ABSTRACT: There has been a statistically significant increase of common primary malignant brain cancers since 1985, and perhaps as early as 1984, according to the National Cancer Institute SEER data. This phenomenon occurred within 1-2 years following licensing of the chemical aspartame for beverages in July 1983. Furthermore, the annual incidence rates of primary brain tumors appear to be increasing. The SEER data also reveal an increased incidence of primary brain lymphoma in 1982- 1984. Others have reported a tripling of the incidence of this condition, previously rare. Again, the licensing of aspartame for "dry" use in July 1981 is relevant. The significance of these associations is underscored by the high incidence of brain tumors in rats after the experimental administration of aspartame.

Food and Drug Administration (FDA) scientists and a Public Board of Inquiry (PBOI) strongly recommended delay in licensure pending further investigation, including repetition of the animal studies, to clarify this matter. To the author's knowledge, these have not been reported. Aspartame containing products are now being consumed by an estimated 200 million persons in over 4,000 products. These data, coupled with an unacceptably large number of aspartame-related seizures reported to the FDA and the writer, appear to warrant an "imminent public health hazard" designation for such products.


The title of this article should disturb seasoned clinicians. It suggests that several major human cancers may be caused or influenced by an additive currently being consumed by more than half the population.

Such an assertion obviously requires epidemiologic and statistical validation, as well as the repetition by corporate-neutral investigators of animal and human studies on which the FDA had relied for licensing products containing this synthesized chemical.
The Rising Incidence of Primary Brain Cancer

The National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) statistics (1) indicate an impressive increase in the age-adjusted incidence rates of primary brain cancer since 1985, and possibly as early as 1984. This phenomenon has been documented in the categories covering all races and both genders.

Disturbing statistically-significant rises in the Estimated Annual Percent Change (EAPC) for brain cancer also were noted in the 1983-1987 period - 1987 being the last year for which complete data are available. For example, SEER Table II-34 contains the five-year trends for all races. The EAPC rose from 2.1 to 8.7 in males, and from 2.1 to 11.7 in females for the time periods 1975-1979 and 1983-1987, respectively.

Although such increases might be attributed to more accurate diagnosis by modern scanning and other diagnostic procedures, three considerations seem to rebut this explanation. First, adequate brain scanning devices were widely available at least one decade ago. Second, the rise in primary brain tumors has been quantitative, and not attributable to changes in nosology. Third, the incidence rates for cancer involving most other systems either remained stable or declined during the 1983-1987 period.

The search for nonoccupational etiologic factors of glioblastoma in adults has proved frustrating (2). Hochberg, Toniolo and Cole (3) were unable to document any significant association with a family history of central nervous system (CNS) malignancies or other neurologic conditions. Exposure to pets, a farm environment, head irradiation, cigarette smoking, alcohol consumption, the intake of cured or smoked meat or fish, or the use of various drugs did not appear to correlate.

The Rising Incidence of Primary Brain Lymphoma

The increasing frequency of primary brain lymphoma of B cell derivation -including reticulum cell sarcoma, microglioma and histiocytic lymphoma - also requires explanation because this subset was previously rare.

Eby et al (4) reported a nearly threefold rise in its incidence among immunologically normal persons in the 1982-1984 SEER data, which they could not explain. Specifically, the rate increased from 2.7 to 7.5 cases per ten million population (p=0.001) in the time periods 1973- 1975 and 1982-1984, respectively. The age-adjusted rise was more striking among women which increased from 4.9 per ten million in 1979- 1981 to 8.9 per ten million in 1982-1984.

Hochberg and Miller (5) reported a tripling of incidence of this tumor in non-immunosuppressed persons during the 5-year interval between 1980 and 1984. Moreover, there was a decrease in the median age of onset by 3.5 years. They projected that the tumor could be the most common neurological neoplasm by 1991 because of its increase both sporadically and in the acquired immunodeficiency syndrome (AIDS) population.

Hardwidge et al (6) noted the increased incidence of primary cerebral lymphoma encountered since 1987 in their neuropathology center in England. They suggested the importance of epidemiological studies to determine any environmental factors that might be implicated.

This phenomenon coincides with two other events: (a) the formal licensing of aspartame in July 1981, and (b) the 3:1 preponderance of women with adverse reactions to aspartame products (7,8). Eby et al (4) suggested other noninfectious environmental exposures as a possible explanation. Although primary brain lymphomas might have a long latency period and result from occupational exposure or other chemical exposures, these investigators regarded occupational exposure to be an unlikely cause in view of the similar increases in incidence among both men and women, particularly in older persons.

