1 secondaries; they have a different natural history in 2 many respects than what you see in the primaries. But 3 it is possible, I imagine, that some -- the small number 4 of individuals that don't present with a pattern that 5 you can find consistent with secondaries could 6 conceivably be primaries. 7 Q. But none has been reported that you know of? 8 A. Well, I mean, we do have negative -- we don't 9 have an absolute concordance of 100 percent cytogenetic 10 aberrations in AML secondary to chemotherapy. Do those 11 represent primaries? I don't think anyone can tell you 12 that. 13 Q. That being the case for the chemotherapeutic 14 agents, would we put radiation in that same category? 15 Is it philosophical there as well? 16 A. Well, you see a lot of the same changes in 17 radiation too. With radiation you're dealing with a 18 different dynamic, because you don't always know 19 precisely what an individual is exposed to, so you're in 20 the same -- you're basically in the same situation you 21 are with occupational exposures. There is some 22 fuzziness with respect to the exposure data, even with 23 respect to radiation. 24 Q. If we're talking about people who have had 25 radiation therapy PATTERSON REPORTING & VIDEO 73 1 A. I mean, there you do have a captive. 2 Q. I'm sorry. If we talk about people who have 3 had radiation therapy where their exposure is pretty 4 well defined from the therapy? 5 A. Right. What you have to keep in mind there is 6 that as we become much more sophisticated in terms of 7 distinguishing and designing effective therapies and 8 figure out where the risk factors are involved, it has 9 become clear that radiation is certainly within the 10 therapeutic range of exposures, not anywhere near as 11 potent as chemotherapeutic agents in inducing AML. It 12 can happen, but it's not -- it is not as effective as 13 chemotherapeutic agents; and, in fact, there is a major 14 difference between combined radiation/chemotherapy and 15 radiation and chemotherapy; and chemotherapy looks to be 16 the big culprit there. 17 Q. None of the work you have done at the 18 university or earlier at CIIT involved radiation; would 19 that be accurate? 20 A. No. We have looked at radiation -- we have 21 done quite a lot of radiation work in the context of the 22 NASA grant, looked at interactions between radiation and 23 leukemogenic agents in the mouse model. 24 We have not directly done any dosimetric 25 studies looking at radiation with human cells, primarily PATTERSON REPORTING & VIDEO 74 1 because the model that we were characterizing for NASA, 2 we basically provided sufficient evidence that the mouse 3 was not a good model for predicting effects in human 4 cells. 5 Q. So in the NASA work, you actually exposed 6 cells to radiation in your research? 7 A. That's correct. 8 Q. But only to mouse cells? 9 A. We have done some human work. We find the 10 mouse cells are much more susceptible. We don't see the 11 same effects. 12 Q. Now, if we go to occupational-exposed groups 13 for benzene exposure, where do you find your pool of 14 exposed individuals that you can say are secondary 15 leukemias that you can study? 16 A. Now, with respect to the human occupational 17 literature, I'll present you with a caveat, and then I 18 will go through it with you. 19 The caveat is, We have talked about some of 20 the problems with respect to assigning or characterizing 21 exposure in an occupational setting, which I'm sure you 22 are intimately familiar with. You can characterize the 23 secondary -- the occupational literature with respect to 24 benzene solvents, and you can characterize the 25 literature with respect to characterization of PATTERSON REPORTING & VIDEO 75 1 cytogenetic aberrations in leukemias in 2 occupational-exposed populations. 3 In general -- generalizations are dangerous 4 but in general, the most sophisticated useful 5 cytogenetic data is in case-controlled studies where the 6 exposure characterization is abysmal, where it's 7 self-selected, it's an anecdotal, no one measured a 8 thing. Where we have the greatest strength of data with 9 respect to characterization of human exposure, 10 cytogenetic analogies are amateurish. 11 Q. If they exist? 12 A. Yes. So we're dealing with, in general, 13 different strengths and weaknesses with respect to the 14 study populations. 15 So there is some inference that is required 16 with respect to looking at the pattern of results that 17 are found in one versus another. Are they consistent? 18 Are they inconsistent? Do they support each other? Do 19 they completely corroborate each other? And often you 20 don't have, across the literature, complete 21 corroboration, but sometimes you do. 22 So, for instance, where we have solvent 23 exposure where benzene is a component and is the only 24 well-characterized and accepted leukemogenic agent, we 25 have consistent cytogenetic findings in AMLs; and the PATTERSON REPORTING & VIDEO 76 1 pattern is the same. 2 In other studies we have -- where we know the 3 exposure is benzene, we have nondescript cytogenetic 4 aberrations that have been described in certain tumor 5 types that are consistent with what is seen in the 6 secondary -- the literature associated with exposure to 7 chemotherapeutic agents. It doesn't provide us with 8 specific banded-chromosomal analytical data, but the 9 types of changes -- the types of chromosomes that are 10 involved are comparable; they're consistent with the 11 same pattern that is seen. So there are variations in 12 the -- in that data that, if you put it all together, is 13 entirely consistent. 14 Having said that, I have recently written a 15 review where I attempted to summarize that literature, 16 and I believe I brought it. 17 Now, I hand this to you, and I don't have it 18 memorized, but basically this is just -- this is not an 19 all-inclusive table. The text has more information in 20 it. 21 This table shows the pattern of cytogenetic 22 characterization in secondary AML. The first several 23 studies involve a combination of either chemotherapeutic 24 agents and/or some occupational data as well. The last 25 few studies involve primarily occupationally exposed PATTERSON REPORTING & VIDEO 77 1 populations where solvent and/or benzene are implicated. 2 And there are other studies as well. It's not meant 3 to be all-inclusive, but the pattern is consistent in 4 either case. 5 Q. I see the title to what you have handed me; it 6 says, "To be published in Environmental Health 7 Perspectives." Has it been accepted now? 8 A. Yes. 9 Q. Any idea when it will come out? 10 A. No, I don't have any idea. That's worse than 11 -- that's even more ambiguous than most journals, 12 because it's a government journal. 13 Q. Since I haven't had a chance to read it, let 14 me see if I can get any questions together, having not 15 read the paper. It may or may not make any sense. 16 Going back to something you said, you looked 17 at the literature, I think you said, for solvent 18 exposure where benzene was in the solvents and where 19 there was no other -- and I couldn't write it down as 20 you said it -- where there is no other well-recognized 21 or well-something. 22 A. There have been a number of studies that have 23 looked at occupational exposure to solvents, and that is 24 in quotations. Obviously, studies vary with respect to 25 the reliability of what that means and what is defined PATTERSON REPORTING & VIDEO 78 1 by that. 2 What I have said is that, and it's my opinion, 3 it's reasonable to include that data in an evaluation 4 such as this if benzene is the only defined human 5 leukemogen that we're talking about. And when we're 6 talking about aromatic solvents, that that's where we 7 are. We're not talking toluene, we're not talking 8 xylenes, we're not talking substituted benzenes. 9 So if we're talking solvents and benzene in 10 the mix, then I think it's fair to assume that if we are 11 dealing with secondary leukemias, they are 12 benzene-derived leukemias for purposes of population 13 analysis and for looking at the characteristics of the 14 tumor types. 15 Now, since they do fit the pattern, I think 16 clairvoyance or hindsight is 20/20. I think I'm 17 justified in doing that because the pattern is the same. 18 It's what is seen with the chemotherapeutic agents. So 19 I think that it's a fair analogy and one can deal with 20 those in the same context. 21 Q. Except for the plants that are manufacturing 22 the chemotherapeutic agents, is there any other 23 occupational exposure to a defined human leukemogen? 24 A. Radiation. Certainly based upon the existing 25 literature, no. PATTERSON REPORTING & VIDEO 79 1 Q. In looking, then, here at the table that you 2 handed me, which was on page 22 of your submitted paper, 3 can you help me go down these and say -- you pointed out 14 some differences between whether these were 5 chemotherapeutic as well as occupational. I think you 6 said the bottom four were occupational only. It's a 7 long table. 8 A. The first one is the -- what do they call it? 9 The first one is one of the first studies done looking 10 at the working group that looked at secondary leukemias 11 that involved both chemotherapy and some occupational, 12 but primarily chemotherapy. 13 Q. Can you tell me what the occupational group 14 was there? 15 A. Not without going back and -- I am not sure 16 that I discussed it in detail. Le Beau and Rowley are 17 chemotherapeutic. Pederson-Bjergaard, the particular 18 one I have referenced there, he has done both and 19 combinations of both -- which one did I reference for 20 this? This is taking a bit, because there are several 21 in here. 22 Q. That's fine. 23 A. Tile first one is the Fourth International 24 Workshops on Chromosomes. That involved both 25 occupational and chemotherapeutic. I cannot PATTERSON REPORTING & VIDEO 80 1 specifically tell you which populations they looked 2 as we sit here this minute. 3 Q. What is the date for that publication? 4 A. I think it's 1984, yes. Well, as we sit here, 5 I am not sure precisely which Bjergaard study is cited 6 in the table, but there are at least half a dozen cited 7 in the chapter, and he has looked at both. 8 Q. Bjergaard, is he an American researcher? 9 A. Danish. 10 Q. Any idea where his benzene-exposed leukemias, 11 his secondary benzene leukemias came from? 12 A. He's looked at solvent-exposed. I'm not sure 13 which population he has looked at, as I sit here. I 14 would have to go back and review that literature to tell 15 you specifically where they came from. 16 Mitelman has looked at occupational exposure. 17 Mitelman, Golomb and Fagioli. 18 Q. Golomb; is that the Golomb that is at the 19 University of Chicago? 20 A. Yes. Those are case-control studies, so they 21 define their populations in different ways, and you have 22 to look at each individual case-control study to see 23 what those are and how they characterize or assign their 24 exposure. 25 Some of them involve occupational exposure to PATTERSON REPORTING & VIDEO 81 1 what is referred to as pesticides -- which is probably 2 the least reliable piece of information you will ever 3 see. It boils down to a questionnaire, Have you ever 4 used a pesticide? -- and organic solvents, which can 5 range from leukemias in individuals who again answer a 6 questionnaire in a self-selected fashion. 7 We may be talking casual use of a solvent or 8 we're talking about occupational-exposed individuals 9 where there is some reasonable characterization in the to exposure stream. It varies from study to study. As I 11 told you before, it ranges from very reliable to very 12 unreliable. 13 Q. Maybe I can help short-circuit some of this, 14 and I don't mean to sound perhaps offensive; but as I do 15 benzene cases as a lawyer, I seldom ever find anybody on 16 the other side of my cases who will ever say that this 17 is a benzene-induced leukemia, no matter what kind of 18 leukemia it is, including AMLs, and no matter what I 19 think the exposure history might have been. And so it 20 strikes me as odd that I could go or you could go to the 21 medical and scientific literature and find this pool of 22 individuals that have been studied that are conceded to 2_) be benzene-induced leukemias, from which you can then 2.4 look at their blood or bone marrow and do cytogenetics 25 to say this is a secondary leukemia. PATTERSON REPORTING & VIDEO 82 1 Can I short-circuit through that and tell me 2 whether or not you have such a group or can find such a 3 group reported in the literature. 4 MR. BELFOUR: Herschel, let me just make a 5 general objection. I don't mind your giving a preface 6 in order to expedite things, but I don't want the record 7 to reflect I have not objected to those comments. Since 8 the objections to the form of the question must be mate 9 at the time of the question, may I make it general, and 10 you may proceed. 11 MR. HOBSON: Yes. 12 A. As I said before, the reliability with respect 13 to the exposure history tends to be at odds with the 14 sophistication of the cytogenetic analysis. The most 15 recent studies that have been done of occupational 16 that have involved occupationally exposed populations, 17 that have sophisticated cytogenetics, have been done in 18 an era in which the exposure to benzene in general is 19 far less than it was in the 1940s, the 1950s, the 1960s. 20 If we look at occupationally exposed 21 populations in that era -- most notably the studies done 22 in Europe where we have an attempt at cytogenetics, 23 superimposed upon a very clear, heavy, consistent, 24 egregious exposure to benzene -- there is a pattern that 25 is seen there. There are cytogenetic aberrations. They PATTERSON REPORTING & VIDEO 83 1 are consistent generically with the types of aberrations 2 that are seen either in case-control studies or in 3 studies of people receiving chemotherapy today where the 4 cytogenetics is much more reliable and much more 5 sophisticated and precise. 6 We do not have, outside of China, to my 7 knowledge, a population that one can point to today, 8 hopefully, that was consistently exposed to high levels 9 of benzene and for which we have leukemias with 10 clear-cut, state-of-the-art cytogenetic 11 characterization. 12 The Chinese study may, in fact, represent the 13 exception to that, and certainly the results that are 14 coming out of that study now show aneuploidy in 15 benzene-poisoned workers involving chromosomes 7, 8, and 16 maybe 5; we don't know yet. Certainly the 17 characterization of cytogenetic aberrations in the AMLs 18 is consistent with what we see in the studies that are 19 -- the chemotherapeutic studies. 20 As we go forward with that study in China, we 21 may very well be able to answer that question directly 22 and precisely. Right now, the evidence certainly points 23 to it. 24 Q. (BY MR. HOBSON) What is the existing evidence 25 for the Chinese workers in this regard? PATTERSON REPORTING & VIDEO 84 1 A. Martha Linet has reported predominant 2 involvement of cytogenetic aberrations, from my 3 understanding and my conversations with her; certainly 4 involvement of chromosome 7 features strongly in those. 5 I do not know what the remainder of the cytogenetic is 6 in those patients, but I do know -- in the population 7 that is being followed now, that have been poisoned by 8 benzene, we do have predominant evidence of aneuploidy 9 involving chromosome 7; and studies are ongoing now 10 looking at 5, as well as other chromosomes. 11 Q. Dr. Linet is where? 12 A. National Cancer Institute, NCI. 13 Q. At which location? 14 A. Bethesda. 15 Q. She is doing the hands-on research herself? 16 A. Some of it, yes. She is coordinating that 17 aspect of the study. I think the study director is 18 Nathaniel Rothman, if he is the overall director of the 19 project. 20 Q. Has any of the work that you referred to 21 through Dr. Linet been published? 22 A. Some of it has. One of the articles -- here 23 is a reference. I have included here a manuscript 24 because I'm a coauthor on it. There are a number of 25 others that are coming out and that are either submitted PATTERSON REPORTING & VIDEO 85 1 or in press at the present time. 2 Certainly, the chromosomal involvement 3 involving chromosome 7 is either in press or close to it 4 at the present time. 5 Q. The work that you're coauthor on, is that the 6 result of the NCI contract that you had? 7 A. Yes. 8 Q. You said that is published now? 9 A. It is in press. 10 Q. I didn't think I remembered seeing it, but I 11 don't read it all. 12 A. No, it's in press. 13 Q. Now, that contract work that you did for NCI, 14 did you say that that was work that was done on workers 15 who had been benzene-poisoned? 16 A. Yes. 17 Q. How do you use the term "benzene-poisoned" in 18 this context? 19 A. A combination of industrial hygiene and 20 personal monitoring data that documents the current 21 exposures, as well as anecdotal and retrospective 22 industrial hygiene data, as well as clinical evidence of 23 bone marrow suppression associated with benzene 24 exposure. 25 Q. So if I can say that in perhaps a little bit PATTERSON REPORTING & VIDEO 86 1 different terminology, you have got industrial hygiene 2 exposure data in recent times that documents extremely 3 high exposure levels, right? 4 A. Yes. 5 Q. And we're talking hundreds, if not thousands, 6 of parts per million? 7 A. We're talking highly heterogeneous exposure: 8 level, but some individuals have documented exposure 9 well into the hundreds, I would say 400 parts per 10 million. 