Analyzing the PIVOT Study: Who are the Best Candidates for Prostatectomy vs. Observation?
James L. Mohler, MD
Associate Director and Senior Vice President for Translational Research, Chair of the Department of Urology, and Professor of Oncology, Roswell Park Comprehensive Cancer Center
An important study published in the New England Journal of Medicine (NEJM) today addresses the use of the prostate-specific antigen test, or PSA test, for early detection and the challenges regarding proper early detection and early treatment of prostate cancer in the PSA era.
The results published today concern the Prostate Cancer Intervention Versus Observation Trial, or PIVOT study — a randomized controlled trial sponsored by the Department of Veterans Affairs and the National Cancer Institute. The study set out to determine which of two strategies — radical prostatectomy (surgery to remove the prostate) or observation — is superior for managing clinically localized prostate cancer.
Prostate cancer mortality has declined 40% in American men since the advent and application of PSA, but this reduction in mortality has created new problems of over-diagnosis and overtreatment. Many investigators skilled in translational and clinical research recognize the conundrum posed to patients by a diagnosis of cancer, and the anxiety that causes, and questions that arise from a recommendation for “active surveillance” — also referred to as watchful waiting, observation or expectant management.
Patients and medical professionals need to understand that prostate cancer is a disease of aging and comes with a wide degree of variation in aggressiveness that can be thought of more clearly as consisting of four basic types:
- “Autopsy” prostate cancer, or clinically insignificant prostate cancer (the kind that would only be found after death from other causes), occurs with approximately the frequency of one’s age.
- Clinical prostate cancer occurs at approximately half the frequency of autopsy prostate cancer if the general population of men were subjected to prostate biopsy; that is, prostate biopsy will uncover half the autopsy cancers.
- Low-risk prostate cancer is detected upon prostate biopsy following an abnormal prostate exam or elevated PSA but is clinically indolent – these cancers grow at a rate slow enough that death will occur from other causes and comprise the majority of the clinical prostate cancers.
- The final type of prostate cancer, often called the lethal phenotype, causes the death of approximately 3% of all men. These cancers often present too advanced or already have metastasized or spread, preventing potentially curative therapy. However, they may be detected as intermediate-risk or high-risk localized prostate cancer using PSA and prostate exam. These prostate cancers represent a minority of the clinical prostate cancers.
Treating all clinically localized prostate cancers the same creates a conundrum: although prostate cancer can be diagnosed in 17% of men using PSA early detection strategies, treatment of all those men will result in unnecessary overtreatment of most men in order to attempt to prevent the deaths of 3% of men. Further complicating the problem is that performance of both treatment for clinically localized prostate cancer and serum PSA for early detection of prostate cancer is compromised by indiscriminate use. For instance, treatment effect will appear less significant than it truly is if treatments are applied to men who are not at a threat of death from prostate cancer — that is, older men with low-risk prostate cancer. By the same token, it would be easier to prove a benefit to an effective treatment if it was applied only to men who were destined to die of prostate cancer — men diagnosed with prostate cancer under age 60 with high-risk prostate cancer. The effect of PSA is ameliorated by the demonstration that as many as one-fourth of American men older than age 85 have had a PSA test within the last year. Since PSA detects prostate cancer an average of 10 years before a clinical diagnosis is possible, most of these older men in whom prostate cancer can be diagnosed with a high frequency have less than a 1% chance of ever dying of prostate cancer if they receive no treatment of any kind.
How does the PIVOT study inform this national problem of overtreatment of men? The study suggests that men older than 60 (the mean age of men entering this study was 67, and only 9.5% in the observation group and 11% in the radical prostatectomy group were under age 60) would be expected to live long enough to derive a survival benefit from a prostate cancer detected early through PSA. In addition, life expectancy for this group of men was compromised by a relatively high prevalence of serious diseases that produced Charlson Co-morbidity Scores of 1 or greater in almost half the men.
Fortunately, these challenges have been recognized by many groups — most prominently by the National Comprehensive Cancer Network Guidelines Panel for Prostate Cancer, which I chair. In 2009, we proposed active surveillance as the only recommendation for men with low-risk prostate cancer whose life expectancy is less than 10 years and for men with very-low-risk prostate cancer whose life expectancy is less than 20 years. This would prevent overtreatment of approximately 40% of men diagnosed with prostate cancer in America and, if applied to the population of men entering the PIVOT trial, would probably have allowed the finding of benefit to radical prostatectomy. In conclusion, I believe this study adds a further burden on urologists, other physicians who diagnose, treat and counsel patients regarding prostate cancer, prostate cancer patients and men who are considering PSA testing to become fully informed regarding the potential risks and benefits of prostate cancer early detection and, once diagnosed, prostate cancer treatment.
However, there is no doubt that men and physicians alike fear a diagnosis of cancer and that overtreatment will remain a serious problem until men enter PSA early detection and prostate cancer treatment studies at a high enough frequency to provide the data that will more intelligently inform decision-making.
Dr. Mohler is Chair of the National Comprehensive Cancer Network Guidelines Panel for Prostate Cancer. To download the NCCN guidelines for various cancer disease sites, go to http://www.nccn.org/professionals/physician_gls/f_guidelines.asp#site.
View a brief video interview with Dr. Mohler on the PIVOT study.