Prostate Cancer: Our previous work suggests that PSA can stratify men at risk 20 years prior to prostate cancer diagnosis;a time when preventive strategies might decrease the risk for life threatening prostate cancer. A major concern with PSA is that diagnosed cancer will be low grade and never threaten the well-being or life of the individual man. The Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial found no impact of randomization to screening on prostate cancer death rate after 7 to 10 years of follow-up. However, a major issue was the high screening rate in the control group. The European Randomized Study of Screening for Prostate Cancer (ERSPC) reported a 20% mortality reduction with PSA screening. However, the number-needed-to-screen (NNS) was 1410 and number-needed-to-treat (NNT) was 48 to prevent 1 prostate cancer death at 9 years. The problem is that most prostate cancer deaths occur more than 10 years after diagnosis. Longer follow-up is needed to verify whether screening is reasonable in reducing mortality. To address this question, we projected the published ERSPC data using a cumulative hazard function based on a piecewise exponential model. Constants for the model were derived from the published ERSPC figure and assumed a constant hazard of 0.0002 for the screening and control groups for years 1 to 7 of the trial and different constant rates of 0.00062 and 0.00102 for the screening and control groups, respectively, for years 8 to 12. According to our model, the NNS and NNT at 9 years were 1,254 and 43, respectively. Subsequently, NNS decreased from 837 at year 10 to 503 at year 12, and NNT decreased from 29 to 18. Despite the seemingly simplistic nature of estimating NNT, the measure is very time dependent. Thus it appears that over a 12 year follow-up, the NNT based on PSA screening is similar to other screening methods currently in use to decrease cancer mortality. Our recent work has focused on identifying men who develop life threatening prostate cancer at a time when the cancer should be curable. We found that PSA velocity was associated with a relative risk of 4.0 per ng/ml/yr for having a life-threatening prostate cancer and this risk is present 10-15 years prior to prostate cancer diagnosis. We have extended this concept using a simple additive approach called risk count where at each evaluation the health care provider sums the number of PSA evaluations that are higher than a certain level. For example, a man who never meets the rule of having a PSA velocity greater than 0.2 mg/year has less than a 2-4% probability of developing life-threatening prostate cancer, while a man who has met this rule on 5 consecutive evaluations has a risk of 19% (CI=8-35%), despite the likelihood that his PSA is still normal. Such models demonstrate that a mans risk for dying of prostate cancer (approximately 3% in the male US population) can range from less than 1% to approximately 20-25%. Another concern is whether prostate cancer screening is appropriate and for which men and at what age. We find that men older than 75 years with a PSA less than 3 ng/ml have essentially zero risk of subsequently developing a life threatening prostate cancer. Thus, stopping PSA testing in the presence of low measurements is a low risk procedure in this older age group. In a number of studies, we and others have found that PSA velocity (PSAV) >0.35 ng/ml per year and >2.0ng/ml per year are associated with increased risk of prostate cancer death more than 10 years prior to diagnosis. Since PSA and PSAV are highly correlated, the question is raised whether PSAV provides additional information to the simple PSA. In examining the distributional relationship between PSA and PSAV, we found the probability of life-threatening cancer could be stratified at a given PSA by PSAV. For example, the risk of life-threatening cancer was 3% (the same as the mortality rate observed in the US male population) at a PSA <3 ng/ml, but increased to 13.6% with PSAV >0.4 ng/ml/year. An unanswered question is to what degree this knowledge could improve the management of men with an elevated PSA. In another analysis, we compared 5 PSAV calculation techniques when PSA<10 ng/ml. We observed that the predictive ability of PSAV is dependent on the method of velocity calculation. Another potential modulator of PSA that could impact diagnosis is body size. In men without prostate cancer, body mass index (as an indicator of body size) was not associated with prostate specific antigen after adjusting for age (p = 0.06). A 10-point increase in body mass index was associated with a prostate specific antigen difference of -0.03 ng/ml. Thus, adjusting prostate specific antigen for BMI does not appear warranted. Benign Prostatic Hyperplasia and Prostate Aging: A second area of interest is benign prostatic hyperplasia (BPH) which is a common problem affecting more than 90% of men by the age of 80 years. We are interested in the natural history of prostate growth, and in understanding the association between aging, prostate growth and the development of BPH and symptoms. While the trend is for prostate enlargement with increasing age, we have observed with serial pelvic magnetic resonance imaging performed at a median followup of 4.3 years prostate size actually was stable or decreased in 38% of men, while for the entire sample the median growth rate was 2.5% per year. The results suggest that changes in prostate size are highly variable among aging men, and a considerable proportion have a stable or decreasing prostate size. Recently, we have been exploring the impact of body mass on the subsequent development of BPH. We examined the relationship between body mass index in men when they were less than 60 years of age and the size of their prostate after age 60. We found that heavier men were more likely to have large prostate glands, and that this effect was most prominent in men when they were less than 35 years of age. Likewise, we have found that men who are heavier at a younger adult age are more likely to have erectile dysfunction after age 60. At present, we continue to examine factors contributing to prostate growth and cancer development. We are interested in strategies for early diagnosis of prostate disease with a focus on the identification of risk for high risk cancers.
Showing the most recent 10 out of 12 publications
