6. Aspartylphenylalanine Diketopiperazine (DKP) The fact that Dr. Liebovitz did not even mention DKP, a chemical which can sometimes be more prevalent than aspartame itself, is rather distressing. At least one of the references he cites, AMA (1985) discusses DKP at length, so he must be aware of this breakdown product. Before aspartame was foisted upon the public, the amount of this particular DKP in the diet was essentially zero (Federal Register 1984). Therefore, no claim can automatically be made that DKP ingestion is safe. Several quality studies would have to be performed in order to conclude that DKP probably does not have a detrimental affect on humans. No such quality studies have ever been done. Most of the controversy surrounding DKP has involved the issue of brain tumors and uterine polyps. While I will limit my discussion in this section to cancer, there are two very important preliminary points that need to be made: a. Cancer is a very serious disease. However, there are countless other serious diseases and therefore no reason to limit the concern regarding DKP to cancer. The FDA and EPA have approved numerous drugs and chemicals that have unexpectedly caused or contributed to a wide range of serious adverse reactions other than cancer. There is no reason to believe that the possible detrimental effects of DKP are limited to cancer. Dioxin is an example of a chemical which was linked to cancer. Now it turns out that the biggest issue is that dioxin has an extremely deleterious effect on the immune system. b. When looking at the aspartame and cancer issue, DKP becomes a likely candidate for a possible cause. However, it is only one of several possibilities. Ingesting methanol or significant quantities of racemized amino acids could be another cause or contribuatory factor. Brain chemistry changes making one slightly more susceptible to brain tumors caused by long- term ingestion of aspartic acid and/or phenyalalnine might be another possibility as to how aspartame could contribute to brain cancer. We will discuss the studies that show aspartame caused cancer in laboratory animals later in this section. However, it should be understood that the pre-approval studies were so poorly designed and conducted that it would be impossible to conclude that aspartame is safe. Some of the flaws which show that the FDA could not possibly have concluded that aspartame does not cause cancer are as follows: Lack of Statistical Power (simplified discussion of statistics) --------------------------------------------------------------- Aspartame is being regularly consumed by over 70 million people in the U.S as discussed earlier. If aspartame was to cause cancerous tumors in 1% of the people who ingest it for several decades (as is what happens with cigarettes), 700,000 people would develop such aspartame-caused tumors in that time. A 5% tumor rate would lead to 3.5 million people developing tumors. etc. Let's say, for arguments sake that aspartame causes cancerous tumors in 2% of the people in their lifetime. That's 2% of 70 million people (in the U.S.) or 1.4 million people. If I wanted to see if this was really the case by testing aspartame in animals, I would feed aspartame to 70 million (test) animals. To another 70 million animals, I would give a normal diet. If the aspartame-ingesting animals had 1.4 million cancerous tumors more than the non-aspartame- ingesting (control) animals, the cancer rate of aspartame would be 2%. Of course, it would be impossible to conduct an experiment on this many animals. So, instead of 70 million test animals and 70 million control animals, I could use 1000 test animals and 1000 control animals. If the cancer rate was 2%, there would be, in theory, 20 more cancerous tumors in the test animals than in the control animals. If there was a 2% cancer rate for aspartame, the reality is that each time I conducted the experiment, an average of 20 more cancers in the test animals would appear than in the control animals. If I ran the experiment ten times I may get something like this: Number of Excess Tumors in Test Animals As Compared to Control Animals Run #1 24 Run #2 14 Run #3 18 Run #4 22 Run #5 28 Run #6 19 Run #7 13 Run #8 20 Run #9 22 Run #10 20 As you can see, the numbers would vary, but the more times I conducted the experiment, the closer the average of all of the runs would be to 20. Instead of running the experiment on 1000 test animals and 1000 control animals, a researcher may choose to use only 100 test animals and 100 control animals. If there was a 2% cancer rate for aspartame, there would be an average of 2 more