To the Editor:

Ajani et al¹ from the Radiation Therapy Oncology Group (RTOG) 0113 trial should be congratulated for their efforts in running a randomized trial of nonoperative therapies dedicated to patients with esophageal cancer (EC).

In addition to failing to detect sufficient efficacy in their two experimental treatments (arm A received fluorouracil, cisplatin, and paclitaxel and then fluorouracil plus paclitaxel with 50.4 Gy radiation; arm B received paclitaxel plus cisplatin and then the same chemotherapy with radiation), they found that both treatments resulted in an unacceptably high level of morbidity, with more than 80% of patients experiencing severe toxicities in both treatment arms. Most of these adverse events were hematologic. Although we are fully aware that it is easier to comment on the final report of a randomized trial than it is to effectively run this kind of trial, we would like to pose the following question: Would this unexpectedly high rate of severe hematologic toxicities not have been anticipated? The following assumptions favor our hypothesis: first, paclitaxel plus cisplatin is a known myelosuppressive regimen²; second, because patients in arm B received a higher dose-intensity of paclitaxel during radiation than did patients treated with regimen A (60 v 50 mg/m²/wk), it is not surprising that they experienced a higher rate of severe hematologic toxicities (69% v 38%); third, compared with routine treatment, an increased number of blood counts (not detailed in the report by Ajani et al) in trials—and by consequence, an increased likelihood of detecting hematologic adverse events—is not uncommon; and fourth, in the late 1990s, investigators showed that paclitaxel-based preoperative treatment for EC produced increased rates of severe neutropenia, febrile neutropenia, and grade 4 thrombocytopenia, compared with a previous cohort from the same institution treated with a fluorouracil-based regimen. These factors led the investigators to stop the use of paclitaxel in their programs dedicated to EC.³

Although Ajani et al¹ concluded that their two regimens were not efficient or safe enough to be selected as the experimental arm of an additional randomized trial versus the standard RTOG regimen (fluorouracil, cisplatin, and concurrent radiation),⁴ their data are conclusive enough to suggest that fluorouracil remain a major component of any radiochemotherapy regimen.¹ Sharing the same view on the key role of fluorouracil, and hypothesizing that oxaliplatin may have a better safety and efficacy profile than cisplatin, we launched a phase II/III trial comparing the standard RTOG regimen with a combination of fluorouracil, folinic acid, and oxaliplatin with radiation (50 Gy; 2 Gy per fraction, five fractions per week) in patients with EC not suitable for surgery. The phase II randomized part of our trial (n = 97; squamous cell cancers, 82%; stage III to IVA, 63%) provided some interesting results.⁵ It is important to note that of 52 patients treated with fluorouracil, folinic acid, and oxaliplatin and radiation, seven (13%), 10 (19%), and six (12%) patients presented with grade 3 thrombocytopenia, grade 3 neutropenia, and febrile neutropenia, respectively. In addition to a good safety profile, our experimental arm demonstrated a high complete response rate (assessed by an independent data monitoring committee) and promising median survival rate, compared with the standard RTOG regimen (21 of 47, 45%; 95% CI, 30 to 60 v 12 of 40, 30%; 95% CI, 17 to 46; and 14.9 v 22.7 months, respectively). The phase III part of our trial is ongoing. It has been designed to detect a 20% absolute difference in event-free survival between our two schedules. To comply with this objective, 266 patients must be included.

Therefore, together with Ajani et al,¹ we believe that “fluorouracil should continue to be incorporated in the treatment of localized esophageal cancer,” at least for a while.

AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The author(s) indicated no potential conflicts of interest.