Association between cage shelf level and spontaneous and induced neoplasms in mice.

Data from a chronic feeding study with 2-acetylaminofluorene [(2-AAF) CAS: 53-96-3; N-fluoren-2-yl-acetamide] done on 20,880 female BALB/c mice were analyzed for associations between cage shelf level and occurrence of induced and spontaneous neoplasms. Each cage was maintained on a rack at a given shelf level throughout the experiment, allowing analysis of data by shelf level. Differences in the crude incidence of 2-AAF-induced liver and bladder neoplasms appeared to be shelf related, but these differences were small and disappeared when shelf-level analyses of time-to-tumor onset distributions were performed. There was evidence for a shelf-level influence on 5 of the 6 major spontaneous neoplasms noted. Time to onset of uterine polyps and reticulum cell sarcomas was significantly delayed on the top shelf of five of six animal rooms. Also, there was a significant delay in onset of lymphomas, adrenocortical adenomas, and lung alveolar cell tumors on the top shelf when data were combined from all six animal rooms, but these delays on the top shelf were significant in no more than two of six animal rooms when rooms were analyzed separately. There was no indication of any shelf-level influence on the development of harderian gland adenomas. In conclusion, shelf level is an environmental factor that should be considered in the design and analysis of carcinogenesis studies.