Re‐evaluation of silicates and talc (E 553b) as food additives

Re‐evaluation of silicates and talc as food additives

Calcium silicate (E 552), magnesium silicate (E 553a) and talc (E 553b) are authorised as food additives according to Regulation (EC) No 1333/2008 on food additives and specifications have been defined in the Commission Regulation (EU) No 231/20121 for calcium silicate (E 552), magnesium silicate (E 553a(i)), magnesium trisilicate (E 553a(ii)) and talc (E 553b).

In 1991, the Scientific Committee on Food (SCF) established a group acceptable daily intake (ADI) ‘not specified’ for sodium silicate, silicon dioxide, calcium silicate, magnesium silicate and potassium silicate.

Although calcium silicate (E 552) and magnesium silicate (E 553(i)) may be described in terms of theoretical oxides, the Panel concluded that they are not mixtures of silicon dioxide and calcium or magnesium oxides; therefore, their definitions in the European Union (EU) specifications should be revised accordingly.

Information on particle size of calcium silicate analysed by laser diffraction (LD) and transmission electron microscopy (TEM) were provided. The Panel noted that these methods measure different particle characteristics, which are reflected in the different numerical size‐values obtained. Taking into account the analysis by TEM provided by industry, calcium silicate falls under the definition of a nanomaterial according to the Commission Recommendation 2011/696/EU. However, the Panel considered that the dispersion method (sonication) of the sample before its analysis by TEM is not representative of the common use of calcium silicate as a food additive. Following an European Food Safety Authority (EFSA) request, no TEM data were provided for magnesium silicate (E 553a(i)) and magnesium trisilicate (E 553a(ii)). Scanning electron microscope (SEM) data for talc indicated a particle size distribution of 20–40 μm.

The Panel considered that crystalline silica, fluoride, aluminium and nickel may be present in associated minerals in talc (E 553b) and, then, they could be present also in the food additive E 553b. Therefore, maximum limits for them should be established in the EU specifications for talc (E 553b).

The Panel considered that silicate anion from both calcium silicate or magnesium trisilicate was absorbed to a limited extent in rats. No data were available for magnesium silicate.

Based on a 2‐year study with calcium silicate in rats, the Panel considered that at high doses (up to 5,000 mg/kg body weight (bw) per day), there was evidence of silicon accumulation in the liver and kidney. The Panel considered that limited data in humans indicated that the silicate anion from magnesium trisilicate is absorbed to a limited extent, then excreted in the urine (as determined from urinary silicon measurements). No human data were available for calcium silicate or magnesium silicate; however, the Panel considered that a read‐across approach was appropriate and considered that silicate anion from both calcium silicate or magnesium silicate would behave similarly.

The Panel noted that studies with synthetically produced talc in mice, rat and guinea pigs as well as talc (baby powder) in hamsters indicated that less than 2% talc was systemically available, with low levels seldom found in the liver.

The Panel considered that calcium silicate (E 552), magnesium silicate (E 553a(i)), magnesium trisilicate (E 553a(ii)) and talc (E 553b) dissociate to a limited extent in the gastrointestinal tract into silicates and their corresponding cations. The resulting low amounts of calcium and magnesium ions were considered not to disturb normal physiological processes and therefore, are not discussed further in this opinion.

The Panel considered that calcium silicate, magnesium silicate and talc have a low acute oral toxicity. No studies were available for magnesium trisilicate.

No adverse effects were observed in short‐term toxicity studies in rats with calcium silicate, magnesium trisilicate or talc. The kidney effects observed in dogs were most probably related to the large amount of test compound consumed as a bolus dose by the animals. The effects on the kidney reported in guinea pigs could be due to higher concentrations of silicate in the primary urine as a consequence of lower glomerular filtration rates in guinea pigs (2.29 mL plasma/min per kg) as compared to rats (4.63 mL plasma/min per kg). The Panel noted that in humans the glomerular filtration rate (3.56 mL plasma/min per kg) is higher than in guinea pigs and, furthermore, kidney effects have not been found in humans in the EudraVigilance database despite the wide and long‐term use of high doses of magnesium trisilicate (up to 4 g/person per day) as an antacid over decades.

