Background: Artificial intelligence (AI) has contributed substantially in recent years to the resolution of different biomedical problems, including cancer. However, AI tools with significant and widespread impact in oncology remain scarce. The goal of this study is to present an AI-based solution tool for cancer patients data analysis that assists clinicians in identifying the clinical factors associated with poor prognosis, relapse and survival, and to develop a prognostic model that stratifies patients by risk. Materials and Methods: We used clinical data from 5275 patients diagnosed with non-small cell lung cancer, breast cancer, and non-Hodgkin lymphoma at Hospital Universitario Puerta de Hierro-Majadahonda. Accessible clinical parameters measured with a wearable device and quality of life questionnaires data were also collected. Results: Using an AI-tool, data from 5275 cancer patients were analyzed, integrating clinical data, questionnaires data, and data collected from wearable devices. Descriptive analyses were performed in order to explore the patients’ characteristics, survival probabilities were calculated, and a prognostic model identified low and high-risk profile patients. Conclusion: Overall, the reconstruction of the population’s risk profile for the cancer-specific predictive model was achieved and proved useful in clinical practice using artificial intelligence. It has potential application in clinical settings to improve risk stratification, early detection, and surveillance management of cancer patients.

Purpose: Complete pathologic response (CPR) correlates with long-term survival after perioperative chemoimmunotherapy (ChIO) in resectable non–small cell lung cancer (NSCLC). We provide a multiomic characterization of B cells and tertiary lymphoid structures (TLS) to dissect the immune landscape associated with CPR. Experimental Design: We integrated B-cell receptor (BCR) repertoire profiling (n = 87 tissue, n = 25 blood), multiplex immunofluorescence (n = 67), and bulk (n = 15), spatial (n = 12), and single-cell transcriptomics (n = 15) from tumor tissue and blood (baseline, surgery, and at 6 months of adjuvant therapy) in 123 patients (NADIM/NADIM II trials, NCT03081689/NCT03838159). Results: CPR tumors exhibited a more clonal baseline BCR repertoire (AUC 0.775; P = 0.030) that was better conserved and reinvigorated during neoadjuvant ChIO. In blood, patients with CPR tumors displayed a repertoire enriched in class-switched clones (AUC 0.833; P = 0.008), characterized by higher diversity, lower clonality, and upregulation of activation-related transcriptional programs. Neoadjuvant ChIO was associated with the induction of B-cell–related genes within TLS regions and with higher TLS density at surgery compared with Ch (P = 0.034). TLS density was not associated with CPR (P = 0.129); however, mature TLS in CPR tumors were enriched in immune activation and antigen-presenting pathways, estimated T follicular helper cells, plasmacytoid dendritic cells, and plasma cells, whereas low B-cell regions from CPR tumors displayed higher inferred infiltration of CD8+ T cells, NK cells, and macrophages, with reduced neutrophils and Tregs. Conclusions: Patients with CPR tumors exhibit a preexisting and more mature B-cell response that develops further during neoadjuvant ChIO. Our findings link B-cell–related features to CPR and highlight BCR metrics as promising predictive biomarkers in NSCLC.

Table S4 and S5 from Decoding B-cell Signatures of Complete Pathologic Response to Perioperative Chemoimmunotherapy in Non–Small Cell Lung Cancer

Supplementary Figure 11. Comparison between Complete Pathological Response (CPR) (N=6) and Non-Complete Pathological Response (NCPR) (N=7) in pre- and post-treatment samples across all other cell types.

Supplementary Figure 5. Areas of interest represent different biological entities and differ by pathological response.