Long-term outcomes of induction chemotherapy followed by chemoradiotherapy using volumetric-modulated arc therapy as an organ preservation approach in patients with stage IVA-B oropharyngeal or hypopharyngeal cancers

Abstract The present study aimed to analyze treatment outcomes after induction chemotherapy followed by chemoradiotherapy (CRT) using volumetric-modulated arc therapy (VMAT) in patients with stage IVA-B oropharyngeal cancer (OPC) or hypopharyngeal cancer (HPC), with long-term observation, including examination of larynx preservation. A total of 60 patients with stage IVA-B OPC or HPC, who underwent induction TPF chemotherapy (a combination regimen consisting of docetaxel, cisplatin, and 5-fluorouracil) followed by CRT using VMAT were analyzed. Overall survival (OS), progression-free survival (PFS), laryngoesophageal dysfunction-free survival (LEDFS), and locoregional control (LRC) were calculated and compared. Univariate and multivariate analyses were performed to determine statistical differences in OS and LEDFS. The median follow-up period at the time of evaluation was 61 months. Twenty-six (43%) patients had OPC and 34 (57%) had HPC. The 5-year OS, PFS, LEDFS, and LRC rates were 57%, 52%, 52%, and 68%, respectively. Response to TPF therapy was the only significant predictive factor of OS and LEDFS in multivariate analyses. Regarding long-term toxicities, grade ≥ 2 late toxicities accounted for 15%. No patients experienced grade ≥ 3 xerostomia, and 5% of all patients developed grade 3 dysphagia. With long-term observation, the OS, PFS, and LEDFS rates were relatively good, and the incidence of late toxicities was low. TPF followed by CRT using VMAT was feasible and more effective in those who responded to induction chemotherapy.


INTRODUCTION
Radiation therapy is a well-established standard treatment for locally advanced head and neck squamous cell carcinomas, and induction chemotherapy followed by chemoradiotherapy (CRT) is one of several treatment options [1]. Although induction chemotherapy does not demonstrate a survival benefit compared with CRT alone [2], interest in the utility of induction chemotherapy persists for a few reasons including a reduction in the likelihood of distant metastases, and improvement of local regional control and organ preservation.
In Japan, several studies have investigated the feasibility and efficacy of induction chemotherapy followed by CRT for locally advanced head and neck carcinomas [3][4][5][6][7][8]. These studies reported good treatment • 554 outcomes after induction chemotherapy followed by CRT, including organ preservation. However, the follow-up duration was only 18-43.3 months; hence, the long-term outcome of this treatment strategy in the Japanese population remains unclear. Long-term observation is important to accurately assess treatment outcome(s) such as larynx preservation rate and swallowing function. In addition, current radiation techniques, such as intensity-modulated radiation therapy (IMRT) and image-guided radiation therapy, were not considered in previous randomized trials.
In our institution, induction chemotherapy followed by CRT has been indicated for stage IVA-B head and neck cancer from 2009. Therefore, in the present study, we analyzed treatment outcomes after induction chemotherapy followed by CRT using volumetric-modulated arc therapy (VMAT) in patients with stage IVA-B oropharyngeal cancer (OPC) or hypopharyngeal cancer (HPC) with long-term observation, including an examination of laryngeal dysfunction.

Patients
Data from patients, who were diagnosed with locally advanced OPC or HPC and underwent definitive radiation therapy (RT) at Hiroshima University Hospital (Hiroshima, Japan) between July 2009 and June 2017, were retrospectively analyzed. Individuals who fulfilled the following criteria were included in the present study: histological diagnosis of squamous cell carcinoma of the oropharynx or hypopharynx; stage IVA-B disease and no distant metastasis; underwent induction chemotherapy consisting of a combination regimen of docetaxel, cisplatin, and 5-fluorouracil (i.e., TPF therapy); and definitive CRT using VMAT. Patients with a follow-up duration < 2 years without death were excluded from this study.
Before induction chemotherapy, all patients underwent clinical evaluation including assessment of medical history, physical examinations and laboratory investigations, endoscopy, ultrasound examination, and radiographic studies. Based on the above information, we discuss the treatment of each head and neck cancer patient on the cancer board by multidisciplinary team for head and neck cancer. Laboratory investigations included complete blood cell count, liver function, renal function, and measurement of electrolyte levels. Radiographic studies included contrast-enhanced computed tomography (CT), 18 F-fluorodeoxyglucose positron emission tomography-CT (PET/CT), and magnetic resonance imaging (MRI). Almost 60% of patients with OPC did not undergo human papillomavirus (HPV) status testing. The clinical TNM stage was defined according to the Tumor Node Metastasis classification (Union for International Cancer Control, 7th Edition). The study was approved by the Human Ethics Review Committee of Hiroshima University Hospital.

