Artificial sweeteners—do they bear a carcinogenic risk?
M. R. Weihrauch* & V. Diehl
Department of Internal Medicine I of the University of Cologne, Cologne, Germany
Received 12 June 2003; accepted 6 January 2004
Artificial sweeteners are added to a wide variety of food, drinks, drugs and hygiene products. Since
their introduction, the mass media have reported about potential cancer risks, which has contributed
to undermine the public’s sense of security. It can be assumed that every citizen of Western
countries uses artificial sweeteners, knowingly or not. A cancer-inducing activity of one of these
substances would mean a health risk to an entire population. We performed several PubMed
searches of the National Library of Medicine for articles in English about artificial sweeteners.
These articles included ‘first generation’ sweeteners such as saccharin, cyclamate and aspartame, as
well as ‘new generation’ sweeteners such as acesulfame-K, sucralose, alitame and neotame. Epide-
miological studies in humans did not find the bladder cancer-inducing effects of saccharin and
cyclamate that had been reported from animal studies in rats. Despite some rather unscientific
assumptions, there is no evidence that aspartame is carcinogenic. Case–control studies showed an
elevated relative risk of 1.3 for heavy artificial sweetener use (no specific substances specified) of
>1.7 g/day. For new generation sweeteners, it is too early to establish any epidemiological evidence
about possible carcinogenic risks. As many artificial sweeteners are combined in today’s products,
the carcinogenic risk of a single substance is difficult to assess. However, according to the current
literature, the possible risk of artificial sweeteners to induce cancer seems to be negligible.
Key words: aspartame, cancer, cyclamate, saccharin, sweeteners
The fondness of humans for sweet foods is inborn: studies
have proved a preference for sweet-tasting nutrition in new-
borns . Therefore, mankind has always added sweet sub-
stances to their food. The first recorded sweetener was honey,
which was used in the ancient cultures of Greece and China
. Honey was later replaced by saccharose, common sugar,
which was originally obtained from sugar cane. During the
World Wars, sugar beets were the major source of saccharose.
The first artificial sweetener was saccharin, which was syn-
thesized in 1879 by Remsen and Fahlberg. It was well
accepted during World Wars I and II because of its low
production costs and the shortcoming of regular sugar . As
economies recovered and living standards increased after the
wars, sugar became affordable. With a growing candy and fast
food industry, obesity increased in the Western societies, as
we know today from our daily clinical practice. Since the
1950s, the reasons for using saccharin have shifted from cost
to calorie reduction. A profitable market for calorie-reduced
‘diet products’ evolved, in which sugar was substituted or
supplemented with artificial sweeteners. However, saccharin
was known not only for its extreme sweetness, but also for its
bitter aftertaste, so that there was a growing need for new
improved taste, calorie-reduced substances. A breakthrough in
the artificial sweetener industry was achieved with cyclamate
in the 1950s, which provided a better taste than saccharin. In
addition, it blended very well with saccharin. Both substances
were mixed together with other additives and were sold as
‘Sweet’n’Low’, which became a huge success in the USA.
Because of its characteristics, cyclamate was not only used in
tablet or liquid form (‘table top sweetener’), but also proved
suitable for sweetening soft drinks.
The first insecurity shook the artificial sweetener market in
1970, when the Food and Drug Administration (FDA) banned
cyclamate from all dietary foods and fruits in the USA. The
FDA had become suspicious of induced cancer in experimen-
tal animals . In all other countries, cyclamate is still used
today, especially in combination with other sweeteners. The
next step in the development of artificial sweeteners was the
approval of aspartame in 1981 and its marketing as ‘Nutra-
Sweet’. For the first time, dairy products such as yogurts were
calorie-reduced and could be sold with the prefixes ‘diet’ or
‘light’ . The first three substances, saccharin, cyclamate and
aspartame, are referred to as ‘first generation sweeteners’.