cancerous tumors for the test group as opposed to the control group. However, since there are variations for each experimental run, the results might look something like this for 10 runs (on 100 test animals and 100 control animals): Test Group Control Group Run #1 2 0 Run #2 4 1 Run #3 0 0 Run #4 3 0 Run #5 4 1 Run #6 0 2 Run #7 5 1 Run #8 3 1 Run #9 5 0 Run #10 0 0 The average over 10 runs is a 2% more cancer rate in the test group, but there is a wide variations for each individual run. As you can see, the smaller the group of animals used, the wider the percentage variation for each run. Instead of 10 runs on 100 test animals and 100 control animals, had I simply done a single run, the results would likely be meaningless because there is such a wide variation for a particular run with this few amount of animals. For example, Run #9 shows 5 cancerous tumors in test animals and 0 in control animals. Whereas Run #6 shows 0 cancerous tumors in test animals and 2 cancerous tumors in control animals. Therefore, in order to more accurately investigate whether aspartame causes a 2% cancer rate (1.4 million in 70 million users), I would have to use many times more than 100 animals or make numerous test runs with 100 animals each. As you can see, if I was to use a small amount of animals, I could not accurately determine whether there was a 2% cancer rate of aspartame. I may be able to conclude that there is not a 20% cancer rate for aspartame use (or 14 million cancers). If I can say that aspartame does not cause 14 million cancers, that does not help me very much. It begs the question as to whether it causes a 5% cancer rate, a 2% cancer rate (1.4 million people) or a 0.2% cancer rate (140,000 people). G.D. Searle used such a small number of animals in their experiments relating to cancer that it would have been impossible to conclude that aspartame was safe even had the experiments been conducted properly and had there been no significant number of tumors found. FDA Chief of Statistical Evaluations Branch, Satya D. Dubey stated the following in a memo to the FDA Commissioner's office (Farber 1989, page 101): "From the design viewpoint, the probability of observing a statistically significant result at 5% significance level with 60 animals in the control group and 40 animals in the treatment group when the true difference in incidence rates of brain tumors is not more than 5% would be less than 27.9% . . . Even 27.9% of statistical power will generally be considered to be very low power. The studies, E33/34 and E70, which I have statistically reviewed, thus possess very low power to detect true significant effect of the kind stated above. . . . Therefore, their results should not be considered confirmatory for decision purposes." What Dr. Dubey is saying is that there is only a 27.9% liklihood that a cancer rate of less than 5% would be found for the experiments discussed above. A 4.9% cancer rate still amounts to 3,430,000 cancers in 70 million aspartame users! E33/34 and E70 are two studies which showed that aspartame caused cancer in laboratory animals and will be discussed later. This is one of the tricks that is sometimes used in animals experiments. If the researcher uses a small amount of animals, the study may not find an increase in cancer rates for the test animal as opposed to the control animal, but the lack of statistical power caused by the use of a small number of animals made such a finding very unlikely. When a researcher does use a small number of animals, the condition of every single animal is very important. If even one animal's cancer is missed, the results of the experiment can be changed dramatically. Had the studies E33/34 and E70 been run perfectly, there would have been a 27.9% chance that the researchers would have found a 5% or more tumor rate. Since the experiments were so incredibly sloppy and the condition of the animals was really a guess, it is likely that a 10% cancer rate would have been missed. G.D. Searle employees attempted to increase the statistical power of the cancer studies by combining the results from different studies such as E33/34 and E70 (Cornell 1984). E70 studies the effects of aspartame on Charles River albino rat offspring by giving adding it to the mother's feed through the period of lactation and then the offspring's feed until death. The offspring were supposed to be examined at 104 weeks (or at death). E33/34 was a chronic dosing study of aspartame given in four different doses to rats for 104 weeks. However, FDA Senior Statistician, Dr. Satya Dubey stated in1981 in a memo to the FDA Commissioner's