Correlation of statin use and duration with increased survival in hormone refractory prostate cancer

• Published in: Journal of clinical oncology Vol. 24; no. 18_suppl; p. 14617
• Main Authors: Rich, N. T., Smith, B. J., Vaena, D. A.
• Format: Journal Article
• Language: English
• Published: 20.06.2006
• ISSN: 0732-183X; 1527-7755
• DOI: 10.1200/jco.2006.24.18_suppl.14617

Abstract only 14617 Background: Prior studies associated statin use with prevention and reduced progression of prostate cancer. (Platz et al, 96 th AACR abst. 4374). However, the effects of statins in hormone refractory prostate cancer (HRPC) are unknown. We evaluated the impact of statins on survival of HRPC patients by comparing actual survival data with predicted survival using the Halabi nomogram (JCO 21:(7);1232). Methods: Using the Iowa City VA (ICVA) prescription database since 2001 along with chart review, we identified 28 consecutive patients who were prescribed statins when they had HRPC. Survival information, duration of statin use while HRPC, and Halabi nomogram variables were collected (Hb, PS, Alk Phos, Gleason, PSA). An LDH value of 90 U/L (low-normal at ICVA) was used for all patients, since LDH values were not available. Log-rank and and Cox regression analyses were performed. The outcome of interest was overall survival. Patients alive at last follow-up were censored. Statin use was treated as a time-dependent covariate in the regression analysis, and subject-specific predicted survival from the Halabi nomogram as another covariate. Results: 17 patients were alive and 11 had died. Median duration of follow-up for patients alive was 23 months. The observed vs. nomogram-predicted estimated median survivals were 36 vs. 17 months, respectively. Study patients significantly outlived their nomogram-predicted survival (p = 0.0004). Four patients had negative bone scans. 12 patients outlived their nomogram, 8 died before, and 8 were indeterminate (censored prior to their Halabi-predicted survival). Risk of death decreased with greater length of statin use (p = 0.02). When covariates for Halabi nomogram and statin use were both included in a multivariate regression model, neither was statistically significant, although similar point estimates for the relative risks were obtained. Conclusions: Statin use, and longer duration of use, significantly improved survival of HRPC patients in univariate analysis. The sample size limits the multivariate analysis. Evaluation of a time-dependent statin effect in HRPC could be considered in larger studies. No significant financial relationships to disclose.