In this issue, Lanciano et al report the results of the Gynecologic Oncology Group’s most recent randomized study of concurrent chemoradiotherapy for cervical cancer (GOG-165). This trial, which compared the results of radiation therapy delivered concurrently with either weekly cisplatin or protracted venous infusion of fluorouracil (FU), was closed well before the planned accrual goal of 416 patients and 150 events because a scheduled interim analysis suggested that patients treated with FU were not likely to have a better outcome than those treated with the control regimen of radiation and weekly cisplatin. Although this negative study does not represent a major advance in the treatment of patients with cervical cancer, it does highlight many of the issues and challenges faced by investigators who seek to improve cervical cancer treatment in the modern era. GOG-165 is best understood in its historical context. The trial was designed and opened in 1997. In that year, the GOG published abstracts describing encouraging preliminary results from GOG trials of cisplatin-based concurrent chemoradiotherapy. However, the inclusion of hydroxyurea in the control arms of these trials and other perceived weaknesses, including problems with the quality of radiation therapy, prevented clinicians from reaching an early consensus about the trials’ importance. Therefore, in an effort to consolidate earlier findings, GOG-165 was designed. The study initially was planned with three arms: a control arm of modern radiation therapy alone, a second arm intended to confirm the value of adding weekly cisplatin, and a third, experimental arm that consisted of treatment with radiation and protracted venous infusion FU. This third arm drew on extensive data demonstrating the value of FU-based chemoradiotherapy in patients with gastrointestinal neoplasms and was intended to follow-up on an inconclusive but encouraging Canadian study of concurrent radiation and protracted venous infusion FU in patients with cervical cancer. However, in 1999, just 18 months after GOG-165 was opened, the near-simultaneous publication of results from five prospective randomized trials, all of which demonstrated significant improvements in local control and survival with cisplatin-based chemoradiotherapy, led to a National Cancer Institute alert establishing chemoradiotherapy as a standard treatment for patients with locoregionally advanced cervical cancer. At that point, the GOG felt compelled to drop the radiationonly control arm of GOG-165 but continued to accrue patients to the two concurrent chemoradiotherapy arms. The results of GOG-165 suggest that radiation plus protracted venous infusion FU was not a better treatment than radiation plus weekly cisplatin, but it does not prove that FU is ineffective as a radiation sensitizer in patients with cervical cancer. More importantly, the study does not answer the potentially more interesting question of whether FU can enhance the beneficial effects of adding cisplatin to radiation therapy. In fact, although randomized trials have demonstrated the benefit of concurrent cisplatin and of concurrent cisplatin with FU, no trial has yet compared the two regimens directly. Although the GOG regimen of radiation plus weekly cisplatin tends to be favored by gynecologic oncologists in the United States, data from the seven randomized trials suggest that radiation plus a combination of cisplatin and FU might be more effective. Of four randomized trials that included radiation therapy plus weekly cisplatin, only one, reported by Keys et al in 1999, demonstrated a benefit for radiation therapy plus cisplatin over radiation therapy alone. In that study, patients with bulky stage IB cancers were randomly assigned to prehysterectomy radiation therapy or chemoradiotherapy. Two other trials—the current trial and the GOG three-arm trial published in 1999—did not have a control arm of radiation therapy only; instead, they included chemotherapy of uncertain value in the noncisplatin-containing JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 23 NUMBER 33 NOVEMBER 2