Biostatistics Shared Resource for Massey Comprehensive Cancer Center
The mission of the Massey Comprehensive Cancer Center Biostatistics Shared Resource (BSSR) is to promote excellence in cancer research by providing outstanding biostatistical support and collaboration to Massey members.
Our faculty and staff at the BSSR play an instrumental role in cancer research at the Massey Comprehensive Cancer Center. They design innovative cancer clinical trials via applications of their methodological research, oversee cancer center study protocols in the Protocol Review and Monitoring System (PRMS), and promote high-quality study conduct in the Data Safety Monitoring Committee (DSMC). The BSSR Core also trains postdoctoral scientists and clinical fellows in drafting pilot grants and K awards and provides training through seminars and individual consultation sessions.
The collaborative nature of this work optimizes the shared resources' effectiveness, resulting in support from peer-reviewed grants and publications at top-tier journals.
Goals
To achieve our mission, we aspire to the following goals:
- Foster long-term collaborations with cancer researchers, beginning with idea conception and continuing through the peer-review process and beyond.
- Promote the use of reproducible and open research practices
- Identify and tailor appropriate biostatistical tools and adhere to evidence-based decisions for each research question
- Provide training in study design, biostatistical analysis, and reproducible research practices.
Who Should Contact the BSSR
Our principal goal is to foster long-term scientific collaborations with Massey investigators, and we encourage prospective cancer researchers to contact us as soon as they conceive of their research question. We are trained to help cancer researchers think through the quantitative aspects of study design, including sampling and cohort identification, measurement definition, analysis set construction, and drafting statistical plans, and the sooner in the process we can begin to offer that assistance, the better both the collaboration and the research will be.
Another of our goals is to provide training in study design, biostatistical analysis, and reproducible research practices, so we are happy to provide consultations to discuss these matters; we can also provide advice on proposals, analysis plans, statistical code and output, and draft manuscripts.
If at any point in these consultations it becomes clear that the relationship needs to expand, we can discuss avenues to support that growth. We strongly encourage requests for assistance with projects to be made early in the research process and with sufficient time for biostatistical work to be completed.
Sound and strategic planning for effective and high-quality cancer research takes months, and it is difficult to guarantee high standards or performance in a few days or weeks. We prioritize requests for assistance with funded support for biostatistical analyses and grant proposals with sufficient time given for completion.
In order to better manage statistical work on projects with external deadlines, the BSSR suggests that users:
- Notify the Core at least 6 months before the grant deadline for competing NIH P series, U series, and SPORE grants (and similar sized foundation grants).
- Notify the Core at least 2 months before the grant deadline for competing NIH R series and K series grants (and similar sized foundation grants).
- Provide the Core at least 1 month before the deadline with data sets for an analysis for an abstract submission to a national or international meeting (e.g., ASCO, AACR, ASH, ASTRO).
Frequently Asked Questions
You should contact BSSR members or make a request for assistance as soon as you develop your research question. Only with sufficient time to support long-term collaboration can we help assure that your proposal or project will be successful and high quality.
Unfortunately, there is no such thing as a “quick” statistical analysis. One of our foundational goals is to promote high-quality, reproducible and translatable research, and we cannot guarantee that our advice or work will adhere to that goal on a compressed timeline without the ability to partake in the necessary long-term planning. Based on our high workload, we will likely not have the time to complete short-term projects and cannot in fairness reprioritize other projects to accommodate these types of requests.
For requests of grant proposal development, you should plan for many weeks to several months of collaboration. Note that each project is unique and brings its own challenges and opportunities, so the actual amount of effort required will likely vary. If your request is approved, we will provide an estimated timeline in our Statement of Work.
We often work in either the SAS and R statistical computing environments, so data need to be stored in a conventional file format (e.g. comma-separated file) that can be imported in those programs. In addition to all project summaries and dissemination material, we will provide annotated copies of all scripts used for data management and analysis.
