Predictors of extracapsular extension and positive margins in African American and white men.

OBJECTIVE: Radical retropubic prostatectomy (RRP) pathology from African American (AA) and White men from 1988 to 1999 was examined to determine if the pre-treatment factors PSA, clinical stage, biopsy grade, age at surgery, and year of surgery (YOS) were predictive of extracapsular extension (ECE) and positive margins for each ethnic group. METHODS: Clinical and pathologic data was collected on 179 AA and 548 white men undergoing RRP from 1988 to 1999 at a tertiary military medical facility. Logistic regression with multivariate analysis was used to determine which pre-operative data-points were predictive of pathologic ECE and positive margins for each ethnic group. RESULTS: PSA, biopsy grade, age, and YOS (more recent years had better surgical pathology) were predictive of ECE for AA and white men. PSA, biopsy grade, and YOS were predictive of positive margins for AA men, while PSA and YOS were predictive of positive margins for white men. PSA continues to be a strong predictor of ECE and positive margins for both AA and white men. However, we describe for the first time, YOS being predictive of ECE and positive margins for both AA and White men, using multivariate regression analysis. CONCLUSION: This is thought to be reflective of the improving public awareness of prostate cancer that has occurred during the PSA-era, resulting in patients participating in screening programs and being diagnosed earlier. Close follow-up of these patients is warranted to determine if the improved pathologic stage of those patients treated more recently translates into improved disease-specific mortality.

Improved prostate cancer-specific survival and other disease parameters: impact of prostate-specific antigen testing.

OBJECTIVES: To determine how the implementation of prostate-specific antigen (PSA) testing has affected disease-specific survival and other characteristics of prostate cancer. METHODS: Data were collected on all patients with prostate cancer diagnosed between 1988 and 1998 and registered in the Center for Prostate Disease Research Database at Walter Reed Army Medical Center. Statistical analyses were used to summarize trends over time in survival, mortality, and clinical stage. RESULTS: Between 1988 and 1998, a total of 2042 patients with prostate cancer were registered at Walter Reed Army Medical Center. The 5-year disease-specific survival rate was 86.9% and 93.7% for patients diagnosed in the respective year groups of 1988 to 1991 and 1992 to 1994, with follow-up through December 1, 2000 (P < 0.001). Prostate cancer was the cause of death for 37.5% of the patients in 1988 to 1989 versus 15.4% in 1999 to 2000. Marked stage migration has occurred; from 1988 to 1998, the percentage of patients presenting with metastatic disease decreased from 14.1% to 3.3% (P < 0.001). CONCLUSIONS: A statistically significant improved 5-year disease-specific survival and a decreased chance of dying from prostate cancer has occurred after the widespread implementation of PSA. We suspect that PSA testing has resulted in fewer patients presenting with metastatic disease and more patients presenting with localized disease amenable to curative treatment. This portends well for the use of PSA screening to improve outcomes for prostate cancer. However, randomized trials are needed to confirm the improvements in survival and mortality.