Deep learning on CT scans to predict checkpoint inhibitor treatment outcomes in advanced melanoma.

Journal: Scientific reports
PMID: 39738216

Abstract

Immune checkpoint inhibitor (ICI) treatment has proven successful for advanced melanoma, but is associated with potentially severe toxicity and high costs. Accurate biomarkers for response are lacking. The present work is the first to investigate the value of deep learning on CT imaging of metastatic lesions for predicting ICI treatment outcomes in advanced melanoma. Adult patients that were treated with ICI for advanced melanoma were retrospectively identified from ten participating centers. A deep learning model (DLM) was trained on volumes of lesions on baseline CT to predict clinical benefit. The DLM was compared to and combined with a model of known clinical predictors (presence of liver and brain metastasis, level of lactate dehydrogenase, performance status and number of affected organs). A total of 730 eligible patients with 2722 lesions were included. The DLM reached an area under the receiver operating characteristic (AUROC) of 0.607 [95%CI 0.565-0.648]. In comparison, a model of clinical predictors reached an AUROC of 0.635 [95%CI 0.59 -0.678]. The combination model reached an AUROC of 0.635 [95% CI 0.595-0.676]. Differences in AUROC were not statistically significant. The output of the DLM was significantly correlated with four of the five input variables of the clinical model. The DLM reached a statistically significant discriminative value, but was unable to improve over known clinical predictors. The present work shows that the assessment over known clinical predictors is an essential step for imaging-based prediction and brings important nuance to the almost exclusively positive findings in this field.

Authors

  • Laurens S Ter Maat
    Image Sciences Institute, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands.
  • Rob A J De Mooij
    Medical Image Analysis, Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands.
  • Isabella A J Van Duin
    Department of Medical Oncology, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands.
  • Joost J C Verhoeff
    Department of Radiation Oncology, University Medical Center Utrecht, Utrecht, The Netherlands.
  • Sjoerd G Elias
    Department of Epidemiology, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands.
  • Tim Leiner
    Departments of Radiology and Nuclear Medicine (C.P.S.B., A.J.N., P.v.O., R.N.P.) and Cardiology (S.M.B.), Amsterdam University Medical Centers, Location Academic Medical Center, Meibergdreef 9, 1105 AZ Amsterdam, the Netherlands; Department of Radiology, Leiden University Medical Center, Leiden, the Netherlands (J.J.M.W.); Department of Research and Development, Pie Medical Imaging BV, Maastricht, the Netherlands (J.P.A.); and Departments of Cardiology (G.P.B., S.A.J.C.) and Radiology (T.L.), University Medical Center Utrecht, Utrecht, the Netherlands.
  • Wouter A C van Amsterdam
    Department of Data Science and Biostatistics, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands.
  • Max F Troenokarso
    Utrecht University, Utrecht, The Netherlands.
  • Eran R A N Arntz
    Utrecht University, Utrecht, The Netherlands.
  • Franchette W P J van den Berkmortel
    Department of Medical Oncology, Zuyderland Medical Center, Heerlen, the Netherlands.
  • Marye J Boers-Sonderen
    Department of Medical Oncology, Radboudumc, Radboud University, Nijmegen, The Netherlands.
  • Martijn F Boomsma
    Department of Radiology and Nuclear Medicine, Isala, P.O. Box 10400, 8000 GK Zwolle, The Netherlands.
  • Fons J M Van den Eertwegh
    Department of Medical Oncology, Amsterdam UMC, Vrije Universiteit Amsterdam, Cancer Center Amsterdam, Boelelaan 1117, Amsterdam, The Netherlands.
  • Jan Willem de Groot
    Isala Oncology Center, Isala Zwolle, Zwolle, The Netherlands.
  • Geke A P Hospers
    Department of Medical Oncology, UMC Groningen, University of Groningen, Groningen, The Netherlands.
  • Djura Piersma
    Department of Medical Oncology, Medisch Spectrum Twente, Enschede, The Netherlands.
  • Art Vreugdenhil
    Department of Medical Oncology, Maxima Medical Center, Veldhoven, The Netherlands.
  • Hans M Westgeest
    Department of Internal Medicine, Amphia Hospital, Breda, The Netherlands.
  • Ellen Kapiteijn
    Department of Medical Oncology, Leiden University Medical Center, Leiden University, Leiden, The Netherlands.
  • Ardine A De Wit
    Department of Public Health, Healthcare Innovation and Evaluation and Medical Humanities, Julius Center Research Program Methodology, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands.
  • Willeke A M Blokx
    Department of Pathology, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands.
  • Paul J van Diest
    Department of Pathology, University Medical Center Utrecht, Utrecht, the Netherlands.
  • Pim A de Jong
    University Medical Center, Utrecht, The Netherlands.
  • Josien P W Pluim
    Medical Image Analysis, Department of Biomedical Engineering, Eindhoven University of Technology, The Netherlands.
  • Karijn P M Suijkerbuijk
    Department of Medical Oncology, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands. k.suijkerbuijk@umcutrecht.nl.
  • Mitko Veta
    Medical Image Analysis Group, Eindhoven University of Technology, Eindhoven, the Netherlands.

Keywords

External Resources

Popular Topics
Recent Journals