Aspartame Consumption

Any attempt to explain this increase in incidence of primary brain cancers must seriously consider the widespread consumption of aspartame products. Aspartame (NutraSweet) was licensed for use as an additive in the Generally Regarded as Safe (GRAS) category by the Federal Drug Administration (FDA), first as a tabletop sweetener in July 1981, and then for "wet" use in beverages in July 1983. Long-term clinical studies in humans were not reported, to our knowledge, before such licensure. Currently, more than 4,000 products containing aspartame are being consumed by over 200 million persons, often in prodigious amounts (7).
Experimental Aspartame-Associated Brain Tumors

An unexpectedly high incidence of primary brain tumors was found in rats experimentally exposed to aspartame during the 1970s (7). Although FDA scientists and others expressed considerable concern, the statutes of limitation on such studies were allowed to expire before regulatory action could be taken. The details were published in the Congressional Record-Senate hearings of May 7 (9) and August 1, 1985 (10), and in a recent text (7).

Park, a Staff Science Advisor for the Office of Health Affairs of the Department of Health and Human Services, concluded an analysis of aspartame safety by a special PBOI relative to brain tumors in aspartame-treated rats (11). He stated that aspartame had not been shown to be safe for the proposed food additive uses. The PBOI accordingly recommended that aspartame should not be approved until additional studies were performed using proper experimental designs.

These studies were never reported, to our knowledge, even though producers of aspartame continue to tout it as "the most thoroughly tested additive in history". If mutagenic, it could be due to the molecule itself, one or more of its three components (phenylalanine, aspartic acid, methyl alcohol), or their breakdown products. The latter include steroisomers of the amino acids and/or multiple metabolites, especially the diketopiperazine derivative (DKP). It is noteworthy that these breakdown products increase during the prolonged storage and exposure to heat to which many aspartame products are exposed (7).

Related Criticisms of FDA Action

Gross (10), a senior FDA pathologist, made a report to the United States Senate hearing held on August 1, 1985. He stated that the cancer- causing potential of aspartame had been established beyond reasonable doubt from a number of indications. He expressed surprise that there was an apparent refusal by the FDA to invoke the Delaney Amendment to the Food, Drug, and Cosmetic Act for this food additive.

Olney, Professor of Psychiatry and Neuropathology at Washington University School of Medicine, wrote to Senator Howard Metzenbaum. He reported that he had discovered a study by the manufacturer of aspartame that had been submitted to the FDA. The FDA apparently accepted it as evidence for the safety of aspartame. However, it had revealed the occurrence of 12 brain tumors in 320 aspartame treated rats. In a similar group of controls there had been no brain tumors.

This investigator noted the fact that spontaneous brain tumors in rats were rare. Although the literature reported an incidence of less than 0.6%, the incidence in these experimental rats was 3.75%, suggesting that much more study of this food additive was required. He expressed the doubt that these data would stand scrutiny by competent disinterested statisticians.

The PBOI panel member who viewed the slides stated to Olney that he "had been surprised at the large size of the brain tumors in the experimental rate". Senator Metzenbaum (9) commented on this matter at the May 7, 1985 Senate hearing: "I do not claim children will develop brain tumors. I do not know that. I do know that the FDA was worried about it. I do know that three of the six FDA scientists advising the FDA Commissioner on final approval were sufficiently worried about it that they were not willing to approve the product. The FDA's own scientists were split on the issue".

Due to concern over neurotoxicity, the Community Nutrition Institute and others filed a petition on August 8, 1983 seeking a public hearing by the FDA concerning its approval of aspartame in liquids because none had been held, and a stay of such approval pending the hearing. The United States Court of Appeals for the District of Columbia Circuit denied both requests (12). The ruling stated that the court was aware of misgivings by scientists and by the PBOI of January 1980 (11), including the Board's plea for further study to establish whether or not a relationship existed between the ingestion of aspartame and brain tumors. The court was under the impression that the manufacturer had not conducted or reported cancer studies on aspartame during the three years following the Board's recommendation, and the FDA had failed to demand them. It made the following pertinent comment (12): "Our scope of review, the exactitude of the fit that we require between the agency's conclusions and the germane facts it investigated, is necessarily deferential. The judiciary is ill-equipped to conduct investigations and analyze facts of the type involved in this case. Because of the agency's expertise and broad discretion in ensuring the safety of food additives, we cannot substitute our judgment for the agency's. The Commission's finding that there were no material issues of fact can be overturned only if an examination of the record discloses that material issues of fact are apparent to any reasonable examiner".

