Journal Logo


A Randomized Phase II Trial of Adjuvant Hepatic Arterial Infusion and Systemic Therapy With or Without Panitumumab After Hepatic Resection of KRAS Wild-type Colorectal Cancer

Kemeny, Nancy E. MD; Chou, Joanne F. MPH; Capanu, Marinela PhD; Chatila, Walid K. MS; Shi, Hongyu BS; Sanchez-Vega, Francisco PhD§,||; Kingham, Thomas Peter MD; Connell, Louise Catherine MBBCh; Jarnagin, William R. MD; D’Angelica, Michael I. MD

Author Information
doi: 10.1097/SLA.0000000000004923


There are almost 150,000 new cases and ∼51,000 deaths from colorectal cancer annually in the US.1 Liver metastases are detected at diagnosis in 15% to 25% of patients, whereas 60% of those who subsequently develop metastatic disease have liver involvement.2 Resection is a potentially curative option for some patients with liver-confined metastases, with 5-year overall survival (OS) of approximately 30% to 50%, yet recurrence occurs in up to 70% of these patients within 2 years, the majority in the remnant liver.3–5 Postresection liver recurrence may be explained by the presence of microscopic residual metastases.6 Tumors that grow to >3 mm derive most of their blood supply from hepatic arterial circulation.7 Therefore, infusion of chemotherapy directly into the hepatic artery may decrease hepatic recurrence. Indeed, hepatic arterial infusion (HAI) of chemotherapy plus systemic therapy (SYS) has significantly improved recurrence-free survival (RFS).8–12

Monoclonal antibodies to epidermal growth factor receptor (EGFR), panitumumab (Pmab), and cetuximab (cetux), have led to improved survival outcomes when added to SYS chemotherapy,13–18 including improved resection rates.19–21 We assessed whether the addition of panitumumab (Pmab) vs no Pmab to an adjuvant regimen of floxuridine (FUDR) plus SYS leucovorin, fluorouracil, and irinotecan (FOLFIRI) would improve RFS after resection of KRAS wild-type colorectal liver metastases.


Study Design

All patients were recruited from Memorial Sloan Kettering Cancer Center (MSK) and signed informed consent; the protocol and informed consent were approved by the MSK institutional review board (NCT01312857). The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. Patients were consented and randomized to the trial after liver surgery and pump placement.

Patients were randomly assigned in a 1:1 ratio to HAI FUDR + SYS FOLFIRI ± panitumumab (arm A, +Pmab; arm B, no Pmab). Patients were registered in the MSK Protocol Participant Registration system and randomized using the MSK Clinical Research Database. Patients were assessed by clinical risk score, which awards 1 point each for node-positive primary disease, <12 months disease-free, >1 metastasis, largest diameter of metastasis >5 cm, and carcinoembryonic antigen level >200 μg/L.3 All randomized patients were included in the analysis, and we did not exclude any patients with elevated alkaline phosphatase after surgery.


Patient assessment for protocol entry is shown in the CONSORT diagram (Fig. 1). All patients had histologically confirmed colorectal adenocarcinoma with completely resected liver metastases and no evidence of extrahepatic disease as well as KRAS codon 12 wild-type tumors. Enrollment was initially limited to KRAS codon 12 wild-type. The enrollment criteria were updated when it became evident that any RAS mutation could influence response to Pmab.13 Tumor tissue was analyzed for RAS-pathway mutations using a next-generation sequencing platform. Pretreatment evaluation included complete medical and surgical histories, physical examination, and appropriate laboratory studies. All patients underwent preoperative computed tomography (CT) liver angiogram with visualization of the celiac and superior mesenteric arteries to determine adequate hepatic arterial anatomy and had MediPort placed before starting systemic treatment.

CONSORT diagram. Later changed to all-RAS mutated.

Guidelines for pump placement have been previously reported.22 A CT scan of the chest, abdomen, and pelvis was performed postoperatively within 3 weeks before the assigned start of adjuvant treatment, to take a baseline measurement of disease. Previous chemotherapy and prior investigational agents were permitted if the last doses were administered ≥21 days and ≥30 days before trial entry, respectively. Prior anti-EGFR antibody therapy was permitted if the patient had demonstrated a response to treatment. The extent of liver resection was characterized as minor or major. A major resection was defined as a resection of 3 or more segments. A wedge resection was counted as a resection of one quarter of a segment. The number of segments resected for patients undergoing atypical resections was estimated by a review of the operative notes and imaging studies.

