KEY TAKEAWAYS
- The study aimed to assess data attributes in RWD to improve methods for capturing and evaluating real-world treatment responses.
- The results concluded that aligning data sources can evaluate real-world response endpoints but needs further refinement.
Real-world data (RWD) offers potential for linking real-world (rw) outcomes to drug interventions, though determining rw-response from RWD can be challenging. Friends of Cancer Research collaborated to evaluate data attributes related to rw-response across RWD sources, aiming to improve methods for capturing, defining, and evaluating rw-response in patients with non-small cell lung cancer (NSCLC).
Brittany A McKelvey and the team aimed to evaluate data attributes in real-world data to enhance methods for capturing and assessing real-world treatment responses.
This retrospective observational study involved 7 electronic health record data companies, which provided summary-level deidentified data from 200 patients with mNSCLC treated with first-line platinum doublet chemotherapy. Data providers assessed the availability and frequency of components for evaluating real-world response, such as images, radiology reports, and clinician assessments.
A standardized protocol was used to evaluate and report real-world response endpoints, including real-world response rate (rwRR), real-world duration of response (rwDOR), and the association of rw-response with real-world overall survival (rwOS), real-world time to treatment discontinuation (rwTTD), and real-world time to next treatment (rwTTNT).
The results showed that the availability and timing of clinician assessments were relatively consistent across datasets, unlike images and image reports. Real-world response analysis, based on clinician assessments (median proportion of evaluable patients, 77.5%), demonstrated the highest consistency in assessment timing. Consistency was also observed in rwRR (median 46.5%) and the median and directionality of rwOS, rwTTD, and rwTTNT. However, there was variability in rwDOR across datasets.
The study concluded that aligning disparate data sources is feasible for evaluating real-world response endpoints using clinician-documented responses in patients with mNSCLC. However, there is variability in the availability of data components for assessing responses and related endpoints, indicating that further work is needed to improve drug effectiveness evaluation within real-world data sources.
Supported by Friends of Cancer Research. Research funding was provided by ConcertAI (Inst), Syapse, Flatiron Health and Merck Sharp & Dohme (Inst), Boehringer Ingelheim (Inst).
Source: https://www.boehringer-ingelheim.com/in
McKelvey BA, Garrett-Mayer E, Rivera DR, et al. (2024). “Evaluation of Real-World Tumor Response Derived From Electronic Health Record Data Sources: A Feasibility Analysis in Patients With Metastatic Non-Small Cell Lung Cancer Treated With Chemotherapy.” JCO Clin Cancer Inform. 2024;8:e2400091. doi:10.1200/CCI.24.00091
