Document 8RbJeo791Yk5gLEJ41Kd340Xm

QRAC - 3- 3I Method 1: based upon fitting bt c (nonincidental analysis) to a 1922-1946 cohort of the Selikcff, et. &1. data. A reasonably simple way to estimate the median life (ML) risk to median survival age 77 (in 1979) for humans exposed 2 yrs. to talcum powder during infancy is given by the product of the following terms: (a) (77 yrs. since first exposure for infants/37 yrs. since first 4-3 1 exposure for 1922-46 cohort as of 1978 ) * * 9.70. (b) (2 yr. infant exposure duration/34 yrs. approx, worker exposure duration for 1922-1946 worker cohort) ** .059. **6 (c) (infant/worker) yearly exposure ratio E 0,3 s 10 (d) 1922-1946 cohort cumulative mesothelioma response of 3.75% (180 mesotheliomas/4,.800 cohort members). This product yields a median life risk of --8 = 0.64 x 10 Method 2: based upon b(t-10) 2.`1 (delayed observation or time lagged nonincidental analysis). Kote that to estimate real mesothelioma incidence (time of mesothelioma induction - the last stage of the multistage cancer process) at age x, the worker must be assumed to have been autopsied or surgically inspected at some average age, say x+10. Thus, assumzning the worker stops exposure 3 years before death, the component relative and absolute risk factors for incidence at age 77 now are the following: (a) ((87 yrs.-10 yrs.)/(37 yrs.-10 yrs.)^** = 9.03. (b) (2 yr. infant exposure duraticn/(37-10) yr. worker exposure duration) = .074. (c) (infant /worker) exposure rate ratio 15 0,3 x 10 (d) 3.75% mesothelioma response in 1922-1946 cohort Thus - 0.75 x 10"8 . QRAC - 4- 1 64 Method 3: based upon bt c (prevalence or incidental analysis)s The relative and absolute risk product factors ares (a) (77 yrSc since first exposure for infant/35 yrs. gince first exposure for the 2,271 deaths to 1 9 7 6 ) - 3.64. (b) (2 yr. infant exposure/34 yr. ave. worker exposure duration for 2,271 deaths to 1976) = .059. -6 (c) (infant/worker) exposure rate ratio ~ 0.3 x 10 (d) 7.71 mesothelioma cumulative prevalence to 1976 (175 mesotheliomas/2,271 deaths). Thus *= 0.50 x 10"8. 3.1 Method 4: based upon bt (nonincidental analysis) and a first stage effect in a generalized multistage process. k-1 Ve assume that bt fits the time-response data of a nonincidental tumor and is consistent with a first-stage-only effect in a generalized multistage process (with K stages), where biological time t starts at age of first exposure and continues until death [9]. Although this is not precisely true for the 1922-46 asbestos worker cohort, it appears to be approximately true. Moreover the time lag from cessation of exposure to end of followup (1976 or 1978+) is assumed to be small compared to total duration of exposure (i.e., exposure duration is a large fraction of time since first exposure). However, the exposure duration for infants is very small compared to median lifespan. Thus, while we fit k-1 worker yearly incidence data to bt we should extrapolate yearly K-1 incidence (I) for exposed infants using the expression I b(t (t-d) K-1 ) for a K stage multistage process with duration of exposure d and time since first exposure t [9]. QRAC - 5- Now K" 1 3 . 1 from Fig. 1 and b can be written as the product of a constant and f where f is the time adjusted yearly dose of asbestos fibers in ml-yrs. is a constant dependent upon the type and dimen sions of the asbestos. Since f * 3.43 f/ml-yr. (15 ave. /mi is workplace (1922-1946) x 8 hrs./24 hrs. x 5 days/7 days x 50 vks/52 wks) 3.1 for the Selikoff study,, K can be computed from, the plot of I B I f i In Wl =A in Fig. 1. At t K 20 yrs, I *= 5.6 x 10 , implying that the InK * In (5.6 x I0~^)-In(3.43)-3.1 (In 20) -7.49-1.23-9.29-- 18.01. Thus K = 1.51 x 10--8 (same as Peto obtains). Continuing, I * m K f(tlv *-(t-d)K *) * K ftK ^(l-(l-d/t)^ *) which roughly mm K ftK-- 1(d/t)(K-l) for d much less than t (using Taylor expansions), m K-2 Thus yearly incidence is approximately I*K fd(K-l)t . Integrating m (without correcting for decreasing survival) over a total of T years yields a cumulative incidence of about K-l = K^fdT . If d = 2 yrs. infant exposure duration, T - 11 yrs., K-l - 3.1, f *= 3.43 f/ml-yr. for worker x 0.3 x 10 ^ (infant/worker exposure ratio) = 1.03 x 10 f/tnl-yr. , and K = 1.51 x 10^, then 1 = 2.2x10 mc However, this figure assumes no mortality from competing causes of death and does not even adjust for the effect of previous mesothelioma related deaths. Factoring in a standard population age-specific mortal ity or corresponding survival function into the above integral would o yield a median life risk of about 75% of 2.2x10 or o c 1.6x10 This correction for survival can vary depending upon the limits of integration and what functional forms are under the integral, but for median life risk estimates the correction ranges from 1.0 dorm to .5 at worst. e also note that integrating I out to 100 yrs. of life with QRAC - 6- respect to a standard mortality curve should yield approximately the same risk as cumulative incidence to median age 77 yrs. without any mortality adjustments. These approximately cancelling effects of two mathematical refinements may support the utility of using the median lifespan