Inclusion criteria

• Patients with epithelial tumours of nasopharynx; nasal cavity and paranasal sinuses; salivary gland cancer in large and small salivary glands; and middle ear (i.e. squamous carcinoma; adenocarcinoma; neuroendocrine; adenosquamous carcinoma, teratocarcinosarcoma, NUT carcinoma, odontogenic tumours) + neuroendocrine and adenocarcinoma in hypopharynx; oropharynx; larynx; oral cavity and lip + odontogenic carcinoma in oral cavity.
• Adult patients (aged ≥18 years).
• Diagnosis performed or verified by the expert centre entering the patient information in the registry.
• Patients entering the HCP at any clinical phase of the disease (diagnosis, treatment of primary cancer, treatment of recurrence, treatment of M+ etc.). The HCP will collect information on the entire course of the disease regardless of when it started to manage the patient. The HCP can decide, based on its resources, the number of patients on whom it can collect data.
• New patients managed by the HCP from 2021 onwards plus patients managed by the HCP, who are actively followed up at the hospital, with year of diagnosis dating back to maximum 2018.

Sample size

This is an observational clinical registry which implies a long-term data collection lasting until all the registry objectives are met. Only HCPs that treat at least 100 cases a year of all rare head and neck cancers are EURACAN member. As of April 2022, the registry is activating 10 HCPs and, with time, is envisioning to at least duplicate the number of data providers. Considering that we expected 6 centres in Italy, 2 centres in Germany, 1 centre in Czech Republic, 1 centre in Spain, 2 centres in France (400 cases per year)and the whole rare head and neck cases in The Netherlands (300 patients yearly, based on incidence estimations), we expect about 1700 patients with a rare head and neck cancer yearly.

Due to the observational nature of the registry, sample size justification is based on the precision of the estimates presented in terms of width of two-sided 95% Confidence Interval (CI) for a single proportion using the Simple Asymptotic method (in case of categorical variables) and in form of normal distribution for means (in case of continuous variables). Thus, for example, for a categorical endpoint (e.g., proportion) a sample size of 80 patients (e.g. middle ear cases in 4 years) will achieve a maximum width of 95% CI on estimated proportions of 23.4% (i.e. estimated proportion +/- 11.7%). For continuous endpoints, a sample size of 80 patients will achieve a maximum width of 95% CI on estimated means of 0.46*SD (i.e. estimated mean +/- 0.23*SD, where SD = Standard Deviation).

Table 1 reports the precision of the estimates for categorical and continuous endpoint under different sample size.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Precision of the estimates expressed as 95% Confidence Interval (CI) for categorical and continuous endpoint accordingly to different sample size.

https://doi.org/10.1371/journal.pone.0283071.t001

For the analytical questions involving several different outputs and variables, it is not possible to define a summary of the sample size calculation. For this reason ad hoc analysis plans for each research question are envisioned.

Data collection and storage

The registry will exploit data available from:

• national or regional registries/databases (DBs) dedicated to rare head and neck cancers (i.e. nasopharynx; nasal cavity and paranasal sinuses; salivary gland; and middle ear cancers)
• HCP registries/DBs;
• ad hoc data collection by HCPs.

The registry is federated thus, data are stored by the data provider (i.e. national/regional registries and/or HCP). At the local level, data are pseudonymised.

Data analyses

The Personal Health Train (PHT) enables data from multiple organizations to be analysed without identifiable data leaving the organization. By keeping data at its source, no copies of datasets are generated and/or shared with third parties. Vantage6 is the open source implementation of the PHT (https://www.vantage6.ai). Vantage6 uses the mathematical principle of “federated learning”, typically applied to horizontally partitioned data (i.e. organizations have data from different patient cohorts, but with similar characteristics/items). Federated learning is based on the mathematical principle of splitting computations into (a) parts at the station (local HCP or registry) and (b) a central part. The stations share sub-computations with the central server only. If federated learning does not work, the data, after quality validation, will be anonymised and sent to the coordination centre (i.e. National Cancer Institute of Milan [Fondazione IRCCS Istituto Nazionale dei Tumori, Milan-INT]).

