A two-cohort phase 1 study of weekly oxaliplatin and gemcitabine, then oxaliplatin, gemcitabine and erlotinib during radiotherapy for unresectable pancreatic carcinoma

Introduction

A significant fraction of patients with pancreatic cancer have unresectable locally advanced disease at diagnosis with a median overall survival (OS) in the range of 9 to 12 months[1]. Despite assiduous research, there has been relatively little advancement in the treatment of locally advanced pancreatic cancer. Recently there has been significant debate about the relative effectiveness of chemoradiotherapy versus chemotherapy alone[2]. However, the consequences of failure to control local disease should not be underestimated; uncontrolled primary disease has been associated with pain, nausea, vomiting, weight loss, obstruction and decreased survival [3].

A number of studies have investigated chemoradiotherapy in locally advanced pancreatic cancer. Cancer and Leukemia Group B conducted a phase II trial of low-dose twice weekly gemcitabine (60mg/m2) and concurrent radiation (XRT, 50.4 Gy) with considerable toxicity and a median OS of 8.2 months.[4] There has been similar toxicity in other phase II trials using higher doses of gemcitabine administered once or twice weekly with relatively small radiation fields. As an example, a European phase II study tested weekly gemcitabine (100 mg/m² as a continuous infusion over 24 hours) plus concurrent XRT (50.4 Gy), followed by five cycles of standard-dose gemcitabine, in patients with locally advanced pancreatic cancer. Grade 3/4 toxicity was observed in half the patients, but the median OS was 15.5 months and 25% of patients were alive at two years[5]

Oxaliplatin has synergistic cytotoxicity when combined with gemcitabine in human cancer cell lines in vitro or in xenografts. [6] Several clinical trials have demonstrated the efficacy of oxaliplatin and gemcitabine in various solid tumors, including pancreatic cancer, with response rates ranging from 10 to 30%, but no improvement in OS has been demonstrated [7–9]. Two recent phase 1 clinical trials in pancreatic cancer demonstrated the feasibility of combining oxaliplatin and gemcitabine with concurrent XRT encompassing narrower radiation fields [10, 11].

Our aim was to define the safety and tolerability of gemcitabine and oxaliplatin given on a once weekly schedule with full doses of XRT for maximum local control, in the treatment of locally advanced and/or low volume metastatic pancreatic cancer. A second phase of the study, which added erlotinib to the combination, was initiated but not completed due to termination of funding; results for the patients enrolled in that substudy are also presented.

Methods

Results

A total of 18 patients were treated on this protocol (Table 2). All had pancreatic cancer. The median age was 60. Four patients had chemotherapy prior to study treatment. Most of the tumors were in the head of the pancreas. One patient had low-volume metastatic disease that had progressed on gemcitabine-based chemotherapy.

Table 3 describes grade 3 and 4 toxicities. One DLT (grade 4 elevated AST without evidence of cholangitis) occurred in a patient at dose level 3 and a dose reduction occurred in one patient. This dose level was expanded to 6 patients and no other patients experienced a DLT or dose reduction. One patient experienced a DLT and one patient a dose reduction at dose level 4, the highest dose level evaluated. After expansion of this dose level, the erlotinib study began, as described below.

Common grade 1 and 2 toxicities were nausea, vomiting, diarrhea, weight loss, thrombocytopenia and leukopenia. The most commonly encountered grade 3 toxicities were nausea (22%), elevated transaminases (17%), leucopenia (17%) and hyperglycemia (17%). Treatment delays and subsequent dose reductions occurred at dose level 1 (n=1), dose level 2 (n=2), dose level 3 (n=2 of 6, one of which was a DLT) and dose level 4. The highest tested combination was gemcitabine 200 mg/m² weekly, oxaliplatin 50 mg/m² weekly and XRT 50.4 Gy over 5 to 6 weeks, which produced manageable toxicity and represents the dose we would recommend for further evaluation.

All 18 patients completed XRT and each was evaluable for response assessment. Responses included one partial response, one complete response, and 14 (78%) with stable disease. Only 2 patients experienced progression at their first evaluation. Median progression free survival was 7.1 months (95% CI, 4.6–11.1 months) and median OS was 10.8 months (95% CI, 7.1–16.7 months).

