Study design

This prospective study was designed as a multicentre, nonrandomised observational study for evaluation of the efficacy of END for patients with cN0 OTSCC. The methodology was developed according to the Strengthening the Reporting of Observational Studies in Epidemiology and standards for reporting diagnostic accuracy statements. The protocol was written based on the Standard Protocol Items: Recommendations for Interventional Trials checklist. Figure 1 contains the study flow chart.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Flow diagram for the trial participants. cN0, clinically node-negative; OTSCC, oral tongue squamous cell carcinoma; END, elective neck dissection.

Study setting

This multicentre study is being conducted by the joint research committee of the Japanese Society of Oral Oncology (JSOO). The participating institutions are JSOO-certified high-level clinical facilities in the provision of treatment for oral cancer and accept compliance with the study protocol. The research secretariat solicited the participation of such facilities. The participating facilities are Nagasaki University, Tokyo Medical and Dental University, Kumamoto University, Nippon Dental University, Tokyo Dental College, Kobe University, Tokushima University, Shinshu University, Kurume University, Nara Medical University, Tokyo Medical University, Hokkaido Cancer Center, Hokkaido University, Fukuoka Dental College, Tokyo Metropolitan Cancer and Infectious Diseases Center Komagome Hospital, Hirosaki University, Keio University, Nagaoka Red Cross Hospital, Yamagata University, Akita University, Tsukuba University, Tsurumi University, Saga University, Hyogo Medical University, Saiseikai Matsusaka General Hospital, Kagoshima University, Wakayama Medical University, Saitama Prefectural Cancer Centre, NTT Medical Centre Tokyo, Tokai University, Sapporo Medical University, Osaka University, Dokkyo Medical University, Kanagawa Dental College and Yamaguchi University. The coordinating centre is located at Nagasaki University, and the data coordinating centre is located at the Centre for Clinical Research, Shinshu University Hospital.

Eligibility criteria

Inclusion criteria

• Histologically confirmed OTSCC.

• cN0 OTSCC based on the Union for International Caner Control eighth edition TNM classification.

• No apparent distant metastasis.

• Aged ≥20 years at the time of informed consent acquisition.

• Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1.

• Patient operable.

• Signed informed consent.

Exclusion criteria

• History of treatment for malignant tumours of the head and neck area.

• History of multiple cancers with a disease-free period of less than 5 years at the date of registration.

• Psychosis or psychiatric symptoms that complicate participation in the study.

Interventions

All patients are diagnosed with enhanced CT or MRI at the minimum. Surgical treatment is performed with the patient under general anaesthesia. Resection of the primary tumour was performed to ensure adequate histopathological margins (>5 mm) in all planes. The choice of whether to perform END at the same time as resection of the primary tumour is based on institutional policy and patient preference. For the END procedure, dissection of at least levels I–III is recommended; however, the type of neck dissection is at the surgeon’s discretion. In the group that undergo resection of the primary tumour alone, comprehensive neck dissection (dissection of levels I–V) is indicated when cervical lymph node metastases are detected during follow-up. If there are high-risk factors for postoperative recurrence, such as positive margins and/or extranodal extension, additional resection of the primary tumour, postoperative radiation therapy (RT) or chemoradiation therapy (CRT) may be recommended.

Outcomes

The primary endpoint of this study is OS. The OS will be calculated from the date of curative treatment (the date of surgery or the end date of postoperative RT/CRT) to the date of death from any cause and censored at the last follow-up day at which the patient is alive.

The secondary endpoints are DSS, relapse-free survival (RFS) and the impact on QoL. The DSS will be calculated from the date of curative treatment to the date of death from OTSCC and censored at the time of death from other diseases or the last follow-up day at which the patient is alive. The RFS will be calculated from the date of curative treatment to the date of any relapse or death from any cause and censored at the last follow-up day at which the patient is alive and relapse-free. The impact of QoL is assessed by using the Functional Assessment of Cancer Therapy-Head and Neck (Version 4; FACT-H&N)13 and the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire.14 The FACT-H&N consists of 11 items specific to head and neck cancer, an addition to the FACT-General, a QoL questionnaire for cancer in general.

Follow-up and assessment schedule

Participants will be followed up for 3 years after the date of curative treatment. The date of curative treatment is the date of surgery or the end date of postoperative RT/CRT. Patients will be followed up every month for the first year, every 2 months in the second year, and every 3 months thereafter. Palpation with/without ultrasonography will be performed at each follow-up visit. In addition, enhanced CT or MRI of the head and neck will be performed every 3 months after curative treatment and chest X-ray examination or CT every 6 months thereafter. Furthermore, the FACT-H&N questionnaire will be used for assessment 1, 3, 6, 12, 24 and 36 months after curative treatment. Simultaneously, the DASH questionnaire will be used for assessment 36 months after curative treatment. The data collection schedule is presented in table 1.