11 Q. By "anecdotal," you mean people telling 12 stories of its being even worse in the past? 13 A. I wouldn't even go so far as having to say 14 "worse." There is a certain conspectus with respect. '_o 15 the descriptions, with respect to what has been seen in 16 recent times. There is a movement to reduce exposure 17 levels, which is wonderful. It obviously -- it 18 obviously complicates the evaluation process, but there 19 is now a very valuable natural experiment, if you will, 20 or unnatural experiment involving a cohort of 21 individuals that have documented toxicity acute 22 subacute toxicity to benzene who are being followed, and 23 there are individuals independent of those who have 24 continuous exposure to benzene that certainly in the 25 I would have to look back to see precisely what their PATTERSON REPORTING & VIDEO 87 1 exposure levels are, but we're talking double digits 2 with respect to ppm. 3 Q. And then another parameter for this group that 4 you looked at from the NCI contract was that they 5 actually had clinical findings? 6 A. Within -- yes, within that subpopulation there 7 are individuals who actually have had some clinical 8 manifestation of benzene exposure: cytopenias, anemias, 9 what have you. 10 Q. Now, did you actually analyze tissue from some 11 of these individuals for NCI? 12 A. I analyzed serum for specific cytokine levels. 13 That was my function. 14 Q. And the serum that you analyzed, did it come 15 from individuals who had a clinical manifestation? 16 A. Excuse me, plasma. I should correct that. 17 It's my understanding ostensibly yes, although 18 I was blinded to the nature of the samples, so I can't 19 tell you which samples were what. 20 Q. Did any of these individuals whose plasma you 21 evaluated as part of the NCI contract have leukemia? 22 A. No, not to my knowledge. 23 Q. If we go back then to my question of, Is there 24 any pool of individuals who have leukemia that anyone is 25 willing to say was secondarily the result of benzene PATTERSON REPORTING & VIDEO 88 1 exposure, does such a group exist that has been looked 2 at, to your knowledge? 3 A. Yes, the Chinese study. There are individuals 4 -- there are a number of individuals, to my knowledge, 5 who are still alive who have been diagnosed with 6 leukemia and/or myelodysplastic syndrome where the 7 exposure is basically a pure play on benzene. 8 Q. And their cytogenetics have been looked at? 9 A. Some of them have and, again, to my knowledge, 10 I have not seen specific data, but my conversations with 11 Martha Linet and the summary data that have been 12 published or is in press suggests a consistency between 13 those findings and the literature that we have been 14 talking about here today. 15 Q. But there is no way for me to go to the 16 literature and see what this information is. 17 A. Travis, 1994, it's No. 23 in the set of 18 documents I brought here this morning. I can just read 19 you one summary sentence from the abstract, which reads 20 -- this is "Hematopoietic Malignancies and Related 21 Disorders Among Benzene-Exposed Workers in China," and I 22 quote, "The hematopoietic characteristics of the 23 benzene-exposed acute nonlymphocytic cases resembled 24 those following chemotherapy or radiotherapy. ANLL in 25 workers exposed to benzene may represent a distinct PATTERSON REPORTING & VIDEO 89 1 clinicopathologic entity, with characteristics similar 2 to treatment-related ANLL, including a preceding 3 preleukemic phase in some patients." So some of that 4 information is characterized and presented in this 5 publication. 6 There is another one that should be in press. 7 I believe the first author is either Lee or Martha 8 Linet. I have not yet seen a copy of that manuscript. 9 Q. And had these Chinese workers been studied 10 epidemiolosically? 11 A. Yes, and these studies are ongoing, but a lot 12 of this is coming out now as the data is analyzed. 13 Q. Is there a current epidemiological report that 14 you're aware of in the published literature that I can 15 go to? 16 A. Well, this is one. Lee is another. There may 17 be a couple by Lee. There is also, as I said, Martha 18 Linet's publication which is in press in the same issue 19 of Environmental Health Perspectives, and that should be 20 available. I expected to have a copy by now, but I 21 don't have one yet. If I had it, I would have provided 22 it. 23 Q. Now, if I heard you right, Travis, in the 24 paper that you were citing from for the ANLLs, what was 25 the modifier he used, similar to, or consistent with? PATTERSON REPORTING & VIDEO 90 1 What did he use? 2 A. It resembled those found in chemotherapy or 3 radiation therapy, and he said, "may represent a 4 distinct clinicopathologic entity, with characteristics 5 similar to treatment-related ANLL." 6 Q. "Similar to" and "resembled." How "similar 7 to" do you find them to be, the benzene-exposed workers 8 with leukemia, or I should say, with ANLL, and those 9 people who get leukemias from secondary to 10 chemotherapeutic agents? 11 A. Certainly with respect to the frequency of any 12 cytogenetic aberrations and the subtypes of leukemias 13 that are associated with the different groups, they are 14 essentially consistent. I certainly can't point to any 15 feature that distinguishes them from AML secondary to 16 chemo- or radiation therapy. 17 Q. Say that last part again. 18 A. I cannot point to any single feature or 19 characteristic that distinguishes them from AML 20 secondary to chemotherapy. 21 Q. So they're the same? 22 A. Essentially they are the same. What I cannot 23 do, because I have not seen it, is give you the specific 24 cytogenetic aberrations that occur in every case. I 25 haven't seen that data. It has been described to me as PATTERSON REPORTING & VIDEO 91 1 consistent with AMLs secondary to chemotherapy. 2 Q. Is Travis reporting on the Chinese workers, 3 the same ones that Lee and Linet are talking about? 4 A. Yes, these are all follow-up studies as part 5 of a continuing retrospective and prospective analysis 6 of benzene-exposed workers in China. 7 Q. But as of yet, you have not seen the actual 8 data analysis? 9 A. I haven't seen specific patient-by-patient 10 cytogenetic aberrations. I have seen data with respect 11 to -- wait a minute. What can I tell you? The subtypes 12 in morphologic characteristics are reported by a 13 patient. 14 Q. So in other words, you're saying that whether 15 it's acute myelogenous leukemia or a variant of acute 16 myelogenous leukemia? 17 A. Nil, M2, so forth and so on. Whether it's 18 evidence of preleukemic changes or myelodysplasia. They 19 simply 20 Q. "They," being Travis, et al.? Are you still 21 on the Travis paper? 22 A. Yes. They speak of them in general in that 23 they have many of the same features that are seen 24 associated with secondary to chemotherapy or radiation. 25 They don't describe the individual aberrations in this PATTERSON REPORTING & VIDEO 92 1 paper. 2 Q. And the features that they mention are not 3 elucidated, they're not set out? 4 A. No. They talk about 5 and 7. They talk about 5 the various populations that have looked -- the various 6 populations that have been looked at in terms of 7 exposure to organic solvents and involvement of those 8 chromosomes, and they characterize those aberrations, 9 but they simply state that the -- their data with the 10 Chinese workers is consistent with those. 11 Q. How many individuals are we talking about here 12 that were looked at? 13 A. Well, I am not sure -- I don't know how many 14 actual cytogenetic analyses they have. They are talking 15 about 32 cases of leukemia. 16 Q. But you don't know, of those 32, how many !d 17 cytogenetic analysis for this statement? 18 A. No, I can't tell you that. 19 Q. Has anyone looked at the other cancers of the 20 blood or bone marrow besides ANLLs? 21 A. Yes. In what context? With respect to 22 association with benzene exposure? 23 Q. No. In these Chinese workers; I'm sorry. 24 A. They characterize virtually all hematopoietic 25 neoplasms in this population. One of the difficulties PATTERSON REPORTING & VIDEO 93 1 that they now are faced with is looking for confounders 2 and other potential exposure scenarios because they're 3 dealing with 74,828 benzene-exposed individuals, and I 4 forget the number of actual sites, but it numbers in the 5 hundreds to thousands, so they have a huge number of 6 occupational-exposure scenarios. They have captured 7 virtually all the tumor types based upon the diagnostic 8 criteria that they are able to use for those 9 populations. 10 Q. So these are 74,000 workers from throughout 11 China doing different occupations that putatively 12 involve benzene? 13 A. Yes. This is probably the largest study I 14 have ever seen. There is no "probably" about that; 15 there is no qualifier necessary, both with respect to 16 industrial hygiene and with respect to employment 17 characterization and disease characterization. 18 Having said that, even this study has some 19 problems with respect to interpretation. 20 Q. For the workers that Travis reports 21 cytogenetic analysis, is there any description of the 22 occupation of those people? 23 A. Not on an individual basis, I don't believe. 24 Q. How about as a group? I mean, do they 25 predomindce from one particular industry? PATTERSON REPORTING & VIDEO 94 1 A. No. They have evaluated painters, they have 2 evaluated petroleum-refinery workers, they have 3 evaluated individuals using benzene as a glue. They 4 have evaluated a variety of different occupational 5 settings in which benzene is either produced or used. 6 They have, as I said, retrospective, prospective, and 7 current industrial hygiene data. So it's a very 8 complicated study with a number of different types of 9 analyses that are going on simultaneously. 10 This is one report. There are many that are 11 coming out in various bits and pieces as this data is 12 sifted through and evaluated. 13 Q. Is there any suggestion, that exposure level 14 being so high in these workers, that t'-~at has the 15 potential to give you a different outcome than if you 16 have exposure levels to benzene that are lower? 17 A. I haven't seen any data that would suggest 18 that. Now, I have not seen everything, by any means. I 19 have seen their characterization of exposures, and one 20 of the problems they have is that their time-weighted 21 average cumulative exposure scenarios that are used for 22 dosimetry purposes include a huge heterogeneity with 23 respect to transients that they have documented. In 24 other words, a time-weighted average exposure of 10 ppm 25 -- for instance, 10 ppm-years in China doesn't mean PATTERSON REPORTING & VIDEO 95 1 anything like it would here in the sense that in this 2 day and age today, of course, we wouldn't be looking at 3 a 10 ppm time-weighted average exposure, or shouldn't 4 be. It would be much closer to the mark than it is in 5 China, where you can have within that context exposures 6 to sub-ppm or low-ppm levels, as well as transients into 7 the hundreds of ppm. So it's a very heterogeneous 8 exposure base, and they are trying to dissect that out. 9 Q. In the work that you have done in your 10 laboratory, is there any way to gather information when 11 you look at the benzene-metabolite exposures that you 12 have had to the different cell populations? Is there 13 any way to correlate the amounts of metabolites that you 14 have used to the exposure level? 15 A. Ideally, I would hope we would be able to do 16 that. We're not in the position where we can do that 17 today. There are a number of issues that have to be 18 dealt with. 19 First of all, in order to look at the precise 20 effects of various metabolites on the cells, some of our 21 early works are, for instance, looking at clonogenic 22 responses, expose the cells for short periods of time in 23 the absence of media protein, so that we were looking 24 directly -- we were delivering, A, dose of the 25 metabolite to the cells we could quantitate, and, B, PATTERSON REPORTING & VIDEO 96 1 assure that that's what they were seeing. 2 Those same cells in bone marrow are going to 3 be bathed in protein, bathed in media -- call it media, 4 if you will -- that is going to compete with abstract 5 and otherwise provide a totally different dosimetric 6 with respect to what a given level of metabolite means 7 to that cell. 8 At some point, and in liquid culture with 9 long-term exposure now, we're beginning to get there, 10 we're beginning to address it. We have to come up with 11 a means of relating those types of exposure levels, 12 concentrations to exposure levels that might be targeted 13 in vivo. 14 Having done that, the next step is to take the 15 pharmacodynamic and pharmacokinetic data that is 16 available today both from animals and humans and try to 17 extrapolate. No one, to my knowledge, has been able to 18 do that adequately. It's obviously something that needs 19 to be done if we are ever going to be able to translate 20 this directly with respect to what it means 21 dosimetrically. We can say some things generically now, 22 but we certainly can't say in precise terms, 23 quantitatively how those relate to in vivo exposures. 24 Q. Your work that you have done in your 25 laboratory with the benzene metabolites, how would you PATTERSON REPORTING & VIDEO 97 1 characterize the amount of work that you have done with 2 individual metabolites as opposed to the combinations of 3 metabolites? 4 A. Our work is centered primarily on individual 5 metabolites. We have done some work on combinations of 6 metabolites. In fact, we have pioneered looking at 7 interactions between metabolites and the potentiation of 8 the effects of hydroquinone biphenol. There have been 9 other students independently that have looked at 10 catechol and hydroquinone, and hydroquinone and the 11 putative -- the ring-opened products -- trans, 12 trans-muconaldehyde, in particular. 13 You focus where we have because we have been 14 trying, first of all, to go from the simple to the 15 complex, and we have seen interesting effects at the 16 simple level. Our focus on the quinone metabolites has 17 been primarily because they have produced effects. 18 We have looked at some of the others, and we 19 haven't been able to produce anything except cell death, 20 which is not a very interesting input, but that doesn't 21 mean they don't have effects. 22 Q. Did I understand you to say that when you used 23 the quinone metabolites along with the phenol, that you 24 have gotten greater effects, or has it been lesser 25 effects, or can you tell? PATTERSON REPORTING & VIDEO 98 1 A. We have described those types of effects both 2 in vivo in mice as well as in vitro. It extends right 3 down to the cell level. Phenol potentiates the effects 4 of hydroquinone. It appears to be related to promoting 5 peroxidase-mediated oxidation of hydroquinone to its 6 terminal end-stage product, para-benzoatd_none. There 7 may be some pharmacokinetic effects, as ~.~11, that 8 various people have proposed, but we certainly can see 9 it at the enzymatic levels, so it may, in fact, explain 10 why you see it at in vivo level. 11 Phenol, itself, doesn't produce bone marrow 12 toxicity in experimental animals; and I would say there 13 is no evidence that phenol in occupational settings is 14 toxic or metatoxic. 15 Hydroquinone, in experimental animals, can 16 produce some effects. It's a very nonphysiological 17 situation when you administer hydroquinone externally, 18 but you put the two together and there is a phenomenal 19 synergy. 20 Q. Externally? 21 A. Yes. 22 Q. Hydroquinone and phenol? 23 A. Injected together will produce hone marrow 24 toxicity that is much greater than hydroauinone by 25 itself. PATTERSON REPORTING & VIDEO 99 1 Q. That's what you meant by "potentiated"? 2 A. Yes. At an enzymatic level, phenol stimulates 3 the terminal hydroxylation by peroxidation, and we can, 4 by adding phenol, know the presence of hydroquinone, 5 although most of our work has not focused on the 6 combinations as of yet. We have been looking at the 7 isolated effects. 8 Q. Is there a way to drive the metabolism of 9 benzene away from quinone in a living organism? 10 A. There are no known selective inhibitors of 11 peroxidase. 12 Q. Can you administer something else along with 13 the benzene, though, that will drive the metabolism of 14 benzene away from quinone or the quinone metabolites? 15 A. I don't know of it, because as soon as you put 16 a substituent on the ring, you're going in a different 17 direction in the first place. 18 Q. I wasn't talking about on the ring. I'm 19 talking about in combination with. 20 A. Right. I simply was saying that in terms of 21 -- if you throw in toluene or xylene, you can protect 22 against benzene toxicity, but that's competition for 23 peroxidaition. After that, we're not talking about the 24 same pathway at all. I don't know how one would 25 selectively block that. PATTERSON REPORTING & VIDEO 100 1 You can look -- what we have done is look in 2 specific cell populations and under specific conditions 3 where you have a different balance of detoxifying and 4 toxifying enzymes, peroxidase versus two electron 5 quinone reductase on DT diaphorase, which is a 6 protective enzyme in that system. And coincidentally, 7 the myeloid progenitor cell has very respectable mycloid 8 peroxidase activity and absolutely no DT diaphorase 9 activity. So susceptibility correlates -- in terms of 10 cell populations, correlates with the enzymatic 11 machinery that drives activation and not detoxification. 12 So I guess what I am saying is that there are 13 cell-specific models that provide a natural experiment 14 to indirectly address what you are saying, but we don't 15 have any major bullets that allow us to go in and 16 specifically block the selective oxidation of those 17 quinones. 18 Q. Going back to a little bit earlier line of 19 questioning, and I hope to wrap it up. Your work that 20 you have done, and the works of others that you are 21 familiar with, still don't allow the exposure of human 22 cells or -- stay with human cells for a minute -- to 23 benzene or benzene metabolites and generate a leukemic 24 cell; is that accurate? 25 MR. BELFOUR: I object to the question as PATTERSON REPORTING & VIDEO 101 1 vague, Herschel. I'm sorry, I didn't understand what 2 you asked. 3 A. What you have defined with that question is 4 basically the holy grail of leukemia research. We are 5 not at the stage where we can cradle-the-grave, 6 nuts-to-bolts, step-by-step, induce a human leukemia in 7 a test tube as a function of exposure to anything. 