The Panel considered that the available data did not raise concern with respect to genotoxicity of calcium silicate (E 552), magnesium silicate (E 553a(i)), magnesium trisilicate (E 553a(ii)) and talc (E 553b) when used as food additives.

In a 2‐year study in rats, not performed according to current standards, calcium silicate had no effect on mortality at a dose up to 5,000 mg/kg bw per day. No gross pathology or histopathological findings that could be attributed to calcium silicate were observed in the 500 and 2,500 mg/kg bw per day groups. However, in the absence of clinical chemistry data, given the respiratory infection of animals and only 15 animals/sex per group, the Panel considered that this study was too limited to conclude on the chronic toxicity of calcium silicate. However, the Panel noted that no carcinogenic effects were reported in this study. There were no data for oral chronic toxicity/carcinogenicity of talc.

No reproductive toxicity studies were available. Prenatal developmental toxicity studies with calcium silicate by gavage during organogenesis in mice, rats and hamsters, and with talc in mice and rats, up to 1,600 mg/kg bw per day (the highest dose tested), showed no dose‐related developmental effects.

The Panel noted that cases of renal calculi were rarely reported considering the high number of exposed humans to magnesium trisilicate used as an antacid. The Panel applied the WHO algorithm for assessing the association between adverse events and drug intake, and found that the association between silicate antacid use and renal calculi was ‘possible’ but not ‘definite’, which does not exclude that the occurrence of renal calculi and intake of silicates would be a chance finding.

In the available data on subacute toxicity, genotoxicity and developmental toxicity studies, no adverse effects were reported for silicates and talc. From the only chronic toxicity study (with calcium silicate) available, there was no indication for carcinogenicity. However, due to the limitations of this study, it was not possible to draw a reliable conclusion on the chronic toxicity and carcinogenicity of calcium silicate. Furthermore, no subchronic and reproductive toxicity studies with silicates or talc were available.

The Panel noted that in 1991, the SCF established a group ADI ‘not specified’ for sodium silicate, silicon dioxide, calcium silicate, magnesium silicate and potassium silicate presumably on the basis that they share a common moiety. The Panel noted that more recent evidence suggested that this assumption might not be valid. Therefore, the Panel considered that on the evidence currently available, there is no mechanistic rationale for a group ADI for silicates and silicon dioxide. Therefore the Panel considered this group ADI obsolete.

Due to the limitations in the available toxicological database for individual silicates, the Panel was unable to derive ADIs for calcium silicate (E 552), magnesium silicate (E 553a(i)), magnesium trisilicate (E 553a(ii)) and talc (E 553b).

Silicates (E 552–553) are authorised in 28 food categories, including FC 0, according to Annex II to Regulation (EC) No 1333/2008. Their use in FC 0 means that they are ‘permitted in all categories of foods excluding foods for infants and young children, except where specifically provided for’. Silicates (E 552–553) are also authorised according to Annex III to Regulation (EC) No 1333/2008 (Parts 1, 2 and 5 A) in food‐improving agents and nutrients, except nutrients intended for foods for infants and young children. As such, silicates (E 552–553) can be found in many foods via carry‐over. The industry provided use levels for silicates (E 552–553) for their use according to Annex II. No analytical data on the concentration of these food additives in foods were made available by the Member States.

Dietary exposure to silicates (E 552–553) from their use as food additives according to Annex II was calculated for different exposure scenarios based on the provided use levels. 98% of the reported use levels referred to the use of talc (E 553b). This was in line with the information from the Mintel’s Global New Products Database (GNPD), showing that 89% of the foods labelled with silicates (E 552–553) were labelled to contain talc (E 553b). Therefore, the Panel noted that the calculated exposure reflects chiefly the exposure to talc (E 553b).

