PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 5, 2026Cancer Research0 citations

Abstract 2970: Limited efficacy of HER3-DXd in colorectal cancer PDX models reveals molecular resistance mechanisms and informs rational combinatorial strategies.

View Full Paper
PKP. KanikarlaASAlexey V. SorokinABAlisha Bent

Key Points

  • To explore the efficacy of HER3-DXd in colorectal cancer and identify resistance mechanisms.
  • Evaluated HER3-DXd in 17 colorectal cancer patient-derived xenograft models.
  • Conducted molecular profiling, including whole-exome sequencing and RNA sequencing.
  • Assessed tumor growth inhibition to measure response.
  • Performed functional studies to examine combinatorial strategies.
  • HER3-DXd showed variable activity, with limited responses in irinotecan-resistant models.
  • HER3 expression was not a reliable predictor of response.
  • Molecular analyses indicated persistent PI3K/AKT and MAPK signaling in non-responders.
  • Combination therapies with MEK, WEE1, or PLK1 inhibitors demonstrated enhanced antitumor effects.

Abstract

Abstract HER3 (ERBB3), a member of the EGFR family, activates PI3K/AKT signaling through dimerization with other ERBB receptors and is frequently implicated in resistance to targeted therapies. Patritumab deruxtecan, a HER3-directed antibody-drug conjugate (ADC), has shown clinical activity in non-small cell lung and breast cancers, and has been recently evaluated in metastatic colorectal cancer (mCRC). Despite frequent HER3 expression in mCRC, its therapeutic relevance in chemotherapy-refractory disease is not well defined. We evaluated HER3-DXd in a panel of 17 molecularly characterized mCRC patient-derived xenograft (PDX) models, the majority of which were derived from patients previously treated with and resistant to irinotecan. Tumor growth inhibition was used to assess response. Molecular profiling (including whole-exome sequencing, RNA sequencing, and Reverse Phase Protein Array) was conducted to elucidate resistance mechanisms and direct rational combination strategies. Single-agent HER3-DXd showed variable activity across models, with limited responses in those exhibiting irinotecan resistance. HER3 expression did not consistently predict response. Interpretation of this association might be limited by the absence of a standardized cutoff for HER3 positivity. The results of the relative weight analysis indicated that 86% of the response was attributable to the payload sensitivity, while only 13% was associated with HER3 expression as determined by IHC. Molecular analyses revealed persistent PI3K/AKT and MAPK signaling in non-responders, along with upregulated receptor tyrosine kinase (RTK) signaling and a correlation with increased expression of drug efflux transporters. Functional studies demonstrated that combining HER3-DXd with MEK, WEE1, or PLK1 inhibitors produced synergistic antitumor effects. Phospho-proteomic data supported adaptive pathway rewiring following HER3 blockade. These findings highlight the complexity of HER3 signaling in chemotherapy-refractory mCRC and support the development of biomarker-driven combination strategies to enhance therapeutic efficacy of HER3-targeted ADCs (and conceivably other ADCs with topoisomerase inhibitor payload) in this setting. Citation Format: Preeti Kanikarla, Alexey Sorokin, Alisha Bent, Kelly Gale, Fengqin Gao, Zhensheng Liu, Kanwal Pratap Singh Raghav, Scott Kopetz. Limited efficacy of HER3-DXd in colorectal cancer PDX models reveals molecular resistance mechanisms and informs rational combinatorial strategies abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 2970.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Kanikarla et al. (2026) studied this question.

synapsesocial.com/papers/69d1fc70a79560c99a0a1fdchttps://doi.org/10.1158/1538-7445.am2026-2970
Ask AI
Helpful
Bookmark
Share
View Full Paper