Integrating Next Generation Sequencing, EHR, and Claims Data to Extend Follow-Up in a Real-World Advanced Lung Adenocarcinoma Biomarker-Treatment Landscape
Author(s)
Joshuah Kapilivsky, BS1, Emma Roth, BS1, Zachary Rivers, PharmD, PhD1, Adam J. Hockenberry, PhD1, Sandeep Jain, MD2, Jeremy L. Warner, MD, MS2, Amy L. Cummings, MD, PhD3, Gary Grad, MD1, Emilie Scherrer, MSc1, Chithra Sangli, MA1;
1Tempus AI, Inc., Chicago, IL, USA, 2Brown University, Providence, RI, USA, 3UCLA Medical Center, Los Angeles, CA, USA
1Tempus AI, Inc., Chicago, IL, USA, 2Brown University, Providence, RI, USA, 3UCLA Medical Center, Los Angeles, CA, USA
Presentation Documents
OBJECTIVES: Comprehensive genomic profiling (CGP) datasets linked with systemic anticancer treatments are key tools in precision oncology, but clinical follow-up may be limited if derived from a single data modality. When supplementing electronic health records (EHRs), administrative claims data may improve the completeness of treatment journeys. In this study, we extended abstracted treatment histories using closed claims data and used the integrated data to characterize real-world biomarker-treatment relationships.
METHODS: We extracted CGP results and abstracted EHRs from a clinico-genomic database for 6487 stage 4 lung adenocarcinoma patients diagnosed between 2020 and 2023. Closed claims data were joined using deterministic patient linkages. Ongoing abstracted lines of therapy (LOTs) were extended using claims until: i) a treatment gap of at least 90 days, ii) a new class of persistent treatment (at least 2 claims), or iii) end of follow-up. New LOTs from claims began at the next persistent treatment, included all unique persistent medications within 30 days, and were extended similarly. Our study assessed the subset of patients with an integrated LOT1 beginning 0-60 days after CGP testing.
RESULTS: Overall, 1597, 380, and 108 patients had abstracted LOT1, LOT2, or LOT3, respectively. Integrating claims increased patients in each LOT by 11.1%, 20.0%, and 28.7%. Claims extended 8.5%, 8.7%, and 3.4% of LOTs previously lost to follow-up by at least 30 days. Patients with a known LOT end date increased by 24.9%, 40.3%, and 50.0%. Integrated LOTs reflected NCCN guidelines: 93% of LOT1s for EGFR-mutated patients included an EGFR inhibitor. For KRAS p.G12C, where targeted therapies are currently approved in second-line, 86% of LOT1s included immunotherapy and 60% of LOT2s included a KRAS inhibitor.
CONCLUSIONS: This study highlights the use of closed claims to extend EHR-abstracted treatment data, demonstrating utility for real-world treatment patterns and outcome analyses.
METHODS: We extracted CGP results and abstracted EHRs from a clinico-genomic database for 6487 stage 4 lung adenocarcinoma patients diagnosed between 2020 and 2023. Closed claims data were joined using deterministic patient linkages. Ongoing abstracted lines of therapy (LOTs) were extended using claims until: i) a treatment gap of at least 90 days, ii) a new class of persistent treatment (at least 2 claims), or iii) end of follow-up. New LOTs from claims began at the next persistent treatment, included all unique persistent medications within 30 days, and were extended similarly. Our study assessed the subset of patients with an integrated LOT1 beginning 0-60 days after CGP testing.
RESULTS: Overall, 1597, 380, and 108 patients had abstracted LOT1, LOT2, or LOT3, respectively. Integrating claims increased patients in each LOT by 11.1%, 20.0%, and 28.7%. Claims extended 8.5%, 8.7%, and 3.4% of LOTs previously lost to follow-up by at least 30 days. Patients with a known LOT end date increased by 24.9%, 40.3%, and 50.0%. Integrated LOTs reflected NCCN guidelines: 93% of LOT1s for EGFR-mutated patients included an EGFR inhibitor. For KRAS p.G12C, where targeted therapies are currently approved in second-line, 86% of LOT1s included immunotherapy and 60% of LOT2s included a KRAS inhibitor.
CONCLUSIONS: This study highlights the use of closed claims to extend EHR-abstracted treatment data, demonstrating utility for real-world treatment patterns and outcome analyses.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
SA71
Topic
Study Approaches
Disease
SDC: Oncology, STA: Personalized & Precision Medicine