Home » Aspartame » Level 3 » Question 4


4. Can aspartame affect human health?

  • 4.1 Can aspartame cause cancer or induce genetic damage?
    • 4.1.1 Laboratory studies
    • 4.1.2 Epidemiological data
  • 4.2 Can aspartame affect reproduction and development?
  • 4.3 Can aspartame produce neurological disorders?
  • 4.4 Can aspartame affect behaviour, cognition or mood?
  • 4.5 Does aspartame cause headaches?
  • 4.6 Does aspartame trigger epileptic seizures?
  • 4.7 Can aspartame cause other health effects?

4.1 Can aspartame cause cancer or induce genetic damage?

    • 4.1.1 Laboratory studies
    • 4.1.2 Epidemiological data

4.1.1 Laboratory studies

The source document for this Digest states:

Genotoxicity and carcinogenicity

The available mutagenicity and long-term carcinogenicity studies on aspartame were recently reviewed by AFSSA (2002). AFSSA noted that:

"Aspartame is not genotoxic in a reverse mutation test on S. typhimurium, in two chromosome aberration tests in vivo on somatic cells and in Rodent dominant lethal test on germ cells (JECFA, 1980). Recently, two studies have confirmed the absence of clastogenic potential (Durnev et al., 1995; Mukhopadhyay et al., 2000) of the compound."

The AFSSA report also noted:

"Trocho et al., (1998) demonstrated that aspartame, radio-labelled on the methanol, induced in the liver stable DNA and protein adducts. According to these authors, the accumulation of these adducts after repeated administration of aspartame could pose problems of toxicity and carcinogenicity in the long term. Besides the fact that aspartame at high doses has never induced liver cancer in rats, Trocho's studies did not identify the radioactivity found in the proteins and DNA. Consequently, the formation of adducts of formaldehyde on the proteins and nucleic acids from aspartame, in vivo, remains to be proved (Tephly, 1999)."

As regards the long-term studies, the AFSSA report noted that:

"In a carcinogenicity study on CD-1 mice (FDA, FR 1981), aspartame administered in feed at doses of 1, 2 and 4 g/kg bw/day for 110 weeks, showed no carcinogenic potential."

"Three carcinogenicity studies were conducted in Sprague Dawley and Wistar rats. In the first study (1973), post-weaning Sprague Dawley rats were fed doses of aspartame corresponding to 1, 2, 4, 6/8 g/kg bw/day for 104 weeks (6/8 i.e. dose of 6 was increased during the study to 8 g/kg bw/day). In the second study (1974), male and female Sprague Dawley rats, from a two-generation study, were exposed during gestation, lactation and after weaning for 104 weeks, to doses of 0, 2 and 4 g/kg bw/day in their food. The results of these two studies have been widely discussed by the scientific community and the regulatory authorities (FDA). In the first study, the incidence of brain tumours in the treated animals was higher than in the control animals but without any dose-response relationship. In contrast, in the second study the incidence of tumours in the treated rats was lower than in the control group. For these reasons, a third study was conducted under conditions of Good Laboratory Practice in order to ensure the reliability of the experimental data. In this third study (Ishii, 1981), groups of male and female Wistar rats were given doses of aspartame of 0, 1, 2, 4 g/kg bw/day for 104 weeks. Under these conditions, aspartame did not cause any increase in the incidence of brain tumours."

AFSSA concluded as follows on carcinogenicity:

"Taking into account all the studies that have been conducted, the frequency of spontaneous tumours in laboratory rats, the types of tumours observed and the absence of a dose-response relationship, it was concluded that aspartame had no carcinogenic potential on the brain in experimental animals (FDA FR, 1981-1984; Koestner, 1984; Cornell et al., 1984; Flamm, 1997)."

Source & ©: EC-SCF   Update on the Safety of Aspartame
AFSSA  Aspartame Assessment Report chap 4

Toxicity of diketopiperazine in terms of effects on the nervous system

The toxicity of diketopiperazine, a degradation product of aspartame, has been studied in laboratory animals. This substance is neither genotoxic nor carcinogenic in rats and mice. The Acceptable Daily Intake of diketopiperazine for humans has been established at 7.5 mg/kg body weight (JECFA, 1980) based on the no observed effect level of 750 mg/kg bw./day obtained in a long term toxicity study on rats, divided by a safety factor of 100.

