Thoughts on optimizing prostate cancer screening.

PSA screening reduces prostate cancer-specific mortality. However, measures to optimize screening programs are indicated. The buzzword is “smarter screening.”

PSA screening has been one of the most discussed topics in prostate cancer screening for years, especially because the large studies “European Randomized Study of Screening for Prostate Cancer” (ERSPC) and “Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial” (PLCO) provided contradictory results at first sight, thus questioning the validity of PSA screening. However, the ERSPC study, which was also conducted in Switzerland, among other countries, confirmed a reduction in prostate cancer-specific mortality through screening and thus paved the way for individual risk-adapted PSA screening.

Overview

Based on the seminal work of Catalona, prostate specific antigen (PSA) has been used for early detection of prostate cancer (PCa) since 1991. As expected, this led to an increase in newly diagnosed PCa; however, some men were also subjected to unnecessary invasive diagnosis and therapy, although they would probably never have had to bear the negative consequences of advancing PCa throughout their lives. On the other hand, the incidence of metastatic PCa and subsequently prostate cancer-specific mortality also halved in the USA within seven years, which was demonstrably attributed to PSA-based early detection. However, the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial (PLCO) [1], published in 2009 and involving over 75 000 subjects, failed to demonstrate a relevant survival benefit for participants in the screening cohort compared with non-screened participants. As a result, in 2012, the United States Preventive Services Task Force (USPSTF) opposed widespread PSA screening in the United States primarily because of the negative consequences of screening, such as overdiagnosis and overtreatment. Despite serious shortcomings of the PLCO study [2], this recommendation was adopted by the Swiss Medical Board here. A recent study by Shoag et al. published re-analysis [3] of the PLCO study methodology, however, showed that over 90% of patients in the control arm had already had a PSA test before or during the first five years of the study period, rendering the comparison to the screening arm meaningless. In retrospect, an assessment of the effectiveness of PSA screening by the PLCO study is thus not possible due to this serious methodological error. In contrast, the largest European PSA screening trial, the “European Randomized Study of Screening for Prostate Cancer” (ERSPC) [4] with more than 180,000 patients, shows a solid methodology and comes to a different conclusion after a duration of now 13 years. Indeed, with increasing statistical significance, a carcinoma-specific mortality reduction of 21% was demonstrated [4]. In addition, the Number Needed to Invite (NNI) and the Number Needed to Detect (NND) decreased as expected to 27 and 781, respectively, as follow-up progressed. However, the reduction in cancer-specific mortality is also associated with overdiagnosis of 27-56%, depending on the screening protocol used [5]. The consequences of resulting overtreatment, such as impotence and incontinence, can be a burden for patients. Several papers based on the ERSPC data have analyzed these advantages and disadvantages of PSA-based screening and have shown an overall gain in quality-of-life-adjusted life-years (QUALYs). Because of the contrasting results of the ERSPC and PLCO studies, the impact of PSA screening on prostate cancer-specific mortality was assessed based on the ERSPC and PLCO data by Tsodikov et al. [6] analyzed again. Here, a Cox regression analysis was performed graded by age and study; in addition, the so-called “mean lead times” (MLTs) were calculated based on unprocessed data and three prostate carcinoma models for diagnosis and progression of the disease. This led to the conclusion that in ERSPC and PLCO, a reduction in prostate cancer-specific mortality would be expected between 25% and 30% and 27% and 32%, respectively, in the screening group compared with the control group. Thus, by evaluating them together, a reduction in prostate cancer-specific mortality was demonstrated in both studies.

Most recently, the results of the largest screening study to date, with over 400,000 patients, were published by Martin et al. [7] published. After a median follow-up of ten years, no significant reduction in prostate cancer-specific mortality was shown in the intervention group, in contrast to the ERSPC trial. In addition, 19% more and a higher proportion (45% vs. 35%) of well-differentiated (Gleason ≤6) and thus less aggressive tumors not initially requiring treatment were diagnosed in the screening group in overall younger patients. It should also be emphasized that screening in this population was performed at “low intensity,” i.e., with only a single PSA test. This, like a PSA test performed multiple times over time, leads to overdiagnosis, but without benefits such as reduction in prostate cancer-specific mortality. Against this background, a one-time PSA test only therefore does not seem reasonable. Also, in the Gothenburg Trial, only organized screening had a positive effect on prostate cancer-specific mortality [8]. Therefore, the question now is how to optimize screening protocols to minimize the potential harm from overdiagnosis and overtreatment and their consequences.

Optimization of risk calculation

Promising approaches already exist to improve the specificity of PSA screening alone. Parameters such as patient age, prostate volume, positive family history, and digital-rectal examination results have been identified as independent risk factors for the development of PCa and are already used in various risk calculators for refined risk stratification. However, the validity of risk calculators outside the respective study population is limited because these models are often based on small, homogeneous patient cohorts. The SWOP/ERSPC Rotterdam risk calculator should be highlighted in this context. Based on its key data with a representative population size, its superior predictive accuracy over other risk models has been confirmed in several comparative studies. Based on Swiss data from the ERSPC study, the “Aarau ProstateCheck app” was recently developed. In addition to the known variables, this also includes the free PSA in the risk calculation and was thus able to further increase the “area under the curve” (AUC), i.e. the accuracy of the prediction.

