Modeling tumor growth kinetics after treatment with pazopanib or placebo in patients with renal cell carcinoma

Abstract

Purpose: The purpose of this study is to characterize tumor growth kinetics in patients with renal cell carcinoma (RCC) after treatment with pazopanib or placebo and to identify predictive patient-specific covariates.

Methods: Different tumor growth models that included patient-specific covariates were fit to tumor growth data from Phase 2 (n = 220) and Phase 3 (n = 423) clinical trials using nonlinear mixed-effects modeling. Logistic regression was used to determine whether individual model parameters or covariates were related to occurrence of new lesions.

Results: A modified Wang model that included a quadratic growth term and a mixture model adequately described the data. Patients in Group 1 (93 %) showed treatment-dependent tumor shrinkage followed by treatment-independent regrowth. Patients in Group 2 (7 %) showed treatment-independent tumor shrinkage that did not regrow. In Group 1, pazopanib 800 mg increased the tumor shrinkage rate by 267 % compared to placebo. Baseline tumor size was dependent on baseline hemoglobin, baseline lactate dehydrogenase, study, and prior nephrectomy. Logistic regression analysis showed that prior radiotherapy, baseline tumor size, tumor shrinkage rate, tumor regrowth rate, study, and treatment (P < 0.01 for all) were all important predictors of new lesions. Patients treated with placebo were approximately twice as likely to develop new lesions than patients treated with pazopanib.

Conclusions: Mathematical modeling of tumor growth kinetics can quantify the effect of anticancer therapies. Pazopanib 800 mg was shown to be an effective treatment for RCC that increased the tumor shrinkage rate by 267 % compared with placebo and reduced the likelihood of developing new lesions.