Deep learning on CT scans to predict checkpoint inhibitor treatment outcomes in advanced melanoma
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...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-12-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-024-81188-2 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559425338310656 |
|---|---|
| author | Laurens S. Ter Maat Rob A. J. De Mooij Isabella A. J. Van Duin Joost J. C. Verhoeff Sjoerd G. Elias Tim Leiner Wouter A. C. van Amsterdam Max F. Troenokarso Eran R. A. N. Arntz Franchette W. P. J. Van den Berkmortel Marye J. Boers-Sonderen Martijn F. Boomsma Fons J. M. Van den Eertwegh Jan Willem de Groot Geke A. P. Hospers Djura Piersma Art Vreugdenhil Hans M. Westgeest Ellen Kapiteijn Ardine A. De Wit Willeke A. M. Blokx Paul J. Van Diest Pim A. De Jong Josien P. W. Pluim Karijn P. M. Suijkerbuijk Mitko Veta |
| author_facet | Laurens S. Ter Maat Rob A. J. De Mooij Isabella A. J. Van Duin Joost J. C. Verhoeff Sjoerd G. Elias Tim Leiner Wouter A. C. van Amsterdam Max F. Troenokarso Eran R. A. N. Arntz Franchette W. P. J. Van den Berkmortel Marye J. Boers-Sonderen Martijn F. Boomsma Fons J. M. Van den Eertwegh Jan Willem de Groot Geke A. P. Hospers Djura Piersma Art Vreugdenhil Hans M. Westgeest Ellen Kapiteijn Ardine A. De Wit Willeke A. M. Blokx Paul J. Van Diest Pim A. De Jong Josien P. W. Pluim Karijn P. M. Suijkerbuijk Mitko Veta |
| author_sort | Laurens S. Ter Maat |
| collection | DOAJ |
| description | 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. |
| format | Article |
| id | doaj-art-495322e408e349a3aff5ac747651cd04 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-495322e408e349a3aff5ac747651cd042025-01-05T12:29:28ZengNature PortfolioScientific Reports2045-23222024-12-0114111110.1038/s41598-024-81188-2Deep learning on CT scans to predict checkpoint inhibitor treatment outcomes in advanced melanomaLaurens S. Ter Maat0Rob A. J. De Mooij1Isabella A. J. Van Duin2Joost J. C. Verhoeff3Sjoerd G. Elias4Tim Leiner5Wouter A. C. van Amsterdam6Max F. Troenokarso7Eran R. A. N. Arntz8Franchette W. P. J. Van den Berkmortel9Marye J. Boers-Sonderen10Martijn F. Boomsma11Fons J. M. Van den Eertwegh12Jan Willem de Groot13Geke A. P. Hospers14Djura Piersma15Art Vreugdenhil16Hans M. Westgeest17Ellen Kapiteijn18Ardine A. De Wit19Willeke A. M. Blokx20Paul J. Van Diest21Pim A. De Jong22Josien P. W. Pluim23Karijn P. M. Suijkerbuijk24Mitko Veta25Image Sciences Institute, University Medical Center Utrecht, Utrecht UniversityMedical Image Analysis, Department of Biomedical Engineering, Eindhoven University of TechnologyDepartment of Medical Oncology, University Medical Center Utrecht, Utrecht UniversityDepartment of Radiotherapy, University Medical Center Utrecht, Utrecht UniversityDepartment of Epidemiology, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht UniversityDepartment of Radiology, Mayo ClinicalDepartment of Data Science and Biostatistics, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht UniversityUtrecht UniversityUtrecht UniversityDepartment of Medical Oncology, Zuyderland Medical CenterDepartment of Medical Oncology, Radboudumc, Radboud UniversityDepartment of Radiology, Isala ZwolleDepartment of Medical Oncology, Amsterdam UMC, Vrije Universiteit Amsterdam, Cancer Center AmsterdamIsala Oncology Center, Isala ZwolleDepartment of Medical Oncology, UMC Groningen, University of GroningenDepartment of Medical Oncology, Medisch Spectrum TwenteDepartment of Medical Oncology, Maxima Medical CenterDepartment of Internal Medicine, Amphia HospitalDepartment of Medical Oncology, Leiden University Medical Center, Leiden UniversityDepartment of Public Health, Healthcare Innovation and Evaluation and Medical Humanities, Julius Center Research Program Methodology, University Medical Center Utrecht, Utrecht UniversityDepartment of Pathology, University Medical Center Utrecht, Utrecht UniversityDepartment of Pathology, University Medical Center Utrecht, Utrecht UniversityDepartment of Radiology, University Medical Center Utrecht, Utrecht UniversityImage Sciences Institute, University Medical Center Utrecht, Utrecht UniversityDepartment of Medical Oncology, University Medical Center Utrecht, Utrecht UniversityMedical Image Analysis, Department of Biomedical Engineering, Eindhoven University of TechnologyAbstract 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.https://doi.org/10.1038/s41598-024-81188-2 |
| spellingShingle | Laurens S. Ter Maat Rob A. J. De Mooij Isabella A. J. Van Duin Joost J. C. Verhoeff Sjoerd G. Elias Tim Leiner Wouter A. C. van Amsterdam Max F. Troenokarso Eran R. A. N. Arntz Franchette W. P. J. Van den Berkmortel Marye J. Boers-Sonderen Martijn F. Boomsma Fons J. M. Van den Eertwegh Jan Willem de Groot Geke A. P. Hospers Djura Piersma Art Vreugdenhil Hans M. Westgeest Ellen Kapiteijn Ardine A. De Wit Willeke A. M. Blokx Paul J. Van Diest Pim A. De Jong Josien P. W. Pluim Karijn P. M. Suijkerbuijk Mitko Veta Deep learning on CT scans to predict checkpoint inhibitor treatment outcomes in advanced melanoma Scientific Reports |
| title | Deep learning on CT scans to predict checkpoint inhibitor treatment outcomes in advanced melanoma |
| title_full | Deep learning on CT scans to predict checkpoint inhibitor treatment outcomes in advanced melanoma |
| title_fullStr | Deep learning on CT scans to predict checkpoint inhibitor treatment outcomes in advanced melanoma |
| title_full_unstemmed | Deep learning on CT scans to predict checkpoint inhibitor treatment outcomes in advanced melanoma |
| title_short | Deep learning on CT scans to predict checkpoint inhibitor treatment outcomes in advanced melanoma |
| title_sort | deep learning on ct scans to predict checkpoint inhibitor treatment outcomes in advanced melanoma |
| url | https://doi.org/10.1038/s41598-024-81188-2 |
| work_keys_str_mv | AT laurensstermaat deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT robajdemooij deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT isabellaajvanduin deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT joostjcverhoeff deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT sjoerdgelias deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT timleiner deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT wouteracvanamsterdam deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT maxftroenokarso deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT eranranarntz deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT franchettewpjvandenberkmortel deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT maryejboerssonderen deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT martijnfboomsma deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT fonsjmvandeneertwegh deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT janwillemdegroot deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT gekeaphospers deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT djurapiersma deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT artvreugdenhil deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT hansmwestgeest deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT ellenkapiteijn deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT ardineadewit deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT willekeamblokx deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT pauljvandiest deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT pimadejong deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT josienpwpluim deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT karijnpmsuijkerbuijk deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma AT mitkoveta deeplearningonctscanstopredictcheckpointinhibitortreatmentoutcomesinadvancedmelanoma |