Exclusion criteria included age <18 years; serious comorbidities including active infection, active hepatitis, history of primary CNS tumor, and Gilbert disease; history of other malignancy except malignancy treated with curative intent without known active disease for ≥3 years or adequately treated nonmelanomatous skin cancer; and Karnofsky performance status <60, white blood cell count <3 × 109/L, absolute neutrophil count (ANC) ≤1.5 × 109/L, platelet count <100 × 109/L, creatinine ≥132 μmol/L, hemoglobin <90 g/L, magnesium <lower limit of normal, calcium <lower limit of normal, aspartate aminotransferase > 5× upper limit of normal, alanine aminotransferase > 5× upper limit of normal, and total bilirubin >26 μmol/L (>1.5 mg/dL). No prior radiation to the liver or prior treatment with HAI FUDR was allowed.

Chemotherapy Administration

Patients commenced chemotherapy 4 to 5 weeks postsurgery. Six 35-day treatment cycles were planned. In each 5-week cycle, HAI FUDR + dexamethasone (Dex) was infused for 14 days (1–14) at a dose of 0.12 mg/kg/pump volume (30 cc) divided by pump flow rate (supplied by the pump manufacturer, Codman Inc, Johnson and Johnson). For patients ≥35% above ideal weight, FUDR dose was calculated using the average of actual weight and ideal weight. Dex was administered concurrently with FUDR at a flat dose of 25 mg, combined with 30,000 units of unfractionated heparin and normal saline in a quantity enough to fill the pump reservoir to 30 cc.

On days 15 and 29 of each cycle, the pump was emptied and filled with 30 cc of heparinized saline. SYS was also administered on days 15 and 29 (Table S1, Patients received irinotecan at 125 mg/m2 to 150 mg/m2 (30–60-minute continuous infusion) concurrently with leucovorin at 400 mg/m2 via a Y connection, followed by 5FU 2000 mg/m2 delivered via an external pump for a 48-hour continuous infusion. No bolus 5FU was administered. Patients randomized to arm A received Pmab at a dose of 6 mg/kg (60-minute continuous infusion).

Toxicity Evaluation

All toxicities were graded according to the National Cancer Institute Common Toxicity Criteria v3.0. Because patients entered the study with varying degrees of hepatic enzyme abnormalities, individual baseline values were used to determine FUDR dose modification. The FUDR dose was lowered or held, as outlined previously, if there were increases of alkaline phosphatase, aspartate aminotransferase, and total bilirubin.23

Patients were required to have a white blood cell count ≥2.5 × 109/L, absolute neutrophil count > 1 × 109/L, platelet count ≥75 × 109/L, creatinine level ≤159 μmol/L, and bilirubin level <26 μmol/L (1.5 mg/dL) for subsequent cycles of systemic chemotherapy. If these levels were not met, treatment was delayed for 1 or 2 weeks. CT scans of the chest, abdomen, and pelvis were done every 2 cycles to assess for recurrence. All patients received prophylactic minocycline and a topical steroid for skin rash and were evaluated by a dermatologist if toxicity could not be managed.

Assessments and Statistical Analysis

For power calculations, the primary endpoint of 15-month RFS was defined as a binary outcome. We chose an unacceptable RFS rate of 50% derived as the average RFS of our recent trials and historical controls.24–29 Based on an exact binomial design, with 39 patients in each arm we could differentiate an unacceptable 15-month RFS of 50% and an acceptable 15-month RFS of 70% with type I and II error of 10% each. If ≥24 patients in 1 arm were alive and disease-free at 15 months that regimen would be considered worthy of further investigation. This design was conducted separately for each arm and therefore no comparison was planned between the 2 arms. If the regimens in each arm were deemed efficacious (ie, both arms had ≥24 patients alive and disease-free at 15 months), a pick-the-winner approach would be used, based on the randomized phase II clinical approaches proposed by Simon et al.30

Secondary endpoints included toxicity, median RFS and OS, and whether tumor-related mutations such as NRAS, BRAF, PIK3CA, AKT1, and MEK1 were prognostic of RFS and OS. Toxicity was assessed separately for each arm using descriptive summary statistics. RFS and OS were calculated from start of treatment and estimated using the Kaplan-Meier method. Surviving patients without disease recurrence were censored at the date of their last follow-up visit for RFS. Overall and progression-free survival rates were reported separately for each arm. Carcinoembryonic antigen tests and CT scans of the chest, abdomen, and pelvis were performed every 3 months for the first 2 years post-treatment, every 4 months after 2 years, every 6 months after 4 years, and yearly after 5 years.

