Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Prediction of early-stage melanoma recurrence using clinical and histopathologic features

Wan, Guihong ; Nguyen, Nga ; Liu, Feng ; DeSimone, Mia S. ; Leung, Bonnie W. ; Rajeh, Ahmad ; Collier, Michael R. ; Choi, Min Seok ; Amadife, Munachimso and Tang, Kimberly , et al. (2022) In NPJ precision oncology 6(1).
Abstract

Prognostic analysis for early-stage (stage I/II) melanomas is of paramount importance for customized surveillance and treatment plans. Since immune checkpoint inhibitors have recently been approved for stage IIB and IIC melanomas, prognostic tools to identify patients at high risk of recurrence have become even more critical. This study aims to assess the effectiveness of machine-learning algorithms in predicting melanoma recurrence using clinical and histopathologic features from Electronic Health Records (EHRs). We collected 1720 early-stage melanomas: 1172 from the Mass General Brigham healthcare system (MGB) and 548 from the Dana-Farber Cancer Institute (DFCI). We extracted 36 clinicopathologic features and used them to predict the... (More)

Prognostic analysis for early-stage (stage I/II) melanomas is of paramount importance for customized surveillance and treatment plans. Since immune checkpoint inhibitors have recently been approved for stage IIB and IIC melanomas, prognostic tools to identify patients at high risk of recurrence have become even more critical. This study aims to assess the effectiveness of machine-learning algorithms in predicting melanoma recurrence using clinical and histopathologic features from Electronic Health Records (EHRs). We collected 1720 early-stage melanomas: 1172 from the Mass General Brigham healthcare system (MGB) and 548 from the Dana-Farber Cancer Institute (DFCI). We extracted 36 clinicopathologic features and used them to predict the recurrence risk with supervised machine-learning algorithms. Models were evaluated internally and externally: (1) five-fold cross-validation of the MGB cohort; (2) the MGB cohort for training and the DFCI cohort for testing independently. In the internal and external validations, respectively, we achieved a recurrence classification performance of AUC: 0.845 and 0.812, and a time-to-event prediction performance of time-dependent AUC: 0.853 and 0.820. Breslow tumor thickness and mitotic rate were identified as the most predictive features. Our results suggest that machine-learning algorithms can extract predictive signals from clinicopathologic features for early-stage melanoma recurrence prediction, which will enable the identification of patients that may benefit from adjuvant immunotherapy.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; ; and , et al. (More)
; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; and (Less)
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
NPJ precision oncology
volume
6
issue
1
article number
79
publisher
Springer Nature
external identifiers
  • pmid:36316482
  • scopus:85140963991
ISSN
2397-768X
DOI
10.1038/s41698-022-00321-4
language
English
LU publication?
yes
id
82b93de5-32aa-4180-ad5f-00cd89a4b702
date added to LUP
2022-12-05 15:28:40
date last changed
2024-04-18 15:57:10
@article{82b93de5-32aa-4180-ad5f-00cd89a4b702,
  abstract     = {{<p>Prognostic analysis for early-stage (stage I/II) melanomas is of paramount importance for customized surveillance and treatment plans. Since immune checkpoint inhibitors have recently been approved for stage IIB and IIC melanomas, prognostic tools to identify patients at high risk of recurrence have become even more critical. This study aims to assess the effectiveness of machine-learning algorithms in predicting melanoma recurrence using clinical and histopathologic features from Electronic Health Records (EHRs). We collected 1720 early-stage melanomas: 1172 from the Mass General Brigham healthcare system (MGB) and 548 from the Dana-Farber Cancer Institute (DFCI). We extracted 36 clinicopathologic features and used them to predict the recurrence risk with supervised machine-learning algorithms. Models were evaluated internally and externally: (1) five-fold cross-validation of the MGB cohort; (2) the MGB cohort for training and the DFCI cohort for testing independently. In the internal and external validations, respectively, we achieved a recurrence classification performance of AUC: 0.845 and 0.812, and a time-to-event prediction performance of time-dependent AUC: 0.853 and 0.820. Breslow tumor thickness and mitotic rate were identified as the most predictive features. Our results suggest that machine-learning algorithms can extract predictive signals from clinicopathologic features for early-stage melanoma recurrence prediction, which will enable the identification of patients that may benefit from adjuvant immunotherapy.</p>}},
  author       = {{Wan, Guihong and Nguyen, Nga and Liu, Feng and DeSimone, Mia S. and Leung, Bonnie W. and Rajeh, Ahmad and Collier, Michael R. and Choi, Min Seok and Amadife, Munachimso and Tang, Kimberly and Zhang, Shijia and Phillipps, Jordan S. and Jairath, Ruple and Alexander, Nora A. and Hua, Yining and Jiao, Meng and Chen, Wenxin and Ho, Diane and Duey, Stacey and Németh, István Balázs and Marko-Varga, Gyorgy and Valdés, Jeovanis Gil and Liu, David and Boland, Genevieve M. and Gusev, Alexander and Sorger, Peter K. and Yu, Kun Hsing and Semenov, Yevgeniy R.}},
  issn         = {{2397-768X}},
  language     = {{eng}},
  number       = {{1}},
  publisher    = {{Springer Nature}},
  series       = {{NPJ precision oncology}},
  title        = {{Prediction of early-stage melanoma recurrence using clinical and histopathologic features}},
  url          = {{http://dx.doi.org/10.1038/s41698-022-00321-4}},
  doi          = {{10.1038/s41698-022-00321-4}},
  volume       = {{6}},
  year         = {{2022}},
}