Preliminary data from a retrospective cohort of 8,669 patients with clinically localized prostate cancer treated with either radical prostatectomy or radiation therapy suggested that short post-treatment PSA doubling time (<3 months in this study) fulfills some criteria as a surrogate endpoint for all-cause mortality and prostate cancer-specific mortality after surgery or radiation therapy.
Likewise, a retrospective analysis (SWOG-S9916 [NCT00004001]) showed that PSA declines of 20% to 40% (but not 50%) at 3 months and 30% or more at 2 months after initiation of chemotherapy for hormone-independent prostate cancer, and fulfilled several criteria of surrogacy for overall survival (OS).
These observations should be independently confirmed in prospective study designs and may not apply to patients treated with hormonal therapy. In addition, there are no standardized criteria of surrogacy or standardized cutpoints for adequacy of surrogate endpoints, even in prospective trials.
Follow-up after radical prostatectomy
After radical prostatectomy, a detectable PSA level identifies patients at elevated risk of local treatment failure or metastatic disease; however, a substantial proportion of patients with an elevated or rising PSA level after surgery remain clinically free of symptoms for extended periods of time. Biochemical evidence of failure on the basis of elevated or slowly rising PSA alone, therefore, may not be sufficient to initiate additional treatment.
For example, in a retrospective analysis of nearly 2,000 men who had undergone radical prostatectomy with curative intent and who were followed for a mean of 5.3 years, 315 men (15%) demonstrated an abnormal PSA of 0.2 ng/ml or higher, which is considered evidence of biochemical recurrence. Among these 315 men, 103 (34%) developed clinical evidence of recurrence. The median time to the development of clinical metastasis after biochemical recurrence was 8 years. After the men developed metastatic disease, the median time to death was an additional 5 years.
Follow-up after radiation therapy
For patients treated with radiation therapy, the combination of clinical tumor stage, Gleason score, and pretreatment PSA level is often used to estimate the risk of relapse.[Level of evidence: 3iDii] As is the case after prostatectomy, PSA is often followed for signs of tumor recurrence after radiation therapy. After radiation therapy with curative intent, persistently elevated or rising PSA may be a prognostic factor for clinical disease recurrence; however, reported case series have used a variety of definitions of PSA failure. Criteria have been developed by the American Society for Therapeutic Radiology and Oncology Consensus Panel.[58,59] It is difficult to base decisions about initiating additional therapy on biochemical failure alone. The implication of the various definitions of PSA failure for OS is not known, and, as in the surgical series, many biochemical relapses (rising PSA only) may not be clinically manifested in patients treated with radiation therapy.[60,61]