Document VpNRbvvevaQNLOXGO7rOOaJo

fthone (914) 331-0113 NATIONAL ASSOCIATION OF FILTER MATERIAL MANUFACTURERS U. P. O. Box 245, Kingston, N. Y. 12401 December 21, 1973 The Hearing Clerk Food and Drug Administration Room 6-86 5600 Fishers Lane Rockville, Maryland 20852 Gentlemen: Ref: Department of Health, Education and Welfare Food and Drug Administration 21 CFR 121, 128, 133 Federal Register Vol. 38, No. 188 Friday, September 28, 1973 The following comments are submitted in accordance with the above referenced proposed rule-making, by the National Association of Filter Media Manufacturers. The NAFMM represents all of the National producers of cellulose/asbestos filter media as well as allied companies. The adverse health effects of inhaling large quantities of asbestos fiber over a substantial period of time has been amply reported. In 1971, the seven-man Panel on Asbestos of the National Academy of Sciences' Committee on Biologic Effects of Atmospheric Pollutants examined all of the relevant data and concluded: "Associations between asbestos exposures and malignancies of the gastrointestinal tract and of other sites have been reported, but the data are still inconclusive." Reviewing the situation two years later, the Advisory Committee on Asbestos Cancers of the International Agency for Research on Cancer of the World Health Organization concluded that, for some exposed populations, such an increase does exist, but that the excess is relatively small, that further research is needed, and that * # 2- - there is no cancer hazard resulting from the ingestion of "asbestos fibers present in water, beverages, food or in the fluids used for the administration of drugs." If the human evidence as to a possible relationship between asbestos exposure and an increased risk of gastrointestinal cancer is inconclusive, the animal experimentation that has been performed to date is even more so. As is pointed out in the proposed rulemaking, the major feeding experiments of L.M. Swinburn (your reference No. 40) and W. E. Smith (your reference No. 41) produced no tumors in thecpstrointestinal tracts of the animals, even though the amounts of fiber fed to the animals were massive when compared with human occupational exposures know to result in respira tory disease. These findings have been further confirmed by the work of J.M.G. Davis of the Institute of Occupational Medicine, Edinburgh, Scotland, whose feeding experiments with rats have thus far produced no GI tumors. A 1967 paper by G.M. Bonser and D.B. Clayton ("Feeding of Blue Asbestos to Rats," 1967 Annual Report, British Empire Cancer Campaign for Research, p. 242) likewise reported no tumors in 40 rats fed crocidolite asbestos throughout their lifetime. It has been suggested that mesothelioma might be caused in some instances by a migration of asbestos fibers from other parts of the body, including the gastro intestinal system, it should be pointed out that no mesothelial tumors were found in any of the animals involved in the various feed experiments. While some injection experiments have produced site and mesothelial tumors in test animals, the amount of fiber used in each case was so large as to render any attempted translation to human experience completely inappropriate. Moreover, the work of M. F. Stanton of the National Cancer Institute ("Some Aetiologic Considerations of Fiber Carcinogenesis," as presented at the World Health Organization Conference on 3- - the Biological Effects of Asbestos, Lyon, France, October 2-5, 1972, paper 48A) and P. Gross ("The Locus of Pathogenicity of Asbestos Dust", Archives of Environmental Health, Vol. 27, October 1973, p. 240) strongly suggests that injection of implantation cancers in test animals are directly related not only to dose but, even more importantly, to the size of the fiber used in the experiment. Electron microscope sized fibers did not produce tumors in the test animals. This is extremely relevant to the proposed rulemaking in that the vast majority, if not all, of the asbestos fibers found in air, water, beverages, drugs and food in the United States are in this apparently non-carcinogenic size range. On the basis of the human and animal studies contained in the literature, one would have to conclude that the relationship between asbestos exposure and an increased risk of gastrointestinal cancer is inconclusive. Because of the known respiratory health hazards of asbestos, there has been a tendency among some researchers to ascribe to asbestos sole blame for an excess mortality from any cause experienced by an exposed cohort, even if that cohort was exposed to other toxic substances in addition to asbestos. It is possible that some other substance or agent is responsible for the excess GI cancer rate found among some asbestos exposed popula tions. This would explain why other exposed cohorts have shown no such increase and also why animal feeding experiments have failed to demonstrate an increased risk. Until further research is completed, this possibility cannot be excluded from consideration. It should also be pointed out that despite the long-term use of asbestos filters in this country, there is not one shred of evidence that anyone in the World has ever contracted either GI cancer or mesothelioma from drinking a beverage filtered through an asbestos filter or by being injected with a drug filtered in the same fashion. In fact, the GI cancer rate in the United States is on the decline, and # 4- - if we were going to see an epidemic of these diseases because of asbestos filter contami nation, certainly we would have had some evidence of it before now, especially in view of the fact that asbestos containing filters have been in use for some 85 years, to filter products that are injected frequently, such as insulin and allergenic products. It is obvious that in dealing with the question of asbestos disease that all of the human groups studies to date have had