Exclusion criterion 21 was changed; it originally referred to patients randomised to treatment with chemotherapy only. As the consent process took place before randomisation, it was necessary to cover both treatment arms with this exclusion criterion. The restricted medications during treatment with Afatinib were changed. It was specified that the list of restricted medications refers to all patients randomised. An additional explanatory paragraph was added that the concomitant use of potent P-gp inhibitors and inducers was to be avoided during treatment with Afatinib. The background was that a trial [1200.79] in healthy volunteers indicated that co-administration of these drugs affected the pharmacokinetics of Afatinib.

Several changes and corrections were introduced with the second amendment to the protocol; major changes are presented. It was specified that the trial had 2 screening visits. During the first screening visit, the patient signed the first informed consent and agreed to the EGFR mutation testing. During the second screening visit patients with positive EGFR mutation testing signed a second informed consent and agreed to participate in the main part of the trial. The strict time window for Afatinib intake was removed to accommodate individual patient’s daily schedule preference because Afatinib has a long half-life. The storage conditions for Afatinib were corrected to match the labelling in the USA and Canada. Tablets were to be stored at the temperature specified in the label [i.e. between 15°C and 30°C in the USA and Canada and not above 25ºC in all other countries]. It was specified how data in HRQOL questionnaires were to be handled for Adverse Event [AE] reporting. The length of the observation period of this trial was specified. The focus of the analysis of HRQOL was broadened to include all summary scales and items measuring cough, dyspnoea, and pain measured by the EORTC QLQ-C30 and QLQ-LC13 questionnaires. The time window of the on-treatment period was modified to match the planned safety analysis with other Afatinib trials. It was specified that the decision of whether to proceed to full accrual was based on the first 40 patients randomised to treatment with Afatinib whether or not they had stopped treatment before the Week 6 assessment. New information was added to the appendix of the protocol. The appendix was updated with the current Summary of Product Characteristics [SPC] of Cisplatin provided for the trial. In addition, the RECIST version 1.1 criteria in the appendix were updated to ensure consistency with the imaging charter for the central independent review of radiological imaging.

The third amendment to the protocol introduced several changes; the key changes are presented. Amendment 3 changed the frequency of Electro Cardio Gram [ECG] assessments from once every third course to “as clinically indicated”. The amendment was released after at least 18 months of collection of centrally assessed ECG data for all randomised patients. The data showed that Afatinib did not have any effect on QTc or other ECG parameters; therefore routine monitoring of ECG was no longer required. The requirements for collection of biopsy and blood samples at the time of Progressive Disease [PD] were changed for both treatment arms. Based on the review of already collected data, further follow-up biopsy and blood samples were not requested. The frequency of collection of observation-period data was changed. A data snapshot could now be requested at any time; after the analysis of OS, the collection of observation-period data could be reduced in frequency or stopped, as decided by the Trial Clinical Monitor [TCM]. The requirement for reporting ‘always serious’ adverse events as per new corporate standard was added. The new corporate standard for monitoring and assessment of potential drug-induced liver injury was added. Based on the new clinical data, the recommendation to avoid the use of P-gp inhibitors or inducers in patients treated with Afatinib was modified to allow for their use with caution if clinically indicated.

Several changes affecting patients still ongoing in the trial were introduced with the fourth amendment to the protocol; the main changes are presented. Routine monitoring of Left Ventricular Ejection Fraction [LVEF] was no longer required after database lock for the analysis of OS. LVEF assessments were now to be performed as clinically indicated. No safety signals indicating an effect of Afatinib on the cardiac contractility had been identified in this trial and using a larger safety database. Routine monitoring of LVEF was therefore no longer required. The frequency of trial visits was reduced after database lock for the analysis of OS. A treatment course now comprised 9 weeks [63 days] and FU visits were to be performed every 9 weeks [63 days] until PD or death. All patients had been on treatment for more than 2 years, therefore the frequency of clinic visits could be reduced. The frequency of imaging assessments was reduced after database lock for the analysis of OS. Assessments were now to be performed every 18 weeks. More frequent assessments were no longer required after assessment of the primary endpoint PFS. Central independent review of tumour imaging was stopped after database lock for the analysis of OS. Central independent review was no longer needed as the primary endpoint PFS had been assessed and reported. Completion of HRQOL questionnaires was no longer requested after database lock for the analysis of OS. Sufficient HRQOL data had been collected for the analysis. Patients could move onto an alternative supply of Afatinib after database lock for the analysis of OS, as Afatinib had been approved for use in patients with EGFR mutation positive Non-Small Cell Lung Cancer [NSCLC] in some countries by the time of the analysis. The follow-up and observation periods were amended to allow for completion of the trial. The follow-up period was to end on 31 Jan 2015 with the exception of FU visit 1 which was still required to assess Adverse Events [AEs].

The follow-up and observation periods were amended to allow continued collection of data after 31 Jan 2015, until all patients had completed study treatment. Follow-up visit 1 was still required for all patients to assess for AEs.