CADTH Common Drug Review Results

CDR did not consider the manufacturer’s weighted average “blended” incremental costs to be valid:

• Eslicarbazepine was only recently introduced onto public formularies in Canada; thus, its market share is likely still growing. Data retrieved by CDR using QuintilesIMS indicated that the Ontario public market share for eslicarbazepine relative to lacosamide and perampanel was between 1.8% and 2.5% in the first quarter of 2016, depending on the method of calculation; this is at least ▬ times the ▬ share estimated by the manufacturer based on QuintilesIMS Rx Dynamics data from February 2015 through January 2016.
• Comparing the cost of brivaracetam to a blended average of multiple comparators weighted by market share, even if the market share is accurate and stable, assumes that brivaracetam will displace these comparators in that same ratio. Given the differences in reported adverse-event profiles, regulatory warnings, and the suitability of each comparator for patients with differing treatment histories (e.g., patients on carbamazepine would be less likely to have eslicarbazepine added, while patients on levetiracetam would be less likely to have brivaracetam added), this seems unlikely.

CDR reviewers noted the manufacturer’s assumption of a dispensing fee every 30 days may be an overestimation, as some patients would be older than 65 years of age, and thus would qualify to receive up to 100 days of medication at a time. However, the impact of this assumption is unlikely to be large, given the relatively young age of patients with epilepsy, the likelihood of dose adjustments, and the absence of the main comparators on the ODB list of chronic medications (which limits the number of dispensing fees for included medications to five annually).¹³ CDR removed dispensing fees and markups (given the variability between jurisdictions) from further analyses.

Using the manufacturer’s assumed dose distribution, but removing markups and dispensing fees, yields a dose-weighted average annual cost of $3,154 per patient for brivaracetam, which is $254 less than that of lacosamide ($3,408 per patient), $296 less than that of perampanel ($3,449 per patient), and $336 less than that of eslicarbazepine ($3,489 per patient) (Table 9).

Table 9

CDR’s Estimated Weighted Average Cost of Brivaracetam Versus Comparators Using Manufacturer’s Dose Distribution (Dispensing Fees and Markup Removed).

The manufacturer’s base-case analysis uses a uniform dose distribution across product monograph–recommended maintenance doses; however, this is not supported by data. Additionally, by only including the higher doses, incremental cost calculations are biased in favour of flat-rate priced comparators relative to those with graduated pricing. In reality, the lower doses of flat-rate priced comparators can be expected to replace at least some of the lower-dose and lower-cost utilization of lacosamide and, thus, these costs should have been considered. In fact, 35% of perampanel claims and 37% of eslicarbazepine claims reimbursed by Ontario in the first quarter of 2016 were for the lowest strengths (2 mg and 200 mg tablets respectively), while 31% of claims were for 50 mg lacosamide tablets, supporting the necessity of considering lower doses in cost calculations. These percentages remain similar when calculated by the proportion of individual patients on each strength within each month of Ontario public claims data (January, February, and March 2016). CDR conducted a reanalysis using QuintilesIMS public claims data for January through March 2016 to estimate the proportion of patients using each daily dose of lacosamide, and assuming cost-efficient dispensing. Under these assumptions, the annual cost of brivaracetam ($3,154 per patient) is $248 more than that of lacosamide ($2,905 per patient) (Table 10).

Table 10

CDR’s Reanalysis of Incremental Cost of Brivaracetam Relative to Lacosamide Based on Utilization-Based Dose Distribution for Lacosamide.

Interestingly, the manufacturer also did not consider the 1,200 mg dose when calculating the average annual cost of eslicarbazepine, a dose included in the recommended range of the product monograph⁶ and estimated to be used by 20.7% of patients in the base case of the CDR review of Aptiom.¹⁴ Using this dose-distribution estimate, and assuming that patients taking 1,200 mg are using one and one-half 800 mg tablets, the weighted average annual cost of eslicarbazepine increases to $3,851 per patient, yielding an incremental savings with brivaracetam compared with eslicarbazepine of $697 per patient per year.

