Assessment of Response to Tyrosine Kinase Inhibitors in Metastatic Renal Cell Cancer: CT Texture as a Predictive Biomarker

Purpose: To assess the changes in tumor CT texture following 2 cycles of TKIs; and to determine if tumor texture correlates with measured time to progression in metastatic renal cell cancer treated with TKIs. Materials and Methods: Institutional review board waiver was obtained for this retrospective analysis. Contrast enhanced CT texture parameters were assessed in 39 patients with metastatic renal cell cancer treated with a TKI. 87 metastases were analysed at baseline and following 2 cycles of treatment: changes in tumor entropy and uniformity, derived using a software algorithm that selectively filters and extracts texture at different scales (fine to coarse detail: 1.0 to 2.5), were recorded. Response assessment was also obtained using RECIST, Choi and modified Choi criteria. The correlation of texture parameters and standard criteria to measured time to progression was assessed by Kaplan-Meier analysis and a Cox regression model. Statistical significance was set at 5%. Results: Tumor entropy decreased by 3-45% and uniformity increased by 5-21% for the different scale values following TKI treatment. Using a threshold change of =-2% for uniformity at a coarse scale value of 2.5, Kaplan Meier curves of the proportion of patients without disease progression were significantly different, and better than for standard response assessment (p=0.008 versus p=0.267, p=0.053, p=0.042 for RECIST, Choi, and Modified Choi criteria respectively). Cox regression analysis showed that texture uniformity was an independent predictor of time to progression (odds ratio, OR =4.02; 95% CI 1.52,10.65; p=0.005). Conclusion: CT texture analysis, reflecting tumor heterogeneity, is an independent factor associated with time to progression, and has potential as a predictive imaging biomarker of response of metastatic renal cancer to targeted therapy.