Five patients were treated in Part 2 of the study, in which erlotinib was added to gemcitabine and oxaliplatin (Table 1). At dose level 1, two DLTs occurred. One patient was hospitalized for nausea, vomiting and diarrhea and another missed more than a week of treatment due to weight loss and cytopenias. This prompted de-escalation to dose level -1. Accrual was slow, however, and the study was closed early because the sponsors withdrew funding. Grade 3 toxicities in the limited number of patients treated included cytopenias, nausea, vomiting, diarrhea and anorexia.

Discussion

There is a persistent controversy over how best to treat locally advanced and metastatic pancreatic cancer and little convincing evidence to recommend any ‘new’ therapy since gemcitabine was approved in the 1990s. However, a strong rationale remains for studying new chemoradiotherapy regimens. Localized disease may profoundly affect quality of life and control of local disease may help palliate some of the most agonizing symptoms of pancreatic cancer-- pain, obstruction and cachexia. Radiotherapy, using chemosensitizing agents, may have an important role in the management of local disease. Gemcitabine acts as a radiosensitizer when combined with standard XRT, however gemcitabine doses must be reduced by approximately 10 to 30% of conventional doses, delivering suboptimal systemic drug concentrations[1]. Oxaliplatin also has radiosensitizing properties, and combinations of oxaliplatin and gemcitabine have demonstrated synergy in various studies. Combining the two agents in the locally advanced setting may improve systemic effect, and at the same time provide local control. We were able to demonstrate that full dose XRT can be integrated safely into this combination regimen. The highest tested combination was gemcitabine 200 mg/m² and oxaliplatin 50 mg/m² weekly with 50.4 Gy XRT over 5 to 6 weeks, which produced manageable toxicity.

Two other recent phase I/II trials have investigated the feasibility of using oxaliplatin and gemcitabine and concurrent XRT in the treatment of pancreatic cancer. One group of investigators administered oxalipatin (escalated to 85 mg/m²), full dose gemcitabine and an attenuated XRT dose. Grade 3 and 4 toxicities included thrombocytopenia, worsening of performance status, GI bleed, and other GI toxicities[10]. Another group administered two cycles of gemcitabine and oxaliplatin followed by escalating weekly doses of oxaliplatin (up to 70 mg/m²), a fixed dose of gemcitabine (300 mg/m²) and radiotherapy to 45 Gy (for a total of 5 weeks) with acceptable toxicity[11]. Median time to progression was 8 months and OS was 17 months

In our study, oxaliplatin and gemcitabine were also combined at less than standard doses but with a conventional dose of radiotherapy. There was no major toxicity at the highest planned doses of the two agents. PFS and OS in this study compares favorably to the survival results found in the Laurent study [11], which was very similar in design. Recent data from a chemotherapy-first approach in locally advanced pancreatic cancer[12] suggest that perhaps a strategy of chemotherapy alone followed by chemoradiation for patients who do not progress systemically is an approach with the best tradeoff between efficacy and minimization of toxicity. A gemcitabine/oxaliplatin/radiotherapy combination would be worth exploring within such a trial.

Our experience with erlotinib was limited by trial closure, but was associated with significant toxicity at the first planned dose level. Others have had experience with erlotinib plus radiotherapy combinations in pancreatic cancer. Iannitti et al utilized a regimen of gemcitabine (75 mg/m2 weekly), paclitaxel (40 mg/m2 weekly) and erlotinib, with the finding that erlotinib MTD was 50 mg daily [13]. Using a lower dose of gemcitabine of 40 mg/m2 twice weekly, Duffy et al. found the MTD of erlotinib to be 100 mg daily with hematologic toxicities comprising the DLTs. OS in that study was particularly interesting at 18.7 mos [14]. Tempering our enthusiasm for further study, however, is the recent experience in rectal cancer, where the addition of oxaliplatin to 5-FU based chemoradiation has produced disappointing results [15].

In summary, gemcitabine (200 mg/m2 weekly) and oxaliplatin (60 mg/m weekly) can be combined with standard-dose RT (1.8 Gy fractions to 5040 Gy). Lower doses of chemotherapy are required if erlotinib is added, even at only 50 mg daily. OS results in this study were promising, although the group was somewhat heterogeneous and the sample size small, tempering our enthusiasm to recommend this regimen for further study.