• Additional safety assessments were added and blood volumes for clinical chemistry and haematology assessments were increased accordingly (Food and Drug Administration [FDA] request). • Inclusion criterion 13 was updated (clarification). • Text added to specify that all adverse events were to be graded according to National Cancer Institute Common Terminology Criteria for Adverse Events Version 4 (clarification at FDA request). • BRAF/NRAS mutation status definitions were added (clarification). • Haematoxylin and eosin stained sections were to be prepared by the central laboratory (to ensure sample processing consistency).

• Primary and secondary objectives were updated (FDA comments): efficacy in the genetic subgroup population of BRAF or NRAS mutation positive patients was considered a secondary objective (previously a co primary objective). It was also clarified that complete remission rate and clinical remission rate at 18 months were to be assessed in the ITT study population. Text describing the analysis of the primary endpoint was updated accordingly and further details of sensitivity analyses were added. Description of the Dunnett and Tamhane step-up procedure was removed as this was no longer applicable. • Exclusion criterion 7 was added (FDA request). • Text was added to clarify that archival tumour analyses may have included, but were not limited to, BRAF V600E and NRAS Q61R, Q61K and Q61L (FDA comments). • It was clarified that randomisation of 228 patients in a 2:1 ratio was expected to provide at least 80% power to show statistical significance for the primary endpoint (FDA comments).

• The treatment plan was updated to clarify that if a patient suspended study drug (selumetinib or placebo) treatment for more than 14 days, they were no longer eligible to re-start treatment (to provide clear instruction on allowed study drug interruption). • It was clarified that the visit window for Visit 4, RAI treatment dose administration (Visit 5) and Visit 6 was +1 week (clarification). • Inclusion criterion 14 was updated (to specify the time frame for adequate organ function assessment). • Exclusion criteria 4, 10 and 22 were updated (to allow patient rescreening, enrolment of patients without external beam radiation therapy in the 6 months before randomisation and to specify the time frame for assessment of blood pressure and adequate organ function). • The maximum daily dose of vitamin E received from selumetinib/placebo was increased to 261.6 mg/day (new calculation). • Text added to allow re-screening. • Instructions on handling patients who had discontinued selumetinib/placebo were updated. • It was clarified that assessment of the percentage uptake in the thyroid bed was to be assessed by the local investigator site according to local clinical practice (to reflect the current situation in clinical practice). • Plasma clinical chemistry assessments and leucocyte differential count as a percentage were added (to allow local laboratories to perform clinical chemistry assessments in plasma). • Guidance for the management of visual symptoms was updated (retinal pigment epithelial detachment added). • Guidance for the management of patients with dyspnoea was updated (dyspnoea is an expected event of selumetinib; however, dyspnoea could also be the symptom of underlying serious lung conditions).

• Study was to be terminated early and all randomised patients who had not yet completed their 3-year follow up were to instead have an end of study phone call. The wording of the 3-years post-RAI follow-up was changed to final study follow-up, in-line with the change. Reason: In the primary analysis (DCO date: 18 May 2018), the primary endpoint was not met and, consequently, collection of further data on efficacy endpoints was no longer considered relevant.