WHO FOOD ADDITIVES SERIES: 52

First draft prepared by Dr D. Benford
United Kingdom Food Standards Agency, London, England

The food colour curcumin (turmeric yellow) is obtained by solvent extraction of turmeric, i.e. the ground rhizomes of Curcuma longa L. (C. domestica Valeton), with purification of the resultant extract by crystallization. The commercial product consists essentially of curcumins: the colouring principle (1,7-bis(4-hydroxy-3-methoxyphenyl) hepta-1,6-diene-3,5-dione) and its desmethoxy and bisdesmethoxy derivatives in varying proportions. The total content of colouring matter (curcuminoids) in curcumin is not less than 90%. Minor amounts of oils and resins that occur naturally in turmeric may also be present.

The term "curcumin" in this monograph refers to the material for which specifications exist. The principal colouring component, 1,7-bis(4-hydroxy-3-methoxyphenyl) hepta-1,6-diene-3,5-dione, is often referred to as curcumin in the literature; to avoid confusion, the Committee decided that this report would not use the term curcumin when referring to this substance. A common synonym for this substance is diferuloylmethane, and this name will be used when it is necessary to refer to the principal colouring component of curcumin.

Turmeric oleoresin, the product of solvent extraction of turmeric containing <90% of total colouring matter (curcuminoids), and curcumin were evaluated by the Committee at its thirteenth, eighteenth, twenty-second, twenty-fourth, twenty-sixth, thirtieth, thirty-fifth, thirty-ninth, forty-fourth, fifty-first and fifty-seventh meetings (Annex 1, references 19, 35, 47, 53, 59, 73, 88, 101, 116, 137 and 154). At its eighteenth meeting, the Committee established a temporary acceptable daily intake (ADI) of 0–0.1 mg/kg bw for curcumin, based on the then existing ADI for turmeric oleoresin (0–2.5 mg/kg bw) and an assumed average concentration of 3% curcuminoids in turmeric. The temporary ADI for curcumin was extended at the twenty-second, twenty-fourth, twenty-sixth, thirtieth, thirty-fifth and thirty-ninth meetings, whereas the ADI for turmeric oleoresin was withdrawn at the thirty-fifth meeting. At its thirty-ninth meeting, the Committee requested the results of studies of carcinogenicity in mice and rats fed with turmeric oleoresin and the results of a study of reproductive and developmental toxicity associated with curcumin. At its forty-fourth meeting, the Committee evaluated the results of studies of carcinogenicity in rats and mice given turmeric oleoresin containing 79–85% curcuminoids, and new biochemical and genotoxicity data. The Committee concluded that data on developmental toxicity were no longer required but reiterated its request for a study of reproductive toxicity. Based on the NOEL of 220 mg/kg bw per day for liver enlargement observed in the study of carcinogenicity in mice, and a safety factor of 200, the Committee increased the temporary ADI to 0–1 mg/kg bw and extended it, pending submission of the results of a study of reproductive toxicity with curcumin to be reviewed in 1998.

At its fifty-first meeting, the Committee evaluated the results of studies of fertility in rats and mice treated with turmeric oleoresin (68.0–76.5% curcuminoids). The low survival rate of pups in the study in mice and the low rates of pregnancy in rats led the Committee to conclude that these studies did not provide assurance that the potential reproductive effects of curcumin had been adequately investigated. The Committee again extended the temporary ADI, pending submission of the results of a study of reproductive toxicity with a substance complying with the specifications for curcumin, for review in 2001. At its fifty-seventh meeting, the Committee was informed that a multigeneration study in the rat was in progress, and thus extended the temporary ADI of 0–1 mg/kg bw until 2003.

The results of the multigeneration study were available to the Committee for evaluation at its current meeting. In addition, the Committee reviewed the results of two new clinical trials investigating either an extract of Curcuma or diferuloylmethane as potential anticancer agents.

