The present study is a phase II, single-arm, multicenter study of panitumumab and irinotecan combination therapy. Approval for the study was obtained from the Ethics Committee of Jichi Medical University (Shimotsuke, Japan) and each facility involved, and written informed consent was obtained from all enrolled patients (trial ID, UMIN000004500).

The inclusion criteria were as follows: Histological diagnosis of unresectable advanced and/or recurrent colorectal cancer; presence of a measurable target lesion; absence of KRAS mutation in exon 2 (codons 12 and 13); disease progression or intolerance to irinotecan- and oxaliplatin-based therapy, and had previously received fluoropyrimidines; Eastern Cooperative Oncology Group performance status of 0–2; age ≥20 years; neutrophil count ≥1,000/mm³; platelet count ≥100,000/mm³; total bilirubin level ≤1.5 mg/dl; aspartate transaminase (10–35 U) and alanine transaminase levels (5–40 U) ≤2.5 times the facility's upper limit of normal (≤5 times the respective facility's upper limit in the case of hepatic metastasis); serum creatinine level ≤2.0 mg/dl; at least 2 weeks having passed with no treatment since completion of any of the following treatments: and radiotherapy, surgical treatment with organ resection, chemotherapy, hormone and immunotherapy. The presence or absence of a UDP-glucuronosyltransferase 1–1 measurement was not taken into consideration.

Panitumumab (Takeda Pharmaceutical Company, Tokyo, Japan) was administered at a dose of 6 mg/kg every 2 weeks by intravenous infusion over 60 min, and irinotecan was administered at a dose of 100–180 mg/m² every 2 weeks by intravenous infusion over 90 min, depending on the preceding treatment dose. As prophylaxis for anti-emesis, topical steroids and 5-HT3 receptor antagonists were administered prior to the chemotherapy regimen. This treatment was continued until disease progression, intolerable adverse events and/or upon patient refusal of treatment. If a treatment course could not be initiated due to a neutrophil count <1,000/mm³, a platelet count <75,000/mm³ and/or other non-hematological toxicity, the treatment was deferred or the dose decreased at the discretion of the attending physician. In the event irinotecan could not be continued due to adverse events, panitumumab monotherapy was continued. To prevent skin toxicities, a moisturizer was applied to the face, arms, legs, neck, back and chest every morning after waking up; sunscreen was applied prior to stepping outside; a topical steroid was applied to the face, arms, legs, neck, back, and chest at bedtime; and oral antibiotics (100 mg minocycline twice daily) were initiated 24 h prior to the first course of treatment. All these treatments were continuously administered for 6 weeks. The decision to continue antibiotics after 6 weeks was left to the discretion of the attending physician.

Patients underwent regular examinations and blood tests as outpatients, and an examination by computed tomography or magnetic resonance imaging was performed every 8 weeks from chemotherapy initiation. The primary endpoint of the present study was the response rate (RR) determined by two independent reviewers. The antitumor effect was determined using the Response Evaluation Criteria in Solid Tumors v. 1.0 (16) by the attending physician, researchers and two independent radiologists. Secondary endpoints included the disease control rate (DCR), the progression-free survival (PFS) time, the OS time and adverse events. PFS time was defined as the period from chemotherapy initiation until progression, and OS time was defined as the period from chemotherapy initiation until mortality. Adverse events were evaluated using the National Cancer Institutes' Common Toxicity Criteria v. 3.0 (Japanese edition) (17). To ascertain each patient's requirement for prophylactic treatment of skin toxicities, patients were asked to keep a diary beginning 24 h prior to the first treatment day until week 6.

With the RR threshold set at 10% and the anticipated RR at 25%, 41 patients with measurable responsiveness were required for the α error to be <0.10 (one-sided) when the true RR was less than the RR threshold and for the statistical power to be >90% when the true RR was above the anticipated RR (18,19). Considering that certain patients would be excluded from the analyses, the aim was to enroll 45 patients. PFS and OS were estimated on a survival curve using the Kaplan-Meier method, and the two-sided 95% confidence intervals (CIs) of the standard deviation and median value were calculated using JMP^® software (SAS Institute Inc., Cary, NC, USA).

A total of 37 patients from 6 Japanese medical institutions were enrolled between October 2010 and March 2012. The institutions were as follows: Tochigi Cancer Center (Utsunomiya, Japan), Jichi Medical University Hospital (Shimotsuke, Japan), Chiba Cancer Center (Chiba, Japan), Saitama Cancer Center (Saitama, Japan), Ibaraki Central Hospital (Mito, Japan) and Saku Central Hospital (Saku, Japan). Delays in the enrollment process made it difficult for data to be maintained by the study coordinator; as a result, enrollment was terminated. Of the 37 patients enrolled, 1 patient was excluded due to the lack of a target lesion, and 1 patient was excluded due to the inability to receive treatment as a result of disease progression. As a result, the chemotherapy regimen was administered to 35 patients in total (Fig. 1). Patient characteristics are shown in Table I; the median age was 61 years (range, 41–76 years), with 25 male and 10 female patients. The initial irinotecan dose was 150 mg/m² in 19 patients (54%) and 180 mg/m² in 1 patient. The median number of treatment courses administered was 6 (range, 3–41).

A central review indicated that, of the 35 patients included, 8 patients (22.9%) presented with a partial response and 6 patients (17.1%) with stable disease, with an RR of 22.9% (95% CI, 12.1–39.0) and a DCR of 40% (Table II). An RR of 17.1% (95% CI, 8.1–32.7) and a DCR of 42.9% were recorded. The objective RR of the patients with standard-dose irinotecan (150 or 180 mg/m²) was 30.0% (6/20), although that of low-dose irinotecan (100–120 mg/m²) was 13.3% (2/15). In 2 patients, the chemotherapy was terminated due to surgery being performed. The median PFS time was 81 days (2.7 months; 95% CI, 71–101) and the median OS time was 189 days (6.3 months; 95% CI, 146–247) (Table II). The survival curve is shown in Fig. 2.

Diaries could be collected from 26 patients, among whom prophylactic treatment for a rash was properly administered in 22 patients (84.6%). Other diaries could not be collected as they had been lost, the patients had refused to complete them or as the patients' symptoms had deteriorated prior to collection. Adverse events are shown in Table III. A grade 3 or above acne-like rash developed in 25.7% of patients, paronychia in 11.4% and dryness in 8.6%. Recovery from the majority of adverse events was possible with the appropriate cessation of the chemotherapy and active interventional therapy. In total, 1 patient succumbed 15 days after the final course of treatment. In this patient, a causal association between mortality and treatment could not be ruled out. In addition, the chemotherapy regimen was discontinued by the attending physician in 4 patients, as these patients refused further treatment due to adverse events.