Agency Response Letter GRAS Notice No. GRN 000094
CFSAN/Office of Food Additive Safety
April 18, 2006
Claire L. Kruger, Ph.D.
Environ International Corporation
4350 North Fairfax Drive, Suite 300
Arlington, VA 22203
Re: GRAS Notice No. GRN 000094
Dear Dr. Kruger:
The Food and Drug Administration (FDA) is responding to the notice, dated December 18, 2001, that you submitted on behalf of Ross Products Division, Abbott Laboratories (Ross) in accordance with the agency's proposed regulation, proposed 21 CFR 170.36 (62 FR 18938; April 17, 1997; Substances Generally Recognized as Safe (GRAS); the GRAS proposal). FDA received the notice on December 21, 2001, filed it on January 2, 2002, and designated it as GRAS Notice No. GRN 000094. Subsequent to the original submission, Ross or Environ provided 10 amendments to address issues raised by FDA during the review of GRN 000094 (amendments dated June 14, 2002; October 28, 2002; December 13, 2002; February 2, 2004; July 30, 2004; August 10, 2004; December 7, 2004; February 1, 2005; July 15, 2005; and August 4, 2005) and several other amendments to clarify confidentiality or non-confidentiality of materials submitted to FDA.
The subjects of the notice are arachidonic acid-rich oil from Mortierella alpina (AA-rich fungal oil(1)) and docosahexaenoic acid-rich oil from tuna (DHA-rich tuna oil). Although Ross refers to these ingredients as AA-rich oil (SUNTGA40S) and DHA-rich oil, FDA is including the sources of these ingredients for clarity. In this letter, FDA refers to these ingredients as AA-rich fungal oil and DHA-rich tuna oil. The notice informs FDA of the view of Ross that AA-rich fungal oil and DHA-rich tuna oil are GRAS, through scientific procedures, to provide a combination of arachidonic acid (AA) and docosahexaenoic acid (DHA) in infant formula at levels of use specified for each of three categories of infant formula, based on the intended subpopulation of infants who would consume the infant formula, as described in Table 1. Although the notice specifically mentions these formulas by their trade names, in this letter FDA refers to them by "category" of use (i.e., preterm infant formula used by hospitalized premature infants, preterm infant formula used by post-discharge premature infants, and term infant formula.)
|Formula||Target Ratio of AA:DHA||Target mean AA concentration|
(g/100g of fatty acids)
from AA-rich fungal oil
|Target mean DHA concentration|
(g/100g of fatty acids)
from DHA-rich tuna oil
|Preterm infant formula used by hospitalized premature infants||1.6:1||0.40||0.25|
|Preterm infant formula used by post-discharge premature infants||2.7:1||0.40||0.15|
|Term infant formula||2.7:1||0.40||0.15|
The use of "AA-rich fungal oil" and "DHA-rich tuna oil" in this letter should not be considered an endorsement or recommendation of these terms as appropriate common or usual names for the purpose of declaring these substances in the ingredient statement of foods that contain these ingredients. Under 21 CFR 101.4, all ingredients must be declared by their common or usual name. In addition, 21 CFR 102.5 outlines general principles to use when establishing common or usual names for non-standardized foods. Issues associated with labeling and the appropriate common or usual name of a food are the responsibility of the Office of Nutritional Products, Labeling, and Dietary Supplements in FDA´s Center for Food Safety and Applied Nutrition.
Data and information that Ross presents to support its GRAS determination
As part of its notice, Ross includes the statement of a panel of individuals (Ross' GRAS Panel) who evaluated the data and information that are the basis of Ross' GRAS determination. Ross considers the members of its GRAS panel to be qualified by scientific training and experience to evaluate the safety of substances added to food.
