Increase transparency and reproducibility of real-world evidence in rare diseases through disease-specific Federated Data Networks.

PURPOSE: In rare diseases, real-world evidence (RWE) generation is often restricted due to small patient numbers and global geographic distribution. A federated data network (FDN) approach brings together multiple data sources harmonized for collaboration to increase the power of observational research. In this paper, we review how to increase reproducibility and transparency of RWE studies in rare diseases through disease-specific FDNs. METHOD: To be successful, a multiple stakeholder scientific FDN collaboration requires a strong governance model in place. In such a model, each database owner remains in full control regarding the use of and access to patient-level data and is responsible for data privacy, ethical, and legal compliance. Provided that all this is well documented and good database descriptions are in place, such a governance model results in increased transparency, while reproducibility is achieved through data curation and harmonization, and distributed analytical methods. RESULTS: Leveraging the OHDSI community set of methods and tools, two rare disease-specific FDNs are discussed in more detail. For multiple myeloma, HONEUR-the Haematology Outcomes Network in Europe-has built a strong community among the data partners dedicated to scientific exchange and research. To advance scientific knowledge in pulmonary hypertension (PH) an FDN, called PHederation, was established to form a partnership of research institutions with PH databases coming from diverse origins.

Haematology Outcomes Network in Europe (HONEUR)-A collaborative, interdisciplinary platform to harness the potential of real-world data in hematology.

INTRODUCTION: There remains a need to optimize treatments and improve outcomes among patients with hematologic malignancies. The timely synthesis and analysis of real-world data could play a key role. OBJECTIVES: The Haematology Outcomes Network in Europe (HONEUR) is a federated data network (FDN) that aims to overcome the challenges of heterogenous data collected from different registries, hospitals, and other databases in different countries. It has the functionality required to analyze data from various sources in a time efficient manner, while preserving local data security and governance. With this, research studies can be performed that can increase knowledge and understanding of the management of patients with hematologic malignancies. METHODS: HONEUR uses the Observational Medical Outcomes Partnership (OMOP) common data model, which allows analysis scripts to be run by multiple sites using their own data, ultimately generating aggregated results. Furthermore, distributed analytics can be used to run statistical analyses across multiple sites, as if data were pooled. The external governance model ensures high-quality standards, while data ownership is retained locally. Twenty partners from nine countries are now participating, with data from more than 26 000 patients available for analysis. Research questions that can be addressed through HONEUR include assessments of natural disease history, treatment patterns, and clinical effectiveness. CONCLUSIONS: The HONEUR FDN marks an important step forward in increasing the value of information routinely captured by individual hospitals, registries and other database holders, thus enabling larger-scale studies to be undertaken rapidly and efficiently.

The Value of Federated Data Networks in Oncology: What Research Questions Do They Answer? Outcomes From a Systematic Literature Review.

OBJECTIVES: Real-world evidence (RWE) plays an important role in addressing key research questions of interest to healthcare decision makers. Federated data networks (FDNs) apply novel technology to enable the conduct of RWE studies with multiple partners, without the need to share the individual partner's data set. A systematic review of the published literature was performed to determine which types of research questions can best be addressed through FDNs, specifically in the field of oncology. METHODS: Systematic searches of MEDLINE and Embase were undertaken to identify the types of research questions that had been addressed in studies using FDNs. Additional information was retrieved about study characteristics, statistical methods, and the FDN itself. RESULTS: In total, 40 publications were included where research questions on the following had been addressed (multiple categories possible): disease natural history (58%), safety surveillance (18%), treatment pathways (15%), comparative effectiveness (10%), and cost/resource use studies (3%)-13% of studies had to be left uncategorized. A total of 50% of the studies were run with data partners in networks of ≤5. The size of the networks ranged from 227 patients to >5 million patients. Statistical methods used included distributed learning and distributed regression methods. CONCLUSIONS: Further work is needed to raise awareness of the important role that FDNs can play in leveraging readily available RWE to address key research questions of interest in cancer and the benefits to the research community in engaging in federated data initiatives with a long-term perspective.

Cardiovascular and general safety of a 24-day regimen of drospirenone-containing combined oral contraceptives: final results from the International Active Surveillance Study of Women Taking Oral Contraceptives.

OBJECTIVES: The "International Active Surveillance Study of Women Taking Oral Contraceptives" investigated the risks of short- and long-term use of an extended 24-day regimen of drospirenone and ethinylestradiol (DRSP24d) compared to established oral contraceptives (OCs) in a routine clinical setting. STUDY DESIGN: Prospective, controlled, noninterventional cohort study conducted in the United States and six European countries with three main cohorts: new users of DRSP24d, DRSP21d (21-day regimens of DRSP-containing OCs), and non-DRSP (OCs without DRSP). All self-reported clinical outcomes of interest (OoI) were validated via attending physicians and relevant source documents. Main OoI were serious clinical outcomes, in particular venous thromboembolism (VTE). Comprehensive follow-up procedures were implemented. Statistical analyses were based on Cox regression models. Primary statistical variable was the VTE hazard ratio (HR) for DRSP24d vs. non-DRSP. RESULTS: A total of 2285 study centers enrolled 85,109 women. Study participants were followed for 2 to 6 years, which generated 206,296 woman-years (WY) of observation. A low loss to follow-up of 3.3% was achieved. DRSP24d, DRSP21d, non-DRSP and levonorgestrel-containing OCs (LNG) showed similar incidence rates of venous and arterial thromboembolism, fatal outcomes, cancer, severe depression and other serious adverse events. VTE incidence rates for DRSP24d, DRSP21d, non-DRSP and LNG were 7.2, 9.4, 9.6 and 9.8 VTE/10,000 WY, respectively. Adjusted HRs for DRSP24d vs. non-DRSP and DRSP24d vs. LNG were 0.8 [95% confidence interval (CI), 0.5-1.3] and 0.8 (95% CI, 0.4-1.5). CONCLUSION: DRSP24d, DRSP21d, non-DRSP and LNG use was associated with similar risks of serious adverse events, and particularly VTE, during routine clinical use. IMPLICATION STATEMENT: The 24-day regimen of drospirenone-containing combined OCs is associated with similar risks of venous and arterial thromboembolism, fatal outcomes, cancer, severe depression and other serious adverse events compared to 21-day regimens of drospirenone-containing combined OCs, OCs without drospirenone and LNGs.

