Denosumab Safety Assessment in Multiple Observational Databases

This is a prospective open-cohort study with annual assessment and reporting of descriptive findings from 5 secondary data sources, US Medicare,Optum Research Database (formerly United HealthCare), Scandinavian national health registry databases, including data from Denmark, Sweden, and Norway. The study period will include up to 10 years in each data system depending on data availability. Descriptive statistics will be used to characterize exposure cohorts with respect to patient characteristics, clinical features, and AESI risk factors. Person-year adjusted AESI incidence rates will be calculated among postmenopausal women, postmenopausal women with osteoporosis, and exposure cohorts. Exploratory analyses comparing incidence rates of AESI in women with PMO adjusting for potential confounders will also be conducted. Descriptive statistics will be used to characterize denosumab utilisation patterns in patients who receive denosumab for PMO and in subjects who receive Prolia therapy for unapproved indications. Subsequent substudies using US data systems were added to describe men with osteoporosis treated with Prolia and men and women who receive Prolia with glucocorticoid exposure.

Criteria: Inclusion Criteria for women with postmenopausal osteoporosis:

• Postmenopausal women: Postmenopausal status will be determined based on age and defined as women ≥ 55 years old. For the Medicare database, only women ≥ 65 years old will be included in the analysis, given that generally all individuals in the US ≥ 65 years old are eligible for Medicare coverage and data on postmenopausal women less than 65 years old will be available for only a small number of women meeting other specialized eligibility criteria.
• Women with Post Menopausal Osteoporosis: The presence of Post Menopausal Osteoporosis will be determined utilizing an algorithm based upon definition of postmenopausal women (≥ 65 years old in Medicare or ≥ 55 years old in other data systems), diagnostic codes indicating osteoporosis, diagnostic codes indicating osteoporotic fracture, and/or relevant postmenopausal osteoporosis treatment codes.
• The inclusion of men and women who have Glucocorticoid Induced Osteoporosis (GIOP) has also been added to the study population. Men and women who receive Prolia with glucocorticoid exposure are included in the ongoing study 20090522 to assess safety in the post-marketing environment

Exclusion criteria

for women with postmenopausal osteoporosis:

• Women with postmenopausal osteoporosis: Women with Paget'sdisease during the 12-month period prior to meeting criteria for inclusion in postmenopausal osteoporosis population will be excluded. Additionally, in US Medicare and United Healthcare, women with a diagnosis of malignancy (excluding non-melanoma skin cancer) or treatment with chemotherapy, hormonal therapy or radiation therapy for cancer up to 12-months before index date will be excluded. In the Scandinavian national registries, women with a diagnosis of cancer according to the patient registry and/or cancer registry up to 12-months prior to meeting criteria for inclusion in postmenopausal osteoporosis population will be excluded.

Inclusion criteria

for men with osteoporosis:

• Men ≥ 65 years old in the Medicare database or ≥ 30 years old in andthe Optum Research Database (formerly United HealthCare)United Healthcare who receive at least one denosumab 60 mg injection will be included. All men need to be continuously enrolled for at least 12 months prior to start of follow-up. A diagnosis code of osteoporosis or osteoporotic fracture would not be required because it is known that male osteoporosis is under-diagnosed. The exclusion of patients with cancer or Paget disease (as described in the Exclusion Criteria Section) will ensure that patients treated with Prolia is for osteoporosis but not for other indications. The index date is defined as the date when a patient first satisfies all inclusion criteria.

Exclusion criteria

for men with osteporosis:

• Men with a diagnosis of malignancy (excluding non-melanoma skin cancer) or treatment with chemotherapy, hormonal therapy or radiation therapy for cancer up to 12 months before the index date will be excluded.
• Men with Paget's disease during the 12-month period prior to the index date will also be excluded.

Inclusion criteria

for Men and Women Who Receive Prolia With Glucocorticoid Exd Exposure • Men and women identified from Medicare or Optum Research Database who are ≥65 years old (or ≥30 years old in Optum) on or before receiving at least 1 denosumab 60 mg injection or fill after approval date of Prolia for treatment of glucocorticoid-induced osteoporosis (18 May 2018) will be included. This includes both new and prevalent users of denosumab. The index date is defined as the date of the first denosumab administration on or after the approval date fulfilling both the database specific age, 12-month enrollment criteria, and has received a 90-day average daily dose equivalent ≥7.5 mg of prednisone at any time during the previous 12 months.

Exclusion criteria

for Men and Women Who Receive Prolia With Glucocorticoid Exposure

• Men and women with a diagnosis of malignancy (excluding non-melanoma skin cancer) or treatment with chemotherapy, hormonal therapy or radiation therapy for cancer up to 12 months before the index date will be excluded. Men and women with Paget's disease during the 12-month period prior to the index date will also be excluded. The exclusion of subjects with cancer or Paget's disease will ensure that subjects included in the substudy are receiving denosumab for the treatment of osteoporosis, and not for other indications.