(1) Background: In advanced non-small cell lung cancer (aNSCLC), programmed death ligand 1 (PD-L1) remains the only biomarker for candidate patients to immunotherapy (IO). This study aimed at using artificial intelligence (AI) and machine learning (ML) tools to improve response and efficacy predictions in aNSCLC patients treated with IO. (2) Methods: Real world data and the blood microRNA signature classifier (MSC) were used. Patients were divided into responders (R) and non-responders (NR) to determine if the overall survival of the patients was likely to be shorter or longer than 24 months from baseline IO. (3) Results: One-hundred sixty-four out of 200 patients (i.e., only those ones with PD-L1 data available) were considered in the model, 73 (44.5%) were R and 91 (55.5%) NR. Overall, the best model was the linear regression (RL) and included 5 features. The model predicting R/NR of patients achieved accuracy ACC = 0.756, F1 score F1 = 0.722, and area under the ROC curve AUC = 0.82. LR was also the best-performing model in predicting patients with long survival (24 months OS), achieving ACC = 0.839, F1 = 0.908, and AUC = 0.87. (4) Conclusions: The results suggest that the integration of multifactorial data provided by ML techniques is a useful tool to select NSCLC patients as candidates for IO.

BackgroundIn advanced Non-Small Cell Lung Cancer (NSCLC), Programmed Death Ligand 1 (PD-L1) remains the only used biomarker to candidate patients to immunotherapy (IO) with many limits. Given the complex dynamics of the immune system it is improbable that a single biomarker could be able to profile prediction with high accuracy. A promising solution cope with this complexity is provided by Artificial Intelligence (AI) and Machine Learning (ML), which are techniques able to analyse and interpret big multifactorial data. The present study aims at using AI tools to improve response and efficacy prediction in NSCLC patients treated with IO. MethodsReal world data (clinical data, PD-L1, histology, molecular, lab tests) and the blood microRNA signature classifier (MSC), which include 24 different microRNAs, were used. Patients were divided into responders (R), who obtained a complete or partial response or stable disease as best response, and non-responders (NR), who experienced progressive or hyperprogressive disease and those who died before the first radiologic evaluation. Moreover, we used the same data to determine if the overall survival of the patients was likely to be shorter or longer than 24 months from baseline IO. For A literature review and forward feature selection technique was used to extract a specific subset of the patients data. To develop the final predictive model, different ML methods have been tested, i.e., Feedforward Neural Network (FFNN), Logistic Regression (LR), K-nearest neighbors (K-NN), Support Vector Machines (SVM), and Random Forest (RF).Results 200 patients were included. 164 out of 200 (i.e., only those patients with PD-L1 data available) were considered in the model, 73 (44.5%) were R and 91 (55.5%) NR. Overall, the best model was the LR and included 5 features: 2 clinical features including the ECOG performance status and IO-line of therapy; 1 tissue feature such as PD-L1 tumour expression; and 2 blood features including the MSC test and the neutrophil-to-lymphocyte ratio (NLR). The model predicting R/NR of the patient achieves accuracy ACC= 0.756, F1 score F1=0.722, and Area Under the ROC Curve AUC=0.82. The use of the PD-L1 alone has an ACC=0.655. The accuracy of the ML models excluding some of the features from the model were as follow: without PD-L1 value (ACC=0.726), MSC (ACC=0.750), and both PD-L1 and MSC (ACC=0.707), i.e., considering only clinical features. At data cut-off (Nov 2020), median Overall Survival (mOS) for R was 38.5 months (m) (95%IC 23.9 - 53.1) vs 3.8 m (95%IC 2.8 - 4.7) for NR, with p<0.001. LR was the most performing model in predicting patients with long survival (24-months OS), achieving ACC=0.839, F1=0.908, and AUC=0.87. ConclusionsThe results suggest that the integration of multifactorial data provided by ML techniques is a useful tool to improve personalized selection of NSCLC patients candidates to IO. In particular, compare to PD-L1 alone the expected improvement was around 10%. In particular, the model shows that the higher the ECOG, NLR value, IO-line, and MSC test level the lower the response, and the higher PD-L1 the higher the response. Considering the difference in survival among R and NR groups, these results suggest that the model can also be used to indirectly predict survival. Moreover, a second model was able to predict long survival patients with good accuracy.

