The primary measure of spleen size was by MRI. MRIs were performed with a body coil because the objective was to measure organ volume only, not to assess for lesions. MRIs were performed by local radiologists who were instructed not to provide a quantitative measure of spleen volume, but could provide a qualitative assessment such as enlarged, smaller, larger, etc. The scans from an individual patient were to be read by a central reader upon transfer from the site radiologist.

Full Analysis Set (FAS): comprised of all patients who received at least one dose of ruxolitinib.

The best response rate was defined as the proportion of patients achieving a ≥ 35% reduction in spleen volume from baseline at any post-baseline assessment. The best response rate was estimated with an associated 95% confidence interval.

Full Analysis Set (FAS): comprised of all patients who received at least one dose of ruxolitinib.

For patients who had at least one ≥ 35% reduction in spleen volume from baseline at postbaseline, the duration of response was calculated. The start date of the duration was defined as the first spleen volume measurement that was ≥ 35% reduction from baseline, and the end date of the response duration was defined as the earliest of the following: death, A ≥ 25% increase in spleen volume by MRI (or CT in applicable patients) compared to baseline, Splenic irradiation, Leukemic transformation as defined by a bone marrow or a peripheral blood blast count of ≥ 20%, Splenectomy. Duration of response is calculated only for participants who achieved at least one measured >= 35% reduction in spleen volume at any time.

Full Analysis Set (FAS): comprised of all patients who received at least one dose of ruxolitinib.

Patient reported outcomes regarding the impact of MF on patients were assessed using the EORTC QLQ-C30. Data from the EORTC QLQ-C30 questionnaire was analyzed using the standardized scores. There were 2 categories to this scale: Functional/QOL scale and Symptom and Other items scale. For each sub-scale, the raw scores were standardized in order to obtain scores ranging from 0 to 100. For Functional/QOL subscales: a higher score represents a higher/better level of functioning. For Symptoms and Other items subscales: a higher score represents worse level of symptoms. The absolute change from baseline was calculated for each scale and summarized descriptively by scheduled visit.

Full Analysis Set (FAS): comprised of all patients who received at least one dose of ruxolitinib.

The Seven-day modified MFSAF v2.0 is a 7-item PRO instrument based on the modified MFSAF v2.0 diary administered at specified visits. Symptoms of myelofibrosis (MF) were assessed using this instrument & included filling up quickly/early satiety, abdominal discomfort, pain under the ribs, night sweats, itching, bone/muscle pain & inactivity. The first 6 items assessed MF symptom severity at its worst as recalled & the seventh captured MF-related inactivity in the 7 days prior to the clinic visit assessment. All 7 items asked subjects to record their answers on an 11-point numeric rating scale (NRS) (0=Absent, 10=Worst Imaginable). The first 6 items of the instrument focus on MF symptoms & are summed to create a Total Symptom score, defined as the sum of the 6 individual symptom scores other than the inactivity score (each with 0-10 point scale) collected on the same week. The total symptom scale ranges from 0 -60 where higher score indicates a worse level of the condition.

Full Analysis Set (FAS): comprised of all patients who received at least one dose of ruxolitinib.