Cocoa, powd., alkalized

Endpoint summary

Description of key information

In the studies documented in CSR, the dietary level(s) of cocoa tested represent amounts far exceeding cocoa consumption by humans.  A recent approximation of cocoa powder consumption in 2011 (using cocoa production data from 2011 and EU27 population statistics) approximates European cocoa powder consumption to be 18.5 mg/kg bw day.  
The above information, as well as the knowledge that a vast majority of the population (including children) regularly consume large quantities of cocoa containing products, clearly indicates that there is no concern regarding the repeated oral toxicity of cocoa.
In accordance with the results obtained from the key and supporting studies, cocoa powder is not classified according to CLP i.e. NOEL (or derived/adjusted LOEL) dose/concentration exceeds the guidance value ranges of 10 < C ≤ 100 mg/kg/bw/day, (Regulation (EC) No 1272/2008). 
 [ Please refer CSR 5.6.3]

Key value for chemical safety assessment

Repeated dose toxicity: via oral route - systemic effects

Endpoint conclusion

Endpoint conclusion:

adverse effect observed

Dose descriptor:

NOAEL

3 000 mg/kg bw/day

Study duration:

subchronic

Species:

rabbit

Quality of whole database:

The subchronic toxicity of cocoa powder has been assessed in a non-guideline study on a hypercholesterolemic strain of rabbits. This study is well-documented and meets basic scientific principles.

Repeated dose toxicity: inhalation - systemic effects

Endpoint conclusion

Endpoint conclusion:

no study available

Quality of whole database:

No standard studies on laboratory animals repeatedly exposed to cocoa by inhalation were available. Symptoms of repeated dose toxicity reported in the limited epidemiological data appear to reflect local, rather than systemic, effects.
The fact that about 90% of cocoa powder consists of substances of only low systemic toxic potential (carbohydrates, fats and protein) is reassuring. The minor components are also expected to have low systemic toxicity following repeated exposure by inhalation.

Repeated dose toxicity: inhalation - local effects

Endpoint conclusion

Endpoint conclusion:

no study available

Quality of whole database:

No standard studies on laboratory animals repeatedly exposed to cocoa by inhalation were available. In the limited available epidemiological data, a statistically significant increase in the incidence of a range of ‘chronic’ respiratory symptoms was reported in exposed workers, compared to controls. However, these are likely to be indicative of generalised local effects typical of organic dusts. The fact that about 90% of cocoa powder consists of substances of only low systemic toxic potential (carbohydrates, fats and protein) is reassuring. The minor components are also expected to have a low local toxicity following repeated exposure by inhalation.

Repeated dose toxicity: dermal - systemic effects

Endpoint conclusion

Endpoint conclusion:

no study available

Quality of whole database:

Please refer CSR 5.6.1.3

Repeated dose toxicity: dermal - local effects

Endpoint conclusion

Endpoint conclusion:

no study available

Quality of whole database:

Please refer CSR 5.6.1.3

Additional information

Tarka et al., (1991) conducted a comprehensive chronic toxicity/carcinogenicity study for a reference cocoa powder under Good Laboratory Practice (GLP), (Tarka et al., (1991). The animals (offspring of a multi-generation reproductive toxicity study., (Hostetler et al., (1990)) were maintained on cocoa powder-containing diets for 104 weeks at the same dietary concentrations (0, 1.5, 3.5, and 5.0%) that had been fed to their predecessors for three generations. The results from this study clearly indicated no evidence of chronic toxicity or carcinogenicity. The incidence of bilateral diffuse testicular atrophy was increased and spermatogenesis was decreased in male rats fed 5% cocoa powder however, these effects were not unexpected since methylxanthines, theobromine, caffeine, and theophylline had been previously documented to exhibit effects on the testis at this level, (Friedman et al.(1978), Gans, (1984), Tarka et al. (1979, 1981), Ettlin et al. (1986), Wang et al.(1992) and Wang and Waller (1994). It should also be noted that the statistically significant effects on the testis were only in male rats fed the highest concentration of cocoa powder (5%), and only occurred after 78 weeks of continuous exposure.

There were no consistent statistically significant treatment and dose-related adverse histopathological findings for other organs examined at any time point. At the highest level tested, there was limited involvement of the heart and kidneys. Increases in the incidence of interstitial fibrosis (for both sexes), in the heart, suggested that this organ could represent another target organ from exposure to continuous intake of 5% cocoa powder in the diet. Similarly, non-suppurative myocarditis was also present in rats of both sexes fed 5% cocoa powder diets. However, the authors noted that the absence of consistent time- and dose associated effects together with reports of similar lesions observed in this strain of rat irrespective of treatment (Anver et al., 1982; Greaves and Faccini, 1984a; Laham et al., 1985) casts doubt on a relationship between dietary cocoa powder intake and the cardiac lesions observed. It was also noted that the likely increased incidence of non-suppurative myocarditis resulted from exacerbation of spontaneous, age-related lesions. Increases in both renal pelvic dilatation (hydronephrosis) and pelvic microcalculi in high cocoa powder-fed rats of both sexes were evident from cumulative incidence data, but these effects were reported to lack convincing temporal and dose-related qualities. The development of renal pelvic dilatation has been shown to be a polygenic heritable trait and, since the F3b generation of rats in a multi-generation study was used in this study, genetic predisposition may have been a contributing factor. Dietary protein has also been implicated as a possible causative factor in the development of pelvic dilatation (Greaves and Faccini, 1984b). The development of pelvic renal microcalculi is also documented to occur spontaneously in aging rats (Woodard and Khan, 1986). Besides age, local factors independent of treatment status are reported to be important in the initiation of stone formation in rats (Heptinstall, 1974a,b). The increase in the incidence of both renal pelvic dilatation and pelvic microcalculi in cocoa powder fed rats was reported to possibly reflect a complex interaction between factors such as gender, diet composition, age, strain, urine production and genetic predisposition and as such an obvious cause and effect relationship between cocoa powder intake and renal lesions observed was not demonstrated in this supporting study.

