Toxicological Principles for the Safety Assessment of Food Ingredients
Chapter IV.B.3. Pathology Considerations in Toxicity Studies
Return to Redbook 2000 table of contents
- Description of the Process for Review of Pathology Data
- Common Problems Encountered During Review of Pathology Data
- General Recommendations for Reporting Pathology Data
Pathology data make up an essential part of the toxicology information submitted to FDA in support of the safe use of the use of food ingredients. The interpretation of pathology data and other safety data forms the basis for judgement about the safety of a product.
Specific recommendations concerning necropsy of test animals and microscopic examination of organs and tissues for short-term toxicity tests with rodents and non-rodents, subchronic toxicity tests with rodents and non-rodents, one-year toxicity tests with non-rodents, carcinogenicity studies with rodents, combined chronic toxicity/carcinogenicity studies with rodents, reproduction studies, and developmental toxicity studies can be found in Chapter IV.B.1.e. In general, these guidelines recommend that all animals in the studies be subjected to complete gross necropsy, all gross lesions and all protocol-required tissues and organs (see Chapter IV.B.1.e.iii.) from all control and high dose animals in the study should be examined microscopically; all gross lesions and target organs from all other dosed groups should also be examined microscopically.
This section on pathology considerations in toxicity studies describes the review process for pathology data, identifies common problems reviewers encounter in reviewing such data, and presents general guidelines for reporting pathology data. Although not addressed in this chapter, CFSAN pathologists also review and provide advice to petitioners on protocols for proposed toxicity studies; requests for such review should be directed to the CSO assigned to the petition (see Chapter II.A.).
Review of pathology data may begin with a request for pathology evaluation from regulatory review scientists or from the CAC. This happens when questions about the interpretation of pathology data arise during the scientific review of the toxicology information submitted in support of the safety of food ingredients . Requests for review are generally limited to specific interpretative questions, directing the reviewing pathologist's attention to findings in a particular organ or tissue. Occasionally, a reviewing pathologist is asked to examine all of the pathology findings in a study.
The pathology portion of the study report usually contains mean and individual organ weight parameters, clinical chemistry results, hematological measurements, summary incidences of observed pathological changes, and gross and microscopic pathology observations for individual animals. An evaluation memorandum from the regulatory review scientist may accompany the material; the memorandum contains summaries of toxicology information, including the results of previous toxicity studies and information from relevant scientific literature.
The reviewing pathologist usually begins his/her review by examining the experimental design and methods. He/she carefully reviews general indices of toxicity in test animals (for example, body weight gain, food consumption, clinical or hematologic findings, and organ weight changes); particular attention is paid to the survival of the animals and the number of animals alive at termination. All this information helps the reviewing pathologist evaluate the relationship of observed pathology changes to treatment.
Although the approach to pathology review may vary, the elements listed below are considered in all reviews. The reviewing pathologist:
- Determines how the percentage of animals with lesions in summary incidence tables has been calculated; for example, was the denominator based on the total number of animals in the study, or was it the number of animals for which a particular tissue or organ was examined microscopically;
- Compares gross and microscopic findings to ensure that all gross observations are accounted for by microscopic findings or by other suitable explanations;
- Examines the diagnostic terminology applied to lesions to determine whether it is contemporary and conventional;
- Checks to see that individual animal data provide adequate information on the location, size, and distribution of reported gross lesions;
- Considers the qualitative characteristics, severity of lesions, and the incidence figures in evaluating treatment-related differences among groups of experimental animals;
- Carefully evaluates control data before interpreting findings;
- Evaluates the discussion of significant pathological findings prepared by the study pathologist; and
- Correlates pathology findings, when appropriate, with other observations about treatment-related effects on test animals during the study.
When the pathology review is completed, a formal written report is submitted to the collaborating regulatory review scientist. The report discusses the pathological findings based on review of submitted material and the relationship of pathological findings to treatment. If questions about the pathology data remain, the report may recommend a request for additional, clarifying material.
A follow-up pathology review requires additional data. The additional information most often requested by the Agency is clarification of the diagnostic criteria used and historical control data for a specific lesion. The Agency may ask to review the original microscope slides; in some cases, the petitioner may be asked to prepare new slides from paraffin blocks or wet tissue for FDA review. The Agency's review of slides from a toxicity study provides an independent characterization of the lesions and enables the incidence of lesions to be verified.
When microscope slides and other materials are requested by the Agency for a follow-up review, the Agency provides instructions for their submission. Usually, microscope slides from an organ or tissue site should be arranged by treatment group, sex, and in the order of pathology accession numbers. If microscope slides are submitted according to the Agency's directions, the follow-up review will be expedited.
The timely review of pathology data is sometimes hindered by missing, inaccurate, or incomplete information. These problems are often encountered in submissions to the Agency; a general discussion of problems resulting from information deficiencies is presented below. A more detailed discussion of this subject has been published.(1)
One of the most common problems causing delay in the review of pathology data is the lack of adequate morphologic descriptions of lesions. It is difficult to assess the significance of reported lesions without information on their diagnostic criteria, distribution, and severity. This is particularly important when the terminology for lesions is controversial.
The use of multiple diagnostic terms without explanation for describing a single type of lesion can present problems for the reviewing pathologist. Further clarification is needed to indicate whether two or more terms are being used interchangeably or the results of the study have been evaluated by more than one pathologist, each using different terms for the same morphologic change. For example, in one study the terms "hepatocellular carcinoma" and "hepatoma, malignant" were used in the same set of diagnoses. In another report, four different terms--"c-cell," "clear cell," "light cell," and "parafollicular cell"--were used to describe rat thyroid lesions. In both instances, reasons for using multiple terms for the same diagnosis were not provided.