8 We are much closer than I ever imagined we 9 would be as recently as 1990. We have come a long way, 10 and I may not have to answer that question in the 11 negative for too many more years. 12 Q. (BY MR. HOBSON) Where is the gap? 13 A. I think the major gap is secondary events that 14 are probably not consequential with respect -- important 15 with respect to order and/or -- that are not probably 16 associated with exposure to benzene or anything else. 17 In other words, it's becoming clear there are multiple 18 steps in the process -- some are independent, some are 19 not -- comparable to other types of tumors that are 20 fairly well described now, such as colon carcinoma, and 21 that some of the subsequent events probably are not 22 associated with exposure. They are the product of 23 proliferation in abnormal cells that leads to secondary 24 events. 25 Q. What do you mean by "the product"? PATTERSON REPORTING & VIDEO 102 1 A. As I said earlier, some changes may predispose 2 to other ones, and if you have three, four, or five 3 independent phenomena that have to occur in some order, 4 some totally independent order, for the actual leukemia 5 phenotype to be expressed, those later events may h~---~ 6 nothing to do with exposure to anything. They may be a 7 function of the probability of a random event occurring 8 in a cell that has already gotten to that stage in the 9 process, for instance, if a late event in AML or a type 10 of AML involves mutation of the retinoblastoma gene. 11 Q. You mean a late event in the formation? 12 A. In an actual history of the disease. 13 Q. Before the disease actually occurs? 14 A. Yes. 15 Q. Okay. I'm sorry. 16 A. It may not be related -- it may be totally 17 separate and distinct from the processes that led to the 18 early changes. 19 Q. We're talking perhaps host-susceptibility 20 here? 21 A. It could be. I mean, that would be 22 speculative. It's more likely a random event. 23 Q. I mean, I'm having trouble thinking of 24 possibilities. What would some of these possibilities 25 be for these random events? PATTERSON REPORTING & VIDEO 103 1 A. We know, for instance, simply that there are a 2 number of human tumors where the only known influence is 3 cell proliferation. You increase the number of times a 4 cell proliferates, you increase the probability of a 5 replication-dependent error; or simply increasing 6 proliferation can increase the risk of probability of 7 cancer developing in humans. 8 If we have a cell that has got two or three 9 aberrations or abnormalities in it that now leads to 10 increased survival and replication in a cell that is 11 capable of transformation, that next event may simply be 12 a function of it or the probability may be related only 13 to the fact that the cells proliferated. That would be 14 a random 7, but with a very high -- with a different 15 rate of probability than it would be in any other cell 16 which doesn't proliferate as fast or in the context that 17 we're talking about. 18 So that would be an example of an event where 19 you have proliferating population that shouldn't be, and 20 that leads to an ultimate mutational change, if you 21 will, or cytogenetic event that is associated with the 22 phenotypic expression of a tumor. 23 Q. Is it possible that what you're saying 24 maybe I am just off in left field; tell me if I am 25 that if you have an individual who has been exposed to PATTERSON REPORTING & VIDEO 104 1 benzene and you have had some of these precursor steps 2 occur, that you have got this cell sitting in a 3 compartment; that but for having to replicate a lot, you 4 would never see leukemia in that individual, and then 5 this individual's bone marrow is affected by a greater 6 benzene exposure which would require that cell then to 7 replicate much more quickly? Is that one of the things 8 you're talking about as a possibility? 9 A. No. 10 Q. I am off in left field then. 11 A. I would say that one has about the least 12 plausible probability of any scenario, and the reason is 13 that benzene, like other chemotherapeutic agents -- end 14 early in the century was actually used for one, God help 15 the patients that got it -- I don't think it would be 16 very effective, and the side effects would be pretty 17 lousy -- but, nevertheless, it tends to target 18 proliferating cells. So if you are Going to imagine a 19 scenario where you have those types of events, I think 20 that high-level benzene exposure might actually be 21 protective in a situation where you have already got 22 or at least is somewhat ameliorated. 23 So high-level benzene exposure may actually 24 in individuals who are already down that .road, 25 high-level benzene exposure, there is no reason, a PATTERSON REPORTING & VIDEO 105 1 priori, to assume that is going to potentiate. It more 2 likely is going to suppress a proliferated cell 3 population. 4 Q. Does the latency period seem to be different 5 for benzene-induced leukemias than it is for 6 chemotherapeutic-induced leukemias? 7 A. It's a little longer. 8 Q. Which way? 9 A. Benzene. 10 Q. By "little," are we talking 11 order-of-magnitude? 12 A. No, no. For AML secondary to benzene, 13 certainly in most populations that have been studied, I 14 think the consensus is that it's on the order of about 15 nine to ten years, 9 1/2 years for a geometric mean for 16 latency. 17 Q. I'm sorry, I didn't get the pronoun. You said 18 "it." What are we talking about there? 19 A. The latency between first exposure and onset 20 of leukemia in occupationally exposed workers. 21 Q. To benzene? 22 A. Yes. 23 Q. What would it be for chemotherapeutic? 24 A. It can vary three to six years, is usually the 25 mean that is tossed out. Now, with some of the newer PATTERSON REPORTING & VIDEO 106 1 agents that don't act like classic alkylating agents, 2 it's a much shorter latency. It's on the order of 3 months, but that I don't think is appropriate for what 4 we are talking about. 5 So benzene -- certainly, occupationally 6 exposed AML, we're talking a mean latency on the order 7 of nine to ten years. 8 Q. Since we're talking statistics and using mean, 9 what would then be the standard deviation for that? 10 What does the curve look like? 11 A. I think you can argue that you're way below 12 the 5 percent confidence level by -- probably between 13 three and five years. So, I mean, that would be the 14 bell at one end, and it's probably close to the bell at 15 the other, although I think you're parsing hairs there. 16 My interpretation of that latency is both a 17 product of the nature of the specific disease and the 18 type of exposure, but it is not clearly one or the 19 other. 20 In the case of patients receiving 21 chemotherapy, I think it's important to remember these 22 people are not receiving trivial doses of drugs, and the 23 bone marrow is the target, and they are poisoned. we're 24 talking toxicity here; we're talking bone marrow damage. 25 Probably not even in Aksoy's Turkish population do you PATTERSON REPORTING & VIDEO 107 1 have that level of toxicity associated with benzene. 2 Arid so the difference between three to six 3 and, say, nine may be related to the intensity of the 4 damage that was done in the first place. 5 Q. Take me, for instance. In my graduate school 6 program I had occasion to work with benzene for a few 7 years, and based on what I know now, my exposures were 8 probably in the hundreds of parts per million several 9 days a week for a couple of those years, at least, 10 because of the work I had to do. That occurred in about 11 1969 through 1972. 12 Can you tell me that I will never have to 13 worry about secondary leukemia from benzene exposure? 14 As far as I know, I was never acutely ill at the time 15 from benzene exposure. 16 A. Well, we're talking 199 17 Q. '5. 18 A. -- '5. 19 Q. 27, 28 years ago. 20 A. I am in the same -- I am almost precisely in 21 the same boat you are. We could each have an oar. 22 From 1965 through 1968 I was a chemistry major 23 with an absolute flaming love of organic chemistry that 24 many people took to be a sign of mental instability, but 25 I used benzene as a solvent, phase extractions, and PATTERSON REPORTING & VIDEO 108 1 everything under God's green earth. 2 Q. Not with hood, not with gloves? 3 A. Not with gloves. Often not in a hood. But 4 based upon the latency we're talking about, I certainly 5 am not worried about that. There are other things in 6 life, I think, that are much more potentially important 7 to me, such as cardiovascular disease and a few other 8 things as I get older; but based on being out 20 years, 9 I don't think it's reasonable to expect that you are 10 going to incur any major probability of developing AML 11 secondary to those exposures. 12 Q. Can you rule it out? 13 A. I mean, absolutely, positively, never, nohow 14 or based on a more-probable-than-not standard. As a 15 scientist, you can't rule out anything absolutely, 16 never, nohow; but I don't think it's a realistic 17 prospect, and it certainly wouldn't meet a standard in 18 terms of probability. You're way out on the curve. 19 It's nonexistent. 20 Q. I hope your prediction becomes reality, not 21 that I want cardiovascular disease. 22 A. Well, I was just speaking in terms of the fact 23 that as we age, we have other things that become a lot 24 more imminent, I think. 25 Q. The lesions that are created by these chemical PATTERSON REPORTING & VIDEO 109 1 agents, be they chemotherapeutic or benzene-induced, 2 some of those lesions are created in the pluripotent 3 stem cell compartment? 4 A. We have no direct evidence of that, and I 5 don't think that that's a very high probability based 6 upon what we have seen. The transient changes we see 7 first involve shifts in differentiation, which, by 8 definition, involve commitment. That appears to be the 9 first thing we can see. It is also consistent with the 10 first types of changes that I have seen both with human 11 cells and in animal models associated with cytokine 12 alterations in cytokine response that are associated 13 with dysplasias. 14 The key, I think, is going to be where those 15 chromosomal aberrations show up. Everything points to 16 myeloid progenitor cell. That is going to be -- at 17 least today, that is the first nonrandom permanent 18 change associated with the evolution of AML. 19 Q. When you do your experiments with the 20 pluripotent stem cell compartment cells, you're exposing 21 those cells to benzene metabolites, correct? 22 A. Yes. 23 Q. And then after the exposure occurs, there is 24 some period time that passes, I take it? 25 A. Hours to days to weeks, depending upon PATTERSON REPORTING & VIDEO 110 1 precisely how the experiment is designed. 2 Q. When you say that you see a shaft to the 3 myeloid compartment, what kind of time period are we 4 talking about to see that shift? 5 A. The exposures can be as short as 30 minutes. 6 Q. When do you see the shift, how far in the 7 future? 8 A. Within a matter of hours. We haven't reduced 9 it -- we have not had as a goal to figure out whether 10 it's minutes or hours. We've been more interested in 11 the fact that it's less than days. 12 Certainly it's easy to see -- see the 13 clonogenic assays. It was difficult to predict what was 14 going on, because the exposures were only for 30 15 minutes. So -- you know, the exposure didn't take long, 16 but the readout is days later. So you're looking at a 17 series of events that have to occur before you see 18 ultimately what has happened. You have a different 19 number of cells -- well, you have to have a different 20 number of cells recruited to respond in the first place. 21 What we can say is that functionally, they're responding 22 like myeloid progenitors. They weren't functioning as 23 earlier cells; they were functioning as myeloid 24 progenitors. 25 Now, in liquid culture, we're beginning to PATTERSON REPORTING & VIDEO 111 1 look directly at marker changes, change in 2 differentiation, and to correlate those things on the 3 time -- on a timing basis. And it appears that that 4 whole shift functionally occurs within hours, for sure, 5 and is over in a matter of days. So you're looking at a 6 window of time in which this particular type of 7 transience occurs. 8 For hydroxycyclophosphamide, which is a 9 metabolite of cyclophosphamide, we've been able to do it 10 in vivo in mice, and it's comparable to what we have 11 seen in vitro and we have used it as a surrogate. We're 12 talking a matter of few hours. The whole process is 13 certainly over in a matter of a couple -- two to three 14 days, something like that. 15 Q. As I envision this in my mind, now, which I'm 16 sure is far too simplistic for the work you are doing, 17 you start off with the pluripotent stem cell compartment 18 cells and you expose them to benzene metabolites for 19 some period, generally a matter of 30 minutes, or is 20 A. Yes, that's in one scenario. We can also look 21 at that with continuous exposure, but that's using our 22 newer culture techniques. 23 Q. But you have the exposure period, and then you 24 wait a period of time to see those cells proliferate? 25 A. That's one type of assay. That's the first PATTERSON REPORTING & VIDEO 112 1 type that we were using. 2 Now, we're looking at changes in their 3 phenotype with time -- that is, changes in their 4 expression of various genes that correlate with their 5 functional characteristics. 6 Q. Now, when you start off with the pluripotent 7 stem cell compartment cells, I take it you run controls 8 as well as having your dosed cells? 9 A. That's the only way we can do this, yes. 10 Q. So in your controls, when you look at how 11 these pluripotent stem cells differentiate, some 12 differentiate into the myeloid compartment and some 13 differentiate into the lymphoid? 14 A. Yes. 15 Q. And then when you look at the exposed 16 pluripotent stem cells, you're seeing a different 17 division, but still some lymphoid, some myeloid? 18 A. We're looking at population shifts in highly 19 purified cell populations. Some of those are ultimately 20 capable of or destined to be lymphoid. The vast 21 majority appear to be destined to ultimately either be 22 erythroid or myeloid in terms of the way they respond in 23 culture early on. 24 Q. Are you saying you do your analysis, then, 25 before this differentiation is completed? PATTERSON REPORTING & VIDEO 113 1 A. Yes, for those types of -- yes, for the liquid 2 culture work, yes. We can monitor -- we can monitor 3 readily erythroid and myeloid, and in the liquid culture 4 we're using they are -- they predispose -- the ones 5 we're using right now are -- they predispose toward 6 looking at the myeloid pathway. 7 We have to use different conditions if we want 8 to look specifically at early differentiated events that 9 would be associated with lymphoid. Clonogenic assays 10 provide us with some handle on that, because those, by 11 definition, coincide with populations that have lymphoid 12 capability or myeloid; but for the different -- what 13 we're doing with markers right now, the work we have 14 done to date has emphasized looking at myeloid and not 15 lymphoid. That's another step in the process. We are 16 not there yet. 17 Q. In your system that you have set up, can you 18 start off with pluripotent stem cell compartment cells 19 and culture those cells long enough to see them 20 differentiate and become either myeloid or lymphoid 21 cells and keep them alive? 22 A. Okay. If you are asking, have we done that in 23 both directions in our laboratory now in the systems 24 we're using, the answer is no. Is that technology 25 available? Yes, it is. If you're asking, Can you, in PATTERSON REPORTING & VIDEO 114 1 the context of what the state of the art is, the answer 2 is yes. With respect to, Have we done that, no. 3 9. I guess in a sense, the experiment that I 4 would be interested in knowing if you could do is, could 5 you take pluripotent stem cell compartment cells, expose 6 those, for whatever the appropriate period of time is, 7 to benzene metabolites or a metabolite; let those cells 8 culture-differentiate; and then do a cytogenetic 9 analysis of both the myeloid compartment differentiated 10 cells and the lymphoid compartment differentiated cells? 11 A. To varying degrees, those are capabilities 12 that exist. We are, to varying degrees, attempting to 13 do that; not precisely as you have defined it; but 14 generically, yes. 15 There are issues with respect to the technical 16 culture and readout of lymphoid versus myeloid. That 17 involves technologies that exist but that are difficult. 18 You can't do them in the same system. You have to 19 figure a way of controlling for what are essentially 20 stochastic events in the system, because the culture 21 conditions are different, although there is at least one 22 technique that has been -- that has been proposed to 23 allow both simultaneous lymphoid and myeloid 24 development, but I have yet to see that. 25 Our approach is not to look for random -PATTERSON REPORTING & VIDEO 115 1 cytogenetic random aberrations, but to look for 2 nonrandom. Those that survive, by definition, are going 3 to be the ones that are most important. They are also 4 by definition ones that are going to be amplified and 5 we're going to be able to see. So there is a certain 6 for long-term culture techniques, that's our strategy. 7 The short-term techniques allow us to look at 8 the potential, and that's what I described for you in 9 terms of exposed to cells in culture to hydroquinone, 10 look at what types of aneuploid changes or chromosome 11 changes are associated with those cultures. 