To assess tumor-related mutations, formalin-fixed paraffin-embedded tumor tissue and matched normal blood samples collected under an IRB-approved protocol were analyzed prospectively in a Clinical Laboratory Improvement Amendments-certified laboratory. Our onsite 341 to 468 gene next-generation sequencing assay, MSK-IMPACT, detects mutations, small insertions and deletions, copy number alterations, and select structural rearrangements.31,32 Genomic alterations were filtered for oncogenic variants using OncoKB.33 To correlate somatic mutations in tumor tissue for prognostic efficacy, samples from arms A and B were combined, genes with mutation frequency ≥10% in samples analyzed by MSK-IMPACT were associated with OS and RFS using a log-rank test.

The accrual target was revised from 78 patients to 75 patients when Codman pumps were no longer available. All analyses were performed in SAS (v.9.4, Cary, NC) or R (v.3.6.1, Vienna, Austria).34,35 All P values are 2-sided and a value of ≤0.05 indicates statistical significance.


Patient Characteristics

Seventy-five patients were randomized into the study, 37 in arm A (+Pmab) and 38 in arm B (no Pmab). Patient characteristics are shown in Table 1. To ensure relatively balanced patient distribution between the 2 treatment arms, prior chemotherapy (yes vs no) and clinical risk score (0–2 vs >2 points) were both used to stratify patients.

TABLE 1 - Patient Characteristics (N = 75)
Arm A + Pmab Arm B no Pmab All Patients
n = 37 n = 38 N = 75
Age ≥50 21 (57) 22 (58) 43 (57)
Karnofsky performance score ≥70 37 (100) 38 (100) 75 (100)
Female 12 (32) 17 (45) 29 (39)
Clinical risk score ≥3 16 (43) 17 (45) 33 (44)
 + Lymph nodes at primary 25 (68) 25 (66) 50 (67)
 Disease-free interval ≤12 mos. 30 (81) 26 (68) 56 (75)
 >1 liver metastasis 22 (59) 29 (76) 51 (68)
 Largest diameter ≥5 cm 11 (30) 7 (18) 18 (24)
 Preoperative CEA >200 μg/L 2 (5.4) 3 (8.0) 5 (6.7)
Synchronous disease 22 (59) 19 (50) 41 (55)
Poorly differentiated 4 (11) 5 (13) 9 (12)
≥4 liver metastases 9 (24) 10 (26) 19 (25)
Median number of liver lesions 2 2 2
Tumor margins <1 cm 22 (59) 17 (45) 39 (52)
Positive tumor margins (<0.1 mm) 2 (5.4) 3 (8.0) 5 (6.7)
Bilobar 18 (49) 18 (47) 36 (48)
Disease site (colon/rectum) 32 (86) / 5 (14) 30 (79) / 8 (21) 62 (83) / 13 (17)
 Left colon 26 (81) 21 (70) 47 (76)
 Right colon 6 (19) 9 (30) 15 (24)
Preoperative CEA >30 μg/L 7 (19) 5 (13) 12 (16)
Postoperative CEA >5 μg/L 7 (19) 6 (16) 13 (17)
Previous chemotherapy 33 (89) 33 (87) 66 (88)
 Adjuvant after primary resection 11 (33) 15 (45) 26 (39)
 Metastatic 22 (67) 18 (55) 40 (61)
Previous oxaliplatin before entry 32 (86) 32 (84) 64 (85)
Prior liver surgery/radiofrequency ablation 5 (14) 2 (5.3) 7 (9.3)
Simultaneous resection of primary and liver mets at time of pump placement 13 (35) 14 (37) 26 (35)
Ablation at time of pump placement 3 (8.1) 7 (18) 10 (13)
Major liver resection 18 (49) 15 (39) 33 (44)
Values are n (%) unless noted. There were no significant differences in any factors between treatment groups.
Systemic therapy for liver metastases before resection.CEA indicates carcinoembryonic antigen; mets, metastases; Pmab, panitumumab.

