PFS was calculated from the time of randomization to either progression of disease, death, or treatment discontinuation because of unsatisfactory therapy results (such as global deterioration of health status). Disease progression was defined as 1 or more of the following: radiographic progression (20 percent [%] increase in measurable lesions, appearance of new lesions or unequivocal progression of evaluable lesions as defined by Response Evaluation Criteria in Solid Tumors [RECIST]); occurrence of new pleural/pericardial effusions or ascites confirmed by positive cytology; persistent hypercalcemia requiring more than 2 IV treatments with bisphosphonates; intervention for any cancer-related events (radiations, surgery) or new symptoms related to tumor growth requiring participant discontinuation; development of brain metastasis; or death for any cause. Median PFS was estimated from the Kaplan-Meier curve. 95% confidence interval (CI) is based on the Brookmeyer and Crowley method.

Full Analysis Set (FAS): all enrolled participants; grouped by randomized arm, where the first 10 participants enrolled but not randomly assigned to treatment were not included.

PFS was calculated from the time of randomization to either progression of disease, death, or treatment discontinuation because of unsatisfactory therapy results (such as global deterioration of health status). Disease progression was defined as 1 or more of the following: radiographic progression (20% increase in measurable lesions, appearance of new lesions or unequivocal progression of evaluable lesions as defined by RECIST); occurrence of new pleural/pericardial effusions or ascites confirmed by positive cytology; persistent hypercalcemia requiring more than 2 IV treatments with bisphosphonates; intervention for any cancer-related events (radiations, surgery) or new symptoms related to tumor growth requiring participant discontinuation; development of brain metastasis; or death for any cause. Median PFS was estimated from the Kaplan-Meier curve. 95% CI is based on the Brookmeyer and Crowley method.

FAS; only participants with baseline HbA1c <5.7% were included in the analysis.

Objective responses were defined using RECIST as CR: disappearance of all target and nontarget lesions. PR: at least a 30% decrease in the sum of the longest diameter (LD) of target lesions, taking as a reference the baseline sum LD. Nontarget lesions may persist provided there is no unequivocal progression in these lesions. SD: measurements demonstrating neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify as progressive disease (PD) during the first 6 weeks after the start of treatment taking as reference the smallest sum LD since the treatment started. During this time, nontarget lesions may persist provided there is no unequivocal progression in these lesions.

FAS

The study was terminated early due to strategic reasons and pharmacokinetic (PK) sampling was discontinued by Protocol Amendment 6. Therefore, the analysis was not performed and an incomplete data set was not reported because it would potentially skew the data, and is not statistically relevant, thus could be misleading.

The study was terminated early due to strategic reasons and PK sampling was discontinued by Protocol Amendment 6. Therefore, the analysis was not performed and an incomplete data set was not reported because it would potentially skew the data, and is not statistically relevant, thus could be misleading.

The study was terminated early due to strategic reasons and PK sampling was discontinued by Protocol Amendment 6. Therefore, the analysis was not performed and an incomplete data set was not reported because it would potentially skew the data, and is not statistically relevant, thus could be misleading.

Negative human anti-human antibodies were defined as <6.64

All randomized participants who started treatment and who had at least 1 sample submitted for the biomarker; collection of samples for this analysis was stopped after Amendment 6. All samples were negative to HAHA.

The study was terminated early due to strategic reasons and biomarker sampling was discontinued by Protocol Amendment 6. Therefore, the analysis was not performed and an incomplete data set was not reported because it would potentially skew the data, and is not statistically relevant, thus could be misleading.

The study was terminated early due to strategic reasons and biomarker sampling was discontinued by Protocol Amendment 6. Therefore, the analysis was not performed and an incomplete data set was not reported because it would potentially skew the data, and is not statistically relevant, thus could be misleading.

EORTC QLQ-C30: included functional scales (physical, role, cognitive, emotional, and social), global health status, symptom scales (fatigue, pain, nausea/vomiting) and single items (dyspnoea, appetite loss, insomnia, constipation/diarrhea and financial difficulties). Most questions used 4 point scale (1 'Not at all' to 4 'Very much'; 2 questions used 7-point scale (1 'very poor' to 7 'Excellent'). Scores averaged, transformed to 0-100 scale; higher score equals (=) better level of functioning or greater degree of symptoms.

The study was terminated early secondary to strategic reasons and the questionnaires were discontinued by Protocol Amendment 6. Therefore, the analysis was not performed and an incomplete data set was not reported because it would potentially skew the data, and is not statistically relevant, thus could be misleading.

EORTC-QLQ-BR23: included functional scales (body image, sexual functioning, sexual enjoyment, and future perspective) and single item symptoms scales (systemic therapy side effects, breast symptoms, arm symptoms, and upset by hair loss). Questions used 4-point Likert scale (1 ‘Not at All’ to 4 ‘Very Much’). Scores averaged and transformed to 0-100 scale. High score for functional scale=high/healthy level of functioning. High score for single item=high level of symptomatology/problems. Change from baseline=Cycle/Day score minus baseline score.

The study was terminated early secondary to strategic reasons and the questionnaires were discontinued by Protocol Amendment 6. Therefore, the analysis was not performed and an incomplete data set was not reported because it would potentially skew the data, and is not statistically relevant, thus could be misleading.