12 We can see the potential for involvement of 7 13 very readily, some evidence for 8. There are other 14 chromosomes we obviously want to look at. Whether those 15 are the ones that ultimately are clonally expanded and 16 become persistent, I can't tell you; but it certainly is 17 my hypothesis that that will be the case. 18 Q. In trying to follow that, then, the long-term 19 culture, when you use that terminology in this context, 20 that's something that you would like to be able to do, 21 but you can't do it yet? 22 A. We can do cultures after four or five weeks. 23 The state of the art for long-term culture of human 24 cells -- human bone marrow cells to date is on the order 25 of four to five weeks and still maintain what is called PATTERSON REPORTING & VIDEO 116 1 long-term repopulating. 2 Q. In this context, are you talking about for 3 long-term culture, starting off with pluripotential stem 4 cell component cells and maintaining them for four 5 weeks? 6 A. Yes, with significant repopulation of all the 7 different compartments, including the pluripotent 8 compartment, not just an exhaustion of pluripotent 9 cells. 10 Q. And if you start off with pluripotent stem 11 cells, is that the beginning phase? I mean, are all the 12 cells that you start off with when you attempt this, 13 long-term culture pluripotential stem cells, or do you 14 have a mix of pluripotential, myeloid, and lymphoid that 15 you're culturing up simultaneously? 16 A. Most of the types of experiments that we have 17 either planned or done are the second variety; however, 18 we now have the technology to separate out very early 19 cells. So I don't think we're far from being able to 20 have at least a highly enriched population of long-term 21 repopulating cells. I could tell you a lot more next 22 year than I can right now. 23 MR. BELFOUR: Excuse me, Herschel. Would this 24 be a good time for a break? We've been going for about 25 two hours. PATTERSON REPORTING & VIDEO 117 1 MR. HOBSON: It's up to you guys, whatever. 2 (Break was taken.) 3 Q. (BY MR. HOBSON) Is it possible, when you look 4 at this four weeks or so, that you can culture cells 5 that start. off as pluripotential stem cells, that you 6 can grow them long enough that they do fully 7 differentiate into the lymphoid or myeloid compartment 8 cells? 9 A. Oh, yes. That's not a -- yes. That's the 10 answer to that one. 11 Q. And you can culture the pluripotential stem 12 cells, expose those to the metabolites of benzene, and 13 then let them culture for four weeks or so before you do 14 any evaluation? 15 A. That is our ultimate -- that's one of the 16 ultimate types of experiments we hope to perform. We 17 are, as I said, trying to take it in stages, and we're 18 looking at both sides, the cytotoxic and cytogenetics, 19 and the altered-growth regulation side, and hope to 20 bring those together in the same context that you're 21 talking about. 22 Q. But you're not there yet? 23 A. No. 24 Q. Can you even go so far as to start off with a 25 cell from the pluripotential stem cell compartment, PATTERSON REPORTING & VIDEO 118 1 expose that cell to benzene metabolites, have it 2 differentiate, and then do the cytogenetic analysis 3 after it's differentiated to a myeloid or lymphoid cell? 4 A. Yes. 5 Q. Have you done that? 6 A. We are in the process of doing those studies 7 now. I don't have any results. 8 I do have results looking at prototype -- as I 9 said, prototype cell lines, but I don't have results on 10 cytogenetic analyses in human stem cells at the present 11 time; but I could have -- I mean, we're talking right 12 now, that type of experiment is underway now. So within 13 a matter of months, if we're successful, we may have 14 some information. 15 Q. I think we're planning on going to trial here 16 in about two or three weeks. Anyway, that would be 17 ready by then? 18 A. I'm sorry; that's a little bit- too challenging 19 a time frame. 20 Q. Is your ultimate goal or -- I should ask it a 21 different way. 22 Is a goal of your work to be able to take 23 cells from the pluripotential stem cell compartment, 24 expose those to benzene metabolites, let them culture 25 and then analyze the -- do cytogenetic analysis on all PATTERSON REPORTING & VIDEO 119 1 of the cells that grow from those pluripotential stem 2 cells? 3 A. I am not sure I understand precisely 4 whether I really understand what you are asking me. 5 Q. Let me do it again, then. 6 You take pluripotential stem cells, expose 7 those to benzene metabolites, culture them for a period 8 of time, you will still have, I would expect, some of 9 those differentiate into myeloid line cells and some 10 into lymphoid line cells. Do you plan on doing 11 cytogenetic analysis of both lineages? 12 A. Eventually, yes. 13 Q. How far away do you think you would be before 14 you could do that? 15 A. Right now we're looking at myeloid. If we go 16 into what are called a bioreactor type of system, we 17 could do both simultaneously, but that increases the 18 complexity of the system by orders of magnitude; so we 19 probably are better off at present trying to do these as 20 separate types of experiments. 21 Q. So months or years, at least? 22 A. Ultimately years, but early events we may be 23 much -- we may be much closer than that. We may be able 24 to say a lot more about dosimetry and early events, both 25 myeloid arid lymphoid, within the span of, say, for PATTERSON REPORTING & VIDEO 120 1 instance, a year or so. We already clearly have 2 differences in susceptibility of populations with 3 respect to seeing any observable response. 4 Q. I'm sorry, I didn't hear you. 5 A. Seeing any observable response at all, and 6 there the myeloid cells are much more acceptable. They 7 have the metabolic machinery to activate the 8 metabolites; lymphoid cells don't. So you are looking at 9 a totally different dosimetric to begin with, as well as 10 very clear-cut differences in cytokine response. 11 Q. The case that we're here on is a multiple 12 myeloma case. Multiple myeloma comes out of the 13 lymphoid stem cell compartment. Would that be accurate? 14 A. I would quibble with that. Because of the 15 nature of B-cell development, we can say a lot about the 16 ultimate clone origin of multiple myelomas, and the cell 17 of origin with considerable degree of confidence. I 18 think it's fair to say that people that are looking19 specializing in looking at the pathogenesis of multiple 20 myeloma are in general agreement that the cell of origin 21 is an early B cell. It's not a stem cell; it's not a 22 clonogenic cell in the context of colony formation. 23 It's a cell that has received its walking papers. 24 Now, the transformation may be at the point at 25 which it receives its graduation certificate as a B PATTERSON REPORTING & VIDEO 121 1 cell, but it is at that level, the level of an early B. 2 Q. Where would you put an early B cell if we 3 started off talking about a pluripotential stem cell? 4 A. We are not even -- we're even past the 5 committed progenitor cell stage. We are further down 6 we are -- that cell has no clonogenic potential for 7 colony formation. So it's not -- that's more -- the 8 last stage of B-cell differentiation that falls into the 9 colony-forming-unit type of progenitor is probably the 10 pre-B. And there are changes that occur, including the 11 loss of CD34, that are commensurate with the maturation 12 process, and that's a functional marker. Unlike many of 13 them, that is consistent with clonogenic potential or 14 colony formation. That's lost before gene 15 rearrangements lead to ultimate commitment as a B cell 16 to the actual epitope-specific B cell. So it has lost 17 its potential to form colonies before it has that 18 capability; and multiple myeloma cells have that 19 capability, so transformation occurs either there or 20 slightly later. 21 Q. And "there" is where again? 22 A. Post -- these cells have lost their ability 23 they're not CFU. It's not a pre-B; it's an early B. 24 Q. Even earlier than pre-B? 25 A. No. It's early B. It's post pre-B and PATTERSON REPORTING & VIDEO 122 1 early B. We have an advantage there because of the gene 2 rearrangements that happen in immunoglobulin gene 3 rearrangement and the cell committing to the 4 responding to a specific epitope, producing a specific 5 immunoglobulin molecule. Those events are very -- they 6 are at a fixed point in time when they occur, and 7 they're consistent with the phenotype of that cell as a 8 B cell. 9 Q. Let's see if I have got this close to right, 10 then. Multiple myeloma occurs as plasma cells or a 11 disease of plasma cells? 12 A. Plasma cell is -- if you recall when we were 13 talking about leukemic cells, I differentiated between 14 the circulating abnormal leukemic cells, which also may 15 divide, and the cell of origin, which was a progenitor 16 cell in the bone marrow. This is a rough analogy here. 17 A cell of origin is an early B cell. The 18 phenotype of the tumor cell is a plasma cell. So that 19 cell differentiates to become an antibody-producing cell 20 that specializes in antibody production, not 21 proliferation, and that's a plasma cell. That's the 22 hallmark of multiple myeloma. 23 So you get a single clone -- expanding clone 24 of plasma cells, they're all the same, they're all a 25 knockoff on each other. And they are the phenotypic PATTERSON REPORTING & VIDEO 123 1 manifestation of the tumor. The origin of the cell is a 2 B cell. 3 Q. A B cell that should have gone to being a 4 plasma cell gets, somewhere in the process, defective, 5 just like we were talking about for a leukemic cell, 6 that ultimately leads to the development of multiple 7 myeloma? 8 A. B cells have a number of facilities -- a 9 normal B c,21I. It can proliferate; it can respond to 10 stimuli and proliferate and also differentiate; and if 11 it differentiates, it differentiates into a plasma cell, 12 and basically becomes an immunoglobulin factory. 13 In multiple myeloma, the transforming cell 14 does that selectively; and so you get -- you get 15 proliferation and terminal differentiation to a plasma 16 cell, and you get all these plasma cells producing a 17 single type of immunoglobulin molecule. You don't have 18 the normal panoply of the responses the cell can undergo 19 and it doesn't need any. 20 So it's no longer an antigen process at all, 21 and those cells continue to produce -- they don't divide 22 rapidly, so they are not a highly proliferative tumor 23 type, but they're highly destructive, because they 24 produce immunoglobulin molecules and they influence the 25 pattern of cytokine release, so they cause lytic bone PATTERSON REPORTING & VIDEO 124 1 lesions and a variety of other symptom=. 2 Q. What would we -- what could we call this cell 3 that becomes the originator of the multiple myeloma 4 disease? 5 A. Well, certainly, the major descriptor that is 6 used by people in the field are early B cell. 7 Q. Is it your view that it is biologically 8 implausible that multiple myeloma can be caused by a 9 chemical agent? Can it be secondary to a chemical 10 agent? 11 A. We're touching on semantics here, and I don't 12 want to -- I don't want to play games. In terms of 13 transformation of lymphocytes, the basic patterns that 14 we see tend to be consistent with things that promote 15 proliferation, antigen stimulation, retroviral or 16 viral-mediated processes. They factor very strongly 17 into lymphoid peripheral diseases that Pre associated 18 with environmental exposure, and I am using 19 "environment" here not to mean -- I mean it in a very 20 generic sense. 21 For instance, the only analogy -- and this is 22 not a close analogy that I know of -- are lymphomas that 23 are associated with exposure to immunosuppressive drugs, 24 not alkylating agents, per se, but agents that are 25 potent immunosuppressive agents. It's seen in organ PATTERSON REPORTING & VIDEO 125 1 transplantation, it's seen in conditions where patients 2 are treated with immunosuppressive agents chronically. 3 Virtually all of those involve -- that I have seen 4 involve Epstein-Barr virus -- virtually every one I have 5 seen has molecular involvement of Epstein-Barr in the 6 tumor. 7 It is fair to say that there is no precedence 8 in that model system for a process like we see in AML, 9 because alkylating agents don't specifically do that. 10 Some patients with alkylating therapy may develop 11 lymphomas, but it's invariably because it's also 12 associated with a potent immunosuppressive component 13 process. We don't have anything comparable for multiple 14 myeloma, even to that level. 15 When we get into the epidemiologic data, there 16 are patterns that are seen. There is a whole lot of 17 inconsistency in the epidemiologic literature associated 18 with multiple myeloma. 19 I don't know of a plausible molecular 20 mechanism ire the lymphoid system to use by way of 21 analogy to say that it's biologically plausible that you 22 can derive. a lymphoid tumor via comparable process. All 23 evidence suggests it's a different process, certainly in 24 humans. 25 Q. is it your opinion that it is biologically PATTERSON REPORTING & VIDEO 126 1 implausible that benzene is a cause of multiple myeluma? 2 A. Based on my knowledge today and my knowledge 3 of the literature, I see no basis upon which to conclude 4 that benzene produces multiple myeloma. I cannot, in my 5 -- as I review the nature of the pathonenesis of the 6 disease, I cannot come up with a paradigm -- biological 7 paradigm for the production of multiple myeloma that is 8 consistent with what I know about the disease when we're 9 talking about benzene or other chemical exposure. 10 Q. So is the answer, then, it is not biologically 11 plausible, or is it that you just haven't been able to 12 figure one out? 13 A. Well, I think plausibility, there may be 14 degrees of plausibility. I do not mean to rule out 15 abjectly that there is no conceivable mechanism that a 16 chemical can cause multiple myeloma. 17 By the same token, I don't see any paradigm 18 that I could hypothesize that I would think is 19 consistent with what the data on either the pathogenesis 20 of the disease or the epidemiology will support. And 21 also by analogy, I cannot describe for you an animal 22 model that is at all satisfactory for predicting the 23 human condition with respect to the development of 24 multiple myeloma. 25 Q. Multiple myeloma, as a disease for rats or PATTERSON REPORTING & VIDEO 127 1 mice, they're not there, are they? 2 A. I don't think there is any model for multiple 3 myeloma as we know it. There are one or two spontaneous 4 B-cell neoplasms that have been defined in the balb/c 5 mouse. There is a polyclonal lesion that can be 6 produced in mouse, not by alkylating agents or agents 7 that produce leukemia in the mouse or even T-cell 8 leukemia in the mouse, but only agents that induce 9 chronic granulomatis inflammation and are highly 10 irritative adjuvants -- molecules that are 11 characteristically used as adjuvants. 12 And even in that model system, it's a 13 polyclonal lesion, to begin with, and it involves again 14 endogenous retroviruses, although agents that 15 classically cause leukemia in other mouse strains don't 16 induce plasmacytomas in that one model. It's, I think, 17 a horrifically lousy model for predicting anything. 18 Some agents are not associated with the 19 development of lymphomas and plasmacytomas, and as I 20 said, these is a retroviral confounder, the fact that it 21 readily appears to be a polyclonal response to 22 granulomatis inflammation. 23 Q. I take it that you have been informed at least 24 about some of what Dr. Goldstein has said about multiple 25 myeloma/benzene exposure. PATTERSON REPORTING & VIDEO 128 1 A. Yes, I reviewed a letter that he recently 2 wrote to Mr. Baggett. 3 Q. Is it your appreciation of what Dr. Goldstein 4 has said that Dr. Goldstein believes that there is 5 biological plausibility for benzene causing multiple 6 myeloma? 7 A. Is it my appreciation that Dr. Goldstein 8 thinks this is biologically plausible? 9 Q. Yes, sir. 10 A. I think that's a given, based upon his letter. 11 Q. Have you seen any of the basis of his 12 statement of biological plausibility enough so that you 13 can be instructive as to criticisms of it? 14 A. Yes, I think so. The criticisms I have of the 15 letter are more -- are the same criticisms in general 16 that I have with his original opinion piece, which he 17 published a few years ago, and I think are more specific 18 with respect to his letter. He has, by his own 19 admission, made an incomplete survey of the literature 20 associated with the development of multiple myeloma. He 21 has focused on certain studies that he is aware of and 22 that support his opinion, but he has not made an 23 exhaustive critical review of the epidemiologic 24 literature or the mechanistic literature associated with 25 the development of multiple myeloma. Prior to 1990, PATTERSON REPORTING & VIDEO 129 1 '91, '92, I could put mice in the same context, where I 2 had selective knowledge of what was out there, and I had 3 some concerns. 4 Having made a critical review of that 5 literature, I have formulated a formal opinion -- that 6 is, that there is no evidence to support that benzene is 7 associated with the development of multiple myeloma. 8 It's based upon that critical evaluation. 9 So, in general, I think an incomplete 10 evaluation of the literature is not useful because it's 11 the overall pattern that one sees in the literature as a 12 whole that I think is important, not focusing on 13 individual studies and in microcosm. 14 Having said that, there are a couple of points 15 that Dr. Goldstein makes that I think are critically 16 flawed with respect to his analysis. The most important 17 deals with his critical review of the Rinsky study, 18 which he places a great deal of weight on with respect 19 to his opinion. 20 The Rinsky analysis -- first of all, the 21 assignment the cohort that was done in the '87 22 analysis that found four multiple myelomas basically 23 provided no critical criteria for exposure duration. 