science advisory team on aspartame (Farber 1989, page 102): "Since the protocol of E70 study is different, its design is deficient, its data exhibit certain peculiar patterns, and the statistical conclusions derived from such data are much different from that of the E33/34 study (compare the statistical results of E33/34 and E70 studies), it does not appear proper to me to combine the results of these two studies." In his treatise on aspartame, Farber (1989, page 102) states in regards to combining E33/34 and E70: "I asked Myrto Lefkopoulou of the Harvard School of Public Health, Biostatistics Department for her comments on the statistical aggregation of E70 and E33/34. She considered the aggregation of these studies inappropriate because the effect of an in utero study [E70] is different than a chronic feeding study [E33/34]. ...she stated that the in utero study is looking for a teratogenic effect, whereas the chronic feeding study was concerned with carcinogenicity -- did aspartame promote brain tumors?" Farber points out that even if the results of these studies are (improperly) combined, there is only a 67% chance that a 5% tumor rate could be discovered (Farber 1989, page 103). Live / Dead Status ------------------ Below are listed a selection of animals from G.D. Searle's studies where the researchers couldn't even determine whether the animal was alive or dead. This information comes from the original data observation sheets as audited by the first FDA Task Force in 1975/1976. While not all of this information applies to aspartame studies, it should serve to give you a sense of the confusion that was rampant throughout the animal observations (Schmidt 1976a, page 21 of US Senate 1976a). J24HM Found dead 3/21/71 Alive 5/19/71 Dead 5/16/71 Alive 7/14/71 Dead 8/11/71 K18LF Alive 4/22/71 vanished (dead ?) 5/20/71 Alive 6/17/71 vanished (dead ?) 7/15/71 M25CF Found dead 3/6/71 Alive 6/18/71 Dead 7/16/71 Alive 9/10/71 Alive 10/8/71 Dead 11/5/71 H28MF Alive 7/13/71 vanished (dead ?) 8/10/71 H15CF Alive 7/13/71 vanished (dead ?) 8/10/71 G 2HM Found dead 3/10/71 Alive 8/9/71 A15MM Found dead 3/13/71 Alive 5/3/71 Dead 6/1/71 Alive 8/23/71 Dead 9/20/71 G16HM Found dead 3/9/71 Alive 8/9/71 Dead 9/7/71 A 6HM Found dead 2/25/71 Alive 5/3/71 Dead 6/1/71 Alive 8/23/71 Dead 9/20/71 G23HM Found dead 3/7/71 Alive 8/9/71 Dead 9/7/71 E15MM Found dead 1/21/72 Alive 2/25/72 G 8MM Found dead 9/3/71 Alive 11/29/71 Dead 12/27/71 B19HF Alive 6/29/71 vanished (dead ?) 7/27/71 Alive 8/24/71 vinished (dead ?) 9/21/71 Alive 10/19/71 vanished (dead ?) 11/16/71 Alive (?) 2/22/72 B21HF Found dead 2/25/71 Alive 8/24/71 Dead 9/21/71 Alive 10/19/71 Dead 11/16/71 Alive 2/22/72 B14MF Killed 7/30/71 Alive 10/19/71 Dead 11/16/71 Alive (?) 2/22/72 B12HF Found dead 9/2/71 Alive 10/19/71 Dead 11/16/71 Alive (?) 2/22/72 B 4CF Found dead 9/12/71 Alive 10/19/71 Dead 11/16/71 Alive (?) 2/22/72 D30LF Found dead 1/22/72 Alive 2/22/72 B15HF Found dead 1/25/72 Alive 2/22/72 C29LM Found dead 3/29/71 Alive 6/2/71 Dead 6/30/71 C12HM Found dead 8/10/71 Alive 10/20/71 Dead 11/17/71 There may have been numerous animals which were listed incorrectly as alive and then dead. There would have been no way for the Task Force to discover errors such as that. An animal's alive or dead status is only one of the variables which goes into the statistical analysis of an experiment testing for cancer. Whether an animal has a tumor is another important piece of information. Tumor Status ------------ There were many instances in which the FDA Task Forces discovered that G.D. Searle had confused animals to such an extent, the tumor status was not known. To give you a sample of the confusion that reigned during this period of time in the G.D. Searle laboratories, here a sample from just one cancer study of one of G.D. Searles drugs. The testimony is that of FDA Toxicologist Dr. Addrienne Gross (Gross 1976b, page 44-49): What may be added here is that the live/dead status of the experimental animals is not the only "careless" type of error present in the Observation for Drug Effects. The following are merely a few samples of the way entries are kept on externally visible tissue masses in these animals; most of such tissue masses turn out to be benign or malignant mammary tumors. 