Abstract
Background
Oxaliplatin-based hepatic arterial infusions (HAI) combined with intravenous therapy is a therapeutic option for colorectal cancer with liver-only metastasis, notably in the palliative setting, either initially or after failure of systemic chemotherapy. Our study aimed to assess efficacy and tolerance between oxaliplatin-naive patients and oxaliplatin-pretreated patients.
Methods
Between 2008 and 2022, single-center consecutive patients presenting with liver metastasis secondary to colorectal cancer who received at least one cycle of HAI-oxaliplatin combined with systemic therapy were included.
Results
The oxaliplatin-naive arm included 63 patients (median age 58 years) and the pretreated arm included 244 patients (median age 53 years). Patient characteristics were well balanced between the groups. All patients in the oxaliplatin-naive arm received HAI-oxaliplatin while 13% of the pretreated patients received HAI-FOLFIRINOX. After a median follow-up of 36 months, median progression-free survival was 14 months in the oxaliplatin-naive group (range 11.8-24 months) and 10.1 months in the pretreated group (range 9.4-12.5 months) (P = 0.016). The objective response rate was 66.7% and the disease control rate was 79.4% in the oxaliplatin-naive group, versus 32.4% and 77.5% (P < 0.001) in the pretreated group. Grade 3-4 toxicities were comparable between the two groups, including neuropathy. Secondary resection/ablation rate was 22.2% in oxaliplatin-naive patients and 17.6% in pretreated patients.
Conclusion
Oxaliplatin use as an intra-arterial hepatic infusion is feasible and efficient after previous systemic oxaliplatin; it showed significant response rates without increased toxicities. It can provide alternative treatments and spare late-setting drugs such as regorafenib and tipiracil–trifluridine for a further palliative intent treatment.
Key words: hepatic arterial infusion, oxaliplatin, colorectal cancer, liver metastases
Highlights
-
•
Intra-arterial hepatic infusion with oxaliplatin is feasible and efficient after previous systemic oxaliplatin.
-
•
It showed significant response rates without increased toxicities.
-
•
It offers alternative options, preserving and late-setting drugs such as regorafenib and tipiracil–trifluridin for later use.
Introduction
Worldwide, colorectal cancer (CRC) is the third leading cause of cancer death.1 The liver remains the most common site for metastatic spread of CRC; ∼40%-60% of patients presenting with CRC will develop liver metastasis, either at the time of diagnosis, or subsequently during the disease.2 Of those, only 25% are considered initially resectable at presentation.3 Resection or ablation of liver metastasis is the only potentially curative treatment of patients with liver-only metastasis, with a current 10-year survival rate of 24%.4, 5, 6, 7
Hepatic arterial infusion chemotherapy (HAIC) is an effective way to deliver drugs that are usually administered intravenously (i.v.), directly into the arterial hepatic circulation through the hepatic artery, resulting in a selective routing of the drug to the tumor, while sparing the normal hepatocyte (mainly perfused by the portal vein).8 Subsequently, this allows a considerable increase of drug concentration locally, resulting in improved efficacy and tolerance, with fewer systemic side-effects.9,10 Historically, the first HAI drug used in colorectal cancer was floxuridine [5-fluoro-2’-deoxyuridine (FUDR)],11 with an objective response rate (ORR) of 42.9% in CRC.12 However, the administration of HAI-FUDR requires a surgically implanted pump and a reservoir to be filled every 2 weeks, and is associated with a significant biliary toxicity, hampering its administration and concurrent use of targeted therapies.13 Later on, the use of oxaliplatin combined with systemic or other HAI chemotherapy prompted an ORR of 64% in patients previously treated with one or more lines of systemic chemotherapy.14, 15, 16 Oxaliplatin demonstrated enhanced efficacy and tolerability, along with a more convenient administration schedule.
HAIC is currently being used in three different settings. The first indication is in the adjuvant setting after complete liver metastasis resection, supported by the results of the phase II PACHA trial that were recently presented.17 HAIC can be used for patients with potentially resectable liver metastases. In studies, it achieved a 24% conversion rate to resection for patients with extensive or initially unresectable metastases,18 47% in another study, and up to 57% for chemotherapy-naive patients.19,20 These results significantly outperform standard systemic chemotherapy (12.5%)3 and systemic chemotherapy combined with targeted therapies (25%-50%).21,22 HAIC is also indicated for nonresectable disease as a palliative treatment,23 aimed solely at increasing the survival of these patients, particularly those with a high intrahepatic tumor burden (liver-dominant) or progressive disease confined to the liver. Indeed, the antitumor activity of HAIC is maintained even in heavily treated patients refractory to multiple lines of systemic standard chemotherapy,24 achieving an 18% conversion rate to resectability in those heavily pretreated patients.25
In summary, metastatic patients receiving this type of regimen are either (i) oxaliplatin-naive and received it in the first-line setting mainly, or (ii) pretreated and received it in the palliative setting or as a rescue treatment for intensification after failure of induction therapy.
In this study, we aimed to prove that oxaliplatin could retain its efficacy as an intra-arterial perfusion in patients who were previously treated with i.v. oxaliplatin, as an ulterior therapeutic line, without increasing the toxicity due to this drug.