24 One day's worth was sufficient to be included in the 25 cohort. So within the original four individuals, we PATTERSON REPORTING & VIDEO 130 1 have one who was exposed -- or had the opportunity of 2 exposure for four days. We had one who had the 3 opportunity of exposure for ten months. We then have 4 varying other durations of exposure higher than that. 5 If you accept that all of those are exposed 6 workers in the context of the definition for the cohort, 7 I have some problems with that. But let's take on the 8 assumption, simply for purposes of this discussion, that 9 they're all members of an exposed cohort. 10 Dr. Goldstein, taking that assumption, then 11 says that the fact that the risk decreases with 12 follow-up is most probably due to a dilution of the 13 exposure-dependent dynamic versus outcome within that 14 cohort, and that with time, you have a dilution 15 phenomenon; that's why it disappears. 16 I think that's a critical flaw, and the reason 17 is, the follow-ups -- first of all, the follow-ups that 18 demonstrate no significant association -- the reduction 19 in association are looking at an additional six-year 20 follow-up, are not looking at several decades. So we're 21 not looking at a huge increase in the latency, number 22 one. 23 Number two, the exposure dynamic, which is 24 predicated on a high-dose-exposure scenario, is the same 25 for all neoplasms in that cohort. At the same time, the PATTERSON REPORTING & VIDEO 131 1 risk of multiple myeloma has decreased in that 2 population with follow-up. 3 There has been a 50 percent increase in the 4 risk of acute myelogenous leukemia, same exposure 5 paradigm, same exposure dynamic. We are not seeing a 6 dilution that is related to the exposure and the nature 7 of the exposure relative to latency of the disease. 8 AML is going up with respect to the risk 9 associated with it. We have seen an increase in 10 collection of AMLs in that cohort. We have lost the 11 significance of the multiple myeloma, but his 12 explanation for why that study does not support his 13 opinion, I think, is critically flawed, because we are 14 not seeing a dilution of that exposure dynamic in terms 15 of the outcome. We certainly aren't seeing it for AML. 16 If he was correct in his evaluation or 17 critique of the follow-up studies, we should be seeing a 18 decrease in the risk of AML with time, and we're not. 19 We have seen an increase. 20 Q. You mentioned that he has been incomplete in 21 the mechanistic literature review. Which pieces from 22 the literature do you find he has not relied upon or not 23 included that he should have? 24 A. Well, he certainly has not taken the time to 25 provide certainly me with enough confidence that he is PATTERSON REPORTING & VIDEO 132 1 talking the same language. He talks about the fact that 2 benzene targets B cells as being commensurate with a 3 potential role in multiple myeloma. Well, the data on B 4 cells relates to experimental data and some clinical 5 data that they're suppressed, and much of that work was 6 conducted in my own laboratory, and we're looking at 7 suppression of proliferation at noncytotoxic doses 8 associated with exposure of benzene or metabolites to 9 circulating B cells. That's going to reduce the number 10 of B cells that are circulating, but it is decreasing 11 proliferation. It's not a cytogenetic event, it's not a 12 cytotoxic event. 13 So I cannot rationally see the connection 14 between that and the potential pathogenesis of multiple 15 myeloma. They are not to me -- an analogy would be if 16 water is overflowing the bucket, there must be a hole in 17 the bottom of it. They're not related with respect to 18 cause and effect. I don't see how that directly 19 follows. He refers to that as though it's a direct 20 association. 21 Basically I don't think he has taken the time 22 to completely evaluate the literature, and I would say 23 that that generically is the same with the early opinion 24 that he had where he did not review the entire 25 epidemiologic and biologic literature. He has an PATTERSON REPORTING & VIDEO 133 1 opinion based, I think largely, on gestalt, and he has 2 selected -- to a certain extent, I think he is guilty of 3 selecting explanations to support that without having to 4 deal with and wrestle with the entire literature that is 5 out there. 6 Q. But which mechanistic literature or citations 7 would you have included that he did not? 8 A. He didn't include any. He didn't include any. 9 He simply refers to the fact that benzene, quote, 10 targets B cells. 11 Q. Which mechanistic literature should he have 12 included or considered, in your view? 13 A. Certainly, what is known about the 14 pathogenesis of multiple myeloma. I have brought 15 several of those references here. Certainly, what is 16 known about the effects of benzene and/or metabolites 17 directly on lymphocyte proliferation. Certainly, 18 distinguishing particularly -- well, the inconsistent, 19 and in some cases, particularly unimpressive data with 20 respect to the effects of benzene exposure and 21 proliferating lymphocyte cytogenetics in humans are B 22 cell, they clearly are not B cells. All the models are 23 looking at T-cell proliferation-dependent processes. 24 None of that is addressed, and albeit where he does talk 25 about -- he does admit to the fact he hasn't reviewed PATTERSON REPORTING & VIDEO 134 1 the entire literature and he does admit to the fact that 2 some of the findings are inconsistent, some support 3 he thinks support his opinion; others he thinks are 4 contrary to it. I would say that that's consistent with 5 my perception of the literature as a whole. It tends to 6 be very inconsistent. 7 Q. Are you finished? 8 A. Yes. 9 Q. I'm sorry. You said "uh," and I thought you 10 were waiting for me there. I had a couple of things I 11 wanted to ask you about. I have what I think is a 12 chapter from a book that you wrote, "Benzene and Other 13 Hemotoxins." 14 A. Yes. 15 Q. I'm not sure when that was published. I 16 didn't bring that particular page, but I guess it was 17 sometime in the 18 A. 1991 is when it came out. 19 Q. You had a statement in here that I wanted to 20 talk to you about. You may have that chapter with you. 21 A. I do someplace. 22 Q. It might help to look at the same page, then. 23 A. Okay. 24 Q. Page 719 in that chapter. 25 A. Here we are. PATTERSON REPORTING & VIDEO 135 1 Q. Page 719 of the chapter. The paragraph on 2 "Study Design," the first paragraph, you make the 3 statement that "a major weakness is that a reliable 4 exposure history is exceedingly difficult to obtain," 5 and you go on there, "rarely, if ever, is chemical 6 exposure documented even remotely as well as the 7 diagnosis." 8 Do you still feel that way? 9 A. Are you talking about case reports? 10 Q. Yes, sir. 11 A. In case reports, certainly retrospective case 12 reports, yes, that chemical exposure is rarely 13 individuals publishing individual case reports rarely 14 provide any amount of information that would allow you 15 to quantitatively assess what the exposure or 16 characterization is. 17 Q. How would you characterize exposure history in 18 a case report as compared to exposure history that you 19 would find in the usual medical practitioner's files? 20 A. I don't understand. What is your question? 21 Q. A person becomes ill, goes to the physician, 22 perhaps they have leukemia, multiple myeloma, 23 some other blood disease, and they go to see a usual 24 oncologist/hematologist, and the oncologist/ 25 hematologist, as most physicians do, takes a history. PATTERSON REPORTING & VIDEO 136 1 Do you have any opinion as to how a reliable exposure 2 history that hematologist/oncologist is going to get, 3 compared to the exposure history you find in case 4 reports? 5 A. I am not sure I understand the question, but I 6 would -- if I understand what you are asking, you are 7 asking -- are you asking me how they differ -- are you 8 asking how they differ? 9 Q. Well, let me ask it this way. If you find 10 that exposure histories are unreliable in case-control 11 studies, wouldn't you expect that exposure histories as 12 found in a general practitioner's or a hematologist;' 13 oncologist's patient files would be even less reliable 14 for exposure history? 15 A. I would say that they are ostensibly 16 comparable. 17 Q. Both unreliable? 18 A. Well, it's degrees of unreliability. It 19 depends what the purpose is with respect to use of the 20 information. I am here making some very serious 21 criticisms with respect to the use -- the use of case 22 reports. That doesn't mean they aren't useful in the 23 aggregate to provide impetus to do quantitative studies, 24 either prospective or retrospective or what have you; 25 but as individual studies, they are very unreliable with PATTERSON REPORTING & VIDEO 137 1 respect to providing an adequate characterization of 2 exposure for use and causation, for assigning causation 3 or rendering a testable hypothesis. 4 The problems associated with a particular 5 history that that practitioner takes are comparable to 6 the same problems with a case report. The only thing 7 different is for reasons either associated with previous 8 case reports, bias, hunch, lightning, whatever, the 9 practitioner has elected to publish it; but in general, 10 the amount of emphasis that is put on exposure 11 characterization is very, very low in case reports. 12 It's understandable in the context in which 13 the practitioner is operating. He's either getting 14 self-selected information from the individual, or if he 15 is doing a case-control study, it may be by 16 questionnaire. I am simply saying that it varies 17 dramatically in terms of how reliable that information 18 is. 19 If it is someone who has been in an 20 occupation, that may be useful as a qualitative 21 indicator of the person's previous exposure history, 22 still not very useful for quantitative purposes. 23 MR. BELFOUR: Herschel, let me object. I 24 think your question was vague, because you asked him a 25 question about quantifiable. Then you asked him a PATTERSON REPORTING & VIDEO 138 1 question about taking a history of exposure, which could 2 mean any exposure, not whether that exposure was 3 quantifiable. With that objection, please proceed. 4 Q. (BY MR. HOBSON) Let me set up a hypothetical, 5 if I might. 6 A hematologist/oncologist has an office in 7 town, and the hematologist/oncologist has been in 8 practice for, let's say 30 years, and so when the 9 hematologist/oncologist around the 1960 time period set 10 up his practice, he believed that benzene and AML were 11 the only cause of these relationships from benzene. So 12 far okay? 13 A. I understand it. 14 Q. A patient comes in who has a disease, be it 15 multiple myeloma, leukemia, lymphoma, various different 16 variants of leukemia. 17 The practitioner's usual history-taking goes, 18 if he has acute myelogenous leukemia, I might ask him 19 about, Have you ever been exposed to benzene? But if he 20 doesn't have acute myelogenous leukemia, I don't take a 21 history of benzene exposure, and that's gone on for 30 22 years. Follow me so far? 23 A. Okay. 24 Q. Now, you go to that hematologist/oncologist in 25 town and say, Do you believe that benzene is causative PATTERSON REPORTING & VIDEO 139 1 of anything other than AML after 30 years of practice? 2 And the benzene -- I mean, the oncologist/hematologist 3 says to you, I have never seen in the histories that I 4 have taken any association between benzene and any of 5 these other blood, lymph, bone cancers that I have ever 6 seen. Does anything about that scenario strike you as 7 being less than reliable? 8 MR. BELFOUR: Objection, vague. 9 A. I definitely don't understand the question. I 10 understand the scenario that you just described. I 11 don't understand the question. I can try to answer what 12 I think -- let me try to restate it. 13 If what you are asking me is, Is that a likely 14 scenario, it's one possibility. There are several other 15 hypotheses within the context of the whole limitation 16 associated with the issue of case reports and that 17 process. It represents a hypothesis. 18 Some physicians might respond to a history of 19 benzene or reputed benzene exposure and another disease 20 as being novel and worthy of publication. Somebody else 21 might ignore it. They all suffer from the same 22 self-selection process, so I would -- you could be 23 looking at a case of omission; or a case of case 24 histories, as I said, can also represent a bias that is 25 not supported by fact or what subsequent study proves to PATTERSON REPORTING & VIDEO 140 1 be real. There is no denominator. 2 So even in the case that you have presented, 3 there is no way of determining whether or not what the 4 appropriate denominator is in the context of publication 5 of a case history. 6 Q. We may not have been communicating on the 7 question. I will try it again a different way. 8 If we're talking about this hypothetical 9 oncologist who sees patients who have various cancers, of 10 the blood and bone marrow, lymphomas, leukemias, 11 non-Hodgkin's -- lymphoma, the whole combination in his 12 practice, and he believes for the 30 years he's been in 13 practice that the only disease that he regularly sees 14 that is associated with benzene exposure is acute 15 myelogenous leukemia, so the only time he ever inquires 16 of a patient and their history about benzene exposure is 17 when he sees an AML. And then the question is presented 18 to that physician, Do you see any evidence from your 19 practice that benzene could ever cause any of these 20 other kinds of disease, other than AML? Would you think 21 his ability to answer that question would be rather 22 unreliable? 23 MR. BELFOUR: Again, objection. I don't think 24 you have given him the entire hypothetical. Herschel, I 25 think I understand what you're asking, but the PATTERSON REPORTING & VIDEO 141 1 deposition that you took lasted a lot more than a couple 2 of questions, but with that, go ahead. 3 A. It depends on a number of real intangibles. 4 If he's a practicing physician for 30 years, he may be 5 in an environment where he has seen individuals who have 6 been occupationally exposed to benzene and he has seen 7 AMLs he thinks are related to that. He may very well 8 have other cues that would be appropriate surrogates for 9 benzene exposure. 10 If an individual is employed, for instance, in 11 an environment where there is opportunity for exposure 12 to benzene, he may not -- if he sees a pattern there, 13 that could very well be -- in fact, that's used in 14 quantitative studies in some circumstances as a 15 surrogate for exposure, as you well know. 16 So I think a whole lot depends on the 17 individual, whether or not he -- what he basically uses 18 as a cue to tip him off to the prospect that benzene 19 could be the epidemiological agent. 20 I think, given the same limitations that we 21 are talking; about with respect to case histories, 22 they're all part and parcel of the same process. 23 Whether he asks for a benzene history may depend upon 24 whether he thinks that the individual's history is one 25 that is related to or could conceivably be associated PATTERSON REPORTING & VIDEO 142 1 with it. In the case of AML, it may be something he 2 specifically does. In the case of anything else, it may 3 just be part of the way he takes the history. 4 That's another one of the problems with 5 respect to an evaluation of case reports, is you're 6 talking about an art form here, not a quantitative 7 science; and it's that practice of the art that makes 8 one case history maybe worth reading and another not, 9 and you don't have a clue. 10 Q. (BY MR. HOBSON) Let me ask it a different 11 way, because I'm not sure yet that I have gotten my 12 question across to you. 13 If I started out, and this is my experiment, 14 my question is, What diseases of the blood, the lymph 15 and the bone marrow can benzene cause? That's the 16 question I am going to start out to answer. 17 And so what I am going to do then is for 30 18 years every time I come across someone who has one of 19 these diseases, I am going to take certain steps, and 20 those steps are along this line: If the person has 21 acute myelogenous leukemia, I am going to ask them about 22 their exposure to benzene and learn as much about it as 23 I can. If they have any other disease, I am not going 24 to inquire at all about their history of exposure to 25 benzene or in their work history enough to know if they PATTERSON REPORTING & VIDEO 143 1 ever had exposure to benzene. 2 Would my method be flawed? It's pretty 3 obvious it would be, wouldn't it? 4 A. It depends -- I don't think it's that clear s cut, and the reason I don't think it's that clear-cut is 6 there are other cues that physicians use. 7 For example, did -- well, let's discuss the 8 total futile exam. If the only patients that individual 9 sees come from a specific occupation, a specific 10 environment, you will never know what a denominator is, 11 ever. All he does is see cases. All he can do is tell 12 you whether he's seen a lot or a little vis-a-vis that 13 one universe, so he has no denominator. There is no way 14 he can possibly relate that. 15 If, however, he sees a cross-section of the 16 community and he sees the diseases that appear as a 17 cross-section of the community, and he doesn't have a 18 cue with respect to an association with an occupational 19 setting that is consistent in his own experience with 20 benzene exposure, then he may not feel that that's an 21 area worth looking at, vis-a-vis the others. 22 If he sees an AML, he may have a totally 23 different criteria based upon the fact he has seen an 24 association. 25 All I am saying is that if he is seeing a PATTERSON REPORTING & VIDEO 144 1 cross-section of a community, there are a number of 2 other surrogates or other behavioral characteristic 3 that may provide cues with respect to the probability of 4 an exposure related to benzene that are not necessarily 5 going to be lost in that process. 