Background: Differentiated thyroid cancer (DTC), although rare, is the commonest endocrine malignancy, with a significant recent increase in incidence. Natural history is well understood: the majority of patients have an extremely good prognosis, but with a tendency for late recurrences, up to twenty years post diagnosis. These features have combined to effect a lack of evidence underlying current practice, the majority of guidelines based around retrospective series and opinion. A number of validated risk stratification systems exist, unique to DTC, some applicable at the point of intervention. Although these systems reliably predict outcome, they are inconsistently used to guide the extent of intervention. As such, the three main facets of treatment, surgery to the thyroid, surgery to the lymph nodes, and radioiodine (RAI), are variably applied to the same extent of disease. This may depend at least partly upon the clinician‟s preference towards use of risk stratification systems. Consequently, low risk patients may be at risk of morbidities, both short and long term from unnecessarily aggressive interventions, despite consistently good prognosis. Rationalisation of the approach is attractive from a patient, provider, and health service perspectives; without controlled or long term data, such change in practice requires justification. Aim The overall aim of this thesis is to describe and compare the clinical effectiveness of risk stratified (RS) management of DTC, compared to non risk stratified (NRS) management, in an adult population. UK context is provided by description of current practice and long term disease specific outcomes. Three strands of work are combined to address this aim. Part One: Systematic Review A comparison of RS/NRS approaches necessitated objective and reproducible definition of practice overall, encompassing the three facets of intervention. Thereafter, rigorous methodology was applied to examine the clinical effectiveness of the two approaches, based on review of 76 datasets identified from the literature, with RS/NRS approach assigned; there were no relevant randomised trials nor prospective comparative series. Patients, disease and practice were described and compared, with outcome data grouped by safety, disease control, and survival. Sensitivity analysis was planned around population and practice items. Study quality vi was objectively assessed. Weighted mean effect sizes were calculated for RS/NRS cohorts and compared across the two intervention groups for each outcome, with findings further tested on sensitivity analysis. Broadly, equivalence was shown across the two treatment groups by outcome, with small differences consistent with disease biology/magnitude of intervention. Risk stratification appeared to be applicable not only to primary intervention, but also follow-up strategies. Variations in outcome definitions were highlighted, and a pragmatic, patient-centred definition favoured as an approach for future work. Lack of time-specific data was a significant limiting factor. Part Two: Description and Analysis of UK Practice With no UK dataset within the systematic review, validity for the findings was required, as well as the need to describe and compare practice and outcomes in a contemporary local context. The ideal dataset was defined and participants for a UK collaborative sought. Five centres contributed datasets of adequate quality, at least in part prospective, with time-specific outcome measures. Overall, UK practice was by definition risk stratified – and appeared increasingly so over time. There was considerable variation in practice across the five centres (2 NRS, 3RS), but all with consistent, equivalent outcomes compared with the review, despite patients of somewhat higher risk. Utilising methodologies from the systematic review, the analysis was repeated and augmented, confirming consistency in effect directions. Time-specific, adjusted outcome measures demonstrated a possible lack of effectiveness of RAI in high risk disease – a new finding worthy of further exploration. The comprehensive datasets further reinforced the need to work towards prospective, long term, time specific data with common agreed definitions, and suggestions are made how this can be achieved. Part Three: National Survey of Practice Preference The current UK preference for RS management, notwithstanding current service provision around thyroid cancer, is little known. Through cross-specialty survey, utilising a number of case scenarios validated by external experts, and gathering data on clinician demographics, this information was sought, in order to give further context to the above findings. A high degree of variation was identified within, and across scenarios; the least RS preference for intervention applied to the lowest risk scenarios, implying risk of over-treatment. RS preference was associated independently with fewer years in practice and high volume practice. Conclusion Based on the best available data, and supplemented by congruent, contemporary UK data and perspectives, this thesis confirms safety and equivalence of effectiveness of a risk stratified approach to the management of differentiated thyroid cancer, demonstrates considerable variations in practice and suggests possible tools towards building a better evidence base for the future.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:577596 |
| Date | January 2013 |
| Creators | Craig, Wendy |
| Publisher | University of Aberdeen |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://digitool.abdn.ac.uk:80/webclient/DeliveryManager?pid=196011 |
Page generated in 0.0017 seconds