This phase II randomized trial (NCT01312350) was conducted to test the efficacy of IC with a TPF regimen followed by concurrent chemoradiotherapy compared to CRT alone. Eligible patients were age ≥ 18 years with histologically confirmed stage III or IV (M0 disease) hypopharyngeal or BOT cancer according to the American Joint Cancer Committee (AJCC) staging classification 7th edition and had one or more measurable lesions. No previous chemotherapy, radiotherapy or surgery was allowed. Other key eligibility criteria included an Eastern Cooperative Oncology Group (ECOG) performance status of ≤ 2 and adequate bone marrow, renal and hepatic function. Patients with distant metastasis were excluded. All patients provided informed consent. This study was approved by the Institutional Review Board of Samsung Medical Center (IRB no.: 2010-10-028).

Patients were randomly assigned to receive either IC followed by CRT (IC arm) or CRT only (CRT arm). IC consisted of two cycles of TPF regimen (docetaxel and cisplatin at 75 mg/m² each on day 1, and fluorouracil [5FU] at 750 mg/m² on D1 to 4, every 3 weeks), followed by CRT 3 to 5 weeks after the completion of IC. CRT was administered with cisplatin 100 mg/m² on day 1 every 3 weeks up to a total of two to three times. Patients received radiotherapy using three-dimensional conformal radiotherapy or intensity-modulated radiotherapy techniques delivered at doses of 63 to 72 Gy in 30 to 35 fractions in 2.0 to 2.4 Gy per fraction over 6 to 7 weeks. For IC + CRT treated patients, neck computed tomography (CT) and positron emission tomography (PET)-CT were performed to assess the response to IC before starting CRT 3 weeks after completion of the TPF chemotherapy. If a patient had disease progression after IC, they were was referred to a multidisciplinary head and neck tumor board for further treatment. After completion of CRT in both the IC and CRT arm, a neck CT was performed after 1month and then every 3 months for 1 year and every 6 months thereafter. PET-CT was performed after 4 months and repeated every 6 months for 2 years and every 2 years thereafter until disease progression or death. If patients showed disease progression during or after study treatment, further treatment could be administered at the discretion of the investigators.

The primary endpoint was PFS, defined as the time from randomization to progression, secondary cancer or death resulting from any cause. Secondary endpoints included ORRs, OS, and toxicity. OS was calculated from the time of randomization to the date of death. PFS was set to be primary endpoint due to a surrogate marker of OS. These endpoints were measured in all registered patients (i.e., the intention-to-treat population). Response Evaluation Criteria in Solid Tumors (RECIST version 1.1) and adverse events (AEs) were assessed according to National Cancer Institute criteria (CTCAE v3).

Survival rates were estimated by the Kaplan-Meier method and compared between the two treatment groups using the log-rank test. The sample size for the trial was derived from following assumption. Given a one-tailed alpha of 0.1 and power of 0.85, the 1-year PFS rate would be 65% and 80% in the CRT arm and IC arm, respectively, with an accrual rate of 29 patients per year and an additional follow-up period of 1.5 years. A 10% ineligible or non-assessable rate was assumed, resulting in the accrual goal of 100 total patients (50 patients for each arm). The significance of statistical tests was evaluated based on a one-tailed alpha value of 10% for the primary endpoint and a two-tailed alpha value of 5% for the other endpoints. Statistical analyses were performed using SPSS version 21 (IBM Co., Armonk, NY, USA). This trial was closed early due to slow recruitment.

Between December 2010 and April 2015, a total of 41 patients were screened. Five patients were excluded during the screening period. We randomly assigned 36 patients to receive IC followed by CRT or CRT only (Fig. 1). Patient baseline characteristics are shown in Table 1 and were well balanced between the two groups. About twothirds of all patients were diagnosed with hypopharyngeal cancer and the rest were BOT cancers. All had squamous cell carcinoma histology, 28 patients (78%) had N2 to N3 disease and approximately 80% of patients showed clinical stage IV (A/B) disease (AJCC 7th edition).

At the time of data analysis (February 2017), a total of 15 patients (42%) had disease progression or recurrent disease after partial response or complete response to treatment. After a median follow-up of 47.2 months, 1-year estimated PFS rates were 63.2% (95% confidence interval [CI], 0.415 to 0.848) in the CRT group and 76.5% (95% CI, 0.563 to 0.966) in the IC group, respectively. The median PFS was 26.8 months in the CRT arm and not reached in the IC arm (p = 0.131) (Fig. 2A). Two-year OS rates were 72.2% (95% CI, 0.515 to 0.929) and 86.9% (95% CI, 0.698 to 1.040) in the CRT and IC arms, respectively. Three-years after randomization, the survival curves widely separated with a plateau: 3-year PFS estimates were 45% and 68% (hazard ratio [HR], 0.55; 95% CI, 0.19 to 1.60), and 3-year OS estimates were 56% and 86% (HR, 0.35; 95% CI, 0.07 to 1.69), in the CRT and IC arms, respectively (Fig. 2B).

In 23 patients with hypopharynx cancer, IC tended to show more improvement in PFS compared to BOT cancer, although neither were statistically significant (p = 0.198 and p = 0.217) (Fig. 3). There was no significant difference in the overall incidence of distant failure between the two groups (p = 0.324). Two patients in the CRT arm and three in the IC arm initially progressed at distant metastatic sites.

The ORR to IC was 94.1%; one patient (5.8%) achieved complete response, 88.2% had partial responses and one patient (5.8%) had stable disease. None of the patients progressed on IC. After completion of CRT, among all 36 patients, 20 reached complete response (55.6%), 14 had partial responses (38.9%) and one had stable disease (2.8%). Overall response rates were 100% in the CRT group and 88.2% in the IC group (p = 0.216). Five patients underwent salvage operations (four CRT arm, one IC arm) and four patients received salvage chemotherapy after progression.

All patients were monitored for AEs and 94.4% (34/36) of them experienced an AE during the study period. There was only one case of grade ≥ 3 AE of stomatitis (5.9%) in the IC group during the CRT phase. The most common AEs experienced in both treatment arms were anemia, stomatitis, dry mouth, and dysphagia (Table 2). The addition of IC before CRT was relatively well tolerated and all AEs were manageable. Dose reductions of chemotherapy were required in 29.4% during IC and 5.9% during concurrent chemoradiotherapy only in the IC arm. Dose delays of at least one cycle of chemotherapy happened in 15.8% and 29.4% of patients during the concurrent chemoradiotherapy period in the CRT and IC arms, respectively (Table 3).

The total number of cisplatin chemotherapy cycles administered was 47 (mean, 2.5) in the CRT arm and 42 (mean, 2.5) in the IC arm. Two cycles of planned IC were performed in all patients on the IC arm. The delivered relative dose intensities of radiation and cisplatin for CRT were similarly high in both groups (98.9%, 98.8% in the CRT am and 93.5%, 92.8% in the IC arm, respectively) (Table 3). The mean cumulative dose of cisplatin during CRT was 242 mg/m² in the CRT arm and 213 mg/m² in the IC arm (p = 0.096). Three chemotherapeutic agents were given to IC group patients with quite high dose intensities (≥ 95%).