Composition and Priorities of Multidisciplinary Pediatric Thyroid Programs: A Consensus Statement

Introduction

The incidence of pediatric thyroid cancer has increased over recent decades, ¹ and the approach to care for affected children can vary by region. Some of these differences reflect socioeconomic disparities in diagnosis or treatment, ^2,3 whereas others may reflect differing protocols, available resources, and standards of practice. Distinctions in clinical, molecular, and pathological features between pediatric and adult differentiated thyroid cancer (DTC) ^{4 –6} underscore the need for tailored recommendations for the pediatric population.

Consensus treatment guidelines, including the 2015 American Thyroid Association (ATA) Management Guidelines ⁵ and the 2022 European Thyroid Association Guidelines, ⁷ advocate for multidisciplinary care teams specializing in pediatric thyroid neoplasms. However, the exact composition and functions of these teams remain unclear. This study aimed to develop consensus statements on essential components and priorities for clinical teams managing pediatric thyroid cancer.

We sought clarity in four areas: defining the patient population and conditions managed by pediatric thyroid programs, identifying key specialties and services, and determining research priorities to address current knowledge gaps and foster collaboration.

Methods

We used the modified Delphi method to distill expert opinion into consensus statements amongst a group of international experts in pediatric thyroidology per established methodology. ^{8 –10} After an initial literature review, a group of five content experts discussed and reviewed statements to be included in the survey, chaired by the senior author (K.D.M.). A methodologist (K.B.) was assigned given his extensive previous experience in creating clinical consensus statements. This study was considered exempt from IRB approval using the Stanford Medicine Center for Improvement Guidelines. ¹¹ The remainder of the development group members are all content experts in the care of pediatric patients with thyroid pathology, have experience with application of evidence-based medicine, and confirmed willingness to participate prior to the initial survey. Content experts were nominated by the corresponding author based on their involvement with established clinical programs, as cited authors of relevant literature, national and international speaking engagements, and relevant society leadership. We aimed for a diversity of geographic representation, specialty training, and demographic backgrounds. The survey was created, maintained, and iterated using a web platform (Welphi.com). Literature search was conducted in February 2023 to include publications in English from PubMed, EMBASE, SCOPUS, and Cochrane Database of Systematic Reviews. Literature was used to identify current evidence regarding pediatric thyroid cancer care models and then used as the basis for writing the initial items. Content experts responded to each statement on a 9-point Likert scale with the following anchors: strongly disagree (1), disagree (3), neutral (5), agree (7), strongly agree (9), which were modified for each subgroup. In the 9-point scale, the anchors are spaced apart at intervals of 2 (1, 3, 5, 7, 9). We identified a priori four content areas: 1) overall program objectives, 2) defining the care team, 3) defining essential services and support, and 4) establishing research priorities. Upon completion of the Round 1 survey, ranked results, summary statistics, and anonymized comments were returned to participants. The original group of content experts reviewed the results and comments and modified the statements accordingly.

Consensus was defined as a mean score of ≥7.0 and no more than 1 outlier; near consensus was defined as a mean score of ≥6.5 and no more than 2 outliers; no consensus was defined as <6.5 or ≥3 outliers. Outliers were defined as any rating by a development group member that is at least 2 Likert points away from the mean score for that item. By completion of the second questionnaire, each item had either achieved consensus, was deemed no consensus, or remained as near consensus with no absolute change in the mean score by ≤0.3, so no further questionnaires were administered. One item (Item 44, “Ability to give rhTSH”) was inadvertently omitted from Round 2 and therefore only data from Round 1 are reported. Regarding research priorities, an average score of 8–9 was interpreted as “very strong/essential” and a score of 7.0–7.99 was interpreted as a “strong” priority based on the Likert scale descriptors. For analysis of the team members, a Likert score of ≥5 was considered to indicate a core team member. ^12,13 This methodology is in keeping with prior program-design Delphi publications. ^12,13 A full list of questions from each round is included as Supplemental Material.

Descriptive statistics were performed in Excel v16 2024 (Microsoft Corp, Redmond, WA) and SPSS (IBM Corp. IBM SPSS Statistics for Macintosh, Version 29.0.2.0 Armonk, NY). The article was drafted and reviewed with participation from each panel member. Three external reviewers considered clinical experts also critically reviewed the data and article.

Results

Thirty experts (representing pediatric endocrinology, pediatric surgery, pediatric otolaryngology, general surgery, pathology, nuclear medicine, pediatric oncology, research, care coordinators/nursing, and genetics) were identified and 24 agreed to participate. These individuals were experts in their fields of pediatric endocrinology (n = 6), nuclear medicine (n = 1), pathology (n = 1), pediatric surgery (n = 3), pediatric otolaryngology (n = 10), and pediatric oncology (n = 3). Survey response rates were 88% and 83% for Rounds 1 and 2, respectively.

