Endostatin lowers blood pressure via nitric oxide and prevents hypertension associated with VEGF inhibition

Abstract

Antiangiogenesis therapy has become a vital part of the armamentarium against cancer. Hypertension is a dose-limiting toxicity for VEGF inhibitors. Thus, there is a pressing need to address the associated adverse events so these agents can be better used. The hypertension may be mediated by reduced NO bioavailability resulting from VEGF inhibition. We proposed that the hypertension may be prevented by coadministration with endostatin (ES), an endogenous angiogenesis inhibitor with antitumor effects shown to increase endothelial NO production in vitro. We determined that Fc-conjugated ES promoted NO production in endothelial and smooth muscle cells. ES also lowered blood pressure in normotensive mice and prevented hypertension induced by anti-VEGF antibodies. This effect was associated with higher circulating nitrate levels and was absent in eNOS-knockout mice, implicating a NO-mediated mechanism. Retrospective study of patients treated with ES in a clinical trial revealed a small but significant reduction in blood pressure, suggesting that the findings may translate to the clinic. Coadministration of ES with VEGF inhibitors may offer a unique strategy to prevent drug-related hypertension and enhance antiangiogenic tumor suppression.

Fig. 1.

Murine Fc-ES resulted in a dose-dependent reduction in systolic BP in C57BL/6 mice. The low-dose group received 4 mg/kg, 2 d/wk (on days 0, 3, 7, 10, 14, and 17) and had the smallest decrease in BP. The high-dose group received 4 mg/kg, 5 d/wk (on days 0–4, 7–11, and 14–18) and had the most stable BP reduction. The concentrated-dose group received the cumulative high weekly dose of 20 mg/kg administered in a single concentrated bolus once a week (on days 0, 7, and 14) and had the greatest declines in BP on the days following treatment; BP levels recovered thereafter. *P < 0.05.

Fig. 2.

Murine Fc-ES administration (4 mg/kg, 5 d/wk) is associated with (A) higher circulating nitrate levels (a measure of NO) and (B) lower systolic BP in C57BL/6 mice as compared with saline-treated controls. (C) Circulating NO levels were higher in ES-overexpressing mice and were lower in collagen XVIII/ES-knockout mice than in wild-type control mice. These findings correlated NO levels with endogenous and exogenous ES levels.

Fig. 3.

ES (4 mg/kg, 5d/wk) caused a significant decrease in systolic BP throughout the treatment period in wild-type C57BL/6 mice (A) but only on day 5 in eNOS-knockout mice (B).

Fig. 4.

ES prevented the marked BP elevations that accompany VEGF inhibition. Normotensive mice treated with anti-murine VEGF antibody MAb G6-31 (5 mg/kg on days 1, 4, 8, 11, and 15) developed significant spikes in BP on the day following treatment. Simultaneous ES treatment (4 mg/kg, 5 d/wk) prevented these BP spikes. Control mice received IgG (5 mg/kg on days 1, 4, 8, 11, and 15) or IgG + ES (4 mg/kg, 5 d/wk). A gradual increase in baseline BP occurred in both control groups and may be related to IgG.

Fig. 5.

ES decreased BP in patients. BP was recorded from 42 patients treated with rhES as part of a phase I clinical trial for neuroendocrine cancer. Measurements from the first three cycles show a small but significant decrease in both systolic BP (A) and diastolic BP (B). Predicted values are indicated by solid lines; 95% confidence intervals for individual patients are indicated by dashed lines.