These were followed by new generation or second genera-
tion sweeteners such as acesulfame-K, sucralose, alitame
*Correspondence to: Dr M. R. Weihrauch, Immunologisches Labor Haus 16, Uniklinik Koeln, Joseph-Stelzmann-Strasse 9, 50924 Koeln, Germany. Tel: +49-221-4784488; Fax: +49-221-4785912; E-mail: email@example.com
Annals of Oncology 15: 1460–1465, 2004
q 2004 European Society for Medical Oncology
and neotame, which have quite different key market areas, as
shown in Table 1 (from Lindley ). However, even the new
sweeteners have similar limitations to the older ones. The
taste is often accompanied by a bitter and metallic aftertaste
and does not provide the ‘realistic’ and ‘voluminous’ mouth-
feel of regular sugar. The combination of many, synergic arti-
ficial sweeteners has led to an improvement of the quality of
sweetened products. In soft drinks, a combination of acesul-
fame-K, aspartame and others has found broad application (as
shown in Figure 1).
Today, many people have mixed feelings when using artifi-
cial sweeteners, because they associate news about possible
cancer risks with these substances. Particularly in the 1980s,
when many sweeteners were newly synthesized and intro-
duced to the food market, the public press reported on the
ostensible carcinogenic effects of sweeteners. News articles
frequently lacked a fundamental scientific background or were
inattentively investigated, and added to a public insecurity.
Even some of the scientific publications in reliable medical
journals, which caught media attention, were not well
researched, and ignored common statistical knowledge as
described later. During the last decade, the cancer-inducing
effect of artificial sweeteners has not been discussed as fre-
quently as in earlier years, although some of the long-term
studies about saccharin and cyclamate have recently been
completed and published.
Several PubMed searches of the National Library of Medicine were per-
formed. Relevant preclinical, clinical and epidemiological studies on artifi-
cial sweeteners and possible health risks were identified. All searches
focused on English language journals only, but were not limited to a cer-
tain period of time. Where appropriate, cited references of articles were
also reviewed. Key words for the PubMed search included ‘artificial
sweetener’, ‘cancer’ and ‘carcinogenic’, as well as all artificial sweetener
names. To present an overview, the studies were sorted by the investigated
artificial sweetener, and will be discussed separately.
Saccharin is the oldest chemical sugar substitute and the best
researched of all sweeteners. More than 50 studies have been
published about saccharin in laboratory rats. Approximately
20 study groups analyzed the effect of saccharin in one gene-
ration of rats, which were exposed to high doses of saccharin
for at least 1.5 years. Usually, the doses administered included
a high concentration of 5% of the various forms of saccharin
in the diet, and in several cases, animals started the study at
6 weeks of age. Except for one study, none of the 20 groups
found significantly more neoplasias in the saccharin-fed ani-
mals than in controls. The positive study reported an increased
incidence of bladder cancers . However, ACI rats were
used in this trial, which are frequently infected with the blad-
der parasite Trichosomoides crassicanda and are therefore sus-
ceptible to saccharin-induced bladder cell proliferation .
After many ‘one generation’ studies, ‘two generation’ stu-
dies were conducted feeding the parent (F0) and the following
Table 1. Current artificial sweeteners and their key market areas
(taken from Lindley )
Sweetener Key market areas
Acesulfame-K North America, Europe and Asia
Alitame Oceania, South/Central America
Aspartame North America, Europe and Asia
Cyclamate Europe and Asia
Neohesperidine DC Europe and Japan
Saccharin Asia, Europe and USA
Sucralose North America
Thaumatin Europe and Asia
Figure 1. Two product labels of a diet soda, taken from the USA (A) and Germany (B). In the USA, only aspartame is used in this soda, whereas the
same product is sold in Germany with an artificial sweetener combination of cyclamate, acesulfame-K and aspartame.
