A phase III German trial also compared induction chemotherapy (three courses of bolus 5-FU, leucovorin, etoposide, and cisplatin) followed by chemoradiation therapy (cisplatin, etoposide, and 40 Gy) followed by surgery (arm A), or the same induction chemotherapy followed by chemoradiation therapy (at least 65 Gy) without surgery (arm B) for patients with T3 or T4 squamous cell carcinoma of the esophagus.[Level of evidence: 1iiA] OS was the primary outcome. The analysis of 172 eligible, randomly assigned patients showed that OS at 2 years was not statistically significantly different between the two treatment groups (arm A: 39.9%; 95% CI, 29.4%–50.4%; arm B: 35.4%; 95% CI, 25.2%–45.6%; log-rank test for equivalence with 0.15, P < .007). Local progression-free survival (PFS) was higher in the surgery group (2-year PFS, 64.3%; 95% CI, 52.1%–76.5%) than in the chemoradiation therapy group (2-year PFS, 40.7%; 95% CI, 28.9%–52.5%; HR for arm B vs. arm A, 2.1; 95% CI, 1.3–3.5; P < .003). Treatment-related mortality was higher in the surgery group compared with the chemoradiation therapy group (12.8% vs. 3.5%, respectively; P < .03).
Preoperative Chemotherapy Alone
The effects of preoperative chemotherapy are being evaluated in randomized trials, as was done in the NCT00525785 trial.[21,22][Level of evidence: 1iiA]. An Intergroup trial randomly assigned 440 patients with local and operable esophageal cancer of any cell type to three cycles of preoperative 5-fluorouracil (5-FU) and cisplatin followed by surgery and two additional cycles of chemotherapy versus surgery alone. After a median follow-up of 55 months, there were no significant differences between the chemotherapy/surgery and surgery-alone groups in median survival (14.9 months and 16.1 months, respectively) or 2-year survival (35% and 37%, respectively). The addition of chemotherapy did not increase the morbidity associated with surgery. The Medical Research Council Oesophageal Cancer Working Party randomly assigned 802 patients with resectable esophageal cancer also of any cell type to two cycles of preoperative 5-FU and cisplatin followed by surgery versus surgery alone. At a median follow-up of 37 months, median survival was significantly improved in the preoperative chemotherapy arm (16.8 months vs. 13.3 months with surgery alone; difference 3.5 months; 95% CI, 1–6.5 months), as was 2-year OS (43% and 34% respectively; difference 9%; 95% CI, 3–14 months). The interpretation of the results from both of these trials is challenging because T or N staging was not reported and prerandomization and radiation could be offered at the discretion of the treating oncologist.