- © 2013 by American Society of Clinical Oncology
Randomized Controlled Trial of Cetuximab Plus Chemotherapy for Patients With KRAS Wild-Type Unresectable Colorectal Liver-Limited Metastases
- Le-Chi Ye,
- Tian-Shu Liu,
- Li Ren,
- Ye Wei,
- De-Xiang Zhu,
- Sheng-Yong Zai,
- Qing-Hai Ye,
- Yiyi Yu,
- Bo Xu,
- Xin-Yu Qin and
- Jianmin Xu⇑
- Corresponding author: Jianmin Xu, MD, PhD, Department of General Surgery, Zhongshan Hospital, Fudan University, 180 Fenglin Rd, Shanghai, 200032, People's Republic of China; e-mail: xujmin{at}yahoo.com.cn.
-
Presented orally at the 10th Annual Meeting of the Japanese Society of Medical Oncology, Osaka, Japan, July 26-28, 2012; at the 13th Korea-Japan-China Colorectal Cancer Symposium, Seoul, Korea, September 8-9, 2012; and in poster format at the 37th Congress of the European Society for Medical Oncology Vienna, Austria, September 28-October 2, 2012.
-
L.-C.Y., T.-S.L., L.R., and Y.W. contributed equally to this work.
Abstract
Purpose To assess the effects of cetuximab plus chemotherapy as first-line treatment for unresectable colorectal liver metastases (CLMs).
Patients and Methods After resection of their primary tumors, patients with KRAS wild-type synchronous nonresectable liver-limited metastases from colorectal cancer were randomly assigned to receive chemotherapy (FOLFIRI [fluorouracil, leucovorin, and irinotecan] or mFOLFOX6 [modified fluorouracil, leucovorin, and oxaliplatin]) plus cetuximab (arm A) or chemotherapy alone (arm B). The primary end point was the rate of patients converted to resection for liver metastases. Secondary end points included tumor response and survival.
Results The intent-to-treat population comprised 138 patients; 70 patients were randomly assigned to arm A and 68 to arm B. After a median of 25.0 months of follow-up, the 3-year overall survival (OS) rate and median survival time (MST) for all patients were 30% and 24.4 months, respectively. The R0 resection rates for liver metastases were 25.7% (18 of 70 patients) in arm A and 7.4% (five of 68 patients) in arm B, which were significantly different (P < .01). Patients in arm A had improved objective response rates (57.1% v 29.4%; P < .01), increased 3-year OS rate (41% v 18%; P = .013) and prolonged MST (30.9 v 21.0 months; P = .013) compared with those in arm B. In addition, in arm A, patients who had resection of liver metastases had a significantly improved MST (46.4 v 25.7 months; P < .01) compared with those who did not undergo surgery.
Conclusion For patients with initially unresectable KRAS wild-type CLMs, cetuximab combined with chemotherapy improved the resectability of liver metastases and improved response rates and survival compared with chemotherapy alone.
INTRODUCTION
The liver is the most common site of metastasis in patients with colorectal cancer (CRC). Approximately half of patients with CRC develop hepatic metastases (LMs) during the course of disease,1 and LM is responsible for death in at least two thirds of those with CRC.2 In the absence of treatment, the prognosis of patients with LMs is dismal, with a 5-year survival rate approaching zero.3,4
Over the past decade, with the introduction of irinotecan- or oxaliplatin-based combination chemotherapy, the median survival among patients with colorectal LMs (CLMs) has increased from 6 to 8 months to > 20 months.5In recent years, drug development has focused on agents targeting molecular pathways that are expressed strongly or exclusively by tumor cells. One of the most promising targets is the epidermal growth factor receptor (EGFR), which is activated in colorectal tumors. As a strong EGFR inhibitor,6 cetuximab (Erbitux; Merck Serono, Darmstadt, Germany) has rapidly gained attention for the treatment of CLM.
Although numerous publications have reported the efficacy of cetuximab in combination treatment for patients with unselected metastatic CRC (mCRC), most of these studies have focused solely on addressing the impact of KRAS mutation status or various combined chemotherapy regimens on efficacy.7,8 In contrast, little work has concentrated on exploring the impact of this approach on the conversion to secondary resection of LM,9 which, if feasible, remains the only chance of cure for patients with CLMs.10,11 In addition, the patients with mCRC enrolled onto these cetuximab-based studies were primarily from Europe or North America rather than from East Asian countries (especially China), which might have led to trial bias in similar settings because of racial or genetic differences.
