Definitive Results of the French FFCE-SFRO 2000-01 Study: Phase III Trial Comparing Chemoradiotherapy (Cisplating and Infusional 5-FU) Followed by Gemcitabine vs. Gemcitabine Alone in Patients with Locally Advanced Non Metastatic Pancreatic Cancer
Reviewer: Christine Hill, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 31, 2007
Presenter: C.H. Crane substituting for F. Mornex Presenter's Affiliation: Centre Hospitalier Lyon Sud, Lyon Pierre Benite, France Type of Session: Scientific
Pancreatic cancer is the fourth most common malignancy in the United States, and the annual incidence rate is almost identical to the mortality rate. In 2007, over 37,000 cases were diagnosed, and over 33,000 deaths documented [Jemal, CA Cancer J Clin, 2007]. The poor prognosis over pancreatic cancer is felt to be partially due to the frequency with which the disease is diagnosed at an unresectable stage.
No consensus currently exists regarding “standard” therapy for locally advanced, non-metastatic pancreatic cancer. The Gastrointestinal Tumor Study Group (GITSG) and Radiation Therapy Oncology Group (RTOG) in the US have previously identified a benefit to use of combined chemoradiotherapy using a 5-fluorouracil (5FU) based regimen, while chemotherapy alone has been advocated by the European Study Group for Pancreatic Cancer (ESPAC).
Gemcitabine is approved for use in treatment of pancreatic adenocarcinoma, but has been associated with severe toxicity when used as a radiosensitizer in phase I and retrospective trials [Crane CH, IJROBP, 2002]. Its use as a single agent has thus been advocated by some groups.
This phase III trial was designed to compare initial chemoradiotherapy with a 5FU based regimen followed by gemcitabine monotherapy to gemcitabine monotherapy alone in patients with locally advanced, non-metastatic, pancreatic cancer.
Materials and Methods
A total of 120 patients from 22 European centers were randomized as part of this trial. Patients were World Health Organization (WHO) status 0 – 2, and had pathologically proven pancreatic adenocarcinoma. All patients had been deemed unresectable by institutional guidelines, and had no evidence of metastatic disease based on CT imaging.
Patients were randomized 1:1 to receive concurrent chemoradiotherapy (CHRT) versus gemcitabine (GEM) alone. Patients in the former arm were treated with 60 Gray (Gy) in 2 Gy fractions with concurrent 5FU (300 mg/ m2/24h, continuous infusion, day 1-5 weekly) and cisplatin (20 mg/ m2/ day, day 1-5 at weeks 1 and 5). GEM was administered at a dose of 1000 mg/ m2 weekly.
All patients received maintenance gemcitabine treatment, 1000 mg/ m2 weekly for 3 of every 4 weeks until disease progression or limiting toxicity.
Patients were stratified according to treatment center, performance status, and initial surgery or lack thereof.
The primary endpoint was overall survival (OS), with secondary endpoints of toxicity and progression free survival (PFS).
This study was designed to include randomization of 176 patients; however, slow accrual prompted this unplanned analysis.
Between 3/2000 and 2/2007, 59 patients were randomized to CHRT and 60 to GEM. Median follow-up was 33 months.
Patient characteristics were similar between the the CHRT and GEM arms, with mean age 60.1 and 62.4 years, respectively, and sex ratio 1 and 1.4, respectively. Almost half (40 and 42%) of patients in both arms had undergone exploratory laparotomy.
The GEM arm included more performance status 2 patients than the CHRT arm (23% versus 9%).
During the induction phase, 83% of patients received at least 75% of the planned radiation dose. At least 75% of planned chemotherapy doses were delivered to 54% receiving 5FU, 51% receiving cisplatin, and 73% receiving gemcitabine.
Median survival was 8.6 months in the CHRT arm versus 13 months in the GEM arm (p = 0.057). Progression free survival was 6 versus 6.7 months, respectively (p = 0.3). One year OS was 32% versus 53% (p = 0.05) in the CHRT arm versus GEM and 1 year PFS was 14% versus 32% (p = 0.09), respectively.
During the induction phase, grade 3-4 hematologic toxicity was significantly more common in the GEM study arm. Non-hematologic toxicity, including nausea, vomiting, and anorexia, was significantly higher in the CHRT arm. One treatment related death was observed in the CHRT arm, and none in the GEM arm.
Maintenance gemictabine was administered in significantly higher doses in the GEM arm versus the CHRT arm (11681 mg/ m2 versus 6464 mg/ m2, p = 0.04).
The authors conclude that concurrent chemoradiation is a feasible treatment strategy, but that outcomes are not superior to gemcitabine monotherapy.
They note that the toxicity of the concurrent regimen employed was significant and limited maintenance gemcitabine dosing.
This study represents a comparison of concurrent chemoradiotherapy to gemcitabine monotherapy in patients with locally advanced pancreatic carcinoma.
The authors did not find benefit to use of the described concurrent regimen when compared to gemcitabine alone, with non-statistically significant trends in improvement in OS and PFS demonstrated with gemcitabine alone.
The presentation of this study offers very little detailed information regarding radiotherapy techniques employed. Larger than necessary volume contouring has been demonstrated by other groups to significantly impact the development of grade 3/ 4 gastrointestinal toxicity [Crane, ASTRO, 2007]. Without more information on the details of the delivered radiotherapy, assessing the potential impact of volume differences within this study is not possible.
Additionally, the authors’ conclusions are largely based on increased toxicity encountered by patients treated with concurrent chemoradiotherapy; however, very little information regarding supportive care was presented. Adequate supportive care is essential to provision of adequate chemo and radiotherapy to patients with gastrointestinal (and other) malignancies, and proactive nutritional support has been demonstrated to improve outcomes in patients with pancreatic adenocarcinoma [Wilson, ASTRO, 2007]. Further information on supportive care provided would certainly be useful in interpretation of this data.
Lastly, the authors report from this phase III trial that the concurrent chemoradiation protocol employed in this trial is feasible, but may not offer benefit. This trial was not in face preceded by a phase II feasibility trial; had a phase II trial investigated the use of this specific protocol, several dose limiting toxicities might have been anticipated and addressed more thoroughly in the phase III trial reported here.
This study represents an investigation of two potential treatments for locally advanced pancreatic cancer. Each appears to be feasible; however, as expected, toxicity was a more important dose-limiting factor in the chemoradiotherapy arm, as evidenced by decreased compliance in this arm. Radiotherapy techniques and supportive care would both be expected to impact this toxicity significantly. Further information regarding these factors would be of assistance in interpreting the results of this trial.
Partially funded by an unrestricted educational grant from Bristol-Myers Squibb.