Population-based prostate cancer screening with Magnetic Resonance Imaging or Ultrasonography: the IP1-PROSTAGRAM study

Abstract

Importance: Screening for prostate cancer using PSA-testing can lead to problems of under- and over-diagnosis. A short, non-contrast MRI or transrectal ultrasound might overcome these limitations. Objective: To compare the performance of PSA, MRI and ultrasound as screening tests for prostate cancer. Design, Setting and Participants: This prospective, population-based, blinded cohort study was conducted at seven primary care practices and two imaging centres in the UK. 2034 community based men aged 50-69 years invited for prostate cancer screening and 408 were consented. Interventions: All participants underwent screening with a PSA test, MRI (T2-weighted and diffusion) and ultrasound (b-mode and shearwave elastography).-The tests were independently interpreted without knowledge of other results. Both imaging tests were reported on a validated 5-point scale of suspicion. If any test was screen-positive, a systematic 12-core biopsy was performed. Additional image-fusion targeted biopsies were taken if the MRI or ultrasound was positive. Main Outcomes and Measures: The proportion of men with screen-positive MRI or ultrasound (defined as either score 3-5 or 4-5) or screen-positive PSA (defined as PSA≥3g/L). Key secondary outcomes were the number of clinically-significant and clinically-insignificant cancers detected if each test was used exclusively. Clinically-significant cancer was defined as any Gleason score≥3+4. Results: The proportion with a screen-positive MRI (score 3-5) was higher than the proportion with a screen-positive PSA (72/406, 17.7%[95%CI 14.3-21.8] vs. 40/406,9.9%[95%CI 7.3-13.2]; p<0.001). The proportion with a screen-positive ultrasound (score 3-5) was also higher than PSA (96/405, 23.7% [95%CI 19.8-28.1];p<0.001). For an imaging threshold of score 4-5, the proportion with a screen-positive MRI was similar to PSA (43/406, 10.6%[95%CI 7.9-13.2];p=0.71), as was the proportion with a screen-positive ultrasound (52/405, 12.8%[95%CI 9.9-16.5];p=0.15). PSA(≥3ng/ml) detected 7 clinically-significant cancers.-MRI (score 3-5) detected 14 and MRI (score 4-5) detected 11. Ultrasound (score 3-5) detected 9 and ultrasound (score 4-5) detected 4. Clinically-insignificant cancers were diagnosed by PSA in 6 cases, by MRI (score 3-5) in 7, MRI (score 4-5) in 5, ultrasound101 (score 3-5) in 13 and ultrasound (score 4-5) in 7. Conclusions and Relevance: When screening the general population for prostate cancer, MRI using a score of 4 or 5 to define a screen-positive test compared to PSA alone at ≥3ng/ml, might lead to more men diagnosed with clinically-significant cancer, without increasing the number of men recommended to have a biopsy or over-diagnosed with clinically-insignificant cancer. There was no evidence that ultrasound would have better performance compared to PSA alone.