Prognostic Significance of Neutrophil-to-Lymphocyte Ratio in Predicting Complete Pathological Response in Rectal Cancer Patients Receiving Neoadjuvant Chemoradiotherapy - Article abstract #943750

Background: Although higher-generation TKIs are associated with improved progression-free survival in advanced NSCLC patients with EGFR mutations, the optimal selection of TKI treatment remains uncertain. To address this gap, we developed a web application powered by a reinforcement learning (RL) algorithm to assist in guiding initial TKI treatment decisions. Methods: Clinical and mutational data from advanced NSCLC patients were retrospectively collected from 14 medical centers. Only patients with complete data and sufficient follow-up were included. Multiple supervised machine learning models were tested, with the Extra Trees Classifier (ETC) identified as the most effective for predicting progression-free survival. Feature importance scores were calculated by the ETC, and features were then integrated into a Deep Q-Network (DQN) RL algorithm. The RL model was designed to select optimal TKI generation and a treatment line for each patient and was embedded into an open-source web application for experimental clinical use. Results: In total, 318 cases of EGFR-mutant advanced NSCLC were analyzed, with a median patient age of 63. A total of 52.2% of patients were female, and 83.3% had ECOG scores of 0 or 1. The top three most influential features identified were neutrophil-to-lymphocyte ratio (log-transformed), age (log-transformed), and the treatment line of TKI administration, as tested by the ETC algorithm, with an area under curve (AUC) value of 0.73, whereas the DQN RL algorithm achieved a higher AUC value of 0.80, assigning distinct Q-values across four TKI treatment categories. This supports the decision-making process in the web-based ‘EGFR Mutant NSCLC Treatment Advisory System’, where clinicians can input patient-specific data to receive tailored recommendations. Conclusions: The RL-based web application shows promise in assisting TKI treatment selection for EGFR-mutant advanced NSCLC patients, underscoring the potential for reinforcement learning to enhance decision-making in oncology care.

Background and Objectives: Glioblastoma (GBM), defined as IDH-wildtype CNS WHO grade 4, remains the most common and aggressive primary malignant brain tumor in adults. Although the extent of resection (EOR), particularly gross total resection (GTR), is considered a potentially modifiable factor, survival comparisons across surgical groups are vulnerable to selection bias and unmeasured biological confounding. We evaluated the association between GTR and survival outcomes in patients with newly diagnosed IDH-wildtype GBM in a dual-center, real-world cohort. Materials and Methods: We conducted a retrospective, dual-center cohort study of 100 adult patients with histopathologically confirmed GBM who underwent primary surgical resection between 2015 and 2021. GTR was defined as no measurable residual contrast-enhancing tumor on early postoperative MRI (≤72 h). All patients received adjuvant chemoradiotherapy according to the Stupp protocol. Survival was analyzed using Kaplan–Meier methods with log-rank tests and explored using univariable Cox regression analysis. Given the missing key prognostic covariates (notably MGMT promoter methylation) and the retrospective design, the analyses were reported as unadjusted and descriptive. Results: Of the 100 patients, 63 (63%) underwent GTR and 37 (37%) non-GTR. The GTR group had a significantly higher rate of radiologic complete response (42.9% vs. 10.8%, p = 0.001). However, no significant differences were observed in overall survival (OS; median 13 vs. 12 months, p = 0.847) or progression-free survival (PFS; 8 vs. 8 months, p = 0.963) between the groups in unadjusted analyses. Long-term Kaplan–Meier estimates (e.g., 5-year OS) should be interpreted cautiously due to the small number of patients at risk and potential selection and biological confounding. Conclusions: In this dual-center cohort, GTR was associated with improved radiologic response but not with longer OS or PFS in unadjusted analyses. These results should be considered hypothesis-generating and not interpreted as evidence against maximal safe resection. The absence of MGMT promoter methylation status, lack of volumetric EOR quantification (including non-contrast-enhancing/FLAIR disease), and lack of standardized functional outcome data substantially limited causal inference. Prospective studies integrating molecular stratification, volumetric resection metrics, and functional outcome assessments are warranted.

