Evaluating the Feasibility of a Network Meta-Analysis Comparing Treatment Options in Polycythemia Vera
Author(s)
Andrew Howe, BA, PharmD1, Noemi Hummel, PhD2, Claudia S. Castro, MS, PharmD3, Agnieszka Kopiec, MSc4, Hung Lun Chien, MPH5;
1Pharmaessentia, Woodstock, GA, USA, 2Certara GmbH, Research, Lörrach, Germany, 3Pharmaessentia, Medical, Burlington, MA, USA, 4Certara, Krakow, Poland, 5Pharmaessentia, Global HEOR, Burlington, MA, USA
1Pharmaessentia, Woodstock, GA, USA, 2Certara GmbH, Research, Lörrach, Germany, 3Pharmaessentia, Medical, Burlington, MA, USA, 4Certara, Krakow, Poland, 5Pharmaessentia, Global HEOR, Burlington, MA, USA
OBJECTIVES: Polycythemia vera (PV), a rare, chronic hematologic malignancy, negatively impacts patient outcomes. The optimal treatment approach is uncertain with head-to-head clinical trials lacking. Therefore, a targeted literature review (TLR) and subsequent feasibility of indirectly comparing ropeginterferon alfa-2b to peginterferon alfa-2a or ruxolitinib using standard of care as the common comparator was performed.
METHODS: A TLR screened clinical comparative evidence on PV treatments from 5/2014-5/2024 from PubMed and relevant conference abstracts. Key endpoints included complete hematologic response (CHR), molecular response, allele burden, event-free survival, and safety. Feasibility of performing a network meta-analysis (NMA) for these endpoints, ensuring homogeneity of trial populations and endpoint definitions, was assessed.
RESULTS: Of the records identified via the PubMed search (193) and screening of conference abstracts (460), 41 records were included, providing comparative evidence from 10 randomized controlled trials and 12 observational studies. Among those, 8 studies formed networks for the considered endpoints. Substantial heterogeneity in the evidence base precluded an NMA: Included studies differed with respect to important treatment effect modifiers, including individuals newly diagnosed or untreated, with high-risk features or low-risk PV, or refractory or intolerant to hydroxyurea (HU); endpoint definitions varied across studies, e.g., CHR required no disease-related symptoms, which was assessed inconsistently in the studies, and molecular response applied different thresholds; follow-up times varied considerably, prohibiting NMAs on safety outcomes; and standard of care was differently defined in the studies, e.g., 100% HU, or a mix of HU and other therapies.
CONCLUSIONS: A NMA for a rare disease like PV was not feasible due to heterogeneity among studies including the population of interest, treatments, outcome definitions, length of follow-up and potential treatment effect modifiers.
METHODS: A TLR screened clinical comparative evidence on PV treatments from 5/2014-5/2024 from PubMed and relevant conference abstracts. Key endpoints included complete hematologic response (CHR), molecular response, allele burden, event-free survival, and safety. Feasibility of performing a network meta-analysis (NMA) for these endpoints, ensuring homogeneity of trial populations and endpoint definitions, was assessed.
RESULTS: Of the records identified via the PubMed search (193) and screening of conference abstracts (460), 41 records were included, providing comparative evidence from 10 randomized controlled trials and 12 observational studies. Among those, 8 studies formed networks for the considered endpoints. Substantial heterogeneity in the evidence base precluded an NMA: Included studies differed with respect to important treatment effect modifiers, including individuals newly diagnosed or untreated, with high-risk features or low-risk PV, or refractory or intolerant to hydroxyurea (HU); endpoint definitions varied across studies, e.g., CHR required no disease-related symptoms, which was assessed inconsistently in the studies, and molecular response applied different thresholds; follow-up times varied considerably, prohibiting NMAs on safety outcomes; and standard of care was differently defined in the studies, e.g., 100% HU, or a mix of HU and other therapies.
CONCLUSIONS: A NMA for a rare disease like PV was not feasible due to heterogeneity among studies including the population of interest, treatments, outcome definitions, length of follow-up and potential treatment effect modifiers.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
SA51
Topic
Study Approaches
Topic Subcategory
Meta-Analysis & Indirect Comparisons
Disease
SDC: Oncology, SDC: Rare & Orphan Diseases