Main pulmonary artery enlargement during bevacizumab therapy in KRAS-mutant metastatic colon cancer.

purposeBevacizumab, a monoclonal antibody targeting vascular endothelial growth factor (VEGF), is a cornerstone of metastatic colorectal cancer treatment. While systemic hypertension is a recognized complication of VEGF pathway inhibition, the effects of bevacizumab on the pulmonary vasculature remain poorly characterized, and the only longitudinal CT data to date derive from hepatocellular carcinoma patients receiving atezolizumab plus bevacizumab. In this exploratory study, we investigated whether serial thoracic computed tomography (CT) demonstrates progressive enlargement of the main pulmonary artery (mPA) during bevacizumab-based chemotherapy in colon cancer.

methodsIn this retrospective single-center study, 150 patients with KRAS-mutant colon cancer treated with bevacizumab between 2017 and 2025 were screened. Patients with a minimum of 3 months of bevacizumab exposure and at least two follow-up thoracic CT scans were included. The mPA diameter, ascending aorta (AAo) diameter, and mPA/AAo ratio were measured at baseline (T0) and at approximately 3 (T1), 6 (T2), and 9 (T3) months. The primary analysis employed a linear mixed-effects model (LMM) with a patient-level random intercept. Adjusted models, paired secondary analyses, and exploratory overall survival (OS) analyses were also performed. All analyses were conducted in R.

resultsSeventy-four patients met inclusion criteria (median age 62.8 years; 71.6% male). The mPA increased significantly from 23.7 ± 3.3 mm at T0 to 25.9 ± 3.3 mm at T3. In the unadjusted LMM, the mPA increased by + 0.94 mm at T1 (p < 0.001), + 1.45 mm at T2 (p < 0.001), and + 2.22 mm at T3 (p < 0.001) relative to baseline. The linear trend estimate was + 0.72 mm per 3-month interval (95% CI: 0.61-0.83; p < 0.001). Time effects remained significant after covariate adjustment. The mPA/AAo ratio also increased over time (p < 0.001 at all timepoints). No cases of probable pulmonary hypertension (mPA ≥ 33 mm) were identified at any assessment. Exploratory Cox regression showed no significant association between early mPA change and OS (HR 0.92; 95% CI: 0.70-1.19; p = 0.508). Progression-free survival could not be analyzed because the dataset lacked a progression variable. Sensitivity analyses consistently confirmed the robustness of the longitudinal mPA increase.

conclusionSerial thoracic CT imaging demonstrated a statistically significant but small, progressive increase in mPA diameter during bevacizumab-based chemotherapy in this selected cohort. The change remained within the normal range, no patient crossed the threshold for probable pulmonary hypertension, and no association with OS was observed. In the absence of a non-bevacizumab comparator and of hemodynamic correlation, these changes cannot be attributed to bevacizumab; whether they carry clinical significance requires prospective validation with hemodynamic and echocardiographic correlation.