252
P. Volarath et al.
includes a summary of the toxicological data submitted to support the estimated
exposure level of the additive under its intended condition of use and the toxicologist reviewer’s conclusions/recommendations on the proposed use of the substance
based on a review of the toxicology information. The e-memos allow the chemical,
study-level toxicity data, and administrative data to be automatically captured into
CERES. Other tools that are under development are toxicity data entry (TDE) and
compound registration (CR) tools. The TDE tool is designed to allow the data harvest
team to directly enter the test-level toxicity data from the submissions that support
the study-level data collected from the e-memo. Similarly, the CR tools will allow
OFAS chemical registrars to directly enter and correct the food ingredients’ chemical
information in the knowledgebase.
12.4 Conclusions
OFAS review scientists evaluate available information about food ingredients
included in industry premarket submissions as well as data available in the public
domain or agency files to determine if there is reasonable certainty that the substances
are not harmful under their intended conditions of use. With recent advancements in
toxicological testing methods and computational science, OFAS review scientists are
expanding the use of in silico methods (SAR/QSAR, TK/PBTK, and bioinformatics)
as additional tools for evaluating the safety of food substances. In addition, in-house
tools (CERES and others) have been developed not only to organize and store the
institutional data, but also to provide modern computational capabilities that may
allow OFAS scientists to fill data gaps using existing information. Although the in
silico methods described in this chapter have been well-utilized in the field of clinical
drug development, they are not as routinely used for assessing safety of food ingredients. Considering that OFAS review scientists evaluate the safety of diverse classes
of food ingredients that cover a different chemical space than drugs, it is important to
understand the limitations of utilizing these computational methods for food ingredient safety assessment. Nevertheless, efforts have been made to overcome some of
the challenges related to different computational approaches used for their safety
assessment. Lastly, it is important to emphasize that there is a need for effective
communication and collaboration among scientists from all sectors: government,
industry, and academia, who are interested in the development and application of
computational methods for supporting an efficient food ingredient safety assessment
process.
Acknowledgements The authors would like to acknowledge the following peer reviewers from
the FDA for their intellectual contribution to the chapter: Dr. Michael Adams (OFAS, CFSAN), Dr.
Kirk Arvidson (OFAS, CFSAN), Dr. Jason Aungst (OFAS, CFSAN), Dr. Omari Bandele (OFAS,
CFSAN), Dr. Supratim Choudhuri (OFAS, CFSAN), Dr. Mary Ditto (OFAS, CFSAN), Dr. Jeffrey
Fisher (OR, NCTR), Dr. Suzanne Fitzpatrick (OCD, CFSAN), Celeste Johnston (OFAS, CFSAN),
Dr. Antonia Mattia (OFAS, CFSAN), Dr. Geoffrey Patton (OFAS, CFSAN), Dr. Catherine Whiteside
(OFAS, CFSAN), and Andrew Zajac (OFAS, CFSAN).
Précédent

- 262/416

Suivant