Table 4.
Performance measures related to clinical usefulness of baseline model, MRI-model, and net differences between both models, in biopsy-naive, prior negative biopsy, and combined setting
Study | Test threshold | Baseline-model | MRI-model | Comparison MRI- and baseline-model | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Performance measures related to clinical usefulness | Performance measures related to clinical usefulness | Performance measures related to clinical usefulness | ||||||||||||||
Avoided Biopsy | Avoided cisPCa | True positive rate (TPR) Sensitivity | False positive rate (FPR) 1-Specificity | Net benefit (NB) | Avoided Biopsy | Avoided cisPCa | True positive rate (TPR) Sensitivity | False positive rate (FPR) 1-Specificity | Net benefit (NB) | Avoided Biopsy | Avoided cisPCa | True positive rate (TPR) Sensitivity | False positive rate (FPR) 1-Specificity | Net benefit (NB) | ||
Biopsy-naive setting | ||||||||||||||||
Alberts et al. [20] | 0% | 0.000 | 0.000 | 1.000 | 1.000 | 0.000 | 0.000 | 0.000 | 1.000 | 1.000 | 0.000 | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
5% | 0.002 | 0.000 | 1.000 | 0.997 | 0.393 | 0.020 | 0.012 | 0.993 | 0.971 | 0.391 | 1.8% | 1.2% | − 0.7% | − 2.6% | − 0.2% | |
10% | 0.006 | 0.000 | 0.995 | 0.993 | 0.357 | 0.143 | 0.112 | 0.967 | 0.776 | 0.359 | 13.7% | 11.2% | − 2.8% | − 21.7% | 0.2% | |
15% | 0.054 | 0.050 | 0.972 | 0.927 | 0.317 | 0.238 | 0.161 | 0.953 | 0.622 | 0.340 | 18.4% | 11.2% | -1.9% | -30.5% | 2.3% | |
20% | 0.157 | 0.099 | 0.948 | 0.766 | 0.291 | 0.325 | 0.323 | 0.920 | 0.496 | 0.318 | 16.8% | 22.4% | − 2.8% | − 27.0% | 2.7% | |
Fang et al. [23] | 0% | 0.000 | 1.000 | 1.000 | 0.000 | 0.000 | 1.000 | 1.000 | 0.000 | 0.0% | 0.0% | 0.0% | 0.0% | |||
5% | ||||||||||||||||
10% | 0.099 | 0.995 | 0.871 | 0.170 | 0.098 | 1.000 | 0.870 | 0.171 | − 0.1% | 0.5% | − 0.1% | 0.1% | ||||
15% | ||||||||||||||||
20% | 0.198 | 0.972 | 0.747 | 0.096 | 0.198 | 0.986 | 0.743 | 0.100 | 0.0% | 1.4% | − 0.4% | 0.4% | ||||
Mehralivand et al. [22] | 0% | 0.000 | 1.000 | 1.000 | 0.000 | 0.000 | 1.000 | 1.000 | 0.000 | 0.0% | 0.0% | 0.0% | 0.0% | |||
5% | ||||||||||||||||
10% | 0 | 1.000 | 1.000 | 0.313 | 0.172 | 0.970 | 0.739 | 0.320 | 17.2% | − 3.0% | − 26.1% | 0.7% | ||||
15% | 0.019 | 1.000 | 0.969 | 0.276 | 0.250 | 0.960 | 0.620 | 0.299 | 23.1% | − 4.0% | − 34.9% | 2.3% | ||||
20% | 0.053 | 0.990 | 0.921 | 0.236 | 0.376 | 0.890 | 0.460 | 0.269 | 32.3% | − 10.0% | − 46.1% | 3.3% | ||||
Radtke et al. [21] | 0% | 0.000 | 1.000 | 1.000 | 0.000 | 0.000 | 1.000 | 1.000 | 0.000 | 0.0% | 0.0% | 0.0% | 0.0% | |||