6 Q. Would you agree that before a physician should 7 express an opinion about whether or not benzene causes 8 any particular disease or not, based on what he has seen 9 in his clinical practice and learned from the histories 10 that he has taken, that he should first know where the 11 potential for benzene exposure is? 12 A. I think the responsible answer to that 13 question lies in the degree. It's not an all-or-none, 14 yes-or-no question. 15 One would hope that there was some knowledge 16 with respect to potential levels of exposure, but at 17 what point that becomes critical, I can't tell you. 18 Q. The physician might not have to know each and 19 every potential exposure known to mankind, but ought to 20 at least have a good grasp of what is generally known 21 about benzene exposure; is that what you are saying? 22 A. Some concept of what the -- in very crude 23 terms, where the opportunities for exposure are. 24 Q. Like there is benzene in gasoline, for 25 instance, if a person is a service station attendant PATTERSON REPORTING & VIDEO 145 1 A. Well, now we're getting -- I disagree with 2 that. I disagree, based upon my knowledge of the 3 benzene content in gasoline, whether that constitutes a 4 health risk with respect to benzene exposure. 5 Q. I wasn't asking about health risk. I'm asking 6 about whether or not -- shouldn't a physician at least 7 have an appreciation that there is some benzene in 8 gasoline? 9 A. Well, I personally do not feel that the amount 10 of benzene that is present in gasoline constitutes a 11 risk with respect to benzene exposure. I think that 12 chronic exposure to gasoline provides many others. 13 If you're talking cumulative exposure, then 14 gasoline -- if you assume that cumulative exposure to 15 low levels of benzene plays a role in benzene 16 leukemogenicity, then gasoline, like smoking and 17 everything else, constitutes a potential source. 18 I do not believe that the amount of benzene 19 present in gasoline constitutes a health risk with 20 respect to benzene exposure. So I would disagree with 21 you on the premise that the amount of benzene that is 22 present in gasoline constitutes something that a 23 physician should be necessarily aware of. 24 Q. Where would you think that a physician should 25 necessarily be aware of benzene being out there in the PATTERSON REPORTING & VIDEO 146 1 world for workers to be exposed to it? Where are those, 2 then, that you would expect a physician to know about, 3 if he is going to give an opinion, based on his clinical 4 practice, that benzene is or is not causative of 5 anything? What should he know? 6 A. Well, without talking specifics, certainly 7 historically, there is no, I think, argument that 8 opportunities for exposure to benzene have been found in 9 the petroleum industry, and have been found in refinery 10 situations. Certainly, in this country, those are major 11 potential sources or opportunities for exposure to 12 benzene. 13 Q. Any others? 14 A. It depends. I mean, we get into the highly 15 specific. I have seen examples of individuals exposed 16 to benzene in a number of scenarios that are not 17 generic; they're highly specific -- using it to clear, 18 metal parts that are hot -- very bizarre types of 19 scenarios where benzene is employed. 20 Historically -- certainly not today, but 21 historically -- I mean, certainly in the '50s and into 22 the '60s, and perhaps through the '60s -- painting was a 23 major source of benzene exposure. In China that 24 constitutes probably the heaviest source of benzene 25 exposure, even given the other horrendous exposures they PATTERSON REPORTING & VIDEO 147 1 have. 2 Do we have to go back and invoke shoe workers 3 in Europe? I mean, using benzene as a glue is a highly 4 hazardous practice. 5 Q. What about using benzene in paint removers? 6 (Mr. Tillson left the deposition room.) 7 A. Consistent with benzene contamination of other 8 solvents and aromatics into the '60s, that is a logical 9 potential source. With the efforts to remove benzene 10 from most aromatic stocks that are used as solvents in 11 the '70s and certainly up through the present, I would 12 say that that's a highly improbable source today. 13 Q. What about up to the mid-1970s, are you aware 14 that 15 A. Any aromatic distillate fraction that 16 contained 10, 15 percent benzene is a potential source 17 of benzene exposure, and at those levels the presence of 18 substantive benzene isn't going to afford effective 19 protection. At lower levels, if you're below 5 percent, 20 then I think you have to get into the issue of what the 21 potential exposure is and whether or not there is a 22 tremendous excess of toluene or xylene present to 23 evaluate the potential hazard associated with benzene. 24 Q. In your chapter here on page 724, you say that 25 gasoline currently ranges from 0.5 to 2 percent benzene PATTERSON REPORTING & VIDEO 148 1 concentration. Any knowledge at all of what 2 historically gasoline has had in it as far as benzene 3 content? 4 A. Certainly 3, 4, 5 percent. I don't know if 5 there were higher proportions in the '50s or '60s. 6 There has been an effort -- historically, I am aware 7 that there has been an effort to reduce benzene 8 certainly below those levels, even prior to the general 9 consensus with respect to the potential hazards of 10 benzene, because of the economic issues. You don't want 11 to throw benzene in gasoline; you want to market benzene 12 as benzene. 13 Q. Would you know, one way or the other, if any 14 of the refiners who had the ability to extract benzene 15 at their refinery would ever blend benzene back into 16 their gasoline stocks when the market for benzene was 17 soft? 18 A. I have absolutely no general or specific 19 knowledge about that. 20 Q. Would that matter to you in knowing about tile 21 benzene content in motor gasolines? 22 A. Again, it doesn't -- that would not tell me 23 what the levels were. To me, what is important with 24 respect to potential for hazard is going to be the 25 amount of benzene that is present. PATTERSON REPORTING & VIDEO 149 1 Q. Have you seen reports of motor gasolines 2 containing 8 to 10 percent benzene? 3 A. No, I have not. 4 Q. That would surprise you, then? 5 A. It would surprise me today. I am not sure it 6 would surprise me back in the '50s. 7 Q. How about in the 1970s, through 1975? 8 A. I would be dumbfounded after 1975 to hear 9 that. 10 Q. I would be dumbfounded after '78. 11 MR. BELFOUR: Herschel, why don't you show him 12 the papers you have, if we're thinking of the same 13 thing, and just ask his opinion. In all fairness, if 14 you have got something in a document, why don't you just 15 show him it. 16 Q. (BY MR. HOBSON) Basically what you're saying, 17 Dr. Irons, is that up until the benzene ETS was 18 proposed, you wouldn't be terribly surprised to find 19 high concentrations of benzene in many petroleum 20 products; is that fair or not? 21 A. With the caveat that "high" could mean many 22 things, that's fair. 23 Q. 10 to 15 percent or more. 24 A. It would depend again on what it is. I would 25 be kind of surprised to find 10 percent in gasoline. PATTERSON REPORTING & VIDEO 150 1 Q. Continuing there on page 724 -- as a matter of 2 fact, in the same paragraph under "Environmental 3 Exposure," you lay out two scenarios: one is no 4 threshold, and the other is that there is a threshold. 5 for benzene exposure. And what I assume you're talking 6 about there is cancer risk; would that be accurate? 7 A. Yes. 8 Q. I take it that you are of the opinion that 9 there is a threshold concentration? 10 A. Yes, and I -- yes. 11 Q. I didn't mean to cut you off. 12 A. I think the more we know about the process of 13 carcinogenesis, the more it becomes clear that even 14 carcinogens have practical thresholds. And certainly, 15 in terms of mechanisms of benzene toxicity and early 16 events associated with the cytogenetic aberrations we've 17 been talking about, I think, both practically and 18 theoretically, there is a threshold for those types of 19 effects. 20 Q. I was curious about your choice of words here, 21 which having come to know you as I do, I think you 22 usually choose your words very carefully. 23 Your scenario for a threshold, you say here, 24 "a threshold concentration below which benzene is not 25 toxic"; you go on though, and you say, "and that PATTERSON REPORTING & VIDEO 151 1 intermittel-it exposures to high concentrations present a 2 much greater risk than low-level exposure." 3 Did you, in fact, choose your words carefully 4 there? Do you still agree that that's what you should 5 have put in this context? 6 A. I need to find this. 7 Q. It's in the paragraph "Environment Exposure," 8 the last sentence on page 724. 9 A. Could you repeat the question. Did I choose 10 my words carefully? 11 Q. Yes. 12 A. I think so. 13 Q. I am curious, then, to know why you put 14 "intermittent." 15 MR. BELFOUR: I'm sorry. Are we reading the 16 same sentence, where the sentence begins, "If, on the 17 other hand, one assumes"? 18 MR. HOBSON: I believe we are. 19 A. Yes. 20 Q. (BY MR. HOBSON) Isn't the word "intermittent" 21 there on your copy? 22 A. What I am saying there -- what I think I am 23 saying and was saying -- and I still agree with -- is I 24 am separating what I think is the biologically plausible 25 mechanism for benzene toxicity and leukemogenesis, a PATTERSON REPORTING & VIDEO 152 1 risk hazard, from the low-level cumulative paradigm that 2 has been the default policy in many risk assessments in 3 regulatory context for benzene. And what I am saying is 4 that if you -- if you assume there is a threshold and 5 that the risk is associated with high-level exposure, 6 even intermittent, but high-level exposure, then the 7 overall impact associated with environmental levels, 8 ambient levels, is negligible. That is to say, 9 low-level exposure to eggs and woods, potentially even 10 cigarette smoke, may, in terms of benzene-risk health 11 effects associated with benzene, be negligible. If you 12 assume cumulative exposure is important, then all of 13 those have to be added in. 14 So all I am saying there is that ambient 15 exposure, the environmental low levels that we encounter 16 in daily life, are negligible if you assume there is a 17 threshold, and that the risk is associated with exposure 18 to high levels. 19 Q. So you're comfortable with leaving in the word 20 "intermittent" in this sentence? 21 A. I think virtually all high level -- I mean, 22 there are exceptions; but I think virtually all 23 high-levels exposures are intermittent exposures. 24 Certainly, the Chinese experience is in that context. 25 People are not uniformly exposed. PATTERSON REPORTING & VIDEO 153 1 I mean, this obviously doesn't apply to the 2 Turkish shoe workers, but people are not historically 3 uniformly exposed to 100 parts per million. They're 4 exposed to one -- to a few hundred parts per million for 5 some period of time, interspersed by periods of time 6 when they're not exposed to -- when they're basically 7 exposed to negligible levels. 8 Q. Where would you then classify your use of 9 "intermittent" here? Are we talking about an exposure 10 that occurs and then there are many days that pass -- is 11 that intermittent -- before the next exposure? 12 A. Could be. It would be -- that's consistent. 13 By "intermittent," my use in this sentence is meant to 14 represent a dynamic, as opposed to a specific, scenario. 15 And I am simply saying that maybe I should have chosen 16 the word "transient" as opposed to "intermittent," and 17 that probably would be more precise, because I do place 18 -- certainly today, based upon the work I have done, I 19 think "intermittent" represents something a little bit 20 more specific than this. 21 But what I am basically -- transient or 22 intermittent, I am simply saying that if you assume 23 there is a risk associated with exposure to benzene, 24 that is associated with a temporal exposure to high 25 levels -- and which I think is the majority of cases PATTERSON REPORTING & VIDEO 154 1 that would be described today -- and from most 2 situations, even in the past, that's where the risk is. 3 If you want to talk about specific 4 intermittency, we can talk about that, but that would be 5 outside this context. I mean, it could be included; 6 it's just that this doesn't refer to a specific regimen 7 Q. Is there a threshold quantitative value that 8 you can give me for benzene? 9 A. For acute or chronic? 10 Q. Exposures? 11 A. Yes. 12 Q. Chronic. "Chronic" meaning over a period of 13 years, as opposed to one or two short-term exposures. 14 A. We have been down this road before. I will, 15 for completeness, go through it, and then I will 16 hopefully provide you with something to clarify the 17 issue. 18 None of us are going to argue abut exposure 19 to 100 parts per million. We can talk about duration, 20 but that's trivial, I think, in the context of most 21 situations. If you're exposed to 200 parts per million 22 routinely, you are going to be exposed for durations of 23 time that are going to be significant. 24 I personally am concerned about exposures 25 the order of 50 parts per million with respect to PATTERSON REPORTING & VIDEO 155 1 potential for bone marrow toxicity and for 2 leukemogenesis. I don't think the data is as strong as 3 it is for 100. There are people who will argue with me 4 about 50. 1 am on the conservative side with respect to 5 50. You can start going below there in terms of the 6 actual exposure levels, and you run out of reliable data 7 very quickly. 8 At the level of below 25 parts per million, 9 certainly below 10 parts per million -- there is no 10 evidence that exposure to 10 parts per million is, in my 11 opinion, associated with a leukemogenic risk. 12 Now, that is separate and distinct, a 13 time-weighted average. Because within the context of 14 the broadest excursions possible, a 10 ppm time-weighted 15 average can include very significant excursions to 1-, 16 even 400 parts per million. 17 So within the context of a time-weighted 18 average of 10 ppm, you can have a scenario that I think 19 is a potential leukemogenic risk, but I don't believe 20 exposure to 10 ppm is associated with that. I don't 21 think there is evidence to support that. 22 Q. Let me see if I can clarify this a little bit. 23 i~ I have got a job that requires me, say, 24 twice a day to be exposed to benzene that is on the 25 order of 100 to 200 parts per million, for 30 minutes, PATTERSON REPORTING & VIDEO 156 1 and I do that job for five years, is that a significant 2 risk factor for developing AML, in your view? 3 MR. BELFOUR: Excuse me, Herschel. You said 4 every day for five years? 5 MR. HOBSON: Every workday, five days a week, 6 that's part of my job. I might get some vacation. 7 A. Well, you have picked a duration that is 8 problematic, in that if you had said 15 or 20 minute=, I 9 would question the significance of it with respect to 10 actual delivery of metabolites to the target organ, the 11 sufficient concentrations to produce damage. There is 12 time required for that. 13 But having said that, I think it is 14 unrealistic to assume that someone is going to have an 15 exposure like that. If they are going to be exposed to 16 100 parts per million, I would imagine, day in/day out, 17 they are going to be exposed for longer periods of time; 18 as you go above 30 minutes, we don't have a disagreement 19 or a question. I think, to any great extent, exposure 20 certainly on the order of an hour or two are going to be 21 sufficient to produce a health risk. 22 Q. (BY MR. HOBSON) Let me give you a 23 hypothetical and get your thoughts about it. 24 Assume for a moment that I am a worker on a 25 benzene unit where benzene is being manufactured, and my 157 1 job as an operator -- I am an outside operator, and I go 2 out, and I routinely go through the unit and do my 3 check; but part of what I do is I have to catch 4 quality-control samples, which each sample may take me, 5 say, five minutes. And I am just like a little 6 honeybee, I go from this point to that point, and I have 7 got a rack of quart bottles. 8 When I get to my point to take my sample, I 9 open up a valve and I bleed benzene on the ground for a 10 while, a concrete pad, until I make sure I have got 11 fresh product. And then I stick a quart bottle 12 underneath the tap, and then I fill it up, turn it off, 13 put the screw tap on it, then I go to the next one. 14 And I collect this tray of benzene samples in 15 this manner. Benzene is hot to the touch; doesn't burn 16 my skin; I don't get third degree burns; but it is hot 17 to the touch. 18 I take the benzene then to a small building 19 that is maybe 4 feet by 4 feet with no ventilation, 20 other than in the summertime; I open up the door, and 21 maybe I can crack the windows, and I do freeze points 22 where I take the benzene, and I pour it into a small 23 beaker with a thermometer, and put that little beaker in 24 a bed of ice, and I sit there, and I shake it until I 25 see frozen benzene forming on the thermometer, and I PATTERSON REPORTING & VIDEO 158 1 note the temperature, checking the purity of the benzene 2 in this manner. 3 Now, all the time I am catching my sample-. 4 all the time I am doing my work in the freeze-point 5 shack, I smell benzene; and I do this at least twice a 6 day, and it takes me approximately an hour total to do 7 all of this. 8 Is that a significant risk exposure for 9 leukemia? That's my job. I do it five days a week, 10 eight-hour shifts. Sometimes I get doubled over, so I 11 do it 16 hours a day. Is that anything that you ought 12 to be worried about? 13 MR. BELFOUR: Just objection; vague. I am not 14 sure you have enough specifics there to form a 15 quantification of exposure, but subject to that 16 objection, you're free to answer the question. 