2.1.1 Absorption, distribution and excretion

Fifteen patients with advanced colorectal cancer received an extract of Curcuma (18 mg of diferuloylmethane and 2 mg of the desmethoxy derivative suspended in 200 mg of essential oils derived from Curcuma spp.) daily for up to 4 months. The doses were equivalent to 26, 72, 108, 144 and 180 mg of curcumin, with three patients receiving each dose. Neither diferuloylmethane, or its glucuronide or sulphate conjugates, or hexahydrocurcumin or hexahydrocurcuminol were detected in plasma or urine after up to 29 days of treatment.

Diferuloylmethane was detected in the faeces of all patients. Diferuloylmethane sulphate was also detected in the faeces of one of the patients receiving diferuloylmethane at a dose of 180 mg/day, which may have been a result of biotransformation in the gut (Sharma et al., 2001).

Twenty-five patients with conditions indicating a high risk of malignancy were given diferuloylmethane (purity, 99.3%) for 3 months. The starting dose was 500 mg/day, which was increased stepwise to 1000, 2000, 4000, 8000 and finally 12 000 mg/day. Pharmacokinetic studies were conducted in patients who agreed to give samples of blood and urine and in normal volunteers. Serum concentrations of diferuloylmethane peaked at 1–2 h after administration of 4000–8000 mg diferuloylmethane and gradually declined within 12 h. A half-life was not determined. Diferuloylmethane was barely detectable in the serum of patients taking 500–2000 mg of diferuloylmethane. No diferuloylmethane could be detected in urine. Similar results were obtained in two patients who had taken diferuloylmethane for more than 1 month, indicating that repeated administration did not alter the pharmacokinetic profile of this substance (Cheng et al., 2001).

2.2.1 Reproductive toxicity

A two-generation study in Wistar rats was conducted according to OECD guideline 416 (May 1983) using curcumin comprising 80% diferuloylmethane, 99% total curcuminoids. The study was conducted in India, which does not have a formal good laboratory practice (GLP) inspectorate, but the laboratory was inspected by European Inspectorates for compliance with European Union and OECD guidelines, and the study was conducted in accordance with GLP.

Groups of 30 male and 30 female rats were fed diets containing curcumin at a concentration of 0, 1500, 3000 or 10 000 mg/kg of diet starting from 10 weeks before the mating period and throughout mating. Treatment of females continued throughout pregnancy and weaning of the offspring. The total periods of treatment were 21 weeks for the parental generation and 24 weeks for the F₁ generation. On postnatal day 4, the litter sizes of the F₁ offspring were standardized to a maximum of eight. After weaning, 30 male and 30 females of the F₁ generation were selected to become the parents of the F₂ generation. Parents were observed for clinical signs, body weights, food intake, cohabitation interval and duration of gestation. Pups were weighed on postnatal days 1, 4, 7, 14 and 21. All parents, F₁ weanlings not selected for mating and all F₂ weanlings were subjected to complete necropsy at terminal sacrifice. The concentrations used corresponded to doses of 0, 130–140, 250–290 or 850–960 mg/kg bw per day in males, and 0, 160, 310–320 or 1000–1100 mg/kg bw per day in females.

The following indices were calculated: male and female fertility index, percentage of matings resulting in pregnancy, number of implantations, percentage of pregnancies resulting in birth of live litters, percentage of live pups born, postimplantation loss, mean litter size and mean viable litter size, live birth index, percentage survival of pups at postnatal days 4, 7, 14 and 21. Ovaries, uterus, vagina/cervix, testes, epididymides, seminal vesicles, prostate, coagulating glands, liver, kidney, pituitary and adrenals were examined histologically.

There was a dose-related decrease in body-weight gain in the dams of the parental generation during days 10–15 of gestation, which was statistically significantly different from that of controls (body-weight gains, >80% that of controls) at the intermediate and highest doses. At this time, body weights were reported to be below the range of values for the historical controls. However, maternal body weights did not differ significantly between groups at the end of gestation. The mean body weights of the F₂ offspring (both sexes combined) were significantly decreased on postnatal days 1 and 7 at the intermediate dose, and on postnatal days 7, 14 and 21 at the highest dose. A dose-related trend was apparent, but the effect was small, with average body weights being >90% that of the control pups, and the observed changes were reported to be within the range of the data for historical controls. There were no other effects on general health, body weight, pup survival and fertility indices in either generation. The effects at the intermediate dose were observed at isolated time-points only and were considered to be incidental; and therefore this dose, equal to 250–320 mg/kg bw per day for the F₁ generation, was the NOEL (Ganiger, 2002).

