Anti-Vascular endothelial growth factor therapy impairs endothelial function of retinal microcirculation in colon cancer patients - an observational study

Abstract

Background: To assess acute effects of bevacizumab (anti-VEGF therapy) on cerebral microvessels and systemic cardiovascular regulation.

Design and subjects: 20 consecutive patients with colorectal cancer (median age: 60.4 years, range 45.5-73.9 years) received bevacizumab intravenously (5 mg/kg) uncoupled of chemotherapy. Prior to and within the first 24 hours after bevacizumab infusion, patients were investigated for retinal endothelial function. A series of a triple 24-hour ambulatory blood pressure measurement was conducted. Retinal endothelial function was determined as flicker light-induced vasodilation. The integrity of baroreflex arc and autonomic cardiovascular control was examined by stimulatory manoeuvres.

Results: Bevacizumab therapy significantly reduced the vasodilatory capacity of retinal arterioles in response to flicker light. A slight decrease in diastolic pressure and heart rate was observed after bevacizumab infusion but this was unrelated to changes in retinal function. The pressure response upon nitroglycerin was largely preserved after bevacizumab infusion. The proportion of patients with abnormal nocturnal blood pressure regulation increased under anti-angiogenic therapy. Autonomic blood pressure control was not affected by bevacizumab treatment.

Conclusions: Bevacizumab acutely impairs microvascular function independent of blood pressure changes. Imaging of the retinal microcirculation seems a valuable tool for monitoring pharmacodynamic effects of bevacizumab.

Trial registration: NCT ID: NCT00740168.

Figures

Figure 1

a-d Functional and structural changes of retinal vessels in colorectal cancer patients in response to bevacizumab infusion. While the diameter of retinal vessels were unaffected by bevacizumab treatment (Figure 1a) it evoked an increase in arteriolar-to-venular ratio (Figure 1b) and a decrease in flicker-induced arterial vasodilation (Figure 1c). The reactivity of retinal arteries to flickering light tended to be lower in patients with a pathological nocturnal blood pressure profile (non-dippers) than in those with physiological nocturnal blood pressure profile (dippers) (Figure 1d). Legend: before (day-1) and after (day + 1) bevacizumab infusion.

Figure 2

a-d 24-h ambulatory blood pressure profiles in colorectal cancer patients in response to bevacizumab infusion. Diastolic blood pressure during daytime but not during nighttime decreased after bevacizumab infusion (Figure 2a,b). The proportion of individuals with pathological blood pressure profile increased after bevacizumab (Figure 2c,d). Legend: systolic blood pressure (broken lines), diastolic blood pressure (smooth lines); proportions are based on the diagnostic criteria of the European Society of Hypertension and the American Hypertension Association before (day-1), during (day 0) and after (day+1) bevacizumab infusion.

Figure 3

a-b Blood pressures and frequency bands of R-R interval during autonomic testing in response to bevacizumab infusion. The blood pressure response to the vasodilator nitroglycerin was largely preserved except for diastolic blood pressure during passive standing (Figure 3a). Cardiovascular perturbation by deep breathing and head-up tilt evoked a physiological shift from high to low frequency oscillations in R-R interval which was unaffected by bevacizumab treatment (Figure 3b). Adequate adjustments of baroreflex sensitivity were only observed in response to postural change but not to deep breathing irrespective of bevacizumab treatment. Legend: BP, blood pressure; BRS, baroreflex sensitivity; DB6, deep breathing at six cycles per minute; HF, high frequency oscillations; LF, low frequency oscillations; Nitro, nitroglycerin; +Nitro, Nitroglycerin application during head-up tilt before (smooth lines) and after (Broken lines) bevacizumab infusion.