AMG 531

No thrombocytopenia-induced modification of any myelosuppressive agent in the second and third cycles of the planned on-study chemotherapy regimen. Thrombocytopenia-induced modifications include chemotherapy dose reduction, delay, omission, or chemotherapy treatment discontinuation due to platelet counts below 100 x 109/L

Participants met the criteria of the primary endpoint if there was no thrombocytopenia-induced modification of any myelosuppressive treatment agent in the second and third cycles of the planned on-trial chemotherapy regimen (cycles were up to 3 weeks). A thrombocytopenia-induced modification was defined as any dose reduction, dose delay, dose omission, and/or early chemotherapy treatment discontinuation due to low platelet counts less than 100 x 10\^9/L. The 95% confidence interval (CI) is based on the exact Clopper-Pearson method.

Platelet response was defined as a platelet count of ≥ 50 x 10⁹/L with no rescue medication use for ITP in the past 4 weeks. Monthly platelet response was calculated based on the median platelet count during each month. For each participant, the percentage of time with platelet response during the first 6 months was calculated as the number of months a platelet response was observed divided by the total number of months response was assessed.

The percentage of participants who developed collagen as evidenced by trichrome staining, defined as a Grade 4 on the modified Bauermeister grading scale: Grade 0: No reticulin fibers demonstrable Grade 1: Occasional fine individual fibers and foci of a fine fiber network Grade 2: Fine fiber network throughout most of the section; no coarse fibers Grade 3: Diffuse fiber network with scattered thick coarse fibers but no mature collagen (negative to trichrome staining) Grade 4: Diffuse, often course fiber network with areas of collagenization (positive trichrome staining)

The percentage of participants with an increased modified Bauermeister grade defined as an increase by ≥ 2 severity grades or an increase to grade 4 (i.e., grade 0 to 2-4, grade 1 to 3-4, grade 2 to 4, or grade 3 to 4 over baseline). The modified Bauermeister grading scale: Grade 0: No reticulin fibers demonstrable Grade 1: Occasional fine individual fibers and foci of a fine fiber network Grade 2: Fine fiber network throughout most of the section; no coarse fibers Grade 3: Diffuse fiber network with scattered thick coarse fibers but no mature collagen (negative to trichrome staining) Grade 4: Diffuse, often course fiber network with areas of collagenization (positive trichrome staining) Participants without an evaluable baseline result were assumed to have a baseline modified Bauermeister score of 0.

The percentage of participants with bone marrow abnormalities (eg, myelodysplastic syndrome, monosomy 7) based on analysis of bone marrow biopsy and aspirate samples using cytogenetics and fluorescence in situ hybridization.

A participant with durable platelet response was defined as achieving at least 6 weekly platelet counts of ≥ 50 x 10\^9/L during the last 8 weeks of treatment (platelet counts obtained from week 18 to week 25). If a platelet count from a participant was not available (missing) in a certain week, that week was imputed as non-response for that participant. Platelet counts were not deemed as a positive response for 4 weeks after the administration of rescue medication.

The adverse event (AE) severity grading scale used was the Common Terminology Criteria for Adverse Events (CTCAE) version 3.0 grading scale, where Grade 1 = Mild AE; Grade 2 = Moderate AE; Grade 3 = Severe AE; Grade 4 = Life-threatening or disabling AE; Grade 5 = Death related to AE. A serious adverse event was defined as an adverse event that met at least one of the following serious criteria: * fatal * life threatening (places the subject at immediate risk of death) * required in-patient hospitalization or prolongation of existing hospitalization * resulted in persistent or significant disability/incapacity * congenital anomaly/birth defect * other medically important serious event. The investigator assessed whether each adverse event was possibly related to the investigational product.

Exposure adjusted rate was defined as the total number of events divided by the duration of time participants were under observation. The adverse event (AE) severity grading scale used was the Common Terminology Criteria for Adverse Events (CTCAE) version 3.0 grading scale, where Grade 1 = Mild AE; Grade 2 = Moderate AE; Grade 3 = Severe AE; Grade 4 = Life-threatening or disabling AE; Grade 5 = Death related to AE. A serious adverse event was defined as an adverse event that met at least one of the following serious criteria: * fatal * life threatening (places the subject at immediate risk of death) * required in-patient hospitalization or prolongation of existing hospitalization * resulted in persistent or significant disability/incapacity * congenital anomaly/birth defect * other medically important serious event. The investigator assessed whether each adverse event was possibly related to the study drug.

