Purpose To evaluate the cost-effectiveness of multiparametric magnetic resonance imaging (mpMRI) to diagnose prostate cancer and direct all low-risk patients into active surveillance (AS).

Materials and Methods A Markov cohort model was developed to assess three scenarios: 1) no mpMRI and current AS; 2) mpMRI and current AS; and 3) mpMRI and increased AS. Men were tracked from diagnosis to end-of-life. Estimates to populate the model were derived from systematic reviews, meta-analyses, epidemiological publications, and national cost reports. An Australian Government perspective was used. Outcomes included healthcare costs, survival, quality-adjusted life years (QALYs), number of biopsies, and significant and insignificant cancers. Extensive sensitivity analyses were undertaken to address possible variation in the modeled inputs.

Results Mean lifetime costs per patient were AU$23,191 for Scenario 1, AU$23,387 for Scenario 2 and AU$21,064 for Scenario 3. Corresponding QALYs were 7.81, 7.77, 7.83 for Scenarios 1, 2, and 3, respectively. At the current uptake of AS in Australia, mpMRI alone does not appear cost-effective (16.9% likelihood). However, mpMRI with AS for all men with low-risk disease is strongly cost-effective (86.9% likelihood) at a willingness-to-pay AU$50,000 per QALY gained. For the mpMRI options, for every 1000 men suspected of prostate cancer, using mpMRI would avoid 340 biopsies, detect an additional 20 significant cancers, and detect 10 fewer insignificant cancers.

Conclusion Diagnosis of prostate cancer through mpMRI technology would be cost-effective if it leads to increased uptake of AS for men with confirmed very-low- or low-risk prostate cancer.v