Radiation therapy
Before treatment, the heads of all patients were non-invasively immobilized using a thermoplastic head-neck-shoulder mask, and subjected to contrast-enhanced CT. CT images were acquired at a slice thickness of 2.5 mm and imported to the Eclipse treatment planning system (Varian Medical Systems, Palo Alto, CA, USA) for VMAT planning (Rapidarc; Varian Medical Systems, Palo Alto, CA, USA). Gross tumor volume (GTV) was based on clinical, endoscopic, and radiological findings according to CT, MRI, and PET/CT. GTV was determined based on radiological findings before induction chemotherapy. Clinical target volume (CTV) 70 was generated by adding a 5-10 mm margin from GTV, including the primary tumor and involved lymph nodes. CTV63 included the high-risk areas and was created individually based on primary tumor site and lymph node metastases. CTV56 included elective nodal regions. Planning target volume (PTV) 70, PTV63, and PTV56 were generated by adding a 5-10 mm margin from CTV70, CTV63, and CTV56. Usually, 2 or 3 axial coplanar arcs were used for VMAT. All treatment plans were designed based on a TrueBeam linear accelerator equipped with 5 mm leaf-width multi-leaf collimators (Varian Medical Systems, Palo Alto, CA, USA). All plans were normalized so that PTV70 D95 (the dose that covers 95% of the PTV70) was equal to 70 Gy using 6-10 MV photon beams. PTV70, PTV63, and PTV56 received a total dose of 70 Gy, 63 Gy, and 56 Gy in 35 fractions, respectively. Cone-beam CT was performed daily for patient set-up and positioning verification.

Chemotherapy
All patients underwent induction chemotherapy followed by CRT. The medication administered was TPF, a combination chemotherapy regimen consisting of docetaxel (70 mg/m 2 ) on day 1, cisplatin (70 mg/m 2 ) on day 4, and 5-fluorouracil (750 mg/m 2 ) by 24 h continuous infusion for 5 days. Two cycles at 28-day intervals were planned.
The concurrent chemotherapy regimen was 3 cycles of cisplatin 100 mg/m 2 at an interval of 3 weeks. If the administration of cisplatin was not suitable due to patient general condition, renal function, or age, carboplatin or cetuximab was administered.

Statistical analysis
The Kaplan-Meier method was used to calculate overall survival (OS), progression-free survival (PFS), laryngoesophageal dysfunction-free survival (LEDFS), and locoregional control (LRC). OS was defined as the date of initiation of CRT to the date of the final follow-up or death from any cause. PFS was defined as the date of initiation of CRT to the date of any tumor progression or death from any cause. LRC was estimated as the CRT start date until the date of any locoregional progression; patients without locoregional progression at the time of death were censored. In estimating LRC and PFS, primary tumor progression was defined as enlargement of the tumor from its state after CRT, as observed by CT or endoscopic findings, even in patients who did not achieve complete response following CRT. In patients in whom the tumor remained evident after CRT and underwent subsequent surgery in which the residual tumor was pathologically assessed, it was included as an event of local progression. LEDFS was analyzed to evaluate organ preservation in the cohort. For LEDFS, the Larynx Preservation Consensus Panel recommends including death, local recurrence, total or partial laryngectomy, tracheotomy at ≥2 years, or feeding tube at ≥2 years [9]. Swallowing function was evaluated with videofluorography. Response to induction chemotherapy was divided into two groups according to the Response Evaluation Criteria in Solid Tumors: responder and non-responder [10]. A responder was defined that both the primary tumor and involved lymph nodes showed a partial or complete response. A non-responder was defined that the primary tumor and/or involved lymph nodes showed stable or progressive disease. The χ2 test was conducted to determine the significant differences between responders and non-responders. Univariate analyses using the Cochran-Mantel-Haenszel test and multivariate analyses using the Cox proportional hazards model were performed to compare and determine statistical differences in OS and LEDFS. BellCurve version 3.20 (Social Survey Research Information Co., Ltd., Tokyo, Japan) for Excel (Microsoft Corporation, Redmond, WA, USA) was used to perform statistical analyses. Differences with p < 0.05 were considered to be statistically significant. Treatment-related toxicities were evaluated according to the Common Terminology Criteria for Adverse Events ver 4.0.
The 5-year LRC rate was 67.7% (95% CI 55.8-79.7%). The 5-year LRC rates in those with OPC and HPC were 60.7% (95% CI 41.7-79.8%) and 73.1% (95% CI 58.1-88.2%), respectively.  Table 3. Response to TPF therapy was the only significant predictive factor of OS and LEDFS in the univariate analysis, and RT, radiation therapy * Five patients did not receive cisplatin concurrent with RT, and 45 responders and 10 non-responders received cisplatin with RT. remained significant in the multivariate analysis. The OS and LEDFS rates of responders and non-responders are presented in Figure 2.

