PURPOSE: To evaluate prostate cancer screening strategies based on multiple biomarker protocols.
METHOD: A partially observable Markov chain was validated and used to simulate patients’ progression through prostate cancer states from age 40 until end of life. Disease states included localized Gleason score <7, 7, and >7, extraprostatic extension or lymph node positive cancer, metastasis, and death. Each year patients received one or more biomarker tests, and the results determined whether a biopsy was performed. Patients diagnosed with Gleason score <7 were treated via active surveillance and patients with Gleason score ≥7 via prostatectomy. Monte-Carlo simulation was used to estimate quality-adjusted survival, metastasis rate and biopsy rate per 1,000 men screened. Screening strategies were based on combinations of previously recommended thresholds and screening intervals for three biomarkers: prostate specific antigen (PSA), prostate cancer antigen 3 (PCA3) and TMPRSS2:ERG (T2:ERG). We also evaluated policies that use the Mi-Prostate Score (MiPS) model, which incorporates PSA, PCA3, and T2:ERG scores in a logistic regression model to estimate the probability of prostate cancer and of high-grade prostate cancer. In total we compared 1,387 different prostate cancer screening strategies that use various combinations of PSA, PCA3, T2:ERG, and the MiPS Score.
RESULT: The best-performing PSA strategy increased the expected quality-adjusted life years (QALYs) by 0.074 compared to no screening. The best performing multi-biomarker strategy was the high-grade MiPS risk-based strategy, which further increased the expected QALYs by 0.104 compared to no screening. The number of men with metastatic cancer and the number of biopsies per 1,000 men are presented in Table 1. The strategy based on high-grade MiPS score resulted in the lowest number of patients with metastatic cancer and the lowest number of biopsies per 1,000 men screened. Table 1 Strategy Range of Ages (yr) Screening Interval (yr) Threshold Number of Biopsies per 1,000 men Number of Metastasis per 1,000 men No Screening 0 50 PSA 55-74 1 2.5 ng/mL 296 36 MiPS High-Grade 55-74 1 0.50 229 32
CONCLUSION: Identifying biomarkers and risk thresholds optimized for identification of high-grade cancers has the greatest impact on quality-adjusted survival in the screening setting. These results suggest that, of the screening strategies tested, the MiPS risk-based was the most efficient, leading to both fewer biopsies and lower rates of disease progression.