However, the more recent Kurosawa et al. (2005) study in which hypercholesterolemic rabbits fed cocoa powder at 10% in the diet for six months failed to show any treatment-related adverse effects.In light of the changes observed during the Tarka (rat) study, the lowest-observed-effect level (LOEL) is approximately 1000 mg/kg bw day. However, no adverse effects were identified in the more recent, well conducted 6-month rabbit study by Kurosawa et al., with determination of a NOAEL for cocoa powder of ca3000mg/kg bw/d (the only dose tested).

The low oral toxicity of cocoa powder was also demonstrated in the short-term study reported by Takyi and Ofori-Mensa (1981). Rats fed either 0 (control) or 350 g/kg (irradiated and non-irradiated) cocoa beans (equivalent to approx. 10 g/kg bw/day), failed to show any significant, treatment-related differences upon haematological or gross pathological examination. The study LOEL was considered to be 10 g/kg bw/day, which was attributed due to the reported significantly lower body weight gains in treated rats. Additionally,

In the studies documented above, the dietary level(s) of cocoa tested represent amounts far exceeding cocoa consumption by humans. A recent approximation of cocoa powder consumption in 2011 (using cocoa production data from 2011 and EU27 population statistics) approximates European cocoa powder consumption to be 18.5 mg/kg bw day. 

The above information, as well as the knowledge that a vast majority of the population (including children) regularly consume large quantities of cocoa containing products, clearly indicates that there is no concern regarding the repeated oral toxicity of cocoa.

In accordance with the results obtained from the key and supporting studies, cocoa powder is not classified according to CLP i.e. NOEL (or derived/adjusted LOEL) dose/concentration exceeds the guidance value ranges of 10 < C ≤ 100 mg/kg/bw/day, (Regulation (EC) No 1272/2008).

 [CSR 5.6.3]

Justification for selection of repeated dose toxicity via oral route - systemic effects endpoint:
No treatment-related adverse effects were seen over the course of the study. Cocoa powder was seen to have a protective effect against the oxidation of LDL cholesterol in the second and third month of administration, and suppressed the development of atherosclerosis.

Justification for selection of repeated dose toxicity inhalation - systemic effects endpoint:
No standard repeat dose inhalation studies in laboratory animals were available for cocoa powder, but particle size measurement and respiratory modelling provided useful insights (British American Tobacco, 2013).

The particle size distributions of three representative samples of cocoa powder were measured using an optical technique (Kulkarni et al). First, the particles in each sample were dispersed by pulsating air jets. Each particle was accelerated through a pair of laser beams. The resulting scattered light was then collected by an avalanche photo-detector. This produced an electrical pulse with a shape that is a function of the particle size. Time-Of-Flight and Time-In-Beam methods were then used to determine the particle size. This technique offers high resolution particle size measurements weighted by number, surface area and volume in the size range of 0.3 - 500 μm, suitable for dry powder and medical aerosol characterisation. Three replicates were used for each sample. The results were used to derive the particles’ aerodynamic diameters (dae). This is an expression of the particles’ aerodynamic behavior as if they were perfect spheres with unit-density. Aerodynamic diameter is commonly used to assess lung deposition efficiency.

The aerodynamic particle sizes and geometric standard deviations were used to estimate regional deposition for the head, bronchial and pulmonary regions of the lung. The assessment was conducted using the Multiple-Path Particle Dosimetry Model (MPPD v2.71, Applied Research Associates, Raleigh, NC) using default tidal breathing parameters in the Yeh-Schum symmetric lung morphology (Anjilvel and Asgharian, 1995; Asgharian et al, 2001). The results are given in the table in CSR 5.6.1.2

Justification for selection of repeated dose toxicity inhalation - local effects endpoint:
No standard repeat dose inhalation studies in laboratory animals were available for cocoa powder, but particle size measurement and respiratory modelling provided useful insights (British American Tobacco, 2013).

