Multicenter phase II study of weekly docetaxel, cisplatin, and S-1 (TPS) induction chemotherapy for locally advanced squamous cell cancer of the head and neck

Patients were eligible if they had locally advanced stage III or IV squamous cell carcinoma of the larynx, oropharynx, or hypopharynx. Patients at least ≥18 years old with adequate bone marrow and organ function were included in this study (i.e., absolute neutrophil count ≥1,500/μL, platelets ≥100,000/μL, serum bilirubin <2.0 mg/dL, creatinine <1.5 mg/dL, and serum transaminase levels less than twice the upper limit of normal). Patients were excluded from the study if they had received previous chemotherapy. Other exclusion criteria included history of another malignancy; pregnancy or lactation; current or history of distant metastasis; history of clinically significant cardiac disease (serious arrhythmia, heart failure, myocardial infarction, or unstable angina) within the last 6 months; active serious infection; or a psychiatric illness that would preclude obtaining informed consent. Patients with nasopharyngeal carcinoma were also excluded from the study.

Pretreatment staging involved examination of the ears, nose, and throat by an otolaryngologist, as well as a computed tomographic (CT) scan or magnetic resonance imaging (MRI) of the primary tumor site and neck. To detect other primary aerodigestive tract malignancies, patients underwent a CT scan of the chest and an esophagogastroduodenoscopy or pharyngoesophagram. Before radiation therapy, all patients received a dental examination to avoid unexpected osteonecrosis or osteomyelitis associated with radiation.

All patients provided written informed consent before being enrolled in the study, which was approved by the institutional review board of each participating hospital.

Docetaxel (30 mg/m²) was given as a 1-h intravenous infusion on days 1 and 8. Cisplatin (60 mg/m²) was given as a 3-h intravenous infusion on day 1. S-1 was given orally twice daily, at 70 mg/m², for 14 consecutive days. The cycles were repeated every 3 weeks. Patients received further cycles of chemotherapy only when the absolute neutrophil count was ≥1,000/mm³ and the platelet count was ≥100,000/mm³. Toxicity was graded according to the National Cancer Institute Common Toxicity Criteria (NCI-CTC), version 3.0. Dose modifications were determined based on hematological and non-hematological toxicity. The dose of docetaxel was reduced by 20% after any episode of febrile neutropenia or grade 4 neutropenia lasting more than 5 days or for grade 4 thrombocytopenia and grade 3 or 4 non-hematological toxicity, except alopecia. If these complications developed after docetaxel dose adjustment, the S-1 dose in the next cycle was reduced by 20%. The dose of docetaxel on day 8 was interrupted in cases of grade 2 or higher neutropenia and thrombocytopenia and postponed to day 10. Docetaxel treatment on day 10 was withheld in cases of grade 4 neutropenia and grade 3 or 4 thrombocytopenia. All patients received pre- and post-cisplatin intravenous hydration. Standard intravenous premedications with dexamethasone, diphenhydramine, and ranitidine were administered 30 min before docetaxel infusion to prevent hypersensitivity reactions. A prophylactic antibiotic (levofloxacin 500 mg) was given orally for 5 days of each chemotherapy cycle. As supportive treatment for grade 4 neutropenia and grade 3 or 4 febrile neutropenia, granulocyte colony-stimulating factor (G-CSF) and antibiotics were given at the investigators’ discretion.

After three cycles of induction chemotherapy, patients had radiotherapy using 6 MV photon beams produced by linear accelerator having multi-leaf collimator. All patients had CT scanning for radiation treatment planning at supine position with a thermoplast immobilization mask. Primary tumor site and upper neck node bearing areas were irradiated by two parallel opposing fields with half beam technique, and lower neck and supraclavicular nodes by matched anterior one field. The spinal cord was shielded after 44 or 45 Gy, and then treatment fields were gradually reduced so that the volume containing the gross disease was irradiated with a curative dose. All patients were treated with standard radiotherapy technique in daily fraction of 1.8 or 2 Gy, 5 days per week. The gross neck node had a boost irradiation with a 9 MeV or 12 MeV electron beam. The primary tumor site and gross neck node area received 65 to 70 Gy over 7 to 8 weeks. A minimum of 45 Gy was delivered to clinically uninvolved neck nodes and supraclavicular nodes.