in simple calculations. Comments on the k Mesothelioma extrapolation methods; First and most importantly, it should be noted that the first 3 methods yield virtually identical median lifespan risks for babies exposed to talc for 2 years (.5-.75x10 ). Thus many of the debates over the "correct model" appear somewhat superfluous. In particular heated debates over whether mesothelioma rates follow given high or low powers of time appear to be superfluous since the power of time is compensatingly related to other poorly defined and difficult to measure conceptual model parameters (e.g., tumor stage initiation and consequent time lag to clinical detection or death, and context of tumor observation (incidental or nonincidental)). Furthermore, small perturbations of the rough estimates of worker exposure or the power of time (K) have only a small effect on the overall.risk. All the above models appear to be reasonable summary descriptors of the observable data and result in simple extrapolatory tools for the given problem of inferring median lifetime risk from Infant exposure. One can always make method 4 computationally more difficult if one avoids use of the approximations. A second observation is that the rough mutual agreement of the results of the 4 extrapolation methods does not necessarily imply that QRAC - 7- the obtained excess median life risk Is accurate even if the Infant and worker exposure were to the same type and dimension of asbestos fiber For example* none of the four models take into account the possibility that accumulated dose rather than yearly dose rate might more accurately reflect the biological burden of asbestos due* for example* to its ability to reside in vivo in the lung* pleural or peritoneal lining for years without being excreted (although, encystment may be possible). Note also that we did not define dose on a mg/kg body weight basis. Although* we prefer such a definition for routine compounds that are ingested and metabolized* we strongly suspect that routine approach to be inappropriate for asbestos. In addition* all 4 methods assume linearity In response vs. dose at all dose levels. However, we have virtually no reliable dose response data from any of the epidemiological studies. Furthermore, some investigators have suggested that the nonconstant accumulated asbestos dose may be as conceptually consistent with a late stage multistage carcinogenic process as the more usually defined yearly asbestos dose rate appears to be consistent with a first stage Armitage-Doll multistage process [9]. Although the theory and computa tions are more complicated for nonconstant exposures* it does appear that median life risks from infant exposure to asbestos affecting only a late stage in the carcinogenic process will generally result in much smaller risks than those calculated above for a first-stage-only effect in the carcinogenic process. QKAC - 8- Our third observation which we have just hinted at is that method 4 above (the firet-etage-only effect in a multistage model) may be just another way of Implementing method I,, but just slightly more computationally difficult and having a slightly higher risk* partially because it substitutes a theoretical risk integration against the current (1979) U.S. population's standard survival function for the implicitly observ able but poorer asbestos worker's cumulative survival of an earlier era in a more toxic environment. For example, the method 4 risk is about 2.6 times greater than the average risk of methods 1-3. There are probably other reasons for this 2.6 fold Increase in risk over methods 1-3. However, since even partial intervention of asbestos fibers at later stages of the carcinogenic process in the Araitage-Doll multistage model imply lower overall risks, we prefer the simpler methods 1-3 at this moment to the more complicated multistage models whose proper application with respect to the stage or stages affected is still very much in doubt. In general, we do not put a lot of faith in mechanical use of sophisticated but unverifiable models, but we will occasionally refer to them as in method 4 where we can suggest implicit and perhaps elucidative connections to apparently more humble and simpler procedures. QRAC All four mathematical methods of modelling the nonlinear mesothelioma response data from the Selikoff study indicate a lifetime added human risk to infants exposed 2 years to talc powdering of at most about 10"8 risk, and quite probably far less risk, if for example, asbestos intervenes in the carcinogenic process at a later stage than the first stage which was assumed in method 4 for the Armitage-Boll multistage process. Robert N . Brown QRAC REFERENCES - 10 - lc L. Taylor* "Request for CAC Evaluation of the Hazard of Asbestos Contamination of Cosmetic Talc*" FDA memo* Rov. 15* 1984. 2. B.W. Cramer* HD* W.R. Welch* R.E. Scully, C.A. Wojciechcweki* "Ovarian Cancer and Talc - A Case Control Study," Cancer* July 15 1982. 3. L. Tollefson* "Revie*? of reports of increased risk of ovarian cancer from talc use," FDA memo* Jan. 30 1985. 4. F. Hartge* R. Hoover* L. Lesher* L. McGowan, "Talc and Ovarian Cancer*" JAMA* Oct. 14, 1983. 