Source & ©: AFSSA   Aspartame Assessment Report section 4.3

4.1.2 Epidemiological data

The source document for this Digest states:

Epidemiological data

Concerning the epidemiological data on brain tumours, the AFSSA (2002) report noted that:

"In 1996, Olney et al. published an article on a possible link between the increase in the frequency of brain tumours in humans and the consumption of aspartame in the United States. Based on the data from the NCI (10% of the population) from 1975-1992, the authors concluded that there was a significant increase in the frequency of brain tumours in the mid- 1980s, that is to say the period following aspartame came onto the market. The conclusions of this epidemiological study have been criticised by a number of scientists who questioned the methodology, the use of the data and their interpretation (Levy et al., 1996; Linet et al., 1999; Ross, 1998; Seife, 1999; Smith et al., 1998). One of the major criticism is that the authors only took into account the frequency of brain tumours during a selected period (1975-1992). When all the epidemiological data are used (1973-1992) a different conclusion is reached, as the frequency of brain cancers began to increase in 1973 and stabilised from the mid-1980s (Levy et al., 1996). Furthermore, Olney et al. did not provide any quantitative or qualitative relationship between the exposure of the population to aspartame and the observed frequency of brain tumours. Finally, an increase in the incidence of the tumours can have many causes including, among others, improvements in diagnostic methods (Modan et al., 1992)."

"More recently, Gurney et al., (1997) published the results of a case-control study on the relationship between the consumption of aspartame and the frequency of brain tumours. The study covered 56 patients affected by tumours in childhood and 94 controls. According to these authors, no relationship could be established between the consumption of aspartame and the frequency of brain tumours."

"In France, data on the incidence of and mortality from brain cancers were supplied by the FRANCIM network (F. Ménégoz et al., 2001). These cancers include meninges tumours and tumours of the brain itself. Between 1980 and 1997, the incidence (number of new cases appearing each year) of cerebral tumours was relatively stable in men and showed a slight increase in women. The trend towards an increase in mortality from cancer of the brain and other parts of the nervous system is a longstanding one, as it first appeared in 1950 and continues to the present day, for both sexes. However, during the last decade, mortality in men stabilised and the increase in mortality from brain cancer in women was less pronounced than during the preceding period."

"In France, the epidemiological data from the cancer registers do not enable a definitive indication to be given on a possible aspartame-brain tumour relationship, but they do show that, at the present time, the sale of this food additive in France is not being accompanied by an increase in the frequency of brain tumours or increased mortality from this disease in the general population.

Source & ©: EC-SCF  Update on the Safety of Aspartame
The 2002 AFFSA Assessment report quoted above is available on:
AFSSA   Aspartame Assessment Report chap 4

4.2 Can aspartame affect reproduction and development?

The source document for this Digest states:

Reproduction and Developmental toxicity

The derivation of an ADI for aspartame by JECFA (1980) and the EEA (1985) included assessment of single- and multi-generation studies in animals that were specifically designed to examine the possible effects of aspartame and its metabolic conversion products on reproduction, and development, including neuro-development.

The data used by JECFA (1980) were discussed in more recent reviews (Kotsonis and Hjelle, 1996; London and Rorick, 1996; Shaywitz, 1997; AFSSA, 2002), but no additional studies were identified which would impact on the no-observed-adverse- effect level (NOAEL)."

Source & ©: EC-SCF  Update on the Safety of Aspartame

"(…) The multigeneration reproduction and teratogenicity studies showed consistent adverse effects on the weight of progeny, both at weaning and at terminal examination, at the highest dose levels tested.

The committee reviewed additional information on the multigeneration reproduction and teratogenicity studies, including recalculations of the actual intake by the offsprings, which suggested the consumption by the pups of levels higher than which would follow from the composition of the test diets. The Committee noted the observed growth depression in the progeny was marginal and related to a decrease in food consumption caused by a high intake of phenylalanine. In the light of these findings the committee concluded that 4,000 mg/kg bw could be considered as the no-adverse-effect level also in these studies. (…)

Source & ©:  1985 Reports of the Scientific Committee for Food, 16th Series, annex 2 page 8.

