Novel biomarkers predicting vascular complications of anti-VEGF therapies: a pilot study

Background: VEGF pathway inhibitors have transformed cancer treatment, but are commonly associated with vascular toxicity including hypertension (HTN) and proteinuria. These adverse effects can lead to treatment interruptions and discontinuation of effective therapies. Few studies have examined whether biomarkers can predict high-grade hypertension and proteinuria in patients treated with VEGFi/TKIs. Our objective was to use stored plasma samples from patients with metastatic renal cell carcinoma (RCC) subsequently treated with VEGFi/TKIs, to identify biomarkers at baseline that are associated with the development of high-grade proteinuria and high-grade hypertension. Methods: We used SomaScan, an aptamer-based, multiplexed proteomic assay, to measure relative levels of over 7,000 circulating proteins at baseline in >200 patients who were treated with VEGFi/TKIs between 2005 and 2023. Our primary outcomes were high-grade hypertension defined as any instance of grade 3 hypertension or higher based on the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0 (systolic blood pressure ≥160 mm Hg or diastolic blood pressure ≥100 mm Hg) and high-grade proteinuria, defined as ≥ 2+ proteinuria on dipstick, or a calculated urinary protein-to-creatinine raEo ≥ 1.0 g/g, in the first year following drug iniEaEon. We considered as controls for the hypertension cohort paEents who developed at most grade 1 CTCAE new-onset hypertension (systolic BP 120-139 mm Hg or diastolic BP 80- 89 mm Hg) or those who developed grade 2 CTCAE hypertension (systolic BP 140-159 mm Hg or diastolic BP 90-99 mm Hg) but had grade 1 hypertension at baseline. For the proteinuria cohort, we considered as controls patients who did not develop proteinuria or who remained at grade 1 CTCAE proteinuria (1+ proteinuria; urinary protein ≥ULN - <1.0 g/24 hrs) as at baseline. We examined the univariable association between the proteins and each outcome of interest. Protein levels were rescaled by a factor of 100 for analysis. Results: After implementing inclusion and exclusion criteria, 97 patients were included in the hypertension analysis, of whom 46 developed high-grade hypertension. For proteinuria, 109 patients were included and 29 developed high-grade proteinuria. We identified 571 proteins associated with high-grade hypertension (p-value ≤0.05). Among the 46 patients who developed high-grade proteinuria, 149 proteins were associated with the outcome, with a p-value <0.05. The 5 proteins with the lowest p-values for the association with proteinuria were: Enhancer of rudimentary homolog (OR=1.07 [95%CI 1.03-1.10], p-value=0.0004), Protein ABHD14A (OR=1.05 [95%CI 1.02-1.08], p-value=0.001), Grancalcin (OR=4.07 [95%CI 1.69-9.81], p-value=0.002), B melanoma anEgen 2 (OR=1.15 [95%CI 1.05-1.26], p-value=0.002), Histone RNA hairpin-binding protein (OR=1.49 [95%CI 1.15-1.92], p-value=0.002). For hypertension, the 5 most strongly associated proteins were: AngiopoieEn-related protein 3 (OR=0.85 [95%CI 0.76-0.95], p-value=0.004), Spermatogenesis-associated protein 24 (OR=0.68 [95%CI 0.53-0.89], p-value=0.004), DCC-interacting protein 13-alpha (OR=0.32 [95%CI 0.15-0.70], p-value=0.004), Prostate acid phosphatase (OR=0.10 [95%CI 0.02-0.50], p-value=0.005), Glutamate receptor ionotropic, delta-2 (OR=0.93 [95%CI 0.88-0.98], p-value=0.005). No p-values were statistically significant after adjusting for multiple comparisons using the Benjamini-Hochberg (BH) procedure. However, BH might be inadequately stringent in this context, as it adjusts for tesEng multiple independent hypotheses, while the proteins included in our analysis are biologically related and don’t reflect independent hypotheses. Conclusions: The results from our exploratory pilot study suggest that there are protein biomarkers at baseline that could be positively or negatively associated with the development of high-grade hypertension and proteinuria for patients treated with VEGFi/TKIs. Our study is hypothesis-generating and needs to be validated in a larger cohort. Identifying patients who are at high-risk for vascular complications may allow for closer monitoring and implementation of prophylactic measures, improve optimal treatment selection, and ideally limit discontinuation of effective therapies.