5. L. Tollefson and F. Cordle, "Review of an assessment concerning asbestos contamination of cosmetic talc," FDA memo, Dec. 17 1984. 6. Chronic Hazard Advisory Panel cm Asbestos, Report to the D.S. Consumer Product Safety Commission, July, 1983. 7. Selikoff, I.J., Hammond, E.C.* Seidman, H., Mortality Experience of Insulation Workers in the United States and Canada, 1943-1976* Annals of the H.Y. Academy of Sciences, 1979* 91-116. 8. Feto, J., Seidman, H . Selikoff, X.J.* Mesothelioma Mortality In Asbestos Workers: Implications for Models of Carcinogenesis and Risk Assessment, Br. Jour, of Cancer (1982) 45, 124-135. 9. Day, R.E., Brown, C.C., Multistage Models and Primary Prevention of Cancer, JNCI, 64, 977-989 (1980). 10. Eiermann, Heinz J., "Health Research Group Inquiry on Talc Safety," FDA memo, Aug. 28, 1978. 11. Wenninger, John A., "Denial of Petition for 'Labelling of Warning of the Hazardous Effects Produced by Asbestos in Cosmetics Talc' from Philippe Douillet," FDA memo, July 11, 1984. 12. Garfinkel, L . , "Time Trends in Lung Cancer Mortality Among Konsmokers and a Rote on Passive Smoking," JNCI, _66_, 1061-1066. 13. Vital Statistics of the United States, Mortality, Part A* 1979, published by the U.S. Dept, of Health and Human Services. 14. U.S. Dept, of Health and Human Services, PHS, "The Health Consequences of Smoking: Cardiovascular Disease," a report of the Surgeon General* 1983. 3 A WS 2 O Al5 CS ^ ai ca SJ E^i '*>) a R <7 / X j 014 o es <& o ii <"s( tA ., . bk ^JL " j*> 'ti'O ' Sj 4 ,"il"5""' I P *5^ '. ^ / V i&s s V 6 S 9 S P C I `v i | .o ssa ^ UUM|% V sq<Sqst> 5^ a @A* *I) ^ ; s q w $ s ^o2>v^okvi0^ S^04 * y 5""to *MTj V|4^5 p\/ ~S i^SiTpS nw/ y t tQVf* JOL 1 1 TS86 BdlllpfM. Doulllet Ohe Bslyoke U i s Stony Brook, New York 11790 i Dm p Mr. Douillet: ''Re: ibstet Fb* 83F-0404 Ihig responds Qo your fevcsahsr ft, 1983, p^cifcioei requesting time cosmetic tftle be labeled with an asbestos warning information on afibestoc particle sis, -id the proportion of talc impurities in the product Yea assert that, because the mining of talc almoet invariably includes the mining of asbestos m well, tale m y contain significant aCRZit of asbestos particles that preacant an inhalation hazard to hussne* Also, toil cite reference* to substantiate th&t significant amounts of asibestoe have been found In coraoercial talc samples, that asbestos inhalation Is hazardous to hunsns, and that aebeetoe contaminants In tele ^11 produce toxicological response when Inhaled Ft& recogni that ssbeirtos inhalation over ttended periods is hazardous to hts&sns. Th agency 1 algo aware tb^t cosmetic talc produced in the 1960s and early 1970s did contain asbestiform mineral a* However, your petition has not persuaded us tbf*t the cosmetic talc that is presently being produced contains slgjiificmt amounts of asbestifor minerals. Ourir^ the early 1970s, FD& bec^ concerned abmit the possibility that cosmetic talc did contain significant a^ijounts of this material. The agency received severe! reports hour such cont*ninatloru However, at that tin, the analytical procedures for determining asbestos in tlc were not fully developed, and most of the analytical work wag conducted without scientific agreement a to vhich methods were well-suited for the Identification of asbestiforcu minerals in talc. Consequently, FB& considered ell analytical results Co be of questionable reliability. This assessment proved to bs correet because many questions were subsequently raised about results reported to the literature in the early 1970s (see enclosed copy of National Bureau of Standards Special Publication 506 entitled "Mlidentification of Asbestos in M e " ) . Because of the questionable nature of the analytical results, the agency was not able to assess reliably the levels of ssbestiform minerals in cosmetic talc then in the marketplace. Mf. PhiXlippe feudliet - F>$ 2 Under foecs circuEStaneee, FOX decided fh&t the most appropriate set!ass that it could take to protect the public health wwild be to make the reports public and to request acsletAnce from the effected inchtry in developing acceptable analytical procedure.. This approach apparently h&L led to aoftsiderable Improvement in the quality of this talc. After FDA tosh the action ,, many cos&stie ks^ufaoturere bc^an to analyse their talc for asbsstifore mir^raia m psrf of their cp&Iity control progrmig, and tele supplier* began to sell higher parity" talcL m the cosmetic industry* By 1976, ashestoe analytical methodology us* sufficiently developed that the Cosmetic, Toiletry7* and Fragrance Association (CXFA) could i&awi a spscification (copy meloz&d) for cosmetic telec Tbit. specification required that such tale be free of fibrour awphibole (eg( asbsstot in th<& f o m of asbestifor* tr&.sllte) u s Ir q a CTFA method of analysis that is capable of detecting 0S percent of aasphiboie asbestos Ihi specification contributed to the improvement of cosmetic talc quality. In addition, FBI surveillgrtc activities, that sre eoaiustGd in the latter portion of the 1970s lowed .that the quality of cosmetic talc had significantly improved, and that w e n vitan asbestos was present, the levels w w o low that no health hazard existed