Project Description

I. OBJECTIVES

To provide a guide for the interpretation of PSA kinetics in men being managed with active surveillance for favorable risk prostate cancer. To identify patients at risk for disease progression based on PSA kinetics, in whom a) ongoing surveillance is warranted, or b) definitive intervention should be considered. To provide a database for the ongoing evaluation of the pattern of disease progression and the relationship between PSA kinetics, the grade of the tumor, and other clinical parameters in favorable risk localized prostate cancer.

II. BACKGROUND AND RATIONALE

1. Epidemiology

Prostate carcinoma (PC) is the most common non-dermalogic malignancy and the second leading cause of cancer deaths in men. Over the last two decades, age-adjusted incidence rates of prostate cancer have risen dramatically (Littrup et al., 1992). This is primarily due to the widespread use of PSA as a screening test for prostate cancer. PSA screening results in the diagnosis of life threatening prostate cancer at a point where it is more amenable to cure. It also results in the diagnosis of many prostate cancers that might previously have remained undiagnosed.

Patients with intermediate to high risk localized prostate cancer (defined as Gleason score 7 or greater, or PSA > 10, or T2b-T4 cancer) who have greater than a 5-10 year life expectancy warrant definitive local therapy in most cases. However, favorable risk patients (defined as a Gleason score of 6 or less, and PSA ≤ 10, and T1c or T2a, are at relatively low risk of dying of prostate cancer, even if they live 20 or more years. In these patients, the approach of active surveillance with selective delayed intervention for the subset with evidence of rapid PSA progression or grade progression on biopsy is an attractive alternative to immediate radical treatment.

2. Natural History and Active Surveillance

Because of the high prevalence of cancer in the prostate found at autopsy of men who die of other causes, the slow progression rate of the tumor and the advanced age of men diagnosed, many of whom have comorbid conditions, patients with PC are said to be more likely to die with rather than from their disease. Estimates from autopsy studies indicate that 50% of men over the age of 50 have prostate cancer, and 65% of men over 70. Approximately 16% of men over 50 years are diagnosed with prostate cancer using PSA screening. The lifetime risk of dying of prostate cancer is about 3%. Thus, approximately 20 times as many men harbour prostate cancer as are likely to die of the disease, and about 5 times as many are diagnosed as are likely to die of it (Smith, 1990).

Early PC detection and treatment programs presume that treatment with radical prostatectomy or radiotherapy prolongs survival in subjects with clinically localized PC. This is not convincingly supported by results from observational studies, case series, structured review of the medical literature, a decision analysis model, and a small clinical trial. These studies demonstrate that the therapeutic approach of active surveillance with selective delayed intervention for favorable risk disease appears to offer survival and prostate cancer mortality comparable to that of radical treatment (surgery or radiation).

During active surveillance, most patients remain stable with respect to PSA and grade of cancer seen on repeat biopsy. However, about 20% demonstrate a rapid rise of PSA over time (defined as a doubling time < 3 years), and 5% show evidence of an increase in Gleason grade. These patients should be offered appropriate definitive therapy. This would consist of radical prostatectomy, brachytherapy, external beam radiation, cryosurgery, or HIFU, depending on patient age, co-morbidity, and preference.

In these patients, the interpretation of PSA kinetics over time is critical. The biological variation in PSA means that modest fluctuations in PSA levels are common. Thus a decision to intervene radically based on PSA doubling time should be based on repeated assays performed over a 6 month to 2 year period. In most patients, 8-9 PSAs performed at 3 month intervals represent an optimal basis for a decision regarding intervention. We have performed the PSA kinetics modeling in a large cohort of patients on active surveillance, using the General Linear Mixed Model (GLMM). This model uses the PSA variation of the entire cohort to reduce the error in calculating the PSA doubling time in an individual patient. It also compensates for the effect of baseline PSA on the PSA doubling time. The model allows the clinician to compare a patient's PSA kinetics with a stable cohort and a cohort who demonstrated biochemical or pathologic progression over time, and determine which phenotype his patient most closely resembles.