Identifying mitigating strategies for endothelial cell dysfunction and hypertension in response to VEGF receptor inhibitors

Abstract

Vascular endothelial growth factor receptor inhibitors (VEGFRis) improve cancer survival but are associated with treatment-limiting hypertension, often attributed to endothelial cell (EC) dysfunction. Using phosphoproteomic profiling of VEGFRi-treated ECs, drugs were screened for mitigators of VEGFRi-induced EC dysfunction and validated in primary aortic ECs, mice, and canine cancer patients. VEGFRi treatment significantly raised systolic blood pressure (SBP) and increased markers of endothelial and renal dysfunction in mice and canine cancer patients. α-Adrenergic-antagonists were identified as drugs that most oppose the VEGFRi proteomic signature. Doxazosin, one such α-antagonist, prevented EC dysfunction in murine, canine, and human aortic ECs. In mice with sorafenib-induced-hypertension, doxazosin mitigated EC dysfunction but not hypertension or glomerular endotheliosis, while lisinopril mitigated hypertension and glomerular endotheliosis without impacting EC function. Hence, reversing EC dysfunction was insufficient to mitigate VEGFRi-induced-hypertension in this mouse model. Canine cancer patients with VEGFRi-induced-hypertension were randomized to doxazosin or lisinopril and both agents significantly decreased SBP. The canine clinical trial supports safety and efficacy of doxazosin and lisinopril as antihypertensives for VEGFRi-induced-hypertension and the potential of trials in canines with spontaneous cancer to accelerate translation. The overall findings demonstrate the utility of phosphoproteomics to identify EC-protective agents to mitigate cardio-oncology side effects.

Keywords: cardio-oncology; endothelial cells; hypertension; proteomics; vascular endothelial growth factor receptor inhibitior.

Figure 1.. VEGFRi treatment increases blood pressure across species and is associated with markers of resistance vessel endothelial dysfunction in mice

Systolic blood pressure (SBP) measurements in: (A) Human cancer patients (n=17) treated with clinically indicated VEGFRis; (B) canine cancer patients (n=31) treated with toceranib; and (C) mice (n=6) treated with sorafenib with SBP measured in mice by radiotelemetry. For each species, the graphs display the SBP before VEGFRi and the maximum SBP recorded while on VEGFRi during 6 days for mice or 6-months for humans and canines. Two-sided paired student t-tests. ***p<0.001, ****p<0.0001. (D-I) Mice were treated for 6 days with vehicle (Veh) or sorafenib (Sor) and markers of endothelial dysfunction were measured in mesenteric arteriolar lysate and in serum. (D, G) Representative immunoblots and quantification (Veh, n=4 pooled samples (8 mice); Sor, n=6 pooled samples (12 mice)) of; (E) phosphorylation of endothelial nitric oxide synthase (eNOS) on serine 1177 (pS1177) relative to total eNOS, one-tailed unpaired student t-test, and (H) endothelin-1 (ET-1) normalized to GAPDH, one-tailed unpaired Welch’s t-test. *p<0.05, **p<0.01. (F) Urine nitrate (Veh n=25; Sor, n=24) and (I) serum ET-1 (Veh, n=17; Sor, n=20) were measured by ELISA. *p<0.05. (A-C, F, I) Dashed lines denote median and dotted lines denote 1^st and 3^rd quartiles, respectively. (E, H) Bars represent mean ± SEM.

Figure 2.. Identification of a phosphoproteomic profile of VEGFRi-treated ECs with validation across species in primary aortic endothelial cells.

(A) Mass spectrometry was performed to quantify 96 sentinel peptides from human umbilical vein endothelial cells treated for 3 hours with tyrosine kinase inhibitors compared to vehicle (DMSO). Marker selection was used to identify a 14 phospho-peptide proteomic signature that differentiates the VEGFRis (n=6) from the non-VEGFRi tyrosine kinase inhibitors (n=7) with p<0.05 and FDR(BH)<0.25. FDR(BH) = False-Discovery Rate with Benjamini-Hochberg correction; SNR = Signal-to-Noise ratio; Primary human (B-D), canine (E-G), and murine (H-J) aortic ECs were treated with vehicle (Veh) or sunitinib (Sun) for 3 hours. Representative immunoblots (B, E, H) and quantification of the impact of sunitinib on serine 222 phosphorylated (pS222)-RBM17 relative to total RBM17; and (C, F, I) serine 1177 phosphorylated (pS1177) nitric oxide synthase (eNOS) relative to total eNOS (n=5 experiments, one-tailed unpaired student t-test except (I) n=3 experiments, one-tailed unpaired Welch’s test. *p<0.05, **p<0.01, ***p<0.001. Veh = Vehicle; Sun = sunitinib. Bars represent mean ± SEM.

Figure 3.. Identification of antihypertensive drugs that produce a phosphoproteomic profile which opposes the impact of VEGFR inhibition in endothelial cells.