Our seiartists recently reviewed data fress these surveillance act!'vitiee a?id concluded that the risk free a worst-case estte&to of eqxmana to asbestos fret, cosmetic tale would be less than the risk from isnrirmmentsl background levels of vprmire to asbestos (non-occupational ^posure) over a lifetime. Consequently, ^ find that there 1s to baris at this time for the agency to conclude that there is a health hazard attributable to asbestos in coexetle talc. Without evidence of such & h^g&rd, the cy concludes that there 1s to nerd to require a warning label on cosmetic talc, FBI should Iso point out th*t, in reviewing y^i\rr petition, we found several problems with the information on vhich you relied. The publication ^Afiheetiform I&spirifcies to Gto^vcrcial Talcus Powders," vhicb you cite Jn your petition, appears; to contain a m w b e r of signlfleant errors that lead us to question the accuracy of the findings that were reported. For your information, we encloc^i a copy of s June 8,, 1973, rebuttal of this publication that vs written by the Chief Mlnerologist of the Golorado School of Minos Heeearch Institute in Golden, Colorado. Also, your petition* 1978 book reforms to the Pfc. Sinai School of Msdlciras finding Is too old to reflect preosmfc cortasination levels. Further, we are rat convinced that the Me. Sinai findings pertained to cosmetic talc, lour reference states that common coHwerci&l talcs were analysed, btrt: it not specify whether these commercial talcs were industrial grade or cosmetic talc. For 11 of ie* reasons, your petition is rerded. 'Sii* donisi is %tbou preludici ts the future filing of & petition an t M s matter, coapaaled by 11 rele^&rst d&tc in support of the petition. Sincerely yours, ^(aJOL^X-^.-- ^ Acting Afeaolfit Gbeeissioncr for Regulatory Affaire cc: HFC-1 HFC-200 (#G-86-182) HFC-220 (Rogers/file) HFF-1 HFF-100 HFF-152 HFF-300 HFF-302 HFF-310 , HFF-440 GCF-1 (Horton/Derfler) HFA-224 HFA-305 Prepared:JRTaylor:5/1 5/86 Initialled: JRTaylor:5/15/86, 6/5/86 EJCampbell:5/1 5/86, 6/5/86 HJEiermann:5/16/86, 6/9/86 JAWenninger:5/19/86 WGFlamn:5/29/86, 6/9/86 LRLake:5/29/86, 6/12/86 RJLenahan:5/29/86, 6/10/86 LBBrock: 6/10/86 HHGill:6/12/86 F/T:JRTaylor:sag:6/4/86 Concurred :EBrisson: 6/27/86 Retype:RLSpencer :cdk: 6/27/86:disk. 26 (#132) Revised :FSDerfler:7/3/86 Retype:RLSpencer:cdk:7/7/86 Concurred:R3erfler:7/8/86 Revised:Concurred:LHorton:7:9/86 F/T:RLSpencer :bka:7/10/86 DEPARTMENT OP HEALTH AND HU M A N SERVICES Acting Directory Office of Compliance Center for Food Safety and Applied Nutrition Denial of Petition to Require Warning Statements on Cosmetic Talc Associate Commissioner for Regulatory Affairs (HFC-1) Through: Directory Center for Food Safety and Applied Nutrition OBJECTIVE To issue a letter of denial (Tab A) for a petition from Hr. Philippe Douillet that requests certain mandatory labelinq on cosmetic talcs to warn consumers of asbestos hazards associated with such products. FACTS Durinci the early 1970's FDA became concerned that significant amounts of asbestiform minerals may be present in cosmetic talc, The agency had received reports of cosmetic talc being contaminated with high levels of asbestiform minerals. However, at that time the analytical procedures for determining asbestos in talc were not fully developed and most of the analytical work was conducted without scientific agreement as to which methods were suited for the identification of asbestiform minerals in talc. FDA then considered all analytical results to be of questionable reliability. As a result, the agency could not assess either the accuracy of the reported results or the extent of the presence of asbestiform minerals in cosmetic talc then in the marketplace. FDA made the reports public and requested assistance from the affected industry in developing acceptable analytical procedures. Subsequently, many cosmetic manufacturers beoan to analyze talc for asbestiform minerals as part of their quality control programs and talc suppliers began to sell higher purity talcs to the cosmetics industry. By 1976 asbestos analytical methodology was sufficiently developed that the Cosmetic, Toiletry and Fraorance Association (CTFA) could issue a specification for cosmetic talc that required such talc to be free of fibrous amphibole (e.g. , asbestos in the form of asbestiform tremolite) by a CTFA method of analysis which was capable of detecting 05 of amphibale asbestos. This specification contributed to the continued improvement of cosmetic talc quality.  * T ^ v "^>T*" A . ^ ,.v . ..-^ ` ^ ,>".*; :;*:v, '?y^- r" - : * ".'-*" - ',, '-*' *:***..'. ' V*T- uS*'*"' V f-`* -V- -jrf^r "V- Ir*- .'