generation (F1) with saccharin. In these studies, an increased
risk for bladder cancer could be consistently proven for the F1 generation. Taylor et al.  showed that especially male rats
developed bladder tumors in up to 30% of all animals at a dose
of 7.5% saccharin of their diet. Later trials, the largest with
2500 F1 generation rats , found that the risk for bladder can-
cers increases with a saccharin concentration of 4%. Because
of these results, saccharin was prohibited in Canada. In the
USA, since 1981, saccharin-containing products have had to be
labeled with a warning that saccharin can cause cancer in lab-
oratory animals. However, the National Institute for Environ-
mental Health Sciences, which issues a biannual report,
removed saccharin as a potential cancer-causing agent, because
it could be shown that the cancer-inducing mechanisms in rats
do not apply in humans. Ascorbic acid (vitamin C), when fed in
similar doses as saccharin, could also cause bladder cancer in
rats; this could be prevented by adding prophylactic ammonium
chloride. Rodents have a high urine osmolarity, which
enhances the precipitation of calcium phosphate-containing
crystals, which are cytotoxic to the superficial layer of the
bladder epithelium, leading to regenerative hyperplasia and
Takayama et al.  in 1998 published a long-term study
on 20 monkeys, of three species, that were treated with
sodium saccharin (25 mg in the diet/kg daily for 5 days a
week) for up to 24 years. Sixteen monkeys served as controls.
None of the animals developed bladder cancer or urothelial
proliferations. The study was criticized for the small number
of monkeys and for the relatively low dosage of saccharin,
which corresponds to a daily diet-soda consumption of 1.5 l in
a 70-kg person . The first studies about the cancerogenous
risk of saccharin in humans were only of descriptive design.
In the UK, a longitudinal study did not show an increase in
bladder cancer incidence during World War II, when saccharin
consumption was high . The same authors analyzed
19 709 death certificates from the UK between 1966 and 1972
and compared the bladder cancer mortality between diabetics,
who used artificial sweeteners more frequently, and non-
diabetics. They did not find any significant differences
between the groups . A Danish study could not detect an
increase of bladder cancer mortality in people aged up to
30 years old, who were born between 1941 and 1945, when
saccharin use was higher than in the years before and after
. The authors concluded that an exposition to saccharin
in utero does not increase bladder cancer incidence during the
first three decades of life. A case–control study from China
published in 1997 analyzed different risk factors in 254
bladder cancer patients and 254 controls . They reported
an odd ratio of 3.9 for bladder cancer in patients with frequent
saccharin use of at least 19 consumptions per year for at least
15 years. However, this study has to be critically assessed as
to its worth, because it was unable to identify the elevated risk
of bladder cancer in smokers, which was proven by other
large trials [16–18]. There are many case–control studies
from the USA and Europe about bladder cancer risk factors,
which not only investigate saccharin as a possible cause, but
also the use of artificial sweeteners in general. Therefore, they
will be discussed later in this review.
Sodium cyclamate entered the US market after its FDA
approval in 1951 . Owing to a study by Wagner in 1970
, which found an increased incidence of bladder carci-
nomas in rats, the use of cyclamate was prohibited in several
countries, including the USA and UK. Further evaluations by
the Cancer Assessment Committee of the Center for Food
Safety and Applied Nutrition of the FDA, by the Scientific
Committee for Foods of the European Union and by the WHO
concluded that cyclamate is not a carcinogen, and readmitted
it to the food market .
Cyclamate is converted to a metabolite, cyclohexylamine,
which has been reported to be rather toxic . In experi-
ments with rats and dogs, cyclohexylamine caused testicu-
lar atrophy and impairment of spermatogenesis [23–26].
Takayama et al.  conducted a long-term toxicity study
with cyclamate in non-human primates, as described before
for saccharin. Twenty-one monkeys were fed either 100 or
500 mg/kg cyclamate per day over 24 years, and compared
with 16 controls. A dose of 500 mg/kg corresponds to � 30 calorie-reduced drinks. In 1994, after 24 years, the remaining
14 cyclamate and 16 control monkeys were killed and
autopsied. In the cyclamate group, three animals showed
malignancies, whereas none were found in the controls. The
tumor stages and histologies of the cancers were a metastatic
adenocarcinoma of the colon (500 mg/kg), a metastatic hepa-
tocellular carcinoma (500 mg/kg) and a local well-differen-
tiated papillary adenocarcinoma of the prostate (100 mg/kg).