Introduction In advanced Non-Small Cell Lung Cancer (aNSCLC), Programmed Death Ligand 1 (PD-L1) remains the only used biomarker to candidate patients (pts) to immunotherapy (IO) even if its predictive accuracy is not satisfactory. Indeed, given the complex dynamics underlying the cross-talk between the tumor and its microenvironment, it is unlikely that a single biomarker could be able to profile prediction with high precision. Artificial Intelligence (AI) and machine learning (ML) are techniques able to analyze and interpret big data, which cope with this complexity. The present study aims at using AI tools to improve response and efficacy prediction in aNSCLC pts treated with IO. Methods A classification task to determine if a pt is likely to benefit from IO was formulated using complete clinical data, PD-L1, histology, molecular data, and the blood microRNA signature classifier (MSC), which include 24 different microRNAs. Pts were divided into responders (R), who obtained a partial response or stable disease as best response, and non-R, who experienced progressive disease. A forward feature selection technique based on the Akaike Information Criterion was used to extract a specific subset of the pts data, being the most informative ones for the task. To develop the final predictive model, different ML methods have been tested: K-nearest neighbors, Logistic Regression, Kernel Support Vector Machines, Feedforward Neural Network, and Random Forest. Results Of 164 enrolled pts, 73 (44.5%) were R and 91 (55.5%) non-R. At data cut-off (Nov 2020), median Overall Survival (mOS) was 10.1 (95%IC 7.0 - 13.2) months (m). mOS for R pts was 38.5 m (95%IC 23.9 - 53.1) vs 3.8 m (95%IC 2.8 - 4.7) of non-R, p&lt;0.001. Overall, the best model was the Logistic Regression and included 5 features (3 clinical, 1 tissue and 1 blood features): ECOG performance status, IO-line of therapy, the neutrophil-to-lymphocyte ratio (NLR), the MSC test and PD-L1 with the following corresponding parameters w= (0.692, 0.718, 1.058, 0.566, -0.471), respectively. The intercept of the model is w_0 = 0.467, and the model achieves a 75% accuracy, computed using a leave-one-out approach. PD-L1 alone has an accuracy of 65%. We also evaluated the accuracy of the models excluding PD-L1 (74% accuracy), MSC (73% accuracy), and excluding both PD-L1 and MSC considering only clinical features (71% accuracy). Conclusions The results suggest that the data integration provided by AI techniques is a powerful tool to improve personalized selection of pts candidates to IO. In particular, the model shows that higher ECOG, NLR value, IO-line, and MSC test level correlate negatively while higher PD-L1 correlates positively with the response. The model confirms PD-L1 and MSC as relevant biomarkers to improve the accuracy of pts response. Considering the difference in survival among R and non-R groups, these results suggest that the model can also be used to indirectly predict OS. Citation Format: Arsela Prelaj, Mattia Boeri, Alessandro Robuschi, Claudia Proto, Giuseppe Lo Russo, Roberto Ferrara, Giulia Galli, Alessandro De Toma, Marta Brambilla, Mario Occhipinti, Sara Manglaviti, Alice Labianca, Teresa Beninato, Marta Bini, Mavis Mensah, Monica Ganzinelli, Nicoletta Zilembo, Filippo de Braud, Gabriella Sozzi, Marcello Restelli, Alessandra Pedrocchi, Marina Chiara Garassino, Francesco Trovo. Artificial intelligence to improve selection for NSCLC patients treated with immunotherapy [abstract]. In: Proceedings of the AACR Virtual Special Conference on Artificial Intelligence, Diagnosis, and Imaging; 2021 Jan 13-14. Philadelphia (PA): AACR; Clin Cancer Res 2021;27(5_Suppl):Abstract nr PO-065.

Introduction: In advanced Non-Small Cell Lung Cancer (NSCLC), Programmed Death Ligand 1 (PD-L1) remains the only used biomarker to candidate patients to immunotherapy (IO) with many limits. Given the complex dynamics of the immune system it is improbable that a single biomarker could be able to profile prediction with high accuracy. A promising solution cope with this complexity is provided by Artificial Intelligence (AI) and Machine Learning (ML), which are techniques able to analyse and interpret big multifactorial data. The present study aims at using AI tools to improve response and efficacy prediction in NSCLC patients treated with IO.MethodsReal world data (clinical data, PD-L1, histology, molecular, lab tests) and the blood microRNA signature classifier (MSC), which include 24 different microRNAs, were used. Patients were divided into responders (R), who obtained a complete or partial response or stable disease as best response, and non-responders (NR), who experienced progressive or hyperprogressive disease and those who died before the first radiologic evaluation. Moreover, we used the same data to determine if the overall survival of the patients was likely to be shorter or longer than 24 months from baseline IO. For A literature review and forward feature selection technique was used to extract a specific subset of the patients’ data. To develop the final predictive model, different ML methods have been tested, i.e., Feedforward Neural Network (FFNN), Logistic Regression (LR), K-nearest neighbours (K-NN), Support Vector Machines (SVM), and Random Forest (RF).Results 200 patients were included. 164 out of 200 (i.e., only those patients with PD-L1 data available) were considered in the model, 73 (44.5%) were R and 91 (55.5%) NR. Overall, the best model was the LR and included 5 features: 2 clinical features including the ECOG performance status and IO-line of therapy; 1 tissue feature such as PD-L1 tumour expression; and 2 blood features including the MSC test and the neutrophil-to-lymphocyte ratio (NLR). The model predicting R/NR of the patient achieves accuracy ACC= 0.756, F1 score F1=0.722, and Area Under the ROC Curve AUC=0.82. The use of the PD-L1 alone has an ACC=0.655. The accuracy of the ML models excluding some of the features from the model were as follow: without PD-L1 value (ACC=0.726), MSC (ACC=0.750), and both PD-L1 and MSC (ACC=0.707), i.e., considering only clinical features. At data cut-off (Nov 2020), median Overall Survival (mOS) for R was 38.5 months (m) (95%IC 23.9 - 53.1) vs 3.8 m (95%IC 2.8 - 4.7) for NR, with p<0.001. LR was the most performing model in predicting patients with long survival (24-months OS), achieving ACC=0.839, F1=0.908, and AUC=0.87.ConclusionsThe results suggest that the integration of multifactorial data provided by ML techniques is a useful tool to improve personalized selection of NSCLC patients candidates to IO. In particular, compared to PD-L1 alone the expected improvement was around 10%. In particular, the model shows that the higher the ECOG, NLR value, IO-line, and MSC test level the lower the response, and the higher PD-L1 the higher the response. Considering the difference in survival among R and NR groups, these results suggest that the model can also be used to indirectly predict survival. Moreover, a second model was able to predict long survival patients with good accuracy.