A prognostic and predictive computational pathology image signature for added benefit of adjuvant chemotherapy in early stage non-small-cell lung cancer
Xiangxue Wang,
Kaustav Bera,
Cristian Barrera,
Yu Zhou,
Cheng Lu,
Pranjal Vaidya,
Pingfu Fu,
Michael Yang,
Ralph Alexander Schmid,
Sabina Berezowska,
Humberto Choi,
Vamsidhar Velcheti,
Anant Madabhushi
Affiliations
Xiangxue Wang
Center for Computational Imaging and Personalized Diagnostics, Case Western Reserve University, OH, USA
Kaustav Bera
Center for Computational Imaging and Personalized Diagnostics, Case Western Reserve University, OH, USA
Cristian Barrera
Center for Computational Imaging and Personalized Diagnostics, Case Western Reserve University, OH, USA
Yu Zhou
Center for Computational Imaging and Personalized Diagnostics, Case Western Reserve University, OH, USA
Cheng Lu
Center for Computational Imaging and Personalized Diagnostics, Case Western Reserve University, OH, USA
Pranjal Vaidya
Center for Computational Imaging and Personalized Diagnostics, Case Western Reserve University, OH, USA
Pingfu Fu
Department of Population and Quantitative Health Sciences, Case Western Reserve University, OH, USA
Michael Yang
Department of Pathology-Anatomic, University Hospitals, OH, USA
Ralph Alexander Schmid
Division of General Thoracic Surgery, University Hospital Berne, Bern, Switzerland
Sabina Berezowska
Institute of Pathology, University of Bern, Bern, Switzerland; Department of Laboratory Medicine and Pathology, Institute of Pathology, Lausanne University Hospital and Lausanne University, Lausanne, Switzerland
Humberto Choi
Department of Pulmonary and Critical Care Medicine, Respiratory Institute, Cleveland Clinic Foundation, OH, USA
Vamsidhar Velcheti
Department of Hematology and Oncology, NYU Langone Health, NY, USA
Anant Madabhushi
Center for Computational Imaging and Personalized Diagnostics, Case Western Reserve University, OH, USA; Louis Stokes Cleveland Veterans Administration Medical Center, Cleveland, OH, USA; Corresponding author at: Center for Computational Imaging and Personalized Diagnostics, Case Western Reserve University, OH, USA.
Background: We developed and validated a prognostic and predictive computational pathology risk score (CoRiS) using H&E stained tissue images from patients with early-stage non-small cell lung cancer (ES-NSCLC). Methods: 1330 patients with ES-NSCLC were acquired from 3 independent sources and divided into four cohorts D1-4. D1 comprised 100 surgery treated patients and was used to identify prognostic features via an elastic-net Cox model to predict overall and disease-free survival. CoRiS was constructed using the Cox model coefficients for the top features. The prognostic performance of CoRiS was evaluated on D2 (N=331), D3 (N=657) and D4 (N=242). Patients from D2 and D3 which comprised surgery + chemotherapy were used to validate CoRiS as predictive of added benefit to adjuvant chemotherapy (ACT) by comparing survival between different CoRiS defined risk groups. Findings: CoRiS was found to be prognostic on univariable analysis, D2 (hazard ratio (HR) = 1.41, adjusted (adj.) P = .01) and D3 (HR = 1.35, adj. P < .001). Multivariable analysis showed CoRiS was independently prognostic, D2 (HR = 1.41, adj. P < .001) and D3 (HR = 1.35, adj. P < .001), after adjusting for clinico-pathologic factors. CoRiS was also able to identify high-risk patients who derived survival benefit from ACT D2 (HR = 0.42, adj. P = .006) and D3 (HR = 0.46, adj. P = .08). Interpretation: CoRiS is a tissue non-destructive, quantitative and low-cost tool that could potentially help guide management of ES-NSCLC patients.