In addition, three benign tumors were found in the treatment
group, an adenoma of the thyroid gland and two leiomyoma
of the uterus, whereas the control group remained free of
tumors. The authors concluded that there is no evidence for
carcinogenicity of sodium cyclamate, because the tumors in
the treatment groups were of different histologies and the
tumors occurred at a rate frequently observed in monkeys. In
particular, no bladder carcinomas were reported as in the rat
study, which had led to the ban of cyclamates. The trial of
Takayama et al.  was critcized for the small number of
animals, which was too low to reach any significance or to
confirm a negative result . In addition, the critics claimed
that the tumor incidence in the treatment group (33%) was
higher than the spontaneous neoplasia rate in respective
monkey strains, and unlikely to be a chance occurrence. There
are no descriptive or case–control studies of cyclamate in
humans, because it was approved after saccharin, and products
contained mixtures of both artificial sweeteners. It has to be
assumed that most consumers have used both saccharin and
cyclamate since the introduction of cyclamate.
Aspartame entered the market in 1981 as the third artificial
sweetener, and was free of any suspicions regarding
carcinogenicity. Animal studies showed that aspartame does
not have any cancer-inducing effects, even in very high doses
[28, 29]. DNA repair assays for the evaluation of genotoxicity
of substances did not show any DNA-damaging properties for
aspartame, cyclamate, saccharin, acesulfame-K or sucralose
. Fifteen years after the approval of aspartame, the Journal
of Neuropathology and Experimental Neurology published an
article by Olney et al.  with the title ‘Increasing brain
tumor rates: Is there a link to aspartame?’, which received tre-
mendous attention from the mass media, as well as the scienti-
fic community. The authors hypothesized that the increasing
rate of brain tumors in humans since 1980 could possibly be
explained by the introduction of aspartame. They supported
their hypothesis with an FDA trial in 320 Sprague–Dawley
rats. Twelve rats developed malignant brain tumors after
receiving an aspartame-containing feed for 2 years . They
argued that another trial had shown that the aspartame
molecule acquires mutagenic activity when nitrosated .
The publication of Olney et al.  led to heavy criticism of
the scientific community, whereas the laymen press suggested
abstaining from aspartame-sweetened products . In an
editorial, Ross  demonstrated the weaknesses of the Olney
study. He explained that Olney et al.  linked two events
that incidentally occurred during roughly the same time
period: the increase of brain tumors and the introduction of
aspartame. This correlation is not admissible in epidemiology,
and is called ‘ecological fallacy’. There was no information
available regarding whether the individuals who developed
brain tumors consumed aspartame. As Ross states, one might
also invoke home computers, VCR usage or the depletion of
the ozone layer to argue trends in brain tumors. In addition,
the introduction of aspartame and the rising brain tumor rate
occurred almost simultaneously. For the development of brain
tumors, a certain latency would have been required. The study
that showed an increased brain tumor incidence in aspartame-
fed rats, which gave rise to the argument of Olney et al.,
could not be confirmed by later trials . Ross 
suggested evaluating the link between aspartame exposure and
brain tumors in a case–control or cohort study.
Indeed, a case–control study on aspartame consumption was
conducted in children with brain tumors . The study group
compared 56 patients with 94 controls in terms of aspartame
use and other known and suspected risk factors, such as
maternal vitamin consumption, cured meat intake, passive
smoke exposure, X-ray exposure and family history of brain
cancer. They observed no elevated brain tumor risk to the child
from maternal consumption of aspartame during pregnany, nor
did they find elevated risks during any trimester of pregnancy
or during breast-feeding. After the questionable study of Olney
et al. , Schwartz  wrote a letter to the Western Journal
of Medicine, which was published in 1999. Schwartz hypoth-
esized a link between aspartame and the increase of breast
cancer. He argued that aspartame is partly metabolized to
methanol, which itself is converted to formaldehyde, which
accumulates within cells and induces cancer . In the same
issue of the journal, Tichopoulos  responded to the letter.