In the clinical trial described in section 2.1.1, 15 patients with advanced colorectal cancer receiving Curcuma extract daily for up to 4 months were examined physically and blood samples were taken on days 1, 2, 8 and 29 of treatment and monthly thereafter. Blood samples were analysed for full blood cell count, concentrations of urea, electrolytes and markers of liver and bone function. Curcuma extract was administered at a dose equivalent to 26, 72, 108, 144 or 180 mg of diferuloylmethane, with three patients receiving each dose. The only adverse effects reported were gastrointestinal symptoms. During the first month of treatment, one patient receiving diferuloylmethane at a dose of 108 mg per day experienced nausea, which resolved spontaneously without discontinuing the treatment. Two patients, who received diferuloylmethane at a dose of 72 or 180 mg per day, respectively, experienced diarrhoea. In the absence of controls, and in view of the clinical conditions of the patients, it is not clear whether these symptoms were related to treatment (Sharma et al., 2001).

Twenty-five patients with conditions indicating a high risk of malignancy were given diferuloylmethane (purity, 99.3%) for 3 months. The starting dose was 500 mg/day, which was increased stepwise to 1000, 2000, 4000, 8000 and finally 12 000 mg/day. The patients received regular followup, including physical examination, weekly haemogram, and measurement of blood electrolytes and biochemistry parameters every 2 weeks. No adverse effects were reported at doses of up to 8000 mg/day. The highest dose of 12 000 mg/day was not acceptable to the patients because of the bulky volume of the tablets (Cheng et al., 2001).

The Committee received an estimate of intake only from Food Standards Australia New Zealand (FSANZ). This estimate, combining maximum curcumin use levels from the draft General Standard for Food Additives (GSFA) with food consumption data, is an unrealistic overestimate of the exposure. The report stated that FSANZ was not able to provide national estimates of intake for Australia or New Zealand owing to the regulatory status of curcumin, which is allowed at levels consistent with good manufacturing practice in all foods. Therefore the Committee concluded that adequate data were not available to accurately assess the exposure.

In a multigeneration study of reproductive toxicity, Wistar rats were fed diets containing curcumin (comprising 80% diferuloylmethane and 99% total curcuminoids) at doses equal to 0, 130–140, 250–290 and 850–960 mg/kg bw per day in males, and 0, 160, 310–320 and 1000–1100 mg/kg bw per day in females. The total period of treatment was 21 weeks for the parental generation and 24 weeks for the F₁ generation. Transient minor decreases in maternal body-weight gain were observed during days 10–15 of gestation in the parental, but not the F₁ generation, at the intermediate and high doses. There were no significant differences in maternal body weights at the end of gestation and no adverse effects were observed in the F₁ offspring. This change was therefore considered to be incidental. Significant decreases in the average weights of the F₂ generation pups were observed at days 1 and 7 at the intermediate dose, and days 7, 14 and 21 at the high dose. These decrements represented <10% of the average weight of the concurrent controls and were reported to be within the range of the data for historical controls. There were no other effects on general health, body weight, pup survival or fertility indices in either generation. The effect on pup weight seen at the intermediate dose (equal to 250–320 mg/kg bw per day) was likely to be incidental. The changes seen at the highest dose (equal to 960–1100 mg/kg bw per day for the F₁ parental generation) could be an indication of a persistent decrement in body-weight gain. The Committee therefore concluded that the NOEL for diferuloylmethane on the basis of decreased body weight in pups was 250 mg of /kg bw per day.

