Panitumumab in combination with modified docetaxel/cisplatin/5-fluorouracil as first-line treatment in gastric and gastroesophageal junction adenocarcinomas: a multicenter phase II study by the Hellenic Oncology Research Group

Eligibility

Patients with histologically documented, inoperable, locally advanced or metastatic adenocarcinoma of the stomach and the GEJ, aged >18 years old, were eligible for the study. No prior chemotherapy for metastatic disease was allowed; adjuvant chemotherapy or chemo-radiotherapy was allowed provided that at least 6 months had elapsed since the completion of the treatment. Other key eligibility criteria were: absence of HER2 expression (score 1 by immunocytochemistry [ICH] or fluorescence in situ hybridization-negative for those with score 2 by ICH); measurable target lesions as defined according to the RECIST criteria (RECIST v.1.1) (patients with non-measurable disease could be enrolled only in the phase I part of the study); ECOG performance status 0-2, adequate bone marrow (defined by hemoglobin ≥8 g/dL, white blood cells ≥3×109/L, neutrophil count ≥1.5×109/L, platelets ≥100×109/L), renal (creatinine clearance ≥50 mL/min) and liver function tests (total bilirubin ≤1.5 times the institutional upper limit of normal, and aspartate/alanine transaminases and alkaline phosphatase ≤2.5 time the upper limit of normal); and estimated life expectancy >3 months. All patients gave written informed consent before their enrollment in the study. The protocol was approved by the Ethics and Scientific Committees of the participating institutions, as well as by the Greek National Organization for Medicines (EOF) and the National Ethics Committee (EED).

Treatment plan

The phase I part of the trial was a dose-escalation study designed to define the recommended dose of docetaxel in combination with 5-fluorouracil, Leucovorin cisplatin, and cetuximab for the second part (phase II) of the study. Patients recruited in the phase I trial were treated in cohorts of 3 (3+3 design), with escalating doses of docetaxel (30 mg/m2 in Level 1, 35 mg/m2 in Level 2 and 40 mg/m2 in Level 3), while the doses of cisplatin, 5-fluorouracil, and panitumumab remained fixed. More specifically, panitumumab was administered as a 1-h iv infusion at a dose of 6 mg/kg on days 1 and 14 of each treatment cycle; docetaxel was administered as a 1-h infusion at a dose of 30-40 mg/m2 on days 1 and 14 of each cycle; cisplatin was given at a dose of 40 mg/m2 as a 2-h infusion on days 1 and 14 of each cycle, after adequate hydration, diuresis with mannitol or furosemide and antiemetic prophylaxis, according to the policies of each participating center; leucovorin was administered at a dose of 400 mg/m2 on days 1 and 14, followed by bolus 5-fluorouracil infusion at a dose of 400 mg/m2 and continuous iv infusion of 1000 mg/m2 per day of 5-fluorouracil on days 1, 2, 14 and 15 of each cycle. Granulocyte-colony stimulating factor (G-CSF) support was administered from day 3-8. Treatment was administered every 2 weeks and one treatment cycle was completed on d28 (2 chemotherapy administrations). Maximum treatment duration was 6 (12 administrations) 4-week cycles. Patients with disease control (complete response [CR], partial response [PR], and stable disease [SD]) during the chemotherapy period received maintenance therapy with panitumumab (9 mg/kg every 3 weeks) until disease progression, intolerance, toxicity, graded according to the CTCAE v3.0 criteria [19], death, or consent withdrawal.

Three patients were enrolled at the first dose level and dose escalation proceeded if no dose-limiting toxicity (DLT) occurred in the first cycle (4 weeks). If DLT occurred in cycle 1, 3 additional patients would be enrolled to the same dose level. If less than 3 of 6 patients experienced DLT, the dose escalation would continue to the next dose level. The patients from the phase I part of the study who received the recommended doses for the phase II study were included in the phase II study and analyzed for efficacy. All patients received at least 6 chemotherapy cycles (12 administrations) at the doses initially allocated. There was no intra-patient dose escalation. All patients were followed until disease progression or death, to assess late or persistent toxicity.

