A Study to Evaluate Patient Preference and Satisfaction of Subcutaneous Administration of the Fixed-Dose Combination of Pertuzumab and Trastuzumab in Participants With HER2-Positive Early Breast Cancer

Question 1 of the Patient Preference Questionnaire (PPQ) was as follows: "All things considered, which method of administration did you prefer?" The available options for a participant's response were IV, SC, or No preference. In Question 2 of the PPQ, participants who reported a preference for one of the two administration routes in Question 1 of the PPQ were asked to rate the strength of their preference (very strong, fairly strong, not very strong).

In Question 3 of the PPQ, participants who reported a preference for one of the two administration routes in Question 1 of the PPQ were asked to provide the two main reasons for their preference. The five available options for a participant's response were: Feels less emotionally distressing; Requires less time in the clinic; Lower level of injection-site pain; Feels more comfortable during administration; and Other reason.

The Therapy Administration Satisfaction Questionnaire (TASQ) is a 12-item, patient-reported questionnaire measuring the impact of each mode of treatment administration (TASQ-IV for IV treatment and TASQ-SC for SC treatment) on five domains: Physical Impact, Psychological Impact, Impact on Activities of Daily Living, Convenience, and Satisfaction. The TASQ-IV/-SC was administered at treatment Cycles 3 and 6 according to the order of treatment received per arm during the Cross-Over Period. Question 1 of the TASQ-IV/TASQ-SC is one of two items in the Satisfaction domain, with participants providing their answers to the following question: "How satisfied or dissatisfied were you with the IV infusion/SC injection?" The five available options for a participant's response were: very satisfied, satisfied, neither satisfied nor dissatisfied, dissatisfied, and very dissatisfied.

The TASQ is a 12-item, patient-reported questionnaire measuring the impact of the mode of treatment administration (TASQ-IV for IV treatment and TASQ-SC for SC treatment) on 5 domains: Physical Impact (3 items: Question [Q]2. Pain, Q3. Swelling, Q4. Redness), Psychological Impact (1 item: Q5. Feeling restricted), Impact on Activities of Daily Living (1 item: Q8. Lost/gained time), Convenience (2 items: Q6. Is it convenient?, Q7. Bothered by the amount of time?), and Satisfaction (2 items: Q1. How satisfied or dissatisfied are you with treatment?, Q12: Would you recommend the way you received the treatment?). In addition, 3 questions in the TASQ (Q9, Q10, Q11) are not part of the domains. Responses for the 3 domains that contain more than 1 item were scored from 0 to 100, with a higher score indicating a better outcome. Responses for the 2 domains with 1 item were scored from 1 to 5, with a higher score indicating a better outcome.

The TASQ is a 12-item, patient-reported questionnaire measuring the impact of the mode of treatment administration (TASQ-IV for IV treatment and TASQ-SC for SC treatment) on 5 domains. In addition, 3 questions (Q.9-11) are not part of the domains. The TASQ-IV/-SC was administered at treatment Cycles 3 and 6 according to the order of treatment received per arm during the Cross-Over Period. Question 9 asked the participant, "When you receive the IV infusion/SC injection treatment, are you able to talk to your nurse and/or doctor as much as you would like about your illness?" There were five available response options: a) Yes, I had more than enough time to talk to my nurse and/or doctor; b) Yes, but I would have liked more time to talk to my nurse and/or doctor; c) It does not matter to me if I have time to talk to my nurse and/or doctor during my treatment; d) No, I did not have enough time to talk to my nurse and/or doctor; and e) No, I did not talk to my nurse and/or doctor at all.

The Therapy Administration Satisfaction Questionnaire (TASQ) is a 12-item, patient-reported questionnaire measuring the impact of the mode of treatment administration (TASQ-IV for IV treatment and TASQ-SC for SC treatment) on 5 domains (questions [Q] 1 to 8 and Q12): Physical Impact, Psychological Impact, Impact on Activities of Daily Living, Convenience, and Satisfaction. In addition, 3 questions in the TASQ-IV/-SC (Q 9-11) are not part of the domains. The TASQ-IV/-SC were administered at treatment Cycles 3 and 6 according to the order of treatment received in each study arm during the Cross-Over Period. Question 10 of the TASQ-IV/-SC asked the participant "Does the IV infusion/SC injection impact the amount of time you have to talk to your nurse and/or doctor about your illness and other concerns?" There were two available options for the participant's response: Yes or No.