Two clinical trials were conducted in patients with cancer, or at high risk of cancer. In pharmacokinetic studies in these patients, diferuloylmethane could be detected in plasma after oral administration of diferuloylmethane (99.3%) at doses of >2000 mg/day (>33 mg/kg bw per day, for a 60-kg adult), but not after lower doses. Diferuloylmethane, and in one patient diferuloylmethane sulfate, was detected in the faeces but not in the urine.

These clinical trials provided limited information of relevance to the assessment of toxicity of curcumin. Fifteen patients receiving an extract of Curcuma (18 mg of diferuloylmethane and 2 mg of the desmethoxy derivative suspended in 200 mg of essential oils derived from Curcuma spp.) containing a daily dose of 26– 180 mg of diferuloylmethane for up to 4 months, were monitored for adverse effects by physical examination and tests for haematological parameters. During the course of the study, one patient (receiving diferuloylmethane at 108 mg/day) experienced nausea and two patients (receiving diferuloylmethane at 72 and 180 mg/day) experienced diarrhoea. There were no reported adverse effects in a study of twenty-five patients taking diferuloylmethane (99.3%) at doses of up to 8000 mg/day for 3 months. The Committee considered that these ancillary studies could not be used to derive an ADI for curcumin.

The new multigeneration study in rats that were fed with curcumin for periods of up to 24 weeks met the Committee’s requirements. Additionally, this material met the specification developed at this meeting. Decreased body-weight gain in the F₂ generation was observed at doses equal to 960–1100 mg/kg bw per day of curcumin; the NOEL was 250–320 mg/kg bw per day. The Committee noted that the previous temporary ADI was derived from a study on turmeric oleoresin (79–85% curcuminoids) that did not comply with the current specification. Taking into account all the data evaluated previously, the Committee withdrew the temporary designation and allocated an ADI for for curcumin of 0–3 mg/kg bw, based on the NOEL of 250–320 mg/kg bw per day in the multigeneration study in rats, and the application of a safety factor of 100.

The new multigeneration study in rats that were fed with curcumin for periods of up to 24 weeks met the Committee’s requirements. Additionally, this material met the specification developed at this meeting. Decreased weight gain in the F₂ generation was observed at doses equal to 960–1100 mg/kg bw per day of curcumin; the NOEL was 250–320 mg/kg bw per day. The Committee noted that the previous temporary ADI was derived from a study on turmeric oleoresin (79–85% curcuminoids) that did not comply with the current specification. Taking into account all of the data evaluated previously, the Committee withdrew the temporary designation and allocated an ADI of 0–3 mg/kg bw for curcumin, based on the NOEL of 250–320 mg/kg bw per day in the multigeneration study in rats, and the application of a safety factor of 100.

Cheng, A.L., Hsu, C.H., Lin, J.K., Hsu, M.M., Ho, Y.F., Shen, T.S., Ko, J.Y., Lin, J.T., Lin, B.R., Wu, M.S., Yu, H.S., Jee, S.H., Chen, G.S., Chen, T.M., Chen, C.A., Lai, M.K., Pu, T.S., Pan, M.H., Wang, Y.J., Tsai, C.C. & Hsieh, C.Y. (2001) Phase I clinical trial with high-risk or pre-malignant lesions. AntiCancer Res., 21, 2895–2900.

Ganiger, S. (2002) Twogeneration reproduction toxicity study with curcumin, turmeric yellow in Wistar rats. Unpublished report No: 3110/10 from Rallis Research Centre, Bangalore, India. Submitted to WHO by Spices Research Foundation, Cochin, India.

Sharma, R.A., McLelland, H.R., Hill, K.A., Ireson, C.R., Euden, S.A., Manson, M.M., Pirmohamed, M., Marnett, L.J., Gescher, A.J. & Steward, W.P. (2001) Pharmacodynamic and pharmacokinetic study of oral Curcuma extract in patients with colorectal cancer. Clin. Cancer Res., 7, 1894–1900.

    See Also:
       Toxicological Abbreviations
       Curcumin (WHO Food Additives Series 35)
       CURCUMIN (JECFA Evaluation)