1. Animal M21 (a control female), is said to have developed a tissue mass in the left cervical area; the mass is said to have been initially detected on 6/18/71; at the next observation period on 7/17/61 this animal is checked off as having no tissue masses; however, the next animal on the list (M22 - an exposed female) is now listed as having its tissue mass "larger" (presumably than at the previous observation period); but this particular animal had not been listed as having exhibited any such masses at any time in the past; at the next observation period on 8/13/71, the tissue mass in the control animal is said to be "larger" while the exposed animal is said to have no tissue mass whatsoever. 2. Animal J16 is said on 2/23/72 under "Tissue Masses - Lesions" to have an abscess in the left inguinal region which is "larger"; no mention of any such abscess is evident for any prior observation. 3. Animal B26 is said on 12/14/71 under "Tissue Masses - Lesions" that its mass is larger. But no tissue mass in this animal is previously reported. Four weeks later on 1/12/72 a tissue mass is said to have been initially detected on that day. 4. Animal B27 is said on 9/21/71 to have developed a tissue mass initially detected on that day; at the next observation period on 10/19/71 the mass is said to be unchanged; at the next observation period on 11/16/71 the mass is said to have regressed; at the next two observation periods on 12/14/71 and 1/12/72 this animal is said to be free of tissue masses; on 2/8/72, the next observation period, the mass for this animal is said to be the "same"(!) 5. Both animals A2 and A3 are said on 9/20/71 to have developed tissue masses initially detected on that day; at the next observation period, on 10/8/71 both of these animals are indicated to be free of any tissue masses; at the next observation period on 11/5/71 it is indicated that both of these masses regressed. 6. Animal E3 is said on 7/1/71 to have developed a tissue mass initially detected on that day; the following are the results of the six subsequent examinations: 7/29/71 - animal is free of any masses 8/26/71 - mass is the same (as what?) 9/23/71 - mass is the same 10/21/71 - animal is free of any masses 11/08/71 - mass is the same 12/06/71 - mass regressed 7. Animal E9 is said on 9/23/71 to have developed a tissue mass initially detected on that day; the following are the results of the four subsequent examinations: 10/21/71 - mass regressed 11/18/71 - animal is free of any masses 12/16/71 - mass regressed 1/13/72 - animal is free of any masses 8. Animal D29 is said on 7/1/71 to have developed a tissue mass initially detected on that day; the following are the results of the seven subsequent examinations: 7/29/71 - animal is free of any masses 8/26/71 - animal is free of any masses 9/23/71 - mass is the same 10/21/71 - animal is free of any masses 11/18/71 - mass regressed 12/16/71 - mass is the same 1/13/72 - animal is free of any masses 9. Animals H26, D12, K25, D5, K17, and D19 each are indicated to have developed more than on tissue mass; in each case, however, observations made subsequently fail to distinguish to which tissue mass they apply. 10. Animal H19 is said on 11/2/71 to have developed a tissue mass initially detected on that date; the subsequent observation dated 11/30/71 indicates this animal to be free of any tissue masses; at the next observation made on 12/28/71 the mass in this animal is said to be the "same." This list of 10 examples involving some 16 animals could be extended further but it is sufficient to make the point that records maintained at Searle on the appearance, persistence or "regression" of tissue masses do not give one much assurance on their reliability. One may ask -- can this sort of thing be shrugged off as merely "careless" observations made by those who were supposed to make such observations? Or was this a situation that could be expected to have occurred, given the policy and practice in force in the Department of Pathology and Toxicology at Searle? A review of the names of the "observers" entered on these sheets referring to "Observations for Drug Effects" reveals different names for subsequent observations. Question: If whoever observed the animals on a given day and who recorded such observation in his or her notebook, is someone else than the one having observed them at the previous observation period, who made similar observations in some other notebook, how can it be said that a certain tissue mass is the "same" or "larger" or "unchanged"? After a certain period in the experiment no names of any observers appear on these records. Searle maintains in their last communication (line 10, page 15) "In the truest sense, the errors identified by the FDA (in these records) were completely irrelevant to the scientific conclusions of the study..." We note this evaluation of "irrelevant" by Searle but we cannot agree with it, and the reason for this is very clear: The title