Overall survival (OS) time was measured as the time from date of randomization to the date of death due to any cause.

The median OS time and its 95.6% confidence interval were estimated using the Kaplan-Meier method. In the absence of confirmation of death, the participant was censored at the last date he/she was known to be alive or the study cut-off date (when 873 deaths have occurred), whichever was earlier.

The analysis was performed on the Intent-to-treat (ITT) population (i.e all randomized participants according to the treatment assigned regardless of the drug actually received).

At the cut-off date, 873 deaths had occurred, 445 in the Placebo group and 428 in the Aflibercept group.

Prostate specific antigen (PSA) response was defined as ≥50% decrease from baseline in serum PSA levels, confirmed at least 3 weeks later. Increases of any magnitude during the first 12 weeks were ignored in determining PSA response.

The analysis was performed on the ITT population evaluable for PSA response (i.e. with a baseline PSA ≥10 ng/mL).

Skeletal Related Events (SRE) included pathological fractures and/or spinal cord compression, need for bone irradiation, including radioisotopes or bone surgery, change in antineoplastic therapy to treat bone pain.

Time to SRE was defined as the time from the date of randomization to the date of occurence of the first event defining a SRE or death due to any cause, whichever occurred first.

The median time to SRE and its 95% confidence interval were estimated using the Kaplan-Meier method. In the absence of SRE, the participant was censored at the last date he/she was known to be alive or the study cut-off date, whichever was earlier.

The analysis was performed on the ITT population.

At the cut-off date, SRE or death had occurred in 1013 participants, 516 in the Placebo group and 497 in the Aflibercept group.

Disease progression was defined as a composite of: Radiological tumor progression (≥20% increase in target lesions, or appearance of at least 2 new bone lesions); PSA progression (≥25% increase in PSA level confirmed 3 weeks later); Pain progression (increase in pain intensity or in analgesic consumption for cancer related pain confirmed 3 weeks later); Radiotherapy for cancer related symptoms; Occurence of Skeletal related events (SRE).

Progression Free survival (PFS) time was measured as the time from the date of randomization up to the date of occurrence of the first event defining a disease progression or death due to any cause, whichever occurred first.

The median PFS time and its 95% confidence interval were estimated using the Kaplan-Meier method. In the absence of disease progression, the participant was censored at the the date of last assessment without evidence of progression or the study cut-off date, whichever was earlier.

The analysis was performed on the ITT population.

At the cut-off date, disease progression or death had occurred in 1184 participants, 592 in each treatment group.

Tumor response was defined as either a Complete Response (disappearance of all target lesions) or a Partial Response (≥30% decrease from baseline in target lesions) as assessed by Response Evaluation Criteria In Solid Tumors (RECIST)version 1.0.

The analysis was performed on the ITT population evaluable for tumor response (i.e. received at least one dose of study drugs (aflibercept/placebo or docetaxel), had no important deviations to protocol and was evaluable for response as per RECIST version 1.0).

Prostate specific antigen (PSA) progression was defined as ≥25% increase in PSA level confirmed 3 weeks later, above the nadir in participants who had achieved a PSA response, or above the baseline in participants who hadn't achieved a PSA response.

PSA progression-free survival (PFS) time was defined as the time from the date of randomization up to the date of the first documented PSA progression or death due to any cause, whichever occurred first.

The median PSA-PFS time and its 95% confidence interval were estimated using the Kaplan-Meier method. In the absence of PSA progression or death, the participant was censored at the the date of last assessment without evidence of progression or the study cut-off date, whichever was earlier.

The analysis was performed in the ITT population evaluable for PSA progression (i.e. with an evaluable baseline PSA).

At the cut-off date, PSA progression or death had occurred in 1138 participants, 571 in the Placebo group and 567 in the Aflibercept group.

Pain progression was defined as either ≥1-point increase in Present Pain Intensity (PPI) score or ≥25% increase in Analgesics Score (AS) confirmed at least 3 weeks later, or requirement for palliative radiotherapy. PPI scale is a self-report 0-5 scale to assess pain intensity - a score 0 reflects no pain, a score 5 reflects excruciating pain. AS is a scoring method to assess analgesics consumption. Each analgesic is scored 1 or 4 depending on the analgesic type and dose. AS is the sum of the analgesic scores.

Pain progression-free survival (PFS) time was measured as the time from the date of randomization up to the date of first pain progression or death due to any cause, whichever occurred first.

The median pain-PFS and its 95% confidence interval were estimated using the Kaplan-Meier method. In the absence of event, the participant was censored at the the date of last assessment without evidence of pain progression or the study cut-off date, whichever was earlier.

The analysis was performed on the ITT population evaluable for pain progression (i.e. with no pain or with stable pain at baseline).

At the cut-off date, pain progression or death had occurred in 507 participants, 263 in the Placebo group and 244 in the Aflibercept group.

Pain response was defined as either a ≥2-point decrease from baseline in Present Pain Intensity (PPI) score without increase in Analgesics Score (AS), or a ≥50% decrease from baseline in AS without increase in the PPI score confirmed at least 3 weeks later. Increases in PPI or AS during the first 12 weeks were ignored in determining pain response.

The analysis was performed in the ITT population evaluable for pain response (i.e. stable analgesia at baseline and, baseline PPI ≥2 and/or baseline AS ≥10 points).

Functional Assessment of Cancer Therapy-Prostate (FACT-P) is a 39-item participant questionnaire that measures the concerns of patients with prostate cancer. It consists of 5 subscales assessing physical well-being, social/family well-being, emotional well-being, functional well-being, and prostate-specific concerns.

FACT-P total score is the sum of the 5 subscores. It ranges from 0 to 156 with higher score indicating better quality of life.

The analysis was performed on the ITT population evaluable for Health related quality of life (i.e. with baseline and at least one post-baseline evaluable FACT-P questionnaire).

Adverse Events (AE) are any unfavorable and unintended sign, symptom, syndrome or illness observed by the investigator or reported by the participant during the study.

AE were collected at regular intervals throughout the study then graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE v.3.0).

The analysis was performed on the safety population (i.e. all randomized and treated participants according to the treatment actually received). Six participants in the Placebo group who received at least one dose of aflibercept in error were considered in the Aflibercept group.

Serum for detection of anti-drug antibodies (ADA) was collected in patients treated in selected centers only. Samples were analyzed using a titer-based, bridging immunoassay developed and validated to detect ADAs in human serum.

Samples with positive antibody levels were further analyzed using a validated, non-quantitative ligand binding assay to detect neutralizing antibodies Ab).

A participant was considered to have positive antibody levels if antibodies were detected above the quantification limits.

The analysis was performed on the safety population evaluable for immunogenicity (i.e. exposed to aflibercept with serum samples evaluable for immunogenicity).