A5040545724- Statistical Methods in Clinical Trials
- Machine Learning in Bioinformatics
- Health Systems, Economic Evaluations, Quality of Life
- Statistical Methods and Inference
- Advanced Causal Inference Techniques
- Multiple Myeloma Research and Treatments
- Protein Degradation and Inhibitors
- Genetic Associations and Epidemiology
- Economic and Financial Impacts of Cancer
- Statistical Methods and Bayesian Inference
- Chronic Lymphocytic Leukemia Research
- Ubiquitin and proteasome pathways
- RNA Research and Splicing
- Genomics and Chromatin Dynamics
- Click Chemistry and Applications
- Chronic Myeloid Leukemia Treatments
- Advanced Control Systems Optimization
- Machine Learning in Materials Science
- COVID-19 and healthcare impacts
- Risk and Safety Analysis
- Statistical and Computational Modeling
- Gene expression and cancer classification
- PI3K/AKT/mTOR signaling in cancer
- Genetic Mapping and Diversity in Plants and Animals
- Bayesian Methods and Mixture Models
Harvard University
2015-2022
La Roche College
2022
Takeda (United States)
2019-2021
Millennium Engineering and Integration (United States)
2017-2021
Roche (Switzerland)
2021
COVID-19 outbreak has rapidly evolved into a global pandemic. The impact of on patient journeys in oncology represents new risk to interpretation trial results and its broad applicability for future clinical practice. We identify key intercurrent events that may occur due trials with focus time-to-event endpoints discuss considerations pertaining the other estimand attributes introduced ICH E9 addendum. propose strategies handle related events, depending their relationship malignancy...
Summary Peripheral neuropathy ( PN ) is the most troublesome adverse event associated with proteasome inhibitor bortezomib. Studies suggest an inflammatory aetiology for bortezomib‐induced (Bi and it has been hypothesized that reducing inflammation concomitant dexamethasone may reduce Bi incidence and/or severity. We retrospectively analysed rates from 32 studies (2697 patients previously untreated multiple myeloma) incorporating bortezomib differing schedules: partnered dosing (days of...
Asymptotic distributions under alternative hypotheses and their corresponding sample size power equations are derived for nonparametric test statistics commonly used to compare two survival curves. Test include the weighted log-rank Wald difference in (or ratio of) Kaplan-Meier probability, percentile survival, restricted mean time. Accrual, loss follow-up allowed follow any arbitrary continuous distribution. We show that Schoenfeld's equation-often by practitioners calculate required number...
An addendum of the ICH E9 guideline on Statistical Principles for Clinical Trials was released in November 2019 introducing estimand framework. This new framework aims to align trial objectives and statistical analyses by requiring a precise definition inferential quantity interest, that is, estimand. explicitly accounts intercurrent events, such as switching anticancer therapies analysis overall survival (OS), gold standard oncology. Traditionally, OS confirmatory studies is analyzed using...
There has been an increasing use of master protocols in oncology clinical trials because its efficiency to accelerate cancer drug development and flexibility accommodate multiple substudies. Depending on the study objective design, a protocol trial can be basket trial, umbrella platform or any other form which investigational products and/or subpopulations are studied under single protocol. Master external data evidence (e.g. controls) for treatment effect estimation, further improve trials....
Abstract Since the first approval of a tumor-agnostic indication in 2017, total seven indications involving six drugs have received from FDA. In this paper, master protocol subteam Statistical Methods Oncology Scientific Working Group, Biopharmaceutical Session, American Association, provides comprehensive summary these approvals, describing their mechanisms action; biomarker prevalence; study design; companion diagnostics; regulatory aspects, including comparisons global requirements; and...
Abstract Co-expression analysis reveals useful dysregulation patterns of gene cooperativeness for understanding cancer biology and identifying new targets treatment. We developed a structural strategy to identify co-expressed networks that are important chronic myelogenous leukemia (CML). This compared the distributions expressional correlations between CML normal states it identified data-driven threshold classify strongly had best coherence with CML. Using this strategy, we found...
