PFS was calculated as the time (days) from date of randomization to date of first observed DP (per modified Response Evaluation Criteria In Solid Tumors [RECIST] or clinical judgment, whichever was earlier: CR, PR, stable disease, progressive disease) or death due to any cause, if death occurred before progression was documented. The actual date of tumor assessments was used for this calculation. PFS for subjects without progression or death was censored at the last date of tumor evaluation. PFS for subjects who had no tumor assessments after baseline and did not die was censored at 1 day.

PFS was analyzed for the intent to treat (ITT) population, defined as all subjects randomized to treatment. For the ITT population, subjects were included in the treatment group assigned at randomization, regardless of the treatment received.

Overall survival (OS) was calculated as the number of days from date of randomization to death date. Subjects who had not died at the time of analysis were censored at their last contact date.

OS was analyzed for the intent to treat (ITT) population, defined as all subjects randomized to treatment. For the ITT population, subjects were included in the treatment group assigned at randomization, regardless of the treatment received.

TTP was calculated as the time (days) from date of randomization to date of first observed disease progression (DP) (per modified RECIST or clinical judgment, whichever was earlier: CR, PR, stable disease, progressive disease). The actual dates of tumor assessments were used for this calculation. TTP for subjects without disease progression at the time of analysis, including subjects with death prior to progression, was censored at the last date of tumor evaluation. TTP for subjects who had no tumor assessments after baseline was censored at 1 day.

TTP was analyzed for the intent to treat (ITT) population, defined as all subjects randomized to treatment. For the ITT population, subjects were included in the treatment group assigned at randomization, regardless of the treatment received.

Duration of response was defined as the time from the first documented objective response of Partial Response (PR: At least a 30% decrease in the sum of the longest diameter [SLD] of target lesions, taking as reference the baseline SLD or better) or Complete Response (CR: Disappearance of all target lesions), whichever was noted earlier, to disease progression or death (if death occurred before progression was documented). Duration of response for subjects who had not progressed or died at the time of analysis was censored at the date of their last tumor assessment.

DOR was analyzed for the intent to treat (ITT) population, defined as all subjects randomized to treatment. For the ITT population, subjects were included in the treatment group assigned at randomization, regardless of the treatment received.

Change in ECOG PS is defined as an improvement (increase) or worsening (decrease) of at least one grade from the baseline ECOG score (from 0 [fully active] to 5 [dead]). Change in ECOG PS was recorded at the visit at which best confirmed response (BCR) using the modified RECIST (PR, CR, stable disease or Progressive Disease (PD)) was first noted (the change was 7% for both Sorafenib and Placebo). The BCR is the BCR recorded from the start of the treatment until DP/recurrence (taking as reference for DP, the smallest measurements recorded since treatment started).

Change in ECOG PS was analyzed for the intent to treat (ITT) population, defined as all subjects randomized to treatment. For the ITT population, subjects were included in the treatment group assigned at randomization, regardless of the treatment received.