Charu Sharma, David Horowitz, John Chabot, Muhammad Wasif Saif
Columbia University College of Physicians and Surgeons. New York, NY, USA
There is no clear consensus on what type of adjuvant therapy should be used for patients with pancreatic cancer. Chemoradiation is the favored treatment modality by many in the United States while gemcitabine based chemotherapy is favored in Europe. Both of these approaches have been shown by large prospective, randomized trials to improve disease free intervals and in some studies overall survival. This year at the American Society of Clinical Oncology (ASCO) Gastrointestinal Cancer Symposium, the randomized phase III study presented by Uesaka et al. from Japan (Abstract #145) represents a newer paradigm of oral adjuvant S-1 chemotherapy in place of the traditional standard of care intravenous gemcitabine in terms of prolonging patients’ survival. Another study by Fan et al. (Abstract #269) examined the value of targeted therapy using erlotinib with adjuvant chemoradiation and chemotherapy. We present the summary of these two studies and discuss the potential impact on our clinical practice on this highly lethal cancer.
Chemotherapy, Adjuvant; Combined Modality Therapy; Deoxycytidine Kinase; gemcitabine; Pancreatic Neoplasms; Radiotherapy, Adjuvant; RRM1 protein, human; S100A2 protein, human; SLC29A1 protein, human
dCK: deoxycytidine kinase; EORTC: European Organization of Research and Treatment of Cancer; ESPAC: European Study Group for Pancreatic Cancer; hENT1: human equilibrative nucleoside transporter 1; RRM1: ribonucleotide reductase subunit 1; RTOG: Radiation Therapy Oncology Group
Only 10-20% of patients have classically resectable disease at the time of pancreatic cancer diagnosis. Surgical resection remains the only curative modality for pancreatic cancer. Nevertheless, the prognosis of patients after complete resection is poor, with 3-year disease-specific survival rate at 27% (95% confidence interval (CI): 23-32%) and median survival of 15-19 months [1, 2, 3]. Therefore, judicious use of efficacious adjuvant therapies is necessary to improve the survival of resected pancreatic cancer patients. There is now substantial high level evidence to support the use of adjuvant chemotherapy in resected pancreatic cancer [4, 5, 6, 7]. However, the role for adjuvant chemoradiotherapy remains more controversial.
There are numerous randomized controlled trials that support the use of adjuvant chemotherapy after resection of pancreatic cancer [4, 5, 6,]. The European Study Group for Pancreatic Cancer (ESPAC-1) trial demonstrated a survival benefit for adjuvant chemotherapy but not adjuvant chemoradiotherapy and even a possible detrimental effect for adjuvant chemoradiation . The Charité Onkologie Clinical- 001 (CONKO-001) study randomized patients with resected pancreatic cancer to gemcitabine for 6 months or observation . Adjuvant chemotherapy showed a trend towards improved overall survival. The use of gemcitabine versus 5-FU was further defined by the ESPAC-3 trial, which demonstrated equivalent survival for both treatments, but more favorable safety profile with gemcitabine . There was also a trend toward improved survival in the gemcitabine arm in patients with node positive disease or those with positive resection margins . To further support the role of adjuvant chemotherapy, the Boeck et al.  metaanalysis demonstrated adjuvant chemotherapy provided a significant increase in median survival.
The role for adjuvant chemoradiation is less well defined as there are conflicting results from trials. Despite the controversy, the level of evidence is strong enough to support the use of adjuvant chemoradiotherapy in the management of resected pancreatic cancer in the United States. However, in Europe it is common practice for patients to receive adjuvant chemotherapy alone. There are three randomized controlled trials investigating the role of adjuvant chemoradiation in resected pancreatic cancer [4, 8, 9].
The Gastrointestinal Study Group (GITSG) study showed a survival benefit in patients who received bolus 5-FU with radiotherapy, but has been criticized for a sample size of 43 patients . The European Organization of Research and Treatment of Cancer (EORTC) trial did not demonstrate a survival advantage for patients treated with adjuvant chemoradiation compared to observation . There was a trend toward survival in the chemoradiotherapy arm compared to observation in the subset of patients with pancreatic ductal carcinoma . Radiation therapy in the EORTC trial was suboptimal as the dose was inadequate (40 Gy) and the radiation was delivered with a split course. The ESPAC-1 evaluated adjuvant concurrent chemoradiation therapy (bolus 5-FU/splitcourse radiation), chemotherapy alone (5-FU/ leucovorin), chemoradiation therapy followed by chemotherapy, and observation . The results demonstrated that the chemotherapy-only arm had a significant benefit over the observation arm in median survival and the chemoradiation therapy arm showed worse median survival compared to the observation arm . This study was criticized for a confusing 2x2 factorial design, possible selection bias and suboptimal radiotherapy (split course/poor quality control). An additional phase 3 trial, Radiation Therapy Oncology Group (RTOG) 9704, showed a benefit of adding gemcitabine to infusional 5-FU combined with radiotherapy at the cost of more grade 4 hematological toxicity .
