BACKGROUND: Surveillance for rare cancers in the postmarketing setting is challenging, especially when the treatment is also rare.
OBJECTIVE: To compare characteristics among four postauthorization safety studies designed to monitor for a potential association between teriparatide and osteosarcoma in adults.
METHODS: We evaluated characteristics of four studies (Study 1 completed in five Nordic countries; Studies 2-4 ongoing in the United States) implemented in the surveillance program including anticipated study duration, exposure ascertainment method, availability of comparator group, and the detectable level of increased risk.
RESULTS: Study 1 used a case series design, ascertained exposure through chart abstraction over a 10-year study period, with the ability to detect a 12-fold increase in risk, if one exposed case had been observed. Study 2 used a case series design, ascertained exposure through telephone interview over a 15-year study period, and was designed to detect a 3- fold increased risk above the background rate. Study 3 used a prospective cohort “registry” design, obtained exposure by patient report, covered a 15-year study period, and is projected to be able to detect a 4-fold increase in risk. Studies 1-3 used an external population comparator. Study 4 used a matched cohort design with exposure identified through prescription claims in the Medicare database, a 3-year study conduct period (using data covering 8 years), and was designed to detect a 4- to 5-fold increase in risk. All studies utilized cancer registries to identify patients diagnosed with osteosarcoma.
CONCLUSION: The database study (study 4) required less time, fewer resources for exposure ascertainment, had similar risk detection capabilities, and a more robust analysis (owing to a matched comparator) than the other studies. The other studies provided surveillance beyond spontaneous reporting to detect a very large increased risk of osteosarcoma, had it existed, early in the product’s lifecycle.