DISCUSSION
Concurrent CRT is regarded to be the standard treatment option for patients with locally advanced head and neck cancers and one of several organ preservation strategies [1,2]. However, this intensification of treatment has resulted in increased short-and late-term adverse events, with a previous study reporting that almost 50% of patients with locally advanced head and neck cancers treated with concurrent CRT experienced long-term dysphagia [11]. In addition, the long-term results of the Radiation Therapy Oncology Group (RTOG) 91-11 trial demonstrated an unexplained increase in deaths unrelated to cancer in patients who underwent concurrent CRT, and indicated that laryngeal dysfunction may be associated with death [2]. Therefore, the preservation of larynx morphology and function is an important focus of contemporary studies, and the Larynx Preservation Consensus Panel has   recommended LEDFS as a new endpoint. Induction chemotherapy is expected to improve larynx preservation. In Japan, several studies have described treatment outcomes after induction chemotherapy followed by CRT, including organ preservation [3][4][5][6][7][8]. However, the follow-up duration was relatively short. Long-term observation is important to accurately assess laryngeal dysfunction. Although long-term results of previous randomized trials have been reported, these studies lacked the examination of LEDFS and current radiation techniques such as IMRT [12,13]. Therefore, long-term outcomes, including LEDFS, of induction chemotherapy followed by CRT using current radiation techniques remain unclear. To address this concern, we analyzed longterm outcomes after induction chemotherapy followed by CRT using current radiation techniques in patients with stage IVA-B OPC or HPC. In our study, the median follow-up period was 61 months. The 5year OS, PFS, LEDFS, and LRC rates were 57%, 52%, 52%, and 68%, respectively.
The clinical results of previous studies investigating CRT for locally advanced head and neck cancers are shown in Table 5. Although most of these studies included stage III head and neck cancers, our study included stage IV disease [7,[12][13][14][15]; nevertheless, our results demonstrated similar efficacy. Regarding larynx preservation, few studies have addressed LEDFS after CRT in patients with locally advanced OPC or HPC with long-term observation. The 5-year LEDFS rate in laryngealpreservation protocols for locally advanced laryngeal cancer was 30-45% [16,17]. The 5-year LEDFS rate in the present study was 52% and was relatively good, even with long-term observation despite the fact that all patients had stage IV disease.
In this study, response to TPF therapy was the only significant predictive factor of OS and LEDFS. Several previous studies have reported that response to induction chemotherapy may be a predictive factor of treatment outcome after CRT for locally advanced head and neck cancers [18]. Accordingly, Gorphe et al. altered the treatment strategy after induction chemotherapy depending on response [16]. Our results also supported their investigations, and we believe that nonresponders to induction chemotherapy should undergo surgery. On the other hand, age, sex, and TNM stage, which have been reported to be correlated with survival, were not significant in this study [19][20][21], although this could simply be explained by the small sample size. We also assessed long-term toxicities. Grade ≥ 2 late toxicities accounted for only 15%, and most other patients experienced asymptomatic or grade 1 toxicities. In the GORTEC 2000-01 trial, grade ≥ 3 late toxicities of the salivary gland accounted for 7% in the TPF arm [13]. In the long-term results of the TAX324 trial, 3% of patients in the TPF treatment arm were dependent on a gastric feeding tube [12]. In our study, no patients experienced grade ≥ 3 xerostomia and grade 2 dysphagia, and only 5% of all patients developed grade 3 dysphagia and required a gastric feeding tube. The incidence of late toxicities in our study were acceptable and lower than that in the RTOG 91-11 trial not using IMRT [2]. Tumor shrinkage by induction chemotherapy and radiation dose reduction in VMAT for organ at risk may have contributed to this result. In addition, deaths unrelated to cancer in this study were only 3 cases, and also lower than those of RTOG 91-11. Since it was indicated that laryngeal dysfunction may be associated with deaths unrelated to cancer, induction chemotherapy followed by CRT using VMAT might contribute to the reduction of the deaths and to the improvement of OS.
The present study had several limitations, including its retrospective single-center design, and the fact that almost 60% of patients with OPC did not undergo HPV status testing and 5 with HPV-positive OPC were included. This is because HPV status testing for OPC is currently not covered by insurance in Japan. In our institution, HPV status testing has been routinely performed since 2014. These facts introduced potential biases. Nevertheless, this study was meaningful in reporting long-term outcomes after induction chemotherapy followed by CRT using current radiation techniques in patients with locally advanced OPC or HPC.
In conclusion, with long-term observation, the OS, PFS, and LEDFS rates in patients with stage IVA-B OPC or HPC after induction chemotherapy followed by CRT using VMAT were relatively good, and the incidence of late toxicities was low. In addition, response to TPF therapy was the only significant predictive factor of OS and LEDFS. CRT using VMAT was more effective in those who responded to induction chemotherapy.