Abstract 6476: Solving the real-world data challenge: A hybrid resourcing model to curate RWD at scale
Bibliographic record
Abstract
Abstract Introduction: As cancer treatments become more complex, the need for structured real-world data (RWD) to support clinical research is critical. However, researchers often lack resources to curate data in-house. Organizations are challenged with timely data curation at the quality and scale needed for real world applications. Methods: As members of the AACR Project GENIE (Genomics Evidence Neoplasia Information Exchange) Biopharma Collaborative (BPC), Memorial Sloan Kettering Cancer Center (MSK) and Dana-Farber Cancer Institute (DFCI) partnered with Vasta Global, an Omega Healthcare company, to build a model for curating RWD at scale, allowing for cost-effective RWD curation while leveraging expert resources and best practices. The approach includes managing a hybrid onshore and offshore team with clear communication and recurring check-ins, cross-training curators on different EHRs to reallocate Vasta Global resources and maintain project pace, building program-specific clinical data curation training, establishing QA and delivery methods, standardizing data curation, assigning productivity standards, aligning data quality practices and benchmarks, and developing data visualizations and analytics to show progress and insights. Results: In Phase 1 of the AACR GENIE BPC, data was curated from four sites: MSK, DFCI, Vanderbilt Ingram Cancer Center (VICC) in Nashville, and Princess Margaret Cancer Centre, University Health Network (UHN) in Toronto. During Phase 1, MSK and DFCI, with Vasta Global’s assistance, delivered high-quality RWD for 6,475 patients (NSCLC = 1,591; CRC = 1,264; Bladder = 639; Breast = 957; Pancreas 989; and Prostate = 1,035). All data for these cohorts will be publicly available by Spring 2024 in the AACR Project GENIE cBioPortal. MSK and DFCI generated usable and fit-for-purpose RWD to help advance oncology clinical decision-making and data sharing. Conclusion: Accelerating RWD curation will empower the oncology community to leverage data-driven insights and advance treatment opportunities for cancer patients. Working alongside lead sites, a partner like Vasta Global enables researchers to leverage extensive expertise in the BPC data model, allowing for consistency and scalability of the BPC program. In this hybrid model, research partners can curate RWD in a cost-effective way, while remaining focused on core research responsibilities. Citation Format: Julia Rudolph, Kenneth L. Kehl, Melanie Bernstein. Solving the real-world data challenge: A hybrid resourcing model to curate RWD at scale [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6476.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.043 | 0.060 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.004 | 0.005 |
| Science and technology studies | 0.004 | 0.007 |
| Scholarly communication | 0.011 | 0.016 |
| Open science | 0.006 | 0.017 |
| Research integrity | 0.004 | 0.003 |
| Insufficient payload (model declined to judge) | 0.008 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".