Hepatic arterial infusion pump chemotherapy for colorectal liver metastases: Revisiting traditional techniques to explore new frontiers

Abstract

Hepatic arterial infusion (HAI) chemotherapy, first introduced in the 1980s, has gained recognition as an effective locoregional treatment for colorectal liver metastasis (CRLM). Initially used for unresectable liver metastases, HAI’s application has expanded to the adjuvant setting following hepatic resection, with early studies indicating improved hepatic disease-free survival. Recent research demonstrates that combining HAI with modern systemic therapies enhances conversion to resectability and prolongs both recurrence-free and overall survival, even in heavily pretreated patients with diverse RAS mutational statuses. Personalization through approaches like microsatellite instability status and dose modifications further optimize outcomes. However, the complexity of HAI requires expertise across multidisciplinary teams, limiting its widespread adoption to specialized centers. Ongoing clinical trials continue to investigate HAI’s role in CRLM management, highlighting its potential to become a cornerstone of liver-directed therapy. We explore how HAI chemotherapy, in combination with personalized medicine, can advance treatment strategies for metastatic colorectal cancer.

Key Words: Hepatic arterial infusion; Chemotherapy; Colorectal cancer; Liver metastases; Loco-regional therapy; Personalized medicine

Core Tip: This letter to the editor highlights the evolving role of hepatic arterial infusion (HAI) chemotherapy in the treatment of colorectal cancer with colorectal liver metastasis. We explore its integration with modern systemic therapies and emphasize the importance of patient stratification for optimizing outcomes. The findings suggest HAI’s potential to enhance personalized treatment approaches, underscoring the need for further research to refine protocols and overcome technical challenges in clinical practice.

TO THE EDITOR

Colorectal cancer (CRC) continues to represent a significant global health burden, with around 25% of patients presenting with metastases at the time of initial diagnosis, and over half eventually developing metastases to the liver throughout their disease progression[1,2]. Metastatic CRC confined to the liver is associated with a poor prognosis, with five-year overall survival (OS) rates hovering around 20%[3]. However, in carefully selected patients undergoing complete resection, five-year survival rates can exceed 50%[4-7]. Despite the potential curative effects of liver resection, only 20% of patients with colorectal liver metastasis (CRLM) are initially considered suitable for surgery, and the majority of CRC-related deaths stem from metastatic spread, with liver metastases accounting for two-thirds of these cases[8,9].

For patients with initially unresectable CRLM, regional therapies have gained increasing interest due to the limited efficacy of systemic chemotherapy, which offers a median survival of around 20 months[10]. Standard first-line systemic regimens include a combination of fluoropyrimidines with oxaliplatin or irinotecan, and targeted agents such as anti-epidermal growth factor receptor or anti-vascular endothelial growth factor therapies[1,3]. While these treatments yield response rates in two thirds of patients in first-line settings, effectiveness declines in second-line therapies, with response rates dropping to 40%[11]. Furthermore, progression-free survival (PFS) decreases from 13 months in first-line treatment to a maximum of 7 months with second-line agents[12]. In light of these challenges, hepatic arterial infusion (HAI) chemotherapy has emerged as a viable locoregional approach to manage liver-dominant disease[9,13,14] and increase the possibility of converting initially unresectable cases into candidates for resection[15-18].

HAI chemotherapy leverages the distinct vascular supply of the liver to deliver high concentrations of chemotherapeutic agents directly to metastatic liver lesions, minimizing systemic toxicity (Figure 1). Unlike normal liver parenchyma, which derives its blood supply predominantly from the portal vein, liver metastases receive their blood supply through the hepatic artery. This difference allows HAI to achieve high local drug concentrations in the liver, with agents like floxuridine (FUDR) reaching concentrations up to 400 times higher in tumors when administered via the hepatic artery compared to systemic delivery[19].

Figure 1 Surgical implantation of a hepatic arterial infusion pump. Hepatic arterial infusion pump placement involves surgically inserting a catheter into the gastroduodenal artery, to deliver chemotherapy directly to the liver. The pump is usually implanted subcutaneously in the abdominal wall, and the catheter is secured to minimize dislodgement, ensuring continuous delivery of chemotherapy to liver metastases.

Over the past 30 years, advancements in both surgical techniques and systemic agents have significantly refined HAI therapy. New developments, such as robotic-assisted pump placement, have improved the precision and safety of the procedure, further enhancing its role in managing unresectable CRLM. Although HAI has proven effective as both a first-line treatment for unresectable CRLM and as an adjuvant therapy in specific cases, its widespread adoption remains hindered by factors such as the technical demands of pump implantation, the need for expertise across a multidisciplinary care team, and the continued preference for systemic chemotherapy as the standard first-line treatment[20].