The Therapy Administration Satisfaction Questionnaire (TASQ) is a 12-item, patient-reported questionnaire measuring the impact of the mode of treatment administration (TASQ-IV for IV treatment and TASQ-SC for SC treatment) on 5 domains (questions [Q] 1 to 8 and Q12): Physical Impact, Psychological Impact, Impact on Activities of Daily Living, Convenience, and Satisfaction. In addition, 3 questions in the TASQ-IV/-SC (Q 9-11) are not part of the domains. The TASQ-IV/-SC was administered at treatment Cycles 3 and 6 according to the order of treatment received during the Cross-Over Period. Question 11 of the TASQ-IV/-SC asked the participant, "There are two ways to get cancer treatment: a) IV infusion given through a port or small tube; b) SC (subcutaneous) injection in your thigh. Which would you prefer?" There were three available options for the participant's response: IV, SC, or No Preference.

At treatment Cycle 6, Day 1, participants were expected to select the study treatment formulation (PH FDC SC or P+H IV) they would receive during the Treatment Continuation Period (starting at Cycle 7) to complete their 18 cycles of neo/adjuvant HER2-targeted treatment. Additionally, for each participant's preference category (SC, IV, and No preference) as per the question 1 of the patient preference questionnaire (PPQ), the percentage of participants who selected each treatment administration route for the Treatment Continuation Period (PH FDC SC or P+H IV) was summarized.

The Healthcare Professional Questionnaire (HCPQ)-Treatment Room Question 1 was completed at each treatment cycle of the Treatment Cross-Over Period by the healthcare professionals (HCPs) who administered treatment to the study's participants. HCPs responded to the following parts of Question 1 that sought to evaluate the amount of time it took to complete activities related to treatment administration: "If new IV access was needed for this cycle of treatment, please indicate what type of IV access was provided (central venous catheter, peripherally inserted central catheter, or peripheral vein cannulation) and how long (in minutes) this took to set up (only for participants receiving IV treatment)? How long (in minutes) did it take to administer the treatment, i.e. total infusion duration? How long (in minutes) was the patient in the Treatment Room for in total?"

HCPs who administered study treatment responded at Cycle 6 of the Treatment Cross-Over Period to the following HCPQ-Treatment Room Question 2: "If all P+H IV infusions are switched to FDC SC injections, please indicate how strongly you agree or disagree with each of the following statements: a) Patients will be moved outside of infusion unit to receive FDC SC; b) FDC SC route will allow more flexible scheduling; c) More patients will be treated in the infusion unit; d) Waiting list for any P+H IV treatment at the infusion unit will be reduced; e) Staff resources will be redistributed to other departments of the hospital; f) There will still be sufficient interaction time between HCPs and patients; g) Staff will spend more time for further education/development; h) Staff will dedicate more time attending to administrative tasks for Perjeta-Herceptin patients; i) Patients will spend less time in the care unit; j) Administration by FDC SC injection is preferred by patients."

Healthcare professionals (HCPs) who administered study treatment responded at Cycle 6 of the Treatment Cross-Over Period to the following HCPQ-Treatment Room Questions 3 to 7: "Looking back over the Perjeta-Herceptin treatment sessions, please indicate based on your opinion which administration method: Q3. Which Method Was Most Convenient for the Patient? Q4. Which Method Was Best for Optimizing Patient Care in Your Centre? Q5. Which Method Took the Least Time from Start to Finish of Administration? Q6. Which Method Required the Least Resource Use for Administration? Q7. Which Method Was Preferred by Patients?" The four available response options were: P+H IV, FDC SC, No Difference, and Unsure.

Healthcare professionals (HCPs) who administered study treatment responded at Cycle 6 of the Treatment Cross-Over Period to the following HCPQ-Treatment Room Question 8: "How frequently would you offer or recommend FDC SC administration to your patients in the future?" The three available response options were: Always, Sometimes, and Never.

The Healthcare Professional Questionnaire (HCPQ)-Drug Preparation Room Question 1 was completed at each treatment cycle of the Treatment Cross-Over Period by the healthcare professionals (HCPs) within the pharmacy/drug preparation area where pertuzumab IV and trastuzumab IV and pertuzumab and trastuzumab FDC SC were prepared and dispensed for treating the study's participants. HCPs responded to the following question: "How long (in minutes) did it take to prepare the treatment for use?"

Healthcare professionals (HCPs) who prepared study treatment within the pharmacy/drug preparation area responded at Cycle 6 of the Treatment Cross-Over Period to the following HCPQ-Drug Preparation Room Question 2: "If all P+H IV infusions are switched to FDC SC injections, please indicate how strongly you agree or disagree with each of the following statements: a) Staff will have increased availability for other tasks in the pharmacy; b) Administrative procedures around FDC SC will require less time; c) FDC SC formulations will provide more flexibility for staff in managing their workload; d) Due to ready-to-use FDC SC formulations, potential dosing errors will be avoided; e) Due to ready-to-use FDC SC formulations, there will be less drug wastage; f) Without having to reconstitute the drug, less storage space for FDC SC related supplies will be required in the pharmacy; g) Preparation procedures and associated time staff time commitment will be reduced."

