To request services, please email the BSR Director, Hao Liu.
The BSR offers the following support to Rutgers Cancer Institute members:
Research Project Design and Development
Biostatisticians discuss statistical considerations of study design, including objectives and aims, realistic endpoints, study population, randomization schema, quality assurance, sample size, and power analyses, and develop the statistical methods for data analyses. These statistical considerations are essential for high-quality scientific studies that are efficient and reproducible.
- For Basic Research: Biostatisticians help the investigator review experiments, hypotheses, methodology, and data characteristics. Methods for analyses depend on whether the data collected will be discrete or continuous, clustered in time or space, and univariate or multivariate, or whether there are additional variables to consider. For example, for the Cancer Metabolism and Immunology Program, Biostatistics provides study designs and data analyses on experiments testing multiple chemopreventive compounds for their synergistic effects in animal models to be translated to human chemoprevention clinical studies.
- For Clinical Research: Investigators are required to utilize Biostatistics for all investigator-initiated clinical trials. Examples of the types of analyses that are performed include:
- Phase 0: Design early-stage studies that sufficiently inform later-phase studies
- Phase I: determination/assessment of maximum-tolerated dose, dose-sequence levels, assumptions of dose-toxicity relationships, dose escalation and/or de-escalation strategies, and statistical properties of sequential dose-adjustment methods;
- Phase II: determination/calculation of the magnitude of treatment effect that the protocol seeks to exclude and with which to power the study, and the associated type-I and type-II error rates;
- Correlative clinical and laboratory studies: planning of regression or association analyses; assessment of conclusions to be drawn from the observed response in tissue and serum assays;
- Calculation of patient characteristics that could affect response, survival, or other biological effects;
- Evaluation of sources of patients available for the study, including a realistic appraisal of sample heterogeneity, projected accrual rates, and length of study.
- The recent increase in precision medicine umbrella and basket trials, often reliant on biomarkers, require new statistical thinking toward efficiency of clinical trial design and analysis.
- For Population Research: Biostatistical support accommodates a wide range of needs, including cancer screening programs and health behavior studies for cancer prevention and control; data analyses based on national databases (such as Surveillance, Epidemiological and End Result (SEER), and Medicare) and data management and analyses of large cohort studies. Recommendations are made regarding the organization of data into appropriate subset databases.
Study Monitoring in Addition to PRMS
Biostatistics supports study monitoring as follows: 1) Biostatistics members who do not serve on the Scientific Review Board (SRB) (Rutgers Cancer Institute’s Protocol Review and Monitoring System) are often called upon by the SRB to review protocols (a Biostatistics reviewer will never be asked to review a protocol for which they wrote or helped write the statistical section); and 2) all Biostatistics members do various analyses in support of monitoring – stopping rules need to be checked, which may require special analyses (e.g., when “response” is defined in terms of rising PSA, statistical analysis is required to calculate the slope of PSA doubling time); for Bayesian toxicity monitoring, Biostatistics is called on for interpretation, and the numbers need to be re-calculated once they are used; when a Phase II trial using Simon’s 2-stage design is slow in enrollment, Biostatistics is called upon to evaluate the statistical justification of possible early termination or extension of the trial.
Developing Novel Supportive Methodology
In some instances, consultation on project design identifies the need to develop unique analytical methods to fully capitalize on the data, conserve resources, monitor the study, or optimize experiments. Biostatistics developed statistical methods for a wide range of needs including analyzing DNA microarray data, small pilot studies for discovering biomarkers, algorithm-based designs, and model-based designs for Phase I cancer studies, and adaptive designs for Phase II cancer studies. During the recent years, additional methods have been developed such as estimation treatment effect following a clinical trial using an adaptive design, prevention and handling of missing data, two-stage winner trials with survival outcomes, survival with cure model, and dose-finding method by joint modeling efficacy and safety endpoints. A Biostatistics seminar introduced Bayesian designs to CINJ researchers, and Biostatistics applied modified CRM to combination therapies in recent studies. The method of Lin and Shih for adaptive phase IIA single arm trials has been used by researchers in other institutions lately for “basket” trials in molecular/genome directed precision medicine such as Vemurafenib in Multiple Nonmelanoma Cancers with BRAF V600 Mutations.
Protocol and Grant Proposal Review
Biostatistics reviews grant proposals from basic, clinical, and population science researchers, including human and non-human, follow-up, and cross-sectional survey studies.
Computation and Statistical Analyses
Methods for appropriate data analyses depend on the objectives and design of the study. To ensure consistency and efficiency, Biostatistics established a standard operating procedure for developing the Data Analysis Plan. In general, to conduct and interpret the results of the statistical analysis used to test the hypotheses of a study, Biostatistics: 1) provides a preliminary data review to ensure that the quality and assumptions required for the analysis are satisfied; 2) completes all statistical analyses required to test the hypotheses considered in the study (including interim analyses when required); 3) applies multivariate logistic regression analyses, Cox models for survival or progression-free survival data, as needed; and 4) creates tables/graphs describing subject demographic characteristics, primary and secondary endpoints and exploratory biomarker data, and range of possible confounders associated with measurements that could affect the response of interest. For biomarker analyses, Biostatistics works with the Biomedical Informatics Shared Resource to identify candidate biomarkers and then with clinical investigators to analyze the association with clinical outcomes.
Biostatistics assists in data reporting: to enhance quality, efficiency, and consistency. Standard operating procedures were developed for data spreadsheet formats and statistical reports. In collaboration with investigators, Biostatistics meets the specific requirements for each study or protocol and for the submission of data for presentations, audits, and publications.
Education, Consultation, and Training Activities
Biostatistics helped develop the curriculum for the training program in clinical protocol development and plays an important role in ongoing grant development training and mentoring. Additional activities include teaching credit-earning courses, seminar series, guest lectures, and individual educational/consulting assistance in study design and data interpretation.
Last updated 01/19/2023