Background: Current cancer treatment strategies target the preservation and enhancement of patient quality of life. Therefore, we aimed to assess the causal impact of hemoglobin (Hb) levels on global quality of life in cancer patients. Methods: We conducted a retrospective analysis using data collected from new cancer patients. The dataset included responses to the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire, demographic, and disease-related variables. We applied the Linear Non-Gaussian Acyclic Model (LiNGAM) algorithm to identify potential causal relationships among those, and their impact on global quality of life. Furthermore, to evaluate the relative importance of Hb levels on global quality of life, we utilized a Random Forest Classifier (RFC). Results: The Random Forest Classifier (RFC) emerged as the most accurate model for classifying global quality of life scores (QL2) in our analysis. RFC analysis showed that Hb level ranked as the 10th most important feature among the 23 predictors in all patients for the global quality of life. However, the LiNGAM algorithm identified the Hb value as the most significant causal factor on global quality of life (total causal effect = 3.5) for anemic patients. Moreover, the Hb value was also a significant causal factor in patients in the “other cancers” and “stage 4” subcategories, with total causal effect figures of 2.9 and 2.5, respectively. Conclusions: This study suggests that Hb levels may exert a beneficial causal effect on global quality of life in cancer patients with anemia and among those with stage 4 disease and cancers other than breast, lung, or colorectal cancer.

Background: Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal malignancy. Next-generation sequencing (NGS) enables molecular characterization and may identify clinically actionable alterations; however, real-world multicenter data linking genomic subgroups to survival outcomes remain limited. We aimed to characterize the molecular landscape of NGS-tested PDAC in a Turkish multicenter cohort and evaluate the association of key molecular alterations, including KRAS status and KRAS variant subgroups, with survival outcomes. Methods: We conducted a multicenter retrospective cohort study including patients with pathologically diagnosed PDAC between 2017 and 2025 who underwent tumor-based NGS in routine clinical practice. Overall survival (OS) was calculated from the date of metastasis, defined as the date of diagnosis for de novo metastatic disease and the date of first documented distant recurrence for recurrent cases. Progression-free survival (PFS) was calculated from the initiation of first-line systemic therapy for metastatic disease to progression or death. Survival was estimated using the Kaplan–Meier method and compared using the log-rank test. Multivariable Cox proportional hazards models were constructed for OS and PFS using clinically relevant covariates selected a priori. Results: A total of 98 patients underwent molecular profiling, and survival analyses were performed in 92 patients with available OS/PFS data. KRAS mutations were detected in 83.7% (82/98) of patients, with predominant variants G12D (47.6%), G12V (30.5%), and G12R (12.2%). TP53 mutations were present in 59.2% (58/98) of tumors, and all tumors were microsatellite stable. Tumor mutational burden data were available for 72 patients; the median TMB was 3.83 mutations/Mb, and 15.3% of evaluable tumors had a TMB ≥ 10 mutations/Mb. Excluding KRAS, clinically actionable alterations were identified in 4.1% of patients, whereas an additional 32.7% harbored potentially actionable or investigational alterations. Median OS was 14.0 months (95% CI, 11.7–16.3), and median PFS was 6.0 months (95% CI, 4.3–7.7). In unadjusted analyses, OS and PFS did not differ significantly according to KRAS mutation status (OS, p = 0.967; PFS, p = 0.652), TP53 mutation status (OS, p = 0.404; PFS, p = 0.510), or KRAS variant subgroup (OS, p = 0.332; PFS, p = 0.194). In multivariable Cox analyses, KRAS mutation status was not independently associated with OS (aHR 1.13, 95% CI 0.56–2.28; p = 0.727) or PFS (aHR 1.09, 95% CI 0.59–2.01; p = 0.780), whereas ECOG performance status remained the strongest adverse clinical factor. Conclusions: In this multicenter real-world PDAC cohort, the molecular landscape was dominated by KRAS and TP53 alterations, whereas clinically actionable non-KRAS alterations were identified in only a minority of patients. After adjustment for major clinical covariates, KRAS mutation status was not independently associated with OS or PFS. Molecular profiling may still be useful for identifying uncommon potentially targetable alterations; however, larger clinically annotated multicenter studies are needed to better define its prognostic and treatment-directing value in routine practice.