Introduction: In 2018, durvalumab was approved by the U.S. Food and Drug Administration as consolidation immunotherapy for patients with stage III NSCLC after definitive chemoradiotherapy (CRT). However, whether durvalumab benefits patients with EGFR-mutated NSCLC remains unknown.

IMPORTANCEEffective treatment options for locally advanced esophageal cancer are limited, and rates of local recurrence after standard chemoradiotherapy remain high.OBJECTIVE To evaluate toxic effects, local control, and overall survival rates after chemoradiotherapy with a simultaneous integrated boost of radiotherapy dose to the gross tumor and nodal disease for patients with unresectable locally advanced esophageal cancer. DESIGN, SETTING, AND PARTICIPANTSA phase 1/2, single-arm trial was conducted in 46 patients from April 28, 2010, to April 9, 2015), at a tertiary academic cancer center. Outcomes of the study patients were compared with those of 97 similar patients treated at the same institution from January 10, 2010, to December 5, 2014, as part of the interim analysis. Statistical analysis was performed from December 15, 2018, to February 12, 2019.INTERVENTIONS Chemoradiotherapy with a simultaneous integrated boost of radiotherapy dose (50.4 Gy to subclinical areas at risk and 63.0 Gy to the gross tumor and involved nodes, all given in 28 fractions) with concurrent docetaxel and capecitabine or fluorouracil.MAIN OUTCOMES AND MEASURES Toxic effects, local (in-field) control, and overall survival rates.RESULTS All 46 patients (11 women and 35 men; median age, 65.5 years [range, 37.3-84.4 years]) received per-protocol therapy, as intensity-modulated photon therapy (39 [85%]) or intensity-modulated proton therapy (7 [15%]); 11 patients (24%) ultimately underwent resection. No patients experienced grade 4 or 5 toxic effects; the 10 acute grade 3 toxic events were esophagitis (4), dysphagia (3), and anorexia (3) and the 3 late grade 3 toxic events were all esophageal strictures. The actuarial local recurrence rates were 22% (95% CI, 11%-35%) at 6 months, 30% (95% CI, 18%-44%) at 1 year, and 33% (95% CI, 20%-46%) at 2 years. Overall, 15 patients (33%) experienced local failure, at a median interval of 5 months (range, 1-24 months). The median overall survival time was 21.5 months (range, 2.3-86.4 months). Exploratory comparison with a 97-patient contemporaneous institutional cohort receiving standard-dose (non-simultaneous integrated boost) chemoradiotherapy showed superior local control (hazard ratio, 0.49; 95% CI, 0.26-0.92; P = .03) and overall survival (hazard ratio, 0.66; 95% CI, 0.47-0.94; P = .02) in the group that received chemoradiotherapy with a simultaneous integrated boost.CONCLUSIONS AND RELEVANCE These findings suggest that chemoradiotherapy with a simultaneous integrated boost of radiotherapy dose for locally advanced esophageal cancer is well tolerated, with encouraging local control, and thus warrants further study.

No abstract

PURPOSE Shapley additive explanation (SHAP) values represent a unified approach to interpreting predictions made by complex machine learning (ML) models, with superior consistency and accuracy compared with prior methods. We describe a novel application of SHAP values to the prediction of mortality risk in prostate cancer. METHODS Patients with nonmetastatic, node-negative prostate cancer, diagnosed between 2004 and 2015, were identified using the National Cancer Database. Model features were specified a priori: age, prostate-specific antigen (PSA), Gleason score, percent positive cores (PPC), comorbidity score, and clinical T stage. We trained a gradient-boosted tree model and applied SHAP values to model predictions. Open-source libraries in Python 3.7 were used for all analyses. RESULTS We identified 372,808 patients meeting the inclusion criteria. When analyzing the interaction between PSA and Gleason score, we demonstrated consistency with the literature using the example of low-PSA, high-Gleason prostate cancer, recently identified as a unique entity with a poor prognosis. When analyzing the PPC-Gleason score interaction, we identified a novel finding of stronger interaction effects in patients with Gleason ≥ 8 disease compared with Gleason 6-7 disease, particularly with PPC ≥ 50%. Subsequent confirmatory linear analyses supported this finding: 5-year overall survival in Gleason ≥ 8 patients was 87.7% with PPC < 50% versus 77.2% with PPC ≥ 50% ( P < .001), compared with 89.1% versus 86.0% in Gleason 7 patients ( P < .001), with a significant interaction term between PPC ≥ 50% and Gleason ≥ 8 ( P < .001). CONCLUSION We describe a novel application of SHAP values for modeling and visualizing nonlinear interaction effects in prostate cancer. This ML-based approach is a promising technique with the potential to meaningfully improve risk stratification and staging systems.

No abstract