A Phase III Randomized, Double-blind, Placebo Controlled Trial Comparing the Efficacy of Gemcitabine, Cisplatin and Sorafenib to Gemcitabine, Cisplatin and Placebo in First-Line Treatment of Patients With Stage IIIb With Effusion and Stage IV Non-Small Cell Lung Cancer (NSCLC)

Overall survival (OS) was defined as the time from date of randomization to death due to any cause. Patients still alive at the time of analysis were censored at their last date of last contact.

Time frame: from randomization of the first patient until 38 months or date of death of any cause whichever came first

Population: Evaluation of OS based on ITT (non-squamous) population. Patients alive at the time of analysis were censored at their last date of follow-up (last visit or contact or at the data cut-off date). In the case of an incomplete date, where day was missing, day 15 (the middle of the month) was used.

DC was defined as the total number of patients whose best response was not PD according to RECIST (version 1.0) by Investigator-assessment (= total number of CR + total number of PR + total number of SD; CR or PR had to be maintained for at least 28 days from the first demonstration of that rating, SD had to be documented at least once more than 6 weeks from baseline). PD: an increase in the sum of tumor lesions sizes or new lesions.

Time frame: from randomization of the first patient until 38 months or date of death or progression whichever came first, assessed until discontinuation every 6 weeks up to 9 months and then every 12 weeks

Population: Evaluation of Disease Control based on ITT (non-squamous) population.

Duration of response was defined as the time from date of first documented objective response of PR or CR, whichever was noted earlier, to date of disease progression or death (if death occurred before progression was documented). Patients without disease progression at the time of analysis or death before progression were censored at the last date of tumor evaluation. Disease progression: increase in the sum of tumor lesion sizes or new lesions.

Time frame: from randomization of the first patient until 38 months or date of death or progression whichever came first, assessed until discontinuation every 6 weeks up to 9 months and then every 12 weeks

Population: Evaluation of duration of response based on ITT (non-squamous) population. No statistical testing performed.

Duration of SD was defined as the time from date of randomization to date that disease progression (radiological or clinical, whichever was earlier) was first documented. Patients without disease progression at the time of analysis or death before progression were censored at the date of their last tumor assessment.(Disease progression: increase in the sum of tumor lesion sizes or new lesions.) Duration of stable disease was only evaluated in patients failing to achieve a best response of CR or PR.

Time frame: from randomization of the first patient until 38 months or date of death or progression whichever came first, assessed until discontinuation every 6 weeks up to 9 months and then every 12 weeks

Population: Evaluation of duration of stable disease based on ITT (non-squamous) population. No statistical testing performed.

The EQ-5D also contains a visual analog scale (EQ-VAS), which records the respondent's self-rated health status on a vertical graduated visual analog scale ranging from 0 (worst imaginable health state) to 100 (best imaginable health state).

Time frame: from randomization of the first patient until 38 months later or death whatever occurs first

Population: All patients valid for the ITT analysis who have a baseline and at least one post baseline value.

The EQ-5D contains a descriptive system which measures 5 health dimensions: mobility, self-care, usual activity, pain/discomfort, and anxiety/depression. These five health dimensions are summarized into a single score, the EQ-5D index score which ranges from -0.594 to 1 when the United Kingdom (UK) weights are applied (0=death, 1=perfect health). Higher index scores represent better health states.

Time frame: from randomization of the first patient until 38 months later or death whatever occurs first

Population: All patients valid for the ITT analysis who have a baseline and at least one post baseline value.

The FACT-L measures health related quality of life (HRQOL) and composes of five domains: the four domains (physical well being, emotional well being, social well being, functional well being) from the Functional Assessment of Cancer Treatment-General scale (FACT-G) and the lung cancer subscale (LCS). The FACT-L total score ranges from 0 to 136, higher scores represent better HRQOL.

Time frame: from randomization of the first patient until 38 months

Population: All patients valid for the ITT analysis who have a baseline and at least one post baseline value.