Sample size

The primary endpoint of this study is 3-year OS, which was used to calculate the required sample size. According to a previous RCT,9 the 3-year OS rate of patients who underwent resection of the primary tumour alone was 67.5%, and the 3-year OS rate of patients who underwent END was 80.0%. The sample size required to detect differences between two independent proportions was calculated by using the SWOG statistical tool (https://stattools.crab.org/), with 80% power and a two-sided significance level of α=0.05. Accordingly, 199 participants are required in each group. As the rate of END for early-stage tongue cancer in Japan is approximately 25%,15 the total sample size should be 800. Since this study is a prospective observational study, the sample size should be acceptable in actual clinical practice, but the sample size was set to be statistically advantageous by referring to previous prospective studies. In addition, the drop-out rate was not taken into account.

Recruitment

Participant recruitment started at the participating facilities in November 2017 and will continue to October 2022. The planned study period is November 2017 to October 2026. All patients with cN0 OTSCC are screened for eligibility and asked to provide written informed consent (see online supplemental file). Principal investigator of the participant institutions will train the participants to explain the purpose and details of the study in the local language, obtain written informed consent, and record relevant information on QoL questionnaire. Participants are informed that they have the right to withdraw from the study at any time without having to give a reason.

Patient and public involvement

No patient involved.

Data collection and management

Data collection will be performed by using the Alliance Clinical Research Support System Electronic Data Capture (EDC) software of the University Hospital Clinical Trial Alliance. Data entry into the case report forms will be performed by investigators using the EDC software at each participating facility. Following the completion of the study, the fixed data will be exported, deleted from the EDC, and stored in a public data repository.

Statistical analysis

Clinicopathological characteristics of participants will be compared between the primary resection-only group and the primary resection with END group by using the χ² or Fisher’s exact test for categorical variables and the t-test or Mann-Whitney U test for continuous variables. Cumulative OS, DSS and RFS rates will be estimated with the Kaplan-Meier method and compared with the log-rank test, followed by multivariate Cox proportional hazards analyses.

Propensity score-matching analysis will be performed to reduce selection biases associated with the institutional policy and patient preferences. The propensity score will be computed via logistic regression for each patient based on the presumed covariates, which includes age, sex, ECOG performance status, smoking status, alcohol consumption, clinical T stage, clinical DOI and oral care management. Propensity score matching will be performed with 1:1 nearest-neighbour matching and a calliper value equal to a width of 0.25 for the SD of the propensity score.

All data analyses will be performed by using IBM SPSS Statistics for Windows V.24.0 (Japan IBM). Two-tailed p values <0.05 will be defined as statistically significant throughout the analyses.

Data sharing

The study investigators will have full access to and ownership of all data. Deidentified data will be made available to interested outside investigators for additional analyses on reasonable request, following reports of primary outcomes, and with an appropriate data use agreement. Anonymised patient data will be made available via the data coordinating centre (Centre for Clinical Research, Shinshu University Hospital; tcend-project@umin.org) to qualified investigators who submit an approved research proposal. The anonymity, reliability and process of the data collected will be explained in the consent form, explaining that the data will be used only for the purpose of the specific study and will be destroyed after completion. To maintain the confidentiality and anonymity of the participants, the identities of the participants will not be mentioned at all. Participants will be coded so that no one but the principal investigator will know whose data are being used. Only the principal investigator will have access to the data.

Perspective/conclusion

Although the superiority of END over primary resection alone in early-stage tongue cancer has been demonstrated by several meta-analyses or systematic reviews,10–12 there is resistance to generalising and interpreting the results of only a few RCTs. In fact, several problems have been pointed out. First, there is the issue of the difference in the significance of DOI according to the primary site of oral cavity, in other words, whether a DOI of 3 mm for tongue cancer and 3 mm for oral floor cancer can be considered in the same way. In addition, the issue of the high rate of neck recurrence, in the RCT by India,9 43% of the primary resection-only group had neck recurrence, and 18% of them were inoperable. Furthermore, the randomisation of END or not may result in a limited number of study participants, which may lead to a study with low external validity, low generalisability, which is the most feared aspect of RCTs. Therefore, a prospective observational study was conceived to see what would happen if the patients were treated without randomisation and uniformly followed up.

Perspectives of this study may clarify the incidence of occult neck metastases in Japan. It may be possible to set a cut-off value for DOI to determine the superiority of END in tongue cancer. It may be possible to clarify the impact of END on QoL. Ultimately, our results will contribute to determining clear criteria for END.