Role of Radiation Therapy in the Management of the Patient With Pancreatic Cancer

Introduction

Treatment strategies for patients with pancreatic cancer differ based on disease stage. Standard treatment strategies for patients with T1-T2,NX,M0 disease include surgical resection and adjuvant or, less commonly, neoadjuvant chemoradiation, whereas therapy for unresectable presentations includes surgical bypass procedures, radiation therapy, and/or chemotherapy [1].

Pancreatic cancer is particularly challenging for the radiation oncologist because of the limited radiation tolerance of adjacent organs in the upper abdomen, including the kidney, liver, stomach, small bowel, and spinal cord [2]. The radiotherapy technique usually used to treat pancreatic cancer is conventional external-beam radiotherapy (EBRT), although more specialized techniques, such as intraoperative radiotherapy (IORT) and brachytherapy, have been described. Intraoperative radiotherapy involves the application of a single high dose of radiation during a surgical procedure, while brachytherapy entails the interstitial implantation of radioactive sources [1,3].

Radiation therapy can be potentiated with the use of radiosensitizing chemotherapeutic agents. One such agent, fluorouracil(Drug information on fluorouracil) (5-FU), has been shown to be a radiosensitizer in vitro [4]. More recently, in vitro experiments reported by Shewach and Lawrence showed that the chemotherapeutic agent gemcita-bine (Gemzar), a deoxycytidine analog, produced increased radiation sensitivity in human colon carcinoma (HT-29) cells [5].

Resectable or Borderline Unresectable Disease

Preoperative Radiation Therapy

Pilepich and Miller examined preoperative (neoadjuvant) radiation therapy as a means of improving the efficacy of surgical resection in patients with borderline resectable or locally unresectable disease [6]. In this study, patients received 4,000 to 5,000 cGy over a 4.5- to 5-week period and were evaluated for surgery 6 weeks after radiotherapy. Of 17 patients, 11 were selected for laparotomy. At surgery, 4 of the 11 patients had metastatic disease, 1 patient was determined to be locally unresectable, and 6 patients underwent resection. Only 1 of the 6 patients experienced a clear conversion from unresectable to resectable disease. The authors concluded that radical resection was possible after preoperative radiation in some patients but that a significant number of patients should be expected to develop metastatic disease during or shortly after the radiotherapy course [6].

Ishikawa et al conducted a retrospective review of the use of preoperative irradiation in 18 patients with resectable or borderline resectable disease [7]. Patients received fractionated radiation to 5,000 cGy, terminating about 1 month prior to exploratory surgery. Measurable tumor shrinkage was noted in all 18 patients, and 16 patients underwent pancreatic resection [7]. Although these results indicate that preoperative radiation is feasible, it is not used routinely [8].

Resectable Disease

Postoperative Radiation Therapy

Historically, surgery alone has had disappointing results in patients with pancreatic cancer. For example, Tepper et al showed that surgery for resectable disease was associated with a 5-year crude survival rate of 15% and a local recurrence rate of 50% [9]. Similar results were reported by Griffin et al [10]. Based on these findings, it was suggested that postoperative radiation therapy might reduce the incidence of local failure after radical surgery [9,10].

A National Cancer Institute (NCI) study compared IORT with standard postoperative radiation therapy in patients with locally confined pancreatic cancer who were undergoing either total or regional pancreatectomy [11]. Patients were randomized to receive either postoperative IORT (2,000 cGy using 9 to 12 MeV to the tumor bed and regional nodal basins) or postoperative EBRT (5,000 cGy over a period of 5 to 6 weeks). Although overall survival was similar between the treatment groups, disease-free survival and local disease control were better in the IORT-treated patients than in the EBRT-treated patients. These results suggested that IORT may provide some benefit in patients with locally confined pancreatic cancer [11].

More recently, a retrospective study by Zerbi et al examined the impact of IORT following surgical resection [12]. Patients either underwent resection alone (47 patients) or resection and IORT at a dose of 12.5 to 20 Gy (43 patients). Whereas 1-, 2-, and 3-year survival rates and median disease-free survival did not differ significantly between the treatment groups, local recurrence was significantly reduced in patients who underwent resection plus IORT (27% of patients), as compared with those who had a resection only (56% of patients; P less than .01 by the chi-square and Mantel-Cox tests).

Although these results provide some support for the concept of adding of IORT to surgical resection to enhance local control, it should be noted that no survival benefit was observed. Selection factors may have contributed to the improved local control observed, and IORT remains investigational in this context [12].

Postoperative Radiation Therapy Plus Chemotherapy

In 1985, the Gastrointestinal Tumor Study Group (GITSG) reported the results of a randomized trial that compared surgical resection alone (N = 22) with surgical resection followed by radiation therapy plus chemotherapy (N = 21) [13]. Postoperative radiation therapy (EBRT) consisted of 4,000 cGy given in two courses of 2,000 cGy each, and chemotherapy consisted of IV 5-FU at 500 mg/m² daily during the first 3 days of each 2,000-cGy course of radiation. One month after the completion of radiation therapy, chemotherapy was continued weekly for up to 2 years or until recurrence.

Accrual into the trial was slow due to the concern that patients would not tolerate surgery plus radiation and chemotherapy [14]. Nevertheless, this study demonstrated that surgery plus adjuvant radiation and chemotherapy significantly improved survival compared with surgery alone (median survival, 20 vs 11 months; P = .03 using a one-sided log-rank test) [13].

The results of the initial GITSG trial led to the design of a confirmatory trial in which an additional 30 patients received surgery plus radiation and chemotherapy [15]. Median survival was 18 months, which was similar to that observed for the adjuvant therapy group (20 months) in the initial trial. Two-year actuarial survival was 46% for the confirmatory trial patients who received adjuvant therapy, 43% for the initial trial patients who received adjuvant therapy, and only 18% for the initial trial patients who received surgery alone (Figure 1). The results of both studies supported the benefit of surgery plus radiation and chemotherapy in patients with resected pancreatic cancer [16].

The GITSG concluded that "... the combined use of radiation therapy and fluorouracil as adjuvant therapy after curative resection is effective and is preferred to no adjuvant therapy [15]." However, the use of adjuvant radiation therapy and chemotherapy has been slow to gain acceptance. Reluctance to use this treatment regimen reflects a variety of concerns, including the possibility that patients with positive surgical margins may not benefit from this treatment regimen [17], and doubt surrounding the GITSG study results due to the low patient numbers (N = 43) and lengthy recruitment time [2].

However, more-recent data continue to support the GITSG results [18]. In addition, efforts at improving postoperative adjuvant radiation therapy by using hepatic radiation and continuous-infusion 5-FU have been initiated, but whether these efforts will result in improved outcomes with acceptable toxicity is unknown [19]. Decisions about whether or not to use adjuvant therapy for patients with resectable pancreatic cancer should be made on a case-by-case basis [1].