Table 2.
Overview of selected programmed cell death ligand 1 (PD-L1) image analysis (IA) algorithms.
| Author | ML method | Tumor type | Scoring type | Sample dataset | Relevant data | Reference |
|---|---|---|---|---|---|---|
| Koelzer et al. | Random forest/supervised learning | Melanoma | %TC | 69 samples of melanoma | Pearson correlation coefficient (r = 0.97, P <0.0001) between pathologist and IA | 42 |
| Kim et al. | Supervised learning | Gastric cancer | CPS | 39 patients with clinical response to pembrolizumab | Correlation of PD-L1 positivity with patient (RFS) outcome [HR 0.536 (95% CI 0.316–0.94), P = 0.0294] | 43 |
| Humphries et al. | Supervised learning | TNBC | % positive PD-L1 | 90 samples with clinical outcome | Correlation of PD-L1 positivity with patient (RFS) outcome [HR 0.536 (95% CI 0.316–0.94), P = 0.0294] | 44 |
| Kapil et al | GAN/semi-supervised learning | NSCLC (biopsies) | TPSa | 270 needle core biopsies; 60 slides used for concordance of manual to IA scores | IA scoring concordance with visual scores (OPA = 0.88, NPA = 0.88, PPA = 0.85; Lin's CCC = 0.94; Pearson CCC = 0.95) | 45 |
| Taylor et al. | Supervised learning with feedback loop | NSCLC | %TC, %IC | 230 cases | Concordance (Lin's CCC) of IA with three pathologists (%TC = 0.81, 0.78, 0.68; %IC = 0.62, 0.53, 0.88) | 46 |
%IC, percentage of PD-L1-positive immune cells; %TC, percentage of PD-L1-positive tumour cells; CCC, concordance correlation coefficient; CI, confidence interval; CPS, combined positive score; GAN, generative adversarial network; HR, hazard ratio; ML, machine learning; NPA, negative percent agreement; NSCLC, non-small cell lung cancer; OPA, overall percent agreement; PPA, positive percent agreement; RFS, relapse-free survival; TNBC, triple-negative breast cancer; TPS, tumor proportion score.
a
TPS calculated from positive and negative pixels.