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

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

Institutions

Publication Details

Journal
Scientific Reports
Published
2024-12-30
DOI
https://doi.org/10.1038/s41598-024-81188-2
Citations
1
Primary Topic
Radiomics and Machine Learning in Medical Imaging
Type
article
Field-Weighted Citation Impact
0.40

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

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

Pim A. de Jong, Martijn F. Boomsma, Isabella A.J. van Duin, Mitko Veta et al.
1 citations
Scientific Reports
Radiomics and Machine Learning in Medical Imaging
0.40
article

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

Pim A. de Jong, Martijn F. Boomsma, Isabella A.J. van Duin, Mitko Veta, Geke A.P. Hospers, Willeke A.M. Blokx, Wouter A. C. van Amsterdam, Djura Piersma, Ellen Kapiteijn, Andrzej Wit, P. J. van Diest, Tim Leiner, Laurens S. ter Maat, Josien P. W. Pluim, Jan Willem de Groot, Hans M. Westgeest, Rob A. J. de Mooij, Eran R. A. N. Arntz, Art Vreugdenhil, Sjoerd G. Elias, Karijn P.M. Suijkerbuijk, Joost J.C. Verhoeff, Max F. Troenokarso, Fons J. M. Van den Eertwegh, Franchette W.P.J. van den Berkmortel, Marye J. Boers‐Sonderen
article en
1 citations

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.

Scientific ReportsVol. 14(1)
University Medical Center Groningen (NL), Radboud University Nijmegen (NL), University of Groningen (NL), Utrecht University (NL), Leiden University Medical Center (NL), Medisch Spectrum Twente (NL), Radboud University Medical Center (NL), University Medical Center Utrecht (NL), Amphia Ziekenhuis (NL), Máxima Medisch Centrum (NL), Zuyderland Medisch Centrum (NL), Isala (NL), Mayo Clinic in Florida (US), Amsterdam University Medical Centers (NL), Eindhoven University of Technology (NL), Vrije Universiteit Amsterdam (NL)
Philips, ZonMw
Reduced inequalities
Openalex Percentile: Top 31%
Radiomics and Machine Learning in Medical Imaging
0.40
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.