Prostate cancer screening and management: Caution against over-interpreting the results of the latest study, ProtecT

Prostate cancer, the most frequently diagnosed solid tumor besides basal cell carcinoma in men, is expected to kill an estimated 34,700 lives in the United States in 2023 [1]. Accordingly, it is not surprising that there have been multiple, large-scale trials to define the optimal screening strategy as well as management. On March 11, 2023, another landmark paper was published by Hamdy et al. [2] from the United Kingdom on the long-term outcome of prostate cancer patients. In this article, the authors report the 15-year outcome of 1,643 men who were diagnosed with prostate cancer between 1999 and 2009 via the blood prostate-specific antigen (PSA) screening strategy. Patients were randomized into three groups: active monitoring, surgery (radical prostatectomy), and radiotherapy. Final data was available in 1,610 men. Such efficiency in capturing the data is a testament to the commitment of the study team led by Dr. Hamdy.

The investigators found that at 15 years of follow-up the incidence of death from prostate cancer was 3% overall, consisting of 45 patients whose death was attributed to the disease. There were 3 additional deaths in the active monitoring group possibly due to prostate cancer. Among the three groups, deaths due to prostate cancer was 3.1% in the active monitoring group, 2.2% in the surgery group, and 2.9% in the radiotherapy group, which did not differ significantly. Based on such low prostate cancer-specific mortality as well as the lack of difference in outcome among the three treatment groups, the investigators concluded that the treatment decision does not impact the survival of men with localized prostate cancer diagnosed using PSA blood test. More provocative, it was suggested that radical treatments likely have no impact on the course of lethal prostate cancer.

The large sample size as well as the long follow-up period of 15 years make ProtecT study an important step forward in helping patients diagnosed with localized prostate cancer make decisions about their management. However, as with any study, care must be exercised in not over-interpreting the results of the trial. Indeed, many public and medical media outlets have run headlines suggesting that the treatment choice in men with localized prostate cancer does not affect survival. Such perception of the ProtecT study’s findings is concerning in that the public may be led to believe that localized prostate cancer is not significant and that many busy primary care physicians will choose to delay care in those with elevated blood PSA levels or forego screening altogether.

I have three major reasons for disagreeing with the conclusions made by the authors of the ProtecT study. First, the entire randomized cohort was 1,643 men. However, of the 82,429 subjects screened by PSA, 2,664 were diagnosed with prostate cancer. Thus, only 61.7% of men diagnosed with prostate cancer via PSA-based screening agreed to the randomization. Since one of the randomization groups includes active monitoring, it is likely that most men who opted out of participating in ProtecT are those with unfavorable disease features. This possibility is supported by the cohort distribution of the study in which 77% had the very favorable Grade Group 1 disease. Given such disease risk distribution of the cohort, it is not surprising that the 15-year prostate cancer-specific mortality was only 3%. Today, an overwhelming majority of men with Grade Group 1 will be treated with active surveillance because of the excellent prognosis. Therefore, the conclusion of the ProtecT study should carefully be limited to prostate cancer patients with favorable features and not be extended to all patients with localized prostate cancer diagnosed via the PSA-based screening strategy.

Second, the overall low prostate cancer-specific mortality of 3.1% in the active monitoring group could be misconstrued when not appreciating high rates of conversion to active treatment. For the public, such report can be interpreted as the outcome of not having a definitive treatment for prostate cancer is same as that of men who chose either surgery or radiation. It should be emphasized that 61% of the men in the active monitoring group eventually were treated definitively with either surgery or radiation. Therefore, the low 15-year prostate cancer-specific mortality rate in the active monitoring group supports the strategy of using time to identify those who require a definitive treatment. Men at risk for prostate cancer must recognize that the results of ProtecT does not support the concept that foregoing definitive treatment for localized prostate cancer altogether is just as good as having surgery or radiation. Rather, the published data suggest that the majority of men on active monitoring eventually will have a definitive intervention.

Third, the conclusion that the trajectory of lethal prostate cancer is not impacted by radical treatment is not strongly supported by the data. In the entire cohort, 83% (1,337 of 1,610 men) had a radical treatment eventually. As aforementioned in the active monitoring group, 61% were treated definitively over the 15-year period. It is reasonable to assume that men with aggressive features in the active monitoring group were the ones who ultimately chose a radical intervention. In short, it is very likely that by the end of the study most men, if not all, with aggressive prostate cancer were treated definitively regardless of the group assignment. Taken together with the very low overall number of deaths from prostate cancer, results of the ProtecT study should not be used to draw any meaningful conclusion on the effectiveness of definitive intervention in men with a localized, yet potentially aggressive prostate cancer.

In summary, I commend the authors of the ProtecT study for executing the trial. The findings are reassuring in that men with localized prostate cancer and favorable features have an excellent prognosis. However, care should be exercised in interpreting the data and the concept of active monitoring. Specifically the low 15-year prostate cancerspecific mortality rate of 3% combined with the overall 83% treatment rate in the entire cohort support the effectiveness of definitive intervention. Unfortunately, the study design of the ProtecT trial does not address whether radiation or surgery constitutes overtreatment in these men. In addition, the overall radical treatment rate of 61% in the active monitoring group suggests that more must be done to better stratify patients. Until new definitive data are available, I still support PSA-based screening followed by active surveillance in men with low-risk disease while definitive intervention is recommended in those with intermediate and high-risk features.