Materials and methods
Patients
All patients who received at least one cycle of HAIC with oxaliplatin for colorectal cancer with liver metastasis in Gustave Roussy between 2008 and 2022 were retrospectively eligible. Oxaliplatin was part of the HAIC regimen, either alone or combined with other molecules such as 5FU and/or irinotecan i.v. or HAI in the OPTILIV regimen)25 with or without targeted therapy [anti-epidermal growth factor receptor (anti-EGFR) or anti-angiogenic antibody (anti-VEGFR)]. HAIC was chosen by a multidisciplinary committee, in the neoadjuvant setting (as an initial regimen or as an intensification treatment after an insufficient response to initial systemic chemotherapy) or the palliative setting for unresectable disease. Patients who received HAIC in the adjuvant setting were excluded from the study. Patients who received non-oxaliplatin-based HAIC were also excluded.
Resectable and limited extrahepatic disease was not a contraindication for HAIC, and local treatments for extrahepatic metastasis could be carried out, if applicable, before treatment initiation or subsequently after. Previous hepatic or extrahepatic local treatment was allowed. The primary tumor was either previously resected, planned to be resected, or kept in place if the disease was unresectable. All clinical, radiological, biological and pathological data were collected from medical files and managed using REDCap electronic data capture tools hosted at Gustave Roussy medical center. This retrospective study complies with the French MR004 methodology regarding general data protection regulation for noninterventional retrospective health research (Délibération n° 2018-155 of 3 May 2018) and was approved by our institutional review board (IRB N°2022-144), in compliance with the Helsinki declaration. Written informed consent requirement was waived.
Treatment scheme and modalities
The hepatic intra-arterial catheter was implanted and linked to an implantable port.26 A digital subtracted angiography during injection of contrast medium through the HAI catheter port was systematically carried out before treatment initiation, and then at regular intervals (at least every two courses of HAIC), to verify the complete perfusion of the liver and the absence of any extrahepatic perfusion or catheter dysfunction.
The chemotherapy protocol consisted of either HAI-oxaliplatin 100 mg/m2 infused over 2 h on day 1, followed by systemic chemotherapy (modified LV5FU2 or FOLFIRI), or HAI chemotherapy according to the OPTILIV protocol, which consisted of the triplet oxaliplatin 85 mg/m2, irinotecan 180 mg/m2 and 5FU 2800 mg/m2, all administered HAI.15 Treatment was repeated every 2 weeks. If targeted therapies were added to the regimen, the following dosages were used: cetuximab 500 mg/m2, or panitumumab 6 mg/kg, or bevacizumab 5 mg/kg. Choice of systemic and targeted therapies was made on the basis of RAS/BRAF mutational status and previous chemotherapy lines. Upon the incidence of neurotoxicity or other HAI-specific adverse events, dose reduction of HAIC was implemented. Oxaliplatin-HAI infusion duration was systematically prolonged to 6 h to avoid pain during administration or mild extrahepatic perfusion. Treatment was continued until disease progression, toxicity, catheter dysfunction, or tumor response allowing curative treatment of the liver metastasis (resection or ablation).
Treatment evaluation
All systemic toxicities related to chemotherapy or HAIC were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) v4.0. The toxicities or complications related specifically to the HAI catheter were graded as follows: grade 0 = absent; grade 1 = mild (no intervention indicated); grade 2 = moderate (intervention needed to maintain HAIC); grade 3 = severe (failure of continuing HAIC in spite of intervention or definitive contraindication); grade 4 = life-threatening; and grade 5 = death. Similarly, the systemic toxicities specifically related to HAIC were graded as follows: grade 0 = absent; grade 1 = mild (nonrelevant and no need to delay HAIC); grade 2 = moderate (reversible upon HAIC delay, no indication to stop treatment); grade 3 = severe (nonreversible upon HAIC delay, and/or definitive contraindication to HAIC); grade 4 = life-threatening; and grade 5 = death. Safety was defined as all HAIC or HAI-catheter-related adverse events, as well as some related to systemic chemotherapy.
Tumor response was evaluated using computed tomography scan or magnetic resonance imaging according to Evaluation Criteria in Solid Tumors (RECIST1.1). As per protocol, radiological evaluation for response assessment was routinely conducted every 2 months for the totality of treatment duration.
Statistical analysis
Qualitative variables were compared with a chi-square test or a Fisher’s exact test, while Wilcoxon signed rank test was used for quantitative variables. Progression-free survival (PFS) was defined from the first day of HAIC until systemic disease progression (hepatic and/or extrahepatic progression), or death for any reason. PFS was determined using the Kaplan–Meier method and compared between the two groups using a log-rank test. To ensure the robustness of our events, a sensitivity analysis was restricted to patients who did not receive the treatment in the adjuvant setting. A Cox proportional hazards model was used for univariate and multivariate analyses of potential prognostic factors. A P value <0.05 was considered to be statistically significant. All statistical calculations were carried out using R Core Team (version 4.3.0; Warwick University, UK).
Results
Patient characteristics
From 2008 to 2022, 417 patients received oxaliplatin-based HAIC in our center. After exclusion of 110 patients who received HAIC in the adjuvant setting, 307 patients were included in our study (Figure 1). The patients were subsequently divided into two groups: group 1 with oxaliplatin-naive patients (who received the molecule for the first time as HAIC) (N = 63), and group 2 with oxaliplatin-pretreated patients (who had received oxaliplatin i.v. during previous lines) (N = 244).
Figure 1.
Study flow-chart.
