Pharmacologic regimens for knee osteoarthritis prevention: can they be cost-effective?

Abstract

Objective: We sought to determine the target populations and drug efficacy, toxicity, cost, and initiation age thresholds under which a pharmacologic regimen for knee osteoarthritis (OA) prevention could be cost-effective.

Design: We used the Osteoarthritis Policy (OAPol) Model, a validated state-transition simulation model of knee OA, to evaluate the cost-effectiveness of using disease-modifying OA drugs (DMOADs) as prophylaxis for the disease. We assessed four cohorts at varying risk for developing OA: (1) no risk factors, (2) obese, (3) history of knee injury, and (4) high-risk (obese with history of knee injury). The base case DMOAD was initiated at age 50 with 40% efficacy in the first year, 5% failure per subsequent year, 0.22% major toxicity, and annual cost of $1,000. Outcomes included costs, quality-adjusted life expectancy (QALE), and incremental cost-effectiveness ratios (ICERs). Key parameters were varied in sensitivity analyses.

Results: For the high-risk cohort, base case prophylaxis increased quality-adjusted life-years (QALYs) by 0.04 and lifetime costs by $4,600, and produced an ICER of $118,000 per QALY gained. ICERs >$150,000/QALY were observed when comparing the base case DMOAD to the standard of care in the knee injury only cohort; for the obese only and no risk factors cohorts, the base case DMOAD was less cost-effective than the standard of care. Regimens priced at $3,000 per year and higher demonstrated ICERs above cost-effectiveness thresholds consistent with current US standards.

Conclusions: The cost-effectiveness of DMOADs for OA prevention for persons at high risk for incident OA may be comparable to other accepted preventive therapies.

Keywords: Cost-effectiveness; Disease-modifying osteoarthritis drugs; Knee osteoarthritis; Prophylaxis.

Figure 1. Sequence of OA Prevention, Incidence, and Treatment

The figure shows the sequence of health states for the prevention, incidence, and subsequent treatment of OA. Subjects who had no diagnosed radiographic evidence of OA either received a DMOAD regimen for the prevention of knee OA or received no prophylaxis. Subjects may have gone on to develop OA, at which point they underwent the standard of care for OA treatment and ceased any prophylactic regimen. Subjects then continued through the model until this guideline-concordant care culminated in the receipt of a total knee replacement.

Figure 2. DMOADs as Prevention in the OAPol Model

This figure depicts the path of a hypothetical subject in the OAPol Model receiving DMOADs before the incidence of OA. If OA developed during or following the discontinuation of a prophylactic DMOAD regimen, subjects continued on with guideline-concordant OA treatment.

Figure 3. Incremental Cost-Effectiveness Ratios of DMOADs as Prevention: ‘Obese with knee injury’ Cohort; Prophylaxis with 0.22% Major Toxicity

The ‘Cost’ column headers apply to every row below them. The ‘Start Age’ row headers on the left of the figure apply to all columns in that row. This figure applies to DMOADs with major toxicity of 0.22%. ICERs improved with increasing efficacy and decreasing toxicity, cost, and late failure. ICERs also improved with increasing start age from initiation at age 30 up to initiation at age 50; ICERs were less favorable at a start age of 60, likely because this was after average age of knee OA onset had been reached. The ICER estimated for the base case DMOAD regimen is highlighted.

Figure 4. Incremental Cost-Effectiveness Ratios of DMOADs as Prevention: ‘Knee Injury Only’ Cohort; Prophylaxis with 0.22% Major Toxicity

Like Figure 3, this figure illustrates the ICERs estimated for a DMOAD regimen at 0.22% major toxicity for the ‘knee injury only’ cohort. ICERs were worse for this cohort than for the ‘obese with knee injury’ cohort. Prophylaxis was unlikely to have ICERs of less than $150,000 when the annual cost of the drug was $2,000 or greater. The ICER estimated for the base case DMOAD regimen is highlighted.