Statistical analyses will be performed based on a study protocol (please refer to the Governance section). Queries will be developed, in collaboration with clinical experts, to interrogate the EURACAN rare head and neck cancers registry to generate the descriptive statistics and relevant information needed to plan the analyses envisaged by the study protocol. Queries may include general description of the characteristics of the patients or of the cancer included in the registry such as: gender, mean age, cancer site, cancer histology, stage, treatment at diagnosis etc. Due to the nature of the registry, supplementary statistical analysis plans are envisioned to reply to specific research questions arising over time. Secondary analyses, developed after making observations, expected or unexpected, on the registry will be identified and highlighted as hypothesis generating. Positive finding could be used for developing a rationale for further studies or grants. Several analyses will be performed, but here we report a general data analysis plan based on the four objectives of the registry.

Data analyses plan

The data analyses will include descriptive statistics showing frequency and patterns of patients’ and cancers’ variables; analytical analyses investigating the association of patients/disease and/or treatment characteristics and health outcomes.

Descriptive statistics will be used to reconstruct the natural history of rare head and neck cancers (e.g. primary tumour growth rate and pattern, its metastatic dissemination, growth of metastases, association with other diseases etc.) and to report indicators about quality of care.

Multivariable Cox’s proportional hazards model and Hazard ratios (HR) for all-cause or cause specific mortality will be used to determine independent predictors of overall survival, recurrence and second primary cancer. Variables to include in the multivariable regression model will be selected based on the results of univariable analysis or using Information Criteria, such as Akaike Information Criterion. According to the specific research question the role of potential confounders and effect modifiers will be evaluated. Confounders will be included in the model and sensitivity analysis will be performed. Stratified analyses will be performed in the presence of effect modifiers that will be detected by testing the interaction between the predictors and the variable.

To assess treatment effectiveness multivariable models with propensity score adjustment that can remove bias due to real world nature of the registry and progression-free survival will be performed.

High proportion of missing data threaten the validity of the inferences/prognostic models. Thus, a maximum of 10% of missing data will be allowed and missing data will be imputed using strategy such as unconditional/conditional mean or expectation maximum.

Outcomes

Primary outcomes with a time frame of 1 year include:

• Overall survival (all cause and/or cause specific)
• Disease free survival
• Treatment response: the percentage of patients whose cancer shrinks or disappears after treatment. Treatment response expressed as a complete response; partial response; stable disease; progression based on clinical judgment on imaging.
• Cancer treatment observed adverse events: incidence of cancer treatment adverse events as assessed by Common Terminology Criteria for Adverse Events (CTCAE)
• Surgical complications: percentage of patients with surgical complications as assessed by the Clavien-Dindo Classification
• Quality of care: percentage of patients treated according to clinical practice guidelines for head and neck cancer.

Predictors of mortality and/or tumor response include but are not limited to:

• Patients’ Characteristics (age, gender, educational level, tobacco, comorbidity, virus status when appropriate)
• Tumor Characteristics (site, histology, T,N,M, clinical and pathological stage, tumour mutation)
• Treatment Characteristics (radiotherapy alone, surgery alone, systemic therapy alone, multimodal therapy)

Exposures will mainly include different type of treatment (radiotherapy alone, surgery alone, systemic therapy alone, multimodal therapy)

Data quality checks

Data quality checks aim to assess whether data value are present, valid and believable in terms of:

• Validity: the data should be in the correct format (e.g. allowed only some values, data formats)
• Plausibility: the data should be acceptable (eg, internal consistency; temporal and atemporal)
• Completeness: the data should not have missing values or miss data records

Validity and plausibility checks are embedded in the electronic case report form (CRF) in the form of alerts and errors during the data input. Additional checks are implemented in R. The R script, including the checks is downloaded locally from an online instruction repository (Fig 1, step 1). The R script extracts from the Research Electronic Data Capture (RedCap) (the IT solution used for the registry CRF) all the completed cases, stores a copy of the DB in the dedicated local server and runs the checks locally (Fig 1, step 2 and 3). The results of these checks are summarized in two reports: a summary and an individual report (Fig 1, step 4). The summary report collects: the number of patients with a complete and valid information on the core data elements; the distribution of treatment by stage for each cancer site included in the registry and the proportion of cases with missing information for each variable. The individual report includes a summary of error for each patient together with a specification of the variables with the errors to be corrected. Thanks to the Vantage6 software, the two reports (not the data) will reach the registry coordination team (INT) to be monitored and discussed with each data provider (Fig 1 step 5, 6 and 7).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Data quality checks implemented in R.