He explained that the increase of the breast cancer rate
occurred before the introduction of aspartame, and has been
declining during the last few years [39, 41]. He concluded that
Schwartz also succumbed to an ecological fallacy.
New generation sweeteners
Except for the toxicological animal data required for FDA
approval, there are no larger studies that investigate the poten-
tially hazardous effects of second generation sweeteners. None
of the substances such as acesulfame-K, neohisperidine,
alitame or sucralose has been suspected to cause cancer or to
Epidemiological studies in humans
After cyclamate and aspartame had entered the food market,
diseases such as bladder cancer could not be linked to the
consumption of saccharin alone, because most consumers used
different artificial sweeteners. Also, substances were mixed in
food products to improve the taste. Therefore, most epidemio-
logical studies in humans relate to sweetener consumption in
general, and not to single substances. The most important pub-
lications in this field are case–control studies. Many of these
trials were conducted with small patient groups of up to 350
bladder cancer patients in the years 1965–1986 [42–46].
None of them showed a significantly increased risk of bladder
carcinoma for artificial sweetener use. A study from the UK
 included 622 existing and 219 new cases of bladder can-
cer, and matched them to hospital-based controls for age and
sex. The study group found an increased relative risk (RR) for
non-smoking males [RR 2.2; 95% confidence interval (CI)
1.3–3.8] and non-smoking females (RR 1.6; 95% CI 0.8–
3.2), but not for smokers. Sweetener use was defined as regu-
lar use for over 1 year at least 5 years prior to diagnosis.
The most recent case–control study was published by
Sturgeon et al.  with 1860 bladder cancer patients and
3934 controls. They examined different factors, among which
were smoking, urinary tract infection, coffee consumption, his-
tory of cystolithiasis and genetic predisposition for the risk of
inducing bladder cancer. Artificial sweetener consumption was
classified as ‘low’ (<1680 mg per day) or ‘heavy’ (>1680 mg
per day). The risk of bladder cancer was not associated with
low sweetener use in 966 patients and 3410 controls. Heavy
sweetener consumption (31 patients, 78 controls) led to a sig-
nificantly increased RR of 1.3 (95% CI 0.9–2.1). Also, high
coffee consumption of >50 cups per week was associated with
an RR of 1.4, and therefore was comparable to heavy artificial
sweetener use or the history of one to two urinary tract infec-
tions (RR 1.3). The authors scrutinized the bladder cancer
histologies. Heavy artificial sweetener use was associated with
higher grade, poorly differentiated tumors.
Owing to the existing studies, the following statements can be
made about the carcinogenic potential of artificial sweeteners.
Saccharin induces bladder cancer in rats, when fed in high
doses. However, rodents react to most sodium salts, such as
sodium ascorbate, with urothel proliferation and neoplasia of
Heavy artificial sweetener use (>1680 mg per day) leads to
an increased relative risk of 1.3 for bladder cancer in humans.
A more precise determination of the exact agents is not
possible, because many artificial sweeteners are combined in
current food products.
Despite unscientific articles in the mass media and scientific
press, there is no evidence that the artificial sweetener aspar-
tame bears a carcinogenic risk.
The approvals of new generation sweeteners (acesulfame-K,
sucralose, alitame and neotame) are too recent to establish any
epidemiological evidence about possible carcinogenic risks.
1. Maone TR, Mattes RD, Bernbaum JC, Beauchamp GK. A new
method for delivering a taste without fluids to preterm and term
infants. Dev Psychobiol 1990; 23: 179–191.
2. Bright G. Low-calorie sweeteners—from molecules to mass markets.
World Rev Nutr Diet 1999; 85: 3–9.
3. Anonymous. FDA extends ban on cyclamates. Science 1970; 169: 962.
4. Lindley MG. New developments in low-calorie sweeteners. World
Rev Nutr Diet 1999; 85: 44–51.
