The 9th Annual World Congress on Gastrointestinal Cancer was held in Barcelona, Spain, June 28 through July 1, 2007 in association with the European Society of Medical Oncology. This meeting continues to grow and is a forum for the review of data recently presented at the 43rd Annual Meeting of the American Society of Clinical Oncology (ASCO) in June as well as new information. The smaller size and gastrointestinal (GI) oncology focus of the meeting allowed for longer and more detailed discussion of the abstracts than at ASCO. Perhaps the most important new data discussed was from the PACCE (Panitumumab Advanced Colorectal Cancer Evaluation) trial, which examined combined EGFR and VEGF blockade in first-line metastatic colorectal cancer.
The ability to predict outcome from tumor samples has attracted a great deal of interest over the past few years. Prognosis as well as prediction of treatment effect would be clinically quite useful. Dr. Roth, et al. from Geneva University presented interesting translational results using samples from the 1,564 patients who participated in the PETACC-3 adjuvant trial. This trial showed minimal improvement of adding irinotecan to infusional 5-FU.,, The investigators used immunohistochemical and molecular techniques to examine a panel of nine markers that have potential utility predicting efficacy or toxicity and found a very high rate of successful examination. Preliminary results showed better prognosis in patients with SMAD4 expression and microsatellite stability consistent with smaller data sets. Further analyses of this and other large data sets have the potential to identify markers that with further validation could lead to truly individualized treatment.
Upper Gastrointestinal Cancer
Dr. Conroy, et al. from Centre Alexis Vautrin, Nancy, France, presented the final results of an important trial of definitive chemoradiotherapy for adenocarcinoma of the esophagus, comparing FOLFOX4 to standard 5-FU and cisplatin., Patients with locally advanced cancer or patients who were unfit for surgery were eligible for the study. What additional benefit surgery brings to chemoradiation remains controversial with only a minority of patients being cured. For decades the standard chemotherapy used as part of combined modality therapy for esophageal cancer has been the combination of infusional high-dose 5-FU and cisplatin. Unfortunately, this regimen has a high rate of gastrointestinal and hematologic side effects. Earlier uncontrolled studies have shown promising activity and toxicity of 5-FU and oxaliplatin as the FOLFOX regimen together with standard radiotherapy.
This Phase II trial randomized 97 adenocarcinoma patients with the primary endpoints being feasibility and endoscopic complete response rate. Seventy-five percent of the FOLFOX/RT and 70% of the 5-FU/cisplatin/RT patients completed the planned treatment. Toxicities were similar except for increased neurotoxicity with FOLFOX, which was mostly grade 1. The FOLFOX arm had a higher endoscopic complete response (CR) rate, time to progression (TTP), and overall survival (OS) rate, though no comparative statistics were presented. The promising activity will be examined in a Phase III trial.
At ASCO 2006 Dr. Chauffert, et al. presented provocative data from the SFRO/FFCD 2000-01 trial, which compared a nonstandard regimen of chemoradiation with 5-FU and cisplatin followed by gemcitabine versus gemcitabine alone in locally advanced pancreatic carcinoma. Surprisingly, the combined modality arm did significantly worse with a median survival of 8.4 months compared with 14.3 months for gemcitabine alone. The trial was stopped early due to this difference and slow accrual. The reasons for this difference remain unclear, perhaps due to less gemcitabine delivery due to delay or toxicity. One possible advantage of chemoradiotherapy is downstaging of unresectable lesions, making potentially curative surgical removal possible. Dr. Rougier presented updated data at the World Congress showing that there was no difference in secondary resections (2 vs. 3) between the arms and corroborated previous nonrandomized studies showing that downstaging is rare. There is an ongoing trial in the United States (PACT) examining a more standard chemoradiotherapy regimen to the same with adenoviral gene therapy encoding the chemoradiosensitizer TNF in this same population of locally advanced pancreatic cancer patients.
The results of the OPUS study, a randomized Phase II comparing FOLFOX with or without cetuximab in first-line metastatic colorectal cancer 9 and the EPIC trial 10 of irinotecan with or without cetuximab in second-line patients were reprised at the conference. The results of the NO16966 trial in previously untreated patients comparing XELOX and FOLFOX and bevacizumab and placebo were also presented again and discussed.
There were two randomized trials in first-line colorectal cancer at the meeting that were presented for the first time. The first examined the use of myeloid growth factors with modern chemotherapy for colorectal cancer.11 The use of filgrastim and pegfilgrastim (Neulasta) has reduced the incidence of neutropenia and neutropenic fever in myelosuppressive chemotherapy regimens. Pegfilgrastim has been approved for use every three weeks, but most modern colorectal cancer regimens are dosed every two weeks. The safety and efficacy of pegfilgrastim has not been examined with these regimens and the shorter 12-day period before chemotherapy. Dr. Hecht, et al. presented for the first time the results of a 252 patient randomized trial showing safety and efficacy of pegfilgrastim in reducing grade 3/4 neutropenia and febrile neutropenia in patients with metastatic colorectal cancer treated with FOLFOX4, FOLFIRI, or FOIL (5-FU, oxaliplatin, irinotecan, leucovorin). Grade 3/4 neutropenia fell from 43% to 13% and febrile neutropenia from 8% to 2% treatment compared to placebo. Other than an increase in bone pain, toxicities were similar. This study validates the feasibility of pegfilgrastim use with every-other-week colorectal cancer regimens.