printed on these "Observation for Drug Effects" is "Statistical Work Sheet"; this says that it is reasonable to expect that these "careless" entries must have formed the basis for input for statistical operation which are crucial to the "scientific conclusions of the study." The methodology used in these statistical operations at Searle (the Horton and Sachs Life-Table procedures) depend completely on the time a certain tissue mass (tumor) is observed and on the time the animals with the mass (and all other animals in that group) died. Now, if the live/dead status of each animals was "carelessly" entered on these "Statistical Work Sheets" as conceded by Searle and if its status as a tumor-bearer at any time was largely in doubt (as demonstrated here) of what value are any of the statistical computations based on this kind of raw input data and would this not affect the "scientific conclusions of the study"? Searle complains (line 2, page 14) that these records "became a subject of considerable levity at the hearing." I believe, however, that the members of the Subcommittee are sufficiently knowledgeable in the ways of the world to realize that animals seldom die more than once. However, I would tend to agree with Searle here, that the state of their records on observations collected during the course of this study is indeed no laughing matter. Given the lack of statistical power of NutraSweet's animals experiments and their inability to be certain whether their test animals are alive or dead and whether they did or did not have tumors, how could any unbiased individual rely on this information to make health policy determinations that would effect an entire country? The items discussed above combine to render the G.D. Searle experiments that tested for cancer totally worthless as far as using them to prove safety. However, these abuses were just the beginning of what was discovered. Other Problems -------------- The following testimony by Dr. Jacqueline Verrett, a former toxicologist of the FDA, who was the Senior Scientist of the FDA Bureau of Foods Task Force describing a few of the other problems with G.D. Searle's cancer studies (Verrett 1987): 1. There was no protocol written until the study was well underway. 2. Animals were not permanently tagged to avoid mixups over the course of the study. 3. Changes were introduced in some laboratory methods during the study with inadequate documentation. 4. There was either sporadic or inadequate reporting and monitoring of both feed consumption and animal weights. 5. In some cases, tumors were removed, and the animals then returned to the study. 6. Animals were recorded as dead and then subsequent records, after varying periods of time, indicated the same animal was still alive--almost a certain evidence of mixup. 7. Many animal tissues, a significant number, were autolyzed, that is, decomposed, before any post mortem examinations were performed. 8. And finally, of extreme importance is that in the DKP study there was evidence, including pictures found in notebooks at Searle, that the diets were not homogeneous, and that the animals could discriminate between feed and the included particles of DKP. In other words, they may or may not have been eating what it was assumed they were eating. Almost any single one of these aberrations would suffice to negate a study designed to assess the safety of a food additive, and most certainly, a combination of many such improper practices would, since the results are bound to be compromised. Raymond Schoeder, a former G.D. Searle employee told the FDA that the particles of DKP (in experiments to test DKP for cancer) were so large that the rats could discriminate between the DKP and their normal diet (Graves 1984, page S5500 of Congressional Record 1985a). After Raymond Schroeder had made his original statements regarding the DKP study, FDA Investigators went to interview him. He was then employed at a different company. When the investigators got there, they noticed that a G.D. Searle company employee had signed in immediately before them. During the interview, Mr. Schroeder retracted his statements about the DKP study (Olney 1987, page 8). The evidence is very strong showing that the amount of DKP ingested was much less than originally intended. This evidence includes the statement by Mr. Schoeder as well as a picture of the large DKP particles. This is a crucial issue as it shows that the uterine tumors and other problems found in NutraSweet-fed rats in the DKP studies may have occurred at a relatively low dose of DKP. Dr. Adriene Gross describes problems with a 115-week study testing DKP in rats (Gross 1987a, page 7): 1. Substitutions of some of the animals in that study. 