RFS and OS Outcomes

At the time of analysis, 25 patients in arm A and 18 patients in arm B were alive and recurrence-free at 15 months; hence, only arm A (+Pmab) met the decision rule of ≥24 patients alive and recurrence-free at 15 months and was worthy of further investigation according to the binomial design described in the Methods section. Arm B (no Pmab) was not deemed efficacious and, therefore, the pick-the-winner design was not used. The 15-month RFS was 69% (95% CI, 53–82) in arm A and 47% (95% CI, 32–63) in arm B. The median follow-up among survivors was 56.6 months. In arms A and B, respectively, 3-year RFS was 57% (95% CI, 43–76) and 42% (95% CI, 29–61; Fig. 2), while 3-year OS was 97% (95% CI, 90–99) and 91% (95% CI, 83–100; Fig. 3).

Relapse-free survival by randomized arm. HAI indicates hepatic arterial infusion; Pmab, panitumumab; SYS, systemic therapy.
Overall survival by randomized arm. HAI indicates hepatic arterial infusion; Pmab, panitumumab; SYS, systemic therapy.


There were similar toxicities in the 2 trial arms, except for Pmab-related rash in arm A, and the majority of toxicities were grade ≤3 (Table 2). Grade 3 liver toxicity with respect to elevated alkaline phosphatase and bilirubin was also similar between patients who underwent major (15%; 5/33) and minor (7%; 3/42) resections. During treatment, patients in arm A were admitted for gastrointestinal problems (cecal volvulus, colonic obstruction, dilated small bowel, abdominal pain) which all resolved with conservative measures (n = 8), colonic anastomotic leak requiring surgery (n = 1), diarrhea (n = 2), and neutropenic fever (n = 1). In arm B, patients were admitted for abdominal problems (n = 5) and neutropenic fever (n = 1). Three patients (2 in arm A and 1 in arm B) were admitted for post-surgical abscesses before starting therapy.

TABLE 2 - Toxicity
Grade 3 Grade 4
Arm A + Pmab Arm B no Pmab All Patients Arm A + Pmab Arm B no Pmab All Patients
Toxicity Grade n = 37 n = 38 N = 75 n = 37 n = 38 N = 75
Elevated liver enzymes
 ALT 12 (32) 11 (29) 23 (31) 0 0 0
 AST 4 (11) 4 (11) 8 (11) 0 0 0
 Bilirubin 1 (2.7) 0 2 (2.7) 0 1 (2.6) 1 (1.3)
 ALP 3 (8.1) 3 (8) 6 (8) 1 (2.7) 0 1 (1.3)
Diarrhea 8 (22) 4 (11) 12 (16) 0 0 0
Rash 9 (24) 0 9 (12) 0 0 0
Hyperglycemia 3 (8.1) 2 (5.2) 5 (6.7) 0 0 0
Neutropenia 2 (5.4) 3 (7.8) 5 (6.7) 0 0 0
Lymphopenia 4 (11) 2 (5.2) 6 (8.0) 0 0 0
Hypophosmatemia 3 (8.1) 0 3 (4.0) 0 0 0
Paronychia 1 (2.7) 0 1 (1.3) 0 0 0
Vomiting 4 (11) 0 4 (5.3) 0 0 0
Values are n (%). Two of 75 (3%) patients had biliary stents placed: 1/37 (3%) in arm A and 1/38 (3%) in arm B. Hospital admissions on trial: 14/37 (38%) in arm A and 7/38 (18%) in arm B.ALP indicates alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; Pmab, panitumumab.

Dosing of HAI FUDR and SYS FOLFIRI in the 2 arms was compared by calculating the fraction of the planned dose that patients received. For SYS, in arms A and B, respectively, 83.8% and 92% of patients received >50% of the planned dose, whereas 40.5% and 47.7% of patients received 100% of the planned dose. In the first 3 cycles in arms A and B, respectively, 77% and 90% of the planned dose was administered, while 37% and 42% of the planned dose was administered in all 6 cycles, respectively. In arm A, 85% of patients received >50% of the planned Pmab dose and 16% of patients received 100%. The reasons for failing to complete therapy in arms A and B, respectively, were the development of metastases (2 vs 2), enteritis in previously radiated pts (3 vs 1), and refusal of therapy (3 vs 0).

Biomarkers Correlated With Recurrence and Survival

Genes with ≥10% mutation frequency in the available IMPACT samples are presented in Table S2, There was no statistically significant association between the most frequently mutated genes and RFS (Table 3). No significant difference was seen between the genetic profile of patients in arms A and B (see Table S2, Before the protocol was updated, 2 patients (1 from each arm) with NRAS alterations and 2 patients (both in arm B) with BRAF alterations were enrolled. One patient with BRAF mutation and 3 others had KRAS exon 4 p.A146T mutations. No mutations in AKT1 or MEK1 were found.