Successfully launching a HAI program requires securing buy-in from both internal teams and external referrers, particularly in the context of positive trial results that have demonstrated significant OS benefits with adjuvant HAI. These results have reignited interest in such programs. To ensure success, it is critical to effectively communicate these benefits to key stakeholders, including surgeons, oncologists, radiologists, and gastroenterologists, while also securing institutional support for acquiring the necessary devices, drugs, and operating room resources. Surgical expertise is central to the success of HAI programs, with essential skills such as navigating variations in the gastroduodenal artery and performing catheter insertions requiring advanced surgical planning. Mentorship from established centers can greatly benefit new programs, and careful patient selection—especially for those with unresectable disease—is crucial. For example, performing pump insertions without concurrent resections can help mitigate complications.

After implantation, medical oncologists play a critical role in managing HAI pump therapy and systemic chemotherapy, ensuring proper dosing and addressing potential complications, such as biliary sclerosis, which affects approximately 5% of patients. Adherence to established dosing protocols is essential to prevent these complications. Building a strong multidisciplinary team-comprising surgeons, pharmacists, nurses, radiologists, and interventional radiologists-is vital to navigate the complexities of HAI. Effective treatment and complication management depend on the coordination and communication among these specialists. Therefore, the success of an HAI program depends on the creation of a skilled, multidisciplinary team, securing institutional and external support, leveraging mentorship from experienced centers, and ensuring close collaboration with meticulous patient selection.

ADJUVANT HAI CHEMOTHERAPY FOR RESECTABLE DISEASE

HAI chemotherapy has been studied extensively as an adjuvant therapy for CRLM to reduce the high rate of hepatic recurrences following resection, which can exceed 50% within two years of surgery[21,22]. Historically, the role of HAI was focused on eradicating micro-metastatic disease within the liver, where systemic chemotherapy alone showed limited efficacy. However, clinical evidence regarding the survival benefit of HAI has been mixed, with early studies yielding conflicting results. One of the initial randomized controlled trials (RCTs) by Lorenz et al[23] in 1998 demonstrated that adjuvant HAI with 5-fluorouracil (5-FU) resulted in worse OS compared to no adjuvant therapy, alongside a significant toxicity burden, including severe gastrointestinal symptoms. A subsequent Cochrane review in 2004 found no consistent evidence of a survival benefit across seven RCTs and, as a result, did not recommend HAI as a standard adjuvant therapy[24].

However, data from several other trials and retrospective studies have suggested that HAI could indeed confer a survival benefit, particularly when combined with modern systemic chemotherapy. A seminal study from Memorial Sloan Kettering Cancer Center (MSKCC) found a significant improvement in both PFS and OS in patients treated with adjuvant FUDR via HAI combined with systemic therapy (OS = 72 months vs OS = 59 months), compared to systemic therapy alone[25]. Similarly, a European Cooperative Group trial in 2002 reported improved median OS (64 months vs 49 months) and higher 4-year hepatic disease-free survival rates (67% vs 43%) in patients receiving HAI with FUDR and systemic therapy[26].

A major retrospective study from MSKCC in 2017 further validated the potential benefits of HAI. This 21-year analysis of 2368 patients who underwent resection of colorectal liver metastases found prolonged 5-year OS in those treated with HAI (53% vs 38%) and a significantly greater 10-year OS (38% vs 24%) compared to patients who did not receive HAI[18]. Notably, subgroup analyses demonstrated a consistent OS benefit from HAI regardless of whether patients received historic or modern systemic chemotherapy, preoperatively or in adjuvant setting. This study reinforced earlier findings, showing that HAI offers additional survival advantages even in the context of contemporary systemic therapies like FOLFOX and FOLFIRI combined with new targeted therapies which were introduced in the early 2000s[25].

Furthermore, genetic factors such as RAS mutation status have also been evaluated in conjunction with HAI therapy. An MSKCC follow-up study revealed that even among patients with KRAS-mutated tumors, HAI with systemic therapy improved 5-year OS rates compared to those who did not receive HAI (59% vs 40%)[17]. This evidence suggests that HAI may mitigate the adverse prognostic effects of certain genetic mutations, offering a more tailored approach to post-surgical treatment.

Despite these encouraging results, skepticism remains about the generalizability of HAI, particularly due to the fact that these results have been predominantly generated in a limited number of high-volume centers with extensive expertise. Current guidelines, such as those from the National Comprehensive Cancer Network (NCCN), remain equivocal regarding the use of HAI in the adjuvant setting. While HAI is considered an option that may improve hepatic disease-free survival, the NCCN advises its use only in experienced institutions, as many studies have not demonstrated a clear survival benefit. Similarly, the European Society for Medical Oncology (ESMO) suggests the use of HAI in clinical trials or as a consolidation treatment in earlier disease lines but does not advocate for its widespread adoption[3].

Ongoing prospective trials may provide further clarity on the role of adjuvant HAI. The PUMP trial, currently underway in the Netherlands, is evaluating the efficacy of adjuvant HAI FUDR in patients at low risk of recurrence, using PFS as the primary endpoint[9]. The PACHA-01 trial in France is another phase II/III study focused on high-risk patients, comparing adjuvant oxaliplatin HAI with systemic FOLFOX chemotherapy[27]. These trials are expected to elucidate whether HAI confers long-term survival benefits when used in conjunction with modern systemic therapies.

