Recent Advances in Biostatistics II

Abstract: Combination therapy is the hallmark of therapies for cancer and other diseases involving complex biological networks. Synergistic drug combinations, which are more effective than predicted from the summing effects of individual drugs, often achieve an increased therapeutic index. Because drug responses involve multiple doses of a single drug, the number of drug combinations increases exponentially. Resulting in a challenging high-dimensional statistical problem. The lack of proper design and analysis methods for multi-drug combination studies has resulted in missed therapeutic opportunities. Although systems biology holds the promise to unveil complex interactions within biological systems, the knowledge of networks remains primarily at the level of topology. We propose a novel two-stage procedure starting with an initial selection by utilizing an in silico model with experimental data of single drugs and systems biology to identify promising combinations and methods to evaluate them. In this talk, I will focus on the maximal power experimental design on multi-drug combinations, statistical modeling of the joint dose effect, and its statistical properties, and an adaptive Bayesian phase I trial design for multidrug combinations with the modeling concept. The development of a histone deacetylase inhibitor combination will be discussed throughout.

Abstract: The instrumental variable method is widely used in the health and social sciences for identification and estimation of causal effects in the presence of potential unmeasured confounding. In order to improve efficiency, multiple instruments are routinely used, leading to concerns about bias due to possible violation of the instrumental variable assumptions. To address this concern, we introduce a new class of G-estimators that are guaranteed to remain consistent and asymptotically normal for the causal effect of interest provided that a set of at least γ out of K candidate instruments are valid, for γ≤K set by the analyst ex ante, without necessarily knowing the identity of the valid and invalid instruments. We provide formal semiparametric efficiency theory supporting our results. Both simulation studies and applications to the UK Biobank data demonstrate the superior empirical performance of our estimators compared to competing methods.

Abstract: We study the domain adaptation problem with label shift in this work. Under the label shift context, the marginal distribution of the label varies across the training and testing datasets, while the conditional distribution of features given the label is the same. Traditional label shift adaptation methods either suffer from large estimation errors or require cumbersome post-prediction calibrations. To address these issues, we first propose a moment-matching framework for adapting the label shift based on the geometry of the influence function. Under such a framework, we propose a novel method named Efficient Label Shift Adaptation (ELSA), in which the adaptation weights can be estimated by solving linear systems. Theoretically, the ELSA estimator is root-n-consistent (n is the sample size of the source data) and asymptotically normal. Empirically, we show that ELSA can achieve state-of-the-art estimation performances without post-prediction calibrations, thus, gaining computational efficiency.