These are partial results from a larger study: "This report presents partial findings from these studies. The occurrences of two tumor types in male Harlan Spr

These are partial results from a larger study: "This report presents partial findings from these studies. The occurrences of two tumor types in male Harlan Sprague Dawley rats exposed to RFR, malignant gliomas in the brain and schwannomas of the heart, were considered of particular interest" So a key question should be: how many variables was the study looking at? If they were testing 40 different things, you'd expect two 'significant' results by chance. The discussion also mentions that pooling other studies in the program, the rate of gliomas in male controls is 11/550, rather than the 0/90 from this study. The incidence in the test group (pooling all dose levels) is 11/540, which is indistinguishable from that larger control group. The numbers for heart schwannomas are a bit more compelling, but that's now just one variable, so the question of how many things the experiment is looking at is crucial.