17 A. There are certain specifics that I would like 18 to know before giving you a precise definitive response 19 to that. 20 For instance, whether or not the individual is 21 standing right over it; how far away they were from it. 22 Smell is only useful in the context, to me, 23 qualitatively, because individuals can detect benzene at 24 concentrations that I think are lower than those that 25 pose a health risk. PATTERSON REPORTING & VIDEO 159 1 Given those reservations 2 Q. (BY MR. HOBSON) The person is at arm's 3 length. I mean, he has the bottle in his hand. 4 A. Given those reservations, the scenario you're 5 painting is one that I think is certainly consistent 6 with the opportunity of exposure to benzene at 7 concentrations over periods of time that would pose a 8 health risk with respect to the development of acute 9 myelogenous leukemia. 10 Q. Have you ever heard that exposure scenario 11 before or anything close to it? 12 A. I have heard descriptions comparable, I think, 13 in the context of washing tools, using benzene as a 14 cleaning solvent, both on bodies, as well as tools, that 15 type of activity, to the degree that they are reliable 16 descriptions of an exposure scenario that constitute a 17 serious health risk. 18 Q. In your book chapter, I notice that -- at 19 least it seemed to me, that you did a fairly good job of 20 referencing many of the statements that you make in the 21 bibliography; but there were a few that I didn't see 22 referenced, and I wanted to ask you about at least one 23 of those. 24 On the same page 724 25 A. You didn't like this page, I can tell. PATTERSON REPORTING & VIDEO 160 1 Q. No. As a matter of fact, I found it very 2 interesting. I did kind of like it. 3 On the right-hand column, about halfway down 4 the first paragraph, after reference 98, it starts ot, 5 "On the other hand, there is considerable evidence to 6 suggest that there is a threshold concentration for 7 benzene leukemogenesis." You didn't give a reference. 8 A. I am referring to both mechanistic, as well. as 9 epidemiologic, studies; and I think it is fair to sp 10 that certainly the most recent evaluations of the 11 epidemiology associated with benzene provide increasing 12 evidence of threshold, and a further clarification of 13 the exposure dynamic that is associated with 14 leukemogenesis. 15 At this particular point in time when I write 16 this, we didn't have the benefit of the most recent 17 updates, but I think, still in terms of the availability 18 of the existing data, what we just were talking about. 19 earlier, with respect to the reliability of the exposure 20 assessment, suggests that there is a threshold. 21 In theoretical terms, there are practical 22 thresholds that are beginning to be accepted by many, 23 even with respect to known genotoxic carcinogens, such 24 as aflatoxins. There is a practical threshold for 25 aflatoxin carcinogenesis that you can see in this PATTERSON REPORTING & VIDEO 161 1 country vis-a-vis southeast Asia, for example. 2 Mechanistically, in terms of what we know 3 about leukemogenesis, the mechanisms of benzene 4 leukemogenesis involve aneuploidy; they involve 5 nondisjunctional events. It's clear that they play a 6 major role in the process, and there is increasing 7 acceptance of that as being a major paradigm in 8 leukemogenesis as opposed to assuming a single-hit type 9 of mechuriism, and in that context we're dealing clearly 10 with threshold phenomenon. 11 Q. Going to page 727. 12 MR. BELFOUR: Excuse me, Herschel. Are we 13 going to stop in a few minutes? I need to make one 14 phone call. Take a break in a little bit. 15 MR. HOBSON: That would be fine. 16 Q. (BY MR. HOBSON) The bottom sentence, the last 17 sentence on the left-hand column, starts out, 18 "Consistent with," you say, "interrupted exposure 19 regimen, i.e., multiple-exposure intervals interrupted 20 by several gays, are more effective in producing bone 21 marrow supdression than repeated daily administration." 22 Now, is that still your view? 23 A. We have seen that at very high levels -- very 24 high levels of exposure in mice, 300 parts per million, 25 100 parts per million. Certainly, in the experimental PATTERSON REPORTING & VIDEO 162 1 data, it was much more difficult to see a difference 2 between continuous and interrupted exposure when you get 3 below 100 parts per million. By 25, you can't see 4 anything. 5 I would say it's at interrupted exposure, :and 6 the interval is important because of the cycle-specific 7 issues; but given the right dynamic with respect to 8 regimen, I think that interrupted exposure is at least 9 as potentially effective as continuous exposure. 10 Q. And, if I understand what you are saying here, 11 is probably more important? 12 A. That's my bias. I would say the experimental 13 data at least supports at least comparable. 14 The problem with this, to be perfectly honest, 15 is the most definitive data that is associated with 16 administration of metabolites -- it gets washed out, to 17 a certain degree, when you are looking at benzene 18 exposure in vivo; but at high doses it is -- certainly 19 the regimen dependence is also dose-dependent. 20 High-dose regimen-dependence appears to be more of an 21 influence than at lower doses, but the duration makes a 22 difference. You are not talking months apart. We're 23 talking a few days apart. We're talking within the 24 context of certainly a week and not more than that. 25 Q. Are we talking exposures of 1- to 200 parts PATTERSON REPORTING & VIDEO 163 1 per million for what duration? When you have the 2 intermittent exposure? 3 A. Metabolism -- primary and secondary metabolism 4 are probably going to be saturated for about three 5 hours. So three hours would be optimal. Obviously, I 6 don't think you need three hours to produce toxicity, 7 but there is a -- there is more to it than simply saying 8 what the cumulative exposure is, because you have got 9 differences -- certainly, this is opinion at this point 10 in time. But we're looking at, at least two types of 11 processes that we discussed here today that appear to be 12 very important in leukemogenesis for AML, in general; 13 and what also benzene metabolites are capable of doing, 14 but they have totally different dose dynamics. Some are 15 low-dose effects; some are high-dose effects. 16 So you're probably look -- it's probably the 17 rate of rise and the duration of exposure that 18 determines whether or not you have precisely the optimal 19 conditions to produce toxicity. 20 So the dynamic with respect to regimen 21 probably plays an important role in the process; it's 22 not just what that cumulative exposure is. So somewhere 23 between 20 minutes and 3 hours you are going to be 24 looking at a paradigm that is important with respect to 25 toxicity. PATTERSON REPORTING & VIDEO 164 1 Q. In this road, from beginning of exposure 2 saturation, is it nonlinear? 3 MR. BELFOUR: I'm sorry. I object as vague. 4 I don't understand what you are asking. 5 Q. (BY MR. HOBSON) The saturation curve or the 6 curve that goes to saturation, concentration of benzene 7 metabolite in the compartment. 8 A. Once you impose upon that, the issue and the 9 dynamics of alterations in stem cell progression 10 proliferation, it is a stochastic process; it is not 11 linear. I am not a mathematician, but it is hard for me 12 to perceive of a linear model for that. 13 Q. If we were to graph out what you believe to be 14 the response or the concentration for dose in the 15 compartment from beginning of exposure to saturation., 16 does it tend to -- does the dose -- or the exposure, I 17 guess would be more proper -- does the exposure tend to 18 go up steeply and then flatten off, you believe, as you 19 get closer to saturation? 20 A. I am not sure I can honestly say one way or 21 the other. I would suspect that, without taking it to 22 an absurd extent, probably the most risky scenario does 23 not involve the steep ramp, because somewhere in the 24 middle and somewhere, depending upon the dynamics of the 25 exposure, you are going to alter stem cell regulation; PATTERSON REPORTING & VIDEO 165 1 and that's probably more important than cytotoxicity 2 because you are increasing the number of cells at risk. 3 And that's far more important than just doing the dirty 4 deed in terms of 5 Q. Saturation? 6 A. -- saturation. So I think that, you know, if 7 you are exposed to 1000 parts per million, like now, for 8 20 minutes, that probably isn't going to do anything; 9 whereas, if you rank -- depending upon what the dynamics 10 are, and whether you were exposed yesterday -- for all I 11 know, exposure to 25 or 30 parts per million yesterday 12 may pose a risk with respect to exposure to 1000 parts 13 per million today. 14 So I think, you know, saying that, yes, a 15 steep curve is the most potentially damaging -- I don't 16 really think there is a basis to conclude that. You 17 might actually be able to come up with speculation that 18 a more complex exposure dynamic would be more damaging. 19 MR. HOBSON: Let's take a break. 20 (Break was taken.) 21 MR. BELFOUR: Herschel, am I correct in 22 assuming that when we talk about the risk of injury, 23 we're talking about AML, because you haven't always said 24 "AML" when we talked about the disease, and I don't 25 want the record to reflect that you have asked a PATTERSON REPORTING & VIDEO 166 1 question without saying "AML," and the doctor has 2 responded about health risks assuming you were asking 3 about AML. 4 MR. HOBSON: Well, I presume that Dr. Irons is 5 always answering "AML" since he said that's all he 6 thinks benzene causes. It may not be what I wanted to 7 ask, but I think that's what was his response. 8 Q. (BY MR. HOBSON) Do you agree that multiply: 9 myeloma is a tumor of plasma cells? 10 A. Given all the discussion we have had today, 11 yes, within the context of what we discussed today. 12 Q. Do you agree that plasma cells are in the body 13 cells derived from B lymphocytes which are located 14 primarily in the bone marrow? 15 A. Yes. All the evidence points to a bone ma,--row 16 origin for multiple myeloma. 17 Q. Do you agree that multiple myeloma in almost 18 all cases begins with a somatic mutation in a single 19 cell? 20 A. It certainly is a clonally derived lesion. 21 What the initial events are, we really don't know. 22 Multiple myeloma is not that well understood. 23 Q. Do you agree that it arises in a single cell, 24 though? 25 A. Yes, I do. PATTERSON REPORTING & VIDEO 167 1 Q. Are you familiar with a monoclonal antibody 2 that is demonstrated on a serum protein electrophoreses? 3 MR. BELFOUR: I'm sorry? 4 MR. HOBSON: That doesn't make any sense to 5 me. Never mind. 6 MR. BELFOUR: Herschel, may I interrupt for a 7 moment. Did you ask whether the cells that divide and 8 become plasma cells are derived from B lymphocytes or B 9 lymphocyte lineage? 10 MR. HOBSON: I don't remember. 11 THE DEPONENT: I think he said "B 12 lymphocytes." 13 MR. BELFOUR: Okay. Thank you. 14 Q. (BY MR. HOBSON) Would you agree that a 15 definitive diagnosis of multiple myeloma depends on the 16 presence of an increased number of plasma cells, 17 sometimes with abnormal morphology in the bone marrow? 18 MR. BELFOUR: Excuse me. I object to that as 19 vague. Hs one component diagnostic criteria? As the 20 definitive diagnostic criteria? 21 MR. HOBSON: Definitive. 22 A. I don't want to get caught up in semantics 23 here. The clinical presentation of multiple myeloma 24 should involve -- the clinical diagnosis, definitive 25 diagnosis, should involve bone marrow examination. PATTERSON REPORTING & VIDEO 168 1 There are other aspects that are consistent with the 2 diagnosis of multiple myeloma that can be found in 3 individuals who don't have multiple myeloma, such as -she 4 presence of a monoclonal gammopathy or presence of 5 M protein, even though they are very -- they are 6 components of the diagnosis. 7 Lytic lesions, certainly on X ray, can always 8 get you there, but ultimately the histopathologic 9 diagnosis of the bone marrow provides you with a 10 definitive conclusion. 11 Q. (BY MR. HOBSON) In some individuals with the 12 benign monoclonal gammopathy, they will occasionally 13 progress to multiple myeloma, will they not? 14 A. Yes. I mean, this is one of the biggest 15 problems in determining what the nature of the clonal 16 progression is for multiple myeloma. I think 1 percent 17 of individuals -- there is a certain -- I think it ice. a 18 very low percentage, but a certain percent of 19 individuals are walking around with monoclonal 20 gammopathies; 3 percent of people over 50 have actually 21 M proteins present in their serum. Obviously the 22 incidence of multiple myeloma doesn't come anywhere 23 close to that. 24 So there are aspects of disregulation of 25 B cells associated with plasma-cell differentiation that PATTERSON REPORTING & VIDEO 169 1 occur in clinically normal individuals and that have no 2 consequence with respect to predicting the probability 3 of the onset of multiple myeloma. 4 On the other hand, there are people with 5 M protein levels that obviously go on to develop 6 multiple myeloma. It's a clonal lesion at the point at 7 which you have got M protein. So there is a clonal 8 gammopathy present. 9 "What differentiates that individual from 10 somebody who develops multiple myeloma" is a -- is the 11 question. 12 Q. You do agree that benzene causes aplastic 13 anemia? 14 A. Yes, with the caveat that the presentation is 15 not what most purists think of as aplastic anemia, 16 necessarily. 17 The difference is the very high-portion 18 preponderance of myeloid proliferative abnormalities in 19 the bone marrow of individuals who develop pancytopenia 20 associated with benzene. A real strict definition of 21 aplastic anemia does not include a hyperplastic bone 22 marrow. I am not parsing hairs. 23 Experts in aplastic bone will distinguish a 24 difference between induced aplastic from hyperplastic on 25 the basis of the fact that if you have got a PATTERSON REPORTING & VIDEO 170 1 hyperplastic bone marrow, it is not really aplastic 2 anemia, but it is certainly a refractory anemia, and it 3 certainly has all the clinical significance of aplastic 4 anemia. 5 Q. If I recall earlier testimony of yours 6 correctly, you have said that a person can have AML that 7 is benzene-induced and you would expect to find a prior 8 effect upon the blood, but it may not necessarily be 9 clinically evident from the usual blood tests that 10 run, such as white blood counts and complete blood 11 counts. Am I right? 12 A. That's a reasonable characterization, yes. I 13 think that the majority of people who develop AML, 14 secondary to benzene or anything else, the literature 15 suggests they are going to have some preexisting 16 evidence. That doesn't mean everybody presents that 17 way. It's probably on the order of -- it's certain1y 18 greater than 50 percent. 19 But one could -- I think it's conceivable that 20 someone could have bone marrow injury and have 21 subclinical presentation. 22 Q. And still develop an AML to be benzene 23 induced? 24 A. I think the majority will have some 25 evidence -- preexisting evidence, but that you cannot PATTERSON REPORTING & VIDEO 171 1 exclude that possibility. 2 Q. And certainly you have opined that benzene is 3 a known cause of acute myelogenous leukemia. We just 4 went through that. 5 A. That is not an issue here. 6 Q. What about acute myelocytic leukemia; is that 7 a different term for the same disease? 8 A. The subtypes of AML within the FAB 9 clarification, with some exceptions, most -- not all 10 most of the subtypes are associated with secondary AML 11 and benzene. There are some exceptions. Myelocytic is 12 the same; myeloblastic is the same as AML. 13 Q. What about for erythroid leukemia? 14 A. M6 is a fairly -- M6 is relatively rare and 15 has a very high predisposition toward secondaries in 16 general and benzene in particular. 17 Q. So you would agree that benzene is a cause of 18 erythroid leukemias? 19 A. Yes. 20 Q. So it would be a correct characterization that 21 benzene is a known human leukemogen, although you would 22 say only for the acute myelogenous? 2:3 A. Yes, I am not going to quibble with you over 24 that leukemia. 25 Q. Would you agree that the literature has shown, PATTERSON REPORTING & VIDEO 172 1 based primarily on case reports, that benzene has also 2 been causally related to a variety of other 3 hematological neoplasms, including CML and non-Hodgkin's 4 lymphoma, as well as multiple myeloma, even though you 5 don't agree that the literature shows it's causative? 6 A. That's an oxymoron. You asked me would I 7 agree that primarily on the basis of case reports, that 8 benzene is causally related, et cetera. 9 Q. No, sir. 10 A. But not causally related? 11 Q. No, I'm not asking you that. I'm asking if 12 you agree that the literature has said that. 13 MR. BELFOUR: Excuse me, Herschel. May I 14 interpose an objection? That any literature, that 10 15 percent, that half? That they show multiple myelomas in 16 deficits, in excesses? They show other diseases in 17 deficits, in excesses? I think it's overly broad, but 18 if the doctor can answer it, I would certainly allow him 19 to. 20 Q. (BY MR. HOBSON) Well, the question is, Do you 21 agree that, based primarily on case reports, that 22 benzene has been causally related to a variety of other 23 hematological neoplasms, including chronic myelogenc us 24 leukemia, non-Hodgkin's lymphoma, as well as multiple 25 myeloma? I'm talking about what the literature has PATTERSON REPORTING & VIDEO 173 1 reported. 2 A. I have a problem with the way you worded that 3 question, because you're saying primarily on the basis 4 of case reports, causally related; and I don't believe a 5 case report constitutes evidence of causation. 6 If you are asking me, are there case reports 7 that hypothesize an association between benzene exposure 8 and the diseases you mentioned, there is probably a case 9 report out there to hypothesize everything that ever 10 was. So if you are saying, are there case reports that 11 propose benzene is associated with those diseases, yes. 12 Q. Do you agree that when we talk about benzene 13 causing acute myelogenous leukemia, and looking at the 14 literature historically, that that connection was 15 originally made through case reports and not 16 epidemiologically? 