OR WAIT null SECS
Drug development in oncology continues to evolve, Sarah Alwardt, VP of Data, Evidence, and Insight Operations for McKesson, discusses the shift to highly personalized therapies, including new modalities like cell and gene therapies.
Drug development in oncology continues to evolve, and we are seeing a shift from broad chemotherapy to highly personalized therapies, including new modalities like cell and gene therapies. Real-World Evidence (RWE) has increasingly become important in oncology where treatment decisions are impacted by the fact that there are many forms of disease and many rare conditions where there is no approved treatment or only conditional approvals.
Build RWE into your regulatory program now
Randomized Controlled Trials (RCT) may always remain the gold standard for evaluating the safety and efficacy of drugs for regulatory decisions, but the industry must identify alternate methods of gaining insight into treatment patterns and performance. We must think about RWE differently – not as trial rescue, where patient recruiting or data collection lags due to small patient populations, but as a planned study designed to capture data on the day-to-day usefulness of drugs. FDA Approval of a first-line therapy for Merkel cell carcinoma is an excellent example of how electronic health record (EHR) data was used in a real-world evidence study to provide essential information about treatment responses and patient outcomes.
A recent article discussed the U.S. Food & Drug Administration’s (FDA) framework for its real-world evidence program (RWE Framework), focusing on the scope of the program, as well as challenges and opportunities for the biopharmaceutical industry. The FDA is predicting that greater use of RWE will result in safety and efficacy information becoming available sooner and helping to further inform regulatory decisions. However, the RWE Framework does not define precisely how and when RWE can be used for regulatory filings. Ultimately, the FDA will be responsible for regulating the use of RWE for regulatory filings; however, the biopharmaceutical industry must be actively engaged in the development of the final guidelines for data collection, analysis and trial design that will advance innovative medicine.
In this article, we will offer actionable insights for biopharma companies looking to build RWE into their drug development and regulatory programs. As the FDA refines the RWE Framework, here are key factors the industry needs to consider in order for RWE to be implemented on a broader scale:
Assess data integrity and verification
To understand the treatment of oncology patients, it is critical to evaluate clinical data points (biomarkers, disease staging, location of metastases and disease progression, among others) as well as elements of the patient experience (adherence rates, reason for treatment discontinuation, toxicities and functional status, among others). This information, along with key demographic and patient characteristics, is widely available through real-time tracking of clinical and claims data elements across unified EHR and reimbursement systems.
According to the Friends of Cancer Research RWE Roundtable Project Proposal, “not every data source can answer all questions, but every data source can answer some questions.” The ability to transition the use of this data from patient management to regulatory filings will be based on validating data quality. McKesson is working with the Friends of Cancer Research, biopharma companies and other industry stakeholders to create rigorous standards for defining regulatory-grade data and quality assessment.
This starts with setting standards for data provenance, documenting the origin and tracing the lineage of the data. This includes validating structured data with chart notes to ensure that the data is consistent, complete and representative of the target patient populations. While data collection is tightly controlled in RCTs and all fields are complete, this is not realistic in RWE studies. In some cases, missing data should not be correlated to poor data quality. For example, with PD-L1 drugs, testing is not required. While it may look like data is missing, it is simply complete data with blank fields. This can be addressed by clearly outlining data requirements and limitations in the study design.
Rank EHR data sets
In addition to rigorously maintaining quality, real-world data (RWD) must also be a fit for its purpose. This becomes critical when it comes to evaluating the appropriateness of EHR data sets. For example, McKesson’s iKnowMedSM oncology EHR, which captures outpatient medical histories from community oncology practices treating approximately 1 million patients per year, has successfully been used to help biopharma companies understand the real-world utilization and outcomes associated with a number of oncology agents. However, that same data set may not have clinically relevant information regarding a cardiovascular therapy. Additionally, there are hundreds of EHRs in clinical use today. While all may be perfectly adequate for patient management and reporting, not all may be suitable for use as a source of RWD.
Aside from demonstrating “meaningful use” under the Medicare regulations, there is no standard format for EHR data sets. Required data fields in one system may not be required in another. Here again, as an industry, we must come to consensus on what is important: what is the minimum data set? Should there be a required data collection process? Other factors to consider when evaluating EHR data sets and RWE providers include required data fields, access to structured data and chart notes, and interoperability across data sources and formats.
Standardize definitions and protocols
Using rigorous RECIST criteria, which is a standard way to measure how well a cancer patient responds to treatment, RCT endpoints are clearly defined, and there is no deviation on how the data is collected or reported. In clinical practice, physicians use documentation methods that are very different than RCT data collection standards. Therefore, the definition of response and outcome measures in RWE may be less uniform than in clinical trials.
Further, in observational studies, we as an industry have not settled on definitions of many endpoints, which can impact analysis and extrapolation of the results. For example, date of death is seemingly simple, but is it when death is reported to SSI or is the date based on when it was entered into the EHR? Date of diagnosis must also be harmonized or there could be significant impacts to survival data results. Is diagnosis when the primary care physician sees something and requests follow up or when the oncologist confirms symptomology? Or, is diagnosis when a biopsy is taken or when the biopsy results come back?
As an industry, we have the opportunity to be thoughtful about how to use this data and establish a standard set of endpoints with common definitions that are critical to applying the benefits of RWE to both regulatory and value-based decisions. In some cases, this is to establish a proxy for traditional RCT endpoints and in other cases to show reproducibility. For example, RWE provided useful data to confirm clinical trial results in the case of pertuzumab, a monoclonal antibody for HER2-positive breast cancer. The data were used to support the use of this regimen for the management of breast cancer in Europe.
Utilize new statistical models
There are currently some 2,000 oncology trials listed on ClinicalTrials.gov, and only 3-4 percent of the patient population qualifies for participation. As a result, clinical trials may not be able to recruit enough participants in a timely manner.
The growth of personalized medicine and the complexity of the oncology treatment landscape offer an opportunity for the industry to pivot to RWE studies. However, statistics applied to large patient populations will not work on small populations. This will necessitate the development and utilization of new statistical tools to answer questions accurately and reliably. According to the FDA, synthetic control arms could help augment new clinical data by allowing biopharma company sponsors to reduce the number of subjects assigned to the control arm in a randomized trial, or to conduct smaller randomized trials.iii.
Synthetic control cohorts must be embraced to support drug development and regulatory approval of breakthrough medicines designed to improve patient care.
Translating RWE into real-world benefits
The FDA is actively supportive of the expanded use of RWE to inform a variety of FDA regulatory decisions. Observational studies using RWE provide significant opportunities to gain insight into treatment patterns and outcomes in clinical practice.
In order to encourage acceptance and trust in RWE by regulatory bodies, patients and clinicians, the biopharmaceutical industry must actively work to create rigorous standards for regulatory-grade data and RWE trial design. This will lead to getting therapies to market faster, which is better for patients and biopharma companies.
i. Miller AB, Hoogstraten B, Staquet M, Winkler A. Reporting results of cancer treatment. Cancer. 1981;47:207–14. [PubMed] [Google Scholar]
ii. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, Verweij J, Van Glabbeke M, van Oosterom AT, Christian MC, Gwyther SG. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst. 2000;92:205–16. [PubMed] [Google Scholar]
iii. Gottlieb, S. Breaking down barriers between clinical trials and clinical care: Incorporating real world evidence into regulatory decision making. fda.gov. 2019 [fda.gov]
Sarah Alwardt, PhD, Vice President, Data, Evidence, and Insights Operations, McKesson