Evaluating eligibility criteria of oncology trials using real-world data and AI.

There is a growing focus on making clinical trials more inclusive but the design of trial eligibility criteria remains challenging1-3. Here we systematically evaluate the effect of different eligibility criteria on cancer trial populations and outcomes with real-world data using the computational framework of Trial Pathfinder. We apply Trial Pathfinder to emulate completed trials of advanced non-small-cell lung cancer using data from a nationwide database of electronic health records comprising 61,094 patients with advanced non-small-cell lung cancer. Our analyses reveal that many common criteria, including exclusions based on several laboratory values, had a minimal effect on the trial hazard ratios. When we used a data-driven approach to broaden restrictive criteria, the pool of eligible patients more than doubled on average and the hazard ratio of the overall survival decreased by an average of 0.05. This suggests that many patients who were not eligible under the original trial criteria could potentially benefit from the treatments. We further support our findings through analyses of other types of cancer and patient-safety data from diverse clinical trials. Our data-driven methodology for evaluating eligibility criteria can facilitate the design of more-inclusive trials while maintaining safeguards for patient safety.

Validation of diagnosis codes to identify side of colon in an electronic health record registry.

BACKGROUND: The use of real-world data to generate evidence requires careful assessment and validation of critical variables before drawing clinical conclusions. Prospective clinical trial data suggest that anatomic origin of colon cancer impacts prognosis and treatment effectiveness. As an initial step in validating this observation in routine clinical settings, we explored the feasibility and accuracy of obtaining information on tumor sidedness from electronic health records (EHR) billing codes. METHODS: Nine thousand four hundred three patients with metastatic colorectal cancer (mCRC) were selected from the Flatiron Health database, which is derived from de-identified EHR data. This study included a random sample of 200 mCRC patients. Tumor site data derived from International Classification of Diseases (ICD) codes were compared with data abstracted from unstructured documents in the EHR (e.g. surgical and pathology notes). Concordance was determined via observed agreement and Cohen's kappa coefficient (κ). Accuracy of ICD codes for each tumor site (left, right, transverse) was determined by calculating the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), and corresponding 95% confidence intervals, using abstracted data as the gold standard. RESULTS: Study patients had similar characteristics and side of colon distribution compared with the full mCRC dataset. The observed agreement between the ICD codes and abstracted data for tumor site for all sampled patients was 0.58 (κ = 0.41). When restricting to the 62% of patients with a side-specific ICD code, the observed agreement was 0.84 (κ = 0.79). The specificity (92-98%) of structured data for tumor location was high, with lower sensitivity (49-63%), PPV (64-92%) and NPV (72-97%). Demographic and clinical characteristics were similar between patients with specific and non-specific side of colon ICD codes. CONCLUSIONS: ICD codes are a highly reliable indicator of tumor location when the specific location code is entered in the EHR. However, non-specific side of colon ICD codes are present for a sizable minority of patients, and structured data alone may not be adequate to support testing of some research hypotheses. Careful assessment of key variables is required before determining the need for clinical abstraction to supplement structured data in generating real-world evidence from EHRs.

An evaluation of the impact of missing deaths on overall survival analyses of advanced non-small cell lung cancer patients conducted in an electronic health records database.

PURPOSE: The aim of this study was to assess the impact of missing death data on survival analyses conducted in an oncology EHR-derived database. METHODS: The study was conducted using the Flatiron Health oncology database and the National Death Index (NDI) as a gold standard. Three analytic frameworks were evaluated in advanced non-small cell lung cancer (aNSCLC) patients: median overall survival [mOS]), relative risk estimates conducted within the EHR-derived database, and "external control arm" analyses comparing an experimental group augmented with mortality data from the gold standard to a control group from the EHR-derived database only. The hazard ratios (HRs) obtained within the EHR-derived database (91% sensitivity) and the external control arm analyses, were compared with results when both groups were augmented with mortality data from the gold standard. The above analyses were repeated using simulated lower mortality sensitivities to understand the impact of more extreme levels of missing deaths. RESULTS: Bias in mOS ranged from modest (0.6-0.9 mos.) in the EHR-derived cohort with (91% sensitivity) to substantial when lower sensitivities were generated through simulation (3.3-9.7 mos.). Overall, small differences were observed in the HRs for the EHR-derived cohort across comparative analyses when compared with HRs obtained using the gold standard data source. When only one treatment arm was subject to estimation bias, the bias was slightly more pronounced, but increased substantially when lower sensitivities were simulated. CONCLUSIONS: The impact on survival analysis is minimal with high mortality sensitivity with only modest impact associated within external control arm applications.

Using Electronic Health Records to Derive Control Arms for Early Phase Single-Arm Lung Cancer Trials: Proof-of-Concept in Randomized Controlled Trials.

Oncology drug development increasingly relies on single-arm clinical trials. External controls (ECs) derived from electronic health record (EHR) databases may provide additional context. Patients from a US-based oncology EHR database were aligned with patients from randomized controlled trials (RCTs) and trial-specific eligibility criteria were applied to the EHR dataset. Overall survival (OS) in the EC-derived control arm was compared with OS in the RCT experimental arm. The primary outcome was OS, defined as time from randomization or treatment initiation (EHR) to death. Cox regression models were used to obtain effect estimates using EHR data. EC-derived hazard ratio estimates aligned closely with those from the corresponding RCT with one exception. Comparing log HRs among all RCT and EC results gave a Pearson correlation coefficient of 0.86. Properly selected control arms from contemporaneous EHR data could be used to put single-arm trials of OS in advanced non-small cell lung cancer into context.

Development and Validation of a High-Quality Composite Real-World Mortality Endpoint.

OBJECTIVE: To create a high-quality electronic health record (EHR)-derived mortality dataset for retrospective and prospective real-world evidence generation. DATA SOURCES/STUDY SETTING: Oncology EHR data, supplemented with external commercial and US Social Security Death Index data, benchmarked to the National Death Index (NDI). STUDY DESIGN: We developed a recent, linkable, high-quality mortality variable amalgamated from multiple data sources to supplement EHR data, benchmarked against the highest completeness U.S. mortality data, the NDI. Data quality of the mortality variable version 2.0 is reported here. PRINCIPAL FINDINGS: For advanced non-small-cell lung cancer, sensitivity of mortality information improved from 66 percent in EHR structured data to 91 percent in the composite dataset, with high date agreement compared to the NDI. For advanced melanoma, metastatic colorectal cancer, and metastatic breast cancer, sensitivity of the final variable was 85 to 88 percent. Kaplan-Meier survival analyses showed that improving mortality data completeness minimized overestimation of survival relative to NDI-based estimates. CONCLUSIONS: For EHR-derived data to yield reliable real-world evidence, it needs to be of known and sufficiently high quality. Considering the impact of mortality data completeness on survival endpoints, we highlight the importance of data quality assessment and advocate benchmarking to the NDI.