Impact of hepatic arterial infusion pump (HAIP) therapy on disease control and survival outcome in metastatic colorectal cancer patients.

Abstract

143 Background: Colorectal liver metastasis (LM) presents a therapeutic challenge in tailoring optimal treatment for patients. HAIP delivers high-dose regional chemotherapy to the liver, maximizing local drug exposure while minimizing systemic toxicity. Real-world impact of HAIP on patient tolerance, and survival outcomes is limited. We report HAIP outcomes in a contemporary cohort. Methods: We performed a retrospective review of CRC patients with LM who received HAIP at UAB between (2021-2025). Demographics, tumor characteristics, and liver function tests (LFTs) were collected, alongside treatment details including dose reductions, treatment breaks, and disease response. Chi-square and Fisher-exact tests were used to evaluate differences between categorical variables. Median progression free survival (mPFS) was estimated via Kaplan-Meier estimates. Results: Among 53 patients, (mean age 54 at diagnosis; 60.4% male; 67.9% Caucasian), 75.5% presented with left sided CRC and (67.9%) had bilobar LM ranging between 1 to 10 lesions. Prior HAIP placement, patients were on chemotherapy for median of 9.4 months. Around 71.6% were on FOLFOX. KRAS (45.2%) and BRAF (30.1%) were the most common mutations detected. 64.1% of patients experienced FUDR dose reductions, while (60.3%) had HAIP treatment interruptions due to LFT changes. These modifications showed no association with disease progression (p = 0.340). 52.8% discontinued HAIP due to LFTs fluctuations, disease progression or infection. Worsening LFTs were not linked to radiologic response, recurrence or disease progression (p = 0.542, 0.844, 0.307, respectively). Stable disease at 6 months was observed in 51%, and curative liver resection was achieved in 17%. After >6 months of HAIP, 83% remained alive. Median progression-free survival (mPFS) was 13 months (95% CI 8.8-17.9). Highest risk of progression was between 10–20 months after HAIP treatment initiation. 20% had extra-hepatic progression mostly to the lung or bones, and this was significantly associated with BRAF, KRAS (p < 0.001). Conclusions: In this real-world cohort, HAIP was associated with prolonged PFS, despite dose modifications in many. Most progression occurred within the first 20 months. Appropriate patient selection for HAIP based on disease biology is vital. It can offer meaningful survival benefit in selected patients, warranting further prospective comparative studies.