Search strategy
The AG identified clinical studies and systematic reviews by searching EMBASE, MEDLINE, PubMed and The Cochrane Library from 1999 onwards. All databases were searched on 10 January 2017. Based on the fact that the FDA approved sorafenib for its first indication in 2005, and lenvatinib in 2015, the AG considered that this date span would allow all relevant clinical evidence to be identified. Searches were restricted to publications in English language. The AG did not use any other search filters. The search strategies used by the AG are provided in Appendix 1. In addition to the electronic database searches, information on studies in progress was sought (on 16 May 2017) by searching the ClinicalTrials.gov website, the International Clinical Trials Registry Platform and the European Union Clinical Trials Register. The references in the systematic reviews included in the AG’s review of systematic reviews, and those listed in the submissions from professional stakeholders that were submitted to NICE, as part of the NICE MTA process, were cross-checked to identify any relevant studies not retrieved from the electronic database searches. Literature search results were uploaded to and managed using EndNote X7.4 [Clarivate Analytics (formerly Thomson Reuters), Philadelphia, PA, USA] software.
Study selection
The eligibility criteria listed in Table 3 were used to identify studies for inclusion in the AG’s literature review.
| Criteria | Inclusion | Exclusion |
|---|---|---|
| Patient population | Adults with progressive, locally advanced or metastatic, differentiated thyroid carcinoma refractory to radioactive iodine | Patients with other types of thyroid cancer or diseases |
| Interventions | Lenvatinib or sorafenib monotherapy (or in combination with BSC) | Lenvatinib or sorafenib in combination with other agents |
| Comparators | Lenvatinib or sorafenib monotherapy (or in combination with BSC), BSC and placebo | A comparator other than lenvatinib, sorafenib, BSC and placebo |
| Outcomes | The outcome measures to be considered include OS, progression-free survival, response rate, adverse effects of treatment and HRQoL | No study was excluded based on outcomes |
| Study design | RCTs, systematic reviews and prospective observational studies | Retrospective cohort studies, case series, case reports, comments, letters, editorials, in vitro, animal and genetic or histochemical studies |
| Restrictions | English language only | Non-English-language studies |
Two reviewers (JH and RH) independently screened all titles and abstracts that were identified in the initial searches (screening stage 1). Based on the titles and abstracts, full-text papers that appeared to be relevant were obtained and assessed for inclusion by the same two reviewers in accordance with the AG’s eligibility criteria (screening stage 2). When necessary, discrepancies were resolved by consultation with a third reviewer (NF). At both stages of screening, studies that did not meet the inclusion criteria were excluded, and, at screening stage 2, the reasons for excluding studies were noted.
The eligibility criteria in Table 3 differ slightly from those specified in the AG’s systematic review protocol.55 The AG, responding to a suggestion from NICE in relation to the final protocol,55 agreed to include evidence from prospective observational studies that had been submitted to the European Medicines Agency (EMA); however, as only reviewing studies included in the EMA submissions26,27 would have introduced selection bias, the AG included all prospective observational studies of patients with RR-DTC identified by its searches.
Data extraction and quality assessment strategy
Data relating to RCT study characteristics and outcomes were extracted by one reviewer (NF) and independently checked for accuracy by a second reviewer (YD). Data relating to study characteristics and outcomes of systematic reviews and observational studies were extracted by one reviewer (JH or NF) and independently checked for accuracy by a second reviewer (JG). In all cases, a consensus was reached. Study data reported in multiple publications were extracted and reported as a single study. Data were extracted into tables in Microsoft Word (Microsoft Corporation, Redmond, WA, USA).
As specified in the AG’s systematic review protocol,55 the quality of included RCTs and systematic reviews was assessed according to the criteria set out in the Centre for Review and Dissemination’s guidance56 for undertaking reviews in health care. The quality of the included RCTs was assessed by one reviewer (YD) and independently checked for agreement by a second reviewer (NF). In all cases, a consensus was reached. The quality of the included systematic reviews was assessed by one reviewer (JG) and independently checked for agreement by a second reviewer (YD). When necessary, discrepancies were resolved by consultation with a third reviewer (MR).
Methods of analysis/synthesis
The AG’s data extraction and quality assessment results are presented in structured tables and as a narrative summary. Data from RCTs are considered to provide primary clinical effectiveness evidence, with data from systematic reviews and observational studies considered to provide supporting evidence.
As the available evidence did not include two or more RCTs comparing the same intervention, the AG was not able to conduct a meta-analysis of RCT data.
The AG assessed the feasibility of conducting an indirect comparison of effectiveness data (including a comparison to assess effectiveness according to previous treatment with TKIs) by evaluating the clinical and methodological heterogeneity of the included RCTs. Heterogeneity was assessed by comparing (1) trial characteristics, (2) participant characteristics, (3) outcome data and (4) study quality.