The petitioners were understandably dismayed by the apparent considerable influence of a study sponsored by a Japanese manufacturer (13) upon the FDA Commissioner in overuling the Board's recommendation. This study was not known to the PBOI, and was not made a part of the hearing process. It concluded that neither aspartame nor its diketopiperazine cause brain tumors. However, one needs to examine the presentation of the facts more closely. In the control group, consisting of 59 male and 60 female rats, one female was found to have a mass seen at necropsy which was described as an "atypical astrocytoma". This yielded a brain tumor incidence of 0.8%. There were 3 groups of rats treated with aspartame; one at 1G/Kg, a second at 2G/Kg, a third at 4G/Kg. A 4th group was treated with a 3:1 mixture of aspartame and DKP. In each group there were 60 males and 60 females.

In group 1 there was 1 oligodendroglioma (0.8%): in group 2 there was 1 ependymoma and 1 astrocytoma (1.7%): in group 3, 1 astrocytoma (0.8%), and in group 4, 1 oligodendroglioma (0.8%). There are two facts which require consideration. The first is that the studies were performed by a manufacturer of aspartame. The second is what is meant by an atypical astrocytoma observed in a control animal, yielding an incidence of brain tumor comparable to the experimental groups. Also, it should be noted that there were 2 brain tumors in group 2, doubling the incidence in comparison with the other groups. This experiment should have raised appropriate concern leading to further study.


Presentation of this epidemiologic hypothesis is predicated upon (a) the animal data, and (b) the progressive increase in incidence of several types of primary brain cancer within one or two years after the marketing of aspartame in both "dry" and "wet" products - namely, 1982 data on primary brain lymphoma, and 1984-1987 data on glioblastoma and astrocytoma. In this regard, Altschule (14) aptly noted that a good hypothesis must bring together scattered data or explain a previously unexplained phenomenon.

These phenomena have not been convincingly explained on the basis of improved diagnostic methods or other factors. The latter include decreased immunocompetence associated with HIV infection, other viral disorders, overall trends in non-Hodgkin's lymphoma, occupational exposure (15), or specific chemical exposures (2,4,16). For example, a 25-year registry of patients with non-Hodgkin's lymphoma, sarcoma, leukemia and aplastic anemia ascribed to occupational or home exposure to pentachlorophenol and its dioxin-dibenzufuran contaminants (17) failed to suggest any increase of central nervous system malignancies.

Similarly, Krinke et al (18) could find no statistically significant evidence for an increased incidence of brain tumors in rats exposed to 14 different types of long-term or life-span studies. They were conducted to determine the carcinogenicity of many drugs, agrochemicals, dyestuffs, plastics, and other industrial chemicals. These investigators also confirmed that neurological tumors occurred spontaneously in only 11 among 8960n aging Sprague-Dawley rats.

A recent review of the value of chemical carcinogenicity studies on laboratory animals by Huff, Haseman and Ral (19) is germane. These authorities underscored the value of laboratory rodents as the key surrogates for presently identifying carcinogenic chemicals in humans. They emphasized the prudent public health responsibility of regarding evidence of experimental carcinogenicity as a plausible basis for inferring a carcinogenic risk in man when adequate human data are not available. In the absence of sufficient and reliable human data, they also anticipated social and political debate when chemicals with economic importance are shown to induce experimental cancer.

Brain Cancer in Females

The apparent increase of these tumors in women is relevant. It has been generally acknowledged that malignant brain tumors in adults occur more often among men (5,15,16). Older male rates also develop more spontaneous brain tumors than females (18).

The increase of fatal brain cancer among women is illustrated by the following death rates per 100,000 populations among females of all ages (Silverberg, E., Department of Epidemiology & Statistics, American Cancer Society, Personal Communication, March 20, 1990): 1979--3.4; 1980--3.5; 1981--3.5; 1982--3.6; 1983--3.6; 1984--3.8; 1985--3.8; 1986--3.8; 1987--4.0.

These increases were more striking among white than non-white women. Socioeconomic and cultural factors pertaining to the consumption of "diet" drinks during the early 1980s might partly explain these discrepancies. For example, the increased incidence of gliomas in Jews could also be correlated with increasing socioeconomic class in a Los Angeles County study (20). The threefold higher incidence of severe reaction to aspartame products in females compared with males (7,8) is also germane.

Admittedly, the writer's personal data base concerning this problem remains limited. For example, the 49-year-old wife of a physician developed fatal cerebellar glioblastoma. As a weight-conscious actress and television personality, she had consumed considerable amounts of aspartame products. She had enjoyed good health previously and did not smoke.

Anecdotal information indicates the potential for bone marrow stimulation by aspartame products. Case examples include eosinophilia in an adult female, and three leukemoid episodes in a girl diagnosed as histiocytic leukemia after repeated aspartame rechallenge (21).