The particle size distributions of three representative samples of cocoa powder were measured using an optical technique (Kulkarni et al). First, the particles in each sample were dispersed by pulsating air jets. Each particle was accelerated through a pair of laser beams. The resulting scattered light was then collected by an avalanche photo-detector. This produced an electrical pulse with a shape that is a function of the particle size. Time-Of-Flight and Time-In-Beam methods were then used to determine the particle size. This technique offers high resolution particle size measurements weighted by number, surface area and volume in the size range of 0.3 - 500 μm, suitable for dry powder and medical aerosol characterisation. Three replicates were used for each sample. The results were used to derive the particles’ aerodynamic diameters (dae). This is an expression of the particles’ aerodynamic behavior as if they were perfect spheres with unit-density. Aerodynamic diameter is commonly used to assess lung deposition efficiency.

The aerodynamic particle sizes and geometric standard deviations were used to estimate regional deposition for the head, bronchial and pulmonary regions of the lung. The assessment was conducted using the Multiple-Path Particle Dosimetry Model (MPPD v2.71, Applied Research Associates, Raleigh, NC) using default tidal breathing parameters in the Yeh-Schum symmetric lung morphology (Anjilvel and Asgharian, 1995; Asgharian et al, 2001). [CSR 5.6.1.2]

Justification for selection of repeated dose toxicity dermal - systemic effects endpoint:
There is no information on the repeated dose dermal toxicity of cocoa powder. However, more than 90-95% of the constituents of cocoa powder have little or no anticipated dermal absorption, with cocoa and cocoa butter having a long and safe history of use in the cosmetics and personal care sectors (CIR 2006, Gasser 2008):
Glycerides (cocoa butter) are generally poorly water soluble, have an estimated log Pow > 6 and molecular weights > 500. As such, uptake into the stratum corneum of skin and further transfer into the epidermis are likely to be low.
Plant proteins, starches and cellulosic fibre are substances that are predominantly listed in Annex IV of Regulation 1907/2006 (REACH) being substances that are considered to cause minimum risk because of their intrinsic properties, and being high molecular weight polymers with low affinity for lipids are expected to have little or no uptake into dermal layers.
The organic acids are intermediaries or components of mammalian metabolism, and if taken up dermally would be expected to enter into normal metabolic pathways. However, dermal uptake of aqueous substances with low affinity for lipids (evidenced by log Kow values <3) is known to be limited. Thus, for citric acid, which has log Kow -0.12, over a 15 minute exposure period, the fraction taken up dermally would be ca 0.006. Also, if the organic acids present in cocoa powder are associated with counter-ions, as would be expected, the dermal uptake is likely to be even lower.
Catechins appear to be poorly absorbed and transferred across adult dermal tissue, but may be more highly absorbed by neonatal skin. However, catechins are not known to be harmful to dermal tissue, and in fact, have demonstrated skin protective properties (Kim, 2001).
There is recorded absorption of caffeine across dermal layers (van der Sandt et al. 2004) but at the levels present in cocoa powder, and limited exposure via the dermal route, uptake via the skin is expected to be very low. [5.6.1.3]

Justification for selection of repeated dose toxicity dermal - local effects endpoint:
There is no information on the repeated dose dermal toxicity of cocoa powder. However, more than 90-95% of the constituents of cocoa powder have little or no anticipated dermal absorption, with cocoa and cocoa butter having a long and safe history of use in the cosmetics and personal care sectors (CIR 2006, Gasser 2008):
Glycerides (cocoa butter) are generally poorly water soluble, have an estimated log Pow > 6 and molecular weights > 500. As such, uptake into the stratum corneum of skin and further transfer into the epidermis are likely to be low.
Plant proteins, starches and cellulosic fibre are substances that are predominantly listed in Annex IV of Regulation 1907/2006 (REACH) being substances that are considered to cause minimum risk because of their intrinsic properties, and being high molecular weight polymers with low affinity for lipids are expected to have little or no uptake into dermal layers.
The organic acids are intermediaries or components of mammalian metabolism, and if taken up dermally would be expected to enter into normal metabolic pathways. However, dermal uptake of aqueous substances with low affinity for lipids (evidenced by log Kow values <3) is known to be limited. Thus, for citric acid, which has log Kow -0.12, over a 15 minute exposure period, the fraction taken up dermally would be ca 0.006. Also, if the organic acids present in cocoa powder are associated with counter-ions, as would be expected, the dermal uptake is likely to be even lower.
Catechins appear to be poorly absorbed and transferred across adult dermal tissue, but may be more highly absorbed by neonatal skin. However, catechins are not known to be harmful to dermal tissue, and in fact, have demonstrated skin protective properties (Kim, 2001).
There is recorded absorption of caffeine across dermal layers (van der Sandt et al. 2004) but at the levels present in cocoa powder, and limited exposure via the dermal route, uptake via the skin is expected to be very low. [5.6.1.3]

Repeated dose toxicity: via oral route - systemic effects (target organ) cardiovascular / hematological: aorta

Justification for classification or non-classification

In accordance with the results obtained from the key and supporting studies, cocoa powder is not classified according to CLP i.e. NOEL (or derived/adjusted LOEL) dose/concentration exceeds the guidance value ranges of 10 < C ≤ 100 mg/kg/bw/day, (Regulation (EC) No 1272/2008).