Patient characteristics were well balanced between the two groups and are described in Table 1. The most common location of the primary tumor was the left colon (41.3% and 47.5%), followed by the right colon (23.8% and 25.8%), and the rectum (30.2% and 24.2%) in the naive and pretreated groups, respectively. Overall, 90.5% and 82.8% presented with synchronous liver metastasis, and 31.7% and 14% had confirmed vascular contact, respectively. Only 3.2% and 13.5% had initially resectable liver disease, and 90.4% and 84.4% had either unresectable or potentially resectable disease, respectively. Most patients presented with a high hepatic tumor burden, defined as eight or more hepatic lesions (44.4% and 38.5%), while 90.4% and 78.3%, respectively, presented with a bilobar distribution. The presence of a KRAS mutation was frequently observed in both the naive (28.6%) and pretreated (40.6%) groups. Furthermore, at the time of HAIC initiation, 28% of the patients in the naive group and 36% in the pretreated group had simultaneous extrahepatic metastasis.
Table 1.
Patient characteristics
| Characteristic | Naive, n (%) N = 63 |
Pre-exposed, n (%) N = 244 |
P value |
|---|---|---|---|
| Gender | 0.0884 | ||
| Male | 31 (49) | 149 (61) | |
| Female | 32 (51) | 95 (39) | |
| Mean age at diagnosis, years | 0.0055a | ||
| 58.4 (10.2) | 53.6 (11.6) | ||
| Performance status (ECOG) at HAI | 0.7445 | ||
| 0 | 29 (46) | 98 (40) | |
| 1 | 31 (49) | 124 (51) | |
| 2 | 3 (5) | 18 (7) | |
| 3 | 0 (0) | 4 (2) | |
| Stage at diagnosis | 0.0609 | ||
| Nonmetastatic | 5 (7.9) | 44 (18) | |
| Metastatic | 55 (87.3) | 196 (80.3) | |
| Not specified | 3 (4.8) | 4 (1.7) | |
| Synchronous or asynchronous liver metastases | 0.0744 | ||
| Synchronous | 57 (90.5) | 202 (82.8) | |
| Asynchronous | 5 (7.9) | 42 (17.2) | |
| Not specified | 1 (1.6) | 0 (0) | |
| Largest diameter of liver metastasis (mm) | 0.0027a | ||
| 63.2 (34) | 48.2 (34.6) | ||
| Number of liver metastases | 0.7255 | ||
| 1-4 | 12 (19.1) | 58 (23.8) | |
| 4-8 | 9 (14.3) | 44 (18) | |
| 8-12 | 5 (7.9) | 15 (6.1) | |
| >12 | 23 (36.5) | 79 (32.4) | |
| Not specified | 14 (22.2) | 48 (19.7) | |
| Distribution | 0.0252a | ||
| Bilobar | 57 (90.4) | 191 (78.3) | |
| Unilobar | 3 (4.8) | 37 (15.2) | |
| Not reported | 3 (4.8) | 16 (6.5) | |
| Vascular contact | 0.0001a | ||
| None | 32 (50.8) | 146 (59.8) | |
| Yes | 20 (31.7) | 34 (14) | |
| Not specified | 11 (17.5) | 64 (26.2) | |
| Resectability | 0.013a | ||
| Resectable | 2 (3.2) | 33 (13.5) | |
| Potentially resectable | 20 (31.7) | 101 (41.4) | |
| Unresectable | 37 (58.7) | 105 (43) | |
| Not clear | 4 (6.4) | 5 (2.1) | |
| Primary tumor location | 0.549 | ||
| Right colon | 15 (23.8) | 63 (25.8) | |
| Left colon | 26 (41.3) | 116 (47.5) | |
| Rectum | 19 (30.2) | 59 (24.2) | |
| Unknown | 3 (4.7) | 6 (2.5) | |
| Molecular biomarkers | |||
| KRAS mutation | 18 (28.6) | 99 (40.6) | 0.3778 |
| NRAS mutation | 2 (3.2) | 5 (2.1) | 0.6362 |
| BRAF mutation | 1 (1.6) | 9 (3.7) | 0.6933 |
| MSI-H status | 1 (1.6) | 5 (2.1) | 0.0311 |
ECOG, Eastern Cooperative Oncology Group; HAI, hepatic arterial infusions.
Significant P-value <0.05.
Treatment characteristics
The mean number of HAIC cycles administered was 6.8 (range 1-31; 8.4 cycles for the oxaliplatin-naive arm and 6.4 cycles for the pretreated arm; P = 0.0002), and the mean number of previously administered therapeutic lines was 1.9 (0.6 for group 1 and 2.53 for group 2; P < 0.0001). In pretreated patients, the mean number of oxaliplatin cycles previously administered was 11.9, and the mean previously received dose of oxalipatin was 1011.5 mg. All 63 patients who were oxaliplatin-naive received oxaliplatin-based HAIC (100%) combined LV5FU or FOLFIRI IV, and in the pretreated group, 87% received HAI-oxaliplatin and 13% received the OPTILIV regimen. Dose reduction was necessary in 71.4% of the oxaliplatin-naive patients, and 58.2% of the pretreated patients; only 64.4% and 72.5% of those dose reductions, respectively, however, included HAIC dose reductions, either alone or combined with systemic chemotherapy dose reduction, the rest being dose reductions related to the systemic chemotherapy (Table 2).
Table 2.