There have been other related observation. During 1987, Roelvink et al (22) reported four patients in whom a primary malignant brain or spinal tumor first became manifest during pregnancy. Their interest was prompted by encountering them within a relatively short time period. They were unable to attribute such neoplasms, however, to hormonal influences of gestation.

Pathogenic Insights

Caldecott (23), a member of the Atomic Energy Commission, warned that by far the most mutagenic agents known to man are chemical, not radiation. He suggested that food additives may pose a greater danger than present levels of fallout. Confirmation of these epidemiologic relationships by others might suggest new approaches to the etiology and pathogenesis of primary brain tumors.

Aspartame and its components or its metabolites might activate one or more oncogenes that potentiate or initiate cell mitosis (2), either by direct or indirect effects -- for example, tissue glucopenia, or the influence of uncommon amino acid dextroisomers. These oncogenes include c-cis, c-er B, N-ras, c-myc, the epidermal growth factor-receptor (EGF- R) gene (24), and the trk proto-oncogeny receptor for nerve growth factor (25).

Another mechanism might be the substitution of no-calorie or low- calorie products for conventional foods and beverages, whether as meals or snacks. This can have serious sequelae in the brain, a point emphasized in prior publications concerning the pathogenesis of multiple sclerosis (26), narcolepsy (27), seizures (27), and migraine (28). Under normal circumstances, the brain is almost totally dependent upon glucose for optimum function.

The initial rise in incidence of primary brain lymphoma in 1982, when the consumption of aspartame was much less than after its "wet" use approval during 1983, is of interest. It might be explained by the influence of a less intense biophysiologic, immunologic, viral or toxicologic stimulus than for the more common types of brain tumor. It is noteworthy that the central nervous system (CNS) generally lacks a lymphatic circulation and endogenous accumulations of lymphoid tissue. Hochberg and Miller (5) suggested that an unknown second event in the local site or sites than transforms a clone of the inflammatory cell population into neoplastic cells. Another scenarious might be the migration of activated "homing" cells or molecules from B lymphocytes elsewhere that carry a CNS-specific binding marker.

An unchecked hyperinsulinized state also appears to be operative in the pathogenesis of other tumors. Prostatic hyperplasia and neoplasia (29) provide examples. Phenylalanine and aspartic acid, the amino acid components of aspartame, are known to stimulate insulin release (30-32).


The relationship between aspartame consumption and the development of primary brain cancers in humans requires careful analysis by corporate- neutral investigators. In the event that such a correlation is shown and brain cancer incidence rates continue to rise, the FDA should declare aspartame products and "imminent public health hazard".

Other findings would have related significance. They include:

(a) A disproportionate increase in incidence of glioblastoma, astrocytoma and primary brain lymphoma among young women who, generally speaking, are known to be consuming considerable amounts of aspartame. Brain tumors heretofore have occurred predominantly in middle-aged men (20).

(b) An increase in the incidence rates of gliomas among children whose mothers consumed aspartame during pregnancy. Phenylalanine concentrates at least fourfold on the fetal side of the placenta (7).

Many reasons already exist for such a declaration by the FDA. This agency has received an unprecedented number of volunteered complaints from at least 5,000 consumers concerning severe reactions attributed to aspartame products - including 250 cases of convulsions. The author's registry of 630 aspartame reactors contains more than 100 individuals with grand mal and other seizures. The list of other central nervous system reactions to aspartame products is long (7,8), again indicating that serious brain dysfunction can be induced or aggravated by these products. In this context, I have urged that a formal diagnosis of multiple sclerosis be delayed in persons consuming aspartame products pending their observation for one or several months after abstinence.


1. National Cancer Institutes. Cancer Statistics Review 1973-87. Bethesda, NIH Publication No. 89-2789.

2. Black, P. McL: Brain tumors, New Engl. J. Med 1991: 324:1471-1476.

3. Hochberg F. Toniolo P, Cole P: Nonoccupational risk indicators of glioblasoma in adults, J Neuro-Oncol 1990; 8:55-60.

4. Eby NL, Grufferman S. Flannelly CM. et al: Increasing incidence of primary brain lymphoma in the US. Cancer 1988; 62:22461-22465.

5. Hochberg FH, Miller DC: Primary central nervous system lymphoma. J Neurogurg 1988: 68:835-853.

6. Hardwidge C, Diengdoh JV, Husfaud D. Nash JRG: Review: Primary cerebral lymphoma - a clinico-pathological study. Clin Neuropath 1990; 9:217-223.

