Machine learning for prediction of immunotherapy efficacy in non-small cell lung cancer from simple clinical and biological data

Benzekry, Sébastien; Grangeon, M.; Karlsen, Mélanie; Alexa, Maria; Bicalho-Frazeto, Isabella; Chaléat, S.; Tomasini, Pascale; Barbolosi, Dominique; Greillier, L.

doi:10.1101/2021.11.30.21267064

2021

DOI: 10.1101/2021.11.30.21267064

|View full text |Cite

Preprint

Machine learning for prediction of immunotherapy efficacy in non-small cell lung cancer from simple clinical and biological data

Sébastien Benzekry

M. Grangeon

Mélanie Karlsen

et al.

Abstract: BackgroundImmune checkpoint inhibitors (ICIs) are now a therapeutic standard in advanced non-small cell lung cancer (NSCLC), but strong predictive markers for ICIs efficacy are still lacking. We evaluated machine learning models built on simple clinical and biological data to individually predict response to ICIs.MethodsPatients with metastatic NSCLC who received ICI in second line or later were included. We collected clinical and hematological data and studied the association of this data with disease control… Show more

Help me understand this report

View published versions

Search citation statements

Order By: Relevance

Paper Sections

Select...

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2022

2024

Publication Types

Select...

Article3

Relationship

Self Cite0

Independent3

Authors

Journals

Cited by 3 publications

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Immunometabolic Markers in a Small Patient Cohort Undergoing Immunotherapy

et al. 2022

View full text Add to dashboard Cite

Although the discovery of immune checkpoints was hailed as a major breakthrough in cancer therapy, generating a sufficient response to immunotherapy is still limited. Thus, the objective of this exploratory, hypothesis-generating study was to identify potentially novel peripheral biomarkers and discuss the possible predictive relevance of combining scarcely investigated metabolic and hormonal markers with immune subsets. Sixteen markers that differed significantly between responders and non-responders were identified. In a further step, the correlation with progression-free survival (PFS) and false discovery correction (Benjamini and Hochberg) revealed potential predictive roles for the immune subset absolute lymphocyte count (rs = 0.51; p = 0.0224 *), absolute basophil count (rs = 0.43; p = 0.04 *), PD-1+ monocytes (rs = −0.49; p = 0.04 *), hemoglobin (rs = 0.44; p = 0.04 *), metabolic markers LDL (rs = 0.53; p = 0.0224 *), free androgen index (rs = 0.57; p = 0.0224 *) and CRP (rs = −0.46; p = 0.0352 *). The absolute lymphocyte count, LDL and free androgen index were the most significant individual markers, and combining the immune subsets with the metabolic markers into a biomarker ratio enhanced correlation with PFS (rs = −0.74; p ≤ 0.0001 ****). In summary, in addition to well-established markers, we identified PD-1+ monocytes and the free androgen index as potentially novel peripheral markers in the context of immunotherapy. Furthermore, the combination of immune subsets with metabolic and hormonal markers may have the potential to enhance the power of future predictive scores and should, therefore, be investigated further in larger trials.

show abstract

Immunometabolic Markers in a Small Patient Cohort Undergoing Immunotherapy

et al. 2022

View full text Add to dashboard Cite

show abstract

CASCADE: Context-Aware Data-Driven AI for Streamlined Multidisciplinary Tumor Board Recommendations in Oncology

Daye,

Parker,

Tripathi

et al. 2024

Cancers

View full text Add to dashboard Cite

This study addresses the potential of machine learning in predicting treatment recommendations for patients with hepatocellular carcinoma (HCC). Using an IRB-approved retrospective study of patients discussed at a multidisciplinary tumor board, clinical and imaging variables were extracted and used in a gradient-boosting machine learning algorithm, XGBoost. The algorithm’s performance was assessed using confusion matrix metrics and the area under the Receiver Operating Characteristics (ROC) curve. The study included 140 patients (mean age 67.7 ± 8.9 years), and the algorithm was found to be predictive of all eight treatment recommendations made by the board. The model’s predictions were more accurate than those based on published therapeutic guidelines by ESMO and NCCN. The study concludes that a machine learning model incorporating clinical and imaging variables can predict treatment recommendations made by an expert multidisciplinary tumor board, potentially aiding clinical decision-making in settings lacking subspecialty expertise.

show abstract

Predictive Markers for Immune Checkpoint Inhibitors in Non-Small Cell Lung Cancer

Ushio

Murakami

Saito

2022

JCM

View full text Add to dashboard Cite

Immune checkpoint inhibitors (ICIs) have dramatically improved the outcomes of non-small cell lung cancer patients and have increased the possibility of long-term survival. However, few patients benefit from ICIs, and no predictive biomarkers other than tumor programmed cell death ligand 1 (PD-L1) expression have been established. Hence, the identification of biomarkers is an urgent issue. This review outlines the current understanding of predictive markers for the efficacy of ICIs, including PD-L1, tumor mutation burden, DNA mismatch repair deficiency, microsatellite instability, CD8+ tumor-infiltrating lymphocytes, human leukocyte antigen class I, tumor/specific genotype, and blood biomarkers such as peripheral T-cell phenotype, neutrophil-to-lymphocyte ratio, interferon-gamma, and interleukin-8. A tremendous number of biomarkers are in development, but individual biomarkers are insufficient. Tissue biomarkers have issues in reproducibility and accuracy because of intratumoral heterogeneity and biopsy invasiveness. Furthermore, blood biomarkers have difficulty in reflecting the tumor microenvironment and therefore tend to be less predictive for the efficacy of ICIs than tissue samples. In addition to individual biomarkers, the development of composite markers, including novel technologies such as machine learning and high-throughput analysis, may make it easier to comprehensively analyze multiple biomarkers.

show abstract

scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.

Contact Info

hi@scite.ai

10624 S. Eastern Ave., Ste. A-614

Henderson, NV 89052, USA

Blog Terms and Conditions API Terms Privacy Policy Contact Cookie Preferences Do Not Sell or Share My Personal Information

Made with 💙 for researchers

Part of the Research Solutions Family.

Machine learning for prediction of immunotherapy efficacy in non-small cell lung cancer from simple clinical and biological data

Cited by 3 publications

References 48 publications

Immunometabolic Markers in a Small Patient Cohort Undergoing Immunotherapy

Immunometabolic Markers in a Small Patient Cohort Undergoing Immunotherapy

CASCADE: Context-Aware Data-Driven AI for Streamlined Multidisciplinary Tumor Board Recommendations in Oncology

Predictive Markers for Immune Checkpoint Inhibitors in Non-Small Cell Lung Cancer

Contact Info

Product

Resources

About