-( 7 7 ^ V _ " *c~1-.z r f* * s ! ' \ i . ''. . - *' . " - ry *> . " ^ " ` ' "*' - t lX^ ^ '? v ' 5 V 'V : ''- - f V<c*^'"% * rv :_' V - " - ' ';,;5si-;.v* D E P A R T M E N T OF HEALTH A N D H U M A N SERVICES TT"-7V^'-rt>t'tf< FDA surveillance activities that were conducted In the latter X t W ' portion of the 19 70 *s showed that the quality of cosmetic talchi' had .signi flean tly iiaproved, and that even when asbestos was/-t;K',r present,, the levels were so low that no health hazard existed.':> Our scientists recently reviewed data from these surveillance activities end concluded that the risk from a worst-case estimate from cosmetic exposure would be less than the risk from v environmental background levels of asbestos (nonoccupational exposure) over a lifetime (Tab B). On November 8 ,, 19S3y Hr. Philippe Douillet submitted s petition (Tab C-Docket No. 83P-0404) requesting that cosmetic talc be labeled with an asbestos warning statement and information on asbestos particle size end the proportion of talc impurities in the product. The petition contends that cosmetic tele is contaminated with asbestos and that this contamination presents an inhalation hazard. The petition's substantiation of this contention consisted only of twenty references. None of the references was recent enough to indicate that cosmetic talc contains any asbestiform minerals at this time. DISCUSSION Because Hr. Douillet's petition contains no substantiation of a health hazard attributable to asbestos in cosmetic talc at this time, we have drafted a letter of denial (Tab A) for your aDproval. The draft letter of denial explains why we do not believe asbestos contamination problems exist at this time. We did not include the recent scientific evaluation of ssbestos risk as an enclosure for the draft letter but we plan to forward the evaluation to the Dockets Management Branch upon issuance of the letter of denial. RECOMMENDATION It is recommended that-the draft letter of denial concerning asbestos warning statements on cosmetic talc be approved and issued. L ,, Robert Lake At tachraents Tab A - Draft Letter of Denial Tab B - Scientific Evaluation of Asbestos Risk Tab C - Philippe Douillet's November 8 * 1983 Petition OFFSCE SURNAM E ' DATE O FFIC E SURNAM E DATE O FFIC E SURNAM E DATE ' v ;r - , - * -*<.^ ... , ,. __ v ` * V i ' - ' ' I* - ,, T f -- .- ;'< # . . > , * ' >>?.- r - * -V '. .* i U > \ u 4c:'L.v V.:'^-- ;.'L .......v ' *'" * ' . n ^ X ' h ^ r Z 'dD E P A R T M E N T O F HEALTH A K D K M A K SERVICES .; './i.f.9 ,">.Associate Commissioner "for vRegtriatory~Af fsi "h, '<' .: : :*'iVsESUswr - :'. - ............................. - ->* - '' ' .-!**j v S i i ' . " *' v '" '- ' ' ,. . . ' " V t ' . r - . 't ' . * '" .' _ ", ;'' DECISION The letter of denial of Philippe Douilletcs petition concerning asbestos warning statements on cosmetic talc is: Approved Disapproved Date Prepared by :HFF-312 :TAYLOR s4B5-0179 HFF-100 H F F - 1 52 HFA-30 5 HFF-300 r/f HFF-302 KFF-312 HFF-312 r/f HFF-440 HFF-310 r/f HFC-1 r/f Draft:3RTaylar:saq:5/15/86 Initialled:3RTaylor:5/15/fl6 F0Car(pbell:5/15/8 H3Fiermann:5/16/36 3AWenninoer:5/19/86 WCF1 arnm :5/29/86 LRLske :5/29/86 R.3 Lenahan:5/30/86 F/T:0RT8ylor:sag:6/5/86 Phillippe Douillet 1 Holyoke Lane Stony Brook* NY 11790 Dear Hr . Douillet: This responds to your November 8* 1983 petition requesting that cosmetic talc be labeled with an asbestos warning statement* information on asbestos particle size* and the proportion of talc impurities in the product* You assert that because the mining of tslc almost invariably includes the mining of asbestos as well* cosmetic talc may contain significant amounts of asbestos particles that present an inhalation hazard to humans. Also* you cite references to substantiate that significant amounts of asbestos have been found in commercial talc samples* that asbestos inhalation is hazardous to humans * and that asbestos contaminants in talc will produce toxicological responses when inhaled. FDA recognizes that asbestos inhalation over extended periods is hazardous to humans. Also* the agency is aware that some cosmetic talc produced in the 1960fs and early 1970's did contain asbestiform minerals. However your petition has not persuaded us that more recently produced cc netic talc contains any significant amounts of asbestiform minerals. During the early 1970cs FDA became concerned that significant amounts of ssbestiform minerals may be present in cosmetic talc. The agency had received reports of cosmetic talc being contaminated with high levels of asbestiform minerals. However, at that time the analytical procedures for determining asbestos in talc were not fully developed and most of the analytical work was conducted without scientific agreement as to which methods were suited for the identification of asbestiform minerals in talc. FDA then considered all analytical results to be of questionable reliability. This assessment was Ister proved correct because many questions were subsequently raised about results reported in the literature (see enclosed copy of National Bureau of Standard Special Publication 506 entitled Misidentification of Asbestos in Talc). As a result, the agency could not assess either the accuracy of the reported results or the extent of the presence of asbestiform minerals in cosmetic talc then in the marketplace. Under these circumstances, FDA decided that the most appropriate course of action to protect the public health was to make the reports public and to request assistance from the affected industry in developing acceptable analytical procedures. This approach appeared to lead to considerable improvement in the quality of this talc. Many cosmetic manufacturers began to analyze