Abstract 1584P
Background
Immune checkpoint inhibitors (ICIs) are widely used in aNSCLC. Although generally well tolerated, their specific toxicity profile requires prompt detection and multidisciplinary management. Markers to predict risk for irAE and to personalize follow-up and decision on rechallenge are missing. Aim of this work is to create a nomogram to predict the risk for irAEs.
Methods
We retrospectively collected aNSCLC patients (pts) consecutively treated with ICI at Veneto Institute of Oncology from 2014 to 2021. We recorded clinico-pathological and hematological data and estimated risk for irAE as odds ratio by logistic regression. Variables statistically associated with risk for irAE (p<0.05) were given a score and scores were combined to build a risk probability model. Likelihood ratio test was used to perform formal assessment of the fit of the model.
Results
The study population consisted of 448 pts: mainly males (n=280, 62.5%) and former/current smokers (n=365, 81.5%). Median age at diagnosis was 69 years (IQR 62.72-75.03). The prevalent histology was adenocarcinoma (n=293, 65.4%). PD-L1 expression was 1-49% in 63 (14.1%), >=50% in 186 pts (41.5%). ICI was administered as first-line in 159 (35.5%), as subsequent line in 289 pts (64.5%). Baseline neutrophil to lymphocyte ratio (NLR) was >=3 (high) in 257 pts (57.4%). At the time of analysis (April 2022), median follow-up was 8.9 months (IQR 3.2-18.9). One hundred ninety-six pts (43.8%) developed irAEs, mainly grade 1–2 (79.1%). Four categorical variables were selected to set up the nomogram: pts with PD-L1 >=50%, low NLR, <=2 metastatic sites and treated in first-line had higher risk of irAEs (p<0.001) and strength of association is shown in the Table. Table: 1584P
| Variables | Scores |
| PD-L1 expression <1% 1-49% >=50% | 0 8.2 10 |
| Baseline NLR <3 >=3 | 9 0 |
| Number of metastatic sites <=2 >2 | 9.3 0 |
| ICI line First-line Second or subsequent line | 3.6 0 |
Conclusions
Association of clinical variables, PD-L1 and NLR could help in predicting the risk for irAE, thus personalizing clinical management and finally improving outcome and quality of life. A validation cohort study will be presented at the Conference.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
Veneto Institute of Oncology IOV – IRCCS.
Funding
Has not received any funding.
Disclosure
G. Pasello: Financial Interests, Personal and Institutional, Advisory Board: AstraZeneca, BMS, Boehringer Ing., Eli Lilly, MSD, Novartis, Pfizer, Roche, Takeda; Financial Interests, Institutional, Principal Investigator: AstraZeneca, Amgen, Roche, Eli Lilly, Novartis; Financial Interests, Institutional, Funding: AstraZeneca, Roche. V. Guarneri: Financial Interests, Personal, Advisory Board: Roche, EliLilly, Novartis, MSD, Gilead; Financial Interests, Personal, Invited Speaker: EliLilly, Novartis; Financial Interests, Institutional, Invited Speaker: EliLilly, Roche, BMS, Novartis, Astra Zeneca, MSD, Synton Biopharmaceuticals, Merck. L. Bonanno: Financial Interests, Personal, Advisory Board: AstraZeneca, MSD, BMS, Roche; Financial Interests, Personal, Invited Speaker: AstraZeneca, MSD, BMS, Roche; Financial Interests, Personal and Institutional, Invited Speaker: AstraZeneca. All other authors have declared no conflicts of interest.