While HAI has shown promise in reducing hepatic recurrences and improving survival in select patients with resectable colorectal liver metastases, its use remains largely limited to specialized centers with extensive experience. Further large-scale RCTs are necessary to determine whether HAI can become a standardized component of adjuvant therapy, particularly in the context of modern systemic chemotherapy.

HAI CHEMOTHERAPY FOR UNRESECTABLE DISEASE

The primary goal of HAI therapy and systemic chemotherapy in patients with unresectable CRLM is conversion to resection, as surgical resection is independently linked with improved survival[6,7]. However, resectability remains highly dependent on the expertise of the surgeon and institution, making reported conversion rates variable across studies. Early systemic treatments for CRLM were limited to 5-FU, which had a modest response rate of 20% and average survival of 11 months[28]. Later developments, such as irinotecan and oxaliplatin, improved response rates to 45% with a median survival of up to 20 months[29]. Modern combination regimens, like FOLFOXIRI with bevacizumab, have significantly increased response rates up to 81%, with conversion to resection reported in up to 60% of cases[30,31]. Despite these advances, long-term survival remains challenging due to liver disease progression, fostering continued interest in HAI therapy and other locoregional treatments[32].

Historically, comparisons of HAI alone to systemic chemotherapy for unresectable CRLM showed higher response rates with HAI, but these studies did not demonstrate consistent improvements in OS[33-36]. Later, a multi-institutional RCT (CALGB 9481) prospectively revealed that adding HAI using FUDR to systemic 5-FU monotherapy significantly improved survival compared to systemic therapy only (24 months vs 20 months)[37]. However, up to 70% of patients treated with HAI later developed extrahepatic disease, highlighting the importance of combining HAI with systemic chemotherapy to manage both intrahepatic and extrahepatic metastases[21].

The advent of modern agents like oxaliplatin and irinotecan further expanded the scope of HAI therapy. A single-arm phase I study by Kemeny et al[38] at MSKCC demonstrated a 92% response rate and 47% conversion to resection using HAI combined with systemic oxaliplatin and irinotecan in patients with adverse prognostic features. Similar findings were reported by Goéré et al[39], where 24% of patients treated with HAI combined with systemic 5-FU and leucovorin underwent resection, with a 5-year survival of 56%. Recent studies continue to report high response rates of up to 76% and conversion to resection rates as high as 52% when combining HAI with modern systemic agents[40-43].

Particularly when combined with systemic chemotherapy, HAI therapy offers promising response rates and potential for resection in select patients with unresectable CRLM. However, careful patient selection based on biomarkers such as tumor burden remains critical for optimizing outcomes. As such, a combination therapy involving HAI with fruquintinib and tislelizumab has shown efficacy in microsatellite-stable CRLM cases that failed multiple treatments, achieving significant disease control rates with manageable toxicity[13]. Additionally, circulating tumor cell (CTC) monitoring has proven valuable in assessing post-HAI chemotherapy outcomes, where decreases in CTC levels correlated with improved PFS[14].

Furthermore, dose adjustments in HAI protocols are becoming crucial. For instance, a modified FUDR regimen reduced dose holds and prolonged treatment duration without compromising the potential for surgical conversion, thus offering a feasible alternative for patients who struggle with high-dose regimens[15]. These modifications suggest that HAI can be adapted to patient tolerance levels, potentially increasing its accessibility and effectiveness for broader patient populations.

These findings underscore the growing shift toward personalizing HAI treatments, incorporating factors such as tumor growth patterns and microsatellite instability status to optimize outcomes. This stratified approach not only seeks to maximize efficacy but also aims to minimize adverse effects and enhance surgical conversion rates, laying the groundwork for more individualized treatment paradigms in CRLM management[13-16].

Finally, recent efforts have explored combining HAI with other treatment modalities, such as transarterial chemoembolization (TACE)[44]. Unlike HAI, which relies on continuous infusion to deliver high local drug concentrations without interrupting blood flow, TACE combines chemotherapy with embolic agents that block the tumor’s blood supply, creating ischemia and promoting tumor necrosis. Although HAI and TACE operate through different mechanisms, they may function as complementary rather than competing therapies. For instance, a recent report on the combination of irinotecan-eluting HepaSphere TACE with HAI for unresectable CRLM indicates potential synergistic benefits[44]. However, further research is necessary to establish optimal combinations and criteria for patient selection.

CONCLUSION

Over the past three decades, the role of HAI chemotherapy has evolved significantly. Initially introduced in the 1980s for patients with unresectable liver metastases, HAI has since expanded to include adjuvant therapy following hepatic resection, reflecting a growing recognition of its potential benefits in CRC management. This evolution highlights the shift from HAI as a niche treatment to a more integrated approach for patients with liver-dominant metastatic disease. Our study contributes to this progress by emphasizing the clinical value of HAI, especially when combined with systemic therapies, and by addressing patient selection criteria critical to optimizing outcomes in CRLM. Future research should focus on refining HAI protocols in combination with emerging systemic therapies, as well as overcoming technical challenges that currently limit broader adoption of this promising treatment. Through these efforts, HAI may continue to solidify its role as a valuable component of care for selected patients with metastatic CRC.