Why shouldn the patient simply pick the prostate cancer treatment with the highest predicted progression-free probability (PFP)?
There are several reasons: • A rising PSA — one of the factors that these models predict — is not the most important endpoint for decision-making purposes. A rising PSA after treatment has no symptoms until the disease progresses, which is generally many years later. What we would like to be able to predict is symptoms of disease progression, which is considerably more difficult to predict. In other words, a rising PSA is not the most important endpoint for decision making purposes. • Treatment complications are different for each patient. The patient needs to discuss potential complications with his doctor. • Disease progression following each of the treatments may have differing prognoses. This means that subsequent progression for the patient who has a rising PSA following surgery may be more or less rapid than the progression for the patient with a rising PSA following a form of radiation. In other words, the endpoint being predicted may mean different things across treatments. W
