Metastatic melanoma: pretreatment contrast-enhanced CT texture parameters as predictive biomarkers of survival in patients treated with pembrolizumab
To determine whether texture analysis features on pretreatment contrast-enhanced computed tomography (CT) images can predict overall survival (OS) and progression-free survival (PFS) in patients with metastatic malignant melanoma (MM) treated with an anti-PD-1 monoclonal antibody, pembrolizumab.
Materials and methods
This institutional-approved retrospective study included 31 patients with metastatic MM treated with pembrolizumab. Texture analysis of 74 metastatic lesions was performed on CT scanners obtained within 1 month before treatment. Mean gray-level, entropy, kurtosis, skewness, and standard deviation values were derived from the pixel distribution histogram before and after spatial filtration at different anatomic scales, ranging from fine to coarse. Lasso penalized Cox regression analyses were performed to identify independent predictors of OS and PFS.
Median OS and PFS were 357 days (range 42–1355) and 99 days (range 35–1185), respectively. Skewness at coarse texture scale (SSF = 6; HR (CI 95%) = 6.017 (1.39, 26.056), p = 0.016), Response evaluation criteria in solid tumors (RECIST) conclusion (HR (CI 95%) = 3.41 (1.17, 9.89), p = 0.024), and body weight (HR (CI 95%) = 0.96 (0.92, 0.995), p = 0.026) were independent predictors of OS. Skewness at coarse texture scale (SSF = 6; HR (CI 95%) = 4.55 (1.46, 14.13), p = 0.0089) and RECIST conclusion (HR (CI 95%) = 10.63 (3.11, 36.29), p = 0.00016) were independent predictors of PFS. Skewness values above − 0.55 at coarse texture scale were significantly associated with both lower OS and lower PFS after administration of pembrolizumab.
Pretreatment CT texture analysis–derived tumor skewness may act as predictive biomarker of OS and PFS in patients with metastatic MM treated with pembrolizumab.
• Pretreatment skewness at coarse texture scale in metastases from malignant melanoma was an independent predictor of overall survival and progression-free survival.
• Skewness values above −0.55 at coarse texture scale were significantly associated with both lower OS and lower PFS after administration of pembrolizumab.
• In patients with metastatic MM, texture analysis performed on pretreatment CT may act as a useful tool to select the best candidates for pembrolizumab therapy.
KeywordsMetastatic melanoma Pembrolizumab Tomography, X-ray computed Biomarkers Survival
Eastern cooperative oncology group
Serum lactate dehydrogenase
Program cell death 1
Response evaluation criteria in solid tumors
Region of interest
Spatial scale image filtration
The authors state that this work has not received any funding.
Compliance with ethical standards
The scientific guarantor of this publication is Carole Durot, MD, Centre Hospitalo-universitaire de Reims.
Conflict of interest
The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.
Statistics and biometry
One of the authors has significant statistical expertise.
Written informed consent was obtained from all subjects (patients) in this study.
Institutional Review Board approval was obtained.
• diagnostic or prognostic study
• performed at one institution
- 10.Tsai KK, Loo K, Khurana N et al (2015) Clinical characteristics predictive of response to pembrolizumab in advanced melanoma. J Clin Oncol 33:9031–9031Google Scholar
- 13.Eisenhauer EA, Therasse P, Bogaerts J et al (2009) New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 45:228–247Google Scholar
- 14.Rao SX, Lambregts DM, Schnerr RS et al (2016) CT texture analysis in colorectal liver metastases: a better way than size and volume measurements to assess response to chemotherapy? United European Gastroenterol J 4:257–263Google Scholar
- 24.Mulé S, Thiefin G, Costentin C et al (2018) Advanced hepatocellular carcinoma: pretreatment contrast-enhanced CT texture parameters as predictive biomarkers of survival in patients treated with sorafenib. Radiology:171320Google Scholar
- 38.Ahn SY, Park CM, Park SJ et al (2015) Prognostic value of computed tomography texture features in non-small cell lung cancers treated with definitive concomitant chemoradiotherapy. Investig Radiol 50:719–725Google Scholar
- 39.Miles KA, Ganeshan B, Hayball MP (2013) CT texture analysis using the filtration-histogram method: what do the measurements mean? Cancer Imaging Soc 13:400–406Google Scholar