Treatment characteristics and description
| Treatment characteristics | Naive, n (%) N = 63 |
Pre-exposed, n (%) N = 244 |
P value |
|---|---|---|---|
| HAI regimen | 0.0024a | ||
| Oxaliplatin | 63 (100) | 212 (86.9) | |
| OPTILIV | 0 (0) | 32 (13.1) | |
| I.v. chemotherapy | 0.0047a | ||
| Irinotecan only | 0 (0) | 1 (0.4) | |
| LV5FU only | 43 (68) | 132 (54.1) | |
| Both (FOLFIRI) | 18 (29) | 66 (27.1) | |
| Other | 2 (3) | 13 (5.3) | |
| None | 0 (0) | 32 (13.1) | |
| Targeted therapy | 0.2747 | ||
| Anti-EGFR | 23 (37) | 71 (29.1) | |
| Anti-VEGFR | 19 (30) | 99 (40.6) | |
| None | 21 (33) | 74 (30.3) | |
| Dose reduction | 0.0412a | ||
| Yes | 45 (71.4) | 142 (58.2) | |
| None | 18 (28.6) | 102 (41.8) | |
| Catheter placement | 0.0044a | ||
| Surgical | 18 (28.6) | 29 (11.9) | |
| Radiological | 43 (68.2) | 209 (85.7) | |
| Combined | 2 (3.2) | 6 (2.4) | |
| Catheter revision | 0.0317a | ||
| Yes | 24 (38.1) | 127 (52) | |
| None | 39 (61.9) | 117 (48) |
EGFR, epidermal growth factor receptor; HAI, hepatic arterial infusion; VEGFR, vascular endothelial growth factor receptor.
Significant P-value <0.05.
Toxicity
Overall, the toxicity profile was comparable between the two groups. Regarding non-HAIC-specific side-effects, the most frequent grade 3/4 toxicity was neutropenia, which occurred in 51% (n = 32) of the patients in the oxaliplatin-naive group, and 31%(n = 77) in the pretreated group, followed by a decreased platelet count in 10% (n = 6) and 13% (n = 32), respectively (Table 3).
Table 3.
Adverse events, HAI-related events and HAI complications
| Adverse events | Naive, n (%) N = 63 |
Pre-exposed, n (%) N = 244 |
||||
|---|---|---|---|---|---|---|
| General adverse events | None | Grade 1/2 | Grade 3/4 | None | Grade 1/2 | Grade 3/4 |
| Nausea/vomiting | 29 (46) | 34 (54) | 0 (0) | 98 (40) | 142 (58) | 4 (2) |
| Diarrhea | 30 (48) | 31 (49) | 2 (3) | 122 (50) | 113 (46) | 9 (4) |
| Oral mucositis | 32 (51) | 28 (44) | 3 (5) | 148 (61) | 84 (34) | 12 (5) |
| Neutropenia | 19 (30) | 12 (19) | 32 (51) | 94 (39) | 73 (30) | 77 (31) |
| Platelet count decreased | 29 (46) | 28 (44) | 6 (10) | 128 (53) | 84 (34) | 32 (13) |
| Anemia | 28 (44) | 34 (54) | 1 (2) | 110 (45) | 128 (52) | 6 (3) |
| Hepatic arterial infusion (HAI)-related events | ||||||
| Pain during HAI | 35 (55) | 25 (40) | 3 (5) | 119 (49) | 111 (45) | 14 (6) |
| Portal hypertension | 58 (92) | 0 (0) | 5 (8) | 226 (93) | 5 (2) | 13 (5) |
| Hepatic failure | 61 (97) | 2 (3) | 0 (0) | 241 (99) | 1 (<1) | 2 (<1) |
| Duodenal ulcer | 59 (94) | 2 (3) | 2 (3) | 224 (92) | 15 (6) | 5 (2) |
| Peripheral sensory neuropathy | 4 (6) | 45 (72) | 14 (22) | 23 (9) | 183 (75) | 38 (16) |
| HAI-related complications | ||||||
| Vascular complicationsa | 38 (60) | 19 (30) | 6 (10) | 118 (48) | 97 (40) | 29 (12) |
| Hepatic/biliary complicationsb | 58 (92) | 5 (8) | 0 (0) | 222 (91) | 19 (8) | 3 (1) |
| Catheter complicationsc | 55 (87) | 7 (11) | 1 (2) | 192 (79) | 39 (16) | 13 (5) |
| Local complicationsd | 59 (94) | 4 (6) | 0 (0) | 206 (84) | 34 (14) | 4 (2) |
Vascular complications included extrahepatic perfusion, incomplete hepatic perfusion, artery dissection, vascular stenosis, thrombosis, arteritis and other.
Hepatic/biliary complications included chemical hepatitis, cholangitis, biliary stenosis, ischemic cholangitis, nodular regenerative hyperplasia and other.
Catheter complications included occlusion, dislodgement or other.
Local complications included hematoma, superficial surgical infection, catheter exposure, drug extravasation, delayed tissue healing and other.
Regarding HAIC-related side-effects, the incidence of pain during HAI was not different between the two groups. In the oxaliplatin-naive group, 25 patients (40%) reported pain of grade 1/2 and only three patients (5%) reported a grade 3/4 pain; while in the pretreated group, 111 patients (45%) and only 14 patients (6%) reported grade 1/2 and grade 3/4 pain, respectively. Regarding the incidence or aggravation of any peripheral sensory neuropathy at the end of the HAIC course, no difference was observed between the two groups. Grade 1/2 neuropathy occurred in 72% (n = 45) of patients in the control group versus 75% (n = 183) in the pretreated group, while grade 3/4 neuropathy was reported in 22% (n = 14) and 16% (n = 38), respectively.
There was, however, a slight increase, though not statistically significant, in the incidence of grade 1/2 duodenal ulcer in the pretreated group (6% versus 3%), and there was a slight increase in the incidence of grade 3/4 portal hypertension in the oxaliplatin-naive group (8% versus 5%) (Table 3).
