ORDER A TEST Log In Contact Us
Back TO events
May 13 — 16, 2025 Montréal Convention Centre, Montréal, QC

Booth #921
Demo our latest AI-enabled technology and end-to-end platform for real-world evidence generation and outcomes research.
6 abstracts

ISPOR 2025

Tempus is a technology company advancing outcomes research and real-world evidence generation with artificial intelligence.

Schedule a meeting with us
Tempus AI & Technology
Wednesday, May 14th - Friday, May 16th

Stop by our booth to discover how Tempus is deploying artificial intelligence to advance HEOR.

See our latest science and experience live demos of Tempus Lens.

Poster Highlights
May 15, 2025
Time
4:00 PM - 7:00 PM EDT

Abstract/Poster
RWD120
First Author
Katie Mo, MS

Comparison of Demographics and Clinical Characteristics using Real World Data from Tempus Multimodal Database and SEER Cancer Registry Across 17 Solid Cancer Cohorts

This study benchmarked the Tempus multimodal database — a real-world data source with clinical and biomarker data from cancer patients — against the SEER cancer registry. Analyzing data from 63,520 patients with solid tumors diagnosed between 2016 and 2021, researchers compared baseline demographic, clinical, and treatment characteristics against the SEER database. The proportion of cancer types of new cancer cases were similar among the two databases. The Tempus database showed a larger representation of patients from the Midwest, whereas the SEER database had a heavy representation from the West. Tempus patients were younger on average but had more advanced cancer staging. Additionally, the Tempus database had more racial diversity based on self-reported race and more complete treatment data. In conclusion, the Tempus and SEER databases show general comparability in demographics and clinical characteristics, but the Tempus database provides greater treatment data granularity and captures more late-stage disease, attributable to sequencing patterns in clinical care.

Time
4:00 PM - 7:00 PM EDT

Abstract/Poster
SA56
First Author
Joshuah Kapilivsky, BS

Assessing the Completeness of Oncology Treatment Data from Administrative Claims: A Benchmarking Study Against Abstracted EHRs Using Patient-Level Linkages

This study benchmarked oncology treatment data from administrative claims against abstracted electronic health records (EHR) for 6,487 stage 4 lung adenocarcinoma patients diagnosed between 2020 and 2023. Claims data (open and closed) were linked using de-identified patient tokens, with EHR data considered the ground truth. Sensitivities and positive predictive values (PPVs) were calculated for 13 infusional and 3 oral medications. Closed claims showed greater sensitivities (50.0-95.3%) than open claims (14.3-54.8%), with infusions having higher sensitivities than orals. PPVs were high for both infusions (closed: 79.1-98.3%; open: 61.5-99.1%) and orals (closed: 84.5-94.2%; open: 91.8-96.8%). Exact matches for abstracted infusion start dates in claims ranged from 45.5-82.5% for closed claims, while 27.6-65.9% of oral start dates matched within 7 days. The team concludes that while EHR remains the gold standard, individual claims may be sufficient for identifying patients receiving specific treatments, and closed claims may be suitable for constructing comprehensive treatment journeys.

Time
4:00 PM - 7:00 PM EDT

Abstract/Poster
RWD113
First Author
Candice Gurbatri, PhD

Impact of NGS Testing Timing on Treatment Patterns and Clinical Outcomes in Colorectal Cancer

Researchers evaluated the impact of next-generation sequencing (NGS) timing on real-world overall survival (rwOS) in 2,293 colorectal cancer (CRC) patients using the Tempus real-world multimodal database. The median age at diagnosis was 58.4 years, with most patients having stage 3 (23%) or stage 4 (68%) disease and receiving first-line (1L) chemotherapy without NGS-informed therapy. Time from biopsy to NGS test order was analyzed, revealing a notable delay in ordering NGS tests post-biopsy. A random forest classifier identified the timeline from biopsy to NGS results receipt as key in determining 1L treatment. Notably, stage 4 patients receiving NGS results within approximately two months of biopsy had a significant survival advantage. The study demonstrates that NGS testing may be associated with increased rwOS in CRC, highlighting the importance of timely NGS for guiding treatment decisions and improving outcomes.

