The changes in protocol were as follows:•Section 3.1 was updated for clarification and consistency with other sections of protocol. For example, now all subjects who discontinued study treatment were followed for OS and subsequent anti-cancer therapies, not only those who discontinued study treatment due to disease progression. The text was also updated to clarify that only subjects who discontinued taselisib/placebo for toxicity could continue single agent fulvestrant at the discretion of the investigator while those who discontinued fulvestrant for an AE, albeit rare, should discuss continuation of single agent taselisib/placebo with the Medical Monitor •Figure 1 was reformatted and key enrollment criteria were updated for consistency with text in section 4.1 •Section 3.3.8.9 was updated to clarify the analysis of optional post-progression biopsies and sharing of the resulting molecular report with investigator who may then share the data with the subject (if the subject agreed) •Sections 4.3.2.1 and 4.3.2.2 were updated to align better with information outlined in Appendix 1 •Section 4.4.2 was clarified relative to specific washouts required in the exclusion criteria in Section 4.1.2 •Section 4.5.5 was updated to be consistent with requirements outlined in Appendix 1 and for clarification. For example, the tumor assessment requirements were clarified based on additional details provided in the Appendix 1 footnotes and inconsistencies were removed regarding the window of the screening tumor assessments. For consistency with Section 3.1, the text was updated to reflect the original intent of the protocol that subjects who discontinue study treatment for reasons other than disease progression will continue to undergo tumor assessments until progressive disease (PD), even if a subject initiates anti-cancer therapy subsequent to study drug discontinuation. The same update was made in Appendix 1 footnote l.

The changes in protocol were as follows:Section 4.5.6 was updated for consistency and to clarify the analysis of post-progression biopsies •Section 4.5.11 was updated for consistency with the relevant changes made in Section 3.1, for example that all subjects who discontinued study treatment were followed for OS and subsequent anti-cancer therapies, not only those who discontinued study treatment due to disease progression •Table 3 was updated to clarify corticosteroid treatment for Grade 3 diarrhoea or colitis •Table 5 was updated for clarification of the pneumonitis guidelines related to local clinical practice •Table 7 was updated with a footnote to clarify for Grade 1 or 2 hyperglycemia that increases in anti-hyperglycemic medications only applied to subjects who initiated these after randomization since subjects with diabetes requiring anti-hyperglycemic medications were not eligible •Section 5.1.1.1.7 was updated for clarification referring to Table 8 for specific management guidelines for subjects who experienced changes in blood counts or showed signs of infections if deemed clinically appropriate by the investigator •Section 5.2.4 was updated for clarification regarding additional supporting data sponsor may request for certain AEs •Section 7.1 was updated for clarification as the study used electronic patient-reported outcome (ePRO) devices but no paper questionnaires •Appendix 1 was updated for consistency with the remainder of the Protocol footnotes n through p were revised so that Cycle 1 Day 1 safety laboratory samples did not need to be redrawn if the corresponding screening samples were taken within 2 days prior to Cycle 1 Day 1. Footnote z was added to the ECG assessments for clarification consistent with the text in Section 4.5.8.

The changes in protocol were as follows:•In Sections 2.1, 6.4.2.1, and 6.4.2.2, the secondary efficacy objectives were reordered to emphasize the change in the hierarchical testing. The testing hierarchy of the secondary endpoints was changed to objective response rate (ORR) first followed by OS after a statistically significant investigator-assessed PFS compared with OS first in the original protocol. Anti-tumor responses have been seen in subjects with estrogen receptor (ER)+breast cancer who have been treated with taselisib. The goal was to now test formally if there was a statistically significant difference in ORR between the treatment arms. OS was still formally tested if both investigator-assessed PFS and ORR reached their significance level •In Sections 2.1, 3.4.1, and 6.4.2.5, a secondary efficacy objective and outcome measure was added: BICR-assessed PFS was intended to show that there was no potential bias in the primary efficacy objective investigator-assessed PFS •In Section 2.5, clinical benefit rate (CBR) was added to the exploratory objectives for consistency with the secondary efficacy objectives •In Sections 3.1, 3.3.2, 6.10, and 9.4, the possible addition of a China extension cohort was introduced. In order to characterize the efficacy and safety profile of taselisib in combination with fulvestrant in Chinese subjects and to potentially support a regulatory submission in China, a China extension cohort was planned in the study. After the global enrollment closes, additional Chinese subjects may continue to be recruited into the China extension cohort. A total of up to 150 Chinese subjects with detectable PIK3CA-mutant tumors may be enrolled as part of the global study population and extension cohort combined •In Section 4.1.1, an additional inclusion criterion was added to define the subject population in the China extension cohort to be from the People’s Republic of China.

The changes in protocol were as follows:•In Section 4.2, further description of blinding criteria for study personnel on the basis of the results of the interim and final analyses for investigator-assessed PFS was added •In Section 4.5.6, a change with regard to timing of optional post-progression biopsies was introduced relative to the start of new anti-cancer treatment. The post-progression biopsy could still be obtained within approximately 14 days of the start of the new anti-cancer treatment as long as it was deemed safe by the investigator •In Section 4.5.10 and Appendix 1, there was an explanation of study drug discontinuation visit (SDDVs) occurring if study treatment was interrupted to allow for a SDDV more than 28 days after the decision to permanently discontinue study treatment •In Section 5.1.1.1.1, there was a clarification of AE management guidelines For diarrhoea, dose resumption was distinguished for certain cases of infectious diarrhoea •In Section 5.1.1.1.2, there was a clarification of AE management guidelines. For pneumonitis, infectious work-up was listed as a relevant investigation•In Section 5.7, the reference document for fulvestrant was added, in Sections 6.1, 6.4.1, and 6.9.1, and Table 11, there was the addition of an interim efficacy analysis of investigator-assessed PFS. This interim efficacy analysis was added to enable an earlier assessment of efficacy that could provide subjects with PIK3CA-mutant tumors with earlier access to a potentially effective targeted therapy should the iDMC recommended stopping the study early on the basis of results from the interim analysis for investigator-assessed PFS and should the Sponsor decided to accept the recommendation and obtained regulatory approval.

The changes in protocol were as follows:•In Section 6.4.2.3, the analysis of CBR was clarified to be performed for subjects with PIK3CA-mutant tumors with measurable disease at baseline and also to be repeated for the group of subjects with PIK3CA-mutant tumors regardless of measurable disease at baseline •In Section 6.7, a time-to-deterioration analysis was included as one of the PRO analyses in order to assess if there was a difference between the treatment arms •In Appendix 2, the table has been updated such that predose blood samples could be drawn within 2 days prior to the cycle visit.