5. Fukushima S, Arai M, Nakanowatari J et al. Differences in suscepti-
bility to sodium saccharin among various strains of rats and other
animal species. Gann 1983; 74: 8–20.
6. Whysner J, Williams GM. Saccharin mechanistic data and risk
assessment: urine composition, enhanced cell proliferation, and tumor
promotion. Pharmacol Ther 1996; 71: 225–252.
7. Taylor JM, Weinberger MA, Friedman L. Chronic toxicity and carci-
nogenicity to the urinary bladder of sodium saccharin in the in utero-
exposed rat. Toxicol Appl Pharmacol 1980; 54: 57–75.
8. Squire RA. Histopathological evaluation of rat urinary bladders from
the IRDC two-generation bioassay of sodium saccharin. Food Chem
Toxicol 1985; 23: 491–497.
9. Cohen SM, Anderson TA, de Oliveira LM, Arnold LL. Tumori-
genicity of sodium ascorbate in male rats. Cancer Res 1998; 58:
10. Takayama S, Sieber SM, Adamson RH et al. Long-term feeding of
sodium saccharin to nonhuman primates: implications for urinary
tract cancer. J Natl Cancer Inst 1998; 90: 19–25.
11. Jacobson MF, Farber E, Clapp R. Re: Long-term feeding of sodium
saccharin to nonhuman primates: implications for urinary tract cancer.
J Natl Cancer Inst 1998; 90: 934–936.
12. Armstrong B, Doll R. Bladder cancer mortality in England and Wales
in relation to cigarette smoking and saccharin consumption. Br J Prev
Soc Med 1974; 28: 233–240.
13. Armstrong B, Doll R. Bladder cancer mortality in diabetics in relation
to saccharin consumption and smoking habits. Br J Prev Soc Med
1975; 29: 73–81.
14. Jensen OM, Kamby C. Intra-uterine exposure to saccharin and risk of
bladder cancer in man. Int J Cancer 1982; 29: 507–509.
15. Yu Y, Hu J, Wang PP et al. Risk factors for bladder cancer: a case-
control study in northeast China [published erratum appears in Eur J
Cancer Prev 1998; 7: 171]. Eur J Cancer Prev 1997; 6: 363–369.
16. Brennan P, Bogillot O, Cordier S et al. Cigarette smoking and bladder
cancer in men: a pooled analysis of 11 case-control studies. Int J
Cancer 2000; 86: 289–294.
17. Mikhailidis DP, Ganotakis ES, Papadakis JA, Jeremy JY. Smoking
and urological disease. J R Soc Health 1998; 118: 210–212.
18. Teschke K, van Zwieten L. Perceptions of the causes of bladder
cancer, nasal cancer, and mesothelioma among cases and population
controls. Appl Occup Environ Hyg 1999; 14: 819–826.
19. Bopp BA, Sonders RC, Kesterson JW. Toxicological aspects of cycla-
mate and cyclohexylamine. Crit Rev Toxicol 1986; 16: 213–306.
20. Wagner MW. Cyclamate acceptance. Science 1970; 168: 1605.
21. Takayama S, Renwick AG, Johansson SL et al. Long-term toxicity
and carcinogenicity study of cyclamate in nonhuman primates. Toxi-
col Sci 2000; 53: 33–39.
22. Renwick AG. The metabolism of intense sweeteners. Xenobiotica
1986; 16: 1057–1071.
23. Gaunt IF, Sharratt M, Grasso P et al. Short-term toxicity of cyclo-
hexylamine hydrochloride in the rat. Food Cosmet Toxicol 1974; 12:
24. James RW, Heywood R, Crook D. Testicular responses of rats and
dogs to cyclohexylamine overdosage. Food Cosmet Toxicol 1981; 19:
25. Roberts A, Renwick AG. The pharmacokinetics and tissue concen-
trations of cyclohexylamine in rats and mice. Toxicol Appl Pharmacol
1989; 98: 230–242.