Extended follow-up intervals and restrictive indication for biopsy

Many patients in the screening studies had PSA levels below 3 μg/l. Baseline PSA is known to be a strong predictor of prostate cancer-specific mortality in men 45 years and older. Therefore, in these patients, it is possible to increase the intervals of PSA follow-up to a safe extent to reduce costs and increase patient comfort. Depending on the respective risk constellation, an individual PSA follow-up interval of up to eight years can be offered. By using the diagnostic part of the “Aarau ProstateCheck-App” for PSA values >3 μg/l and depending on the risk cut-off, a further reduction of diagnostic measures is possible. In addition, the indication for a prostate punch biopsy can be made in a more targeted manner. With a recommended risk cut-off of 11%, invasive procedures can already be reduced by 17% without missing clinically relevant tumors. At risk >11%, imaging of the prostate is recommended first, preferably with multiparametric MRI (mpMRI). If a malignancy-suspicious lesion (classified as so-called PIRADS ≥3) is detected on image morphology, MRI-TRUS fusion biopsy would be indicated for further evaluation [9]. Another measure to counteract the problem of overtreatment is controlled observation (active surveillance). It is an adequate form of therapy for patients with a life expectancy of more than ten years and with a very low risk of tumor progression, i.e. well-differentiated tumor entity (Gleason score 3+3), small tumor volume and PSA level <10 μg/l.

The future of screening for prostate cancer

In recent years, there has been an intensive search for new biomarkers. Various tools such as the 4K score or the Prostate Health Index have been introduced to the market, but have never been able to establish themselves over the conventional PSA test. A new approach being pursued in Sweden is the integration of serum biomarkers in combination with genetic data in the form of single-nucleotide polymorphisms (SNPs) into risk calculation. In patients with a PSA level >3 μg/l and a given indication for prostate biopsy, the STHLM-III model was able to reduce the number of necessary biopsies by 32% compared to a conventional PSA test with the same sensitivity, thus arithmetically potentially saving 44% of biopsies with a benign histological result. What other measures can help improve the harm-benefit ratio of a screening program? The long-term results of the PIVOT Trial [10] have shown that only patients with a life expectancy of more than twelve years benefit from active therapy in the form of prostatectomy. Unfortunately, this study had numerous methodological flaws. The SPCG-4 study, which was of impeccable quality for this purpose, was able to clearly demonstrate the benefit of active therapy, particularly in men under 65 years of age. The recently published interim results of the PROTECT trial showed no mortality difference between prostatectomy, radiation, and active observation for low-risk PCa, but the latter group had a twofold higher number of advanced and metastatic PCa. These results indicate that active observation is a valid option in low-risk PCa. However, proper patient selection is a challenge.

Nevertheless, men over 70, although likely to benefit least from active therapy, remain the most prevalent group in screenings. Therefore, to further minimize the negative consequences of screening, ultimately only men should be screened and actively treated if a survival benefit results. The application of improved diagnostic measures such as mpMRI in the context of prostate cancer screening is also the subject of current large-scale studies, including in Sweden and Finland, particularly in light of the fact that low-level PSA determination has insufficient sensitivity for clinically relevant PCa.

Take-Home Messages

• The reduction in prostate cancer-specific mortality by PSA screening has been demonstrated. Measures are needed to optimize the harm-benefit ratio of a screening program.
• “Smarter Screening” with optimized risk calculators enables risk-adjusted progression intervals and reduces unnecessary prostate biopsies without missing clinically relevant prostate cancers.
• Active Surveillance offers a valid alternative to active treatment for patients at low risk of progression.
• For early PCa diagnosis, mpMRI of the prostate and the so-called MRI/ultrasound-fused prostate stump biopsy are increasingly gaining acceptance.

Literature:

1. Andriole GL, et al: Prostate cancer screening in the randomized Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial: mortality results after 13 years of follow-up. J Natl Cancer Inst 2012; 104(2): 125-132.
2. Kwiatkowski M, et al: Comment on the US Preventive Services Task Force’s draft recommendation on screening for prostate cancer. Eur Urol 2012; 61(4): 851-854.
3. Shoag JE, Mittal S, Hu JC: Reevaluating PSA Testing Rates in the PLCO Trial. N Engl J Med 2016; 374(18): 1795-1796.
4. Schroder FH, et al: Screening and prostate cancer mortality: results of the European Randomised Study of Screening for Prostate Cancer (ERSPC) at 13 years of follow-up. Lancet 2014; 384(9959): 2027-2035.
5. Draisma G, et al: Lead time and overdiagnosis in prostate-specific antigen screening: importance of methods and context. J Natl Cancer Inst 2009; 101(6): 374-383.
6. Tsodikov A, et al: Reconciling the Effects of Screening on Prostate Cancer Mortality in the ERSPC and PLCO Trials. Ann Intern Med 2017; 167(7): 449-455.
7. Martin RM, et al: Effect of a Low-Intensity PSA-Based Screening Intervention on Prostate Cancer Mortality: The CAP Randomized Clinical Trial. JAMA 2018; 319(9): 883-895.
8. Arnsrud Godtman R, et al: Opportunistic testing versus organized prostate-specific antigen screening: outcome after 18 years in the Goteborg randomized population-based prostate cancer screening trial. Eur Urol 2015; 68(3): 354-360.
9. Ahmed HU, et al: Diagnostic accuracy of multi-parametric MRI and TRUS biopsy in prostate cancer (PROMIS): a paired validating confirmatory study. Lancet 2017; 389(10071): 815-822.
10. Wilt TJ, et al: Radical prostatectomy versus observation for localized prostate cancer. N Engl J Med 2012; 367(3): 203-213.

InFo ONCOLOGY & HEMATOLOGY 2018; 6(2): 11-13.