Chemotherapy for Advanced Pancreatic Cancer


The vast majority of patients with pancreatic cancer present with surgically unresectable locally advanced disease or metastases and have a median survival of less than six months. For those patients who do not have resectable disease, the benefits of chemotherapy have generally been thought of as sufficiently modest, and the toxicities sufficiently prevalent and noxious that many opt for the best supportive care rather than antitumor treatment. Gemcitabine was recently approved for use in patients with pancreatic cancer, not on the basis of antitumor effects, but because about one-fourth of treated patients show some objective clinical benefit (decrease in pain, weight gain, improved performance status). However, evidence is emerging that some regimens may have significant antitumor effects in a subset of patients.

Auerbach and colleagues recently published phase II data on a regimen consisting of protracted infusional 5-fluorouracil (5FU) 300 mg/m2/d for 70 days and carboplatin 100 mg/m2 weekly given on weeks 1-10 of a 12-week cycle. No therapy was given during weeks 11 and 12. Cycles were continued until disease progression. Among 54 patients with advanced pancreatic adenocarcinoma treated with this regimen, there were two complete and seven partial responses (overall response rate, 17%). Another 40% experienced stable disease for at least 12 weeks. Median overall survival was 22 weeks, and responders survived a median of 61 weeks. Grade 3 toxicities were encountered in 10% of patients. (Auerbach M, et al. Ann Oncol 1997;8:439-444.)


It is certainly understandable that some authors have advocated the abandonment of chemotherapy for patients with advanced pancreatic cancer. There have been prior instances of regimens that appeared active in one study1 but not in another.2 Furthermore, heterogeneous definitions of response in this difficult cancer have led to inconsistent results. However, several developments are changing the uniformly gloomy prospects of patients with advanced pancreatic cancer. Gemcitabine has been shown to improve quality of life without apparent effect on tumor regression.3 A matrix metalloproteinase inhibitor called marimastat has been suggested to have antitumor efficacy.4 Octreotide has been said to improve survival based on a small study of 32 patients who had failed prior therapy.5 Larger, randomized studies of octreotide are underway. The FAM regimen (5FU, doxorubicin, mitomycin C) has been shown to double the survival of patients compared to best supportive care.6 The taxanes and irinotecan may have antitumor activity in pancreatic cancer.

The regimen described by Auerbach and colleagues is notable for its significant antitumor effects. More than half of the treated patients had stable disease or tumor regression. Yet, the toxicities were mild—substantially milder than most regimens based upon bolus 5FU including FAM. This was not a randomized study; however, other studies using protracted infusions of 5FU have demonstrated similar results. For example, Nicolson et al7 have reported a 16% response rate and a 33% one-year survival rate using protracted infusion 5FU plus three weekly injections of cisplatin. The clear advantage of the Auerbach et al regimen would appear to be the use of carboplatin, which is certainly less likely to induce nausea than cisplatin.

The magnitude of the observed effects of protracted 5FU infusion plus carboplatin is not large, but it is clear that those who did not get a tumor response were not made worse by the treatment. Of course, this is likely due at least in part to their good performance status at the time treatment was begun—86% of patients were ECOG scale 0-1, 14% were ECOG 2. Pancreatic cancer is associated with malabsorption, gastric outlet obstruction, and cachexia that can rapidly diminish performance. Intervention with chemotherapy in seriously ill patients still seems unwise. However, those patients with preserved performance status may well enjoy good functional status longer with therapy such as prolonged 5FU infusion plus weekly carboplatin.


1. Mallinson CN, et al. BMJ 1980;281:1589-1591.

2. Cullinan S, et al. Cancer 1990;65:2207-2212.

3. Moore MJ, et al. Proc Am Soc Clin Oncol 1995;14:198.

4. Rosemurgy A, et al. Proc Am Soc Clin Oncol 1996; 15:A470.

5. Cascinu S, et al. Br J Cancer 1995;71:97-202.

6. Palmer KR, et al. Br J Surg 1994;81:882-885.

7. Nicolson M, et al. Ann Oncol 1995;6:801-804.