talc for asbestiform minerals as part of their quality control programs and talc suppliers began to sell higher purity talcs to the cosmetic industry. By 1976 asbestos analytical methodology was sufficiently developed that the Cosmetic, Toiletry and Fragrance Association (CTFA) could issue e specification (copy enclosed) for cosmetic talc that required such talc to be free of fibrous emphibole (e.g., asbestos in the form of esbestiform tremoiite) by a CTFA method of analysis Which was capable of detecting 0.5^ of amphibole asbestos This specification contributed to the continued improvement of cosmetic talc quality. In addition* FDA surveillance activities that were conducted in the latter portion of the 1970's showed that the quality of cosmetic talc had significantly improved* and that even when asbestos was present, the levels were so low that no health hazard existed. Our scientists recently reviewed data from these surveillance activities and concluded that the risk from a worst-cas^/ estimate from cosmetic exposure would be less than the risk from environmental background levels of asbestos (nonoccupational exposure) over a lifetime. Accordingly, we believe it would be inappropriate to consider implementing your requested labeling requirements for cosmetic talc without adequate substantiation of a health hazard attributable to asbestos in cosmetic talc at the present time. Your petition contains no such substantiation. The publication "Asbestiform Impurities In Commercial Talcum Powders" appears to contain a number of significant errors that lead us to question the accuracy of the findings that were reported For your information, we have enclosed a copy of a June 8, 1973 rebuttal of this publication that was written by the Chief Mineralogist of the Coloredo School of Hines Research Institute in Golden Colorado. Also* your petition's 1978 book reference to the M t . Sinei School of Kedicine findings is too old to reflect present contamination levels. Further* we ere not convinced that the Ht . Sinei findings pertained to cosmetic tele. Your reference states that common commercial talcs were analyzed* but it does not specify whether these commercial talcs were industrial grade or cosmetic talc. In view of these facts* your petition is denied. This denial is without prejudice to the future filing of a petition in this matter* accompanied by all relevant data in support of the petition. Sincerely yours * Joseph P. Hile Associate Commissioner for Regulatory Affairs cc : HFF-1 HFF-100 HFF-152 HFA-305 HFF-300 r/f HF F -440 HFF-310 r/f (-- HFC-1 r/f Draft:JRTayl or i3/1 Initialled bysJRTa, EJCambell :5/15/86 HJEiermann:5/16/86 JAWenninger:5/19/86 WGFlamm;5/29/86 L R L a k e :5/29/86 RjLenahan5/30/86 F/T ?JRTaylorisag;6/4/86 Y^ Hon. Mark Novitch Acting Commissioner Food and Drug Administration 5600 Fishers Lane Rockville, MD 20A60 Petition for labeling of warning of the hazardous effects produced by asbestos in cosmetic talc. The purpose of this petition is to request a labeling of warning as well as a detailed list of components of the commercial cosmetic talcsc Because of its geological formation cosmetic talc may contain significant amounts of asbestos particles producing hazardous effects by its continuous use. Since it is a cosmetic article its production and commercialization is regulated by the Food, Drug and Cosmetic Act. I am a graduate student of Marine Environmental Sciences and I am deeply concerned by the toxic effects produced by the constant and periodic use of talc by the public, specially children. Because commercial talc deposits consist of natural admixtures of mineral, a number of mineralogically different materials have been used as commercial talc. Asbestos is a generic term for a variety of natural minerals which have the ability to be separated in filaments (1 Since the mining of talc rock almost invariably includes the mining of asbestos as well, the asbestos contaminant is carried over into the consumer product and thus introduces the risk of asbestos disease (2). Knowledge of diseases associeted with the use of asbestos apparen tly dates back 2000 years and detailed medical reports with the clas sification of asbestos as a harmful substance began about 1900 (3)* It has been widely proven that the inhalation of asbestos is the direct cause of-hemolysis (^,56) and diseases such as asbestosis, bronchial cancer, pleural mesothelioma and peritoneal mesothelioma (3 ^ ,5 *7 3,9 {.1' Asbestosis is a diffuse pulmonary fibrosis initiated by the inha lation of asbestos particles. The particles above 10 microns are fil tered in the passage through the nose and trachea to the lungs, but the inhaled air reaching the respiratory bronchioles and alveoli contains thesmall-sized particles (1-5 microns) that would tend to deposit (-3). Tirntrell found that the fibres of 3 microns diameter are the thickness that are likely to be deposited in the alveolar regions (11), and the fibres with a smaller diameter (less than 3 microns) -especially those below 1 micron- are carcinogenic if introduced into the pleural or peritoneal cavity (9 ), Neither the "cleanning" (mucociliatory escalator) nor the inmunological system (macrophages) are able to expel or destroy these particles (3*4,12). Then fibroblasts in an irreversible process produce the collagen, which forms the characteristic fibrosis of the asbestosis (5);these can lead torespiratory disability (5) and death may result from pul monary hypertension and cardiac failure (3). About 50$ of the asbestosis patients may develop carcinoma, such as mesothelioma, a diffusive cancer that rapidly spreads over the surface of the lungs, abdominal organs and heart (3)* The exposure to talc dust has been shown to produce lung scarring, termed talcosis, and asbestos bodies are observed in lung tissues of individuals who die of talcosis (14). The "fibrous" talc appear to be more pathogenic than "platy" talc, so this disease is due to asbestos rather than talc (14,15). Timbrell found that due to their fibrous shape, asbestos particles remain airborne longer and show less ten dency to sediment than granular microparticles of equivalent weight (16); this physical factor will increase the asbestos exposure of a person in an environment where talc has been recently used. The asbestos diseases are due to occupational and non-occupational exposures (8,17); only 40.8$ of the mesotheliomas found in patients of London Hospital from 19 17 to 1964 were due to occupational expo sures (8). Every body is affected by this air pollution, as has been demonstrated in lung studies done in France (18) and New York City (19 Permissible occupational exposure limits exist in several countries, in the USA, the Occupational Safety and Health Administration (OSHA) proposed- in 19 7 5 a non-occupational exposure limit of 5 fibers/ml for a period of 1 5 minutes (9 ). Several mineralogical analyses have been made on commercial talc, and all the samples contained asbestosform mineral impurities (5*20). Snider'et al in 1972 found that in eighteen commercial talcum powders, the asbestos impurities varied in amounts from 4 to 46$ meanwile the labels listed no impurities (20). In 54 common talcs analysed at Mount Sinai Hospital, the asbestos content ranged as high as 8 7% (5). The purity 'of any commercially available talc in the U.S. is related to both the nature of the original talc deposit and the extent to which the rock is upgraded to eliminate contaminant minerals. As the percentage of asbestos impurities is not related to price (20), this labeling will probably force the producers to control their talc com position to maintain their revenues. There have been many lawsuits relating to the health aspects of asbestos, and the causes of action in product liability lawsuits generally involve the operative allegations: -Failure to warn: thus the consumer was unaware of the danger. -Failure to test: by the producers to test their products to pro perly ascertain its hazard, risk and dangers. -Failure to remove: by the producers to stop selling, or remedy (make safe) the asbestos product (3) Warning of the health risks/ of exposure to asbestos has been recommended as a measure to be taken within- the European Communities in 1977, and stated: "Asbestos containing products should be clearly labelled" (9) The Food, Drug and Cosmetic Act is clear about these facts: "If an article is alleged to be misbranded because the labe ling is misleading, then in determining whether the labeling is misleading there shall be taken into account (among other, things)...the extent to which the labeling fails to reveal facts material in light of such representations or material with respect to consequences which may result from the use prescribed in the labeling thereof or under the conditions of use as are customary or usual". (21 U3C3 i 3^1*n) "A cosmetic shall be deemed to be misbranded- (b)If in package form unless it bears a label containing (2) an accurate state ment of the quantity of the contents in term of weight, measu re, or numerical count." (21 USC3 362) "A cosmetic shall be deemed to be adulterated- (a)If it bears or contains any poisonous or deleterious substance which may .render it injurious to users under the conditions of use pres cribed in the labeling thereof, or under such conditions of use as are customary or usual." (21 U3C3 361) Taking into account the widespread use of cosmetic talc, which starts at birth (mostly used in the first years of life) and continues in a great number of people as a periodic exposure throughout their lifespan, I address this petition to request the obligatory establishment of lables of quality (asbestos particle' size) quantity (proportion of impurities) of components as 'well as a label of 'warning of the hazardous effects produced by asbestos v/ith the continuous use of cosmetic talc. Respectfully submitted* November 8 ,19 8 3 Philippe Douillet 1 Holyoke Lane Stony Brook, NY, 11790 (516) 751-5350 References 1 SpeilfS and J.P. Leinev/eber. 19 6 9 * Asbestos Minerals in Modern' Technology. Environ. Res. 2:166-208. 2. Rohl,A.N. 1974. Asbestos in Talc. Environ. Hlth. Perspec.9129132. 3. Peters,G.A and B.J.Peters. I98O. Sourcebook on Asbestos Diseases: medical,legal and engineering aspects. Garland Press,N Y .443p. 4. Harington,J.S.,A.C Allison and D.V. Badami. 