2. The presence of intercurrent disease amongst the test animals and the administration of drugs to combat this, neither of which were completely reported to the FDA. 3. Incomplete examination of tissues from the experimental animals. 4. Excision of tissue masses likely to be tumors from live animals during the study. 5. Absence of batch records for the mixing of the test substance into the diet of the test animals. 6. Incomplete stability studies for the agent on test. 7. Absence of homogeneity studies for the agent on test. 8. Deficiencies in the methods of chemical assay for the actual DKP that was mixed into the diet of the experimental rats. 9. Problems with the dosage of the DKP that was given to those rats. 10. Problems with the fixation-in-toto and autolysis (decomposition of tissue). 11. Failure to report to the FDA all tissue masses (likely to be tumors) which were found in the experimental rats. 12. Failure to report to the FDA all internal tumors present in the experimental rats, eg., polyps in the uterus (Animal K9MF), ovary neoplasms (Animals H10CF, H19CF, and H7HF) as well as other lesions (Animal D29CF). 13. Inconsistencies between different parts of the report on this study submitted by GD Searle & Co. to the FDA on the precise nature of the lesions manifested by the test rats. 14. Numerous transcription errors in that report. Brain Tumors ------------ The pre-approval studies submitted to the FDA by G.D. Searle were so bad that it would be impossible to determine safety of aspartame from them. However, statistically significant increases in cancer rates in several of the pre-approval experiments are an indication that aspartame may cause cancer. Two pre-approval studies showed an unusually large number of brain tumors in the test animals. Those studies where called, E33/34 and E70. Before discussing these studies in detail, it is useful to see how Dr. Andrian Gross prefaced his discussion of brain tumors in G.D. Searle's pre-approval studies (Gross 1987b, page 1-2): "However, having said all of this, let us assume that in fact those studies were of an acceptable quality; let us pretend that the test animals were actually exposed qualitatively and quantitatively to what G.D. Searle & Co. would have us believe that they were exposed; that there was no post- mortem autolysis [decay] of their carcasses rendering vast numbers of their tissues to a state unsuitable for pathology examination; that the technicians involved in the conduct of those studies were fully trained, competent, and adequately supervised to make observations on those animals prior to their death; that the same was true with respect to the observations made after their death; that in fact those technicians actually made proper such observations; that the proper samples of tissues with grossly observed lesions were in fact collected for additional microscopic examination; that the identity of such tissue specimens corresponded (as they should) to the identity of each animal that was their source, etc. In short, let us make believe in a spirit of Halloween that nothing which was uncovered for the aspartame studies by the FDA investigations of 1975 and 1977 was actually true, i.e., that in fact we are dealing here with studies of an absolutely perfect quality or reliability. Of course, such assumptions belong to the domain of Fantasyland, but nevertheless, let us play this little game for a while." "Under such highly speculative hypothetical conditions, let us now ask again whether aspartame can be viewed as being safe with "reasonable certainty." E33/34 ------ E33/34 was a 104-week study of Charles River CD rats. There were four experimental groups each consisting of 80 rats. Each experiment group was given a different dose of aspartame. The Control group had 119 rats. Twelve brain tumors were found in the experimental rats and zero in the control rats (Gross 1987b, page 2-3): Group Sex Animal # Type of tumor 1 M 83-651 Astrocytoma 1 M 83-745 Astrocytoma 1 F 83-769 Astrocytoma 1 F 83-766 Astrocytoma 2 M 83-837 Astrocytoma 3 M 83-919 Astrocytoma 3 M 83-888 Oligodendroglioma 3 M 83-892 Astrocytoma 3 M 83-895 Astrocytoma 3 F 83-934 Astrocytoma 4 F 84-010 Medulloblastoma 4 F 84-019 Astrocytoma The UAREP pathologists found only 11 brain tumors in the experimental rats and 1 brain tumor in the control rat. Dr. John Olney had this to say about that discrepancy (Olney 1987, page 6-7): "There were other problematic aspects of the brain tumor data. In the pre-1975 records that I