October 2, 2023
In today’s rapidly evolving landscape of accelerated drug development, reliance solely on data derived from traditional randomized control trials (RCTs) is shifting toward a combination of RCT data complemented by real-world data (RWD).
Background
A growing body of literature1 suggests RWD is a promising data source to advance our understanding of critical aspects within clinical pharmacology and translational science. Clinical pharmacology plays an important role in the drug development process to determine the optimal dosing regimen, taking into consideration both intrinsic factors (e.g., liver and renal impairments, genetic polymorphisms) and extrinsic factors (e.g., dietary habits, drug clearance of concurrent medications).
However, there is a lack of dosing information on patients often excluded from clinical trials, including protected vulnerable populations, such as patients with organ impairment. Clinical trials typically enroll both younger and healthier patients, which may not fully represent the population for whom the drug will be indicated. Therefore, uncertainty in clinical management remains for patient populations who were not studied in clinical trials.
Using RWD for regulatory PMCs / PMRs and label updates
RWD collected during routine clinical care can help bridge this knowledge gap by informing benefit/risk assessments and enabling the monitoring and evaluation of dosing, compliance, and adherence — ultimately improving patient management in these subpopulations. Safety and efficacy data in these patients can be helpful in determining if dose adjustments and/or contraindications are necessary. RWD can also be leveraged to streamline the eligibility criteria of clinical trials and inform dosing for dedicated organ impairment studies (i.e., to fulfill a post-marketing requirement or commitment).
Including information on patients with organ dysfunction in labeling promotes the safe and effective use of medical products across a broader patient population likely to use the drug in clinical practice. That is why health authorities may issue post-marketing requirements (PMR) and/or commitments (PMC) to further evaluate dosing in patients with organ dysfunction. For the 153 U.S. Food and Drug Administration (FDA) oncology drug approvals from January 1, 2021 to December 31, 2022, there were 272 clinical PMCs/PMR issued; of these, 24 were PMRs related to further evaluation of dosing in patients with kidney or liver dysfunction.
Study objectives
To evaluate the generalizability of clinical trial findings to patients with organ dysfunction, Flatiron Health and the FDA collaborated to conduct a retrospective observational study2 in patients with advanced non-small cell lung cancer (NSCLC) and moderate or severe hepatic or renal impairment — a population of patients often excluded from clinical trials. This study was designed to assess the association between baseline organ function and overall survival (OS) in patients treated with PD-1/PD-L1 blocking antibodies. Patient-level RWD from Flatiron’s electronic health records (EHR) and pooled clinical trial data submitted to the FDA were used as data sources for the study and FDA’s analyses. Flatiron and the FDA have collaborated on regulatory science research projects since 20163.
Study methods
The real-world cohort was generated using Flatiron’s longitudinal database of de-identified patient data, consisting of patient-level structured and unstructured data. This data was curated through technology-enabled abstraction, from electronic health records (EHRs) at ~280 US cancer clinics (~800 sites of care). The FDA pooled clinical trial data from nine randomized clinical trials supporting approvals of four PD-1/PD-L1 blocking antibodies as single-agent therapy for NSCLC. The trials were initiated from September 2012 through February 2018. Patients participating in the trials had advanced NSCLC (stage III or stage IV) and were treated with a PD-1/PD-L1 blocking antibody administered as monotherapy.
Study findings
NSCLC patients with moderate to severe kidney or liver impairment were represented in the real-world cohort, but not in the pooled clinical trial data. In the real-world cohort, 9,974 NSCLC patients were treated with anti-PD-1/PD-L1 monotherapy. 7,770 patients had kidney impairment, 2,018 of whom had moderate and 183 of whom had severe kidney impairment. 7,769 patients had liver impairment, 40 of whom had moderate impairment and 12 of whom had severe liver impairment. The pooled clinical trial cohort was composed of 3,214 patients with NSCLC treated with anti-PD-L1/PD1 monotherapy. Of this cohort, 3,089 NSCLC patients had kidney impairment and 2,681 had liver impairment. Patients with missing kidney function information (n=116), severe kidney impairment (n=3), and end-stage kidney disease (ESRD) (n= 6) were excluded from the analysis due to small sample size. Patients with missing liver impairment status (n=529) and moderate liver impairment (n=4) were excluded due to small sample size. The clinical trial dataset had no patients with severe liver impairment.
Baseline kidney impairment did not appear to be associated with real-world overall survival (rwOS) in the real-world cohort or OS in the pooled clinical trial cohort. In contrast, patients with baseline liver impairment had shorter rwOS in the real-world cohort and OS in the pooled clinical trial cohort.
Key takeaways
While there are limitations with RWD, such as sample size, data missingness, and unmeasured confounding, this exploratory study demonstrates the potential opportunity for RWD to support drug development by characterizing outcomes for a subset of patients not included in clinical trials. These results may inform clinical management decisions for a broader population and provide evidence supporting the expansion of eligibility criteria in future trials. This collaborative research shows promise that RWD can complement clinical trial data to better understand populations underrepresented in clinical trials and support drug development and communications to healthcare practitioners.
Source: https://flatiron.com/resources/fda-and-flatiron-healths-collaborative-study-to-evaluate-real-world-outcomes-in-patients-with-organ-impairment