https://doi.org/10.1371/journal.pone.0283071.g001

After the corrections made by the data providers, all checks will be re-run and quality reports reviewed by INT. Interaction with data providers will be reiterated until sufficient data quality is achieved. The DB with sufficient data quality will be saved and used for the federated learning analysis. These checks will be performed annually and will ensure high data quality within a federated DB. The degree of data completeness will be summarized and available for the researchers and consumers of the analysis of the registry.

Data to be collected

Core data elements will include the "Set of common data elements for Rare Disease Registration" developed by the Joint Research Centre (JRC) (https://eu-rd-platform.jrc.ec.europa.eu/set-of-common-data-elements) to address specificities of rare cancers as compared to rare diseases.

1. Following the EURACAN registry objectives, data will be prospectively collected on patient characteristics, exposure and outcomes. Patient characteristics are descriptive patient data, such as patient demographics, including lifestyle, medical history, health status, etc. The registry will not collect genetic data. Exposure data focus on the disease, devices, procedures, treatments or services of interest. Outcome data describe patient outcomes (e.g. survival, progression, progression-free survival, death, etc.). In addition, data on potential confounders (e.g. comorbidity; functional status etc.) will be also collected (details in the registry code book on the registry site) [6]. Clinical experts and patients representatives in the global head and neck cancer community were involved in the definition of the included variables. Even if it is impossible to envision all the possible counfounding factors in planning a registry, the community deeply discussed each variable and defined the necessary data to be collected.

Patient and public involvement

EURACAN has defined European-Patient Advocacy Group (e-PAGs) for each of its domain. G7 e-PAGs were involved in the definition of the registry objectives and of the variables to be collected. They were invited to all the G7 meetings dedicated to the registry since the beginning. Furthermore, an e-PAG Working Group (WG) dedicated to the registry was established to discuss the registry structure, its governance and consent form. G7 e-PAGs representatives were included in this WG. e-PAGs are included in the Registry Steering Committee (please refer to the registry governance section).

Ethical considerations

The registry was approved by the ethic committee of the Fondazione IRCCS Istituto Nazionale dei Tumori (INT) with protocol number (INT 43/21).

INT is the coordinator of the EURACAN registry as well as a data provider. At the INT, and at each HCP involved, responsible investigators ensure that the EURACAN registry will be implemented in compliance with the protocol, following the instructions and procedures described herein. Each HCP is a controller and will identify a data processor.

Patient protection

Personal data will be recorded and stored in pseudonymised format in a DB installed at the HCP level. Either the PHT solution will enable data sharing among multiple organizations without identifiable data leaving the HCP, or anonymised data will be sent to the INT.

All parties involved in registry development will maintain strict confidentiality to ensure that neither personal privacy nor the privacy of the families of patients participating in the registries will be violated. Data will be processed exclusively by authorized personnel participating in the EURACAN registry (a data processor will be identified at each HCP involved). Access to computer systems and the premises where they are kept will be controlled by appropriate security measures which comply with privacy regulation requirements. The processing of patients’ personal data taking part in the EURACAN registry, and specifically in relation to consent-related data, will comply with local privacy legislation and the General Data Protection Regulation 2016/679 (GDPR) of the European Union.

The registry protocol is submitted to the ethics committees (EC) of the HCPs involved. Furthermore, the ECs of the HCP involved will authorize in advance any research carried out using EURACAN registry data.

Informed consent

Written informed consent is the legal basis of the EURACAN registry and so is required from all participants.

Status and timeline of the registry

The registry initiated recruiting in May 2022. The estimated completion date is December 2030 upon agreement on the achievement of all the registry objectives. As of August 2022, the registry is recruiting.