Ross' GRAS panel reviewed the data on the short-term and long-term toxicity of AA-rich fungal oil and DHA-rich tuna oil, human exposure to AA and DHA through human milk, and human clinical studies evaluating infants who consumed AA-rich fungal oil and DHA-rich tuna oil. Ross' GRAS panel concluded that AA-rich fungal oil and DHA-rich tuna oil are GRAS under the intended conditions of use. Following Ross´ submission of several amendments to its original GRAS notice, Ross' GRAS panel re-confirmed its support of Ross' determination that AA-rich fungal oil and DHA-rich tuna oil are GRAS under the intended conditions of use.
Ross describes published information about the identity and characteristic properties of AA-rich fungal oil (SUNTGA40S), including published information about the source for AA-rich fungal oil and the method of manufacture of this oil. The source of the AA-rich fungal oil is the fungal species M. alpina Peyronel 1S-4. Ross cites the use of M. alpina in the production of arachidonic acid single-cell oil, an ingredient that is the subject of GRN 000041. Ross also notes that there is no evidence from the literature that M. alpina is pathogenic or toxigenic, and provides a mycotoxin batch analysis to demonstrate that there is no detectable mycotoxin contamination in the AA-rich fungal oil.
The notice describes the conditions under which the fungal culture is grown and fermented, producing a biomass which is filtered, dried, and stored until hexane extraction and further purification of the oil. Ross provides critical control points for the manufacturing and purification processes. Ross states that a fermentation batch not meeting process specifications would be rejected, and that both the seed and main cultures at the end of cultivation are checked for contamination. Ross provides results of batch analyses (including fatty acid composition) and specifications for the food-grade AA-rich fungal oil. Specifications include quality indicators, AA concentration (greater than or equal to 40 percent of total fatty acids), and residual hexane. Ross provides a statement of compliance that it will not use, in infant formula, any AA-rich fungal oil that does not meet its specifications.
Ross describes generally available information about the identity and characteristic properties of DHA-rich tuna oil, including published information about the sources and the method of manufacture of this oil. The sources of the DHA-rich tuna oil are various species of tuna (Thunnus tonggol, T. obesus, T. albacares, T. alalunga, and Katsuwonus pelamis), currently caught in the waters off Peru. Ross notes that there is a substantial body of literature addressing the physiological and biochemical effects, and the safety, of diets rich in fish or fish oils. Ross further notes that DHA-rich oil from tuna is used in infant formulas, yogurt, bread, and fish sausage in Japan; and in infant formulas, bread, and fish sausage in Korea.
The notice describes the manufacture of the DHA-rich tuna oil by standard oil processing techniques that include hot water extraction, separation of liquid and solid fractions by filter press, and collection of the crude tuna oil extract by centrifugation. The crude tuna oil is further processed by 1) degumming, 2) neutralization, 3) bleaching, 4) molecular distillation, 5) steam deodorization, and 6) blending with mixed tocopherols or ascorbyl palmitate and monoglycerides. Ross provides results of batch analyses (including fatty acid composition) and specifications for the food-grade DHA-rich tuna oil. Ross provides specifications for quality indicators, DHA concentration (greater than or equal to 22 percent of total fatty acids), eicosapentaenoic acid (EPA) concentration (less than or equal to 8 percent of total fatty acids), and DHA:EPA ratio (greater than or equal to 3.1:1), and for potential environmental contaminants such as mercury, dioxins, organochlorine compounds (including polychlorinated biphenyls (PCBs)), and polycyclic aromatic hydrocarbons (PAHs). Ross states that if there are any detectable quantities of PAHs, then the batch will either be rejected for use in infant formula or subjected to activated charcoal purification and then re-analyzed to assure compliance with the PAH specification. Ross provides a statement of compliance that it will not use, in infant formula, any DHA-rich tuna oil that does not meet its specifications.
In the notice, Ross notes that three organizations, (the British Nutrition Foundation, the Food and Agricultural Organization/World Health Organization, and the International Society for the Study of Fatty Acids and Lipids) have published recommendations for fatty acid addition to infant formulas. Ross notes that a comparison of the recommendations to the estimated daily intakes and maximum intakes of AA and DHA from the formulas used in the clinical trials reveals that the AA and DHA present in infant formulas as a result of its proposed addition of AA-rich fungal oil and DHA-rich tuna oil are at, or slightly below, the recommendations of these organizations.