Additionally, 91% of the reported use levels were related to the use of silicates in food supplements. The Panel noted that the main food category labelled with silicates (E 552–553) in the Mintel GNPD was also food supplements (Appendix B). Therefore, the Panel considered the food supplements consumers only scenario as the most appropriate scenario for risk characterisation of silicates (E 552–553). Dietary exposure to silicates (E 552–553) via this exposure scenario was up to 31 mg/kg bw per day at the mean level in children and up to 46 mg/kg bw per day at the high (P95) level in the elderly.

The exposure assessment was hampered by several uncertainties. Overall, it was considered that the exposure was most likely overestimated due to the use levels used and assumptions made in the exposure assessment. Furthermore, the Panel noted that no foods belonging to an important contributing food category in all population groups, i.e. ripened cheese, were labelled to contain silicates (E 552–553) according to the Mintel GNPD.

The Panel noted that:

  • the absorption of silicates and talc was very low;
  • there was no indication for genotoxicity or developmental toxicity for calcium and magnesium silicate and talc;
  • no confirmed cases of kidney effects have been found in the EudraVigilance database despite the wide and long‐term use of high doses of magnesium trisilicate up to 4 g/person per day over decades.

However, the Panel considered that accumulation of silicon from calcium silicate in the kidney and liver was reported in rats, and reliable data on subchronic and chronic toxicity, carcinogenicity and reproductive toxicity of silicates and talc were lacking. Therefore, the Panel concluded that the safety of calcium silicate (E 552), magnesium silicate (E 553a(i)), magnesium trisilicate (E 553a(ii)) and talc (E 553b) when used as food additives cannot be assessed.

Based on the food supplement scenario considered as most representative for risk characterisation, exposure to silicates (E 552–553) for all population groups was below the maximum daily dose of magnesium trisilicate used as an antacid (4 g/person per day).

The Panel noted that there were a number of approaches which could decrease the uncertainties in the current toxicological database. These approaches include – but are not limited to – toxicological studies as recommended for Tier 1 approach as described in the EFSA Guidance for the submission of food additives (EFSA ANS Panel, 2012) and conducted with an adequately characterised material. Since the available data showed that nano particles are present in calcium silicate (E 552), the studies should be conducted with a material that contains a fraction of nanoparticles typical for silicates used as food additives.

The Panel recommended that:

  • the European Commission considers the revision of the EU specifications for calcium silicate (E 552), magnesium silicate (E 553a(i)), magnesium trisilicate (E 553a(ii)) and talc (E 553b) in order to include better definitions, assays in line with the definitions, and characterisation of particle size distributions (using appropriate statistical descriptors (e.g. range, median, quartiles) as well as the percentages (in number and by mass) of particles in the nanoscale (with at least one dimension < 100 nm). With regard to the characterisation of the particle size distribution, the analytical methodologies applied should comply with those recommended in the EFSA Guidance on risk assessment of the application of nanoscience and nanotechnologies in the food and feed chain (EFSA Scientific Committee, 2018)).
  • in first instance, toxicological studies as recommended for a Tier 1 approach as described in the EFSA Guidance for the submission of food additives (EFSA ANS Panel, 2012) should be conducted with adequately characterised material(s) in order to decrease the uncertainties in the current toxicological database.
  • more data on the actual usage and use levels of silicates (E 552, E 553a(i)), E 553a(ii)) should be provided because most of the data submitted were for talc (E 553b).
  • the European Commission considers lowering the current limits for toxic elements (arsenic, lead and mercury) in the EU specifications for calcium silicate (E 552), magnesium silicate (E 553a(i)), magnesium trisilicate, (E 553a(ii)) and talc (E 553b) in order to ensure that the food additives will not be a significant source of exposure to these toxic elements in food.
  • the European Commission considers inclusion of maximum limits for aluminium, nickel, fluoride and crystalline silica (alpha‐quartz) in the EU specifications for talc (E 553b).

Click here for the full opinion


Was This Post Helpful:

0 votes, 0 avg. rating

Share:

Ria Van Hoef