When looking at the HAI-related complications, there was a trend toward an increase of vascular complications in the pre-exposed group versus the control group (52% versus 40%) (mainly grade 1/2 events). No grade 3/4 hepatic/biliary complications or local complications were observed in the oxaliplatin-naive group, and only 1% and 2% were observed in the pretreated group, respectively (Table 3).
Two patients in the pre-exposed group experienced grade 3/4 hepatic failure, one of which resulted in patient death. One death occurred in each of the groups due to severe portal hypertension secondary to HAIC, and one more death occurred in the pretreated group due to biliary complications.
Of note, no specific signal of toxicity was observed when HAI was combined with bevacizumab (Supplementary Table S1, available at https://doi.org/10.1016/j.esmogo.2025.100173).
Response rate and progression-free survival
Among the 244 pre-exposed patients, 2.9% (n = 7) achieved complete response, 29.5% (n = 72) had a partial response, 45.1% (n = 110) had a stable disease, and 17.2% (n = 42) had progressive disease, resulting in an ORR of 32.4% and a disease control rate (DCR) of 77.5%. On the other hand, among the 63 oxaliplatin-naive patients, a complete response was observed in 1.6% (n = 1) of patients and a partial response in 65.1% (n = 41), while 12.7% (n = 8) had stable disease and 7.9% (n = 5) had progressive disease, thus resulting in an ORR of 66.7% and a DCR of 79.4%. Response was not reported in 21 patients, either for loss to follow-up or death soon after HAIC initiation (Table 4).
Table 4.
Best response comparison between the two populations
| Characteristic | Overall, n (%) N = 307a |
Naive, n (%) N = 63 |
Pre-exposed, n (%) N = 244 |
P valuea |
| Best response (RECIST) | <0.0001b | |||
| Complete | 8 (2.6) | 1 (1.6) | 7 (2.9) | |
| Partial | 113 (36.8) | 41 (65.1) | 72 (29.5) | |
| Stable | 118 (38.4) | 8 (12.7) | 110 (45.1) | |
| Progressive | 47 (15.3) | 5 (7.9) | 42 (17.2) | |
| Not reported | 21 (6.9) | 8 (12.7) | 13 (5.3) |
RECIST, Response evaluation criteria in solid tumors.
Fisher's exact test.
bSignificant P-value <0.05.
Among the initial 307 patients, 270 patients were assessable for survival analysis. Median PFS for the overall population (n = 270) was 11 months [95% confidence interval (CI) 9.76-13.6 months], with 47% of the patients alive without progression at 12 months of follow-up, 15% at 36 months, and 11% at 60 months (Figure 2A).
Figure 2.
(A) Progression-free survival for the overall population. (B) Progression-free survival between the oxaliplatin-naive population and the pretreated population.
When compared between the two groups, median PFS was 14 months in the oxaliplatin-naive group (95% CI 11.8-24 months) versus 10.1 months in the pretreated group (95% CI 9.4-12.5 months, P = 0.016). The percentages of patients who were alive and free of any disease were 60% and 44% at 12 months of follow-up, 32% and 21% at 24 months, and 25% and 12% at 36 months, respectively (Figure 2).
Patient outcome
At the time of data collection, death had occurred in 255 patients (83.1%), 45 patients (71.4%) in the control group, and 210 patients (86.1%) in the pretreated group (P = 0.0396). When death occurred, the causes of death in the oxaliplatin-naive group and in the pretreated group were as follows: disease progression in 64% and 67%, complications due to treatment in 8% and 7.1%, cause unknown in 24% and 22%, and intercurrent cause in 4% and 1.1%, respectively. It would be important to specify that the treatment-related deaths described in this paragraph represent the deaths that occurred while receiving the last line of treatment, mostly chemotherapy, and not the treatment-related deaths specific to HAIC line.
Regarding an end result of the HAIC treatment line, the main motive for HAIC discontinuation was progressive disease in 19% of oxaliplatin-naive patients and 30.7% of pretreated patients. Other common reasons for treatment discontinuation were: switch to best supportive care (mostly due to clinical progression) (7.9% versus 11.9%), initiation of a new line due to toxicity (4.8% versus 10.3%), initiation of new line due to catheter dysfunction (9.5% versus 9%), surveillance (6.4% versus 7.8%), switch to another therapy for maintenance (17.5% versus 4.9%), stopping therapy due to catheter dysfunction (0% versus 2.9%), or implementation of a noncurative local treatment (1.6% versus 1.2%).
A total of 18.5% of patients in the overall population were able to undergo curative intent surgery or ablation of their metastatic liver disease after response to HAIC; the oxaliplatin-naive patients had a secondary resection/ablation rate of 22.2%, while pretreated patients had a secondary resection rate of 17.6%. Among patients who successfully underwent secondary tumor resection, the tumor regression grade (TRG) showed 93.3% and 73.4% good response rates, respectively (Supplementary Table S2, available at https://doi.org/10.1016/j.esmogo.2025.100173).
Discussion
In this study, oxaliplatin infusion through an HAIC route in patients previously treated with i.v. oxaliplatin appeared to be feasible, and the molecule is shown to retain efficacy, all while maintaining an acceptable safety profile.
When compared between the two groups, median PFS was 14 months in the oxaliplatin-naive group and 10.1 months in the pretreated group. The first exposure to the drug in the HAI setting resulted in more complete or partial responses. The molecule did not, however, completely lose its efficacy in patients who had previously received oxaliplatin in the adjuvant/metastatic setting; one-third of the patients still showed either a complete or partial response (32.4%), and stable disease was still achieved in almost one-half of the patients (45.1%). We can conclude that oxaliplatin may be considered as a therapeutic option through arterial infusion, even after systemic oxaliplatin failure in the case of liver-only metastases.