Healthcare professionals (HCPs) who prepared study treatment within the pharmacy/drug preparation area responded at Cycle 6 of the Treatment Cross-Over Period to the following HCPQ-Drug Preparation Room Questions 3 and 4: "Looking back over the Perjeta-Herceptin treatment sessions, please indicate based on your opinion which administration method: Q3. Was quickest from start to end of preparation to finish of administration (excluding observation period)?; Q4. Required less resource use for preparation and administration, for example nursing time, facility costs, equipment etc?" The three available response options were: P+H IV, FDC SC, and No Difference.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

The EORTC QLQ-C30 questionnaire consisted of 30 questions covering treatment-related symptoms, functioning, and GHS/HRQoL. Questions were answered by participants on a scale from 1 to 4 or 1 to 7. Raw scores were transformed to scale scores that ranged from 0 to 100 where a higher scale score indicating a higher response (i.e., on the functioning and GHS/QoL scales a higher score meant better functioning/QoL, whereas on the symptom scales a higher score meant greater [worse] symptoms). The EORTC QLQ-C30 data was scored according to the EORTC scoring manual (Fayers 2001). In the event of incomplete data, for all questionnaire subscales, if more than 50% of the constituent items are completed, a pro-rated score was to be computed consistent with the scoring manuals and published validation reports. For subscales with less than 50% of the items completed, the subscale was to be considered as missing.

Investigators used the NCI CTCAE v4.0 grading scale for assessing adverse event (AE) severity; if not listed, AE severity was graded as follows: Grade 1 = mild; Grade 2 = moderate; Grade 3 = severe or medically significant; Grade 4 = life-threatening; Grade 5 = death related to AE. Severity and seriousness are not synonymous and investigators independently assessed these criteria for each AE. Investigators also determined whether an AE was considered to be related to the study drug. Adverse events to monitor were defined based on known risks associated with the study drugs and included: hypersensitivity reactions, administration-related reactions (ARRs), cardiac dysfunction, diarrhea grade ≥3, rash/skin reactions, mucositis, interstitial lung disease (ILD), (febrile) neutropenia, pulmonary events that may occur as a result of an ARR, and pregnancy/neonatal related. Multiple occurrences of AEs were counted only once per participant. LVEF = left ventricular ejection fraction

Investigators used the NCI CTCAE v4.0 grading scale for assessing adverse event (AE) severity; if not listed, AE severity was graded as follows: Grade 1 = mild; Grade 2 = moderate; Grade 3 = severe or medically significant; Grade 4 = life-threatening; Grade 5 = death related to AE. Severity and seriousness are not synonymous and investigators independently assessed these criteria for each AE. Investigators also determined whether an AE was considered to be related to the study drug. Adverse events to monitor were defined based on known risks associated with the study drugs and included: hypersensitivity reactions, administration-related reactions (ARRs), cardiac dysfunction, diarrhea grade ≥3, rash/skin reactions, mucositis, interstitial lung disease (ILD), (febrile) neutropenia, pulmonary events that may occur as a result of an ARR, and pregnancy/neonatal related. Multiple occurrences of AEs were counted only once per participant. LVEF = left ventricular ejection fraction

Heart failure is defined as a disorder characterized by the inability of the heart to pump blood at an adequate volume to meet tissue metabolic requirements, or, the ability to do only at an elevation in the filling pressure. Any adverse event of symptomatic left ventricular systolic dysfunction (LVSD; also referred to as heart failure) occurring during the study was to be reported as a serious adverse event.

Left ventricular ejection fraction (LVEF) is the measurement of how much blood is being pumped out of the left ventricle of the heart (the main pumping chamber) with each contraction. All participants who enrolled in this study must have had a baseline LVEF ≥55%. Verbatim description of adverse events was mapped to Medical Dictionary for Regulatory Activities (MedDRA) version 25.1. The MedDRA preferred term of 'ejection fraction decreased' is defined as an LVEF decrease of at least 10 percentage points from baseline and to below 50%.

The number of participants at any post-baseline timepoint with abnormal readings outside the normal range for vital signs of diastolic and systolic blood pressure, pulse rate, respiratory rate, and body temperature were summarized according the specified direction of the abnormal reading (high or low). The number analyzed (denominator) in the results table represents participants without the specified abnormal vital sign at baseline. Not every vital sign abnormality qualified as an adverse event (AE). A vital sign result must have been reported as an AE if it met any of the following criteria: was accompanied by clinical symptoms; resulted in a change in study treatment (e.g., dosage modification, treatment interruption, or treatment discontinuation); resulted in a medical intervention or a change in concomitant therapy; or, was clinically significant in the investigator's judgment.

Laboratory data for targeted chemistry and hematology parameters were classified according to the NCI CTCAE v4.0; Grade 0 is normal and Grades 1 to 4 represent worsening levels of the parameter outside of the normal range in the specified direction of the abnormality (high and low are above and below the range, respectively). The results show the shifts in the number of participants with Grade 0-2 at baseline to Grade 3-4 post-baseline; those with missing baseline values were counted as Grade 0-2 at baseline. Not every laboratory abnormality qualified as an adverse event, only if it met any of the following criteria: was accompanied by clinical symptoms; resulted in a change in study treatment (e.g., dosage modification, treatment interruption, or treatment discontinuation); resulted in a medical intervention or a change in concomitant therapy; or, was clinically significant in the investigator's judgment. SGOT/AST = aspartate aminotransferase; SGPT/ALT = alanine aminotransferase

Overall survival (OS) is defined as the time from randomization to death due to any cause. The number of participants who had an OS event (i.e., died) while on study is reported.

Overall survival is defined as the time from randomization to death due to any cause. Participants who were not reported as having died at the time of analysis were censored at the date when they were last known to be alive. Participants who did not have post-baseline information were censored at the date of randomization +1 day. Kaplan-Meier methodology was used to estimate the percentage of participants who were alive (event-free) at 12, 24, and 36 months.

Invasive Disease-Free Survival is defined as the time from randomization to the first occurrence of one of the following events: ipsilateral invasive breast tumour recurrence, ipsilateral local-regional invasive breast cancer recurrence, distant recurrence, contralateral invasive breast cancer, and death from any cause. Second primary non-breast invasive cancer (with the exception of non-melanoma skin cancers and in situ carcinoma of any site) was included as an event.

Invasive Disease-Free Survival is defined as the time from randomization to the first occurrence of one of the following events: ipsilateral invasive breast tumour recurrence, ipsilateral local-regional invasive breast cancer recurrence, distant recurrence, contralateral invasive breast cancer, and death from any cause. Second primary non-breast invasive cancer (with the exception of non-melanoma skin cancers and in situ carcinoma of any site) was included as an event. Participants who had not experienced invasive disease at the time of analysis were censored: i) at the time of the last clinical breast examination if they had post-baseline clinical breast examination; ii) on the date of randomization +1 day if no post-baseline clinical breast examination. Kaplan-Meier methodology was used to estimate the percentage of participants who were event-free at 12, 24, and 36 months.

Invasive Disease-Free Survival is defined as the time from randomization to the first occurrence of one of the following events: ipsilateral invasive breast tumour recurrence, ipsilateral local-regional invasive breast cancer recurrence, distant recurrence, contralateral invasive breast cancer, and death from any cause. Ipsilateral or contralateral in situ disease and second primary non-breast cancers (including in situ carcinomas and non-melanoma skin cancers) were not counted as recurrence.

Invasive Disease-Free Survival is defined as the time from randomization to the first occurrence of one of the following events: ipsilateral invasive breast tumour recurrence, ipsilateral local-regional invasive breast cancer recurrence, distant recurrence, contralateral invasive breast cancer, and death from any cause. Ipsilateral or contralateral in situ disease and second primary non-breast cancers (including in situ carcinomas and non-melanoma skin cancers) were not counted as recurrence. Participants who had not experienced invasive disease at the time of analysis were censored: i) at the time of the last clinical breast examination if they had post-baseline clinical breast examination; ii) on the date of randomization +1 day if no post-baseline clinical breast examination. Kaplan-Meier methodology was used to estimate the percentage of participants who were event-free at 12, 24, and 36 months.

Distant disease-free survival (DDFS) is defined as the time from randomization to the date of distant breast cancer recurrence (i.e., evidence of breast cancer in any anatomic site other than for ipsilateral [loco-regional] invasive breast cancer recurrence that has either been histologically confirmed or clinically diagnosed as recurrent invasive breast cancer).

Distant disease-free survival (DDFS) is defined as the time from randomization to the date of distant breast cancer recurrence (i.e., evidence of breast cancer in any anatomic site other than for ipsilateral [loco-regional] invasive breast cancer recurrence that has either been histologically confirmed or clinically diagnosed as recurrent invasive breast cancer). Participants who had not experienced invasive disease at the time of analysis were censored: i) at the time of the last clinical breast examination if they had post-baseline clinical breast examination; ii) on the date of randomization +1 day if no post-baseline clinical breast examination. Kaplan-Meier methodology was used to estimate the percentage of participants who were event-free at 12, 24, and 36 months.