FIGURE 2 from Weakly Supervised Deep Learning Predicts Immunotherapy Response in Solid Tumors Based on PD-L1 Expression

• DOI: 10.1158/2767-9764.24982255.v1
• Publication Date: 2024-01-11T14:20:38Z
• Authors: Marta Ligero; Garazi Serna; Omar S.M. El Nahhas; Irene Sansano; Siarhei Mauchanski; Cristina Viaplana; Julien Calderaro; Rodrigo A. Toledo; Rodrigo Dienstmann; Rami S. Vanguri; Jennifer L. Sauter; Francisco Sanchez...; Sohrab P. Shah; Santiago Ramón y ...; Elena Garralda; Paolo Nuciforo; Raquel Perez-Lopez; Jakob Nikolas Kather

<p>Performance overview of the model for predicting PD-L1 status and response to immunotherapy in NSCLC-MSK and pan-cancer-VHIO cohort. AUC curves for the model to predict PD-L1 status (TPS ≥ 1%) in the training (NSCLC-MSK cohort; <b>A</b>) and in the test cohort (pan-cancer-VHIO cohort; <b>B</b>) for the 5-folds cross-validation. All trained models were deployed in the test cohort. Kaplan–Meier curves for the predicted PD-L1 status (high/low) from the model differentiates patients with longer PFS to immunotherapy from patients with shorter survival in both NSCLC-MSK (<b>C</b>) and pan-cancer-VHIO cohort (<b>D</b>).</p>