Definition of DLT and maximum tolerated dose (MTD)

DLT was defined as any of the following events during the first treatment cycle: (i) hematological toxicity (febrile neutropenia, grade 4 neutropenia lasting ≥7 days, grade 3 thrombocytopenia lasting ≥7 days or grade 4 thrombocytopenia); (ii) non-hematological toxicity grade 3 or more with the exception of fatigue grade 3, which had to last for more than 7 days; (iii) failure to recover from related toxicities to grade ≤1 or baseline severity (or grade ≤2 at investigator and sponsor discretion) after delaying the next cycle up to 7 days; and (iv) failure to complete the first treatment course (≤75% of planned dose). MTD was defined as the highest dose level at which 2 out of 3 or 3 out of 6 patients experienced a DLT. The phase II part of the study would be initiated at the doses of the previous MTD level. If the MTD level corresponded to the doses of level 1 (i.e., 3 of 3, or >3 of 6 patients developing any LTD in the first cycle) the schedule and the doses of the study drugs would have to be revised.

Dose modifications

The panitumumab dose was withheld for severe skin or nail toxicities (skin toxicity requiring narcotics, systemic steroids, or felt to be intolerable by the subject, skin infection requiring iv antibiotic or iv antifungal treatment, need for surgical debridement, or any skin- or nail-related serious adverse event). Panitumumab was discontinued for subjects who missed more than 2 consecutive scheduled doses because of toxicity. In patients with grade 3 and 4 neutropenia, G-CSF support was recommended on days 8 -13 in subsequent cycles. In patients with febrile neutropenia and/or grade 3 or 4 neutropenia, despite the administration of G-CSF, or grade 3-4 thrombocytopenia, subsequent cycles were administered with a 20% dose reduction in the doses of all chemotherapeutic agents.

Regarding non-hematological toxicities, 5-fluorouracil was decreased by 50% for grade 3-4 hepatotoxicity, while 5-fluorouracil and docetaxel were decreased by 25% for grade 3 diarrhea and stomatitis. For grade 3 neuropathy the doses of cisplatin and 5-fluorouracil were decreased by 25%.

If more than 2-week delay was needed for complete hematologic recovery, despite the administration of G-CSF and/or predefined doses adjustments and/or in case of persistent grade 3 or 4 non-hematologic toxicity after the completion of the two weeks of delay, the patient was withdrawn from the study and was subsequently treated at the physician's discretion.

Statistical analysis

This was a phase I/II multicenter, single-arm trial of the combination mDCF plus panitumumab in patients with locally advanced or metastatic adenocarcinomas of the stomach and GEJ. The primary endpoint of the phase I part of the study was to determine the MTD of docetaxel when given in combination with cisplatin, 5-fluorouracil and panitumumab for the subsequent phase II study. The primary endpoint of the phase II trial was to assess the efficacy and the tolerance of the regimen and to investigate whether it could achieve an objective response rate of >40% (ECIST 1.1 criteria; [20]) Secondary endpoints included the safety profile, progression-free survival (PFS) and OS. The sample size was determined based on the overall response rate, according to Fleming's two-stage minimax design [21]. According to the design, a sample of 41 patients for the first step and 28 additional patients for the second step were needed in order to prove the initial hypothesis (at least 17 objective responses in the first 41 enrolled patients) at a statistical significant level of α=0.05 (one-sided) and with a power of 80%.

PFS was defined as the interval from the date of registration into the study until disease progression or death from any cause. OS was measured from the date of registration until death from any cause. The follow-up time was measured from the day of first treatment administration to the study's cutoff date or the date of death. The PFS and OS rates were estimated using the Kaplan-Meier method, and the confidence intervals (CI) for response rates were calculated using methods for exact binomial CIs. All efficacy and toxicity results were assessed for the enrolled patients on an intention-to-treat basis. To test significant associations between continuous variables, the Mann-Whitney test was used. The chi-square test was used to test for associations between response and other dichotomous variables. Statistical significance was set at P=0.05. Tolerability was graded according to the CTCAE v3.0 criteria [19].

Patient demographics

All patients in both phase I and II parts of the study were enrolled at 9 Greek hospitals. Between July 2010 and March 2012, 15 patients with locally advanced or metastatic gastric and GEJ carcinoma were enrolled in the 3 cohorts of the phase I study. Patients’ baseline demographics and clinicopathological data are presented in Supplementary Table S1. At the date of data cutoff, 14 patients had discontinued or completed treatment and one was still on treatment. Seven patients (46.7%) had completed study treatment, while five (33.3%) had discontinued treatment because of disease progression and two (13.3%) because of treatment-related adverse events.