Time
10:30 AM - 1:30 PM EDT

Abstract/Poster
RWD102
First Author
Zach Rivers, PharmD, PhD

Impact of Adverse Event Definitions on Real-World Detection of Immune-Related Adverse Events

Researchers investigated the impact of varying definitions on the identification of immune-related adverse events (irAEs) in real-world data (RWD) from non-small cell lung cancer (NSCLC) patients treated with immune checkpoint blockade (ICB). The research utilized Tempus clinico-genomic data linked to Komodo Health’s claims to analyze irAEs within one year of ICB treatment in patients with stage 3C+ NSCLC. Three peer-reviewed irAE definitions—differing in included irAEs, ICD-10 codes, and pre-treatment washout periods—were applied to the cohort of 4,831 patients. The overall prevalence of irAEs varied significantly across definitions: 41.0% (n=1,981) for Study A (9 irAEs), 75.4% (n=3,849) for Study B (10 irAEs), and 5.4% (n=264) for Study C (3 irAEs). This study demonstrates that irAE identification in RWD varies based on the definitions used, which can affect post-market surveillance, clinical practice guidelines, and patient care. The authors emphasize the need for researchers to accurately communicate the definitions used and conduct sensitivity analyses.

Time
10:15 AM - 11:15 AM EDT

Abstract/Poster
P32
First Author
Natalie Levy, PhD

Oncology Trial Emulation Using Real-World Electronic Health Record Data: Results of the Coalition to Advance Real-World Evidence through Randomized Controlled Trial Emulation (CARE) Initiative

The Coalition to Advance Real-World Evidence through Randomized Controlled Trial (RCT) Emulation (CARE) Initiative seeks to advance understanding of when real-world data (RWD) can generate valid treatment effectiveness estimates by emulating RCTs. This study presents findings from three oncology emulations. The KEYNOTE-189 (metastatic NSCLC) and PALOMA-2 (advanced breast cancer) trials were emulated using electronic health record datasets. Trial entry criteria were applied, and treatment status was based on first-line medications. Inverse probability of treatment weighting controlled for baseline confounding, and Kaplan-Meier and Cox models estimated primary outcomes. In the KEYNOTE-189 emulation, the real-world progression-free survival (rwPFS) hazard ratio (HR) in one dataset was similar to the RCT finding, while the other was closer to the null. PALOMA-2's rwPFS HR was also closer to the null. Real-world overall survival estimates in KEYNOTE-189 also varied across datasets. The researchers conclude that RWD oncology emulation conclusions depend on dataset features, route of administration, and real-world follow-up characteristics.

May 16, 2025
Time
9:00 AM - 11:30 AM EDT

Abstract/Poster
SA71
First Author
Joshuah Kapilivsky, BS

Integrating Next Generation Sequencing, EHR, and Claims Data to Extend Follow-Up in a Real-World Advanced Lung Adenocarcinoma Biomarker-Treatment Landscape

The research team explored the use of closed claims data to enhance electronic health record (EHR)-derived treatment histories for stage 4 lung adenocarcinoma patients with comprehensive genomic profiling (CGP) and a diagnosis between 2020 and 2023. By linking closed claims data to EHRs, researchers extended abstracted lines of therapy (LOTs), defining new LOTs based on treatment gaps, persistent treatments, and follow-up duration. Integrating claims data increased the number of patients in LOT1, LOT2, and LOT3 and extended LOTs previously lost to follow-up. The integrated LOTs reflected NCCN guidelines, with EGFR inhibitors frequently used in EGFR-mutated patients and immunotherapy and KRAS inhibitors used in KRAS p.G12C patients. The study concludes that using closed claims to extend EHR-abstracted treatment data is valuable for real-world treatment pattern and outcome analyses.

Schedule a meeting with us

We'll be in touch shortly.