OBJECTIVES: Improved sensitivity and specificity using Magnetic Resonance Imaging (MRI) to aid the detection of prostate cancer was found by a recent Cochrane Review and the screening-by-invitation STHLM3-MRI trial (NCT03377881). Our aim was to assess and compare the cost-effectiveness of MRI-based screening for prostate cancer in Sweden using test characteristics based on (a) the Cochrane Review and (b) the STHLM3-MRI trial.

METHODS: Cost-utility analyses from a lifetime healthcare perspective were conducted for (i) no screening, (ii) quadrennial screening using prostate-specific-antigen (PSA) test and (iii) quadrennial screening using PSA and MRI. Test characteristics were based on the Cochrane Review and the STHLM3-MRI trial. Costs were calculated for the 2020 calendar year and health state values were collated from the literature.

RESULTS: Compared with no screening, the screening strategies reduced lifetime prostate cancer mortality by 6%-10%. MRI-based screening using test characteristics based on the Cochrane Review and STHLM3-MRI trial resulted in ICERs of €54,831 and €47,162 per quality-adjusted life-year (QALY) gained, which are classified as high and moderate costs per QALY gained in Sweden, respectively. This can be primarily explained by the proportion of negative MRI results from the screening-by-invitation STHLM3-MRI trial being double that from the diagnostic patient cohorts from the Cochrane Review. Compared with PSA screening, MRI-based screening reduced the numbers of lifetime biopsies by approximately 37% and 50% based on evidence from the Cochrane Review and the STHLM3-MRI trial, respectively. MRI-based screening dominated the PSA screening based on the evidence from the Cochrane Review.

CONCLUSIONS: MRI-based screening for prostate cancer is considered as cost-effective in Sweden based on the evidence from the STHLM3-MRI trial. A screening-by-invitation trial is associated with a lower proportion of MRI-positive results which leads to a modest reduction of the ICER compared with estimates based on diagnostic patient cohorts.