4.3 Can aspartame produce neurological disorders?

The source document for this Digest states:

Neurological effects

Much of the recent interest in the safety of aspartame has explored whether its consumption is linked with neurological effects. Therefore this end point has been given special consideration in this review.

Shortly after the widespread marketing of aspartame, there were a number of anecdotal reports of health effects, which some consumers related to their consumption of aspartame-containing products (Hull, 1999). Most of the earlier complaints and reports of aspartame-related adverse reactions were analysed by experts at the Centres for Disease Control (CDC) in Atlanta on behalf of the FDA, who concluded that there was no symptom complex that could be assigned to the ingestion of aspartame (Janssen and Van der Heijden, 1988; Tollefson, 1988).

A number of complaints were of a neurological or behavioural type (Tollefson, 1988) and these received special consideration, in part because experiments in animals have shown that high doses (1000mg/kg bw and above in rats) can alter the concentrations of neurotransmitters and their precursors within the central nervous system (Lajtha et al., 1994).

As regards the potential effect of aspartame on neurotransmitter levels, the underlying hypothesis was that aspartame, as a source of Phe without the other large neutral amino acids (LNAA) (i.e. tryptophan, valine, leucine, methionine, histidine) which compete for transport across the blood-brain barrier, would increase the serum ratio of Phe to the other LNAA, thereby selectively increasing Phe concentrations in brain. It was further suggested that such increased entry of Phe into the brain may result in disturbances in monoaminergic neurotransmission (Wurtman, 1985).

A number of animal studies were conducted to determine whether increases in plasma Phe concentrations secondary to large doses of aspartame may result in changes in brain concentrations of norepinephrine, dopamine, or serotonin and their metabolites (reviewed by Schomer et al., 1996; Lajtha et al., 1994). Although effects on neurotransmitter levels were noted in some acute and repeat-dose studies at high doses in rodents, it was apparent that these effects were not consistent or reproducible. For instance, acute doses of up to 2000 mg/kg bw/d and repeated doses of up to 863 mg/kg bw/d (for 28 days) failed to induce significant changes in brain serotonin or dopamine levels and had no effect on seizure severity in rats genetically prone to epilepsy (Dailey et al., 1991).

Some changes in neurotransmitter levels in rodents were also identified in some of the older studies on aspartame (Lajtha et al., 1994). In mice given aspartame orally at 13,130 and 650 mg/kg bw, increases of 12, 49 and 47% respectively in norepinephrine were found after 3 hours in the hypothalamus; significant increases in norepinephrine in the medulla oblongata (in the low- and high-dose group animals) and corpus striatum (in the low-dose group animals) were also observed (Coulombe and Sharma, 1986). However, these increases were not dose-related and were accompanied by non-significant changes in serotonin levels. Lack of any significant effects on biogenic amine levels, following higher bolus doses (1000 mg/kg bw) of aspartame, have also been reported in both Sprague-Dawley and Fischer 344 rats (Freeman et al., 1990).

Glutamic and aspartic acids act as excitatory neurotransmitters at glutamate receptor sites to which aspartic acid also shows affinity. A more recent study evaluated brain glutamatergic receptor kinetics following perinatal exposure to large doses of aspartame (500 mg/kg bw/day) (Reilly and Lajtha, 1995). In this study aspartame in drinking water was administered to Sprague-Dawley rats throughout gestation and lactation. The kinetics of the N-methyl-D-aspartate receptor and total glutamatergic binding in cerebral cortex and hippocampus of the offspring (20-22 days old) were found to be unaffected by perinatal exposure to aspartame. However, statistically significant but reversible reductions in glutamic acid levels in both brain regions and of aspartate in the hippocampus were noted. The same group of workers reported an absence of effects on dopaminergic, adrenergic and serotonergic receptor binding kinetics in adult rat brain with chronic exposure to aspartame (Reilly et al., 1989)."

