Personalized medicine asks if a new treatment will help a particular patient, rather than if it improves the average response in a population. Without a causal model to distinguish these questions, interpretational mistakes arise. These mistakes are seen in an article by Demidenko [2016] that recommends the "D-value," which is the probability that a randomly chosen person from the new-treatment group has a higher value for the outcome than a randomly chosen person from the control-treatment group. The abstract states "The D-value has a clear interpretation as the proportion of patients who get worse after the treatment" with similar assertions appearing later. We show these statements are incorrect because they require assumptions about the potential outcomes which are neither testable in randomized experiments nor plausible in general. The D-value will not equal the proportion of patients who get worse after treatment if (as expected) those outcomes are correlated. Independence of potential outcomes is unrealistic and eliminates any personalized treatment effects; with dependence, the D-value can even imply treatment is better than control even though most patients are harmed by the treatment. Thus, D-values are misleading for personalized medicine. To prevent misunderstandings, we advise incorporating causal models into basic statistics education.

The National Cancer Institute-led multidisciplinary Comparative Brain Tumor Consortium (CBTC) convened a glioma pathology board, comprising both veterinarian and physician neuropathologists, and conducted a comprehensive review of 193 cases of canine glioma. The immediate goal was to improve existing glioma classification methods through creation of a histologic atlas of features, thus yielding greater harmonization of phenotypic characterization. The long-term goal was to support future incorporation of clinical outcomes and genomic data into proposed simplified diagnostic schema, so as to further bridge the worlds of veterinary and physician neuropathology and strengthen validity of the dog as a naturally occurring, translationally relevant animal model of human glioma. All cases were morphologically reclassified according to a new schema devised by the entire board, yielding a majority opinion diagnosis of astrocytoma (43, 22.3%), 19 of which were low-grade and 24 high-grade, and oligodendroglioma (134, 69.4%), 35 of which were low-grade and 99 were high-grade. Sixteen cases (8.3%) could not be classified as oligodendroglioma or astrocytoma based on morphology alone and were designated as undefined gliomas. The simplified classification scheme proposed herein provides a tractable means for future addition of molecular data, and also serves to highlight histologic similarities and differences between human and canine glioma.