1

Tuesday, May 22, 2012 8:00 AM-10:00 AM Prostate Cancer: Detection and Screening II Moderated Poster None 1451: The association between male pattern baldness and second to forth finger ratio with prostate cancer- a prospective cohort study David Margel Seetha Venkateswaran Abbas Darwish Antonio Finelli John Trachtenberg Karen Chadwick Neil Fleshner

Introduction and Objectives Androgens play a role in the development of androgenic alopecia, commonly known as male pattern baldness. The ratio of the lengths of index and ring fingers (2D:4D) is also a marker of exposure to sex hormones, with low 2D:4D being indicative of high prenatal androgen action. Retrospective case control studies have demonstrated an association between male pattern baldness and 2D:4D ratio and prostate cancer.The aim of this study was to validate these findings in a prospective cohort. Methods Upon approval from our local ethical review board we prospectively enrolled 196 consecutive patients referred to a prostate biopsy. Finger lengths were measured using a digital vernier calliper, and the 2D:4D ratio was calculated. Male pattern baldness was assessed on a scale of 0-4 using the standardized Norwood classification (0= no balding, 1= frontal balding, 2= Mild vortex, 3=moderate vortex and 4=severe vortex). We performed all measurements prior to the biopsy thus blinded to the pathology outcome._x000D_ We performed a univariable and multivariable analysis to associate 2D:4D ratio and male pattern baldness with prostate cancer diagnosis at biopsy. The multivariable model included the two main predictors (male pattern baldness and 2D:4D ratio) as well age, digital rectal examination ( normalvs abnormal) and PSA. Results The median (IQR) age and PSA of our cohort was 64 (59-70) and 5.8 (4.1-8.4), respectively. Overall 109 patients (55%) were diagnosed with prostate cancer. _x000D_ On univariable analysis male pattern baldness was associated with prostate cancer (p for trend=0.03). However 2D:4D ratio was not. On multivariable analysis male pattern baldness remained a significant predictor of prostate cancer. Furthermore, we noted a dose response effect- the more severe balding patterns were more strongly associated with prostate cancer (Frontal balding OR 2.0 (95%CI 1.1-6.6); mild vortex OR 2.1 (95%CI 1.5-5.2); moderate vortex OR 2.5 (95%CI 1.2-7.1); severe vortex OR 2.9 (95%CI 1.1-4.3). Conclusions In a prospective cohort we found that male pattern baldness was an independent predictor of prostate cancer. Further studies are needed in order to assess whether the inclusion of male pattern baldness can contribute to existing models to predict prostate cancer prior to biopsy.