Source & ©: EC-SCF  Update on the Safety of Aspartame


Methanol accounts for approximately 10% of aspartame in terms of weight. It is metabolised into formaldehyde, formic acid and CO2. One litre of "diet" drink containing aspartame produces approximately 48 mg of methanol whereas a litre of fruit or vegetable juice contains approximately 200 to 280 mg of methanol. This indicates that the quantities of methanol provided by aspartame as a food additive are lower than those provided by certain natural foods (Maher, 1986). In humans, ingestion of methanol at a dose of 200-500 mg/kg body weight is required to induce an accumulation of formate in the blood and toxic effects on the vision and the central nervous system. These doses are more than 100 times greater than the maximum dose of methanol provided by aspartame (FDA, FR 1984). These data put into perspective the potential toxic effects of the metabolites of methanol (formaldehyde and formic acid) following exposure to aspartame.

Source & ©: AFSSA   Aspartame Assessment Report, section 4.3.

4.4 Can aspartame affect behaviour, cognition or mood?

The source document for this Digest states:

Behaviour, Cognition and Mood

Some years ago, it was hypothesised that aspartame, primarily due to its content of Phe, could have an effect on human behaviour, cognition, and possibly on measures of physiological function (Wurtman, 1985). However, no consistent and reproducible effects were observed in a number of older animal studies investigating the effects of aspartame on neurotransmitter levels.

Only a limited number of studies on behavioural aspects in animals have been published in the last ten years. A proportion of these focused on seizure activity but a causal link with aspartame could not be established; no adverse effects on other aspects of behaviour and cognition were reported in experimental animals when aspartame was given at oral dose levels of up to 2000 mg/kg bw/day (Yirmiya et al., 1989; Tilson et al., 1991; Mullenix et al., 1991; Vitulli et al., 1996; LaBuda and Hale, 2000; Goerss et al., 2000).

A number of anecdotal reports in humans were received by the manufacturers of aspartame in early to mid 1980's relating to a variety of symptoms following the marketing of aspartame in the USA. About two-thirds of these symptoms fell into the neurobehavioural category (Butchko and Stargel, 2001). These consisted mostly of headaches (see below), mood alterations, insomnia, and dizziness. More than 500 reports were received by CDC, and almost half underwent follow-up and evaluation. A post-marketing surveillance system was developed by the NutraSweet company (Butchko and Kotsonis, 1994; Butchko et al., 1996), which was followed by scientific research on these neurological symptoms (see below).

A number of scientific studies were carried out in healthy and potentially sensitive individuals, including children, to test whether the consumption of aspartame was associated with behavioural and cognitive changes. The potentially sensitive individuals studied were, heterozygotes for PKU, individuals suffering from depression, Attention Deficit Disorder (ADD), Parkinson’s Disease, epilepsy or other suspected seizures. They included double-blind studies in children (Saravis et al., 1990; Shaywitz et al., 1994) in which no effects were observed on behaviour, mood or learning when aspartame was given as a drink at single and multiple doses of 34 mg/kg bw/day for up to two weeks. The longer term study of Shaywitz et al. (1994) examined the effect of aspartame in children with ADD and included an assessment of liver function as well as measurement of plasma levels of amino acids, serotonin and monoamine metabolites. Treatment-related effects were also absent in a study of pre- school children who were given aspartame at 32 mg/kg bw/day and described as sugar sensitive by their parents (Wolraich et al., 1994).

A number of double-blind behavioural studies of variable quality in healthy adults, involving single and repeated administrations of aspartame have also been conducted. No treatment-related effects were noted in tests on a range of cognitive parameters in studies employing single administrations of aspartame at doses of up to 60 mg/kg bw/day (Lieberman et al., 1988; Lapierre et al., 1990; Pivonka and Grunewald, 1990; Stokes et al., 1991, 1994). However, it can be argued that single dosing studies employing high amounts of aspartame do not reflect typical consumption patterns.

A number of longer term studies with a double-blind design involving multiple dosing in healthy individuals also failed to highlight any treatment-related adverse effects on behaviour (Spiers et al., 1998; Leon et al., 1989). As noted with shorter-term studies, no treatment-related effects on behaviour were noted even when aspartame was tested at 74 mg/kg bw/day for periods extending up to 24 weeks. Although Phe concentrations increased significantly as a result of treatment with aspartame, there were no significant effects noted on behaviour, mood or electroencephalogram (EEG) patterns, nor on a comprehensive battery of clinical laboratory tests. Headache was the most frequently reported adverse effect in placebo- and aspartame-treated groups but there were no significant differences noted between groups.