TABLE 3 - Recurrence-free Survival (RFS) by Genomic Alteration
Variable Group Obs. Events 36 Months (95% CI) P value
Overall 75 40 48.9 (36.9, 59.9)
Any APC alteration WT 12 8 33.3 (10.3, 58.8) 0.426
Altered 38 20 48.2 (31.2, 63.2)
Missing 25
Any TP53 alteration WT 10 5 50.0 (18.4, 75.3) 0.576
Altered 40 23 42.7 (26.7, 57.8)
Missing 25
Any PIK3CA alteration WT 64 33 49.9 (36.9, 61.6) 0.185
Altered 11 7 45.5 (16.7, 70.7)
Any SOX9 alteration WT 42 25 40.3 (25.0, 55.1) 0.282
Altered 8 3 62.5 (22.9, 86.1)
Missing 25
Any SMAD4 alteration WT 44 26 40.8 (25.8, 55.2) 0.347
Altered 6 2 66.7 (19.5, 90.4)
Missing 25
Any CDK8 alteration WT 44 23 48.0 (32.2, 62.1) 0.124
Altered 6 5 16.7 (0.8, 51.7)
Missing 25
Any BCL2L1 alteration WT 43 23 45.5 (30.0, 59.7) 0.506
Altered 7 5 34.3 (4.8, 68.5)
Missing 25
Any FLT3 alteration WT 44 24 46.1 (30.6, 60.2) 0.589
Altered 6 4 33.3 (4.6, 67.6)
Missing 25
Any DNMT3B alteration WT 44 23 45.9 (30.4, 60.1) 0.381
Altered 6 5 33.3 (4.6, 67.6)
Missing 25
Any SRC alteration WT 44 23 48.6 (32.9, 62.6) 0.236
Altered 6 5 16.7 (0.8, 51.7)
Missing 25
P value calculated by log-rank test.Obs. indicates observations; missing, no data; 0, no mutation; 1, mutated.

The only gene associated with OS in this study was PIK3CA. Patients with alterations in PIK3CA had a worse prognosis, with 3-year OS of 68% (95% CI, 29–89), as compared to PIK3CA wild-type (98% (95% CI, 87–99.7 (P < 0.001; Table S3, An exploratory, unplanned analysis showed that patients with altered PIK3CA who were treated with Pmab (arm A) appeared to have better OS than those with altered PIK3CA not treated with Pmab (arm B) (Fig. S1,; however, interpretation of this should be made cautiously because there were small numbers of patients.


In this trial, patients who underwent liver resection of metastatic colorectal cancer were randomized to receive Pmab or no Pmab in combination with systemic FOLFIRI and HAI therapy. The addition of Pmab led to improved outcomes, with 3-year RFS of 57% (95% CI, 43–76), while the arm not receiving Pmab did not meet the prespecified decision rule, with 3-year RFS of 42% (95% CI, 29–61).

Approximately half of all patients whose disease recurs experience the recurrence within in the first 2 years post-treatment.36 Furthermore, patients with metastatic colorectal cancer to the liver usually have a shortened survival. Those who can undergo resection of their liver metastases have improved survival, with 5-year OS ranging from 30% to 50%.3 In this trial, 3-year survival was dramatically increased to 97% (95% CI, 90 to 99) and 91% (95% CI, 83 to 100) for the ±Pmab groups, respectively.

SYS alone has been evaluated in the adjuvant setting after liver resection in several studies.37 In an EORTC study, patients with 1 to 4 resectable liver metastases were randomized to 6 cycles of FOLFOX4 pre- and post-surgery versus surgery alone. Three-year disease-free survival increased from 28% to 36% in patients receiving perioperative chemotherapy (P = 0.041) but there was no increase in OS.38,39

Randomized studies evaluating adjuvant HAI therapy after liver resection have found increased disease-free survival and OS rates when using HAI. A randomized study at MSK comparing SYS ± HAI, which found 89% versus 73% 2-year OS (P = 0.02) and median RFS of 31 and 17 months, respectively, in the groups receiving HAI or not.8,9 Similarly, a retrospective study of 2368 consecutive patients who underwent liver resection of colorectal liver metastases at MSK showed that those treated with HAI had a significantly better survival (median 67 vs 44 months; P < 0.001) despite more advanced disease in the HAI group.40