26. Roberts A, Renwick AG, Ford G et al. The metabolism and testicular
toxicity of cyclohexylamine in rats and mice during chronic dietary
administration. Toxicol Appl Pharmacol 1989; 98: 216–229.
27. Huff J, Tomatis L. Re: Long-term toxicity and carcinogenicity study
of cyclamate in nonhuman primates (Takayama.Toxicol Sci. 53, 33–
39). Toxicol Sci 2000; 57: 186.
28. Hagiwara A, Fukushima S, Kitaori M et al. Effects of three sweet-
eners on rat urinary bladder carcinogenesis initiated by N-butyl-N-
(4-hydroxybutyl)-nitrosamine. Gann 1984; 75: 763–768.
29. Ishii H. Incidence of brain tumors in rats fed aspartame. Toxicol Lett
1981; 7: 433–437.
30. Jeffrey AM, Williams GM. Lack of DNA-damaging activity of five
non-nutritive sweeteners in the rat hepatocyte/DNA repair assay.
Food Chem Toxicol 2000; 38: 335–338.
31. Olney JW, Farber NB, Spitznagel E, Robins LN. Increasing brain
tumor rates: is there a link to aspartame? J Neuropathol Exp Neurol
1996; 55: 1115–1123.
32. Study E33-34 in Master file 134 on aspartame. On file at the FDA
Hearing Clerk’s Office 2001.
33. Shephard SE, Wakabayashi K, Nagao M. Mutagenic activity of
peptides and the artificial sweetener aspartame after nitrosation. Food
Chem Toxicol 1993; 31: 323–329.
34. Schwabe U. Health risk caused by sugar substitutes. Internist (Berl)
1998; 39: 869.
35. Ross JA. Brain tumors and artificial sweeteners? A lesson on not
getting soured on epidemiology. Med Pediatr Oncol 1998; 30: 7–8.
36. Anonymous. Aspartame. Review of safety issues. Council on Scienti-
fic Affairs. JAMA 1985; 254: 400–402.
37. Gurney JG, Pogoda JM, Holly EA. Aspartame consumption in
relation to childhood brain tumor risk: results from a case-control
study [published erratum appears in J Natl Cancer Inst 1997; 89:
1460]. Natl Cancer Inst 1997; 89: 1072–1074.
38. Schwartz GR. Aspartame and breast and other cancers [published
erratum appears in West J Med 2000; 172: 233]. West J Med 1999;
39. Adami HO, Signorello LB, Trichopoulos D. Towards an understand-
ing of breast cancer etiology. Semin Cancer Biol 1998; 8: 255–262.
40. Trichopoulos D. Response. West J Med 1999; 171: 301.
41. Chu KC, Tarone RE, Kessler LG et al. Recent trends in U.S. breast
cancer incidence, survival, and mortality rates. J Natl Cancer Inst
1996; 88: 1571–1579.
42. Simon D, Yen S, Cole P. Coffee drinking and cancer of the lower
urinary tract. J Natl Cancer Inst 1975; 54: 587–591.
43. Wynder EL, Goldsmith R. The epidemiology of bladder cancer: a
second look. Cancer 1977; 40: 1246–1268.
44. Wynder EL, Stellman SD. Artificial sweetener use and bladder
cancer: a case-control study. Science 1980; 207: 1214–1216.
45. Morrison AS, Buring JE. Artificial sweeteners and cancer of the
lower urinary tract. N Engl J Med 1980; 302: 537–541.
46. Piper JM, Matanoski GM, Tonascia J. Bladder cancer in young
women. Am J Epidemiol 1986; 123: 1033–1042.
47. Cartwright RA, Adib R, Glashan R, Gray BK. The epidemiology of
bladder cancer in West Yorkshire. A preliminary report on non-occu-
pational aetiologies. Carcinogenesis 1981; 2: 343–347.
48. Sturgeon SR, Hartge P, Silverman DT et al. Associations between
bladder cancer risk factors and tumor stage and grade at diagnosis.
Epidemiology 1994; 5: 218–225.