As reported by Hideki Ueno, MD, PhD, of National Cancer Center Hospital, Tokyo, and colleagues in Journal of Clinical Oncology, treatment with the oral fluoropyrimidine derivative S-1 was associated with noninferior overall survival and reduced hematologic toxicity compared with gemcitabine treatment in a phase III trial (the GEST study) in patients with advanced pancreatic cancer in Japan and Taiwan.¹ Gemcitabine plus S-1 provided no survival advantage over gemcitabine alone, although benefit was observed in some patient subgroups with the combination.

Study Details

In the trial, 832 chemotherapy-naive patients with locally advanced or metastatic pancreatic cancer were randomized to receive gemcitabine (n = 277), S-1 (n = 280), or the combination (n = 275). Gemcitabine alone was given at 1,000 mg/m² on days 1, 8, and 15 of a 28-day cycle, and S-1 alone was given orally at 80, 100, or 120 mg/d based on body surface area on days 1 to 28 of a 42-day cycle. In combination, gemcitabine was given at the same dose on days 1 and 8, and S-1 was given at 60, 80, or 100 mg/d on days 1 to 14 of a 21-day cycle.

The primary outcome measure was overall survival, with the study testing the noninferiority of S-1 to gemcitabine alone and the superiority of the combination to gemcitabine alone.

The patient groups were well balanced for baseline characteristics. For the gemcitabine, S-1, and combination groups, 61%, 61%, and 57.5% of patients were male; 52%, 48%, and 50% were aged 65 years or older; 65%, 64%, and 62.5% had ECOG performance status of 0; 76%, 76%, and 75% had metastatic disease; 98%, 99%, and 99% had adenocarcinoma; 92%, 94%, and 90% had not undergone pancreas excision; and 73%, 77.5%, and 76.0% did not have biliary drainage.

Overall Survival Outcomes

The median durations of treatment were 2.6 months in the gemcitabine group, 2.6 months in the S-1 group, and 4.3 months in the combination group. The median duration of follow-up for surviving patients was 18.4 months. Median overall survival was 8.8 months in the gemcitabine group, 9.7 months in the S-1 group (hazard ratio [HR] = 0.96, P < .001 for noninferiority compared with gemcitabine), and 10.1 months in the combination group (HR = 0.88, P = .15 compared with gemcitabine).

Subgroup analyses of overall survival according to sex, age, extent of disease, ECOG performance status (0 or 1), and biliary drainage (no or yes) showed no significant interactions between S-1 treatment and gemcitabine treatment. Although the P values for interaction were not significant, combination therapy showed favorable hazard ratios compared with gemcitabine alone for the subset of patients with locally advanced disease (HR = 0.67, 95% confidence interval [CI] = 0.46–0.99) and the subset of patients with ECOG performance status of 1 (HR = 0.69, 95% CI = 0.51–0.92).

Median progression-free survival in the S-1 group was noninferior to that in the gemcitabine group (3.8 vs 4.1 months, HR = 1.09, P = .02 for noninferiority). Median progression-free survival in the combination group (5.7 months) was significantly longer than that in the gemcitabine group (HR = 0.66, P < .001). Objective response rates were 13.3% with gemcitabine, 21.0% with S-1 (P = .02 vs gemcitabine), and 29.3% with the combination (P < .001 vs gemcitabine).

Second-line chemotherapy was given to 66% of patients in the gemcitabine group, 66% of S-1 patients, and 62.5% of combination patients. Treatments such as erlotinib (Tarceva), oxaliplatin, and irinotecan were not approved for use in pancreatic cancer in Japan or Taiwan at the time of the study. Thus, second-line therapy consisted primarily of S-1 or S-1–based regimens in the gemcitabine group and gemcitabine or gemcitabine-based regimens in the S-1 group. Second-line therapy consisted primarily of gemcitabine alone, gemcitabine and S-1 in combination, or S-1 alone in the combination group.

S-1 Better Tolerated

Of grade 3 or higher adverse events, patients in the S-1 group had significantly (all P ≤ .01) lower rates of leukopenia (3.7% vs 18.7%), neutropenia (8.8% vs 41.0%), thrombocytopenia (1.5% vs 11.0%), increased ALT (5.9% vs 15.0%), and increased AST (7.7% vs 15.0%), and a significantly higher rate of diarrhea (5.5% vs 1.1%) compared with patients in the gemcitabine group. Compared with the gemcitabine group, patients in the combination group had significantly (all P ≤ .02) higher rates of leukopenia (37.8%), neutropenia (62.2%), rash (4.1% vs 0.7%), diarrhea (4.5%), mucositis/stomatitis (2.2% vs 0%), and vomiting (4.5% vs 0.7%).

The authors note that it is uncertain whether the results of this study can be extrapolated to Western patients, since the pharmacokinetics and pharmacodynamics of S-1 may differ between Western and East Asian patients.

The authors concluded, “[T]his study has verified the noninferiority of S-1 to gemcitabine, thereby suggesting that S-1 can be used as first-line therapy for locally advanced and metastatic pancreatic cancer. Because S-1 was confirmed to be a key treatment for pancreatic cancer, S-1–based regimens are expected to be developed in the future to improve the management of this formidable disease.” ■

Disclosures as published in J Clin Oncol May 1, 2013 Vol. 31 No. 13 1640-1648. Early release online April 1, 2013.

Reference

1. Ueno H, Ioka T, Ikeda M, et al: Randomized phase III study of gemcitabine plus S-1, S-1 alone, or gemcitabine alone in patients with locally advanced and metastatic pancreatic cancer in Japan and Taiwan: GEST study. J Clin Oncol. April 1, 2013 (early release online).