Several studies have found a significant increase in treatment response and progression-free survival when anti-EGFR agents are added to SYS.13–21 A meta-analysis demonstrated that more patients whose disease was initially considered unresectable lived long enough to undergo liver resection with the use of anti-EGFR agents.41 However, the New EPOC study that randomized patients to chemotherapy ± cetux before and after resection found that cetux did not improve patient outcomes.42 There were some problems with this study; for instance, deaths and disease recurrence occurred very quickly in the cetux arm, which is unusual for patients undergoing liver resection. Additionally, almost 90% of patients received FOLFOX rather than FOLFIRI. In some patients, anti-EGFR agents may be more useful when combined with irinotecan rather than with oxaliplatin.43

Our results suggest the use of Pmab in the adjuvant setting after liver resection is favorable. This benefit may be explained by the use of an irinotecan-based chemotherapy backbone (FOLFIRI not FOLFOX) with the anti-EGFR drug and by the incorporation of an anti-EGFR drug postresection rather than in the neoadjuvant setting. After resection, growth factors required for liver regeneration can also stimulate tumor cell proliferation and may enhance growth in micro-metastases in the remnant liver.44,45 Indeed, many growth factors, including epidermal growth factor, are upregulated in liver regeneration and play a major role in hepatocyte proliferation during regeneration.46 In one study of liver regeneration after portal vein embolization, liver metastases exceeded the growth of the normal human liver cells.47 It may be that the growth factors have a greater stimulatory effect on tumor cells than on normal hepatocytes.45,48 Anti-EGFR agents such as Pmab arrest the cascade of intercellular signals activated by the EGFR receptor, thereby inhibiting tumor cell proliferation.49,50

Another difference in our trial compared with others is the use of HAI therapy. The antiproliferative effect of Pmab combined with HAI therapy after liver resection may have improved patient outcomes. Both protocol arms achieved a 3-year survival of >90%. In an earlier study using bevacizumab with HAI therapy,29 we found increased biliary toxicity that was not found with Pmab. By combining Pmab with FOLFIRI and HAI, we demonstrated increased 15-month RFS (the primary objective) without increased biliary toxicity; therefore, this approach should be further investigated in a larger study.


Editorial support in the preparation of this manuscript was provided by Hannah Rice, BA, ELS.