occupational, para-occupational, neighborhood or household exposures of a magnitude far greater than any that could reasonably be expected to exist with regard to the general public. While precise comparisons are impossible because of the different measurement techniques used to quantify asbestos concentrations in the workplace and in ambient air and water (occupational exposures are measured in fibers per cc while ambient exposures are measured in nanograms per cubic meter of air or micrograms per liter of water), rough approxi mations indicate that even safe occupational exposures are many thousands of times higher than those to which the general public is or has been exposed. In addition, asbestos is ubiquitous in our environment. It exists naturally in small though detectable amounts in serpentine rock formations throughout the world and is constantly being released into the atmosphere through erosion by wind and water. Asbestos is thus natural to our environment and, as analysis of dated ice from the polar caps has revealed, has been for thousands or even millions of years. Man has evolved into his present form in an atmosphere containing a natural back ground of asbestos, both in the air he breathed and in the water he drank. While precise comparisons are impossible, today's atmospheric levels of asbestos, even in urban settings, are still only marginally higher than in the distant past. Most speculation as to the health hazards of the minute amounts of asbestos that have been demonstrated to exist in air, water and in beverages or drugs completely ignores these two most important factors, the differences in magnitude between occupational -* and ambient concentrations, and the ubiguitous nature of asbestos in the environment. To conclude, that because exposure to heavy concentrations of asbestos fiber in the workplace is a recognized health hazard, general public exposure to vanishingly smaller quantities of the same material must also be a hazard, is incredibly ridiculous, especially since man has survived such exposures for millions of years. One would have to conclude that the general public is not at risk, and this, of course, is precisely the conclusion contained in the "Report of the Advisory Committee on Asbestos Cancers to the Director of the International Agency for Research on Cancer" (your reference No. 52). On the question of fiber release from asbestos-containing filters into the finished product, it is important once again to comment upon the ubiguitous nature of asbestos in our environment. A number of studies have been undertaken in the United States, Canada and in the United Kingdom to determine the so-called "natural background" level of asbestos in ambient air and water. Because of the different analytical techniques employed, the results of the various studies are not entirely comparable. However, it would appear that the average background level in water in the U.K. is in the vicinity of 4x10"^ grams per gallon; and in the United States and Canada, some where between 2 and 5 micrograms (lxl0"^ grams) per gallon. Because of the complexity of the analytical techniques used to measure the asbestos content in water, an error factor of from t 250 percent must be assumed in evaluating the results. The higher background found in North America is probably due to the greater quantity of asbestos bearing serpentine rock found on the continent, especially on the Eastern seabord, where the samples were collected. To place the results in perspective, one microgram per gallon equals one ounce in 2.84 billion gallons. If these background levels can be considered "normal" for human ingestion, the question must then be raised: how do these levels compare with beverages, drugs, etc., # # 6- - filtered through asbestos-containing filter pads? Research into this aspect of the problem has been conducted by Cunningham and Pontefract (your reference No. 43); Nicholson, Maggiore and Selikoff (your reference No. 45); Badami and Rickards (unpublished data submitted to FDA on May 1, 1973, by S. Holmes of the Asbestos Research Council); J. P. Leineweber (unpublished summary of asbestos fiber content in the United States commercial beers, copy attached); Dr. Philip McGrath (FDA memorandum to Dr. Armand Casola); and Dr. J. R. Crout (FDA Preliminary Results of Special Survey). The results of the Cunningham et a l , Badami et a l , and Leineweber studies all showed the asbestos fiber content of the various soft drinks, beers, wines and parenteral drugs tested to be no more than or, in many cases, less than the natural background level of asbestos that one would expect to find in that particular country. Badami and Leineweber could find no correlation between the use of asbestos-containing filter pads and an increase of asbestos fiber in the finished product. Only Nicholson et al found levels of asbestos in parenteral drugs higher than would be considered normal. This is not particularly surprising, because the Mount Sinai Laboratory has consistently reported higher asbestos levels in ambient air and water than have been found by other laboratories in the United States and abroad. Professional speculation is either that the Mount Sinai electron microscopy technique is faulty to some extent, or that improper attention is being paid to achieving true clean room conditions so that the samples are not contaminated prior to measuring. When seeking to obtain accurate measurements of such infinitesimally small quantities, even the slightest contamination could throw the results off considerably. Of special interest with regard to the proposed rulemaking are the reports of McGrath and Crout. While preliminary in nature, the Crout survey shows that the use of asbestos-containing filters bears no relationship to the presence of asbestos fibers in the finished product. In this respect, it confirms the studies of Badami and Leineweber, and throws considerable doubt on the proposition that the elimination 7- - of asbestos filters for parenteral drugs will result in any reduction in asbestos content of the finished product. In fact, the MaGrath survey could lead one to the conclusion that asbestos filters should be required for all filtration systems, since, to quote the McGrath memorandum, "Filtration through an asbestos filter de monstrated the remarkable capacity of this device to reduce the amount of asbestos in a given sample." It would be highly ironic, in light of the controversy that this subject has generated, if it turned out that the best method for the reduction of natural background levels of asbestos in base liquids would be the use of an asbestos filter. In Summary 1. The evidence for a relationship between asbestos exposure and an increased risk of gastrointestinal cancer is inconclusive. 