495 – Random Effects and Mixed Models
Drug Interaction Assessment with Repeated Measurements in Fixed Ray Design
Shouhao Zhou
The University of Texas MD Anderson Cancer Center
Keila E. Torres
The University of Texas MD Anderson Cancer Center
Lan V. Pham
Chan Shen
The University of Texas MD Anderson Cancer Center
We considered the problem of assessing the joint effects of combined therapies in in-vitro studies with repeated measurements. To better estimate the confidence interval of combination index t a combination dose with observed effect, we proposed the mixed effects linear regression in the estimation of the dose-effect curve for each single drug and their combination. With unbiased estimation of variance components in median effect model, our approach improves the accuracy in construction of confidence interval of combination index for hypothesis testing of synergistic effects.