Definite irradiation was scheduled with concurrent administration of cisplatin in all patients except for those whose performance status or residual toxicities precluded the co-administration of chemotherapy. Cisplatin was given every 3 weeks at a dose of 100 mg/m², depending on creatinine clearance. Doses of cisplatin were delayed if there was evidence of dehydration, renal toxicity, neurotoxicity or ototoxicity. For patients with grades 3–4 mucositis or dysphagia, radiation therapy was delayed until recovery to less than grade 2 toxicities.

After three cycles of induction chemotherapy and 6–8 weeks after completion of CCRT, clinical responses were assessed. Patients underwent examination by an otolaryngologist, as well as CT or MRI imaging of the primary tumor and neck. A biopsy of the primary site was recommended if possible. Tumor response was assessed according to the Response Evaluation Criteria in Solid Tumors (RECIST). For all patients with a complete response (CR) on the physical examination and CT or MRI scans, an [¹⁸ F] fluorodeoxyglucose positron emission tomography (¹⁸ F-FDG-PET) scan was performed as a confirmation at 3 months after CCRT. When CCRT was completed, patients were observed on a monthly basis by a physician examination to evaluate the status of the disease and toxicity; CT or MRI scanning was performed every 3 months until disease progression. Toxicity was assessed according to the NCI-CTC, version 3.0. Quality of life (QOL) measures were assessed using the European Organization for Research and Treatment of Cancer Core Quality of Life questionnaire (EORTC QLQ C-30 version 3) at baseline and after three cycles of induction chemotherapy. During radiation therapy, the toxicities were evaluated by a physician every 2 weeks.

The dose intensity was calculated as the ratio of the total dose per square meter of body surface area divided by the total treatment duration (presented as mg/m²/week). In this calculation, the end of treatment was considered to be 21 days after day 1 of the last cycle of chemotherapy. The relative dose intensity was calculated as the ratio of the dose intensity actually delivered to that planned.

The primary endpoint was objective response, and secondary endpoints were safety, QOL, progression-free survival (PFS), and overall survival (OS). The study was conducted using a Simon’s two-stage phase II design. A sample size of 35 patients was required to accept the hypothesis that the true response rate was greater than 90% with an 85% power and to reject the hypothesis that the response rate was less than 70% with 5% significance. Initially we planned to enroll 18 patients in the first stage. If 13 or more responses were observed, we planned to continue to the second stage for a total of 32 patients in the analysis. Assuming a dropout rate of 10%, the total number of enrolled patients needed was calculated to be 35. The OS was measured from the start of chemotherapy until the date of death or the last confirmed date of survival. The PFS was defined as the time from the start of chemotherapy to the first appearance of progressive disease or death from any cause. The survival analysis was performed using the SPSS software (version 18.0; SPSS Inc., Chicago, IL, USA), and 95% confidence intervals (CIs) were calculated for all relevant estimates using StatXact (version 8; Cytel, Cambridge, MA, USA).

Based on the previously described Simon’s two-stage design, 18 patients who met the criteria were enrolled, and 15 patients (83.3%) achieved an objective response rate (ORR). Therefore, the study proceeded to the next stage, and a total of 35 patients were enrolled.

Patient baseline characteristics are given in Table 1. Thirty-five patients with locally advanced HNSCC were enrolled between October 2008 and October 2011 as the intention-to-treat (ITT) population. The enrolled population consisted of 33 males and two females, with a median age of 57 years (range, 29–72 years). Most patients (80%) had stage IV cancer; 20% had stage III HNSCC. The most common primary tumor site was the oropharynx (48.6%), followed by the hypopharynx (28.6%) and the larynx (22.9%). All 17 patients with oropharyngeal cancer were smokers. Twelve of these patients had evaluable specimens, none of which were positive for human papillomavirus (HPV) DNA by in situ hybridization (ISH).

