Germline predictors for bevacizumab induced hypertensive crisis in ECOG-ACRIN 5103 and BEATRICE

Fei Shen, Guanglong Jiang, Santosh Philips, Erica Cantor, Laura Gardner, Gloria Xue, Geneva Cunningham, Nawal Kassem, Anne O’neill, David Cameron, Thomas m. Suter, Kathy d. Miller, George w. Sledge, Bryan p. Schneider

Research output: Contribution to journalArticlepeer-review

Abstract / Description of output


Bevacizumab is a beneficial therapy in several advanced cancer types. Predictive biomarkers to better understand which patients are destined to benefit or experience toxicity are needed. Associations between bevacizumab induced hypertension and survival have been reported but with conflicting conclusions.


We performed post-hoc analyses to evaluate the association in 3124 patients from two phase III adjuvant breast cancer trials, E5103 and BEATRICE. Differences in invasive disease-free survival (IDFS) and overall survival (OS) between patients with hypertension and those without were compared. Hypertension was defined as systolic blood pressure (SBP) ≥ 160 mmHg (n = 346) and SBP ≥ 180 mmHg (hypertensive crisis) (n = 69). Genomic analyses were performed to evaluate germline genetic predictors for the hypertensive crisis.


Hypertensive crisis was significantly associated with superior IDFS (p = 0.015) and OS (p = 0.042), but only IDFS (p = 0.029; HR = 0.28) remained significant after correction for prognostic factors. SBP ≥ 160 mmHg was not associated with either IDFS or OS. A common single-nucleotide polymorphism, rs6486785, was significantly associated with hypertensive crisis (p = 8.4 × 10−9; OR = 5.2).


Bevacizumab-induced hypertensive crisis is associated with superior outcomes and rs6486785 predicted an increased risk of this key toxicity.
Original languageEnglish
JournalBritish Journal of Cancer
Publication statusPublished - 12 Feb 2024


Dive into the research topics of 'Germline predictors for bevacizumab induced hypertensive crisis in ECOG-ACRIN 5103 and BEATRICE'. Together they form a unique fingerprint.

Cite this