Ross provides a mean estimated daily intake of the AA-rich fungal oil and the DHA-rich tuna oil and the corresponding fatty acids (respectively, AA and DHA).(2) Ross bases these estimates on formula consumption data from its two studies (i.e., one for preterm infants and one for term infants). The estimates listed in Table 2 show the highest mean intake value, on a milligrams per kilogram body weight per day (mg/kg bw/d) basis, for each population of infants receiving the particular infant formula.(3)
|Formula||AA-rich fungal oil*|
|DHA-rich tuna oil|
|Preterm infant formula used by hospitalized premature infants|
|Preterm infant formula|
used by post-discharge
(2 months CA)
|Term infant formula|
|* Ross calculates its estimate of the intake of SUNTGA40S based on a percentage (62.5 percent) of the actual consumption of SUNTGA25 in its two clinical studies.|
** For preterm infants, corrected age (CA) is considered as the age relative to 40 weeks gestational age. For example, an infant born at 32 weeks gestational age would be at term CA at 8 weeks of age.
In the notice, Ross acknowledges that FDA has previously raised concerns about the consumption of high levels of two fatty acids (EPA and DHA), which may increase bleeding time, increase levels of low-density lipoprotein cholesterol, and have an effect on glycemic control in non-insulin dependent diabetics (menhaden oil final rule; 62 FR 30571; June 5, 1997). Ross notes that FDA, in affirming the GRAS status of fish oil from menhaden, concluded that the use of menhaden oil as a direct food ingredient is GRAS, provided that the combined daily intake of EPA and DHA from consumption of menhaden oil does not exceed 3 grams per person per day (g/p/d). Ross calculates that an intake level of 3 g/p/d for a 60 kg person corresponds to an intake level of 50 mg/kg bw/d. On the basis of highest mean estimated daily intake values (see Table 2), Ross notes that the combined intake of EPA plus DHA from infant formula containing DHArich tuna oil is below 50 mg/kg bw/d, for both the higher (0.25 percent of fatty acids as DHA) and the lower (0.15 percent of fatty acids as DHA) intended use level of fatty acids.
Ross describes two published clinical studies that were designed to determine whether adding AA and DHA from different sources to infant formulas support growth and selected developmental outcomes. Study AF92 and Study AG38 involved 404 term and 470 preterm infants, respectively. In all, 240 of the term or preterm infants enrolled in these studies consumed formulas containing SUNTGA25 AA-rich fungal oil and DHArich tuna oil. Infants were fed formula from the time of the first enteral feeding though 12-months term corrected age for preterm or through the first 12 months of life for term infants. During the study, growth, development, and safety parameters were measured. Ross concludes from these two studies that the addition of an AA-rich fungal oil and DHA-rich tuna oil at study levels had no adverse effect on the infant growth, multiple indices of development, morbidity, incidence of adverse events, antibiotic use, clinical chemistries, or antioxidant status compared to infants fed the unsupplemented formulas. Ross concludes that intake of AA-rich fungal oil and DHA-rich tuna oil from infant formulas, at levels and ratios used in these studies,(4) is safe and well tolerated by infants.
Ross discusses a published neonatal piglet study in which piglets were fed AA-rich fungal oil and DHA-rich tuna oil. The piglets were fed formula containing either AArich fungal oil, DHA-rich tuna oil, or AA-rich fungal oil + DHA-rich tuna oil at levels targeted to provide intakes 3-fold to 5-fold higher than the targeted estimated intake levels in infants. After Day 19 of the feeding course, hematology and clinical chemistry parameters were measured and samples of various tissues were analyzed with no adverse test article-related effects observed. Ross notes that several statistically significant differences in absolute organ weights were observed in neonatal piglets. Ross discusses its reasons for concluding that these differences were not adverse effects and that it is reasonable to assume that no adverse effects would occur from ingestion of AA-rich fungal oil and DHA-rich tuna oil at the intended levels of exposure for human infants.