reviewed, it was clear that several competent pathologists had carefully examined the original microscopic slides from the first study and agreed that there were 12 brain tumors in the NutraSweet- fed rats and zero brain tumors in the controls. When the FDA conducted a task force investigation of these laboratories in 1975, they singled out these studies for further investigation and ordered that all laboratory records, including microscopic slides etc. be impounded under FDA seal. Several years later when a group of pathologists (UAREP) was sent to authenticate these studies, they could not find the microscopic slides. The UAREP pathologists were finally taken to a laboratory where the slides were not supposed to be and there they found some but not all of the original slides. Clearly they had not been kept under FDA seal and by mysterious coincidence the slides that were finally presented to the UAREP pathologists contained evidence for 11 brain tumors in Nutrasweet-fed rats and 1 tumor in contols. It is important to recognize that if there are zero tumors in the controls, it is very difficult to argue that the tumor incidence in the control and Nutrasweet-fed rats is the same. But if there is 1 tumor in the control group, it is possible with statistical acrobatics to reach the conclusion that the incidence is the same. And, indeed, this is exactly the argument that the manufacturer and the FDA Bureau of Foods pressed at the Public Board of Inquiry. They accepted the finding of 1 brain tumor among the controls even though the authentic record showed zero brain tumors in the controls, then they proceeded to break down the animals into smaller and smaller sub groups according to sex, dose level etc. and finally the 1 brain tumor in one sub group of control animals appeared to be not significantly different from 2 or 3 tumors in each of the experimental sub groups. I seriously doubt whether this method of data analysis would stand the scrutiny of competent disinterested statisticians. Even more seriously I wonder why FDA allows microscopic slides to disappear (while supposedly impounded) and why they do not question the de novo emergence of a brain tumor among the controls when the slides reappear." In addition, the tumors that were found were large, leading one to believe that they were not normal "spontaneous" tumors. As Dr. John Olney stated (Olney 1987, page 7): "Being a neuropathologist, I know that spontaneous brain tumors in laboratory rats are extremely rare. The archival literature documents an incidence not exceeding 0.6%. Since the above incidence in Nutrasweet-red rats is 3.75%, this suggests that Nutrasweet may cause brain tumors and certainly suggests the need for additional in depth research to rule out that possibility. .... "The PBOI panel member who was primarily responsible for reviewing the brain tumor issue was Peter Lampert, M.D., Neuropathologist and chairman of the pathology department at Univ. of Calif. San Diego. Dr. Lampert personally examined the microscopic slides pertaining to the brain tumor studies and told me a year or so after the PBOI report was completed that he had been surprised at the large size of the brain tumors in the Nutrasweet-fed rats. This reinforced his impression that they had been caused by some tumorigenic agent since spontaneous brain tumors are not only rare in laboratory rats but when they do occur they are usually not so large. Dr. Lampert is now deceased; he died in 1986 of cancer. At the time he participated in the PBOI, he was the President of the American Association of Neuropathologists." It is also important to note that there may have been more brain tumors in the E33/34 than reported. As Dr. Adrian Gross discovered (Gross 1987b, page 4-5): "Furthermore, Appendix IV-20 on page 391 of that same UAREP report reveals in the first row of the table on that specific page that GD Searle & Co. or their agents had provided to the subcontracting EPL pathologists, i.e., to those whose report that firm had originally submitted to the FDA:- a) only 8 (or only 10%) of the brain sections for the 80 animals in Group [1]. b) only 7 (or only less than 9%) of the brain sections for the 80 animals in Group [2]. c) only 5 (or only less than 7%) of the brain sections for the 80 animals in Group [3]; and the UAREP were provided with the brain sections of 2 fewer animals than were provided to the EPL. ... This, quite by itself, is sufficiently eloquent on just how G.D. Searle & Co. saw fit to discharge their responsibilities in reporting fully and completely their results of the Two Year Rat Study with aspartame to the FDA;"