LCS is a subscale of FACT-L measuring lung cancer specific symptoms. The LCS scores range from 0 to 28, higher scores represent fewer lung cancer symptoms.

Time frame: from randomization of the first patient to 38 months later or death whatever occurs first.

Population: All patients valid for the ITT analysis who have a baseline and at least one post baseline value.

OS was defined as the time from date of randomization to death due to any cause. Patients still alive at the time of analysis were censored at their last date of last contact.

Time frame: from randomization of the first patient until 38 months or date of death of any cause whichever came first

Population: Evaluation of OS based on ITT (both non-squamous and squamous) population. Patients alive at the time of analysis were censored at their last date of follow-up (last visit or contact or at the data cut-off date). In the case of an incomplete date, where day was missing, day 15 (the middle of the month) was used.

OS was defined as the time from date of randomization to death due to any cause. Patients still alive at the time of analysis were censored at their last date of last contact.

Time frame: from randomization of the first patient until 38 months or date of death of any cause whichever came first

Population: Evaluation of OS based on ITT (squamous) population. Patients alive at the time of analysis were censored at their last date of follow-up (last visit or contact or at the data cut-off date). In the case of an incomplete date, where day was missing, day 15 (the middle of the month) was used. No statistical testing performed.

Tumor response (= Best Overall Response) of a patient was defined as the best tumor response (confirmed Complete Response (CR: disappearance of tumor lesions), confirmed Partial Response (PR: a decrease of at least 30% in the sum of tumor lesion sizes), Stable Disease (SD: steady state of disease), or Progressive Disease (PD: an increase in the sum of tumor lesions sizes or new lesions)) observed during trial period assessed according to the RECIST criteria (version 1.0) based on Investigator-assessment.

Time frame: from randomization of the first patient until 38 months or date of death or progression whichever came first, assessed until discontinuation every 6 weeks up to 9 months and then every 12 weeks

Population: Evaluation of Tumour Response based on ITT (non-squamous) population.

PFS was defined as the time from date of randomization to disease progression (radiological or clinical, whichever was earlier, based on Investigator-assessment using Response Evaluation Criteria in Solid Tumors (RECIST), version 1.0) or death due to any cause, whichever occured first. Patients without progression or death at the time of analysis were censored at their last date of tumor evaluation. Disease progression: increase in the sum of tumor lesion sizes or new lesions.

Time frame: from randomization of the first patient until 38 months or date of death or progression whichever came first, assessed until discontinuation every 6 weeks up to 9 months and then every 12 weeks

Population: Evaluation of PFS based on ITT (non-squamous) population. PFS for patients with no tumour assessments after baseline was censored at one day.

TTP was defined as the time from date of randomization to disease progression (radiological or clinical, whichever was earlier, based on Investigator-assessment using RECIST version 1.0). Patients without progression at the time of analysis or death before progression were censored at their last date of tumor evaluation. Disease progression: increase in the sum of tumor lesion sizes or new lesions.

Time frame: from randomization of the first patient until 38 months or date of death or progression whichever came first, assessed until discontinuation every 6 weeks up to 9 months and then every 12 weeks

Population: Evaluation of TTP based on ITT (non-squamous) population. TTP for patients with no tumour assessments after baseline was censored at one day.

TTR for patients who achieved a best response (CR or PR) was defined as the time from date of randomization to the earliest date that response was first documented.

Time frame: from randomization of the first patient until 38 months or date of death of any cause whichever came first

Population: Evaluation of TTR based on ITT (non-squamous) population. No statistical testing performed.

TSD is defined as the time from randomization to the date of symptomatic deterioration (≥3 point decline in the LCS score that is maintained for at least 2 consecutive cycles) or death if death occurs before these 2 consecutive cycles are completed.

Time frame: from randomization of the first patient to 38 months later or death whatever occurs first

Population: All patients valid for the ITT analysis who have a baseline and at least one post baseline value.