The sodium-glucose cotransporter 2 inhibitors (SGLT2i) empagliflozin, canagliflozin, and dapagliflozin reduce the risk of heart failure (HF) events in patients with diabetes mellitus (DM) at high risk for HF. Differences in HF outcomes between SGLT2i were demonstrated in a recent-published meta-analysis. Nevertheless, comparative cost-effectiveness analyses of SGLT2i provided for this indication have not been published yet. Therefore, we aimed to provide a preceding economic comparison of the costs required for improving HF outcomes by these three SGLT2i. The primary outcome was the cost needed to treat (CNT) to prevent one event of hospitalization for HF or cardiovascular mortality. CNT is calculated by multiplying the annualized number needed to treat to prevent one event by the annual cost of therapy. Clinical outcome data were extracted from pre-specified cohorts of HF-naïve patients in the pivotal randomized controlled trials (RCTs). Costs of interventions were estimated as 75% of the US National Average Drug Acquisition Cost listing. Sensitivity analysis was performed to mitigate differences between the RCT’s populations. We figured the CNT for the primary prevention of HF events in DM patients to be $542,328 ($409,044-$905,412) for empagliflozin, $2,347,488 ($1,066,208-∞) for canagliflozin and $2,128,374 ($1,204,740-$48,140,518) for dapagliflozin. Sensitivity analysis confirmed the cost benefit of empagliflozin. Our findings suggest that between the available SGLT2i, the cost of primary prevention of HF in patients with DM at high risk for HF is lowest with empagliflozin. These findings may help choose an SGLT2i until head-to-head RCTs, and comprehensive cost-effective analyses for this indication are available.
Empagliflozin, canagliflozin, and dapagliflozin, three sodium-glucose cotransporter 2 inhibitors (SGLT2i), reduce the risk of hospitalization for heart failure (hHF) or cardiovascular (CV) related mortality in patients with diabetes mellitus (DM), in patients at high risk for HF. Accordingly, the cardiology, heart failure, and diabetes associations in the US , and Europe , have expanded their recommendations to include SGLT2i for primary prevention for patients at risk for HF. Although all published guidelines refer to SGLT2i for HF prevention as a class, a recent meta-analysis has demonstrated differences in the HF outcomes between the three SGLT2i. The prevalence of patients with DM that may require primary prevention for HF exceeds 10,000,000 only in the US, thus requiring multi-billion-dollar investments. Head-to-head RCTs and cost-effectiveness analyses of SGLT2i for primary prevention of HF have not been published yet. Therefore, we aimed to provide a preceding economic measure of the costs of the different SGLT2i required to improve this population’s outcomes.
Methods
Data on outcomes of SGLT2i therapy in DM patients without established HF at baseline were extracted from the published RCTs: EMPA-REG OUTCOME for empagliflozin, CANVAS PROGRAM for canagliflozin, and the DECLARE-TIMI 58 trial for dapagliflozin.
The primary outcome was the cost needed to treat (CNT) to prevent an event of hHF or CV mortality. The CNT was calculated as the product of the annualized number of patients needed to treat (NNT) to prevent one event by the therapy’s annual cost.
The annualized NNT was figured by utilizing the annualized absolute risk reductions (ARR) of each intervention’s outcome events. , In cases that the therapy had no benefit compared to the placebo, the NNT was set to be infinity.
Drug prices were extracted in June 2020 from the US National Average Drug Acquisition Cost (NADAC) website. We adopted the method suggested by Levy et al. in assessing drug costs for cost-effectiveness analysis in the US, using 75% of the NADAC price as the base price for cost calculations.
The sensitivity analysis simulated each drug’s effect while using the other two drug’s control arm’s event rates to mitigate the differences in the risk of an HF event in the control arms. Regarding the cost of therapy, we used the full NADAC price as an upper bound and 50% of NADAC price as the lower bound as recommended for use in US cost-effectiveness analyses.
Results
The key patient characteristics and demographics of the patients in the RCTs are compared in Table 1 . The critical difference between the RCTs was the rate of patients with prior CVD. In EMPA-REG OUTCOME, the rate was 99%, compared to 63% in CANVAS PROGRAM and only 37% in DECLARE-TIMI 58. However, the patients’ rates without prior HF were comparable: 80%-90% of the patients in all three trials.
| Variable/ trial | EMPA-reg outcome n= 7,020 * | Canvas-program n = 8,681 | Declare TIMI-58 n = 15,436 |
|---|---|---|---|
| Intervention | Empagliflozin 10 or 25 mg | Canagliflozin 100 or 300 mg | Dapagliflozin 10 mg |
| Follow up years (mean) | 3.1 | 2.4 | 4.2 |
| Mean age (years) | 63 | 63 | 64 |
| Women | 28% | 34% | 38% |
| Mean BMI (kg/m 2 ) | 31 | 32 | 31 |
| EGFR< 60 ml/min/1.73 m2 | 26% | 21% | NA |
| Prior cardiovascular disease | 99% | 63% | 37% |
⁎ Data for EMPA-REG OUTCOME includes all the trial’s patients, including 10% that had prior HF.
Table 2 details the step-by-step calculations of the annualized NNT and CNT. The annualized NNT of empagliflozin is the lowest (118 versus 513 for canagliflozin and 477 for dapagliflozin). The cost of the three therapies is comparable in the US. Therefore, the CNT of empagliflozin is the lowest.
| Parameter | EMPA-reg (Empagliflozin) | Canvas program (Canagliflozin) | Declare TIMI 58 (Dapagliflozin) |
|---|---|---|---|
| Number of patients in control arm | 2,089 | 3,689 | 7,706 |
| Follow up | 3.1 | 2.4 | 4.2 |
| Patient years of therapy in control arm | 2,089*3.1=6,476 | 3,689*2.4=8,854 | 7,706*4.2=32,365 |
| Number of events in the control arm | 149 | 134 | 324 |
| Annualized Event Rate in the control arm | 149/6,476=2.30% | 134/8,854=1.51% | 324/32,365=1.00% |
| Number of patients in intervention arm | 4,225 | 4,992 | 7,730 |
| Patient years of therapy in intervention arm | 4,225*3.1=13,098 | 4,992*2.4=11,981 | 7,730*4.2=32,466 |
| Number of events in the intervention arm (95% CI) | 190 (154-235) | 158 (130-192) | 257 (205-322) |
| Annualized Event Rate in the intervention arm (95% CI) | 190/13,098=1.45% (1.17%-1.79%) | 163/11,981=1.31% (1.08%-1.60%) | 275/32,466=0.79% (0.63%-0.99%) |
| Annualized Absolute Event Rate Reduction (95% CI) | 2.30%-1.45%=0.085% (0.51%-1.12%) | 1.51%-1.31%=0.19% (-0.09%-0.43%) | 1.00%-0.79%=0.21% (0.01%-0.70%) |
| Annualized Number Needed to Treat (95% CI) | 1/0.0085=118 (89-197) | 1/0.0019=513 (233-∞) | 1/0.0021=477 (270-10,789) |
| Annual drug cost (US) | $4,596 | $4,576 | $4,462 |
| Cost Needed to Treat to prevent one event (95% CI) | 118*$4,596= $542,328 ($409,044-$905,412) | 513*$4,576= $2,347,488 ($1,066,208-∞) | 477*$4,462= $2,128,374 ($1,204,740-$48,140,518) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