The BSSR and the CCTR should be acknowledged for all consultations and collaborations with BSSR members. Particular BSSR members should be included as co-authors for any manuscripts and presentations resulting from collaborations, as specified in the Statement of Work. Please see the section on Guidelines & Acknowledgement Policies for specific details.
Indeed, BSSR members adhere to the professional and ethical guidelines from the American Statistical Association (ASA). Some of our ethical principles include but are not limited to the following conventions:
- BSSR members will take a scientific and impartial approach to interpreting results and drawing conclusions; we are not here to prove or disprove any desired outcome and will resist all efforts and encouragements to do so.
- BSSR members will establish all analyses and tests before data are collected in the Statistical Analysis Plan (SAP), and strictly adhere to that plan; we will not conduct additional analyses or deviate from the agreed-upon scope of work (i.e. phishing or p-hacking) once the SAP is established.
- BSSR members will also clearly list any potential limitations resulting from our study design choices.
To discuss biostatistical support to meet your scientific goals, contact Nolan Wages, Ph.D., at wagesn@vcu.edu. Please note that in addition to the BSSR director, we have also assigned biostatisticians specifically to support each cancer center program area.
The BSSR is located in the Goodwin Research Laboratory, room 119, on VCU's MCV Campus. BSSR faculty and staff also reside in the Department of Biostatistics at One Capitol Square (seventh floor).
Nolan Wages, Ph.D.
BSSR director
wagesn@vcu.edu
Expertise: Phase I clinical trials; adaptive clinical trial designs; Bayesian modeling
Roy Sabo, Ph.D.
Biostatistician
rsabo@vcu.edu
Expertise: Phase I and II clinical trials; adaptive clinical trial designs; Bayesian modeling
Dipankar Bandyopadhyay, Ph.D.
Senior advisor, data and cancer modeling
dbandyop@vcu.edu
Expertise: development of biostatistical methods and computation algorithms for spatial cancer epidemiology
Bassam Dahman, Ph.D.
bdahman@vcu.edu
Expertise: Cancer prevention and control; integrating, linking and utilizing large datasets
Nitai Mukhopadhyay, Ph.D.
ndmukhopadhy@vcu.edu
Expertise: Bayesian methods, causality inference, network data analysis, image data analysis
Robert Perera, Ph.D.
rperera2@vcu.edu
Expertise: Cancer prevention and control, structural equation models
Priyam Das, Ph.D.
dasp4@vcu.edu
Expertise: Bayesian methods in biostatistics, EHR Data modeling, Cluster analysis
Xiaoyan Deng, M.S.
xdeng@vcu.edu
Jian He, M.S.
hej7@vcu.edu
Collaboration Process
The collaboration process begins when you submit a request through our electronic submission portal (provide link). At that point a team member will reach out to you electronically to set up a time to discuss your request. At the initial meeting – which may take place virtually or in-person – we will deeply discuss your request, revise if necessary, develop an action plan, and set scope-of-work and expectations. If your request is approved, then a Statement of Work detailing all agreed-upon activities (including any of the following: drafting of grant/proposal sections, sample size justifications, drafting informatics and database management plan, drafting of statistical analysis plans, and dissemination plan) will be delivered for review and signature within one week. Once the signed Statement of Work is received, BSSR team members will be assigned to your project and collaborations will proceed as planned.
Upon project completion, collaborators will receive a summary report of all activities and all agreed upon research materials (including, as applicable, any grant/protocol sections, manuscript and presentation sections, as well as all database and software scripts). Collaborators will also receive a link to complete an online evaluation form of the services provided by the BSSR team. Note that two important metrics used to evaluate the BSSR are the numbers of awarded research/grant proposals and peer-reviewed publications. As such, BSSR members will contact you periodically to check on the status of pending grants and publications.
For more details, please see the ASA guidance document, When You Consult a Statistician...What to Expect.