Department of Gastroenterology and Hepatology, Hokkaido University Hospital, Sapporo, Japan
Satoshi Yuki , Yu Sunakawa , Kentaro Yamazaki , Hiromichi Shirasu , Hiroya Taniguchi , Toshiki Masuishi , Manabu Shiozawa , Hideaki Bando , Tomohiro Nishina , Hisateru Yasui , Takashi Ohta , Naoki Takahashi , Tadamichi Denda , Taito Esaki , Hisato Kawakami , Hironaga Satake , Atsuo Takashima , Yukiko Abe , Shogo Nomura , Takayuki Yoshino
Background: Angiogenesis factors have been reported as prognostic and predictive biomarkers of angiogenesis inhibitors for mCRC (Weickhardt AJ. Br J Cancer 2015. Tabernero J. Ann Oncol 2018). We investigated whether plasma angiogenesis factors could predict the efficacy of biologics combined with chemo in 1L treatment in patients (pts) with RAS wild-type mCRC. Methods: Serial plasma samples were prospectively collected at the time points of pre- and post-treatments in mCRC pts receiving biologics in either 1L or 2nd-line (2L) chemo. From Sep 2017 to Dec 2020, 497 pts were enrolled [1L chemo plus bevacizumab (1L BEV, n=102), 1L chemo plus anti-EGFR antibody (1L aEGFR, n=100), 2L chemo plus bevacizumab (n=100), 2L FOLFIRI plus RAM (n=99), 2L FOLFIRI plus aflibercept (n=85) and other treatment (n=11)]. Total of 17 plasma angiogenesis factors (HGF, PlGF, VEGF-A, VEGF-D, Angiopoietin-2, IFN-γ, IL-6, IL-8, sNeuropilin-1, TSP-2, OPN, sVEGFR1, sVEGFR2, sVEGFR3, sICAM-1, sVCAM-1, and TIMP-1) were analyzed by the multiplex assay with Luminex® technology. Interactions of their pre-treatment measurements with treatment groups on PFS and OS were assessed via Cox proportional hazards model. The strength of interactions was estimated using a propensity score weighting analysis, and the continuous plasma angiogenesis variables were categorized according to the median. The significance level in the interaction was defined as p≤0.1. Results: 133 pts were included in adjusted RAS wild-type 1L cohort (1L BEV: n=33, 1L aEGFR [reference]: n=100). Baseline characteristics of adjusted RAS wild-type 1L cohort were as follows; median age 64 years; male 62.4%; left-sided tumor 88.7%; triplet chemo 15.0%. Propensity-score weighted Cox model for OS showed significant interactions in IL-8 (median 8.03 pg/mL, high: HR 1.738, p=0.0838, Low: HR 0.479, p=0.2624, interaction p=0.0283), sVEGFR-1 (median 1,350 pg/mL, high: HR 0.333, p=0.1770, Low: HR 1.311, p=0.3004, interaction p=0.0777), and sVCAM-1 (median 1,020,000 pg/mL, high: HR 0.100, p=0.0558, Low: HR 1.616, p=0.1765, interaction p=0.0011). In terms of PFS, there were significant interactions in IL-8 (high: HR 1.322, p=0.0418, Low: HR 0.517, p=0.0528, interaction p=0.0752) and sVCAM-1 (high: HR 0.285, p=0.0414, Low: HR 1.200, p=0.7725, interaction p=0.0156). Conclusions: Pre-treatment plasma IL-8 and sVCAM-1 could be predictive biomarkers for efficacy of biologics combined with chemo in 1L treatment of RAS wild-type mCRC. Clinical trial information: UMIN000028616.
Below median | Below median | Above median | Above median | Interaction p-value* | ||
---|---|---|---|---|---|---|
Median | 1L BEV | 1L aEGFR | 1L BEV | 1L aEGFR | Below vs. above median | |
IL-8 | OS | NE | NE | 21.9 m | 29.7 m | 0.0283 |
PFS | 37.1 m | 10.9 m | 8.3 m | 14.1 m | 0.0752 | |
sVCAM-1 | OS | 21.9 m | 29.7 m | NE | NE | 0.0011 |
PFS | 11.5 m | 12.6 m | NE | 14.5 m | 0.0156 |
Note: NE, not estimable: *defined as p≤0.1.
Disclaimer
This material on this page is ©2024 American Society of Clinical Oncology, all rights reserved. Licensing available upon request. For more information, please contact licensing@asco.org
Abstract Disclosures
2023 ASCO Gastrointestinal Cancers Symposium
First Author: Sebastian Stintzing
2023 ASCO Annual Meeting
First Author: Kathrin Heinrich
2024 ASCO Gastrointestinal Cancers Symposium
First Author: Kohei Shitara
2022 ASCO Annual Meeting
First Author: Zhichao Jiang