Cost-Effectiveness Analysis of Arsenic Trioxide in Combination with All-Trans Retinoic Acid in Acute Promyelocytic Leukemia with Pretreatment White Blood Counts <10G/L

Background: Chemotherapy-free treatment with arsenic trioxide and all-trans retinoic acid (ATO/ATRA) of patients with acute promyelocytic leukemia (APL) with presenting white blood counts (WBC) < 10 G/l has been shown to be at least equivalent or even better with regard to survival and quality of life compared to standard treatment according to the AIDA scheme which includes idarubicin, mitoxantrone in combination with ATRA followed by a two-year maintenance therapy with 6-mercaptopurin, methotrexate and ATRA (Lo Coco F et al., N. Engl. J. Med. 2013;369(2):111-21; Efficace F et al J Clin Oncol. 2014 accepted).

Aims: To evaluate costs in relation to benefits in a cost-effectiveness analysis comparing ATO/ATRA to standard treatment with AIDA in newly diagnosed APL with WBC<10G/l from a third party payers perspective

Methods: The study included 55 patients with newly diagnosed APL treated in Germany within the APL0406 study (NCT00482833), n=29 in the experimental arm (ATO) and n=26 in the standard arm (AIDA). Costs were calculated based on data of treated patients to determine average estimates as of 2014. Costs were calculated based on the background of the German Diagnosis-Related Groups system including additional reimbursement for platelet transfusions and ATO as well as payment for out-patient treatment and care. Costs for 1 mg arsenic trioxide were 46.17€ both for in-patient and out-patient treatment.

Results: For induction therapy all patients were hospitalized. The median hospital stay, the proportion of patients with severe complications comprising coagulopathy, differentiation syndrome, and fever triggering a higher DRG grouping, as well as median amount of platelet transfusion were 30 days (range, 16-47 days) and 33 days (range, 21-80 days) (p=0.13), 67% (20/29) and 81% (21/26) (p=0.37), 5 (range, 0-32) and 10 (range, 0-30) (p=0.13) for the ATO-arm and the AIDA-arm, respectively. The median amount of administered ATO was 290mg (range, 100-780mg) in the ATO-arm. Total costs for induction therapy were calculated with 27,211€ and 15,472€ for the ATO-arm and the AIDA-arm, respectively.

During the 4 consolidation cycles in the ATO-arm, 21 of 94 cycles were administered on an in-patient basis (median duration of hospital stay, 20 days; range, 1-30 days). In 8 patients initially treated on an out-patient basis hospitalization due to fever/infection was necessary with a median duration of 19 days (range, 3-51days). One patient received a single platelet transfusion during 94 consolidation cycles. Total costs for all 4 consolidation cycles including in- and out-patient treatment were calculated with 56,305€.

During the 3 consolidation cycles in the AIDA-arm, 36 of 67 cycles were administered on an in-patient basis (median duration of hospital stay, 9 days; range, 1-30 days). In 12 patients initially treated on an out-patient basis a total of 14 hospitalizations due to fever/infection were necessary (median duration, 13 days; range, 6-31days). In 10 patients, 11 platelet transfusions were given during 67 consolidation cycles. Combining the in-patient and out-patient treatment cost of all 3 consolidation cycles including rare complications of fever and platelet transfusion, total costs were calculated with 17,159€. In addition, a maintenance therapy of 2 years with 24 cycles was intended; n=15 patients received a median of 10 cycles (range 7-14 cycles). The preterm cessation of maintenance therapy in all patients was due to cytopenias and drug intolerance. Therefore, costs for 10 cycles were estimated as average costs of maintenance therapy with a total amount of 4,264€.

The higher total costs in the ATO/ATRA-arm of 85,516€ (49,402€ attributed to ATO) per patient as compared to the AIDA-arm of 36,895€ per patient were accompanied by a significantly better event-free and overall survival reported in the original report (Lo Coco F et al. N Engl J Med. 2013;369(2):111-21) at 2 years of 97% and 99% compared to 86% and 91%, respectively. Based on the published overall survival data with ATO/ATRA we calculated numbers-needed-to-treat to save one life based on follow-up data of 2 years. By using the average risk difference approach a number of 14 patients has been calculated (95%-confidence interval, 12-17).

Conclusions: Treatment with ATO/ATRA in newly diagnosed APL with WBC<10G/l was associated with 2.3-time higher costs. At least 14 patients have to be treated with ATO to save one additional life.