Purpose: Osteoporosis is a common and costly disease. Parathyroid hormone [1-34] (PTH) is a promising new agent for the treatment of osteoporosis although its cost-effectiveness is unknown. The purpose of our study was to evaluate the cost-effectiveness of PTH-based treatment strategies compared to the current market leader, alendronate, for the first-line treatment of high-risk postmenopausal osteoporosis.

Methods: We developed a microsimulation incorporating fracture type and individual risk to compare placebo vitamin D and calcium sufficiency to three treatment strategies: life-long alendronate (ALN), two years of PTH (PTH-Alone), and two years of PTH followed by life-long alendronate (PTH->ALN). We used studies identified in a search of MEDLINE and U.S. data sources whenever possible. Long-term anti-fracture efficacy for PTH and alendronate was extrapolated from available data. Our reference population was treatment-naïve 70-year old osteoporotic Caucasian women with pre-existing vertebral fracture. We employed a lifetime horizon and societal perspective. Our primary outcome was cost per quality-adjusted life-year (QALY) gained. We discounted costs and QALYs at a 3% annual rate, and present data in 2003 U.S. dollars.

Results: In our reference case, ALN and PTH->ALN appeared on the cost-effectiveness frontier, while PTH-Alone was dominated even if its anti-fracture efficacy was assumed to persist for 20 years after cessation of PTH. The incremental cost-effectiveness ratio (ICER) was $42,000/QALY for ALN compared to placebo and $92,000/QALY for PTH->ALN compared to ALN. Compared to placebo, the ICER of PTH-Alone was $275,000/QALY. In scenario analysis, the ICER of PTH->ALN was consistently higher than the ICER of ALN; under reference case assumptions for ALN, the ICER of PTH->ALN would be higher than that of ALN even if PTH->ALN could eliminate fractures. Our results were sensitive to the cost of medication. If the cost of PTH was reduced by 50%, the ICER of PTH->ALN would be lower than that of ALN; if the cost of alendronate was reduced by 75%, ALN would be cost-saving.

Conclusions: The cost-effectiveness of PTH-Alone is inferior to ALN and PTH->ALN. ALN appears cost-effective at a threshold of $50,000/QALY, while PTH->ALN may be cost-effective at a threshold of $100,000/QALY. Significant reductions in the cost of PTH could substantially improve the cost-effectiveness of the PTH->ALN treatment strategy.