After 2 modified Delphi method surveys of 66 total items, 29 met the predefined criteria for consensus, 7 were near consensus, and 25 did not achieve consensus. Five items were modified between Rounds 1 and 2 and therefore are counted in the total (n = 66) but not reflected in the end results (n = 61). Items that did not reach consensus are summarized in Table 1. Due to minimal change between rounds 1 and 2, the third round was foregone. The means and standard deviations had stabilized after two rounds of surveys.

Discussion

The results of this Delphi study point to wide consensus among clinicians from various subspecialities and across geographic regions of North America on the overall goals, team members, essential services and support, and research priorities for Pediatric Thyroid Programs. This study was motivated by the 2015 ATA Management Guidelines for Children with Thyroid Nodules and Differentiated Thyroid Cancer, ⁵ which broadly called for multidisciplinary care but did not provide additional specificity as to composition. This study is the first to define the structure and function of Pediatric Thyroid Programs. It aims to guide these programs on essential care elements and advance collaborative research in pediatric thyroid cancer. The data offer an initial framework based on expert opinion and should be adapted to local settings and resources. This work may help establish care standards and support clinics in advocating for multidisciplinary care and resources.

Integrated care of children has been shown to result in improved clinical outcomes, patient satisfaction, provider satisfaction, and value across a range of pediatric conditions. ^{14 –17} Across the four subcategories in the current study, an emphasis on multidisciplinary, integrated care for children with thyroid cancer emerged. The panel agreed that pediatric patients with thyroid cancer should be offered care coordination, have access to multidisciplinary tumor boards, be cared for by dedicated endocrinologists and surgeons, and represent a distinct patient population when compared with patients with noncancerous diagnoses. Furthermore, the panel agreed that collaboration amongst programs is essential in developing standardized quality metrics.

As outlined in the results above, the panel did not reach consensus in defining a specific age range for a Pediatric Thyroid Program. This reflects the diversity of patients and variable institutional practice patterns. Historically, the standard cut-off age for adult versus pediatric care has been 18 years, but several publications that suggest a lower age cut-off (e.g., 14 years) at which patients may be appropriately cared for by adult providers. ^{18 –21} Such patients may benefit from care in a high-volume adult thyroid cancer setting rather than in a low-volume pediatric setting. In addition, some pediatric institutions may care for adult patients, such as in survivorship clinics. Such adolescents and young adults may be served in either an adult or pediatric setting, depending on local expertise and resources. Programs should consider their own practice settings and patterns when defining their age guidelines.

There was a positive correlation between increasing complexity and the perceived benefits of a multidisciplinary Pediatric Thyroid Program, with MTC having the highest correlation, followed by DTCs, thyroid nodules, and thyroid tumor predisposition syndromes. This reflects current literature and multimodal, evolving treatment options for MTC and advanced DTC, which emphasizes the importance of a multidisciplinary approach. ^5,22,23 It is unclear if the panel was assuming that pediatric MTC would be in the setting of multiple endocrine neoplasia syndrome, which tends to have a more nuanced management approach than nonsyndromic MTC, more likely to occur in adults. As such, we acknowledge that many patients with syndromic or nonsyndromic MTC are likely to be managed in adult-focused programs. Nonetheless, pediatric thyroid programs should aim to have the expertise and resources capable to treat syndromic and nonsyndromic pediatric MTC.

Literature indicates that coordination between surgeons and pediatric endocrinologists leads to better outcomes in pediatric thyroidectomies ^21,24,25 and greater confidence in monitoring patients when surgical intervention may not be necessary. ²⁶ Pediatric endocrinologists and thyroid surgeons are considered core team members for all patients. Pediatric endocrinologists, often directing care for pediatric thyroid cancer, differ from oncology teams that lead in other pediatric cancers. They should be involved preoperatively and collaborate with the thyroid surgeon. Although the panel agreed on the necessity of a pediatric-experienced thyroid surgeon, consensus on specific training requirements was not reached. Surgeons, while crucial for primary management, may not be needed for long-term care. The consensus did not address surgeon volume or minimum procedural requirements. There is controversial literature surrounding this topic, and this area of research was outside the scope of the present work. ^{27 –31} Nonetheless, the inverse correlation between surgical volume and complication rates is well-defined and should be incorporated into programmatic consideration. ^{32 –35}