5% | 0.025 | 0.980 | 0.970 | 0.421 | 0.011 | 1.000 | 0.980 | 0.430 | − 1.4% | 2.0% | 0.9% | 0.9% | ||||
10% | 0.053 | 0.980 | 0.919 | 0.394 | 0.101 | 0.980 | 0.830 | 0.399 | 4.8% | 0.0% | − 8.9% | 0.5% | ||||
15% | ||||||||||||||||
20% | 0.179 | 0.941 | 0.720 | 0.334 | 0.281 | 0.919 | 0.550 | 0.347 | 10.2% | − 2.2% | − 17.0% | 1.3% | ||||
Prior negative biopsy setting | ||||||||||||||||
Alberts et al. [20] | 0% | 0.000 | 0.000 | 1.000 | 1.000 | 0.000 | 0.000 | 0.000 | 1.000 | 1.000 | 0.000 | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
5% | 0.003 | 0.000 | 1.000 | 0.996 | 0.252 | 0.265 | 0.261 | 0.976 | 0.637 | 0.258 | 26.2% | 26.1% | − 2.4% | − 35.9% | 0.6% | |
10% | 0.099 | 0.141 | 0.969 | 0.873 | 0.211 | 0.361 | 0.387 | 0.948 | 0.513 | 0.233 | 26.2% | 24.6% | − 2.1% | − 36.0% | 2.2% | |
15% | 0.28 | 0.211 | 0.924 | 0.637 | 0.187 | 0.414 | 0.401 | 0.924 | 0.449 | 0.211 | 13.4% | 19.0% | 0.0% | − 18.8% | 2.4% | |
20% | 0.398 | 0.408 | 0.827 | 0.511 | 0.148 | 0.475 | 0.465 | 0.886 | 0.378 | 0.189 | 7.7% | 5.6% | 5.9% | − 13.2% | 4.1% | |
Radtke et al. [21] | 0% | 0.000 | 1.000 | 1.000 | 0.000 | 0.000 | 1.000 | 1.000 | 0.000 | 0.0% | 0.0% | 0.0% | 0.0% | |||
5% | 0.020 | 0.980 | 0.980 | 0.359 | 0.024 | 1.000 | 0.960 | 0.368 | 0.4% | 2.0% | − 2.0% | 0.9% | ||||
10% | 0.064 | 0.960 | 0.920 | 0.320 | 0.108 | 0.970 | 0.841 | 0.330 | 4.4% | 1.0% | − 8.0% | 0.9% | ||||
15% | ||||||||||||||||
20% | 0.194 | 0.920 | 0.731 | 0.256 | 0.313 | 0.879 | 0.560 | 0.266 | 11.9% | − 4.0% | − 17.1% | 1.0% | ||||
Combined biopsy-naive and prior negative biopsy setting | ||||||||||||||||
Lee et al. [28] | 0% | 0.000 | 1.000 | 1.000 | 0.000 | 0.000 | 0.000 | 1.000 | 1.000 | 0.000 | 0.0% | 0.0% | 0.0% | 0.0% | ||
5% | 0.044 | 0.038 | 0.996 | 0.931 | 0.353 | |||||||||||
10% | 0.106 | 0.113 | 0.992 | 0.833 | 0.325 | |||||||||||
15% | 0.180 | 0.213 | 0.987 | 0.714 | 0.303 | |||||||||||
20% | 0.259 | 0.288 | 0.983 | 0.590 | 0.288 | |||||||||||
Van Leeuwen et al. [29] | 0% | 0.000 | 0.000 | 1.000 | 1.000 | 0.000 | 0.000 | 0.000 | 1.000 | 1.000 | 0.000 | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
5% | 0.074 | 0.022 | 0.995 | 0.884 | 0.348 | 0.186 | 0.066 | 1.000 | 0.700 | 0.356 | 11.2% | 4.3% | 0.5% | − 18.4% | 0.8% | |
10% | 0.155 | 0.057 | 0.987 | 0.758 | 0.322 | 0.282 | 0.128 | 0.974 | 0.562 | 0.330 | 12.7% | 7.1% | − 1.3% | − 19.6% | 0.9% | |
15% | 0.254 | 0.125 | 0.960 | 0.615 | 0.297 | 0.379 | 0.199 | 0.926 | 0.435 | 0.303 | 12.5% | 7.4% | − 3.4% | − 18.0% | 0.7% | |
20% |