Machine Learning to Improve Treatment Selection for NSCLC Patients Treated with Immunotherapy using Real World and Translational Data

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.

Case Report: Laryngospasm as Initial Manifestation of Amyotrophic Lateral Sclerosis in a Long-Survival Patient With Heterozygous p.D90A – SOD1 Mutation

Amyotrophic Lateral Sclerosis (ALS) is a fatal neurodegenerative disease affecting motor neurons. Although its etiology is still unknown, many genes have been found to be implicated in ALS pathogenesis. The Cu/Zn superoxide dismutase (SOD1) gene was the first to be identified. Currently, more than 230 mutations in the SOD1 gene have been reported. p.D90A (p. Asp90Ala) is the most common SOD1 mutation worldwide. It shows both autosomal and recessive inheritance in different populations. To date, five Italian patients with the heterozygous p.D90A mutation have been reported. None of them complained of laryngological symptoms as the initial manifestation of ALS, although they had atypical clinical features. We describe a long-survival patient carrying heterozygous p.D90A mutation who presented with severe laryngospasm due to bilateral vocal cord paralysis. We suggest that genetic analysis may help to diagnose ALS with insidious onset like hoarseness, laryngospasm, and other type of voice disturbances.

Machine Learning to Improve Treatment Selection for NSCLC Patients Treated with Immunotherapy Using Real World and Translational Data

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.

When SUV Matters: FDG PET/CT at Baseline Correlates with Survival in Soft Tissue and Ewing Sarcoma

Introduction: The role of positron-emission tomography/computed-tomography (PET/CT) in the management of sarcomas and as a prognostic tool has been studied. However, it remains unclear which metric is the most useful. We aimed to investigate if volume-based PET metrics (Tumor volume (TV) and total lesions glycolysis (TLG)) are superior to maximal standardized uptake value (SUVmax) and other metrics in predicting survival of patients with soft tissue and bone sarcomas. Materials and Methods: In this retrospective cohort study, we screened over 52′000 PET/CT scans to identify patients diagnosed with either soft tissue, bone or Ewing sarcoma and had a staging scan at our institution before initial therapy. We used a Wilcoxon signed-rank to assess which PET/CT metric was associated with survival in different patient subgroups. Receiver-Operating-Characteristic curve analysis was used to calculate cutoff values. Results: We identified a total of 88 patients with soft tissue (51), bone (26) or Ewing (11) sarcoma. Median age at presentation was 40 years (Range: 9–86 years). High SUVmax was most significantly associated with short survival (defined as <24 months) in soft tissue sarcoma (with a median and range of SUVmax 12.5 (8.8–16.0) in short (n = 18) and 5.5 (3.3–7.2) in long survival (≥24 months) (n = 31), with (p = 0.001). Similar results were seen in Ewing sarcoma (with a median and range of SUVmax 12.1 (7.6–14.7) in short (n = 6) and 3.7 (3.5–5.5) in long survival (n = 5), with (p = 0.017). However, no PET-specific metric but tumor-volume was significantly associated (p = 0.035) with survival in primary bone sarcomas (with a median and range of 217 cm3 (186–349) in short survival (n = 4) and 60 cm3 (22–104) in long survival (n = 19), with (p = 0.035). TLG was significantly inversely associated with long survival only in Ewing sarcoma (p = 0.03). Discussion: Our analysis shows that the outcome of soft tissue, bone and Ewing sarcomas is associated with different PET/CT metrics. We could not confirm the previously suggested superiority of volume-based metrics in soft tissue sarcomas, for which we found SUVmax to remain the best prognostic factor. However, bone sarcomas should probably be evaluated with tumor volume rather than FDG PET activity.

Long survival in metastatic rectal cancer: Due to modulation of the microenvironment?

2 cases of patients with remarkable survival time in stage IV rectal cancer are presented. Histopathological data attributable to malignant tissues may deliver possible reasons for the favorable course. Arguments to acknowledge the value of the modulation of the intra- and peritumoral microenvironment in the treatment of metastatic dieases are presented.