In this study, we compared first-line chemotherapy plus cetuximab with chemotherapy alone in Chinese patients with initially unresectable liver-limited KRAS wild-type synchronous mCRC. The principal goal was to evaluate the effects of cetuximab-containing treatment on the radical resectability of LM; however, tumor response and patient survival were also comparatively analyzed.
PATIENTS AND METHODS
Patient Eligibility
Patients were eligible for inclusion if their primary tumors had been resected and if they had histologically confirmed wild-type KRAS colorectal adenocarcinoma with synchronous liver-confined metastases deemed nonresectable by a local multidisciplinary team, which included > three liver surgeons and one radiologist. Other criteria for eligibility were age ≥ 18 and ≤ 75 years, evidence of tumor EGFR expression, and one measurable tumor. Included patients also had Eastern Cooperative Oncology Group performance status of 0 to 1, life expectancy ≥ 3 months, and adequate hematologic, hepatic, and renal function.
Patients were excluded if they had previously been exposed to any targeted therapy, chemotherapy, radiotherapy, or intervention therapy for mCRC. Patients who had suspected brain metastases or other cancers (with the exception of squamous cell carcinoma of the skin and cervical cancer in situ) within the previous 5 years were also excluded.
In this controlled trial, patients who met these eligibility criteria were randomly assigned to chemotherapy plus cetuximab (arm A) or chemotherapy alone (arm B). The choice of chemotherapy (mFOLFOX6 [modified fluorouracil, leucovorin, and oxaliplatin] or FOLFIRI [fluorouracil, leucovorin, and irinotecan]) was decided before random assignment. All patients provided written and oral informed consent.
Treatment
Treatment was planned to commence between 2 and 4 weeks after primary surgery. For arm A, on day 1 of a 14-day treatment cycle, patients received cetuximab once per week (with an initial loading dose of 400 mg/m2 and thereafter 250 mg/m2) or once every 2 weeks12 (performed after 2011; 500 mg/m2 on day 1 and once every 2 weeks thereafter) followed after 1 hour by either mFOLFOX6 (day 1: oxaliplatin 85 mg/m2, folinic acid 400 mg/m2, and fluorouracil 400 mg/m2 intravenous bolus, then 2,400 mg/m2 over 46 hours continuous infusion) or FOLFIRI (day 1: irinotecan 180 mg/m2 and folinic acid and fluorouracil, administered as with mFOLFOX6). In arm B, patients received the same regimen of mFOLFOX6 or FOLFIRI alone.
Treatment continued until tumor response indicated suitability for surgery for LM or until disease progression or unacceptable toxicity. In the event of predefined toxicity related to chemotherapy or cetuximab, protocol-specified treatment modifications were permitted. Patients with progressive disease were treated at the discretion of the investigators.
Trial Design
The primary end point was the conversion rate to radical resection for LM, which was assessed by multidisciplinary team after four cycles and then every other two cycles up to 12 cycles. LM resectability was determined using four criteria13–15: one, ability to obtain a complete resection (negative margins); two, preservation of two contiguous hepatic segments; three, preservation of adequate vascular inflow and outflow as well as biliary drainage; and four, ability to preserve adequate future liver remnant (> 20% in healthy liver). Nonresectability was defined as not meeting any of these criteria. To provide an objective assessment of changes in resectability, radiologic images were presented by a radiologist to > three liver surgeons, who were blinded to the clinical data. Patients were considered to have resectable disease if ≥ 50% of surgeons voted for radical resection of LM. For patients whose LMs were assessed resectable, resection was scheduled to be performed within 2 to 3 weeks of the last treatment cycle. After resection, patients were advised to continue the same therapeutic regimen until treatment reached a total of 12 cycles. Disease-free survival was calculated from the date of liver surgery to date of disease recurrence or last tumor assessment.
Secondary outcomes included tumor response, overall survival (OS), and progression-free survival (PFS). OS and PFS were further analyzed separately by incidence of LM resection. A retrospective subgroup analysis was used to investigate the associations between PFS, OS, or response rate and the BRAF gene mutation status of tumors or whether the patient had undergone hepatectomy. Adverse events were categorized according to the National Cancer Institute Common Toxicity Criteria, version 3.0.