Ross describes several unpublished studies, including in vitro and rodent toxicology studies of AA-rich fungal oils (SUNTGA25 and SUNTGA40S) and DHA-rich tuna oil. The in vitro studies include tests for estrogenic activity and mutagenicity; the rodent toxicology studies include acute, subacute, and 90-day subchronic toxicity studies.
Based on all of these data and information, Ross concludes that its intended uses of AA-rich fungal oil and DHA-rich tuna oil are GRAS.
FDA´s evaluation of the data and information in Ross´ notice and amendments
FDA asked Ross to address chemical, toxicological, and clinical issues related to FDA´s review of Ross´ GRAS notice for use of AA-rich fungal oil and DHA-rich tuna oil in term and preterm infant formulas. These issues focused on two areas: (1) the method of manufacture of the DHA-rich tuna oil and the effects of this method of manufacture on the chemical composition of the oil, particularly in relation to removal of potential fish oil contaminants; and (2) the design, conduct, and results of Ross´ clinical studies, particularly the preterm infant study (Study AG38).
In reviewing Ross´ information about the DHA-rich tuna oil, FDA considered whether the specifications and associated analytical limits of detection for potential contaminants, as stated in the notice, were sufficient to assure safety. FDA considered the method of manufacture of the DHA-rich tuna oil and the availability of literature addressing the ability of manufacturing processes described by Ross to remove potential contaminants from fish oils and produce a food-grade ingredient. Ross described a molecular distillation (also known as short-path distillation) process. Briefly, the oil is heated under very low pressure and presented in a continuous thin film allowing the physical removal of volatile and semi-volatile contaminants in a short period of time, thus minimizing exposure of the oil to heat. Ross asserts that contaminants removed by this process include PCBs, PAHs,(5) and other potential environmental pollutants. Ross also notes that the method has been validated and that the distillation conditions exceed those described by Bimbo (1990) for vacuum stripping of fish oils to remove cholesterol and chlorinated hydrocarbons (Ref. 1). Ross provided updated tables of batch analyses, published methods of analyses with low limits of detection, tables of product specifications for contaminants, and an updated statement of compliance with these specifications.
During FDA´s review of information in the notice that described the numbers and types of serious adverse events (SAEs) reported in Study AG38, FDA noted a near doubling of the incidence of apnea reported for infants ingesting formula supplemented with AA-rich fungal oil and DHA-rich tuna oil. FDA asked Ross to address questions pertaining to the design, conduct, and results of Study AG38, as these specifically relate to the significance of the SAEs. Ross provided additional data and information about Study AG38, including: the methods of defining and reporting SAEs, both in-hospital and postdischarge; multiple statistical approaches for analyzing the significance of reported SAEs and discussions of the appropriateness of each method; and neonatal history, acute event(s) data, and emergency room and/or in-patient evaluations for the infants that experienced apnea reported as an SAE. After receiving the additional information, FDA conducted its own independent analysis of the reports of apnea.
In light of the additional data and information that Ross submitted, FDA asked to hear the views of Ross´ GRAS panel as they relate to Ross´ notice that AA-rich fungal oil and DHA-rich tuna oil are GRAS for use in infant formula. In a meeting with FDA on May 17, 2004, Ross´ GRAS panel discussed its evaluation of (1) the method of manufacture as it relates to the suitability of DHA-rich tuna oil in infant formulas and (2) the incidence of apnea in the preterm infant study. In its evaluation of the method of manufacture and suitability of DHA-rich tuna oil in infant formulas, the panel considered the nature of the production process and the analytical results of production batches. In its evaluation of the incidence of apnea in the preterm infant study, the panel considered the clinical antecedents to and consequences of apnea described in Ross´ February 2, 2004, amendment as well as time lines of in-hospital feeding initiation; initial occurrence of in hospital vomiting; onset of in-hospital apnea; and feeding history events. After the meeting, the panel submitted to Ross, who transmitted to FDA, a statement dated July 30, 2004, that, in light of all the information provided by Ross, the panel re-confirms its original conclusion that AA-rich fungal oil and DHA-rich tuna oil are GRAS under the intended conditions of use.