A total of 97 cycles of weekly TPS therapy were given to 35 patients, and 30 completed the scheduled CCRT. Three patients received only one cycle and two patients received two cycles of TPS due to toxicological effects. The median duration from day 1 of cycle 1 to day 1 of cycle 3 of induction chemotherapy was 6.6 weeks (range, 5.8–9.4 weeks), and the median duration from day 1 of cycle 3 of chemotherapy to day 1 of CCRT was 5.3 weeks (range, 2.5–11.2 weeks). The mean dose intensities relative to the target dose of docetaxel, cisplatin, and TS-1 were 97.7%, 98.6%, and 97.7%, respectively. The most common reasons for a decrease in dose intensity were neutropenia and diarrhea. During CCRT, the mean dose intensities relative to the target dose of cisplatin was 69.6%. RT was interrupted in 10 patients (33.3%); of these, six had grades 3–4 mucositis, two had grade 2 renal toxicity, and two had grade 3 fatigue. However, RT was continued and completed in these patients after no more than 2 weeks of rest.

Thirty patients (85.7%) were evaluable for a response. The five patients that were not evaluable were included in the ITT analysis and kept in the denominator for calculation of the response rate. Five patients discontinued induction chemotherapy; three due to adverse events (grade 4 neutropenia or febrile neutropenia), one was lost to follow-up (this patient showed a near clinical CR), and one died of asphyxia associated with vocal cord paralysis due to his oropharyngeal cancer.

After three cycles of induction chemotherapy, 30 patients (85.7%) achieved an objective response (CR in 9 patients, 25.7% [95% CI 11.2–40.2%]; and a partial response [PR] in 21 patients, 60% [95% CI 43.8–76.2%]). Consistent response rates across primary tumor sites were observed in a subgroup analysis and included the oropharynx (CR 17.6%, PR 70.6%), hypopharynx (CR 30%, PR 50%), and larynx (CR 37.5%, PR 50%). After sequential CCRT, 22 patients (62.9%) achieved a CR (95% CI 46.8–78.9%) and eight (22.8%) achieved a PR (95% CI 8.9–36.8%) (Table 2). The mean follow-up duration was 30.3 months. The estimated 2-year PFS and OS rates were 73.2% (95% CI 52.4–93.6%) and 79.3% (95% CI 58.4–99.6%), respectively (Figure 1).

Global QOL was assessed at baseline and at 9 weeks (after 3 cycles of induction chemotherapy) using the EORTC QLQ C-30, version 3. Completed questionnaires were available for 30 patients (Table 3). There was a non-significant decrease in global QOL, from a median score of 56.1 at baseline to 53 at 9 weeks. Emotional functioning was significantly impaired, but other functional QOL scores (e.g., physical, role, cognitive, and social functioning) showed no significant differences. There were no significant differences in most treatment-related symptoms, except nausea and vomiting. Nausea and vomiting were significantly increased.

Among 30 patients who showed a complete or partial response after CCRT, five had locoregional recurrence. Four of the five patients showed a PR and one showed a CR after CCRT. All five patients showed initially advanced stage IV disease. Except one patient who showed PD among the patients showing CR, 21 patients had no recurrence until this analysis was done. Among the progressive patients from PR, One patient underwent salvage surgery and chemotherapy, three received only palliative chemotherapy, and one refused additional treatment. Distant relapse was not observed in this study.

The toxicity of weekly induction TPS chemotherapy was assessed in all 35 patients (Table 4). Grades 3–4 neutropenia occurred in 28.5% of patients, and 8.5% developed febrile neutropenia during induction chemotherapy. The most common grades 3–4 nonhematologic toxicities were diarrhea (17.1%), anorexia/nausea/vomiting (11.4%), and mucositis (8.5%). Anorexia/nausea/vomiting, fatigue/asthenia, mucositis, and diarrhea of less than grade 2 developed in 74.2%, 65.7%, 42.8%, and 42.8% of patients, respectively.