Several subpopulations of individuals who may potentially be sensitive to aspartame have also been studied. From a double-blind study with a cross-over design in 13 depressed patients, Walton et al. (1993) concluded that aspartame (30 mg/kg bw/day for 7 days) increased the frequency and severity of adverse experiences in these individuals. These authors concluded that the use of aspartame in individuals with mood disorder should be discouraged. However, it is difficult to interpret this study since the authors numerically combined unrelated adverse effects to show a statistically significant result in depressed patients and only a limited number of subjects were available for evaluation due to premature termination of the study.

The effect of aspartame on behaviour, cognition and EEG patterns has also been investigated in PKU heterozygotes. Older studies in PKU homozygotes and those heterozygous for the condition have been reviewed elsewhere (de Sonneville and Benninger, 1996, and references therein). Overall, the authors concluded that aspartame did not affect cognitive function and EEG profiles in either the general population or those heterozygous for PKU. In a more recent double-blind study, which included assessment of plasma amino acid levels and EEG patterns (Trefz et al., 1994), the subjects ingested aspartame (15 or 45 mg/kg bw/day) and placebo for 12 weeks. The battery of behavioural tests included tests for short-term memory, reaction time and various attention tasks. Although headaches were among the mild adverse symptoms reported, there was no statistically significant difference between treatments. There was a significant rise in Phe in the high-dose group in contrast to the low-dose group and this was also the case for the ratio of Phe to LNAA. However, aspartame had no significant effect on cognitive function or EEG profiles.

Source & ©: EF-SCF  Update on the Safety of Aspartame

4.5 Does aspartame cause headaches?

The source document for this Digest states:


Headache was one of the more common symptoms that was reported to the FDA and evaluated by the CDC (Janssen and Van der Heijden, 1988; Tollefson, 1988). Several studies were carried out to test the potential association between aspartame intake and the onset of headaches. Although the results of a questionnaire-based study (Lipton et al., 1989) and two double-blind out-patient investigations (Koehler and Glaros, 1988; Van Den Eeden et al., 1994) employing daily doses of up to 30 mg/kg bw/day indicated a potential association between aspartame intakes and headache, it was not possible to deduce causality as the effect of diet had not been adequately controlled for and the interpretation of the data was complicated by a high drop out rate and a limited experimental design.

Another study employing a controlled environment, which was also a randomised double-blind placebo-controlled cross-over trial, concluded that aspartame was no more likely than placebo to trigger headaches (Schiffman et al., 1987). This study consisted of 40 subjects who complained of aspartame-related headaches. Subjects received aspartame challenges on days three or five at a total dose of 30 mg/kg bw (for a 70 kg person); subjects received placebo on the other days. While 35% of subjects developed headaches while on aspartame, 45% developed headaches while on placebo. In addition, no treatment related effects were detected in blood pressure, or plasma concentrations of cortisol, insulin, glucagon, histamine, epinephrine or norepinephrine. The subjects who had headaches had lower plasma concentrations of norepinephrine and epinephrine just before the development of headache. This study has been criticised for using tightly controlled experimental conditions which did not mimic normal life (Edmeads, 1988), but Schiffman et al. (1987) argued that the nature of the study and the primary focus of the questions raised by CDC dictated that they use carefully controlled conditions at a hospital setting.

Source & ©: EF-SCF  Update on the Safety of Aspartame

4.6 Does aspartame trigger epileptic seizures?

The source document for this Digest states:


The AFSSA (2002) report noted that

"Among the possible adverse effects of aspartame, researchers have paid particular attention to seizures. Several studies have suggested a relationship between the consumption of large amounts of aspartame and the triggering of epileptic seizures. In an old study (1972), on new-born monkeys (2-3 animals per group) treated with doses of aspartame of 1, 3 and 4g/kg bw/day for 52 weeks, epileptic seizures were recorded at the highest doses, after 218 days of treatment. Thereafter, sporadic convulsions were observed during handling of the animals. These symptoms were identical with those observed in young monkeys treated with phenylalanine."

"In contrast, in a similar study also conducted on young monkeys, no effect was observed at doses of aspartame of 2 and 2.7 g/kg bw/day. The different results observed in the two studies could be explained by differences in the exposure conditions, the food and the state of health of the animals (JECFA, 1980)."