Preclinical models of colorectal cancer indicate that the combination of EGFR and VEGF inhibitors increased growth inhibition.12 Randomized chemorefractory, biological naïve patients found that adding bevacizumab to either cetuximab or irinotecan, cetuximab, and bevacizumab increased response rates compared to historical controls. On the basis of this information, the PACCE (Panitumumab Advanced Colorectal Cancer Evaluation) trial was performed in first-line metastatic colorectal cancer patients. Panitumumab is a fully human IgG2 antibody against the EGFR with single agent activity in EGFR positive and negative colorectal cancer.13,14 Surprisingly, an interim analysis in April 2007 lead to discontinuation of the panitumumab treatment. The World Congress meeting was the site of the first public presentation of the data.
The trial had two separate cohorts to which patients were assigned by investigators’ choice. All patients received bevacizumab and either oxaliplatin-based (bev/ox-CT) or irinotecan-based (bev/iri-CT) chemotherapy at the discretion of the investigator. Within each cohort, patients were randomized 1:1 to receive panitumumab 6 mg/kg every other week or no additional therapy. Due to the rash caused by EGFR inhibitors, the study was unblinded. The primary analysis was progression free survival (PFS) in the planned 800 patient bev/ox-CT cohort and was centrally reviewed. Secondary endpoints were safety and other efficacy parameters in the bev/ox-CT cohort as well as safety and efficacy in the 200 patient bev/iri-CT cohort. Ongoing review continued by the independent data monitoring committee (IDMC) after 800 patients revealed an increase in diarrhea, dehydration, and infection with panitumumab and this was added to the consent form in January of 2007. In March 2007, a planned interim review of the data after 231 PFS events showed worse outcomes in the panitumumab treated patients, and within 48 hours panitumumab treatment was discontinued, though data collection continues.
Interim toxicity data for the bev/ox-CT cohort were presented including an increased risk of grade 4 adverse events (28% vs. 18%) in the panitumumab group. There was an increase in diarrhea, dehydration, infections, and pulmonary embolism, but not deep venous thrombosis, in the patients who received panitumumab.
Interim efficacy results for the bev/ox-CT cohort were presented. Response rates by central review were similar between arms (39% with panitumumab, 41% without). More patients in the panitumumab arm had potentially curative resections (7% vs. 3%). A planned interim analysis of the primary endpoint of PFS was performed as of the data cutoff of Oct 30, 2006. The panitumumab arm was inferior to the control arm with a median PFS of 8.8 months compared with 10.5 months with a HR of 1.44, p=0.004. These data resulted in the discontinuation of panitumumab at the end of March 2007. An update of PFS, in April 2007 capturing the time before discontinuation of panitumumab showed median PFS of 9.0 vs. 10.5 months with a HR of 1.29. Unplanned overall survival analyses at the October 30, 2006 and April 2007 data cutoffs showed decreased survival in the panitumumab arm. OS as of the April 2007 data cutoff showed median OS in the panitumumab arm of 18.6 months versus not reached in the control arm with a HR of 1.44.
Additional analyses were performed to try to understand these unexpected results. No significant differences in initial response or waterfall plots were seen between the arms. Certain vulnerable groups such as those over 80 years of age, higher ECOG, or those with comorbidities had appreciably worse outcome. The delivery of bevacizumab and chemotherapy was decreased in the panitumumab arm with more dose reductions and delays. As in the recently reported Roche NO16966 trial, the majority of patients in both arms discontinued therapy due to reasons other than progressive disease. Time to treatment failure was virtually identical between the arms (HR 1.03).
The presentation was of only the initial data from the PACCE trial. Clearly, the combination of the anti-EGFR antibody panitumumab to bevacizumab and oxaliplatin-based chemotherapy results in increased toxicity. This has been seen with the addition of EGFR inhibitors to chemotherapy in other studies. Furthermore, there were no signs of synergistic anti-tumor activity and, in fact, evidence of less activity. Possible explanations include increased toxicity from treating everyone, particularly vulnerable populations, with a five drug regimen when only a minority of patients may benefit from adding EGFR inhibition. Further analyses of more mature data including molecular analyses and data from the irinotecan cohort may help explain the unexpected results.