The issue of sensibility to HAI-oxaliplatin after previous i.v. treatment with oxaliplatin was previously raised in two studies. The first was a retrospective study that explored the efficacy of HAI-oxaliplatin and systemic LV5FU2 without targeted therapy. Median PFS was 7 months and a 18% conversion rate to resection was observed in a heavily pretreated population of 44 patients, of whom 77% had already received an oxaliplatin-based regimen.25 Another more recent retrospective study evaluating HAI-oxaliplatin and systemic chemotherapy in addition to targeted therapy showed similar results. The median PFS was 9 months, the ORR was 45% and the DCR was 94%, in a population including 89 heavily pretreated patients, with strictly unresectable liver metastasis, of whom 78% had previously received oxaliplatin in the adjuvant or metastatic setting.27 However, no direct comparison between the two groups was conducted.
Our findings are important because ORR is correlated to the conversion rate to resection and subsequently an improvement in long-term survival in patients who initially present with unresectable liver metastasis secondary to colorectal cancer.27 Of note, standard second-line systemic chemotherapy resulted in a modest 7% conversion to resection.28 Increasing ORR using a combination of HAIC and systemic chemotherapy should always be a therapeutic option if available. In this study, HAIC achieved secondary curative intent resection or ablation in 17.6% of pretreated patients, a population that is considered heavily pretreated and received a median of 2.5 previous chemotherapy lines, and for whom the metastatic liver disease was considered unresectable (43%) or potentially resectable (41.4%) before HAIC initiation.
Several drugs are currently available for administration as an intra-arterial infusion. HAI-FUDR is the most commonly used in the United States; when combined with i.v. oxaliplatin and/or irinotecan, it resulted in an ORR between 33% and 44% in pretreated patients, with an non-negligible incidence of hepatic toxicity.24,29,30 In our study, the toxicity profile was comparable between the two groups. The most frequent grade 3/4 systemic toxicities observed were neutropenia (51% in the oxaliplatin-naive group and 31% in the pretreated group), and a decreased platelet count (10% and 13%, respectively). However, hematological toxicity was not limiting and was easily managed with dose reduction or increasing the interdose interval. Regarding HAIC-related side-effects, the incidence of pain during the intra-arterial infusion was identical between the two groups. Interestingly, no increase in the incidence of grade 3/4 neuropathy was reported, as it occurred in 22% (n = 14) and 16% (n = 38), respectively. This could be due to pre-exposed patients receiving smaller starting doses of oxaliplatin, especially in the case of persistent pre-existing neuropathy. Furthermore, the increase in vigilance for screening this limiting toxicity in routine follow-up resulted in rapid dose adaptation if needed. On the other hand, this study confirms the lack of biliary and hepatic toxicity of HAI-oxaliplatin: none of the oxaliplatin-naive patients experienced grade 3/4 biliary or hepatic complications, while they occurred in only three patients in the pretreated group. This underlines the safety of oxaliplatin when used as an intra-arterial infusion in comparison with other more toxic molecules like HAI-FUDR, even if the patients had received the drug previously. Finally, as was already discussed by Boilève et al.,31 it should be mentioned that HAI-oxaliplatin can also be safely combined with bevacizumab as no particular additional toxicity was observed in this study.
Among the limitations of this study is its retrospective monocentric setting, but its robustness lies in the large number of patients included. A second limitation is that a direct comparison of the efficacy of oxaliplatin in these two populations cannot be made; although the two populations have mostly similar characteristics, some characteristics remain heterogenous, as is the case of the extent for extrahepatic disease, the presence of vascular contact, the initial surgeon evaluation of resectability, concurrent HAIC and systemic chemotherapy/target therapy. Thirdly, due to its retrospective design, some data were incomplete. Nevertheless, this is the first study to directly compare the efficacy and safety of oxaliplatin between two populations based on whether the patients previously received the molecule or not. This issue had been previously raised and discussed in previous publications, but this is one study that can objectively confirm that the efficacy of oxaliplatin is maintained, at no cost of increased adverse events. However, randomized trials are still needed to determine the optimal setting of HAIC. Several studies are ongoing concerning this subject. Among them is the OSCAR-PRODIGE49 phase III study, evaluating HAI-oxaliplatin versus i.v. oxaliplatin, combined with LV5FU and targeted therapy according to RAS status in the first-line setting.32
In this study, we aimed to expand upon the prior findings by providing a more nuanced understanding of how patient selection and prior oxaliplatin exposure influence outcomes with HAIC therapy. These findings could help refine selection criteria for HAIC, potentially guiding clinicians in identifying patients most likely to benefit from this approach. Additionally, understanding how prior oxaliplatin exposure affects resectability outcomes can provide valuable insights into treatment sequencing and surgical decision making. Finally, our findings highlight key factors that could be integrated into future trial design, including patient stratification based on prior oxaliplatin response and the potential role of combination therapies.
Conclusion
HAI-oxaliplatin combined with systemic chemotherapy and/or targeted therapies is active and safe in patients presenting with liver metastasis secondary to CRC even after failure of previous lines containing i.v. oxaliplatin. The DCR was comparable with patients who were not previously exposed to the molecule. No increase in high-grade peripheral neuropathy was observed.
Acknowledgments
Funding
None declared.