Despite the lack of randomized controlled trials, evidence supporting the role of adjuvant chemoradiotherapy in resected pancreatic cancer, several single institution and retrospective series demonstrate a benefit for adjuvant chemoradiotherapy [11, 12, 13, 14]. The Johns Hopkins-Mayo Clinic Collaborative Study demonstrated that adjuvant chemoradiation (5- FU based chemo- and radio-therapy to 50.4 Gy) following pancreaticoduodenectomy was associated with improved survival compared to observation alone in their two institutional trial of 1,092 patients .
Furthermore, a retrospective review of 472 patients at the Mayo Clinic found a survival benefit for adjuvant chemoradiation after R0 pancreaticoduodenectomy . Several Surveillance, Epidemiology and End Results (SEER) analysis have also demonstrated efficacy for radiation therapy in pancreatic cancer [13, 14].
There may be a role for chemoradiation in the treatment of patients with R1 resections [15, 16]. A meta-analysis by Stocken et al. demonstrated a 25% significant reduction in the risk of death with chemotherapy with no significant reduction in the risk of death with adjuvant chemoradiation . However, their subgroup analyses demonstrated that chemoradiotherapy was more effective than chemotherapy alone in patients with positive resection margins . Similarly, the meta-analysis by Butturini et al. demonstrated a possible benefit to chemoradiation in patients with positive resection margins .
The optimal time to initiate adjuvant chemotherapy after pancreatic cancer surgery is unknown. Adjuvant chemotherapy or chemoradiotherapy has been started from 2 to 10 weeks after surgery, with most trials starting adjuvant therapy within 8 weeks [1, 2, 5]. Once adjuvant chemotherapy has been initiated, predicting the response of patients has been the subject of intensive research. The calcium-binding protein S100A2 has been validated in a retrospective cohort of patients treated with pancreatectomy for pancreatic cancer as an independent predictor of survival, with high expression correlated with disease progression and poor outcome [17, 18]. Additionally, with gemcitabine-based chemotherapy being a mainstay of therapy, markers of the efficacy of gemcitabine such as expression of the human equilibrative nucleoside transporter 1 (hENT1), deoxycytidine kinase (dCK) and ribonucleotide reductase subunit1 (RRM1) proteins have been tentatively identified in vitro and in vivo .
Despite the use of adjuvant chemotherapy or adjuvant chemoradiotherapy after surgical resection, survival still remains poor. Future studies in pancreatic cancer will help to further define the role of adjuvant chemoradiotherapy, elucidate the most efficacious chemotherapeutic and biologic agents, optimize dosing and timing of chemotherapy/radiation therapy and individualize treatment based on predicting response to chemotherapy and radiation therapy.
Chemotherapy versus Chemoradiotherapy
Drudi et al. (Abstract #4042)  conducted a pooled analysis of all randomized controlled trials from 1966- 2010 investigating the role of adjuvant treatments, including both adjuvant chemotherapy and adjuvant chemoradiation, in resected pancreatic cancer patients. The purpose of the analysis was to determine if adjuvant treatment, adjuvant chemotherapy or adjuvant chemoradiation confer a survival benefit at 5 years compared with no adjuvant treatment. There were a total of 2,410 pooled patients from 12 randomized controlled trials; 1,337 patients were treated with adjuvant treatment (1,008 with adjuvant chemotherapy and 329 with adjuvant chemoradiation), and 1,073 received no adjuvant treatment. The authors demonstrated significant 5 year survival benefit for adjuvant treatment and adjuvant chemotherapy (odds ratio equal to 0.62, P=0.001 and odds ratio equal to 0.63, P=0.021, respectively), but not for adjuvant chemoradiation (odds ratio equal to 0.92, P=0.71). This pooled analysis demonstrated that adjuvant chemotherapy improves 5-year survival in resected pancreatic cancer patients but not adjuvant chemoradiotherapy. This study served to further strengthen the role of adjuvant chemotherapy in the management of resected pancreatic cancer. These results concur with the results of previous meta-analyses demonstrating a survival benefit for adjuvant chemotherapy [7, 15].