2. Should a relationship be established, the evidence to date casts serious doubt on the proposition that the ingestion of asbestos fibers causes either gastrointestinal cancer or mesothelioma. 3. The size range of fiber most commonly found in beverages, drugs, etc. does not appear to be carcinogenic. 4. Asbestos is ubiguitous in our invironment and has been for millions of years. 5. With the exception of the Nicholson et al study, which has been the object of serious criticism, none of the other studies done to date show an unnatural level of asbestos fibers in the finished product. 6. No correlation has been demonstrated between the use of asbestos filters and the presence of excess asbestos fibers in the finished products. 7. It is unlikely that the elimination of asbestos filters for parenteral drugs will result in a reduction in the asbestos content of the drugs. 8. The levels of asbestos found in beverages, drugs etc. have not been shown to be hazardous to human health. * 8- - On the basis of the above analysis of the situation, the following recommendations are offered: 1. The FDA acted properly in not placing restrictions of any type on the use of asbestos filters in the food and beverage processing industries. This decision should not be altered. 2. In light of (1) the available medical evidence (2) the proven value of asbestos filters in removing pyrogens from parenteral drugs, and (3) the unlikelihood that asbestos filters contribute significantly if at all to the asbestos content of the final product, it is our recommendation that the FDA delete from Section 133.8 (j) of the proposed regulations that portion prohibiting the use of asbestos filters "unless it is not possible to manufacture that drug or drug ingredient without the use of such a filter". 3. Only for the reason of Good Manufacturing Practice, we support the precautionary measure of requiring a non-asbestos-containing after-filter whenever an asbestos filter is used for filtering parenteral drugs, even though it is highly unlikely that this is necessary either from the public health or drug purity standpoint. 4. Because of the extreme complexity of the technology involved and the possibility of substantial error, we further recommend that the requirement for proof of reduction be eliminated from the proposed regulation. Since the sampling method as described in the NIOSH document is useful only for detecting light-sized fibers, and not those in the EM range, which constitute the vast majority of all asbestos fibers likely to be found in a parenteral drug, the method is inappropriate and misleading for evaluating the asbestos burden of a particular drug. In addition, it would appear from the studies described above that the asbestos content of the pre filtered solution is more important in determining the content of the final product than whether or not an asbestos filter with or without an after-filter is used. 5. We support that portion of Section 133.8 (j) with regard to the use of an asbestos-containing filter without an after-filter,  9- - We believe that these recommendations will not only provide for the public the same degree of health protection as would the proposed regulations, but in addition, the eTfnination of the sampling requirement and of that portion of the regulations prohibiting the use of an asbestos filter unless it is not possible to manufacture the drug in any other fashion, will considerably ease FDA's administrative enforcement of the standards. It would be much simpler for an FDA inspector to merely have to check to see that an after-filter was being used to assure compliance with the regulations than it would be to set up some type of complicated permit system that would end up creating tremendous confusion and pleasing neither the FDA nor the drug companies involved. We trust that these recommendations and supporting comments will be given fullest consideration by FDA in the development of the final standards on the use of asbestos-containing filters. Sincerely yours Gerald Marconis, Secretary National Association of Filter Material Manufacturers :su m m a o. JriuO * ;r f^r^- "\Trrv? :n COMMI > J--r :srs Brand____________________ SCHLITZ (regular) Sample 1 SCHLITZ (regular) Sample 2 SCHLITZ draft Fiber Content ( f q q / c a-l-.-)--*- (Primary - diatornite only (Polish -- asoesuos-cexaulose precoat + diacomit body feel C .2 3 2.50 oame as regular t Ara repo re RHEINGOLD extra dry (Primary - diatomite only (Polish - asbestos-diatomite ' precoat + diatomite body feel 0.61 SCHMITZ . Same as Schlitz regular 1.39 NA11RAGAN SET (Primary - diatomite only ' 1.14 (Polish - diatomite only OLD BOHEMIAN , (Primary - two stage diatomite (No polish ,;. . 3.07 KAIERS (Primary - diatomite only (Polish - diatomite only 0.24 PIELS regular . PIELS draft ? 7 * 3.96 0.74 *lyg/gal. = I gram in 1,000,000 gallons = 1 ounce in 2,840,000,000 gal. All values determined by electron microscope observation. No fibers ' were visible in any of the samples under optical microscopic observation at 450X magnification. ' ' J. P. -LEINEWEBER ' 10-15-68 . ,rsr~ft P . i! *