Prostate-specific antigen in prostate cancer: a case study in the development of a tumor marker to monitor recurrence and assess response

Abstract

The serum marker known as prostate-specific antigen (PSA) has established itself as the most important tool for the early detection of prostate cancer. However, more recent data indicate that (post-treatment) PSA and PSA kinetics can be used to predict the outcome of a variety of therapeutic interventions including radical prostatectomy, radiation therapy, androgen deprivation, and treatment of hormone-refractory prostate cancer. PSA recurrence after radiation therapy is now accepted as a harbinger of developing metastatic disease. The American Society for Therapeutic Radiation Oncology (ASTRO) consensus definition is the most widely accepted definition of failure after radiation therapy. Rather than using a specific PSA cutoff, three consecutive PSA rises was felt to be a more reliable indicator of biochemical failure. The PSA nadir (the lowest PSA level achieved after therapeutic intervention) also appears to correlate with the likelihood of remaining disease-free. Similarly, a rapid doubling time is a significant predictor of developing distant metastases. The most appropriate definition for biochemical (PSA) failure following radical prostatectomy is usually considered to be a non-zero value. As is the case after radiotherapy, there appears to be a relationship between the rate of rise of the PSA and the risk of distant failure following radical prostatectomy. In patients with metastatic disease, multiple studies appear to indicate that a fall in PSA, however measured, appears to be predictive of improved outcome in prostate cancer patients treated with androgen deprivation. Multiple reports of trials in the treatment of hormone-refractory prostate cancer (HRPC) appear to substantiate the observation that patients who have a greater than 50% decline in PSA have an improved survival. Correlation of PSA declines with other markers of clinical benefit, including clinically significant "subjective" end points such as pain control, have strengthened the argument that a PSA decline can serve as an intermediate endpoint in clinical trials involving HRPC patients.