Those clinicians considered core team members reflect a changing landscape in the management of pediatric thyroid cancer: pediatric endocrinologist, thyroid surgeon, proceduralist for US-FNA, pediatric pathologist, pediatric nuclear medicine, nurse coordinator, pediatric radiologist, pediatric oncologist, pediatric geneticist/genetic counselor, and pediatric social worker. There is an increased use of oncogene-specific and tyrosine kinase inhibitors, reflecting the expanding role of oncology, pathology, and genetics in disease management. Given the rapid advances in the field with molecular targets, we emphasize that pediatric oncology is a core member of the team and likely did not reach 100% consensus because not every patient requires such treatments based on current guidelines. In this survey, “pathologist” was intended to also represent the field of cytology, although it is acknowledged that these are distinct board-certified specialties. Similarly, the nuances of pediatric nuclear medicine, including considerations for dosimetry and the ongoing research into appropriate use of radioactive iodine in the pediatric population supports the essential role of an expert in nuclear medicine. Patients with more advanced disease are likely to require every member of the team outlined in Table 3 at some juncture. The presentation of the team subgroup follows a different methodology from the other subgroups, which can raise questions of the methodology and validity of the reporting. Reporting as a supermajority is the model used in prior program-designed Delphi publications. ^12,13 This is resultant of an inherent flaw in rigid Delphi methodology being applied to a biased and skewed dataset: respondents likely will not exclude their own subspecialty, for example, and may favor ratings reflective of their own program. This is suggested by the number of outliers for obvious core members such as nuclear medicine, geneticists, and oncologists. Responses were blinded, so respondent bias toward one’s own specialty could not be accounted for. However, modification of the Delphi methodology allows for agreement on a more comprehensive care team. These outliers are still important to consider, as they may suggest flexibility in the proposed model. This becomes especially important in community and private practice settings. Best practice would include all individuals according to this survey, but there may be flexibility in providing adequate care including partnering with larger programs on an as-needed basis to access to multidisciplinary opinion and expertise. In addition, outliers can serve as valuable indicators of areas that require further exploration across practice settings to augment the design of truly effective pediatric thyroid programs and build thyroid programs of the future, which are able to respond to clinical innovations such as dosimetry and telehealth. The results of this alternative method aligns with accepted practice and clinical judgement of this author group, as it relates to current evaluation and treatment practices. The composition of essential team members may require ongoing exploration, the incorporation of emerging evidence, or focused questioning in future rounds. This approach could help identify whether differing perspectives arise from alternative data interpretations, gaps in current understanding, potential new methods, or biases and misunderstandings in the questions asked.

Support and services crucial for quality and safety include intraoperative neural monitoring, postoperative calcium monitoring, and institutional clinical care pathways. Molecular testing on cytology and surgical pathology is essential for managing invasive or progressive disease and integrating these data into treatment plans. In addition, 95% of respondents endorsed US-FNA as essential, aligning with current guidelines. ⁵

The identified research priorities collectively emphasize the need for comprehensive, evidence-based approaches to care and evaluation of outcomes reflective of these approaches. The recent development of the Child and Adolescent Thyroid Consortium (https://www.thyroidcatc.org/) will now allow for pooling of data and resources to hopefully address some of these research priorities. There is also a call in the literature for the incorporation of molecular information into management paradigms, given the robust data that the genotypic differences in subpopulations of pediatric thyroid cancer are associated with clinically distinct phenotypes, such as fusion oncogenes having greater metastatic potential ³⁶ and the successful incorporation of selective fusion gene inhibitors into treatment plans. ³⁷ The results of this study highlight the importance of identifying future directions for investigation. To advance these research priorities, including a clinical research coordinator into those programs with a research agenda is crucial, as their involvement in clinical trials and research can drive advancements in care and lead to the development of new treatment options for pediatric thyroid diseases.

Limitations

A major limitation is the recruitment of the expert panel as a convenience sample. This could lead to selection bias. Furthermore, there is an overrepresentation of surgical specialists, which may lead to skewed results, especially as it relates to team structure. There was a strong representation of tertiary/quaternary care centers on the panel, which may also introduce bias regarding team structure. There was also no survey participation from important nonphysician members such as nurses, social work, research coordinators, and nonphysician scientists despite indicating initial interest during the recruitment phase. In light of this, these consensus items are considered an ideal state, especially in the academic or tertiary setting, but we recognize that it is not feasible in every setting. Finally, this panel of respondents did not include representation from patients or families whose views about receiving care are undoubtedly important.

Disclaimer

These statements, based on expert opinions, are for informational and educational purposes only. They aim to guide the development and collaboration of Pediatric Thyroid Centers where literature is lacking. The data do not cover all aspects of pediatric thyroidology or establish care standards. Items not reaching consensus remain relevant for patient care and should not be disregarded. This report does not influence financial coverage for specific providers, therapies, or diagnostics.

In conclusion, we propose that there is a definable model for the care of children with thyroid neoplasia that warrants continued refinement and evaluation. This model comprises multidisciplinary care of patients with thyroid cancer, thyroid nodules, and cancer predisposition syndromes; defined team members, services, and resources; and emphasized collaborative research priorities. This model has the potential to standardize various aspects of clinical care and enhance our ability to study patient outcomes and improve health care delivery and scholarly collaboration. In the future, this may lead to opportunities for accreditation of care centers, facilitation of learning health systems, and expansion of multi-institutional research.