Optimization of bevacizumab therapy by predicting which patients are at higher risk of developing hypertension and proteinuria

Like Comment
Read the paper

Increased angiogenesis is a mechanism by how tumor cells grow and spread in distant sites, and cancer patients with aberrant angiogenesis usually develop advanced and metastatic diseases. Anti-angiogenic drugs targeting the vascular endothelial growth factor (VEGF) signaling pathway, the key signaling pathway involved in angiogenesis, are an effective strategy to treat advanced and metastatic cancer. Bevacizumab, a VEGF-pathway inhibitor, is the first anti-angiogenic drug approved by the FDA in 2004, and it has been used widely in combination with other therapies to treat several types of cancer. Despite the clinical use of bevacizumab for many years and its demonstrated efficacy, patients frequently experience hypertension and proteinuria (a sign of kidney toxicity) during treatment, limiting the efficacy of the cancer therapy and posing a threat to patient health and quality of life.

My main scientific interest is the optimization of cancer therapy, which can be achieved by different approaches, ranging from precision medicine approaches to developing strategies for novel anti-cancer agents. I am currently working as Medical Director at AbbVie in Oncology Early Developments, but during my tenure at the UNC Eshelman School of Pharmacy, I was leading efforts towards the discovery of genomic determinants of toxicity of VEGF-pathway inhibitors. In that effort, my primary goal was to identify new genetic variants that could be assessed in patient samples before treatment with these drugs to identify those patients at higher risk of developing toxicities. Dr. Quintanilha joined my lab in 2019, and since then, we have combined efforts toward this goal. In fact, we both are coinventors of a patent for methods of identifying the risk of bevacizumab-induced proteinuria and hypertension, resulting from one of our most exciting findings that have been recently described in BJC1. In this BJC publication1, we describe the results of the largest genome-wide association study (GWAS) of bevacizumab-induced hypertension and proteinuria. To perform this analysis, we have used data from 1,039 patients of European descent with advanced or metastatic pancreatic, prostate, or breast cancers treated with bevacizumab from four clinical trials from the CALGB/Alliance (CALGB 80303, 40503, 90401, and 40502). In addition, in collaboration with researchers from Indiana University, we have validated our top hit for hypertension in a fifth randomized phase III clinical trial (ECOG-ACRIN E5103) of 582 patients of European descent with breast cancer treated with bevacizumab from the ECOG-ACRIN group.

Interestingly, the genes related to the most statistically significant genetic variants selected based on pre-specified criteria have consistent biology with the phenotypes herein investigated. For example, our top hit discovered in the GWAS analysis for bevacizumab-induced hypertension and validated in the additional cohort was rs6770663, an intronic genetic variant in a gene that encodes a subunit of the K+ channel, KCNAB1, the deletion of which is associated with elevated blood pressure2. From a mechanistic perspective, the serum response factor (SRF), a cardiac-enriched transcription factor, is more likely to bind the major allele A than the minor allele G of rs67706633, possibly activating the transcription of KCNAB14. Our proposed mechanism is that in patients with the G allele of rs6770663, lower activation of KCNAB1, impaired activation of the K+ voltage-activated channel, and increased vasoconstriction results in an increased risk of bevacizumab-induced hypertension. We have illustrated this putative model in the figure below.

Proposed mechanism of rs6770663 in KCNAB1 in an endothelial cell for bevacizumab-induced hypertension. (Created with BioRender.com.)

For proteinuria, the top hit was the variant rs339947 located close to TRIO, which codes to a Rho guanine nucleotide exchange factor. Bevacizumab-induced proteinuria is characterized by glomerular damage, and TRIO is highly expressed in the specialized glomerular cells, podocytes. TRIO coded protein contributes to podocyte injury by inducing Rac15. Genetic variants in moderate linkage disequilibrium with rs339947 have been associated with TRIO expression in the glomerulus6. Thus, we postulate that patients with the minor allele A of rs339947 might have an increased TRIO expression in the glomerulus compared to patients with the major allele C. Higher TRIO expression leads to an increased podocyte injury that would result in an increased risk of bevacizumab-induced proteinuria.  We have illustrated this putative model in the figure below.

Proposed mechanism of rs339947 close to TRIO in a podocyte for bevacizumab-induced proteinuria. (Created with BioRender.com.)

                The genetic variants identified in our study can be used as predictor biomarkers of bevacizumab-induced toxicities and might impact a significant proportion of patients treated with bevacizumab. For example, variant rs6770663 in KCNAB1 associated with bevacizumab-induced hypertension was present in about 10% of patients included in our study, in accordance with the frequency reported for subjects of European Ancestry7. The highest prevalence of rs6770663 is about 70% in subjects of African Ancestry7. Variant rs339947 close to TRIO associated with bevacizumab-induced proteinuria was present in about 13% of patients included in our study, also in accordance with the frequency reported for subjects of European Ancestry, and its highest prevalence is 15% in subjects of Latin American Ancestry7. More efforts are needed to further evaluate these biomarkers in patients of different ancestries than Europeans.

The use of anti-angiogenic drugs for cancer treatment will continue to increase in the following years, as these drugs have been recently added to immunotherapy-based regimens in the treatment of metastatic tumors. We hope our results help guide decisions on the risk assessment of patients treated with bevacizumab and other anti-angiogenic drugs. Moreover, we hope that the mechanisms proposed based on our analysis might give insights to a clearer understanding of the underlying pathophysiology of these toxicities.



  1. Quintanilha JCF, Wang J, Sibley AB, Jiang C, Etheridge AS, Shen F, et al. Bevacizumab-induced hypertension and proteinuria: a genome-wide study of more than 1000 patients. Br J Cancer (2021). https://doi.org/10.1038/s41416-021-01557-w.
  2. Tur J, Chapalamadugu KC, Padawer T, Badole SL, Kilfoil PJ, Bhatnagar A, et al. Deletion of Kvβ1.1 subunit leads to electrical and haemodynamic changes causing cardiac hypertrophy in female murine hearts. Exp Physiol. 2016;101:494–508.
  3. Kulakovskiy IV, Vorontsov IE, Yevshin IS, Sharipov RN, Fedorova AD, Rumynskiy EI, et al. HOCOMOCO: Towards a complete collection of transcription factor binding models for human and mouse via large-scale ChIP-Seq analysis. Nucleic Acids Res. 2018;46:D252–D259.
  4. Rouillard AD, Gundersen GW, Fernandez NF, Wang Z, Monteiro CD, McDermott MG, et al. The harmonizome: a collection of processed datasets gathered to serve and mine knowledge about genes and proteins. Database (Oxford). 2016;2016:baw100.
  5. Maier M, Baldwin C, Aoudjit L, Takano T. The role of trio, a rho guanine nucleotide exchange factor, in glomerular podocytes. Int J Mol Sci. 2018;19:E479.
  6. Gillies CE, Putler R, Menon R, Otto E, Yasutake K, Nair V, et al. An eQTL landscape of kidney tissue in human nephrotic syndrome. Am J Hum Genet. 2018;103:232–444.
  7. Sherry ST, Ward MH, Kholodov M, Baker J, Phan L, Smigielski EM, et al. dbSNP: the NCBI database of genetic variation. Nucleic Acids Res. 2001;29:308–311.



Federico Innocenti, M.D., Ph.D.

Medical Director, Oncology Early Development, AbbVie