IDFS event was defined as the time from randomization until the date of first occurrence of one of the following: Ipsilateral invasive breast tumor recurrence (an invasive breast cancer [bc] involving the same breast parenchyma as the original primary lesion); Ipsilateral local-regional invasive bc recurrence (an invasive bc in the axilla, regional lymph nodes, chest wall, and/or skin of the ipsilateral breast); Contralateral or ipsilateral second primary invasive bc; Distant recurrence (evidence of bc in any anatomic site [other than the three sites mentioned above]) that has either been histologically confirmed or clinically/radiographically diagnosed as recurrent invasive bc; Death attributable to any cause, including bc, non-bc, or unknown cause. 3-year IDFS event-free rate per randomized treatment arms in the ITT population was estimated using the Kaplan-Meier method and estimated the probability of a participant being event-free after 3 years after randomization.

The ITT lymph node positive population was a subpopulation of the randomized participant population including patients with positive lymph node.

IDFS event was defined as the time from randomization until the date of first occurrence of one of the following: Ipsilateral invasive breast tumor recurrence (an invasive breast cancer [bc] involving the same breast parenchyma as the original primary lesion); Ipsilateral local-regional invasive bc recurrence (an invasive bc in the axilla, regional lymph nodes, chest wall, and/or skin of the ipsilateral breast); Contralateral or ipsilateral second primary invasive bc; Distant recurrence (evidence of bc in any anatomic site [other than the three sites mentioned above]) that has either been histologically confirmed or clinically/radiographically diagnosed as recurrent invasive bc; Death attributable to any cause, including bc, non-bc, or unknown cause. 3-year IDFS event-free rate per randomized treatment arms in the ITT population were estimated using the Kaplan-Meier method and estimated the probability of a patient being event-free after 3 years after randomization.

The ITT Population comprised all randomized patients, whether or not they received any study treatment or completed a full course of study treatment. Patients were analyzed according to their randomized treatment.

IDFS including second primary non-breast cancer was defined the same way as IDFS for the primary endpoint but including second primary non breast invasive cancer as an event (with the exception of non-melanoma skin cancers and carcinoma in situ (CIS) of any site).

The ITT Population comprised all randomized patients, whether or not they received any study treatment or completed a full course of study treatment. Patients were analyzed according to their randomized treatment.

DFS was defined as time between randomization and first occurrence of IDFS, second primary non-breast cancer and contralateral or ipsilateral ductal carcinoma in situ (DCIS).

The ITT Population comprised all randomized patients, whether or not they received any study treatment or completed a full course of study treatment. Patients were analyzed according to their randomized treatment.

DRFI was defined as time between randomization and first occurrence of distant breast cancer recurrence.

OS was defined as the time from randomization to death due to any cause.

The ITT Population comprised all randomized patients, whether or not they received any study treatment or completed a full course of study treatment. Patients were analyzed according to their randomized treatment.

An adverse event is any untoward medical occurrence in a participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with the treatment. An adverse event can therefore be any unfavorable and unintended sign (including an abnormal laboratory finding, for example), symptom, or disease temporally associated with the use of a pharmaceutical product, whether or not considered related to the pharmaceutical product. Preexisting conditions which worsen during a study are also considered as adverse events. AEs were reported based on the national cancer institute common terminology criteria for AEs, Version 4.0 (NCI-CTCAE, v4.0).

The safety population included all randomized participants who received at least one full or partial dose of any study treatment. 8 participants were randomized but did not receive any study treatment (5 in the Anthracycline, Trastuzumab, Pertuzumab, and Taxane [AC-THP] arm, 3 in the Anthracycline, Trastuzumab Emtansine and Pertuzumab [AC-KP] arm). 16 participants in the AC-THP and 13 participants in the AC-KP arm only received AC but no HER2 targeted therapy and were assigned to the AC-THP arm.

LVEF was assessed using either echocardiogram (ECHO) or multiple-gated acquisition (MUGA) scans.

The safety population included all randomized participants who received at least one full or partial dose of any study treatment.

The EORTC QLQ-C30 included global health status, functional scales (physical, role, emotional, cognitive, and social), symptom scales (fatigue, nausea/vomiting, and pain) and single items (dyspnoea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties). Most questions used a 4-point scale (1 'Not at all' to 4 'Very much'; 2 questions used 7-point scale [1 'very poor' to 7 'Excellent']). Scores were averaged and transformed to 0 - 100 scale, whereby higher scores indicate greater functioning, greater quality of life, or a greater degree of symptoms, with changes of 7 - 15 points considered to be a clinically meaningful detioration to participants. A positive value means an increase, while a negative value means a decrease, in score at the indicated time-point relative to the score at baseline (Cycle 1, Day 1).

The Patient Reported Outcome (PRO) Population includes participants with both a baseline assessment and at least one post-baseline assessment in any PRO items. The variation of the number of participants in the table reflects the number of participants with corresponding scores at the given visits and at baseline.

EORTC-QLQ-BR23 is a 23-item breast cancer-specific companion module to the EORTC-QLQ-C30. There are four functional scales (body image, sexual enjoyment, sexual functioning, future perspective [FP]) and four symptom scales (systemic side effects [SE], upset by hair loss, arm symptoms, breast symptoms). Questions used 4-point scale (1=not at all, 2=a little, 3=quite a bit, 4=very much). Scores averaged and transformed to 0-100 scale. High score for functional scale indicated high/better level of functioning/healthy functioning. Higher scores for symptom scales represent higher levels of symptoms/problems. For functional scales, positive change from baseline indicated improvement in quality of life (QOL) while negative change from baseline indicated a deterioration. For symptom scales, positive change from baseline indicated deterioration and negative change indicated improvement.

The Patient Reported Outcome (PRO) Population includes participants with both a baseline assessment and at least one post-baseline assessment in any PRO items. The variation of the number of participants in the table reflects the number of participants with corresponding scores at the given visits and at baseline.

The time to clinically meaningful deterioration in the global health status/Quality of life and Functional (Physical, Role, and Cognitive) subscales of the the QLQ-C30 was assessed from the time of the HER2-Targeted treatment to the worsening in the respective scales. Clinically meaningful deterioration is defined as a decrease in score of 10 points in Physical functioning and HRQoL; decrease of 7 points in Cognitive functioning, and decrease of 14 points in Role functioning.

The Patient Reported Outcome (PRO) Population includes participants with both a baseline assessment and at least one post-baseline assessment in any PRO items. This subset population includes only participants who received HER2-targeted therapy.