Pooled efficacy analysis of phase II clinical trials of recurrent Glioblastoma

Abstract

Recurrent glioblastoma require more research due to its wide spread prevalence among people. Researchers are working tirelessly to enhance the accessibility, ease the evaluation as well as ensure the validation of the end points in clinical trials. In this study, we analyzed 391 Phase II clinical trials endpoints [236 Progression Free Survival (PFS), 225 Overall Survival (OS), 214 Overall Response Rate (ORR) and 37 Duration of Response (DOR)] in order to examine the efficacy and impact of anti-cancer agent in recurrent glioblastoma. We assessed the treatment effects for OS, PFS, ORR and DOR by using appropriate statistical methods. We observe statistically significant moderate positive correlation between PFS and OS (r = 0.48, 95% CI = 0.37-0.57, p < 0.00001). Similarly, there is a statistically significant weak positive correlation between ORR and OS (r = 0.14, 95% CI = 0.0012-0.28, p = 0.047). In contrast, there is no significant correlation between DOR and OS (r = 0.19, 95% CI = -0.161-0.502, p = 0.283). Moreover, linear regression analysis performed on full model (adjusted 𝑅2 = 0.2) showed that the independent variables (mPFS, ORR, wECOG, Age, Treatment Size and Targeted Agent) predicted the OS with 20% accuracy. However, in reduced model the independent variables (mPFS, ORR, Treatment Size & Targeted Agent) predicted the OS with 13% accuracy. Furthermore, mean PFS and mean OS of chemotherapy are greater than targeted therapy but p value in both the cases came higher than 0.05 which means obtained result is not statistically significant. On the other hand, mean ORR of targeted therapy is greater in comparison to chemotherapy but the p value (0.176) shows the obtained result is not statistically significant. Therefore, further studies with a larger datasets are required to validate our findings.