Combination of Whole-Body Baseline CT Radiomics and Clinical Parameters to Predict Response and Survival in a Stage-IV Melanoma Cohort Undergoing Immunotherapy.
artificial intelligence and machine-learning
biomarkers for immunotherapy
checkpoint blockade
imaging biomarkers
melanoma
prognostic biomarkers
Journal
Cancers
ISSN: 2072-6694
Titre abrégé: Cancers (Basel)
Pays: Switzerland
ID NLM: 101526829
Informations de publication
Date de publication:
17 Jun 2022
17 Jun 2022
Historique:
received:
28
05
2022
revised:
13
06
2022
accepted:
15
06
2022
entrez:
24
6
2022
pubmed:
25
6
2022
medline:
25
6
2022
Statut:
epublish
Résumé
This study investigated whether a machine-learning-based combination of radiomics and clinical parameters was superior to the use of clinical parameters alone in predicting therapy response after three months, and overall survival after six and twelve months, in stage-IV malignant melanoma patients undergoing immunotherapy with PD-1 checkpoint inhibitors and CTLA-4 checkpoint inhibitors. A random forest model using clinical parameters (demographic variables and tumor markers = baseline model) was compared to a random forest model using clinical parameters and radiomics (extended model) via repeated 5-fold cross-validation. For this purpose, the baseline computed tomographies of 262 stage-IV malignant melanoma patients treated at a tertiary referral center were identified in the Central Malignant Melanoma Registry, and all visible metastases were three-dimensionally segmented ( The extended model was not significantly superior compared to the baseline model for survival prediction after six and twelve months (AUC (95% CI): 0.664 (0.598, 0.729) vs. 0.620 (0.545, 0.692) and AUC (95% CI): 0.600 (0.526, 0.667) vs. 0.588 (0.481, 0.629), respectively). The extended model was not significantly superior compared to the baseline model for response prediction after three months (AUC (95% CI): 0.641 (0.581, 0.700) vs. 0.656 (0.587, 0.719)). The study indicated a potential, but non-significant, added value of radiomics for six-month and twelve-month survival prediction of stage-IV melanoma patients undergoing immunotherapy.
Sections du résumé
BACKGROUND
BACKGROUND
This study investigated whether a machine-learning-based combination of radiomics and clinical parameters was superior to the use of clinical parameters alone in predicting therapy response after three months, and overall survival after six and twelve months, in stage-IV malignant melanoma patients undergoing immunotherapy with PD-1 checkpoint inhibitors and CTLA-4 checkpoint inhibitors.
METHODS
METHODS
A random forest model using clinical parameters (demographic variables and tumor markers = baseline model) was compared to a random forest model using clinical parameters and radiomics (extended model) via repeated 5-fold cross-validation. For this purpose, the baseline computed tomographies of 262 stage-IV malignant melanoma patients treated at a tertiary referral center were identified in the Central Malignant Melanoma Registry, and all visible metastases were three-dimensionally segmented (
RESULTS
RESULTS
The extended model was not significantly superior compared to the baseline model for survival prediction after six and twelve months (AUC (95% CI): 0.664 (0.598, 0.729) vs. 0.620 (0.545, 0.692) and AUC (95% CI): 0.600 (0.526, 0.667) vs. 0.588 (0.481, 0.629), respectively). The extended model was not significantly superior compared to the baseline model for response prediction after three months (AUC (95% CI): 0.641 (0.581, 0.700) vs. 0.656 (0.587, 0.719)).
CONCLUSIONS
CONCLUSIONS
The study indicated a potential, but non-significant, added value of radiomics for six-month and twelve-month survival prediction of stage-IV melanoma patients undergoing immunotherapy.
Identifiants
pubmed: 35740659
pii: cancers14122992
doi: 10.3390/cancers14122992
pmc: PMC9221470
pii:
doi:
Types de publication
Journal Article
Langues
eng
Subventions
Organisme : Deutsche Forschungsgemeinschaft
ID : #428216905
Références
Br J Dermatol. 2002 Jul;147(1):62-70
pubmed: 12100186
Front Oncol. 2021 Oct 07;11:704607
pubmed: 34692481
Radiology. 2018 Aug;288(2):407-415
pubmed: 29688159
Eur Radiol. 2019 Jun;29(6):3183-3191
pubmed: 30645669
J Clin Oncol. 2015 Jun 10;33(17):1889-94
pubmed: 25667295
Cancers (Basel). 2022 Jan 29;14(3):
pubmed: 35158980
Lancet. 2018 Sep 15;392(10151):971-984
pubmed: 30238891
N Engl J Med. 2015 Jan 1;372(1):30-9
pubmed: 25399551
Eur J Cancer. 2017 Sep;82:80-91
pubmed: 28651159
Nat Methods. 2021 Feb;18(2):203-211
pubmed: 33288961
Cancer Res. 2017 Nov 1;77(21):e104-e107
pubmed: 29092951
Radiographics. 2017 Sep-Oct;37(5):1483-1503
pubmed: 28898189
Br J Cancer. 2016 Feb 2;114(3):256-61
pubmed: 26794281
Invest Radiol. 2015 Nov;50(11):757-65
pubmed: 26115366
N Engl J Med. 2014 Nov 13;371(20):1867-76
pubmed: 25265494
Lancet. 2015 Aug 1;386(9992):444-51
pubmed: 26037941
N Engl J Med. 2015 Jan 22;372(4):320-30
pubmed: 25399552
Radiol Artif Intell. 2019 Mar 13;1(2):180014
pubmed: 33937787
Radiology. 2020 May;295(2):328-338
pubmed: 32154773
Mach Learn. 2018;107(12):1895-1922
pubmed: 30393425
Radiology. 2020 Mar;294(3):487-489
pubmed: 31891322
Phys Med Biol. 2016 Jul 7;61(13):R150-66
pubmed: 27269645
Comput Struct Biotechnol J. 2014 Nov 15;13:8-17
pubmed: 25750696
Front Oncol. 2020 Aug 25;10:1524
pubmed: 32984000
Front Oncol. 2021 Jun 10;11:670726
pubmed: 34178657
BMC Med Imaging. 2021 Apr 13;21(1):69
pubmed: 33849483
N Engl J Med. 2017 Oct 5;377(14):1345-1356
pubmed: 28889792
Cancer Med. 2020 Mar;9(5):1603-1612
pubmed: 31951322
Lancet Oncol. 2018 May;19(5):603-615
pubmed: 29573941
PeerJ. 2019 Feb 7;7:e6339
pubmed: 30746306
Ann Oncol. 2019 Jun 1;30(6):998-1004
pubmed: 30895304
Clin Cancer Res. 2020 Aug 15;26(16):4414-4425
pubmed: 32253232
Eur J Cancer. 2015 Dec;51(18):2785-91
pubmed: 26597444
AJR Am J Roentgenol. 2015 Sep;205(3):W283-93
pubmed: 26295664
IEEE Trans Pattern Anal Mach Intell. 2005 Aug;27(8):1226-38
pubmed: 16119262