OBJECTIVES

Several combination therapies are licensed for multiple myeloma (MM) across treatment lines, emphasizing the importance of selecting optimized treatment sequences. Our aim was to explore the impact of using alternative treatment sequences on patient outcomes.

METHODS

In the absence of true sequencing data, we developed a state-transition model using patient-level data (PLD) to model the efficacy of treatment sequences in MM. Efficacy for all European Medicines Agency-licensed treatments was derived either directly using trial data or indirectly using published comparative evidence. Due to their availability and maturity, progression-free survival (PFS) curves informed transition probabilities between treatment lines, with a separate distribution used to determine the split of PFS events between progressions and deaths. The model considers four lines of treatment and uses overall survival (OS) curves to model mortality in fourth-line (4L).

RESULTS

We observed that time spent progression-free decreases with each subsequent line of treatment. Optimizing efficacy in the first two treatment lines is important as these are most impactful on clinical outcomes. Using the most effective treatment sequences delays progression and deaths, and results in more time spent progression-free. Most regimens show a similar distribution of PFS events, with progressions being ~90% of observed PFS events. No combined network was available to model survival in 4L patients, and naïve comparison did not lead to plausible results. Therefore, a single overall survival curve is used for all patients.

CONCLUSIONS

In the absence of data from sequencing trials, we developed a robust, evidence-based model to provide valuable insights into optimal treatment decisions. In terms of clinical outcomes, using the best treatment early is beneficial to delaying progression. Results indicate that efficacy and time on treatment in early lines are the strongest drivers of long-term outcomes.