17 A. That is an overly -- I think that's an overly 18 simplistic characterization of the history of benzene 10 literature. By the time -- if we take from the 20 beginning of the century up through Hunter in the 1930s, 21 '40s, to say, let's say 1960, one can find on the order 22 of 250 cases of benzene poisoning in the literature. 23 Depending upon what you call a leukemia and 24 what you don't, there were between 5 and 12 case reports 25 of leukemia in that literature vis-a-vis a background of PATTERSON REPORTING & VIDEO 174 1 literally hundreds, between 200 and 300 cases of 2 individual bone marrow suppression, aplastic anemia, 3 pancytopenia, thrombopenia, lymphocytopenia, et cetera, 4 et cetera. That, in a few minds, constituted evidence 5 that benzene probably caused leukemia. Hunter was sure 6 of that in 1939. No one else was. 7 Between 1960 and 1970 we now see the 8 collections of Vigliani and Aksoy, which were case 9 reports; but they were collections, so they're not 10 quantitative, but they certainly constituted a much 11 richer set of concentrated examples of benzene toxicity 12 than we had anywhere else. And by the end of the 13 we now have -- probably 25 percent of the cases that 14 have been reported in the literature are leukemias. 15 That obviously provided a foundation for the 16 epidemiologic studies that occurred during the '70s and 17 which led to a general consensus with respect to the 18 relationship between bone marrow toxicity and 19 leukemogenesis. 20 That also was superimposed on our knowledge 21 that chemotherapeutic agents that cause bone marrow 22 suppression could lead to leukemogenesis, which wasn't 23 appreciated by hematologists in the '5os; and it wasn't 24 certainly appreciated in the 30s and '40s. 25 So there was no connection that bone marrow PATTERSON REPORTING & VIDEO 175 1 suppression could be associated with an increase in the 2 number of cells, which is what you see in a leukemia. 3 All of that came about between the late '60s and the 4 late '70s. 5 Case reports I don't think provided the 6 definitive evidence necessary to provide -- that led to 7 the realization of that relationship; certainly not the 8 early ones. There were very few, and if you added them 9 up, there were really few. 1u I do think that the collections that were 11 provided to us primarily by the Italians and the Turkish 12 studies provided reasonable evidence of some 13 relationship there before the retrospective 14 epidemiologic studies. 15 The fact of the matter is, if all we had was 16 the Pliofilm study and nothing else, it might not have 17 had the impact it did. It's much more of a milestone 18 than it is a great epidemiology study. It simply 19 punctuated the cumulative knowledge that we had with 20 respect to benzene and leukemogenesis at that point in 21 time. 22 Q. Have you ever seen a document from the 1950s 23 where a representative of the American Petroleum 24 Institute and a representative of the Chemical 25 Manufacturers Association, or in those days called the PATTERSON REPORTING & VIDEO 176 1 Manufacturer Chemists Association, signed off on the 2 proposition that the relationship between benzene and 3 cancer of the blood had been established? 4 A. I have seen something certainly similar to 5 that, and I don't think it's reliable. I think, based 6 upon the -- certainly not authoritative. Based upon the 7 information that was available to them at the time, that 8 certainly was not a general consensus on the part of the 9 medical community. 10 Q. What are you thinking about? 11 MR. BELFOUR: Excuse me, Herschel. It might 12 be better if you start with what were you asking about, 13 and maybe your question was objectionable, because it 14 was overly broad and vague. 15 Q. (BY MR. HOBSON) You said that there was a 16 document that came to mind, or something that came to 17 mind. I was just curious what it was that came to mind. 18 A. I do not remember the specifics, but I have 19 seen a letter written by a representative of, I believe 20 it was the American Petroleum Institute or a petroleum 21 company employee that said, in essence, what you are 22 saying. I have seen something like that. 23 Q. Would you agree that animal studies have 24 demonstrated that benzene is a carcinogen producing a 25 variety of solid tumors and lymphatic tumors in animals? PATTERSON REPORTING & VIDEO 177 1 A. In animal -- in various animal models, 2 specifically the mouse, benzene causes a variety of 3 different tumors -- also in the rat. Some in glands we 4 don't even have. 5 I would like to extend that to indicate that 6 even the solid-tissue tumors that are seen in rodents 7 have a very similar characteristic in that they all have 8 enzyme profiles that are typical of those found in human 9 bone marrow with respect to the ability to induce 10 peroxidated metabolism of the secondary metabolites. 11 Q. Is it correct that, among the early effects of 12 benzene on laboratory animals and in occupationally 13 exposed groups, there is a decrease in lymphocyte count? 14 A. It's been reported -- absolute decreases in 15 lymphocyte counts have been seen in one occupational 16 setting priluarily, and that was in the rotogravure 17 workers, pressmen from the Goldwater studies. 18 MR. BELFOUR: Excuse me, Herschel. I'm sorry, 19 Doctor. Did you ask "laboratory animals"? 20 THE DEPONENT: He mentioned both. 21 MR. BELFOUR: I'm sorry. I just didn't 22 understand the question. 23 A. Animals are much more susceptible to those 24 changes than humans are. Rodents are particularly 25 susceptible to those changes, and rodent lymphocytes PATTERSON REPORTING & VIDEO 178 1 tend to be very susceptible to conditions that are 2 associated with benzene. 3 Q. (BY MR. HOBSON) And for the laboratory 4 animals exposed to benzene, the effects would be on both 5 the B and the T lymphocytes? 6 A. There have been -- we presented some real 7 early data on B cells in rabbits, but the methodology, 8 quite frankly, was byzantine compared to what we have 9 today. 10 In the mouse model, I don't think it's clear 11 what we're talking about in terms of cytopenia. It is 12 clear with respect to the types of tumors that are seen 13 in the -- in most mouse strains, it's a T cell lymphoma. 14 High-dose benzene exposure -- and by "high dose," I mean 15 injections of very high amounts -- will cause 16 immunosuppression, including B cell. 17 Q. Would you agree that the basic cell type of 18 plasma cells, the B lymphocyte, is affected by benzene, 19 and it's probably in the circ -- it's probably the 20 circulating lymphocyte cell found to have been -- found 21 -- let me start over. I can't even see it here. 22 MR. BELFOUR: Herschel, is that one of my 23 questions I butchered? 24 MR. HOBSON: It could be. I have done a 25 better job of butchering it than you ever would, though. PATTERSON REPORTING & VIDEO 179 1 Q. (BY MR. HOBSON) Let's see if I can say it 2 accurately this time. 3 Would you agree that the B lymphocyte is 4 affected by benzene and is probably the circulating 5 lymphocyte cell found to have benzene-induced 6 cytogenetic abnormalities? 7 A. No, absolutely not. It's a compound question. 8 The answer was to the last part. At some concentrations I wouldn't be surprised 10 to see some effects on B lymphocytes, just like I 11 described qualitatively, maybe similar, to what is seen 12 in rodents with respect to B-cell effects, but those are 13 very high doses. 14 All the cytogenetic studies that I know that 15 have been done, both those that I think provide some 16 useful data and those that are flawed, virtually all the 17 human studies are looking at cytogenetic aberrations in 18 cells and culture after induction with a T-cell mitogen. 19 So some of the ones looking at peripheral lymphocytes in 20 humans have directly examined B cells. 21 Q. What about laboratory animals? 22 A. Very few have done that. I think LPS has 23 probably been used as a mitogen in one or two studies; 24 but then again, the induction of B-cell mitogen involves 25 very artificial concentrations of very toxic substances, PATTERSON REPORTING & VIDEO 180 1 so it's a very unreliable system for evaluation. That's 2 why in the human studies the vast majority of those 3 things are done with PHA, which is a T-cell mitogen. 4 It's stimulating proliferation under conditions that are 5 certainly not normal, but at least they aren't killing 6 off half the cells in the culture. 7 The use of endotoxins and lipopolysaccharides 8 to simulate B cells is a very artificial situation arid 9 it creates a lot of damage just in the process. 10 Q. So if someone said that probably the 11 circulating lymphocyte cell is a cell that is found to 12 have benzene-induced cytogenetic abnormalities, that 13 person would just be wrong? 14 A. Based on the techniques that have been used, 15 whether you're talking about Pacciano or Yardley-Jones, 16 to look at peripheral cytogenetic aberrations in workers 17 potentially exposed to benzene, and looking at T cell, 18 they're not looking at B cells, by definition, based on 19 the methodologies that were used. I'm simply saying 20 they can't say anything with respect to B cells, because 21 they're looking at T cells. 22 Q. Does radiation exposure cause multiple 23 myeloma? 24 A. You know, that has certainly been a presumed 25 paradigm for many years, but some of the more recent PATTERSON REPORTING & VIDEO 181 1 studies of survivors of the atomic bomb blast in Japan 2 suggest that multiple myeloma has not increased. I 3 would say that whereas I have assumed that radiation was 4 a potential cause in the past, I think there is some 5 question about that now. 6 Q. Is it enough of a question that you say 7 radiation is not more likely than not a cause of 8 multiple myeloma? 9 A. Up until recently I would have said it's more 10 likely than not. As of now, I am not sure. I certainly 11 don't see it as a major outcome of radiation therapy, 12 per se. Most of the data is concentrated in the atomic 13 bomb survivors, and that's come under question. 14 Q. Do you agree that the Chinese studies confirm 15 that DNA damage is discernible in human lymphocytes 16 following benzene exposure? 17 A. If you want to include, within the broader 18 context of DNA damage, cytogenetic aberrations and/or 19 evidence of deletions, that type of thing, yes. If you 20 want to refer to adductology, no. 21 Q. Is a human fetus more susceptible to 22 benzene-induced leukemia than a developed individual? 23 A. There is actually no basis upon which to 24 render an opinion on that. There is no evidence for it 25 or against it, that I am aware of. I know of no PATTERSON REPORTING & VIDEO 182 1 leukemia model that involves either -- exposure that 2 would be associated with benzene, either experimental or 3 any evidence in humans. 4 Q. Are you aware of anybody ever looking? 5 A. Certainly reproductive studies have looked at 6 multigenerational studies in rodents. I don't know of 7 any evidence associated with the lymphoma or the 8 leukemic model in those studies associated with exposure 9 in utero. 10 The predominant leukemia type in young 11 children is an acute lymphocytic leukemia. It has not 12 been associated with exposures to anything that I am 13 aware of, certainly not benzene, and it doesn't fit 14 paradigm that is consistent with secondary 15 leukemogenesis in adults. 16 Q. Did I understand earlier that benzene 17 metabolites are much more likely to affect rapidly 18 dividing cells? 19 A. In terms of cytogenetic or cytotoxic effects, 20 yes. 21 Q. Wouldn't the cells in a fetus be more rapidly 22 developing -- I'm sorry, rapidly dividing than you would 23 have in an infant that has been born? 24 A. On a theoretical basis, certainly they're more 25 rapidly dividing, but at the same time you have also got PATTERSON REPORTING & VIDEO 183 1 comparable differentiation-dependent protective 2 mechanisms at work during fetal development as well. 3 It's a very complex issue. 4 Certainly, replicating cells constitute a 5 greater -- a more susceptible target with respect to the 6 cytogenetic effects. Whether those cells, in fact, are 7 potential origin -- constitute a potential origin, or 8 cell of origin for leukemia, is another question. I 9 know of no paradigm or model that would provide me with 10 a basis for saying that would be anything but 11 speculation. 12 Q. And the same would be true: You know of no 13 paradigm or model that would disprove or show that there 14 is no effect on a fetus? Would that be true? 15 A. Well, no. As I said, there are reproductive 16 studies which don't demonstrate persistent hematological 17 problems in the offspring of animals exposed to benzene. 18 Q. But not humans? 19 A. I don't know anybody that has done a study of 20 in utero exposure in benzene. 21 Q. Or if not even in in utero, of blood or bone 22 marrow that comes from a fetus? 23 A. We're doing that. 24 Q. A human fetus? 25 A. We're doing core-blood studies and comparing PATTERSON REPORTING & VIDEO 184 1 core blood with adult bone marrow. Core blood 2 constitutes a source for fetal blood, fetal bone marrow. 3 Q. And you're exposing that core-blood component 4 to benzene metabolites? 5 A. Yes. We're looking for differences in stem, 6 cell regulation and increased susceptibility or 7 decreased susceptibility. 8 We see major differences in early stem cell 9 hierarchical differentiation between mice and humans 10 that certainly play a role in the susceptibility of the 11 mice to T-cell neoplasms, and we are investigating 12 whether those differences are species-related or 13 age-dependent. And some effects appear to be very 14 definitely species-related, and some effects are 15 age-dependent. 16 So far I don't have any evidence of 17 differences associated with exposure to benzene 18 metabolite. 19 Q. Have you published any of that? 20 A. No, not yet. 21 Q. Do you still believe that there is more 22 evidence that benzene causes multiple myeloma than 23 lymphoid leukemias? 24 A. I think you asked me once before, and it was 25 -- I think my answer to that question was related to PATTERSON REPORTING & VIDEO 185 1 degrees of implausibility. You're asking me, Is it less 2 plausible that benzene causes lymphoid leukemia than 3 multiple myeloma? 4 Q. I am asking you, Do you still believe that 5 there is more evidence that benzene causes multiple 6 myeloma than lymphoid leukemias? 7 MR. BELFOUR: Excuse me, Herschel. Do you 8 have a reference to where that came from? 9 MR. HOBSON: Page 55 of his prior deposition 10 taken in the Ellis case. 11 A. I'm sorry. Could you ask me the question once 12 again? 13 Q. (BY MR. HOBSON) Do you still believe there is 14 more evidence that benzene causes multiple myeloma than 15 lymphoid leukemia? 16 MR. BELFOUR: And that deposition in Ellis was 17 taken in approximately 1991 or so? 18 MR. HOBSON: Late '91, I think, without being 19 held to it. 20 A. I don't think, as we sit here today, that I 21 could distinguish between the levels of evidence for 22 each. They don't meet any test of consistency for 23 either, and I don't know of a plausible mechanism that 24 would explain a difference in the potential for either. 25 Q. (BY MR. HOBSON) So today you would say PATTERSON REPORTING & VIDEO 186 1 they're holding their own, being equally bad? 2 A. My -- I think the evidence against benzene 3 being associated with multiple myeloma has accrued 4 significantly in the last several years. And so what. 5 certainly was my opinion in Ellis is much stronger 6 today. And from that standpoint, I would say that there 7 is not more evidence that benzene causes multiple 8 myeloma than lymphocytic leukemia. 9 Q. Have you seen the most recent epidemiological 10 studies for butadiene from coal and diesel? 11 A. Yes, I have. 12 Q. Do you believe that butadiene is a human 13 carcinogen? 14 A. I believe that that study raises some very 15 important questions; but based upon my evaluation of 16 that study, in the form I have seen it in, which is 17 probably not going to be its final form, it raises a 18 number of biological plausibility issues with respect to 19 their classification of leukemias. 20 I think it is probably -- it is evident from 21 those studies that exposure to butadiene and/or 22 something in the polymerization process is influencing 23 the development of leukemias. The question is, Which 24 ones and how? And right now, I think there are some 25 major tests of biological plausibility that are going to PATTERSON REPORTING & VIDEO 187 1 have to be made before I have any idea what that means. 2 Q. What form was the study in when you saw it? 3 A. I peer-reviewed an original -- the original 4 rough draft, and it has undergone review, and I have no 5 idea what the final is going to look like. 6 There are a number of problems that were 7 raised, that I certainly found, with respect to the 8 characterization and analysis, and I have no idea what 9 the final form is going to look like. 10 Q. Your view of the conclusions, then, from that 11 study, are different than the authors'? 12 A. Not necessarily. I think it's dependent; I 13 think it raises some very important issues with respect 14 to what is seen. I think there are questions with 15 respect to the exposure -- the relationship between 16 exposure and the -- and leukemia, in general, and there 17 are questions with respect to the classifications of the 18 leukemias. I think many of those concerns of mine are 19 shared by authors. I don't know how they are going to 20 wrestle with those in terms of publication. 21 MR. HOBSON: I would like to attach all the 22 stuff you have brought today, and I am done. 23 MR. BELFOUR: You want to read and sign, I 24 believe. 25 THE DEPONENT: Definitely. PATTERSON REPORTING & VIDEO 188 1 (The deposition concluded at 4:10 p.m.) 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 PATTERSON REPORTING & VIDEO