Cetuximab plus Gemcitabine
The addition of cetuximab to adjuvant gemcitabine was investigated in an open label, multi-center, phase II trial reported by Fensterer et al. (Abstract #4039) . Patients underwent R0 or R1 resection for pancreatic cancer, then were treated with adjuvant chemotherapy consisting of 6 cycles of gemcitabine with weekly cetuximab for 24 weeks. There were 76 patients enrolled, and 73 patients received at least one dose of cetuximab. Median age was 64 years; 22.4% had R1 resection and 69.1% had K-ras mutation. Median disease free survival was 11.9 months, and the disease free survival rate at 18 months was 33.5%, failing to demonstrate superiority over 35% as hypothesized by the authors. Median overall survival was 21.5 months (95% CI: 16.9-28.2 months). Grade 3 or 4 toxicities were neutropenia in 11% of patients, thrombocytopenia in 8.2%, dermatologic in 6.9%, and allergic reaction in 6.9%. The authors conclude that the addition of cetuximab to gemcitabine in the adjuvant treatment of pancreatic cancer does not improve disease free survival over the use of gemcitabine alone (Table 1).
Gemcitabine, Cisplatin with Radiation
Kwon et al. (Abstract #4094)  conducted a phase II trial of adjuvant gemcitabine and cisplatin chemotherapy followed by chemoradiation with gemcitabine and 5,040 cGy of radiation, then 4 cycles of maintenance gemcitabine. There were 74 patients with stage IB-IIB pancreas cancer who had undergone resection enrolled between 2005 and 2009. The median age was 61 year and the median follow-up was 45 months (range: 10.2-64.6 months). Of the patients enrolled, 57 completed chemotherapy followed by chemoradiation. One-year disease free survival (DFS) rate was 62.1%, median disease free survival was 17.4 months, and median overall survival was 33.6 months. The majority of recurrences (66.2%) were distant metastases. Increasing stage and involved lymph nodes were associated with reduced disease free survival (P<0.001 and P=0.01, respectively). Fifty-three of 74 patients (71.6%) had grade 3 or 4 hematologic toxicity, with 4 patients experiencing febrile neutropenia. These finding suggest promising efficacy with acceptable toxicity for adjuvant multimodality therapy (Table 1).
Pisa et al. (Abstract #e14519)  evaluated a cohort of 29 consecutive patients with resected nonmetastatic pancreatic cancer who received adjuvant chemotherapy with gemcitabine to attempt to identify a relationship between time to adjuvant chemotherapy and survival . The median time to adjuvant chemotherapy was 47 days (range: 22-183 days), and the most common reason for delay of adjuvant chemotherapy was postoperative complications. No difference in age, gender, stage or palliative chemotherapy used at progression was identified between patients who started adjuvant chemotherapy within 56 days (8 weeks) of surgery versus those who started adjuvant chemotherapy after 56 days post-surgery. Median overall survival was 26.4 months in those who started adjuvant chemotherapy within 56 days of surgery versus 14.8 months in those who started adjuvant chemotherapy more than 56 days after surgery (P=0.015). No significant difference was seen in median progression free survival between the groups. No patients died of toxicity or post-operative complications. This underscores the need to keep time to adjuvant chemotherapy under 8 weeks after surgery, as has been the case in most clinical trials.
S100A2 Expression as a Prognostic Marker
Tompero et al. (Abstract #4118)  performed a secondary analysis of a subset of patients with head of pancreas lesions treated adjuvantly on RTOG 9704 in an attempt to validate S100A2 expression as a prognostic marker in 150 specimens from patients receiving adjuvant chemotherapy for pancreatic cancer. Tissue microarray was used to quantify S100A2 expression, and patients were then stratified into four groups based on the level of expression. For high vs. no/low expression of S100A2, disease specific survival was not significantly different at 1 or 2 years (P=0.09; Table 2). While S100A2 was not validated as a prognostic marker in this cohort of patients, the authors recommend further study to try to resolve the conflicting data about the role of S100A2 as a prognostic biomarker.
hENT1, dCK, and RRM1
Marechal et al. (Abstract #4024)  conducted a study using expression of hENT1, dCK, and RRM1 in 434 patients receiving adjuvant gemcitabine after curative-intent resection of pancreatic cancer in an attempt to associate expression of these proteins with efficacy of gemcitabine. Among patients not treated with gemcitabine-based chemotherapy, hENT1, dCK, and RRM1 expression was not associated with overall survival. In contrast, among the patients treated with gemcitabine, hENT1 and dCK expression levels were associated with changes in overall survival after adjusting for tumor grade, size, lymph node involvement and resection margin. In tumors with high hENT1, gemcitabine was associated with better overall survival (HR: 0.44; 95%CI: 0.28-0.69; P<0.001); in tumors with high dCK, gemcitabine was associated with better overall survival (HR: 0.57; 95% CI: 0.41- 0.78; P=0.001). In tumors with low hENT1or dCK levels, gemcitabine was not associated with improved overall survival. These findings suggest that expression levels of hENT1 and dCK may predict response to gemcitabine-based chemotherapy after curative-intent surgery.
The authors have no potential conflicts of interest