The World Congress is developing into an international forum for gastrointestinal oncology much like the Gastrointestinal Cancers Symposium is becoming in the United States. While many new agents have become available for the treatment of GI cancers, how to choose patients and therapies remains an area of intense research.
 Roth AD, Tejpar S, Yan P, et al. Tissue biomarkers (BIOM) in colon cancer (COC): The translational study on the randomized phase III trial comparing infused irinotecan/5-fluorouracil (5-FU)/folinic acid (FA) to 5-FU/FA in stage II-III COC patients (pts). Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4022.
 Van Cutsem E, Labianca R, Hossfeld D, et al. Randomized Phase III trial comparing infused irinotecan / 5-fluorouracil (5-FU)/folinic acid (IF) versus 5-FU/FA (F) in stage III colon cancer patients (pts). (PETACC 3). Proceedings from the American Society of Clinical Oncology Conference. Orlando, FL. 2007. Abstract #8.
 Roth A, Teipar S, Yan P, et al. Tissue biomarkers (BIOM) in colon cancer (COC): The translational study on the randomized phase III trial comparing infused irinotecan/5-fluorouracil (5-FU)/folinic acid (FA) to 5-FU/FA in stage II-III COC patients (pts). Proceedings from The Ninth World Congress on Gastrointestinal Cancer. Barcelona, Spain. 2007. Abstract # 0036.
 ConroyT, Hua A, Etienne P, et al. A randomized Phase II study evaluating definitive chemoradiotherapy (CRT) with FolFox 4 or 5FU Cisplatin as first line treatment of patients (PTS) with inoperable Esophageal Cancer (IEC). Proceedings from the The Ninth World Congress on Gastrointestinal Cancer. Barcelona, Spain. 2007. Abstract # 0014.
 Conroy T, Yataghene Y, Etienne PL, et al. Definitive chemo-radiotherapy (CRT) with folfox 4 or 5FU-cisplatin as first line treatment for patients (pts) with inoperable esophageal cancer (IEC): Final results of a randomized Phase II study. Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4532.
 Khushalani NI, Leichman CG, Proulx G, et al. Oxaliplatin in Combination with protracted-Infusion fluorouracil and radiation: Report of a clinical trial for patients with Esophageal Cancer. Journal of Clinical Oncology, Vol 20, Issue 12 (June), 2002: 2844-2850.
 Chauffert B, Mornex F, Bonnetain F, et al. Phase III trial comparing initial chemoradiotherapy (intermittent cisplatin and infusional 5-FU) followed by gemcitabine vs. gemcitabine alone in patients with locally advanced non metastatic pancreatic cancer: A FFCD-SFRO study. Proceedings from the American Society of Clinical Oncology Conference. Atlanta, GA. . 2006. Abstract # 4008.
 Rougier P, Mornex F, Chauffert B, et al. Unresectable Pancreatic Cancer: Survival and secondary resections after chemoradiotherapy followed by gemcitabine compared to gemcitabine alone (SFRO/FFCD 2000-01 PHASE III TRIAL). Proceedings from the Ninth World Congress on Gastrointestinal Cancer. Barcelona, Spain. 2007. Abstract # 08.
 Schuch G, Staroslawska E, Nowacki M, et al. Cetuximab plus 5-Fu/FA/Oxaliplatin (FOLFOX-4) In the first-line treatment of MCRC: Opus, a Phase II study. Proceedings from the Ninth World Congress on Gastrointestinal Cancer. Barcelona, Spain. 2007. Abstract # 022.
 Sobrero A, Hochester H, Luppi G, et al. Cetuximab plus irinotecan in patients with MCRC who have failed prior oxaliplatin-based therapy: The Epic Trial. Proceedings from the The Ninth World Congress on Gastrointestinal Cancer. Barcelona, Spain. 2007. Abstract # 030.
 Hecht J, Malik I, Gollard R, et al. Pegfilgrastim reduces the indidence of Neutropenia in patients with locally advanced Ormetastatic colon cancer receiving chemotherapy administered every 2 weeks as first-or second-line treatmeant: A Phase II, double -blind, placebo-controlled study. Proceedings from the Ninth World Congress on Gastrointestinal Cancer. Barcelona, Spain. 2007. Abstract # 025.
 Saltz LB, Lenz H, Hochster H, et al. Randomized Phase II Trial of Cetuximab/Bevacizumab/Irinotecan (CBI) versus Cetuximab/Bevacizumab (CB) in Irinotecan-Refractory Colorectal Cancer. Proceedings from the American Society of Clinical Oncology Conference. Orlando, FL. 2005. Abstract # 3508.
Hecht J, Mitchell E, Baranda J, et al. Panitumumab antitumor activity in patients (pts) with metastatic colorectal cancer (mCRC) expressing low (1-9%) or negative (<1%) levels of epidermal growth factor receptor (EGFr). Proceedings from the American Society of Clinical Oncology Conference. Atlanta, GA. 2006. Abstract #3547.