Timo Soeterik

68 CHAPTER 4 TABLE 3. Diagnostic accuracy measures of DRE and mpMRI for the detection of non-organ-confined disease DRE Sensitivity (%) Specificity (%) PPV (%) NPV (%) Accuracy (95% CI) Low 0/15 (0) 54/54 (100) N/A 54/69 (78) 0.50 (0.33 – 0.67) Intermediate 8/116 (7) 159/164 (97) 8/13 (62) 159/267 (60) 0.52 (0.34 – 0.59) High 21/103 (20) 52/54 (96) 21/23 (91) 52/134 (39) 0.58 (0.49 – 0.67) Overall 29/234 (12) 265/272 (97) 29/36 (81) 205/470 (44) 0.55 (0.50 – 0.60) MRI Sensitivity (%) Specificity (%) PPV (%) NPV (%) Accuracy (95% CI) Low 6/15 (40) 44/54 (82) 6/16 (38) 44/53 (83) 0.61 (0.44 – 0.78) Intermediate 59/116 (51) 136/164 (83) 59/87 (68) 136/193 (71) 0.67 (0.50 – 0.74) High 54/103 (52) 42/54 (78) 54/66 (82) 42/91 (46) 0.65 (0.56 – 0.74) Overall 119/234 (51) 222/272 (82) 119/169 (70) 222/337 (66) 0.66 (0.61 – 0.71) DRE = digital rectal examination, MRI = magnetic resonance imaging, Low: ISUP 1 and PSA ≤10 ng/ mL, intermediate: ISUP 2 or 3 and/or PSA between 10– 20 ng/mL, high: ISUP >3 and/or PSA > 20 ng/mL. Impact of mpMRI on Treatment Intensification The cohort included 193 patients low- and intermediate risk patients treated with EBRT. Among these, 101 of 193 (53%) migrated to a higher risk group after mpMRI. Upstaging was also associated with a higher percentage of patients with treatment intensification (addition of androgen deprivation therapy): 52 of 101 (52%) of patients with upstaging versus 23 of 92 (25%) patients without upstaging, p <0.001. Upstaging was also associated with a higher OR for treatment intensification (OR: 3.5 95% CI 1.9-6.5). Association of DRE- and mpMRI T-stage with biochemical recurrence after RARP A total of 300 patients underwent RARP between 2017 and mid-2018. Of these, PSA follow- up data was available in 293 (98%) patients. BCR occurred in 46 (16%) of these patients. Median overall follow-up was 567 days (IQR 432-706 days). BCR-free survival patterns for both DRE- and mpMRI-based risk classification are presented in Figure 1A and B. As shown, migration of patients to different risk groups led to a change in BCR-free survival patterns. Although log-rank tests showed statistically significant difference for both risk group stratification systems ( p < 0.001 for both DRE and mpMRI), relative differences regarding progression-free survival between risk groups decreased if mpMRI T-stage was used. (Figure 1B). Stratifying for organ-confined (stage ≤T2) and stage ≥T3a disease on mpMRI within the EAU risk groups did not show any significant differences regarding BCR-free survival (Figure S2).