Past and future of pancreas cancer: are we ready to move forward together?

N. Merchant,J. Berlin

DOI: https://doi.org/10.1200/JCO.2008.17.0811

IF: 45.3

2008-07-20

Journal of Clinical Oncology

Abstract:In this issue of Journal of Clinical Oncology, four articles are published on perioperative therapy of resectable pancreatic cancer. In two articles, we review the past efforts to improve the outcomes for postoperative patients and in the other two we get a glimpse at a future that may provide new hope for improving outcomes. Many questions are raised by these for articles, but few are answered. The history of randomized trials to understand the benefit of adjuvant therapy for pancreas cancer began with a trial from the Gastrointestinal Study Group (GITSG). The study was small and fraught with many flaws that opened the study to criticism. However, with a statistically significant benefit for the adjuvant therapy arm, the GITSG trial established combined modality adjuvant therapy as a commonly employed standard within the United States, but less so elsewhere. Since the GITSG trial, other trials have not been able to convincingly prove or disprove the role of radiation therapy and have suggested that adjuvant chemotherapy alone may be of benefit. This issue of the Journal contains both a retrospective analysis from the Mayo Clinic and an evaluation of a prospective database from Johns Hopkins, among the highest volume pancreas cancer research programs supporting the use of combined modality chemoradiotherapy as a standard of care. We laud the efforts of these investigators for their excellent analyses. There are two possible ways to interpret these articles. First, they add information to the literature on the outcomes of patients treated at two excellent institutions and also provide perspective through multivariate analyses on prognostic factors. Second, they come as a condemnation to pancreas cancer researchers throughout the country. While we have learned a lot about the disease, these articles demonstrate that we have made little to no progress in the treatment of pancreas cancer during the past 30 years. Similarly, several randomized trials suggest that we improve the survival of resected patients from approximately 10% for surgery alone to slightly more than 20% for patients receiving some form of adjuvant therapy. However, this means the vast majority of patients with seemingly localized disease still die of pancreas cancer. New radiographic techniques have been developed since GITSG for better staging. Operative morbidity and mortality have lessened. Techniques for delivery of radiation and chemotherapy have improved. Yet the outcomes for our patients remain unchanged. The article from Corsini et al evaluates only patients with R0 resection, self-selecting the optimally surgically treated patients while the Herman et al study includes all resected patients. In both studies, there remains a significant element of selection bias toward healthier patients and patients with more adverse prognostic factors receiving adjuvant therapy. The Corsini et al analysis reviews patients from 1975 onward. This means the included patients were treated through 30 years of refinement of imaging technology, as well as improvements in surgical, radiation, and medical management. With adjuvant chemoradiotherapy, 28% of these selected patients were alive at 5 years compared with 17% without chemoradiotherapy, although the comparison was not derived from a randomized population. The Herman et al study evaluates patients from a more modern era (1993 to 2005), and shows that patients receiving chemoradiotherapy followed by chemotherapy experienced an improved median survival compared with a cohort of patients undergoing surgery alone (21.2 months v 14.4 months, P .001). Intriguingly, these results are nearly identical to those reported for patients from an earlier time also by Johns Hopkins (median survival, 20 months with CRT v 14 months with surgery alone). The outcomes in the Yeo article, as well as the two retrospective analyses in this issue of the Journal are similar: patients who were selected for and elected to receive chemoradiotherapy therapy after surgical resection of pancreas cancer lived longer than those patients who did not receive adjuvant chemoradiotherapy. How do we react to the reality the Corsini et al and Herman et al have made us face? We could regret the past, during which we conducted trials that failed to establish a true standard. In recent years, three important randomized trials have been published. The first, The European Study Group for Pancreatic Cancer-1, has been very heavily criticized. In a 2 2 randomization, the ESPAC-1 investigators attempted to resolve the merits of the chemoradiotherapy therapy portion of adjuvant treatment and of the chemotherapy portion. In fact, ESPAC-1 established that chemotherapy with fluorouracil (FU) can increase the median and long-term survival for patients with resected pancreas cancer. However, several legitimate criticisms have been leveled against the study design and the chemoradiotherapy results. The Charite Onkologie CONKO-001 trial compared postoperative gemcitabine with observation, demonstrating an improvement in disease-free survival, and a greater number of 5-year survivors with adjuvant therapy. The analyses by Herman et al and Corsini et al appear to be consistent with these data. It seems reasonable, therefore, that when treating patients with resected pancreas cancer, adjuvant therapy of some form should be given to improve outcomes. To analyze the data further, the components of adjuvant therapy should be reassessed. First, in both the Corsini et al and Herman et al JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 26 NUMBER 21 JULY 2