Impact of VEGF inhibition on the development of erythrocytosis in hereditary hemorrhagic telangiectasia Free

Introduction

In hereditary hemorrhagic telangiectasia (HHT), erythrocytosis is typically attributed to chronic hypoxemia from right-to-left shunting through pulmonary arteriovenous malformations (PAVMs). Embolization of PAVMs is expected to correct hypoxemia and resolve secondary erythrocytosis. However, in clinical practice, erythrocytosis has been observed in some HHT patients treated with bevacizumab, including those without PAVMs or in those who had already undergone embolization. Bevacizumab inhibits vascular endothelial growth factor (VEGF), thereby reducing the abnormal angiogenesis that drives the formation and persistence of telangiectasias and arteriovenous malformations responsible for chronic bleeding in HHT (Al-Samkari, 2022). The underlying mechanism of erythrocytosis in this context remains unclear. Some studies suggest that VEGF inhibitors can stimulate hepatic erythropoietin (EPO) production via a hypoxia-independent pathway (Tam BY et al., 2006). To date, there are no published experiences describing the effects of systemic bevacizumab use and erythrocytosis in HHT patients.

Methods

We conducted a retrospective chart review to identify patients diagnosed with HHT based on the Curacao criteria and/or genetic testing at an academic medical center, from January 2016 through January 2025. Baseline and peak hemoglobin levels from the start of bevacizumab treatment through January 2025 were collected. Additional data included gender, race, ethnicity, age at diagnosis, and known genetic mutations. Records were reviewed for EPO levels and JAK2 mutation testing. Additional variables included smoking status, PAVM history, prior PAVM embolization, obstructive sleep apnea (OSA), thrombosis, and evidence of right-to-left shunting on bubble echocardiogram. Concurrent use of medications linked to erythrocytosis, including steroids, Sodium-glucose cotransporter-2 (SGLT2) inhibitors, and erythropoiesis-stimulating agents, was also noted.

Results

Among 42 HHT patients, 34 received intravenous (IV) bevacizumab, with the earliest treatment starting in 2016, up until January 2025. Of the 34 patients treated with IV bevacizumab, 10 (29.4%) developed erythrocytosis, defined as peak Hemoglobin level exceeding 16.5 g/dL in males and 16 g/dL in females, measured from treatment initiation up to January 2025. Of these 10 patients, 7 (70%) were male and 3 (30%) were female. One (10%) was a former smoker, one (10%) a current smoker, and eight (80%) never smoked. Six (60%) had a history of PAVMs and evidence of right-to-left shunting on bubble echocardiography; four (40%) had a prior embolization. Five (50%) had a history of OSA. Genetic testing showed ACVRL1 mutations in 2 patients (20%) and ENG mutation in 1 patient (10%).

Venous thrombosis was observed in 20.0% (2/10) of patients with erythrocytosis and 17.6% (6/34) of all IV bevacizumab-treated patients. No arterial thrombotic events are reported in the cohort. Among all 42 patients with HHT, 4 underwent JAK2 mutation testing. Three of these patients had erythrocytosis and all tested negative. One patient, not on IV bevacizumab, had a positive JAK2 V617F mutation with a very low allele frequency of 0.18. EPO levels were obtained for 9 of the 42 HHT patients studied. Of these, 6 (66.7%) had levels above the reference range (2.6-18.5 mIU/mL).

Conclusion

Erythrocytosis was observed in nearly one-third of HHT patients treated with IV bevacizumab, including those without PAVMs and those who had undergone embolization. The absence of other known causes, along with negative JAK2 mutation testing in many cases, suggests alternative mechanisms. These may include microscopic arteriovenous shunting not detectable on imaging but sufficient to cause hypoxia-induced erythropoiesis, or a potential VEGF-inhibition mediated upregulation of hepatic EPO production. Further evaluation is underway to assess the relationship between the duration of bevacizumab exposure and the development of erythrocytosis to better understand how this process occurs and if it persists. Overall, these findings highlight the importance of monitoring hemoglobin levels during treatment and support a more comprehensive evaluation of patients who develop erythrocytosis which may have been masked by prior frequent bleeding. Detailed evaluations of erythrocytosis should include JAK2 mutation testing, EPO measurement, and consideration of other causes of secondary erythrocytosis.