In this trial, the assessable population was defined as patients who received the first four cycles of study treatment beyond the first tumor assessment. Tumor response was assessed by multidisciplinary team with the use of contrast-enhanced computed tomography or magnetic resonance imaging (and optional positron emission tomography scan), according to RECIST criteria.16,17 When a complete (CR) or partial remission (PR) was observed, a confirmatory assessment was performed 4 weeks later. The total number of CRs and PRs was reported as the overall response rate (ORR), whereas the disease control rate included CRs, PRs, and stable disease. PFS was defined from date of random assignment to date of disease progression or death. Patients without disease progression who discontinued the study for any reason were censored at the last on-study tumor assessment date. OS was calculated from random assignment to death resulting from any cause or date of last follow-up, at which point the data were censored. The protocol was approved by the local ethic committees and was registered with ClinicalTrials.gov.
Statistical Methods and Considerations
Patient baseline characteristics and disease factors were summarized using descriptive statistics. The categorical parameters were compared using two-sided Pearson's χ2 test or Fisher's exact test, as appropriate. All summary statistics on time-to-event variables were calculated according to the Kaplan-Meier method and were compared by means of the log-rank test. SPSS software (version 16.0; SPSS, Chicago, IL) was used for statistical analyses. A P value < .05 was considered significant.
RESULTS
Baseline Patient Characteristics
From June 2008 to December 2011, 204 patients were enrolled at Zhongshan Hospital. After screening, 138 patients were randomly assigned (70 in arm A; 68 in arm B) and accordingly formed the intent-to-treat (ITT) population. Of the ITT population, 22 patients were considered early dropouts because they discontinued study treatment within the first four treatment cycles, and they were observed every 2 to 3 months in the study. Thus, the assessable population consisted of 116 patients (59 in arm A; 57 in arm B; Fig 1). There were no major imbalances between the two groups in terms of baseline characteristics18,19 (Table 1). The cutoff date for survival data was June 2012, with a median potential follow-up time for the entire cohort of 25.0 months.
CONSORT diagram. (*) mFOLFOX6 (modified leucovorin, fluorouracil, and oxaliplatin) or FOLFIRI (leucovorin, fluorouracil, and irinotecan).
Baseline Patient Demographic and Clinical Characteristics (ITT population)
Efficacy
For the ITT population, the median PFS, median survival time (MST), and 3-year OS rate were 7.3 months, 24.4 months, and 30%, respectively. After evaluation of resectability by multidisciplinary team, 20 patients in arm A and nine in arm B were determined to be eligible for radical surgery for LM. However, three patients (one in arm A; two in arm B) refused further surgical intervention, and for another three patients (one in arm A; two in arm B), R0 resection could not be obtained at exploration. Ultimately, 18 patients in arm A and five in arm B achieved an R0 resection. None of the patients who were early dropouts underwent liver surgery. Nevertheless, the rate of resection with curative intent was significantly higher in the cetuximab-containing treatment group than in the chemotherapy-alone group (odds ratio, 4.37; P < .01).
All 18 patients who underwent radical resection for LM in arm A had objective responses. The median number of treatment cycles administered before surgery was six (range, four to 12). Of these 18 patients who underwent R0 hepatic resection, 11 (61.1% of all patients) had multiple wedge resections, four (22.2%) had right hepatectomies, and three (16.7%) had left hepatectomies. There were no serious complications during the perioperative period except for mild abnormalities of liver function in eight patients. At a median follow-up of 37.0 months, the median DFS and MST were 10.7 and 46.4 months, respectively. A total of 12 patients (66.7%) experienced relapse; recurrence primarily involved the hepatic remnant (eight [66.7%] of 12 patients). Other sites of recurrence included lung (two [16.7%] of 12 patients) and abdomen/peritoneum (two [16.7%] of 12 patients). Among those patients with liver- or lung-only recurrent disease, three underwent another metastasectomy and continued without recurrence. At the cutoff date, six of these 18 patients had died, and the MST was 36.0 months.
Furthermore, patients in arm A experienced greater survival benefit than those in arm B in terms of OS (3-year survival, 41% v 18%; median, 30.9 v 21.0 months; hazard ratio, 0.54; P = .013) and PFS (median, 10.2 v 5.8 months; hazard ratio, 0.60; P = .004; Fig 2). Among patients who did not undergo liver surgery, those in arm A also experienced increased survival benefit compared with those in arm B, with regard to MST (P = .050) and PFS (P = .028). However, among patients undergoing liver surgery, there were no statistically significant differences between arm A (n = 18) and arm B (n = 5) in terms of OS and PFS (Table 2).
