Adjuvant chemotherapy for gastric cancer or not: a dilemma?

In this issue of the Journal, Di Costanzo et al. ( 1 ) report the result of a randomized phase III trial of the Italian Oncology Group for Cancer Research. They conclude that despite very promising previous results in the setting of advanced gastric cancer ( 2 ), there are no positive effects of cisplatin, epirubicin, leucovorin, and 5-fluorouracil (PELF) as adjuvant chemotherapy on overall and disease-free survival of gastric cancer patients who have undergone complete surgical resection. In the trial, 258 patients were randomly assigned to chemotherapy or surgery alone over a 5-year period. Fifty-five percent of the patients underwent D2 or more extended nodal dissection, and the median number of lymph nodes harvested per patient was greater than 15. The study not only demonstrates that PELF as an adjuvant regimen cannot improve patient survival but also suggests that data obtained in the setting of advanced gastric cancer may not be applicable to the adjuvant setting. Here we do not think it meaningful to criticize the shortcomings of this study (including, as always, the low rate of D2 nodal dissection that is common to Western studies), but it would be helpful to rethink the strategies on which this and similar studies are based. The usefulness of adjuvant chemotherapy for gastric cancer has been debated for a long time. Randomized controlled clinical trials that investigated the effi cacy and toxicity of different adjuvant regimens among different gastric cancer patient groups have yielded inconsistent results ( 3 – 7 ), and meta-analyses of data regarding the benefi t of adjuvant chemotherapy for patients with gastric cancer have found hazards reduction of death to be 0.12 – 0.20 ( 8 – 11 ). We are still not able to draw a defi nite conclusion as to whether adjuvant chemotherapy confers advantages to gastric cancer patients who have undergone curative resection. It is appropriate to ask how these negative results can help patients ( 12 ) and how we as professionals in the fi eld of gastric cancer management can resolve the dilemma as to whether adjuvant chemotherapy should be used in gastric cancer patients after curative surgical resection. More and more studies on adjuvant chemotherapy of gastric cancer are forthcoming, and the dilemma persists. The negative overall result may refl ect a positive effect in one subgroup and a negative effect in another. Therefore, one strategy is to identify the subgroup(s) that actually benefi t from adjuvant chemotherapy. Di Costanzo et al. ( 1 ) analyzed the interaction between covariates and adjuvant treatment and showed that the benefi t of treatment, if any, was confi ned to older patients over 60 years old. In this article, the authors also reported some suggestive results from a parallel study that sought to determine the prognostic value of additional biomolecular factors (P53, c-erbB2, MIB-1, TS, and Herg1), and the strategy of using biologic markers to identify subgroups of patients that would benefi t from therapy has been pursued in other kinds of human cancers. The National Surgical Adjuvant Breast and Bowel Project investigated the value of high-degree microsatellite instability (MSI-H) as a marker to predict benefi t from adjuvant chemotherapy in colon cancer and found that there was a strong inverse relationship between MSI-H and mutant p53 status ( P < .001), although the data did not support the use of MSI-H as a predictive marker of chemotherapy benefi t. Hazard ratios (HRs) for relapse-free survival for MSI-H vs microsatellite-stable and low-degree MSI patients were 0.77 (95% confi dence interval [CI] = 0.40 to 1.48) in the untreated patient group and 0.60 (95% CI = 0.30 to 1.19) in the treated group ( 13 ). With molecular classifi cation, gastric cancer patients with a high risk of recurrence may eventually be identifi ed, just as breast cancer patients with a high risk of recurrence are now identifi ed based on HER-2/neu overexpression as candidates for specifi c adjuvant therapy regimens, as recommended by the National Comprehensive Cancer Network. It has been hypothesized that some cytoxic regimens or therapeutic models are not effective in the adjuvant setting. However, recently, Sakuramoto et al. ( 7 ) reported the very promising result of S-1, an oral fl uoropyrimidine, as adjuvant chemotherapy for stage II and III gastric cancer patients who underwent curative gastrectomy and extended (D2) lymph node dissection. A total of 1059 patients were randomly assigned to treatment with S-1 or surgery alone. The 3-year overall survival rate was 80.1% in the S-1 group and 70.1% among those treated with surgery alone. The hazard ratio for death in the S-1 group compared with the surgeryonly group was 0.68 (95% CI = 0.52 to 0.87, P = .003). Adverse events of grade 3 or 4 were not common ( 7 ). However, before we translate this promising result into clinical practice in Western countries, further clinical trials will be needed to confi rm its effi cacy in this new setting. Furthermore, subdivision of stage II and III gastric cancer patients to identify those most likely to benefi t from S-1 adjuvant chemotherapy is also needed, as suggested above. A US intergroup study has provided evidence as to the potential value of postoperative adjuvant chemoradiation in that treatment consisting of 5-fl uorouracil plus leucovorin combined with radiotherapy led to an improvement in overall survival relative to surgery alone ( 14 ). The median overall survival in the surgery-only