Based on all of the information provided by Ross, as well as other information available to FDA, the agency has no questions at this time regarding Ross´ conclusion that AA-rich fungal oil and DHA-rich tuna oil are GRAS under the intended conditions of use (i.e., when added to preterm infant formulas, for use by hospitalized premature infants at target mean concentrations (g/100g fatty acids) of 0.40 percent AA and 0.25 percent DHA; when added to preterm infant formula, for use by post-discharge infants, and to term infant formula, at target mean concentrations of 0.40 percent AA and 0.15 percent DHA), provided that the ingredients, as produced and analyzed, are in compliance with the specifications and associated analytical limits of detection for potential contaminants, as stated in Ross´ notice. The agency has not, however, made its own determination regarding the GRAS status of the subject use of AA-rich fungal oil and DHA-rich tuna oil. As always, it is the continuing responsibility of Ross to ensure that food ingredients that the firm markets are safe, and are otherwise in compliance with all applicable legal and regulatory requirements.
Under section 412 of the Federal Food, Drug, and Cosmetic Act, Ross, or any other manufacturer of a new infant formula, must make a submission to FDA, providing required assurances about the formula, at least 90 days before the formula is marketed.
FDA´s response to Ross GRAS notice does not eliminate the responsibility of any infant formula manufacturer who intends to market an infant formula that contains AA-rich fungal oil and DHA-rich tuna oil to submit the notification about the new formulation required by section 412.
In accordance with proposed 21 CFR 170.36(f), a copy of the text of this letter, as well as a copy of the information in the notice that conforms to the information in proposed 21 CFR 170.36(c)(1), is available for public review and copying on the homepage of the Office of Food Additive Safety (on the Internet at http://www.cfsan.fda.gov/~lrd/foodadd.html).
Laura M. Tarantino, Ph.D.
Office of Food Additive Safety
Center for Food Safety and Applied Nutrition
1. Bimbo, A.P. 1990. “Processing of Fish Oils” in Fish Oils in Nutrition (Ed. M.E. Stansby). Van Nostrand Reinhold, New York, NY, pp. 181-225.
(1)The AA-rich fungal oil that is the subject of GRN 000094 contains 40 percent arachidonic acid. Ross refers to this oil as “SUNTGA40S” to distinguish this ingredient from “SUNTGA25,” a related oil produced from M. alpina, that contains only 25 percent arachidonic acid.
(2)FDA independently calculated an estimate of mean intake of AA and DHA from Ross’ intended uses of AA-rich fungal oil and DHA-rich tuna oil, respectively. FDA’s estimates of mean intake for AA and DHA were in general agreement with Ross' corresponding estimates of intake of these fatty acids for preterm and term infants.
(3)FDA notes that these estimates are based on actual intakes of infant formula and reflect some variation in the mean level of addition of AA and DHA as compared to target levels. For example, Ross notes that for preterm hospitalized infants, a mean level of 0.43 percent AA (range of 0.42 - 0.45 percent) and 0.27 percent DHA (range of 0.24 - 0.32 percent) was provided.
(4)The levels and ratios used in these studies are equal to or slightly higher than the levels and ratios that Ross determines to be GRAS in its notice, GRN 000094.
(5)FDA notes that less volatile, heavy PAHs would not be expected to be removed by the molecular distillation process as described by Ross in its notice. However, Ross has provided assurances that if PAHs are detected in the tuna oil, an additional purification step with activated charcoal will be used.