(A) A reduced-representation phosphoproteomics assay was performed on human umbilical vein endothelial cells treated with cardiovascular agents (n=25) for 3 hours and the Pearson correlation between each VEGFRi and each cardiovascular agent was generated using the 14-phosphopeptide VEGFRi EC signature. Cardiovascular drugs were rank-ordered from least (blue) to most (red) correlated with the VEGFRi signature. (B-E) HUVECs were pretreated with vehicle (Veh), doxazosin (Dox), or lisinopril (Lis) for 21 hours and then treated with vehicle or the VEGFRi sunitinib (Sun) for 3 hours. Representative immunoblots (B, D) and quantifications of; (C) pS222-RBM17 relative to total RBM17, and (E) pS1177-eNOS relative to total eNOS. n = 4 experiments. Two-way ANOVA with Holm-Sidak post hoc. *p<0.05, **p<0.01, ***p<0.001. ET_AR = Endothelin receptor A; NO = nitric oxide; Veh = Vehicle; Dox = Doxazosin; Lis = Lisinopril; Sun = sunitinib. All error bars represent mean ± SEM.

Figure 4.. Doxazosin prevents VEGFRi-induced endothelial dysfunction in primary endothelial cells across species.

(A, B) Human (n=4), (C, D) canine (n=6), and (E, F) mouse (n=7) primary aortic ECs were pretreated with vehicle or doxazosin for 21 hours and then vehicle (V), sunitinib (Su), sorafenib (So), or toceranib (To) for 3 hours. Representative immunoblots (A, C, E) and corresponding quantifications (C, D, F) for pS1177 eNOS normalized to total eNOS and ET-1 normalized to GAPDH for each species. One-way ANOVA with Holm-Sidak post-hoc comparisons to vehicle. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. Bars represent mean ± SEM.

Figure 5.. Differential impact of doxazosin versus lisinopril on sorafenib-induced resistance arteriole EC dysfunction, hypertension, and renal glomerular endotheliosis in mice.

Mice were treated with vehicle (Veh) or sorafenib (Sor) for 3 days (VEGFRi) and then the sorafenib group was randomized to treatment with Sor, sorafenib + doxazosin (S+D), or sorafenib + lisinopril (S+L) for 3 more days. (A) Representative immunoblots of mesenteric arteriole tissue lysate and corresponding quantifications of; (B) pS1177-eNOS normalized to total eNOS, and (C) endothelin 1 (ET-1) normalized to GAPDH (n=6 for vehicle and sorafenib, n=8 for Sor+Dox and Sor+Lis). (D-F) Systolic blood pressure (SBP) measured by radiotelemetry. (D) The SBP from prior-to-treatment (day 0) and over the 6 days of the study. The arrow indicates the day of randomization to antihypertensive therapy (Anti-HTN Tx). Quantification of; (E) the increase in SBP between Day 0 and Day 3 (Change in SBP while on VEGFRi), and (F) the change from Day 3 to Day 6 (Change in SBP after Anti-HTN while on VEGFRi). (G) Representative glomeruli from H&E-stained kidney sections. (H) Quantification of glomerular endotheliosis (n=13 mice/treatment group, 10-12 glomeruli evaluated/mouse). One-way ANOVA with Holm-Sidak post-hoc. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. Bars represent mean ± SEM.

Figure 6.. A randomized study of doxazosin versus lisinopril for toceranib-induced hypertension in canine cancer patients.

Canine cancer patients receiving toceranib (n=31 recruited) that developed hypertension with systolic blood pressure (SBP) > 170 mmHg (n=10) were randomized to co-treatment with either doxazosin (toc+dox, n = 5) or lisinopril (toc+lis, n = 5). (A) Graph of SBP of 10 canines randomized in the study: SBP before starting cancer therapy (Before VEGFRi), SBP at the time of randomization (On VEGFRi (toceranib)), and the minimum blood pressure during the 6-month period after initiating antihypertensive therapy (VEGFRi + Anti-HTN). p = ns at each time point by unpaired Student’s t-test. Quantification of the change in SBP; (B) from Before VEGFRi to HTN Diagnosis on VEGFRi, and (C) from VEGFRi to VEGFRi + Anti-HTN. (B-C) Paired Student’s t-test. *p<0.05, **p<0.01. (D-F) Urine and serum biomarker measurements of all canines recruited for; (D) urine nitrate (n=27), (E) serum ET-1 (n=25), (F) Urine total protein (n=26) before VEGFRi and the last available measurement during the 6-month treatment interval. Dashed lines denote median and dotted lines denote 1^st and 3^rd quartiles, respectively. One-tailed paired student t-test. (G-I) Urine and serum biomarkers available for 7 of the canines randomized to anti-HTN therapy (n=3-4/group) measured Before VEGFRi, On VEGFRi, and on VEGFRi + anti-HTN. (G) urine nitrate, (H) serum ET-1, and (I) urine total protein. Paired t-tests. *p<0.05. (J) Canine cancer patient survival from initial recruitment to the last known alive date or date of death in the groups that developed hypertension and were randomized to toceranib (n=5, toc+dox) or lisinopril (n=5, toc+lis) relative to those that were not randomized (n=21, toceranib). “∣” lines indicate right censoring and each stepwise drop in the survival curves represents a death. The dotted vertical line represents the median survival time. Log-rank test. All error bars represent mean ± SEM.