(A) Overall survival curves and (B) Kaplan-Meier progression-free survival for the intent-to-treat population in the cetuximab plus chemotherapy and chemotherapy-alone treatment arms. HR, hazard ratio.
Efficacy and Adverse Events (ITT population)
Subgroup Analysis
Patients undergoing liver surgery had a significantly improved MST compared with those who did not undergo surgery, whether in arm A (46.4 v 25.7 months; P = .007) or arm B (36.0 v 19.6 months; P = .016). For subsets of populations in arm A (Table 3), regardless of whether the combined chemotherapy regimen was mFOLFOX6 or FOLFIRI, neither treatment regimen yielded a significant survival benefit. The association between BRAF mutation status and cetuximab-based treatment, with regard to ORR and median PFS, was not significant (P > .05). Ten patients in arm A harboring mutated BRAF tended toward worse MST (P = .043); this was not certified in arm B (P = .45).
Subgroup Analysis
Safety Analyses
Overall, the observed toxicity was mostly mild in both arms, and no deaths were attributable to cetuximab. Grades 3 and 4 toxicities are listed in Table 2, with no significant differences between the arms. For patients who underwent secondary surgery, our study revealed no serious liver toxicities from chemotherapy or cetuximab during the perioperative period.20 With cetuximab plus chemotherapy, there was a higher incidence of acne-like rash, but this effect was reversible in all patients. Moreover, regarding the assessable population, patients with grade 2 to 3 skin reactions experienced greater benefit than those with grade 0 to 1 skin reactions in terms of ORR and MST, although the latter was not significant (Table 3).
DISCUSSION
For patients with CLMs, when complete removal of liver-limited metastases can be achieved, reported 5-year OS rates can reach ≥ 55% under optimal conditions in specialized settings.10 However, < 20% of patients with LMs are candidates for surgery.21 As such, the current goal of medical treatment for initially unresectable CLM is to maximize the rate of potentially curative resection, as recommended by the National Comprehensive Cancer Network (version 1.2012). To reach this aim, over the past 10 years, one strategy has been to treat these patients with neoadjuvant systemic chemotherapy, also referred to as downsizing chemotherapy.21 Although traditional chemotherapeutic agents (including irinotecan and oxaliplatin) for CLM have demonstrable capacity to reduce tumor burden in the neoadjuvant setting,21 the rate of secondary liver resection remains poor (3.3%).22 As our data from arm B demonstrate, chemotherapy alone yielded a low rate of metastasectomy (7.4%). With the introduction of cetuximab, a higher rate of metastasectomy was achieved for CLM with this neoadjuvant strategy. For instance, in the CRYSTAL (Cetuximab Combined With Irinotecan in First-Line Therapy for Metastatic Colorectal Cancer) trial,8 the addition of cetuximab to FOLFIRI resulted in an increase in the resection rate from 4.5% to 9.8% in a subgroup of patients with liver-only disease. Similarly, in the OPUS (Oxaliplatin and Cetuximab in First-Line Treatment of mCRC) study,23 with FOLFOX plus cetuximab, the resection rate for LM doubled from 2.4% to 4.7%. It must be noted that in both studies, decisions to offer secondary liver resection were made in many cases by general oncologists, without the input of experienced liver surgeons, so it may be that many resectable patient cases were missed.24 In our trial, where entry and end points were determined by liver surgeons, the cetuximab combination yielded a high secondary resectable rate (25.7%) compared with that in the chemotherapy-alone cohort (7.4%). These results are in agreement with reported resection rates for patients in similar settings.25
For 18 patients in arm A and five patients in arm B who reached the goal of radical metastasectomy, there were few postoperative complications, a reflection of the multidisciplinary team and the experienced hepatic surgeons. The multidisciplinary team approach has been mandatory for CLM since 2005 within the trial center and meets weekly, providing objective assessments of changes in resectability. The hepatic surgeons, members of this multidisciplinary team, perform more than 2,000 hepatic surgeries each year in their liver center and have accumulated a wealth of experience in the resection of LM.
Several large studies have provided evidence that cetuximab-based anticancer agents can increase MST from 20.0 to 23.5 months or from 21.0 to 24.9 months in patients with mCRC.7,8 Our data indicate that MST was significantly prolonged (from 21.0 to 30.9 months) with the addition of cetuximab and improved more than in previous trials in similar settings, which was partially because of the high converted metastasectomy rate in arm A. The patients who underwent liver surgery achieved a longer MST (46.4 months). Regarding tumor response, the ORR in arm A (57.1%) was consistent with previously reported data in patients with wild-type KRAS mCRC, showing rates of 59% to 61%.7,23
Recently, the combination regimen of FOLFOX and cetuximab was removed as a treatment option in the National Comprehensive Cancer Network guideline (version 1.2012) for mCRC because of recent European and US studies.26–28 One of these was the COIN (Cetuximab Combined With Oxaliplatin-Based Chemotherapy in First Therapy for Advanced CRC) trial,26 which reported no OS (17.9 v 17.0 months; P = .67) or PFS benefit (8.6 months in both groups; P = .60) for patients with wild-type KRAS when cetuximab was added. However, there are concerns that many patients in the COIN study underwent significant dose reductions during treatment because of adverse events, so the full therapeutic benefit of the study arm may not have been realized. Our study indicated possibly improved MST and PFS after adding cetuximab to mFOLFOX6 (Table 2); however, these comparisons and conclusion should be interpreted with caution because of the small sample size.
Although tumor response to cetuximab is restricted to wild-type KRAS tumors,8,29,30 whether patients with only a BRAF wild-type tumor benefit from cetuximab remains ambiguous.31 According to our study, patients with wild-type BRAF status in arm A (n = 10) seemed to experience greater benefit in MST but not in PFS or ORR compared with those with a BRAF mutation (n = 60). Thus, BRAF mutation should not be used to exclude patients from cetuximab therapy; however, the data should be interpreted carefully because of the limited number of patients in this study, which is by no means conclusive.
There was no evidence to suggest that cetuximab increased the frequency or severity of the known toxicities of oxaliplatin, irinotecan, or fluorouracil. One concern was that cetuximab-containing chemotherapy might lead to inhibition of posthepatectomy hepatic regeneration.20 Although the results of our study did not substantiate this concern, it is important to anticipate and minimize the potential liver toxicities of the systemic treatment used for tumor shrinkage. Furthermore, in agreement with previous studies,8,29 we found that higher-grade skin toxicities were associated with an improved ORR. Although the mechanism of this relationship has not been fully clarified, skin toxicity may be currently regarded as a surrogate marker of therapeutic effectiveness for cetuximab.
Our study has other limitations. First, the number of patients analyzed was limited, and follow-up time was short. Therefore, certain subgroup analyses are unconfirmed, and the 5-year OS rate has not yet been reached. Second, cetuximab is a high-cost treatment not approved for reimbursement in China. On one hand, patients treated with additional cetuximab mostly had better economic circumstances than those treated with chemotherapy alone; on the other hand, the financial burden led to treatment discontinuation (seven patients in arm A abandoned the cetuximab-based regimen for economic factors) and discouraged patients who wanted to receive cetuximab, which might have resulted in minimal study bias.
In summary, despite these drawbacks, our study confirmed that for Chinese patients with initially unresectable CLMs (wild-type KRAS), the addition of cetuximab to FOLFIRI or mFOLFOX6 was associated with facilitated radical resections of LMs with minimal perioperative complications, providing improved long-term survival. Further investigation in this group of patients, if possible, could shed additional light on performing a multi-center clinical study in China.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The author(s) indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Conception and design: Li Ren, Jianmin Xu
Financial support: Jianmin Xu
Administrative support: De-Xiang Zhu, Jianmin Xu
Provision of study materials or patients: Li Ren, Ye Wei, Sheng-Yong Zai, Bo Xu, Jianmin Xu
Collection and assembly of data: Le-Chi Ye, Tian-Shu Liu, Li Ren, Ye Wei, De-Xiang Zhu, Sheng-Yong Zai, Yiyi Yu, Bo Xu, Jianmin Xu
Data analysis and interpretation: Le-Chi Ye, Tian-Shu Liu, Ye Wei, De-Xiang Zhu, Qing-Hai Ye, Xin-Yu Qin, Jianmin Xu
Manuscript writing: All authors
Final approval of manuscript: All authors
Footnotes
-
Processed as a Rapid Communication manuscript. See accompanying editorial on page 1913
-
Supported by Key Projects of the Clinical Disciplines, administered by the Ministry of Health.
-
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
-
Clinical trial information: NCT01564810.
