Economic evaluation of sequential treatments for follicular non-hodgkin lymphoma.

BACKGROUND

The cost-effectiveness analyses of follicular lymphoma (FL) treatments have focused on the second-line rituximab maintenance in patients with relapsed FL. The assessment of full FL treatment chain has been lacking.

OBJECTIVE

The aim of this study was to assess the cost-effectiveness of FL treatment sequences.

METHODS

Transitions between progression-free first-line treatment (PF1), progression-free second-line treatment (PF2), progression, and death health states were simulated with a probabilistic Markov model with half-cycle correction. At first, patients were assumed to be receiving rituximab with cyclophosphamide, doxorubicin, vincristine, and prednisone (RCHOP) induction. The first-line RCHOP induction responders continued without (RCHOP) or with (RCHOPR) the first-line rituximab-maintenance treatment. In the case of PF1 failure, patients received RCOPR/bendamustine or RCOPR/COP according to the European Society for Medical Oncology guidance. In the case of PF2 failure, patients were expected to receive the best supportive care (BSC). The survivals and adverse events were estimated with direct and indirect comparisons. Health outcomes and Finnish payer (drug, drug administration, monitoring, test, progression, serious adverse event) costs valued in 2010 euros were discounted with 3% per annum.

RESULTS

The mean discounted lifetime overall survival with FL was 9.6 to 11.5 years, quality-adjusted survival was 7.2 to 8.8 quality-adjusted life-years (QALYs), progression-free time was 7.7 to 10.2 years, and costs were €153,425 to €168,549, depending on the treatment sequence. The incremental cost-effectiveness ratios for RCHOPR→RCOPR/bendamustine→BSC, RCHOPR→RCOPR/COP→BSC, and RCHOP→RCOPR/bendamustine→BSC were €9575/€8014/€5900, €9881/€8310/€6013, and €8812/€7194/€5808, respectively, per QALY/life-year/progression-free year gained in comparison with RCHOP→RCOPR/COP→BSC. According to the cost-effectiveness acceptability frontier, the treatment of 61.8% to 72.7% patients with RCHOPR→RCOPR/bendamustine→BSC was cost effective at €20,000 to €30,000/QALY gained (expected value of perfect information [EVPI], €1287 to €1976/patient). The relative results were found to be robust in sensitivity analyses, and, in the direct comparison that included only head-to-head data, the first-line rituximab maintenance had 93.1% cost-effectiveness probability at €20,000/QALY gained (EVPI, €282/patient).

CONCLUSION

Sequences that included first-line rituximab maintenance is and second-line bendamustine are potentially cost effective in the treatment of FL.

LIMITATIONS

Because of data available, health outcomes of the first-line rituximab induction were excluded, the second-line patients on COP were assumed to incur the cost of COP, and the efficacy and adverse events of CHOP and the efficacy and adverse events of bendamustine were estimated indirectly according to a comparison of rituximab+bendamustine and RCHOP, and treatment benefits were truncated.