"Walton et al. (1993) reported, in a study conducted on 13 patients suffering from depression, that the administration of 30 mg/kg bw/day of aspartame for 7 days caused severe side effects in these patients which led the authors to conclude that the use of this sweetener in depressive patients should be avoided. The same author (Walton, 1986) reported a case of 7 epileptic seizures and serious behavioural problems in a woman being treated with anti- depressants who ingested large quantities of tea containing aspartame."

"Wurtman (1985) indicated that the administration of aspartame, due to an increase in phenylalanine absorption in the brain, could affect the synthesis of catecholamines or serotonin and cause seizures. He based his findings on three examples of heavy consumers of "diet" drinks and on experimental studies on animals demonstrating that the consumption of aspartame reduced the threshold of sensitivity to chemically induced seizures (Maher et al., 1987; Guiso et al.,1988; Pinto et al., 1988). Finally Camfield et al. (1992) demonstrated that aspartame could increase the duration of certain types of epileptic seizure in children."

The ATIC on the Internet reported a large amount of evidence from people who have identified aspartame as the cause of their health problems and in particular of seizures. These statements should be taken into account but with the reservation that they have not been examined according to any academic standard. They may, however, in certain cases, reflect the hypersensitivity of certain individuals to aspartame or its metabolites. Effects on seizures have been reported with phenylalanine, aspartic acid and methanol but these were under specific conditions (high doses, individual sensitivity, types of seizures, etc.) which are not representative of the general population and of current use of this sweetener in food (Anderson et al., 1996). This causal relationship between aspartame and epileptic seizures has been refuted by a large number of scientists who base their opinions on numerous experimental studies conducted on laboratory animals or on clinical or tolerance studies in humans (Anderson et al., 1996; Gaull, 1985; Rowan et al., 1995; Shaywitz et al., 1994; Tollefson et al., 1992; 1993; Dailey et al., 1991; Zhi et al., 1989; Sze, 1989; Tilson et al., 1989)."

"The Epilepsy Institute in the USA has also concluded that aspartame is not the cause of epileptic seizures (Congressional Record, June 20, 1986). In the United States various consumer complaints about aspartame have been collected by the Special Nutritionals Adverse Event Monitoring System (SN/AEMS). The sources of these reports were the FDA, federal and local health agencies, consumers and health professionals. Of 2621 side effects reported, concerning 3451 products, some ten cases concerned preparations concerning aspartame (mixtures also containing vitamins, amino acids and various nutritional supplements). The effects reported included seizures, death, nervous and cardiac symptoms, oedema and fever. Still in the United States, the Center for Disease Control assessed 517 complaints about aspartame (1983). The symptoms reported were headaches, mood changes, insomnia, abdominal pain, nausea, convulsions, etc. These symptoms are observed frequently in the general population. Although it might be possible that certain individuals are particularly sensitive to aspartame, these data, which relate to a large number of people, have not enabled any relationship to be demonstrated between the consumption of aspartame and the occurrence of convulsive seizures.

Source & ©: EF-SCF  Update on the Safety of Aspartame
The 2002 AFFSA Assessment report quoted above is available on:
AFSSA   Aspartame Assessment Report chap 4

4.7 Can aspartame cause other health effects?

The source document for this Digest states:

Idiosyncratic reactions described as allergic-like (hives, rashes) were reported by some consumers to CDC in response to aspartame (Tollefson, 1988). However, the results of a multi-centre, randomised, double-blind, placebo-controlled, cross-over study in individuals who were convinced they were allergic to aspartame indicated that aspartame and its conversion products are no more likely than placebo to cause urticaria and angio-oedema (Geha et al., 1993). This finding was supported by the outcome of another study, which also demonstrated that alleged allergic reactions to aspartame were not reproducible under blinded conditions (Garriga et al., 1991). However as with the Geha et al. (1993) study, the authors reported major difficulties in enrolling subjects with a history of allergy/hypersensitivity reactions to aspartame.

A number of other studies focused on the effects of aspartame on hunger and food intake (Rolls and Shide, 1996) and in the control of body weight (Kanders et al., 1996). Sensory and post-ingestion experience with aspartame was reported by these reviewers not to be associated with increased energy intake or increases in body weight.

Source & ©: EF-SCF  Update on the Safety of Aspartame

Themes coveredLeaflets