The association between depth of response in multiple myeloma (MM) and long-term outcomes is well recognized [1][2][3].Thus, clinicians patients are often encouraged by a rapid decrease M-protein when treatment initiated, achieving ≥very-good partial (VGPR) 4 months initial diagnosis has been associated with decreased mortality [4].However, little known about the kinetics outcomes.While some reports suggest that early responders may have compromised compared late [5,6], these studies were...
ABSTRACT Case‐control association studies often collect from their subjects information on secondary phenotypes. Reusing the data and studying between genes phenotypes provide an attractive cost‐effective approach that can lead to discovery of new genetic associations. A number approaches have been proposed, including simple computationally efficient ad hoc methods ignore ascertainment or stratify case‐control status. Justification for these relies assumption no covariates correct...
Conditional power (CP) is a commonly used tool to inform interim decision-making in clinical trials, but the conventional approach using only primary endpoint data calculate CP may not perform well when requires long follow-up period, or treatment effect size changes during trial. Several methods have been proposed use additional short term auxiliary observed at analysis improve estimation these situations, however, they rely on strong assumptions, limited applications, ad hoc choices of...
Abstract Duration of response (DOR) and time to (TTR) are typically evaluated as secondary endpoints in early‐stage clinical studies oncology when efficacy is assessed by the best overall presented rate. Despite common use DOR TTR particular single‐arm studies, definition these questions they intended answer remain unclear. Motivated estimand framework, we present relevant scientific interest for propose corresponding definitions. We elaborate on how deal with intercurrent events which...
An important goal of precision medicine is to identify biomarkers that are predictive, and tailor the treatment according biomarker levels individual patients. Differentiating prognostic versus predictive impacts decision makings for patients treating physicians. Using Hazard Ratio (HR) can mistake a purely one leading disheartening possibility depriving beneficial as demonstrated in OAK trial. This stems from illogical issue HR at population level where marginal overall be larger than those...
There has been an increasing use of master protocols in oncology clinical trials because its efficiency and flexibility to accelerate cancer drug development. Depending on the study objective design, a protocol trial can be basket trial, umbrella platform or any other form which multiple investigational products and/or subpopulations are studied under single protocol. Master external data evidence (e.g., controls) for treatment effect estimation, further improve trials. This paper provides...
Innovation, defined as the successful implementation at scale of a new invention, is key for continued success drug development enterprise. In this article we focus on statistical innovation in pharmaceutical industry. We discuss both components, invention and scale, typically called commercialization, methods. These concepts are illustrated using three examples that successfully implemented our company. summarize factors foster or hinder commercialization. A discussion mindset consider...
We revisit the question of what randomization ratio (RR) maximizes power logrank test in event-driven survival trials under proportional hazards (PH). By comparing three approximations (Schoenfeld, Freedman, Rubinstein) to empirical simulations, we find that RR is balances number events across treatment arms at end trial. This contradicts common misconception implied by Schoenfeld's approximation 1:1 power. Besides power, consider other factors might influence choice (accrual, trial...
Advances in anticancer therapies have significantly contributed to declining death rates certain disease and clinical settings. However, they also made it difficult power a trial these settings with overall survival (OS) as the primary efficacy endpoint. A number of statistical approaches therefore been recently proposed for pre-specified analysis OS safety endpoint (Shan, 2023; Fleming et al., 2024; Rodriguez 2024). In this paper, we provide simple, unifying framework that includes...
COVID-19 outbreak has rapidly evolved into a global pandemic. The impact of on patient journeys in oncology represents new risk to interpretation trial results and its broad applicability for future clinical practice. We identify key intercurrent events that may occur due trials with focus time-to-event endpoints discuss considerations pertaining the other estimand attributes introduced ICH E9 addendum. propose strategies handle related events, depending their relationship malignancy...
Abstract Attempts to identify and prioritize functional DNA elements in coding noncoding regions, particularly through use of silico annotation data,continue increase popularity. However, specific roles may vary widely from one variant another, making it challenging summarize different aspects function. Here we propose Multi-dimensional Annotation Class Integrative Estimation (MACIE), an unsupervised multivariate mixed model framework capable integrating annotations diverse origin assess...