1975- Mineral fibres: Chemical, Physicochemical and Biological Properties. Advan, Pharmacol. Chemother. 12:291-402. 5 . Selikoff,I.J and D.H.K. Lee. 1978. Asbestos ans Disease. Academic Press, NY. 549p. 6. Light,Y/.G and E.T.V/ei. 1977. Surface Charge and Hemolytic Acti vity of Asbestos. Environ. Res. 13:135-1^5* 7 . Sawyer,R.N. 1977* Asbestos Exposure in a Yale Building. Environ. Res. 13:146-169. 8. Commitee on Biological Effects of Atmospheric Pollutants. 1971Asbestos. The need for and feasibility of air pollution controls. National Academie of Sciences. 40p. 9* Report for the Commission of the European Communities. Public Health Risks of Exposure to Asbestos. 1977* Pergamon Press.I49p. 10. Preger ,L. ,D. T. Aral. ,p. Kotin. ,H. V/eill and J. Y/erchick. 1973. Asbestos-related Disease. C-rune & Stratton Inc. 2'71p. 11. TimbrellfV. 1973. Physical factors as etiological mechanisms. In: Biological effects of Asbestos. Ed. ?. Bogovski e.a. 295p. 12 Suzuki,Y. 1'97^. Interaction of Asbestos with Alveolar Cells. Environ. Hlth. Perspec. 924l-252. 13 Langer, A.M. 1974 The subject of continuous vigilance. Environ. Hlth. Perspec. 9-53-56. 14. RohlfA.N and A.M. Langer. 1974. Identification and quantification of Asbestos in Talc. Environ. Hlth. Perspec. 9**95-109 15- Smither,W.J. 1974. Asbestos in the Work place and the Community. Environ. Hlth. Perspec. 9*32 7-329. 16. Timbrell,V. 1965- The inhalation of fibrous dusts. Ann. NY. Acad. Sci. 132,255. 17 ZoloViC.jT. Bourilkov and L. Badadjov. 19 6 7 . Pleural Asbestosis in Agricultural Workers. Environ. Res. Is28 7-2 9 2 . 18. Fondimare,A and J. Desbordes. Asbestos Bodies and Fibres in Lung Tissues. Environ. Hlth. Perspec. 9 147-148. 19 Langer,A.M.Selikoff I.J and A. Sastre. 1971 Chrysolite asbestos In the lungs of persons in New York City. Arch, Environ. Health. 22.-348. 20. Snider,D.W.fD. Epfeifier and J.J. Mancuso. 1972. Asbestosform impurities in commercial talcum powders. Compass. 4 9 ,//2:65-6 7 . DEPARTMENT OF HEALTH & HUMAN SERVICES 0 / I L* Public Heekh Service Date From Subject To Memorandum Kay 21, 1985 Robert Brown BKB* Division of Mathematics (HFF-118) Four methods of quantitating mesothelioma risk based on the Selikoff* et. al, insulation workers asbestos study* Technical support for QRAC's asbestos risk assessment* QRAC In fig. I we have plotted on a log-log scale Selikoffrs original mesothelioma Incidence data vs. years since first exposure to asbestos* Incidence is defined as number of mesotheliomas/man-years exposure* The data do not seem to fit a single straight line. Uncertainties of exposure in the early part of the century and the general decline In Intensity of asbestos exposure after World War II ere possible sources of error. For these reasons, as well as general lack of fit of both recent data and distant past data, Peto recommended use of a more homogeneous subset of workers for quantitative purposes, namely those workers first exposed between 1922 and 1946 [8]. It can be inferred from Selikoff's report that this subset consists of about 4800 workers. Peto reports 180 mesotheliomas (3.75%) among this subgroup out of a total of 236 mesotheliomas for all 17,800 workers followed from 1967 until about 1978 or 1979. Note that Selikoff only reported 175 mesotheliomas total; however, his reported follow-up period was also shorter (1967-1976). Plotting Peto's homogeneous 1922-46 cohort subset, we see that 31 bt * nicely fits the data (expressed as a straight line on log-log 2.1 paper with a slope of 3.1). We also see that b(t-10) f nicely fits the data (with a different value for the constant b) and may be a reasonable QRAC - 2- way of looking at mesotheliomas since the time lag from mesothelioma Induction to death is not zero. The time of mesothelioma Induction Is not even a ell defined concept and may be intimately entertained with the concept of stage definition In* for example, a multistage cancer process. Nevertheless, both these model fits assume mesothelioma to be a ncnincidental tumor (l.e., & life table where incidence Is the ratio ftumor bearers/fsurvivorst re-expressed in man-years, per time Interval). If we assume mesothelioma annual incidence to be better approximated by a prevalence or Incidental definition, (ftuaor bearers/fdead in Interval), I 64 then bt ' seems to be a rough though not very tight fit to the original Selikoff data. Petofs reported 1922-1946 data set does not easily allow determination of a prevalence fit. However, since the prevalence denominator is defined In terms of deaths per time Interval rather than the much larger number of survivors to date, the first 2,271 deaths (12.72 of 17,800 workers) reported by Selikoff are very heavily weighted with the 1922-1946 cohort used exclusively in the two nonincidental curve fits above. Therefore comparisons of slightly different cohort subsets may still be useful. We estimate that the average time since first exposure for the Peto subset (1922-1946 first exposure) is about 37 years (Peto's 1978* follow-up) or 35 years (SelikoffTs 1976 follow-up). This compares to 25 years average time since first exposure usually reported for all 17,800 workers. We also make the assumption that workers ceased exposure on average 3 years before death.