1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin 2019; 69:7–34.
2. Engstrand J, Nilsson H, Stromberg C, et al. Colorectal cancer liver metastases - a population-based study on incidence, management and survival. BMC Cancer 2018; 18:78.
3. Tomlinson JS, Jarnagin WR, DeMatteo RP, et al. Actual 10-year survival after resection of colorectal liver metastases defines cure. J Clin Oncol 2007; 25:4575–4580.
4. Fiorentini G, Sarti D, Aliberti C, et al. Multidisciplinary approach of colorectal cancer liver metastases. World J Clin Oncol 2017; 8:190–202.
5. Fong Y, Fortner J, Sun RL, et al. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999; 230:309–318. discussion 318-321.
6. Hughes K, Scheele J, Sugarbaker PH. Surgery for colorectal cancer metastatic to the liver. Optimizing the results of treatment. Surg Clin North Am 1989; 69:339–359.
7. Ackerman NB. The blood supply of experimental liver metastases. IV. Changes in vascularity with increasing tumor growth. Surgery 1974; 75:589–596.
8. Kemeny N, Huang Y, Cohen AM, et al. Hepatic arterial infusion of chemotherapy after resection of hepatic metastases from colorectal cancer. N Engl J Med 1999; 341:2039–2048.
9. Kemeny NE, Gonen M. Hepatic arterial infusion after liver resection. N Engl J Med 2005; 352:734–735.
10. Kemeny MM, Adak S, Gray B, et al. Combined-modality treatment for resectable metastatic colorectal carcinoma to the liver: surgical resection of hepatic metastases in combination with continuous infusion of chemotherapy--an intergroup study. J Clin Oncol 2002; 20:1499–1505.
11. Lygidakis NJ, Ziras N, Parissis J. Resection versus resection combined with adjuvant pre- and post-operative chemotherapy--immunotherapy for metastatic colorectal liver cancer. A new look at an old problem. Hepatogastroenterology 1995; 42:155–161.
12. Lorenz M, Muller HH, Schramm H, et al. Randomized trial of surgery versus surgery followed by adjuvant hepatic arterial infusion with 5-fluorouracil and folinic acid for liver metastases of colorectal cancer. German Cooperative on Liver Metastases (Arbeitsgruppe Lebermetastasen). Ann Surg 1998; 228:756–762.
13. Douillard J-Y, Oliner KS, Siena S, et al. Panitumumab–FOLFOX4 treatment and RAS mutations in colorectal cancer. N Engl J Med 2013; 369:1023–1034.
14. Van Cutsem E, Lenz HJ, Köhne CH, et al. Fluorouracil, leucovorin, and irinotecan plus cetuximab treatment and RAS mutations in colorectal cancer. J Clin Oncol 2015; 33:692–700.
15. Ji JH, Park SH, Lee J, et al. Prospective phase II study of neoadjuvant FOLFOX6 plus cetuximab in patients with colorectal cancer and unresectable liver-only metastasis. Cancer Chemother Pharmacol 2013; 72:223–230.
16. Folprecht G, Gruenberger T, Bechstein WO, et al. Tumour response and secondary resectability of colorectal liver metastases following neoadjuvant chemotherapy with cetuximab: the CELIM randomised phase 2 trial. Lancet Oncol 2010; 11:38–47.
17. Levi F, Karaboue A, Gorden L, et al. Cetuximab and circadian chronomodulated chemotherapy as salvage treatment for metastatic colorectal cancer (mCRC): safety, efficacy and improved secondary surgical resectability. Cancer Chemother Pharmacol 2011; 67:339–348.
18. Garufi C, Torsello A, Tumolo S, et al. Cetuximab plus chronomodulated irinotecan, 5-fluorouracil, leucovorin and oxaliplatin as neoadjuvant chemotherapy in colorectal liver metastases: POCHER trial. Br J Cancer 2010; 103:1542–1547.
19. Lv W, Zhang GQ, Jiao A, et al. Chemotherapy plus cetuximab versus chemotherapy alone for patients with KRAS wild type unresectable liver-confined metastases colorectal cancer: an updated meta-analysis of RCTs. Gastroenterol Res Pract 2017; 2017:8464905.
20. Ye LC, Liu TS, 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; 31:1931–1938.
21. Douillard JY, Siena S, Peeters M, et al. Impact of early tumour shrinkage and resection on outcomes in patients with wild-type RAS metastatic colorectal cancer. Eur J Cancer 2015; 51:1231–1242.
22. Kemeny N, Daly J, Oderman P, et al. Hepatic artery pump infusion: toxicity and results in patients with metastatic colorectal carcinoma. J Clin Oncol 1984; 2:595–600.
23. Power DG, Kemeny NE. The role of floxuridine in metastatic liver disease. Mol Cancer Ther 2009; 8:1015–1025.
24. Kornprat P, Jarnagin WR, Gonen M, et al. Outcome after hepatectomy for multiple (four or more) colorectal metastases in the era of effective chemotherapy. Ann Surg Oncol 2007; 14:1151–1160.
25. Kemeny N, Capanu M, D’Angelica M, et al. Phase I trial of adjuvant hepatic arterial infusion (HAI) with floxuridine (FUDR) and dexamethasone plus systemic oxaliplatin, 5-fluorouracil and leucovorin in patients with resected liver metastases from colorectal cancer. Ann Oncol 2009; 20:1236–1241.