Two validated assays were used to test for antibodies to romiplostim / the thrombopoietin-mimetic peptide component of romiplostim (TMP). The first was an immunoassay to confirm the presence of antibodies. The second was a cell-based bioassay to detect neutralizing or inhibitory effects in vitro. If a sample was positive in both assays, a participant was defined as positive for neutralizing antibodies. Transient antibodies are those positive post-baseline but negative at the last time point tested. Persistent antibodies were those positive at the last time point tested.

Two validated assays were used to test for antibodies to endogenous thrombopoietin (TPO). The first was an immunoassay to confirm the presence of antibodies. The second was a cell-based bioassay to detect neutralizing or inhibitory effects in vitro. If a sample was positive in both assays, a participant was defined as positive for neutralizing antibodies. Transient antibodies are those positive post-baseline but negative at the last time point tested. Persistent antibodies were those positive at the last time point tested.

Number of weeks with weekly platelet response. A weekly platelet response is defined as a platelet count of ≥ 50 x 10\^9/L on a weekly scheduled dose day from week 2 to week 13.

One or more occurences of one or more adverse events within the participant during the study. Participants with more than one event were only counted once

defined as PLT \< 50,000 cells/mcl) at the end of cycle 1 (C1D21±2d); presence of grade 4 thrombocytopenia (\<25,000 cells/mcl) at any time throughout the treatment cycle; or indication for a PLT transfusion for thrombocytopenia or for bleeding at any time throughout the treatment cycle.

evaluation of safety, defined by the incidence of grade \>/= 4 adverse events (NCI CTCAE v. 4.02 Dec 2009)

The primary endpoint was the number of months a participant achieved a platelet response during the 12-month treatment period. A platelet response for any 1 month was defined as the median of platelet counts measured in the month ≥ 50 x 10\^9/L. Platelet counts within 4 weeks following a rescue medication use or following splenectomy were considered non-response. Months without any platelet count measurement were considered as months with no platelet response.

A clinically significant bleeding event is defined as any bleeding event of grade ≥ 2 per the modified World Health Organization (WHO) bleeding scale: • Grade 0 = no bleeding • Grade 1 = petechia or mucosal or retinal bleeding not requiring intervention • Grade 2 = melena, hematemesis, hematuria, hemoptysis • Grade 3 = bleeding required red cell transfusion • Grade 4 = retinal bleeding with visual impairment • Grade 5 = non-fatal cerebral bleeding • Grade 6 = fatal cerebral bleeding • Grade 7 = fatal non-cerebral bleeding. Bleeding events that continue for more than 7 days were counted as separate events every eighth day. Multiple events that arose from one organ system on one day were collapsed into one single event. Bleeding events with a start date between the first dose date and the last dose date of the test treatment period+7 days are included.

Occurrence of one or more clinically significant thrombocytopenic events, defined as either Common Terminology Criteria for Adverse Events (CTCAE) v. 3 grade 3 or 4 thrombocytopenia starting from week 3 of cycle 1 or receipt of platelet transfusions starting from week 1 of cycle 1 and continuing through the end of treatment visit.

The number of participants experiencing one or more adverse events during the treatment phase or extension phase of Part A.

The number of participants experiencing one or more adverse events during the treatment phase or extension phase of Part B.

The development of antibodies to romiplostim or to endogenous thrombopoietin (eTPO) was assessed using a neutralizing bioassay. Participants positive for neutralizing antibodies at any of the assessments during the study are reported.

Change in platelet nadir from the previous qualifying cycle to the first treatment cycle.

The presence or development of antibodies to romiplostim and endogenous thrombopoietin was assessed using a neutralizing bioassay. Antibody analyses were conducted on study days 29 and at the end-of-study visit (day 78). The number of participants with positive antibody binding at any time during the study is reported.

Categorize and quantify AEs per CTCAE version 5.0

Improvement in at least one of the cell blood lineages by 24 weeks of therapy: 1. Platelet response (increase to 10 X 103/mL above baseline or stable platelet counts with transfusion independence for a minimum of 2 weeks in those who were transfusion dependent on entry into the protocol) 2. Erythroid response (when pretreatment hemoglobin was, below 7 g/dL, defined as an increase in hemoglobin by 1.5 g/dL or, in transfused patients, a reduction in the units of packed red blood cell transfusions by an absolute number of at least 4 transfusions for 8 consecutive weeks, compared with the pretreatment transfusion number in the previous 8 weeks) 3. Neutrophil response (when pretreatment absolute neutrophil count \[ANC\] of 0.5 x103 /mL as at least a 100% increase in ANC, or an ANC increase by 0.5 x103 /mL)