Neoadjuvant Radiation Is Associated With Improved Survival in Patients With Resectable Pancreatic Cancer: An Analysis of Data From the Surveillance, Epidemiology, and End Results (SEER) Registry

Purpose

Cancer of the exocrine pancreas is the fifth leading cause of cancer death in the United States. Neoadjuvant chemoradiation has been investigated in several trials as a strategy for downstaging locally advanced disease to resectability. The aim of the present study is to examine the effect of neoadjuvant radiation therapy (RT) vs. other treatments on long-term survival for patients with resectable pancreatic cancer in a large population-based sample group.

Methods and Materials

The Surveillance, Epidemiology, and End Results (SEER) registry database (1994–2003) was queried for cases of surgically resected pancreatic cancer. Retrospective analysis was performed. The endpoint of the study was overall survival.

Results

Using Kaplan-Meier analysis we found that the median overall survival of patients receiving neoadjuvant RT was 23 months vs. 12 months with no RT and 17 months with adjuvant RT. Using Cox regression and controlling for independent covariates (age, sex, stage, grade, and year of diagnosis), we found that neoadjuvant RT results in significantly higher rates of survival than other treatments (hazard ratio [HR], 0.55; 95% confidence interval, 0.38–0.79; p = 0.001). Specifically comparing adjuvant with neoadjuvant RT, we found a significantly lower HR for death in patients receiving neoadjuvant RT rather than adjuvant RT (HR, 0.63; 95% confidence interval, 0.45–0.90; p = 0.03).

Conclusions

This analysis of SEER data showed a survival benefit for the use of neoadjuvant RT over surgery alone or surgery with adjuvant RT in treating pancreatic cancer. Therapeutic strategies that use neoadjuvant RT should be further explored for patients with resectable pancreatic cancer.

Introduction

Adenocarcinoma of the exocrine pancreas ranks as the fifth most common cause of cancer death in the United States, with approximately 32,000 deaths each year (1). The overall 5-year survival rate is estimated to be less than 5%, and no significant improvement in survival rates has been observed over the past 25 years (1).

Surgical resection is considered to be the only potentially curative treatment and is recommended to approximately 15% to 20% of all patients with pancreatic adenocarcinoma (2). The most commonly performed resection, known as the Whipple procedure, involves removal of the proximal jejunum, gallbladder, common bile duct, and distal stomach, in addition to pancreaticoduodenectomy (3). However, 5-year survival rates after resection remain exceedingly poor, with locoregional or distant recurrence observed in more than 80% of patients 4, 5, 6, 7.

In 1969, Moertel et al. published the first study investigating the use of 5-fluorouracil (5-FU) in combination with radiation therapy (RT) as multimodality treatment for patients with locally advanced pancreatic cancer (8). Since then, chemotherapy and RT have been commonly used for palliation in cases in which surgical resection was not deemed possible. In recent years, chemoradiotherapy was also used as postoperative adjuvant therapy for patients with resectable disease, but the effectiveness of this approach in improving survival outcomes remains controversial (9). Although the use of multimodality therapy has been found to confer a survival advantage in both prospective 10, 11 and retrospective (6) studies, results from a recent randomized trial indicated no advantage to using adjuvant chemoradiotherapy compared with adjuvant chemotherapy alone 12, 13.

The use of preoperative, or neoadjuvant, chemoradiotherapy has been explored as a way to downstage locally advanced disease to allow for surgical resection. Several trials have reported benefits of using this approach to achieve local disease control, to increase resectability rates, and to decrease rates of distant metastases 14, 15, 16, 17. Yet, because of the small sample size and short follow-up in these single-institution studies, no conclusions could be reached regarding the effect of preoperative radiation on long-term survival.

In a recent report, the effects of pre- or postoperative RT on patient survival were compared using the Surveillance, Epidemiology, and End Results (SEER) registry of the National Cancer Institute from 1988 to 2002 (18). The authors found a survival advantage for patients receiving RT but no statistically significant difference between pre- and postoperative RT. Importantly, only 23 of 3008 patients in that study were treated with neoadjuvant radiation. Here we re-examine the results using SEER data from 1994 to 2003. In our analysis we discovered that data sets from 1988 to 1993 contained no cases in the neoadjuvant treatment group. On the other hand, the inclusion of data from 2003 increased the number of patients in the neoadjuvant group to 70. Also, unlike the authors of the previous study, we focus on those patients whose disease was found to be resectable at diagnosis. We exclude those with locally advanced or metastatic (Stage III or Stage IV) pancreatic cancer, because for these patients “surgery” is likely to represent an incisional biopsy or a palliative procedure rather than a curative resection.

Thus the purpose of this study is to reassess the effect of preoperative RT on survival outcomes in patients with resectable pancreatic cancer. Using SEER data from 1994 to 2003, we compare survival in patients who received preoperative RT with those who received postoperative RT, as well as to those who underwent surgical resection without receiving radiotherapy.