26. Kemeny N, Jarnagin W, Gonen M, et al. Phase I/II study of hepatic arterial therapy with floxuridine and dexamethasone in combination with intravenous irinotecan as adjuvant treatment after resection of hepatic metastases from colorectal cancer. J Clin Oncol 2003; 21:3303–3309.
27. Choti MA, Sitzmann JV, Tiburi MF, et al. Trends in long-term survival following liver resection for hepatic colorectal metastases. Ann Surg 2002; 235:759–766.
28. Nordlinger B, Guiguet M, Vaillant JC, et al. Surgical resection of colorectal carcinoma metastases to the liver. A prognostic scoring system to improve case selection, based on 1568 patients. Association Francaise de Chirurgie. Cancer 1996; 77:1254–1262.
29. Kemeny NE, Jarnagin WR, Capanu M, et al. Randomized phase II trial of adjuvant hepatic arterial infusion and systemic chemotherapy with or without bevacizumab in patients with resected hepatic metastases from colorectal cancer. J Clin Oncol 2011; 29:884–889.
30. Simon R, Wittes RE, Ellenberg SS. Randomized phase II clinical trials. Cancer Treat Rep 1985; 69:1375–1381.
31. Cheng DT, Mitchell TN, Zehir A, et al. Memorial sloan kettering-integrated mutation profiling of actionable cancer targets (MSK-IMPACT): a hybridization capture-based next-generation sequencing clinical assay for solid tumor molecular oncology. J Mol Diagn 2015; 17:251–264.
32. Zehir A, Benayed R, Shah RH, et al. Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients. Nat Med 2017; 23:703–713.
33. Chakravarty D, Gao J, Phillips SM, et al. OncoKB: a precision oncology knowledge base. JCO Precis Oncol 2017; 2017:PO.17.00011.
34. R. R Development Core Team. A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2020.
35. SAS Vija [computer program]. Version 9.4. Cary, NC, USA2020.
36. House MG, Kemeny NE, Gonen M, et al. Comparison of adjuvant systemic chemotherapy with or without hepatic arterial infusional chemotherapy after hepatic resection for metastatic colorectal cancer. Ann Surg 2011; 254:851–856.
37. Mitry E, Fields AL, Bleiberg H, et al. Adjuvant chemotherapy after potentially curative resection of metastases from colorectal cancer: a pooled analysis of two randomized trials. J Clin Oncol 2008; 26:4906–4911.
38. Nordlinger B, Sorbye H, Glimelius B, et al. Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial. Lancet 2008; 371:1007–1016.
39. Nordlinger B, Sorbye H, Glimelius B, et al. Perioperative FOLFOX4 chemotherapy and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC 40983): long-term results of a randomised, controlled, phase 3 trial. Lancet Oncol 2013; 14:1208–1215.
40. Groot Koerkamp B, Sadot E, Kemeny NE, et al. Perioperative hepatic arterial infusion pump chemotherapy is associated with longer survival after resection of colorectal liver metastases: a propensity score analysis. J Clin Oncol 2017; 35:1938–1944.
41. Modest DP, Martens UM, Riera-Knorrenschild J, et al. FOLFOXIRI plus panitumumab as first-line treatment of RAS wild-type metastatic colorectal cancer: the randomized, open-label, phase II VOLFI study (AIO KRK0109). J Clin Oncol 2019; 37:3401–3411.
42. Primrose J, Falk S, Finch-Jones M, et al. Systemic chemotherapy with or without cetuximab in patients with resectable colorectal liver metastasis: the New EPOC randomised controlled trial. Lancet Oncol 2014; 15:601–611.
43. Aderka D, Stintzing S, Heinemann V. Explaining the unexplainable: discrepancies in results from the CALGB/SWOG 80405 and FIRE-3 studies. Lancet Oncol 2019; 20:e274–e283.
44. Krause P, Flikweert H, Monin M, et al. Increased growth of colorectal liver metastasis following partial hepatectomy. Clin Exp Metastasis 2013; 30:681–693.
45. Harun N, Nikfarjam M, Muralidharan V, et al. Liver regeneration stimulates tumor metastases. J Surg Res 2007; 138:284–290.
46. Lukomska B, Dluzniewska J, Polanski J, et al. Expression of growth factors in colorectal carcinoma liver metastatic patients after partial hepatectomy: implications for a functional role in cell proliferation during liver regeneration. Comp Hepatol 2004; 3: (Suppl 1): S52.
47. Elias D, De Baere T, Roche A, et al. During liver regeneration following right portal embolization the growth rate of liver metastases is more rapid than that of the liver parenchyma. Br J Surg 1999; 86:784–788.
48. Jiang WG, Hallett MB, Puntis MC. Hepatocyte growth factor/scatter factor, liver regeneration and cancer metastasis. Br J Surg 1993; 80:1368–1373.
49. Said NA, Williams ED. Growth factors in induction of epithelial-mesenchymal transition and metastasis. Cells Tissues Organs 2011; 193:85–97.
50. Argast GM, Campbell JS, Brooling JT, et al. Epidermal growth factor receptor transactivation mediates tumor necrosis factor-induced hepatocyte replication. J Biol Chem 2004; 279:34530–34536.

chemotherapy; colon cancer; hepatic arterial infusion; liver toxicities; systemic

Supplemental Digital Content

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.