Disclosure
The authors have declared no conflicts of interest.
Supplementary data
References
- 1.Rawla P., Sunkara T., Barsouk A. Epidemiology of colorectal cancer: incidence, mortality, survival, and risk factors. Przegląd Gastroenterol. 2019;14(2):89–103. doi: 10.5114/pg.2018.81072. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Siegel R.L., Miller K.D., Jemal A. Cancer statistics, 2016. CA Cancer J Clin. 2016;66(1):7–30. doi: 10.3322/caac.21332. [DOI] [PubMed] [Google Scholar]
- 3.Adam R., Delvart V., Pascal G., et al. Rescue surgery for unresectable colorectal liver metastases downstaged by chemotherapy: a model to predict long-term survival. Ann Surg. 2004;240(4):644–657. doi: 10.1097/01.sla.0000141198.92114.f6. discussion 657-658. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Creasy J.M., Sadot E., Koerkamp B.G., et al. Actual 10-year survival following hepatic resection of colorectal liver metastases: what factors preclude cure? Surgery. 2018;163(6):1238–1244. doi: 10.1016/j.surg.2018.01.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Dhir M., Sasson A.R. Surgical management of liver metastases from colorectal cancer. J Oncol Pract. 2016;12(1):33–39. doi: 10.1200/JOP.2015.009407. [DOI] [PubMed] [Google Scholar]
- 6.House M.G., Ito H., Gönen M., et al. Survival after hepatic resection for metastatic colorectal cancer: trends in outcomes for 1,600 patients during two decades at a single institution. J Am Coll Surg. 2010;210(5):744–752. doi: 10.1016/j.jamcollsurg.2009.12.040. 752-755. [DOI] [PubMed] [Google Scholar]
- 7.Cummings L.C., Payes J.D., Cooper G.S. Survival after hepatic resection in metastatic colorectal cancer: a population-based study. Cancer. 2007;109(4):718–726. doi: 10.1002/cncr.22448. [DOI] [PubMed] [Google Scholar]
- 8.Breedis C., Young G. The blood supply of neoplasms in the liver. Am J Pathol. 1954;30(5):969–985. [PMC free article] [PubMed] [Google Scholar]
- 9.Fallik D., Ychou M., Jacob J., et al. Hepatic arterial infusion using pirarubicin combined with systemic chemotherapy: a phase II study in patients with nonresectable liver metastases from colorectal cancer. Ann Oncol. 2003;14(6):856–863. doi: 10.1093/annonc/mdg247. [DOI] [PubMed] [Google Scholar]
- 10.Collins J.M. Pharmacologic rationale for regional drug delivery. J Clin Oncol. 1984;2(5):498–504. doi: 10.1200/JCO.1984.2.5.498. [DOI] [PubMed] [Google Scholar]
- 11.Laface C., Laforgia M., Molinari P., et al. Hepatic arterial infusion of chemotherapy for advanced hepatobiliary cancers: state of the art. Cancers. 2021;13(12):3091. doi: 10.3390/cancers13123091. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mocellin S., Pasquali S., Nitti D. Fluoropyrimidine-HAI (hepatic arterial infusion) versus systemic chemotherapy (SCT) for unresectable liver metastases from colorectal cancer. Cochrane Database Syst Rev. 2009;(3) doi: 10.1002/14651858.CD007823.pub2. [DOI] [PubMed] [Google Scholar]
- 13.Kemeny N., Seiter K., Niedzwiecki D., et al. A randomized trial of intrahepatic infusion of fluorodeoxyuridine with dexamethasone versus fluorodeoxyuridine alone in the treatment of metastatic colorectal cancer. Cancer. 1992;69(2):327–334. doi: 10.1002/1097-0142(19920115)69:2<327::aid-cncr2820690209>3.0.co;2-u. [DOI] [PubMed] [Google Scholar]
- 14.Ducreux M., Ychou M., Laplanche A., et al. Hepatic arterial oxaliplatin infusion plus intravenous chemotherapy in colorectal cancer with inoperable hepatic metastases: a trial of the gastrointestinal group of the Federation Nationale des Centres de Lutte Contre le Cancer. J Clin Oncol. 2005;23(22):4881–4887. doi: 10.1200/JCO.2005.05.120. [DOI] [PubMed] [Google Scholar]
- 15.Lévi F.A., Boige V., Hebbar M., et al. Conversion to resection of liver metastases from colorectal cancer with hepatic artery infusion of combined chemotherapy and systemic cetuximab in multicenter trial OPTILIV. Ann Oncol. 2016;27(2):267–274. doi: 10.1093/annonc/mdv548. [DOI] [PubMed] [Google Scholar]
- 16.Pernot S. Chimiothérapie intra-artérielle hépatique: un apport importantpour le traitement des métastases hépatiques des cancers colorectaux - Actualités aux 26es FFCD. La Lettre de l’Hépato-gastroentérologue. 2015;(5) https://www.edimark.fr/revues/la-lettre-de-lhepato-gastroenterologue/n-5-octobre-2015-copy/chimiotherapie-intra-arterielle-hepatique-un-apport-important-pour-le-traitement-des-metastases-hepatiques-des-cancers-colorectaux-actualites-aux-26es-ffcd Available at. Accessed August 22, 2023. [Google Scholar]
- 17.Gelli M., Ewald J., Tanguy M.L., et al. Postoperative hepatic arterial chemotherapy after resection of colorectal liver metastases in patients at high risk of hepatic recurrence: a multicenter randomized phase II trial (PRODIGE 43 - PACHA-01) J Clin Oncol. 2023;41(suppl 16) 3515-3515. [Google Scholar]
