LCatlas: A Point-Of-Care Tool for Predicting Immunotherapy Outcomes in Non-Small Cell Lung Cancer

Non-small cell lung cancer (NSCLC) remains a leading cause of cancer mortality, with a minority of patients experiencing durable responses to first-line immune checkpoint inhibitors (ICIs). Currently, there are no easily accessible tools to predict clinical benefit to first-line immunotherapy. To address this gap, we developed LCatlas (https://www.lcatlas.org), a point-of-care web application designed to predict immunotherapy outcomes and inform evidence-based treatment decisions. Using data from 79 NSCLC patients receiving first-line immunotherapy, we extracted demographic, clinical, and molecular features from electronic medical records and compared three independent approaches, statistical model (logistic regression), machine learning (XGboost), and AI model (TabTransformer) to identify the most effective predictor. LCatlas accepts nine variables, gender, smoking history, bone/pleura/liver metastasis, PDL1 tumor staining, albumin level, hemoglobin level, lymphocyte count, neutrophil count, and platelet count, to predict responses to both immunotherapy and immunotherapy plus chemotherapy. LCatlas cross-validation yielded an area under the curve (AUC) of 79% [72% - 85%], with sensitivity of 87% [72% - 95%] and specificity of 78% [64% - 89%], outperforming previous studies including high-volume data-driven deep-learning approaches. In a validation cohort consisting of 28 individuals, it demonstrated a prediction accuracy of 82%. LCatlas may serve as a valuable point-of-care tool by identifying patients likely to experience extended immunotherapy or combination therapy using readily available medical data. Its evidence-based approach can optimize personalized treatment planning and thereby promote quality and safety of care.

Authors

Institutions

Publication Details

Journal
medRxiv
Published
2026-11-20
DOI
https://doi.org/10.64898/2026.11.18.26368446
Primary Topic
Cancer Immunotherapy and Biomarkers
Type
preprint

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

LCatlas: A Point-Of-Care Tool for Predicting Immunotherapy Outcomes in Non-Small Cell Lung Cancer

Alexis Leyba, Robert C. Hsu, Ha Youn Lee, Jorge Nieva et al.
3 links
medRxiv
Cancer Immunotherapy and Biomarkers
preprint

LCatlas: A Point-Of-Care Tool for Predicting Immunotherapy Outcomes in Non-Small Cell Lung Cancer

Alexis Leyba, Robert C. Hsu, Ha Youn Lee, Jorge Nieva, Wilson Yeung, Sung Yong Park
preprint en

Abstract

Non-small cell lung cancer (NSCLC) remains a leading cause of cancer mortality, with a minority of patients experiencing durable responses to first-line immune checkpoint inhibitors (ICIs). Currently, there are no easily accessible tools to predict clinical benefit to first-line immunotherapy. To address this gap, we developed LCatlas (https://www.lcatlas.org), a point-of-care web application designed to predict immunotherapy outcomes and inform evidence-based treatment decisions. Using data from 79 NSCLC patients receiving first-line immunotherapy, we extracted demographic, clinical, and molecular features from electronic medical records and compared three independent approaches, statistical model (logistic regression), machine learning (XGboost), and AI model (TabTransformer) to identify the most effective predictor. LCatlas accepts nine variables, gender, smoking history, bone/pleura/liver metastasis, PDL1 tumor staining, albumin level, hemoglobin level, lymphocyte count, neutrophil count, and platelet count, to predict responses to both immunotherapy and immunotherapy plus chemotherapy. LCatlas cross-validation yielded an area under the curve (AUC) of 79% [72% - 85%], with sensitivity of 87% [72% - 95%] and specificity of 78% [64% - 89%], outperforming previous studies including high-volume data-driven deep-learning approaches. In a validation cohort consisting of 28 individuals, it demonstrated a prediction accuracy of 82%. LCatlas may serve as a valuable point-of-care tool by identifying patients likely to experience extended immunotherapy or combination therapy using readily available medical data. Its evidence-based approach can optimize personalized treatment planning and thereby promote quality and safety of care.

medRxiv
University of Southern California (US), USC Norris Comprehensive Cancer Center (US)
University of Southern California
Good health and well-being
Cancer Immunotherapy and Biomarkers
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon