All patients provided written informed consent. Main inclusion criteria were age ≥ 18 years, Karnofsky performance status (KPS) ≥ 80, histologically confirmed diagnosis of adenocarcinoma of the colon or rectum, metastatic disease not suitable for curative treatment, immunohistochemical evidence of EGFR expression in the primary tumor or metastasis, previous treatment with 1 of 4 pre-defined irinotecan regimens for ≥ 6 wk as the most recent chemotherapy, disease progression on or within 3 mo of the most recent irinotecan regimen, able to tolerate irinotecan continuation, no more than 3 previous lines of chemotherapy, and adequate hepatic, renal and bone marrow function. Main exclusion criteria were known or suspected brain metastases, radiotherapy or major surgery within the 4 wk prior to study entry, previous EGFR-targeted therapy, coronary heart disease or history of myocardial infarction or uncontrolled arrhythmia, other malignancy within 5 years (except basal cell carcinoma of the skin or pre-invasive carcinoma of the cervix), and pregnancy.

This was an open-label, uncontrolled, multicenter, phase II study carried out in 25 centers in 8 countries in the Asia-Pacific region. Patients initially received an infusion of cetuximab 400 mg/m² over 2 h followed by weekly infusions of 250 mg/m² over 1 h in addition to irinotecan treatment at the same dosage regimen (including previous dose reductions) on which the subject had become refractory to irinotecan. Irinotecan was administered intravenously over 30-90 min with at least 1 h between the end of cetuximab and the start of infusion of irinotecan. The irinotecan regimens allowed as pre-study treatment and continued during the study were 100 or 125 mg/m² weekly for 4 wk then 2 wk rest (1 cycle); 100 or 125 mg/m² weekly for 2 consecutive wk out of 3 (1 cycle); 180 or 210 mg/m² every 2 wk (1 cycle); and 300 or 350 mg/m² every 3 wk (1 cycle). The first 3 pre-study irinotecan regimes could have been administered either as single-agent irinotecan or in combination; however, only single-agent irinotecan was accepted for the fourth regimen. Study treatment was continued until disease progression or unacceptable toxicity. In case of an irinotecan-related toxicity, irinotecan could be stopped and treatment could be continued with cetuximab monotherapy. Two irinotecan dose reductions were allowed for toxicity prior to study entry and during the study for all regimens except for patients receiving 100 mg/m² weekly where only 1 dose reduction was allowed.

The primary objective was the assessment of PFS rate after 12 wk of treatment. Secondary objectives included further evaluation of PFS, and the assessment of best overall confirmed response rate, duration of response, time to treatment failure (TTF), overall survival time, safety and toxicity.

The study was conducted in accordance with the ethical principles of the Declaration of Helsinki as well as the International Conference on Harmonization-Good Clinical Practice (ICH-GCP, 1996).

Pre-screening of patients was performed within 3 mo after confirmation of disease progression on irinotecan-based chemotherapy and included informed consent, collection of demographic data, tumor diagnosis and the determination of tumor EGFR expression which was performed by designated pathology laboratories in Hong Kong, India and China by immunohistochemical staining of tumor tissue slides using a standardized protocol with the DAKO EGFR pharmDX kit (DAKO Corporation, Carpinteria, CA, USA). Baseline evaluations were performed within 21 d prior to the start of study treatment; pre-study staging by computed tomography (CT) or magnetic resonance imaging (MRI) was performed within 28 d. After the start of treatment, imaging (CT or MRI) was performed every 6 wk. Based on these images, the tumor response was assessed according to modified World Health Organisation (WHO) criteria every 6 wk during study treatment. Follow-up for survival occurred every 12 wk from the end of the study visit and included recording of subsequent treatments.

Adverse events (AEs) were reported weekly at each infusion visit, at the time of disease progression or study discontinuation, and up to 6 wk after the last infusion of study treatment. AEs were reported by severity according to National Cancer Institute Common Toxicity Criteria (NCI-CTC) version 2 using Medical Dictionary for Regulatory Activities (MedDRA) version 10 preferred terms as event category and MedDRA body system/primary system organ class (SOC) as SOC category.

Safety and efficacy analyses were carried out considering the intention-to-treat (ITT)/safety population, which comprised all patients who had received any dose of study treatment. As patients were not randomized to treatment and the number of patients per individual treatment regimen was expected to be small, all data were analyzed independently of the on-study irinotecan regimen. The primary objective of this study was to determine the PFS rate at 12 wk after initiation of cetuximab treatment. Time-to-event analyses were based on Kaplan-Meier estimates. Confidence intervals (CIs) for the median were calculated according to Brookmeyer and Crowley[16]. CIs for the event-free rate estimates, including the primary variable, were derived from the Kaplan-Meier curve at defined time points. The estimate of the standard error was computed using Greenwood’s formula[17]. PFS was defined as the time from the first study medication to the first observation of radiologically confirmed disease progression, symptomatic deterioration leading to discontinuation of study treatment (unless imaging confirmed absence of progressive disease), or death due to any cause within 60 d of the last on-study tumor assessment. The best overall confirmed response was defined as the best confirmed response from start of treatment until disease progression. Duration of response in patients with a complete response (CR) or partial response (PR) was defined as the time from first assessment of CR/PR to the first time disease progression was documented or death within the trial or until death within 60 d after last tumor assessment, whichever occurred first. The time to TTF was defined as the time from the start of the first cetuximab treatment until the date of the first occurrence of an event defining treatment failure, i.e. withdrawal of any study treatment due to AEs or progressive disease or withdrawal of consent or death, whichever occurred first. The overall survival time was defined as the time from the day of first infusion of study medication to death. For subjects who were still alive at the last survival follow-up, or who were lost to follow-up, survival was censored at the last recorded date that the subject was known to be alive.

A subgroup analysis that used a landmark method for the analysis of time-dependent covariates was carried out to examine whether the development and severity of acne-like rash that occurred during the first 21 d of treatment was associated with prolonged survival time. To avoid selection bias as a consequence of the early exclusion of patients, this analysis was limited to all patients in the ITT/safety population who were still under on-study treatment at day 21. Consequently, overall survival time was analyzed for this subgroup by using the landmark of study day 22 (and not day 1) as the start day. Survival was further analyzed in other subgroups based on known prognostic variables.

To ensure adequate precision of estimates, the sample size determination was based on a confidence limit approach. The expected rate of progression-free patients 12 wk after start of study treatment was 50%. It was calculated that with the enrollment of 150 patients, the observed 95% CI for this point estimate would have a range of approximately ± 8%.

Patient characteristics are summarized in Figure 1. Between November 2005 and September 2006, 188 patients were screened at 25 centers in Australia (3 centers), China (6), Hong Kong (2), India (3) Korea (5), Singapore (1), Taiwan (3), and Thailand (2). Of these patients, 145 (77%) had EGFR detected in their tumor sample (data were missing for 4 patients), 129 patients were enrolled and of these, 123 patients were treated and formed the ITT/safety population. The main reason for study discontinuation was disease progression (86.2%). There were 6 patients who were treated until death (5 of these deaths due to disease progression, 1 due to chemotherapy-related AEs, 4 patients stopped treatment due to AEs and 5 patients withdrew consent.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Disposition of patients at each stage of the study. EGFR: Epidermal growth factor receptor; PD: Progressive disease.

Patient and disease characteristics at baseline are summarized in Table 1. The majority of patients were male (59.3%), the median age was 56 years (range, 24-78 years) and 74.8% of patients were aged < 65 years. The most common primary tumor site was the colon (57.7%), and 74.8% of patients had liver metastases, 56.1% had lung metastases and more than 30% of patients had other organs involved. The median duration of mCRC at baseline was 8.8 mo (range, 1-56).

The most common recent irinotecan regimen used was 180 mg/m² every 2 wk, which had been used in 93 patients. Approximately half of the patients (49.6%) had a time interval of 1-3 mo between the last pre-study dose of irinotecan and the first dose of cetuximab and the interval was < 1 mo in 35% of patients. All other patients had an interval of less than 6 mo between the most recent irinotecan dose and the start of study treatment.

Treatment exposure is summarized in Table 2. At least 50% of patients received cetuximab treatment for ≥ 12 wk and were administered ≥ 12 infusions. The longest treatment duration was 98 wk and almost 5% of patients received more than 50 infusions of cetuximab. Relative dose intensity (RDI) of cetuximab of ≥ 90% was achieved in 89.2% of patients. Furthermore, 118 patients (95.9%) were treated at the planned standard dose of cetuximab. Four patients (3.3%) had at least 1 dose reduction of cetuximab and information was missing for 1 patient. Dose reductions were required at the occurrence of second or third incidences of grade 3 skin toxicity according to the protocol.

The median duration of irinotecan exposure was 12 wk, with patients receiving a median of 6 cycles. RDI of ≥ 90% of irinotecan was achieved in 65% of patients. Approximately three-quarters of patients (74%, n = 91) received ≥ 80% of the planned dose of irinotecan. In total, 74 patients (60.2%) required at least 1 dose reduction of irinotecan, which included pre-study dose reductions.

Study results are summarized in Table 3. The 12-wk PFS rate was found to be 50% (95% CI: 41-59). The median PFS time was 12.1 wk (95% CI: 9.7-17.7) (Figure 2A). In general, there were no marked or unexpected differences in PFS time for any of the prognostic variables or characteristics of previous treatment which were considered in the pre-specified subgroup analyses (data not shown).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Kaplan-Meier curve. A: Progression-free survival (PFS); B: Overall survival (OS) in the intention-to-treat/safety population. CI: Confidence interval.

The median overall survival time was 9.5 mo (95% CI: 7.5-11.6) (Figure 2B), with a 1-year survival rate of 38% (95% CI: 29-46) (Table 3). The median follow-up time for survival was 22.4 mo (95% CI: 21.4-24.4). There were generally no marked or unexpected differences for any of the prognostic variables or characteristics of previous treatment which were considered in pre-specified subgroup analyses of overall survival time (Table 4). A landmark analysis investigated the impact of early occurrence of acne-like rash (during the first 21 d of treatment) on overall survival time. Median overall survival was longer in patients with acne-like rash compared with those patients who experienced no acne-like rash (9.5 mo vs 6.2 mo). Median overall survival time was shorter in patients with grade 3/4 early acne-like rash (4 patients only) compared with grade 1/2 acne-like rash. As the sample size in the subgroup with grade 3/4 acne-like rash was very small, this observation is likely to be an artifact due to differences between patients with and without early acne-like rash with regard to other prognostic factors, such as gender, age and KPS.

The objective response rate was 13.8% (95% CI: 8.3-21.2) (Table 3). The median duration of response was 31.1 wk (95% CI: 18.0-42.6). In addition, 44 patients (35.8%) had stable disease and the disease control rate was 49.6% (95% CI: 40.5-58.8). A higher response rate of 25.0% (5/20 patients) was observed for patients with CR and PR as best response to the most recent irinotecan schedule as compared with subjects with progressive disease and stable disease who had a response rate of 11.8% (12/102 patients). In addition, patients with KPS ≤ 80 at baseline had lower response rates [7.7% (95% CI: 1.6-20.9)] than the KPS > 80 subgroup [16.7% (95% CI: 9.4-26.4)]. There were generally no marked or unexpected differences in response rates between other pre-specified subgroups (data not shown). Disease control rates in the subgroups were consistent with those observed for response rate. The median TTF was 11.7 wk (95% CI, 9.1-17.4).

Ninety patients (73.2%) received post-study anti-cancer treatments, consisting of best supportive care (66 patients; 53.7%), chemotherapy (59 patients, 48.0%), surgery (6 patients, 4.9%), radiotherapy (15 patients, 12.2%) and other treatments (15 patients, 12.2%).

Treatment was generally well tolerated. The most common AEs of any grade according to MedDRA SOC were skin and subcutaneous tissue disorders (85.4%), gastrointestinal disorders (85.4%) and general disorders and administration site conditions (44.7%). The most common AEs according to MedDRA preferred terms occurring in at least 20% of the patients were diarrhea (71 patients, 57.7%), rash (70 patients, 56.9%), nausea (55 patients, 44.7%), vomiting (48 patients, 39.0%), anorexia (35 patients, 28.5%), neutropenia (30 patients, 24.4%), constipation (27 patients, 22.0%) and stomatitis (26 patients, 21.1%). Grade 3/4 AEs occurred in 68 patients (45.5%) (Table 5). The most common grade 3/4 AEs were diarrhea in 17 patients (13.8%), neutropenia in 11 patients (8.9%) and rash and vomiting both in seven patients (5.7%) (Table 5). Composite AE categories were considered for AEs known to be related to cetuximab, namely infusion-related reactions and acne-like rash (Table 5). Grade 3/4 infusion-related reactions occurred in 3 patients (2.4%; 2 patients with Grade 3, 1 patient with Grade 4). Ten patients (8.1%) had grade 3 acne-like rash (Table 4), with no occurrences of grade 4 rash reported. Overall, 7 patients (5.7%) had AEs that led to the discontinuation of cetuximab and 13 patients (10.6%) had AEs leading to the discontinuation of chemotherapy. Three patients (2.4%) had AEs leading to the discontinuation of both.

At the time of database closure (February 27th, 2009), 105 patients in the ITT/safety population had died, of which 8 died within 30 d after the last dose of study medication. Seven deaths (5.7%) occurred due to disease progression and 1 patient (0.8%) died due to events related to irinotecan-based therapy. There were no deaths related to cetuximab treatment.

The incidence and mortality rates for colorectal cancer (CRC) in Eastern and South-East Asia are increasing. Patients with metastatic CRC (mCRC) who progress following first-line therapy have few treatment options. This open-label, phase II study evaluated the efficacy and safety of cetuximab plus irinotecan in irinotecan-refractory mCRC patients from South-East Asia and Australia.

Research in the field of mCRC has recently focused on the development of targeted therapies in disease treatment. Epidermal growth factor receptor (EGFR) is frequently expressed in mCRC and is associated with tumor growth and metastasis. Cetuximab is an EGFR-targeted monoclonal antibody which binds EGFR inhibiting its activity. Registration of cetuximab was based on the pivotal BOND trial in which patients with mCRC progressing on an irinotecan-containing regimen achieved a higher overall survival rate when cetuximab was added to irinotecan compared with irinotecan alone. The subsequent Monoclonal Antibody Erbitux in a European Pre-License Study (MABEL) and Latin American Erbitux Pre-License Study (LABEL) studies confirmed the efficacy of this treatment combination in a community-based setting.

The safety and efficacy of cetuximab in combination with irinotecan in mCRC patients failing on irinotecan-containing regimens has been demonstrated in phase II studies. The ELSIE study supports these findings in patients from the Asia-Pacific region. Median overall survival was comparable with that reported in the LABEL and MABEL studies and the combination arm of the BOND study. There were slightly lower tumor response rates, median PFS and PFS rate at 12 wk in the ELSIE study. Patients were unselected in the ELSIE and the observed treatment benefit may be greater in patients selected for KRAS wild-type tumors.

The results of this study confirm the activity of cetuximab in combination with irinotecan in patients with mCRC from Asia and Australia. In the future cetuximab-containing regimens are likely to be tailored to selected patients in this setting with KRAS wild-type tumors.

The paper on the whole is well written and the abstract reflects the basic design and findings.