- 18.Goéré D., Deshaies I., de Baere T., et al. Prolonged survival of initially unresectable hepatic colorectal cancer patients treated with hepatic arterial infusion of oxaliplatin followed by radical surgery of metastases. Ann Surg. 2010;251(4):686–691. doi: 10.1097/SLA.0b013e3181d35983. [DOI] [PubMed] [Google Scholar]
- 19.Kemeny N.E., Melendez F.D.H., Capanu M., et al. Conversion to resectability using hepatic artery infusion plus systemic chemotherapy for the treatment of unresectable liver metastases from colorectal carcinoma. J Clin Oncol. 2009;27(21):3465–3471. doi: 10.1200/JCO.2008.20.1301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.DʼAngelica M.I., Correa-Gallego C., Paty P.B., et al. Phase II trial of hepatic artery infusional and systemic chemotherapy for patients with unresectable hepatic metastases from colorectal cancer: conversion to resection and long-term outcomes. Ann Surg. 2015;261(2):353–360. doi: 10.1097/SLA.0000000000000614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ye L.C., Liu T.S., Ren L., et al. Randomized controlled trial of cetuximab plus chemotherapy for patients with KRAS wild-type unresectable colorectal liver-limited metastases. J Clin Oncol. 2013;31(16):1931–1938. doi: 10.1200/JCO.2012.44.8308. [DOI] [PubMed] [Google Scholar]
- 22.Gruenberger T., Bridgewater J., Chau I., et al. Bevacizumab plus mFOLFOX-6 or FOLFOXIRI in patients with initially unresectable liver metastases from colorectal cancer: the OLIVIA multinational randomised phase II trial. Ann Oncol. 2015;26(4):702–708. doi: 10.1093/annonc/mdu580. [DOI] [PubMed] [Google Scholar]
- 23.Malka D., Boige V., Faron M., et al. Hepatic arterial infusion (HAI) of oxaliplatin plus intravenous (IV) fluorouracil (FU), leucovorin (LV) and cetuximab for first-line treatment of unresectable colorectal liver metastases (CRLM): final results of a multicenter phase 2 study (CHOICE) Ann Oncol. 2012;23 [Google Scholar]
- 24.Cercek A., Boucher T.M., Gluskin J.S., et al. Response rates of hepatic arterial infusion pump therapy in patients with metastatic colorectal cancer liver metastases refractory to all standard chemotherapies. J Surg Oncol. 2016;114(6):655–663. doi: 10.1002/jso.24399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Boige V., Malka D., Elias D., et al. Hepatic arterial infusion of oxaliplatin and intravenous LV5FU2 in unresectable liver metastases from colorectal cancer after systemic chemotherapy failure. Ann Surg Oncol. 2008;15(1):219–226. doi: 10.1245/s10434-007-9581-7. [DOI] [PubMed] [Google Scholar]
- 26.Meyblum L., Faron M., Deschamps F., et al. Safety and efficacy of percutaneous arterial port Implantation for Hepatic Arterial Infusion Chemotherapy. Eur Radiol. 2025;35:1022–1033. doi: 10.1007/s00330-024-10887-1. [DOI] [PubMed] [Google Scholar]
- 27.Jones R.P., Hamann S., Malik H.Z., Fenwick S.W., Poston G.J., Folprecht G. Defined criteria for resectability improves rates of secondary resection after systemic therapy for liver limited metastatic colorectal cancer. Eur J Cancer. 2014;50(9):1590–1601. doi: 10.1016/j.ejca.2014.02.024. [DOI] [PubMed] [Google Scholar]
- 28.Adam R., Aloia T., Lévi F., et al. Hepatic resection after rescue cetuximab treatment for colorectal liver metastases previously refractory to conventional systemic therapy. J Clin Oncol. 2007;5(29):4593–4602. doi: 10.1200/JCO.2007.10.8126. [DOI] [PubMed] [Google Scholar]
- 29.Gallagher D.J., Capanu M., Raggio G., Kemeny N. Hepatic arterial infusion plus systemic irinotecan in patients with unresectable hepatic metastases from colorectal cancer previously treated with systemic oxaliplatin: a retrospective analysis. Ann Oncol. 2007;18(12):1995–1999. doi: 10.1093/annonc/mdm405. [DOI] [PubMed] [Google Scholar]
- 30.Cercek A., D’Angelica M., Power D., et al. Floxuridine hepatic arterial infusion associated biliary toxicity is increased by concurrent administration of systemic bevacizumab. Ann Surg Oncol. 2014;21(2):479–486. doi: 10.1245/s10434-013-3275-0. [DOI] [PubMed] [Google Scholar]
- 31.Boilève A., De Cuyper A., Larive A., et al. Hepatic arterial infusion of oxaliplatin plus systemic chemotherapy and targeted therapy for unresectable colorectal liver metastases. Eur J Cancer. 2020;138:89–98. doi: 10.1016/j.ejca.2020.07.022. [DOI] [PubMed] [Google Scholar]
- 32.Pernot S., Pellerin O., Mineur L., et al. Phase III randomized trial comparing systemic versus intra-arterial oxaliplatin, combined with LV5FU2 +/- irinotecan and a targeted therapy, in the first-line treatment of metastatic colorectal cancer restricted to the liver (OSCAR): PRODIGE 49. Dig Liver Dis. 2022;54(3):324–330. doi: 10.1016/j.dld.2021.12.012. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.