Supplementary Table S1 Demographic data (Phase I study)

From April 2012 to January 2015, 35 additional patients were enrolled in the phase II part of the study, in addition to 6 patients from phase I who had received the recommended doses for phase II; thus, a pre-planned interim analysis was performed according to the statistical considerations. One patient did not fulfill the eligibility criteria and was excluded from the final analysis. The characteristics of patients enrolled in both the phase I and phase II parts of the study are summarized in Table 1.

Table 1 Demographic data (all patients: Phase I and II)

DLT and MTD determination

At the first dose level (docetaxel: 30 mg/m2) there was one DLT event (grade III transaminasemia); hence, 3 additional patients had to be enrolled at this dose level. In this second cohort of 3 patients one additional patient developed DLT (grade IV rash); however, the criteria for MDT (DLT at 2 of 3 or 3 of 6 patients) were not met; thus, dose escalation continued and 3 patients were enrolled at the second dose level (docetaxel: 35 mg/m2) with no DLT events. At the third dose level (docetaxel: 40 mg/m2) one patient presented febrile neutropenia, requiring hospitalization for intravenous antibiotics and G-CSF support; the patient recovered uneventfully after 4 days. Because of this DLT event, 3 additional patients were enrolled at this dose-level and an additional patient developed a DLT event (grade IV neutropenia); nevertheless, the MTD was not reached since DLT events occurred in only 2 of 6 patients. Therefore, the dose of 40 mg/m2 of docetaxel was recommended as MTD for the subsequent phase II part of the study.

Efficacy

In the phase II part of the study, 40 of the 41 enrolled patients were evaluable for response (one patient did not met the eligibility criteria for metastatic or locally advanced disease after surgery). Two patients who died during the first treatment cycle were included in the analysis and were considered as progressors. One (2.5%) CR and 13 (32.5%) PR were achieved (overall response rate: 35%), while 16 (40%) patients experienced SD and 10 (25%) progressive disease. Since less than 17 responses were recorded during the first step of the phase II part of the trial, the study was terminated due to futility.

At the time of analysis, 28 patients died because of disease progression and two from other causes. The median PFS was 6.9 months (95%CI 3.5-10.3) and the median OS 11.3 months (95%CI 7.7-14.8). There was no significant difference in terms of PFS and OS survival according to the different localization of the primary tumor (data not shown).

Treatment administration and toxicity

In the phase II part of the study, a total of 326 treatment cycles were administered with a median number of 10/patient (range, 1-13). Forty-nine cycles (15%) were delayed because of hematologic (5.5%; n=18 cycles) or non-hematologic (2.1%; n=7 cycles) toxicity and for other reasons unrelated to treatment (i.e., patients’ personal reasons, prescheduled imaging evaluation; n=26 cycles). The median time of delay was 7 days (range, 2-40 days). Dose reduction was required in 36 cycles (11%) because of hematologic (n=12 cycles) or non-hematologic (n=23 cycles) toxicity, or for an unknown reason (n=1 cycle; Table 2).

Table 2 Compliance with treatment

At the time of analysis, 39 patients (97.5%) had discontinued treatment for the following reasons: 16 patients (41%) completed treatment as per protocol (maximum treatment duration of 6 chemotherapy cycles followed by maintenance panitumumab until disease progression, toxicity, death or consent withdrawal in patients with no disease progression); 17 (42.5%) showed disease progression; four (10.3%) withdrew consent; and two (5.1%) suffered toxicity (one patient developed grade III mucositis and another presented a grade II allergic reaction during the infusion of docetaxel). There were 30 deaths due to the following reasons: disease progression (n=28), myocardial infarction occurring during the administration of the first chemotherapy cycle (n=1), and respiratory failure because of infection (n=1) occurring during the 4th treatment cycle. Fatigue, allergic reactions, hematological toxicities (neutropenia, anemia, and thrombocytopenia) and mucositis were the most common adverse events noted during the study. With the exception of fatigue, these were also the most common serious adverse events. All adverse events and serious adverse events observed during the study are summarized in Table 3.

Table 3 Adverse events related to study treatment