7. Roberts HJ. Aspartame (NutraSweet): Is It Safe? Philadelphia The Charles Press, 1989.

8. Roberts HJ. Reactions attributed to aspartame-containing products: 551 cases. J Appl Nutr 1988: 59:85-94.

9. Congressional Recond-Senate. Saccharin Study and Labeling Act Amendments of 1985. May 7, 1985, pp.S5489-5516.

10. Congressional Record-Senate. Aspartame Safety Act of 1985. August 1, 1985, pp.S10820-10847.

11. U.S. Government Printing Office. Aspartame; Availability of Board of Inquiry Decision. Fed Reg 1980; 45:69558.

12. Community Nutrition Institute, et al. v. Dr. Mark Novitch, Acting Commissioner, Food and Drug Administration, United States Court of Appeals for the District of Columbia Circuit, No. 84-1153 and No. 84- 5253 (D.C. Civil Action No. 83-03846, decided September 24, 1985).

13. Ishu H.: Incidence of brain tumors in rats fed aspartame. Toxicol Letters 1981: 7:433-437.

14. Altschule MD: Hypothesis in medicine. Med Sci 1966; 17:94

15. Salcman M: The morbidity and mortality of brain tumors, Neurol Clin 1985; 3:229-257.

16. Cole GC, Wilkins PR, West RR: An epidemiological survey of primary tumors of the brain and spinal cord in South East Wales. Br J Neurogurg 1989; 3:487-493.

17. Roberts HJ: Pentachlorophenol-associated aplastic anemia, red cell aplasis, leukemia and other blood disorders. J Florida M A 1990; 77:86- 90.

18. Krinke G. Naylor DC. Schnid S. Frohlich E. Schnider K: The incidence of naturally-occurring primary brain tumors in the laboratory rat. J Comp Path 1985; 95:175-192.

19. Huff J. Haseman J. Rall D: Scientific concepts, value, and significance of chemical carcinogenesis studies. Ann Rev Pharmacol Toxicol 1991; 31:621-652.

20. Preston-Martin S: Descriptie epidemiology of primary tumors of the brain, crainial nerves and cranial meninges in Los Angeles County. Neuropidem 1989; 8:283-295.

21. Roberts HJ: Aspartame, tryptophan, and other amino acids as potential hazardous experiments. South M J 1990; 83:1110-1111.

22. Roelvink NCA. Kamphort W, van Alphen, HAM. Rao BR: Pregnancy-related primary brain and spinal tumors. Arch Neurol 1987: 44:209-215.

23. Caldecott R: Cited by Sc Newsletter 1961: November 18.

24. Hoi Sang U. Kelley PY, Hatton JE, Shew JY: Proto-oncogene abnormalities and their relationship to tumorigenicity in some human gliblastomas. J Neurosurg 1989: 71:83-90.

25. Kaplan DR, et al: The trk proto-onogeny product: A signal transducing receptor for nerve growth factor. Science 1991:252-554-558.

26. Roberts HJ: An inquiry into the pathogenesis, rational treatment and prevention of multiple sclerosis, with emphasis upon the combined role of diabetogenic hyper-insulinism and recurrent edema. J Am Geriat Soc 1964; 12:926-976.

27. Roberts HJ: The syndrome of narcolepsy and diabetognenic ("function") hyperinsulinism, with special reference to obesity, diabetes, idiopathic edema, cerebral dysrhythmias and multiple sclerosis (200 patients), J Am Geriat Soc 1964; 12:926-976.

28. Roberts HJ: Migraine and related vascular headaches due to diabetogenic hyperinsulinism: Observations of pathogenesis and rational treatment in 421 patients. Headache 1967; 7:41-62.

29. Roberts HJ: Pathogenesis of prostatic hyperlasia and neoplasia. Geriat 1967; 22:85-92.

30. Floyd JC Jr., Fajans S. Conn JW, Knopf RF, and Rull J: Stimulation of insulin secretion by amino acids. J Clin Invest 45:1487-1502.

31. Floyd JC Fajans SS, Pek S. et al: Synergistic effect of certain amino acid pairs upon insulin secretion in man. Diabetes 1970; 19:102- 108.

32. Reitano G. Distefano G. Viro R. et al: Effect of priming of amino acids on insulin and growth hormone respone in the premature infant. Diabetes 1978.

H. J. Roberts, MD (H.J. Roberts MD is Director, Palm Beach Institute for Medical Research. He is Senior Active Staff, St. Mary's Hospital and Good Samaritan Hospital, West Palm Beach. He is author of six texts and was selected the "The Best Doctors in the U.S." Address correspondence to H. J. Roberts MD, Palm Beach Institute for Medical Research, 6708 Pamela Lane, West Palm Beach, FL 33405 FAX 561-547-8008