UC Berkeley

In the past few decades, the role of basic science and controlled clinical trials has changed the communication of function and utility of diagnostics and therapies. This resulted in efforts promoting clinical decisions to follow evidence-based medicine, precision medicine and personalized medicine. The communication concerning a drug’s utility, and the advantage it provides to an individual patient influence patients and physicians attitudes. How? The history of Tamoxifen’s development and its use in the treatment of breast cancer will serve as the example for the changes in medical practice. The rationale for tamoxifen’s development is the growth dependency of breast cancer cells on estrogen. Yet, the development originated from the understanding of the physiological relationship between the ovaries and the breasts, prior to the discovery of the hormone estrogen and the estrogen receptor (ER) its target protein. Does the fact that majority (70%) of women with breast cancer express ER and/or progesterone receptor (PR) while only a fraction (25%- 50%) respond to the treatment, puts the generalizability of the rationale in question? The empirical discoveries and the explanations provide an important platform to interrogate the relationship between the science utilized by contemporary medicine and the practice of medicine. Claims of specificity and robustness made in scientific work may not support causal relatedness. Therefore, such claims may not support explaining or predicting medical problems or generate plausible hypothesis to paths to treatment. While a mechanistic causality model is hailed rhetorically for supporting scientific and clinical rationale, in practice associative descriptions, at times inductive and at times deductive, are employed in practice. When correlated or combined they form weak probabilities in support of predictive outcome in practice. In summary, understanding the limits each data set provides, may help develop specific quantitative models to test utility, context and predictive value.