General Information
Conflict of Interest
Request Clinical Trials
Associated Trial(s):- NCT03180736 - A Phase 3 Study Comparing Pomalidomide and Dexamethasone With or Without Daratumumab in Subjects With Relapsed or Refractory Multiple Myeloma Who Have Received at Least One Prior Line of Therapy With Both Lenalidomide and a Proteasome Inhibitor
- NCT03277105 - A Phase 3 Randomized, Multicenter Study of Subcutaneous vs. Intravenous Administration of Daratumumab in Subjects With Relapsed or Refractory Multiple Myeloma
- NCT02252172 - A Phase 3 Study Comparing Daratumumab, Lenalidomide, and Dexamethasone (DRd) vs Lenalidomide and Dexamethasone (Rd) in Subjects With Previously Untreated Multiple Myeloma Who Are Ineligible for High Dose Therapy
Request Clinical Trials
Data Request Status
Status: OngoingResearch Proposal
Project Title: Covariate-adjusted response-adaptive designs and their properties for censored survival responses
Scientific Abstract:
Background: Covariate-adjusted response-adaptive clinical trial (CARA) is a type of adaptive
clinical trial design that aims to make the study more efficient and informative by dynamically
adjusting the treatment allocation based on accumulated data while accounting for relevant covariate
profile of patients[1]. The simulation of progression free survival (PFS) for sample size
calculation, in real clinical trials, is often assumed to be from an exponential or a Weibull model.
In this proposed study, we want to extend the CARA designs based on all parametric two-parameter
location-scale distribution models like log-normal or log-logistic distribution of the right-censored
survival response.
Objective: To propose new adaptive designs for precision medicine; To evaluate the performance of a novel patient randomization method (CARA)
by redesigning clinical trials.
Study Design and Participants:NCT03180736, NCT03277105, NCT02252172, Part of
data will be used for model development, remaining part will be used for external validation.
Main Outcome Measures: Allocation proportions (standard deviation), number of events
and power of the designs, CR (Complete Response), PFS (progression free survival time) and OS (overall survival time).
Statistical Analysis: The main competing designs include the complete random design(CR),
two response adaptive designs: doubly-adaptive biased coin design(DBCD) (Hu and Zhang, 2004)[2],
efficient randomized-adaptive designs (ERADE) (Hu et al.,2009)[3] and the covariate adjusted
version of these two designs: CADBCD (Zhang and Hu, 2009)[8] and CAERADE (Hu et al.,2009)[3] under the optimal target allocation minimizing the total cumulative hazard subject to
the constraint that the asymptotic variance is constant. The Wald test is used to compare across
treatments.
Brief Project Background and Statement of Project Significance:
Covariate-adjusted response-adaptive designs modify the allocation of treatments in a clinical
trial based on interim responses, favoring the treatment identified as most beneficial for a patient’s
covariate profile. Such designs are developed for censored survival responses with exponential or
Weibull distribution. Empirically, the treatment allocation proportion for these designs converges
to the expected target value. Through extensive simulation studies assessing their operational
characteristics, it is evident that these designs can serve as viable alternatives to conventional balanced
randomization designs, assuming responses related to the primary endpoint are available for
adaptations during the recruitment phase. Earlier CARA designs for time-to-event outcome have
been developed for exponential regression models(Sverdlov et al., 2013)[6]. A more robust CARA
design considering censored Weibull time-to-event response was proposed, making it more relevant
for practical applications for increasing and decreasing hazards(Mukherjee et al.,2023)[5]. The
ambit of applications of the CARA designs can be enhanced further if the derived CARA designs
can be developed to include all models in the two-parameter location-scale class of distributions.
Specific Aims of the Project:
(i) To propose new CARA (covariate-adjusted response-adaptive) designs that can achieve better properties (more ethical and efficient)
(ii) To compare the allocation proportion to the better trial arm, number of events(progress in
RRMM or death) in each design(ethics) and their power for detecting the treatment effect(efficiency).
Study Design: Methodological research
What is the purpose of the analysis being proposed? Please select all that apply.: Participant-level data meta-analysis Meta-analysis using only data from the YODA Project Develop or refine statistical methods Research on clinical trial methods Research on clinical prediction or risk prediction
Software Used: R
Data Source and Inclusion/Exclusion Criteria to be used to define the patient sample for your study:
We will use the data (participant-level data from YODA trials) as motivated examples to design a better clinical trials. The propose designs are call CARA (covariate-adjusted response-adaptive) designs. Then we will redesign the clinical trials based on the features of the real clinical trials. The advantages of proposed designs are also demonstrated by redesigning these real clinical trials.
