The Prostate Cancer Intervention Versus Observation Trial (PIVOT)

Introduction

Despite dramatic efforts to increase early detection and intervention for prostate cancer, it is
not known whether radical prostatectomy or expectant management provides superior length and quality of life for men with clinically localized disease. The Prostate Cancer Intervention Versus Observation Trial (PIVOT) is a multicenter randomized trial designed to answer the fundamental question, Does early intervention with radical prostatectomy reduce all-cause and prostate cancer-specific mortality and morbidity, when compared with expectant management?

At present, PIVOT is still enrolling patients, with results expected in 2008. This article discusses the rationale for conducting PIVOT, describes the trial's study design, and reports early baseline results from the first group of screened and enrolled men.

Trial Rationale

Many experts have emphasized the importance of conducting randomized trials to determine whether early intervention with radical prostatectomy reduces morbidity and mortality in men with clinically localized prostate cancer, as compared with expectant management.[1-10] In 1997, over 400,000 new cases of prostate cancer and 40,000 deaths due to this disease were expected among American men.[11] The number of patients diagnosed with localized prostate cancer has increased dramatically in recent years due, in large part, to enhanced early detection techniques.

Despite widespread utilization of prostate-specific antigen (PSA) testing for early detection and a marked rise in the rate of radical prostatectomy, however, deaths due to prostate cancer have increased. This has led to scrutiny of the effectiveness of early detection and treatment.

Theoretically, prostate cancer confined within the prostate gland should be curable with the removal of the entire gland by surgery (radical prostatectomy) or curative radiotherapy. Screening tests can now detect disease that is localized to the prostate more often than would be the case among men presenting with symptoms. Therefore, it is tempting to speculate that screening for prostate cancer will result in curative treatment of cancers destined to cause future morbidity and mortality. However, this hypothesis has not been tested and may or may not be correct.

Barry and colleagues conducted a policy analysis to determine the cost and effectiveness of early detection and treatment of prostate cancer in Medicare-aged men.[12] They concluded that a strategy of early detection (with PSA testing) and treatment (with radical prostatectomy) was cost-effective, but only if unproven assumptions that were highly favorable with respect to treatment necessity, efficacy, and morbidity were used.

Lack of Randomized Data

Both the US Preventive Services Task Force and the Canadian Task Force on the Periodic Health Examination have recommended against routine screening for prostate cancer, in part, because the effectiveness of treatments in reducing morbidity or mortality, compared to expectant management, is unproven.[9] The only randomized controlled trial comparing radical prostatectomy with expectant management reported no difference in cumulative survival rates over 15 years.[10] However, this study was conducted in the 1970s and suffered from several design flaws.

The Clinical Guideline Panel for the Treatment of Early Stage Prostate Cancer of the American Urological Association (AUA) recently published the results of a comprehensive survey and analysis of outcomes data for methods of treating clinically localized prostate cancer.[8] The panel concluded that the outcomes data were inadequate for valid comparisons of treatments. In particular, panel members cited major limitations that hindered their attempts to analyze the data; these included the scarcity of randomized controlled trials, insufficient data, and data that may not be representative because many treatment results, especially negative or equivocal ones, are not published. The data were obtained from case series, which are confounded by the lack of control groups, the use of older diagnostic and therapeutic techniques, and variability in patient populations, tumor grade, and stage. Reports from nonrandomized studies cannot provide the unbiased information necessary to determine treatment effectiveness.

The panel recommended that treatment alternatives should be presented as options and should include expectant management, radical prostatectomy, and radiation therapy. The term "options" was utilized because the panel concluded that "the health and economic outcomes of the interventions are not sufficiently well known to permit meaningful decisions, preferences among the outcomes are not known, patient preferences are divided among alternative interventions and/or patients are indifferent about the alternative interventions".[8] They emphasized the need for large randomized, prospective, controlled studies utilizing modern diagnostic and therapeutic approaches to investigate surveillance compared to active treatment of localized prostate cancer.

PIVOT is the only randomized trial currently being conducted in the United States comparing radical prostatectomy with expectant management. PIVOT will compare overall and prostate cancer-specific mortality, morbidity, quality of life, and treatment costs in the two treatment groups.

Radical prostatectomy was chosen as the early intervention to be tested because it is the most common therapeutic strategy recommended for patients with clinically localized prostate carcinoma. Almost 100,000 radical prostatectomies will be performed in 1996. Rates of this procedure have increased by almost 100% from 1984 and can be expected to rise further with earlier and more frequent prostate cancer detection.[13] In addition, radical prostatectomy appears to be the early intervention strategy most likely to provide complete tumor eradication. Radiation therapy was not included as an initial treatment option because of sample size, cost, feasibility, and data suggesting that radiation is not superior to prostatectomy in providing prostate cancer-free survival.[8,14,15]

Early Intervention vs Expectant Management

Both early intervention and expectant management involve potential risks and benefits. Radical prostatectomy and radiation therapy offer the possibility of complete tumor eradication and cure. They may reduce patient anxiety, the likelihood of cancer metastasis, and the need for subsequent interventions for disease progression. However, prostate cancer may be unique in that the majority of individuals die "with" not "because of" cancer. Furthermore, in individuals with poorly differentiated prostate cancer, early intervention is least likely to completely eliminate cancer.[16,17] Therefore, although early intervention is often considered curative, it may not be necessary in many patients and may not be effective in the remainder.

However, refinements in early detection and treatment may now provide an opportunity for necessary and effective early intervention. Prostate cancer detection has improved due to the widespread use of PSA testing. Cancers detected through PSA-based screening programs are more likely to be confined to the prostate gland than are those detected by digital rectal examination (DRE).[18] In addition, these tumors exhibit pathologic and clinical characteristics similar to tumors detected by DRE. Nonetheless, because of the lead and length biases associated with cancers detected by PSA testing, it is not known whether early intervention will reduce mortality and morbidity.[19-21] (see note) Finally, despite technical advances, early intervention with either radical prostatectomy or radiation therapy can result in iatrogenic morbidity and mortality, adversely affecting quality as well as quantity of life.[2,22-24]

In contrast, with the expectant management strategy, palliative therapy is given only if men develop symptomatic or metastic disease. Thus, in many men, treatment will not be necessary if expectant management is utilized. If men develop systems due to local disease progression, relief can be provided by minor interventions associated with minimal morbidity. Expectant management may also include hormonal therapy at the time of diagnosis or upon evidence of disease progression (eg) from PSA testing or physical examination. While not of proven benefit, such an approach is widely utilized. The rate of metastasis may also be similar regardless of whether expectant management may result in improved quality of life, as compared with early intervention.[20,24,25]

However, expectant management does not attempt to eradicate the tumor and may miss an opportunity to prevent metastatic disease and prostate cancer-related death. Morbidity associated with progression of prostate cancer is substantial and includes urinary tract obstruction, bone pain, and other sequelae of metastatic disease. Although palliative therapies generally relieve symptoms, they may only delay disease progression and are not curative. Palliative therapy can also result in iatrogenic morbidity and may not have been required if either radical prostatectomy or radiation therapy had been utilized.

Thus , currently available evidence is consistant with either early intervention or expectant management being the preferred treatment option. However, there may be a clinically important difference between the two options with respect to both quality and length of life. Determining whether or not such a difference exists would have great implications for a large segment of society. Because randomized trials are the most scientifically rigorous, clinically reliable method for evaluating the relative efficacy of two treatment approaches, such trials are ethical and necessary.