ASCO GI 2024

Presenters Emily Teslow (Tempus), Amit Mahipal (UH Seidman)

The research team sought to leverage the Tempus multimodal database in order to characterize BRAF alterations in CRC compared to other gastrointestinal (GI) cancers (non-CRC) among 51,560 patients. BRAF alterations were observed in 8.9% of patients in the CRC cohort, and in 2.2% of the non-CRC cohort. BRAF V600E was the most common alteration type in all GI tumors, but was found in a higher percentage of CRC vs non-CRC cancers. Other BRAF variants such as amplifications and certain fusions were more frequent in non-CRC GI tumors.

Presenters Rachel Berg (Tempus), Binyam Yilma (Tempus), Melissa Stoppler (Tempus), William P. Harris (Fred Hutchinson Cancer Center, University of Washington), Gentry King (U Washington)

The research team utilized the Tempus multimodal database to explore relationships between ARID1A, PBRM1 and BAP1 alterations with immunotherapy biomarkers and genomic co-alterations in 696 biliary tract cancer patients (BTC). Co-alteration profiles were found to be distinct between ARID1A-altered vs. PBRM1-and/or BAP1-altered BTC groups, with macrophages being observed as the predominant type of immune cell present in the tumor microenvironment (TME) among all groups.

Presenters Kristiyana Kaneva (Tempus), Christine Lo (Tempus), Daniel Neems (Tempus), Jonathan Freaney (Tempus) Hala Boulos (Tempus), Seung won Hyun (Tempus), Chithra Sangli (Tempus), Rick Blidner (Tempus), Kate Sasser (Tempus), Hall Nimeiri (Tempus), Takayuki Yoshino (National Cancer Center Hospital Japan), Yoshiaki Nakamura (National Cancer Center Hospital East (Japan)

Tempus sought to address the unmet need for a tissue-naive approach to detect minimal residual disease (MRD) in early stage colorectal cancer (CRC) after curative surgical resection in order to delineate high risk patient populations who are likely to relapse. Plasma samples from GALAXY, an observational arm of the CIRCULATE-Japan study, were analyzed with Tempus xM, a novel tumor-naive MRD assay. 53% (20/38) of recurrent (R) patients had detectable xM MRD + and 94% (30/32) of nonrecurrent (NR) patients had no detectable (xM MRD-), demonstrating the strong clinical performance of xM. Additionally among the xM MRD+ status, a 74% chance of accurately classifying a true relapse. Adjusted median DFS at the LMT for MRD+ patients is 39.3wks (9.8 mos) vs. >72 wks (18 mos) for MRD- patients, a clinically meaningful difference (Adj. HR 5.09).

Presenters Ymke van der Pol (Tempus), Zach Rivers (Tempus), Justin Guinney (Tempus)

Previous research had indicated that patients with KRAS G13 or G12 alterations had improved (G13) or worse (G12) outcomes when treated with trifluridine-tipiracil (TAS), but it was unknown if these patients would see similar benefits with TAS plus bevacizumab (Bev), which is now more commonly used. Leveraging both the Tempus multimodal database and institutional data, the team looked to compare outcomes for metastatic colorectal cancer (mCRC) patients with codon-specific KRAS mutations and RAS/RAF WT status that received TAS + Bev. In this cohort of mCRC treated with TAS-Bev, codon-specific KRAS mutations did not appear to predict survival benefit, unlike previously reported for TAS treated patients.