Original Research| Volume 153, P179-189, August 2021

Clinical decision support algorithm based on machine learning to assess the clinical response to anti–programmed death-1 therapy in patients with non–small-cell lung cancer


      • Programmed death-ligand 1 expression alone may not reflect the response to programmed cell death protein 1 (PD-1) inhibitors.
      • Various clinical characteristics are related to the anti–PD-1 response.
      • We established a machine learning–based algorithm to predict the anti–PD-1 response.



      Anti–programmed death (PD)-1 therapy confers sustainable clinical benefits for patients with non–small-cell lung cancer (NSCLC), but only some patients respond to the treatment. Various clinical characteristics, including the PD-ligand 1 (PD-L1) level, are related to the anti–PD-1 response; however, none of these can independently serve as predictive biomarkers. Herein, we established a machine learning (ML)–based clinical decision support algorithm to predict the anti–PD-1 response by comprehensively combining the clinical information.

      Materials and methods

      We collected clinical data, including patient characteristics, mutations and laboratory findings, from the electronic medical records of 142 patients with NSCLC treated with anti–PD-1 therapy; these were analysed for the clinical outcome as the discovery set. Nineteen clinically meaningful features were used in supervised ML algorithms, including LightGBM, XGBoost, multilayer neural network, ridge regression and linear discriminant analysis, to predict anti–PD-1 responses. Based on each ML algorithm's prediction performance, the optimal ML was selected and validated in an independent validation set of PD-1 inhibitor–treated patients.


      Several factors, including PD-L1 expression, tumour burden and neutrophil-to-lymphocyte ratio, could independently predict the anti–PD-1 response in the discovery set. ML platforms based on the LightGBM algorithm using 19 clinical features showed more significant prediction performance (area under the curve [AUC] 0.788) than on individual clinical features and traditional multivariate logistic regression (AUC 0.759).


      Collectively, our LightGBM algorithm offers a clinical decision support model to predict the anti–PD-1 response in patients with NSCLC.

      Graphical abstract


      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'


      Subscribe to European Journal of Cancer
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect


        • Pardoll D.M.
        The blockade of immune checkpoints in cancer immunotherapy.
        Nat Rev Canc. 2012; 12: 252-264
        • Herbst R.S.
        • et al.
        Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients.
        Nature. 2014; 515: 563-567
        • Karlsson A.K.
        • Saleh S.N.
        Checkpoint inhibitors for malignant melanoma: a systematic review and meta-analysis.
        Clin Cosmet Invest Dermatol. 2017; 10: 325-339
        • Liu B.
        • et al.
        Recent development in clinical applications of PD-1 and PD-L1 antibodies for cancer immunotherapy.
        J Hematol Oncol. 2017; 10: 174
        • Garon E.B.
        • et al.
        Pembrolizumab for the treatment of non-small-cell lung cancer.
        N Engl J Med. 2015; 372: 2018-2028
        • Bagley S.J.
        • et al.
        Pretreatment neutrophil-to-lymphocyte ratio as a marker of outcomes in nivolumab-treated patients with advanced non-small-cell lung cancer.
        Lung Canc. 2017; 106: 1-7
        • Diem S.
        • et al.
        Neutrophil-to-Lymphocyte ratio (NLR) and Platelet-to-Lymphocyte ratio (PLR) as prognostic markers in patients with non-small cell lung cancer (NSCLC) treated with nivolumab.
        Lung Canc. 2017; 111: 176-181
        • Li B.
        • et al.
        Impact of smoking on efficacy of PD-1/PD-L1 inhibitors in non-small cell lung cancer patients: a meta-analysis.
        Onco Targets Ther. 2018; 11: 3691-3696
        • Pantano F.
        • et al.
        Prognostic clinical factors in patients affected by non-small-cell lung cancer receiving Nivolumab.
        Expet Opin Biol Ther. 2020; 20: 319-326
        • Conforti F.
        • et al.
        Cancer immunotherapy efficacy and patients' sex: a systematic review and meta-analysis.
        Lancet Oncol. 2018; 19: 737-746
        • Funazo T.
        • et al.
        Liver metastasis is associated with poor progression-free survival in patients with non–small cell lung cancer treated with nivolumab.
        J Thorac Oncol. 2017; 12: e140-e141
        • Shiroyama T.
        • et al.
        Clinical characteristics of liver metastasis in nivolumab-treated patients with non-small cell lung cancer.
        Anticancer Res. 2018; 38: 4723-4729
        • Garassino M.C.
        • et al.
        Italian nivolumab expanded access Program in nonsquamous non-small cell lung cancer patients: results in never-smokers and EGFR-mutant patients.
        J Thorac Oncol. 2018; 13: 1146-1155
        • Svaton M.
        • et al.
        Chronic inflammation as a potential predictive factor of nivolumab therapy in non-small cell lung cancer.
        Anticancer Res. 2018; 38: 6771-6782
        • Wiesweg M.
        • et al.
        Machine learning-based predictors for immune checkpoint inhibitor therapy of non-small-cell lung cancer.
        Ann Oncol. 2019; 30: 655-657
        • Heo J.
        • et al.
        Machine learning based model for prediction of outcomes in acute stroke.
        Stroke. 2019; 50: 1263-1265
        • Cruz J.A.
        • Wishart D.S.
        Applications of machine learning in cancer prediction and prognosis.
        Canc Inf. 2006; 2 (117693510600200030)
        • Kourou K.
        • et al.
        Machine learning applications in cancer prognosis and prediction.
        Comput Struct Biotechnol J. 2015; 13: 8-17
        • Rech A.J.
        • et al.
        Radiotherapy and CD40 activation separately augment immunity to checkpoint blockade in cancer.
        Canc Res. 2018; 78: 4282-4291
        • Chen T.
        • Guestrin C.
        XGBoost: a scalable tree boosting system.
        2016 (arXiv e-prints arXiv:1603.02754)
        • Ke G.
        • et al.
        LightGBM: a highly efficient gradient boosting decision tree.
        2017: 3146-3154
        • Krogh A.
        What are artificial neural networks?.
        Nat Biotechnol. 2008; 26: 195-197
        • Lee Y.
        Support vector machines for classification: a statistical portrait.
        Methods Mol Biol. 2010; 620: 347-368
        • Abello J.
        • Cormode G.
        Discrete methods in epidemiology.
        American Mathematical Society, 2007
        • Ryback R.S.
        • et al.
        Quadratic discriminant analysis as an aid to interpretive reporting of clinical laboratory tests.
        JAMA. 1982; 248: 2342-2345
        • Gogtay N.J.
        • et al.
        Principles of regression analysis.
        J Assoc Phys India. 2017; 65: 48-52
        • Lundberg S.
        • Lee S.-I.
        A unified approach to interpreting model predictions.
        2017 (arXiv e-prints arXiv:1705.07874)
        • Bao X.
        • et al.
        Immune landscape and a novel immunotherapy-related gene signature associated with clinical outcome in early-stage lung adenocarcinoma.
        J Mol Med (Berl). 2020; 98: 805-818
        • Duhazé J.
        • et al.
        A machine learning approach for high-dimensional time-to-event prediction with application to immunogenicity of biotherapies in the ABIRISK cohort.
        Front Immunol. 2020; 11: 608