11MO - AI-based prediction of long-term survival in NSCLC patients treated with immunotherapy: Insights from the multicentric APOLLO11 study

Background

Predicting long-term immunotherapy (IO) efficacy in non-small cell lung cancer (NSCLC) is a critical challenge in personalized treatment. Accurate IO prognosis is key, particularly for younger patients, as it supports life planning. IO offers better quality-of-life survival than chemotherapy, emphasizing the need for reliable predictions. Recent advances in analyzing real-world clinical and laboratory data with AI technologies have demonstrated their role in developing predictive models for IO outcomes.

Methods

We analyzed data from 1,031 stage IIIB-IV NSCLC patients treated with IO across six Italian institutions in the APOLLO11 trial (NCT05550961). Machine Learning (ML)-based survival analysis (Cox-ML and survival ML models) was used to evaluate overall survival (OS), and classification models (logistic regression and random forest) predicted long-term survival (OS ≥ 24 months). Multiple models incorporating cost-sensitive adjustments were implemented to optimize performance in identifying long-term survival patients. Explainability techniques identified key features.

Results

The best models were COX-ML and Extra Survival Trees, both achieving a c-index of 0.7 (±0.02) and 0.68 on the cross-validation and test set, respectively. High ECOG PS, Pleural effusion, Liver and Bone Metastases, PD-L1 negative, and Squamous histology were confirmed through the SHAP analysis as strongly associated with higher risk. The classification model predicted long survival with 0.78 accuracy and 0.77 AUC. SHAP identified low ECOG PS, younger age, high lymphocyte count, and low NLR as key variables for long survival.

Conclusions

Our results, based solely on clinical and laboratory data, align with established scoring systems like LIPI and EPSILON. AI methodologies enhanced performance, demonstrating the value of real-world data and its potential for further optimization. Identifying IO long-term survival patients is critical, enabling clinicians to offer tailored guidance, particularly to younger patients with substantial life plans. Developing cost-effective, explainable AI models ensures accessibility and transparency and fosters co-decision-making between clinicians and patients.

Clinical trial identification

APOLLO11 trial (NCT05550961).

Legal entity responsible for the study

Fondazione IRCCS Istituto Nazionale Tumori Milano.

Funding

Fondazione IRCCS Istituto Nazionale Tumori Milano.

Disclosure

V. Miskovic: Financial Interests, Personal, Other, Honoraria: Novartis. L. Mazzeo: Financial Interests, Personal, Other, Honoraria: Novartis, MSD; Financial Interests, Personal, Other, Conference Grants: Sanofi, Daiichi Sankyo, Roche. A. Ferrarin: Financial Interests, Personal, Other, Honoraria: Novartis. C.M. Della Corte: Financial Interests, Personal, Other, Honoraria: Roche, MSD, AstraZeneca, Regeneron; Financial Interests, Personal, Other, Conference Grants: Amgen. T. Beninato: Financial Interests, Personal, Other, Honoraria, conference grants: MSD; Financial Interests, Personal, Other, conference grants: Sanofi; Financial Interests, Personal, Other, Conference Grants: Pfizer, Llilly. M. Occhipinti: Financial Interests, Personal, Other, Honoraria, Consulting: AstraZeneca, BMS, MSD; Financial Interests, Personal, Other, Conference Grants: Eli Lilly. M. Brambilla: Financial Interests, Personal, Other, Travel Grant: Eli Lilly. C. Proto: Financial Interests, Personal, Other, Conference Grants, Research Funding, Consulting: AstraZeneca, Roche, MSD, BMS; Financial Interests, Personal, Other, Consulting: Janseen; Financial Interests, Personal, Other, Honoraria: Pfizer, Celgene, Daiichi Sankyo. D. Signorelli: Financial Interests, Personal, Other, Honoraria: AstraZeneca, Roche, MSD, Sanofi, Novartis, Amgen, Daiichi Sankyo, Boeringher Ingelheim, BMS, Johnson & Jonhson; Financial Interests, Institutional, Other: Eli Lilly, AstraZeneca, BMS, MSD, Roche. G. Lo Russo: Financial Interests, Personal, Advisory Board: MSD, Novartis, AstraZeneca, BMS, Sanofi, Pfizer, Roche, Lilly, GSK, Daiichi, Johnson and Johnson, Regeneron, Merck, Pierre Fabre; Financial Interests, Personal, Invited Speaker: Italfarmaco, Merck, BMS, Lilly, Sanofi; Financial Interests, Institutional, Other, Contribute for meeting organization: Janssen; Financial Interests, Institutional, Other, Contribute for meeting organization: Bayer; Financial Interests, Personal, Other, Travel accommodation: Amgen, MSD; Financial Interests, Institutional, Other, Contribute to meeting organization: Beigene; Financial Interests, Institutional, Invited Speaker: MSD, BMS, Roche, GSK, Celgene, Novartis, AstraZeneca, Amgen, Lilly. A. Prelaj: Financial Interests, Personal, Other, Training of personnel: AstraZeneca, Italfarma; Financial Interests, Personal, Invited Speaker, The Hive Project: Discussant: Roche; Financial Interests, Personal, Advisory Board, Advisory board in Lung Cancer project: BMS; Financial Interests, Personal, Other, Travel Grant: Janssen; Financial Interests, Personal, Advisory Board: Janssen, AstraZeneca; Financial Interests, Personal, Invited Speaker: Medsir, Novartis, Lilly; Financial Interests, Institutional, Invited Speaker: Bayer, BMS, AstraZeneca, Lilly, MSD, Spectrum, Roche. All other authors have declared no conflicts of interest.