Spleen volume was assessed by magnetic resonance imaging (MRI) or by computed tomography (CT) scans if MRI was not suitable, and analyzed by a blinded central laboratory reader. Patients who withdrew, crossed over to ruxolitinib prior to the visit, or had a missing value at the visit were considered non-responders.

Intent-to-treat (ITT) population included all subjects randomized in the study. Treatment groups for this population were defined according to the treatment assignment at randomization. One patient was not included in the analysis due to a missing baseline spleen volume value.

The maintenance of ≥ 35% reduction from Baseline in spleen volume was assessed up until the data cutoff date using the Kaplan-Meier method for patients who had at least one measured ≥ 35% reduction, and who either had at least one subsequent measurement or who subsequently dropped out prior to another assessment.

Patients who had at least 1 measurement of ≥ 35% reduction from Baseline in spleen volume at any time during the study and had at least 1 subsequent measurement or withdrew prior to another assessment.

The duration of ≥ 35% reduction from Baseline in spleen volume was defined as the longest duration of consecutive measurements of ≥ 35% reduction observed before the data cut-off date for patients who had at least one measured ≥ 35% reduction, and who either had at least one subsequent measurement or, who subsequently dropped out prior to another assessment. The duration of a ≥ 35% reduction from Baseline in spleen volume was analyzed using the Kaplan-Meier method.

Patients who had at least 1 measurement of ≥ 35% reduction from Baseline in spleen volume at any time during the study and had at least 1 subsequent measurement or withdrew prior to another assessment.

Symptoms of myelofibrosis were assessed using a modified Myelofibrosis Symptom Assessment Form (MFSAF) Version 2.0 diary. Using the diary, patients rated the following symptoms on a scale from 0 (absent) to 10 (worst imaginable): night sweats, itching, abdominal discomfort, pain under ribs on left, feeling of fullness (early satiety), and muscle/bone pain. The total symptom score ranged from 0-60 and was calculated as the sum of the 6 symptom scores. A higher score indicates worse symptoms.

ITT evaluable population included patients with Baseline data and who did not have a 0 total score at both Baseline & Week 24; data measured after the cross over date were excluded. Patients who withdrew, met cross over criteria prior to Week 24 or had a 0 Baseline score & a nonzero/missing score at Week 24 were considered not meeting the endpoint.

Symptoms of myelofibrosis were assessed using a modified Myelofibrosis Symptom Assessment Form (MFSAF) Version 2.0 diary. Using the diary, patients rated the following symptoms on a scale from 0 (absent) to 10 (worst imaginable): night sweats, itching, abdominal discomfort, pain under ribs on left, feeling of fullness (early satiety), and muscle/bone pain. The total symptom score ranged from 0-60 and was calculated as the sum of the 6 symptom scores. A higher score indicates worse symptoms hence a negative change from baseline indicates improvement.

This analysis only includes patients who had a non-missing change from Baseline to Week 24. Data collected after the date of treatment cross over were not included in this analysis.

Overall survival is reported here by the number of deaths from randomization until the data cut-off. Patients were censored at the time of database cut-off or the later of either the date of withdrawal or the date of last follow-up for patients who withdrew from study before the date of data cut. The survival time was analyzed using the Kaplan-Meier method.

ITT population

Overall survival was assessed by the time to death or censoring. Patients were censored at the time of database cut-off or the later of either the date of withdrawal or the date of last follow-up for patients who withdrew from study before the date of data cut. The survival time was analyzed using the Kaplan-Meier method.

ITT population

Overall survival is reported here by the number of deaths from randomization until 01 March 2011. Patients were censored at this time or the later of either the date of withdrawal or the date of last follow-up for patients who withdrew from study before the date of data cut. The survival time was analyzed using the Kaplan-Meier method. This outcome reports data from August 2009 through March 2011 to coincide with a pre-planned New Drug Application (NDA) 120-day safety update.

ITT population

Overall survival was assessed by the time to death or censoring up until 01 March 2011. Patients were censored at this time or the later of either the date of withdrawal or the date of last follow-up for patients who withdrew from study before the date of data cut. The survival time was analyzed using the Kaplan-Meier method. This outcome reports data from August 2009 through March 2011 to coincide with a pre-planned New Drug Application (NDA) 120-day safety update.

ITT population

Overall survival is reported here by the number of deaths from randomization until the data cut-off. Patients were censored at the time of database cut-off or the later of either the date of withdrawal or the date of last follow-up for patients who withdrew from study before the date of database cut-off. The survival time was analyzed using the Kaplan-Meier method.

ITT population

Overall survival was assessed by the time to death or censoring. Patients were censored at the time of database cut-off or the later of either the date of withdrawal or the date of last follow-up for patients who withdrew from study before the date of database cut-off. The survival time was analyzed using the Kaplan-Meier method.

ITT population