Differences in the use of diagnostic terms have been encountered when more than one pathologist has examined slides: for example, a study was submitted in which tissues from about one-third of the animals were evaluated by the study pathologist and the remainder were evaluated by a consulting pathologist. The diagnostic terminology was not consistent between pathologists and no attempt was made to explain the inconsistencies in the study report. Although the data appeared to show treatment-related effects, these were subsequently attributed to the way different categories of lesions were summarized.
Incomplete gross descriptions have made it difficult to correlate gross pathology findings with microscopic diagnoses. When microscopic findings do not correlate with gross descriptions, the reviewer must attempt to determine if important information is missing. The report should describe steps taken to resolve discrepancies between gross findings and microscopic diagnoses (for example, recuts of paraffin blocks or additional samples taken from wet tissues).
Inaccurate summary numbers resulting from incorrect counts or calculations have caused difficulty in reviewing pathology data. When pathology data are summarized, all experimental animals should be accounted for and incidence figures should be based on the numbers of animals, organs, and tissues actually examined.
Often, submissions fail to adequately discuss the significance of the results of pathology evaluations. Some reports summarize conclusions but do not explain how the conclusions were deduced from the available pathology data. Some reports base conclusions solely on the results of statistical analyses of data, ignoring broader conclusions that may be discerned from considering all relevant biological information from a study.
The pathology section in the report of a toxicity study generally includes an introductory statement and sections on materials and methods, results and discussion, and summary and conclusions.
When pathology data are reported separately from the toxicity study, adequate information about the experimental design and methodology of the toxicity study should be included. This information should include the species and strain of the experimental animals, details about the administration of the test compound, number of experimental and control groups, number of animals in each group, type and frequency of in-life observations including clinical chemistry measurements and hematological examinations, and the scope of gross and microscopic evaluation of tissues. In general, information provided should be sufficient to enable a reviewer to evaluate the quality of the pathology data.
Deviations from the original protocol should be explained. For example, if tissues from low- and mid-dose groups were not scheduled for microscopic examination but were examined, the appropriate protocol amendment or reason for this deviation should be given.
The arrangement of tabular information in an easily comprehensible format is especially important for facilitating review. Table titles and row and column headings should be brief but informative. In the tables showing the individual animal findings, descriptive diagnostic categories should be informative. Redundancy of categories of lesions should be avoided. Morphologic diagnoses should reflect currently accepted criteria. Whenever multiple categories of lesions are grouped under a common "diagnosis," the rationale for grouping should be provided. When multiple diagnoses are not grouped under a common diagnosis, it will be assumed that morphologic differences preclude grouping. Severity grades as well as information on the distribution of a lesion within an organ or tissue should be provided; these observations are particularly important when progression of lesions and effects of different dosages are being studied. In paired organs such as adrenal glands, gonads, and kidneys, certain lesions, when appropriate, should be indicated as unilateral or bilateral. All gross lesions should be accounted for by microscopic findings or a written explanation.
Summary tables in the results section of a report should clearly indicate the number of animals, organs, and tissues actually examined. Unless the number of tissues examined in animals of each group is indicated, the incidence figures or mean values indicating effects are subject to question. Summary tables should be free of double counting. In determining incidence, the denominators should reflect actual numbers of animals whose tissues were examined, not just the number of animals originally assigned to each group. The figure for the number of tissues examined should clearly show any adjustments that reflect loss, autolysis, or missing tissue: For example, the accurate incidence of lesions involving the adrenal medulla should be based upon how many adrenal sections (for an animal) from both adrenal glands contained sufficient medullary tissue for microscopic examination. In summarizing lesions that are disseminated, e.g., tumors of the lymphoreticular tissue, the incidence figures should reflect the number of animals with these lesions, not just the presence of the disease in individual organs.
A cross-reference table that lists individual lesions on the vertical axis and individual animal numbers along the horizontal axis should be included, if possible. This is convenient both for reviewing lesions within an animal and for comparing lesions across animals in a group or among different groups.
The report should generally contain an animal disposition table that provides the pathology accession number, sex, group designation, number of days on the study, and fate of the animals (for example, interim sacrifice, moribund sacrifice, found dead, or terminal sacrifice). This serves as a ready reference for the Agency's scientific reviewers and eliminates the need to develop this information from individual animal data.
Finally, the report should include a section that specifically discusses the pathology data. This pathology narrative should provide an overview of the pathology findings from the study pathologist's perspective. A discussion that includes qualitative description of lesions and that highlights differences among treated and control groups is an essential part of the interpretation and evaluation of pathology data. The description of morphologic characteristics of lesions is particularly important where terminology may be controversial or misunderstood. Remarks about possible pathogenesis, strengthened by references to the scientific literature, could be an important part of the pathologist's narrative. Significant events, such as a disease outbreak during the study, and the impact of such events on the study outcome should be discussed. Differences in the incidence of key histopathologic findings among groups should be discussed; if observed differences are not regarded as treatment-related, then the basis for this conclusion should be provided.
- Dua, P.N., and B.A. Jackson (1988). Review of Pathology Data for Regulatory Purposes. Toxicologic Pathology. 16:443-450.