Estimating the Number of US Patients with Multiple Myeloma (MM) at 5 or More Lines of Treatment (LOT)
Author(s)
Nikolaou A1, Maiese EM2, Samyshkin Y3, Sansbury L4, Oguz M5, Cid Ruzafa J5, Sapra S6, Kapoor R1, Wang F6
1Modelling and Simulation, Evidera, London, UK, 2Value Evidence and Outcomes, GlaxoSmithKline, Philadelphia, PA, USA, 3GlaxoSmithKline, Brentford, UK, 4Value Evidence and Outcomes, GlaxoSmithKline, Research Triangle Park, NC, USA, 5Real-World Evidence, Evidera, London, UK, 6Value Evidence and Outcomes, GlaxoSmithKline, Upper Providence, PA, USA
OBJECTIVES: As the MM treatment paradigm evolves with the availability of new treatments, patients may receive additional LOTs. Estimates of patient numbers at each LOT are limited, although this data may be important to inform health coverage and resource allocation decisions. This study aimed to assess the number of US patients with MM by LOT in 2020. METHODS: Differential equations were used to develop a five-compartment model: LOT1–LOT5+. Patients in LOT1–LOT4 could die or transition to the next LOT. Patients in LOT5+ could remain in that compartment or die. Each LOT was sub-divided based on stem-cell transplant (SCT) eligibility and treatment type (anti-CD38 therapy or ‘other’; patients may switch treatment with LOT progression). Patient numbers were stratified by age (<65 and ≥65 years) and cytogenetic risk. Results were aggregated to provide patient numbers by LOT. For ‘other’ treatments, mortality and time to next treatment (TTNT) depended on LOT, SCT, age and cytogenetic risk and were informed from published epidemiological/real-world studies. Anti-CD38 (i.e. daratumumab) use data were based on relative treatment effects in randomised controlled trials. Anti-CD38 use per LOT was based on 2020 market-research data. Probabilistic and deterministic sensitivity analyses quantified the impact of input data variability. RESULTS: The model predicted an MM prevalence of 128,326 US patients, with 106,176 actively receiving treatment (LOT1: 53.9%; LOT2: 25.8%; LOT3: 10.7%; LOT4: 4.9%; LOT5+: 4.7%). An estimated 69.7% (3511/5034) in LOT5+ (standard deviation=861) had prior exposure to proteasome inhibitors, immunomodulatory agents and/or anti-CD38 antibodies. LOT5+ model results were most sensitive to mortality, TTNT and anti-CD38 use. CONCLUSIONS: This novel model predicted the number of patients with MM by LOT with an overall prevalence within a 10% deviation of SEER estimates (140,779 US patients); and provides a framework that can be adapted to other countries and healthcare systems. Funding: GlaxoSmithKline: (213290)
Conference/Value in Health Info
2020-11, ISPOR Europe 2020, Milan, Italy
Value in Health, Volume 23, Issue S2 (December 2020)
Code
PCN182
Topic
Epidemiology & Public Health
Disease
Oncology