By using the participant-level data, we can decide the important covariates, then use these covariates in the CARA designs.
Primary and Secondary Outcome Measure(s) and how they will be categorized/defined for your study:
CR (Complete Response), PFS (progression free survival time) and OS (overall survival time) are the three outcome measures.
We will use these three outcomes to demonstrate the advantages of the new adaptive designs and proposed methods.
Main Predictor/Independent Variable and how it will be categorized/defined for your study: This is a main statistical and biostatistical research project. We will select the independent variables based on the data.
Other Variables of Interest that will be used in your analysis and how they will be categorized/defined for your study: We will select the important covariates based on the data from the three studies (the participant-level data). These covariates are then used in the designs and inference.
Statistical Analysis Plan:
We will first use Cox modeling approaches to identify some important covariates. Then use these important covariates in the
covariate-adjusted response-adaptive (CARA) designs. Further, the two parameter location-scale distribution of the PFS and give the MLE(maximum
likelihood estimator) of the location and scale parameters. For a clinical trial with censored
parametric survival times, the optimal target allocation proportions can be obtained by using the
maximum likelihood approach. Consider minimizing the average hazard in the trial subject to
the constraint that the asymptotic variance of the difference in the logarithms of the estimated
scale parameters is a constant (Zhang and Rosenberger, 2007)[7]. This would ensure that, for any
choice of the number of patients allocated to each treatment, the power for testing for a difference
in the treatment effects would remain fixed. The given optimal allocation target also skewed
allocations towards the better treatment, yet the degree of skewing depends on the parameters of
the distribution. NCT03180736 can be re-designed to compare across the treatments using Wald
Test.
We need the participant-level data to decide the important covariates and their estimations to redesign the clinical trials by using CARA designs.
Narrative Summary:
The CARA(Covariate-adjusted response-adaptive) designs take patients’ covariate profiles
and skew the allocation proportion to the better arm with fewer events and comparable power
with RAR(response adaptive randomization) designs and CAR(covariate adaptive randomization).
Based on the parametric distribution of PFS(progress free survival) time, the optimal allocation
can be given for both CADBCD and CAERADE designs using maximum likelihood method and
can be adjusted using the patients’ covariate profile. The Wald Test is then used to test the
treatment effects and power is calculated. The meaning of this study is to find a more ethical,
more personalized, more efficient, that allocates more patients to the superior treatment based on
their patient characteristics by statistical validating.
Project Timeline:
Key milestone dates:
(i) project start date, March 30, 2024
(ii) analysis completion date, Dec 15, 2024
(iii) date manuscript drafted and first submitted for publication, Feb 15, 2025
and (iv) date results reported back to the YODA Project. March 30, 2025
Dissemination Plan:
We will publish two to three papers for biostatisticians and clinicians.
Here are some possible journals: Statistics in medicine; Biometrics; Statistical methods in medical studies, Statistics in biopharmaceutical research, etc.
Bibliography:
[1] F. Hu and W. F. Rosenberger. The theory of response-adaptive randomization in clinical trials.
John Wiley & Sons, 2006.
[2] F. Hu and L.-X. Zhang. Asymptotic properties of doubly adaptive biased coin designs for
multitreatment clinical trials. The Annals of Statistics, 32(1):268–301, 2004.
[3] F. Hu, L.-X. Zhang, and X. He. Efficient randomized-adaptive designs. The Annals of Statistics,
pages 2543–2560, 2009.
[4] J. Hu, H. Zhu, and F. Hu. A unified family of covariate-adjusted response-adaptive designs
based on efficiency and ethics. Journal of the American Statistical Association, 110(509):357–
367, 2015.
[5] A. Mukherjee, D. S. Coad, and S. Jana. Covariate-adjusted response-adaptive designs for
censored survival responses. Journal of Statistical Planning and Inference, 225:219–242, 2023.
[6] O. Sverdlov, W. F. Rosenberger, and Y. Ryeznik. Utility of covariate-adjusted responseadaptive
randomization in survival trials. Statistics in Biopharmaceutical Research, 5(1):38–53,
2013.
[7] L. Zhang and W. F. Rosenberger. Response-adaptive randomization for survival trials: the
parametric approach. Journal of the Royal Statistical Society Series C: Applied Statistics,
56(2):153–165, 2007.
[8] L.-X. Zhang and F. Hu. A new family of covariate-adjusted response adaptive designs and
their properties. Applied Mathematics-A Journal of Chinese Universities, 24(1):1–13, 2009.