The high number of claims for lower strengths appears at least partially driven by cost-inefficient dispensing (i.e., use of multiple low-strength tablets to achieve a daily dose that could be achieved in a less costly manner by using higher strengths). For all three comparators where utilization data are available, the lowest strengths are used twice as often per day as would be expected if dispensing patterns were perfectly cost-efficient. An exploratory analysis of the potential ramifications of this can be found in Table 11, using QuintilesIMS public plan data from January 2016; the use of February or March 2016 data yields similar results.

Table 11

CDR’s Exploratory Analysis on the Cost Impact of the Use of Multiple Low-Dose Tablets to Achieve Target Dosing.

While inefficient from a cost perspective, the apparent overuse of some strengths may be related to the recommendations for initiating therapy within individual product monographs. The lacosamide monograph⁴ recommends an initial dose of 50 mg twice daily, increasing by 50 mg twice daily every week up to 200 mg twice daily depending on individual efficacy and tolerability; the average use of 3.93 units/day rather than 2 per day for the 50 mg strength may reflect continued use of this strength into the maintenance phase when dispensing 100 mg tablets would have been more cost-efficient. Similarly, the initial dose of perampanel is 4 mg daily, increasing in 2 mg increments based on response and tolerance, leading to increased use of both the 2 mg and 4 mg tablets (2.35 and 1.29 tablets per day respectively, rather than 1 per day). Eslicarbazepine is more complicated, given that the recommended starting dose is 400 mg daily, increasing up to 800 mg based on response and tolerance, with a maximum recommended dose of 1,200 mg per day for some patients.⁶ While the monograph suggests the 1,200 mg daily dose be dispensed as one and one-half 800 mg tablets, it is likely that some patients are receiving two 600 mg tablets daily as a maintenance dose, increasing the use of that tablet strength, along with use of the 200 mg and 400 mg strengths.

Of note, the cost-inefficient use of lower-strength tablets disproportionately increases the weighted average daily costs of flat-priced comparators relative to lacosamide (weighted average cost of $9.86 per patient per day versus $14.18 and $16.69 for perampanel and eslicarbazepine, respectively). There were insufficient utilization data available at the time of this review to perform a similar analysis based on brivaracetam utilization. However, given that the initial dose recommended in the monograph is 50 mg twice daily, adjusted to 25 mg to 100 mg twice daily based on response and tolerability,¹ it is likely that the 25 mg and 50 mg strengths of brivaracetam will be inefficiently utilized from a cost perspective to a similar extent as the lower strengths of its comparators, which would increase the incremental cost of brivaracetam relative to lacosamide.

In short, depending on the method of analysis, the daily cost of lacosamide ranged from $7.96 to $9.86 per patient, while the submitted and lowest possible daily cost of brivaracetam was $8.64 per patient. The real-world daily cost of brivaracetam is likely to be higher than this due to cost-inefficient dispensing patterns observed for its comparators.

The patents for lacosamide (Vimpat) are due to expire in early 2017, which may lead to the availability of less expensive generic versions in the near future,⁷ although lacosamide’s inclusion on the Register for Innovative Drugs in effect grants the manufacturer of Vimpat market exclusivity until September 2018.^8,9 CDR conducted analyses to explore the incremental cost of brivaracetam versus its comparators under various price-reduction scenarios in Appendix 1.

A published indirect treatment comparison of levetiracetam with brivaracetam, analogues with similar mechanisms of action, found few statistically significant differences between these drugs in patients with refractory focal seizures (refractory POS).³ Although similar to the results of the manufacturer’s NMA, these findings are not necessarily indicative of clinical equivalence or non-inferiority. Due to the need for a varying number of tablets per day to achieve levetiracetam doses within the range recommended in the product monograph (1,000 mg to 3,000 mg in two divided doses), it is not possible to accurately estimate a dose-weighted, daily-average cost using QuintilesIMS data by tablet strength, as was done for lacosamide, perampanel, and eslicarbazepine. The annual cost of levetiracetam ranges from $397 to $1,098 per patient.