Long-term Safety and Efficacy of Momelotinib in Subjects With Primary Myelofibrosis, Post-polycythemia Vera Myelofibrosis, Post-essential Thrombocythemia Myelofibrosis, Polycythemia Vera or Essential Thrombocythemia

The number of subjects achieving a spleen response, defined as a reduction of 50% or more in palpable splenomegaly of a spleen that was at least 10 cm below the LCM at baseline, or a spleen that was palpable at \> 5 cm and \< 10 cm below the LCM at baseline becoming not palpable for at least 56 days, using baseline of the parent study as the reference.

The interval from the first onset of splenic response (in the parent study or Study GS-US-352-1154) to the earliest date of loss of splenic response. Loss of response was defined as the reduction of splenomegaly by \< 50% among responders (with splenomegaly ≥ 10 cm below the LCM at baseline) that lasts ≥ 56 days, or the recurrence of \> 0 cm splenomegaly among responders (with splenomegaly \> 5 and \< 10 cm at baseline) that lasts ≥ 56 days.

Duration of splenic response was measured by descriptive statistics. Data from responders who maintained their response was censored at the last assessment date.

The number of transfusion dependent subjects at entry to a parent study who became transfusion-independent for ≥ 12 weeks at any time from the first dose of momelotinib in the parent study until the end of Study GS-US-352-1154.

The interval from the first onset date of transfusion independence (in the parent study or Study GS-US-352-1154) to the earliest date of loss of response for participants who are transfusion dependent at baseline in the parent study. Loss of TI response was defined as receiving an RBC transfusion after achieving a TI response.

Duration of transfusion independence response was measured by descriptive statistics. Data from responders who maintained their response was censored at the last assessment date.

The number of subjects achieving an anemia response, defined as:

* Achieving transfusion independence for ≥ 12 weeks, for subjects who were transfusion-dependent at baseline in the parent study, or
* Having ≥ 2 g/dL increase in Hgb from baseline for ≥ 12 weeks, for subjects with Hgb \< 10 g/dL at baseline in the parent study who were not transfusion-dependent (Cohort 1) or who were transfusion-independent (Cohort 2).

The interval from the first onset of anemia response (in the parent study or Study GS-US-352-1154) to the earliest date of loss of anemia response. Loss of anemia response was defined as having any RBC transfusion after achieving an anemia response.

Duration of anemia response was measured by descriptive statistics. Data from responders who maintained their response was censored at the last assessment date.

The average number of RBC units per subject month during the parent study and/or Study GS-US-352-1154.

The interval from the first dose of momelotinib in the parent study until death from any cause.

Overall survival was analyzed using the Kaplan-Meier method. Data from subjects who were lost to follow-up or remained alive until the end of the study were censored at the date of last contact or last response assessment.

The interval from the first dose of momelotinib in the parent study until the first documentation of definitive progressive disease as defined in 2006 IWG-MRT or death due to any cause.

Subjects who were free of progression were censored at the last assessment date.

The interval from the first dose of momelotinib in the parent study until the first documented leukemic transformation or death from any cause. Leukemic transformation was documented in the adverse event electronic case report form.

Leukemia-free survival was analyzed using the Kaplan-Meier method. Subjects who were free of leukemia transformation were censored at the last assessment date.