Levonorgestrel-releasing and copper intrauterine devices and the risk of breast cancer.

BACKGROUND: This study compares the risk of breast cancer for levonorgestrel-releasing intrauterine devices (LNG(IUD)) versus copper IUDs (CU(IUD)) in women younger than 50 years of age. STUDY DESIGN: Retrospective, population-based, case-control study using cancer registers in Finland and Germany, powered to exclude a 1.5-fold risk of breast cancer. RESULTS: Analysis of 5113 breast cancer cases diagnosed 2000-2007 and 20,452 controls - matched by year of birth and area of residence - yielded relative risk estimates approaching unity with 95% CI crossing 1.0 for all comparisons, including ever-use of LNG(IUD) versus CU(IUD) (adjusted OR, 0.99; 95% CI, 0.88-1.12) and current use at time of diagnosis (adjusted OR, 0.85; 95% CI, 0.52-1.39), as well as for sub-analyses by country, age, tumor characteristics and period, recency and duration of use prior to diagnosis. CONCLUSION: This study does not indicate an increased risk of breast cancer for users of LNG(IUD). No indications for tumor promotion or tumor induction were found.

Effectiveness of oral contraceptive pills in a large U.S. cohort comparing progestogen and regimen.

OBJECTIVE: To estimate real-life effectiveness of oral contraceptive pills by progestogen, length of pill-free interval, and body mass index while focusing on the effect of progestogens with a long half-life and on 24-day oral contraceptive pills regimens. METHODS: Outcome data from 52,218 U.S. participants in the International Active Surveillance of Women Taking Oral Contraceptives­a large, prospective, controlled, noninterventional, long-term cohort study with active surveillance of the study participants­were used to analyze contraceptive failure in association with oral contraceptive pills use. Low loss to follow-up is ensured by a comprehensive follow-up procedure. Contraceptive failure rates are described by Pearl Index and life-table analysis. Inferential statistics for contraceptive failure are based on Cox regression models. RESULTS: Analyses are based on 1,634 unintended pregnancies during 73,269 woman-years of oral contraceptive pills exposure. Life-table estimates of contraceptive failure for a 24-day regimen of drospirenone and ethinyl estradiol and 21-day regimens of other progestogens were 2.1% and 3.5% after the first study year, and 4.7% and 6.7% after the third year. The adjusted hazard ratio was 0.7 (95% confidence interval 0.6­0.8). Direct comparisons of the 24-day and 21-day regimens of drospirenone and norethisterone, respectively, showed also lower contraceptive failure rates for 24-day regimens. Contraceptive failure rates adjusted for age, parity and educational level showed a slight increase with higher body mass index. CONCLUSION: The 24-day oral contraceptive regimens containing a progestogen with a long half-life show higher contraceptive effectiveness under routine medical conditions compared with conventional 21-day regimens. Obesity seems to be associated with a slight reduction of contraceptive effectiveness. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov, www.clinicaltrials.gov, NCT00335257. LEVEL OF EVIDENCE: II

International Active Surveillance Study of Women Taking Oral Contraceptives (INAS-OC Study).

BACKGROUND: A 24-day regimen of contraceptive doses of drospirenone and ethinylestradiol (DRSP/EE 24d) was recently launched. This regimen has properties which may be beneficial for certain user populations (e.g., women suffering from premenstrual dysphoric disorder or acne). However, it is unknown whether this extended regimen has an impact on the cardiovascular risk associated with the use of oral contraceptives (OCs). The INternational Active Surveillance study of women taking Oral Contraceptives (INAS-OC) is designed to investigate the short- and long-term safety of the new regimen in a population which is representative for the typical user of oral contraceptives. METHODS/DESIGN: A large, prospective, controlled, non-interventional, long-term cohort study with active surveillance of the study participants has been chosen to ensure reliable and valid results. More than 2,000 gynecologists in the US and 5 European countries (Austria, Germany, Italy, Poland, and Sweden) will recruit more than 80,000 OC users. The two to five year follow-up of these women will result in at least 220,000 documented women-years. The main clinical outcomes of interest for the follow-up are deep venous thrombosis, pulmonary embolism, acute myocardial infarction and cerebrovascular accidents. Secondary objectives are general safety, effectiveness and drug utilization pattern of DRSP/EE 24d, return to fertility after stop of OC use, as well as the baseline risk for users of individual OC formulations. Because of the non-interference character of this study, potential participants (first-time users or switchers) are informed about the study only after the decision regarding prescription of a new OC. There are no specific medical inclusion or exclusion criteria. Study participation is voluntary and a written informed consent is required. After the baseline questionnaire, follow-up questionnaires will be mailed to the participants every 6 months for up to 5 years after baseline. Self-reported serious adverse events will be validated by contacting the relevant physician and by reviewing relevant source documents. At the end of the study an independent blinded adjudication of relevant clinical outcomes will be conducted. Meanwhile, this study has received ethical approval from the Western Institutional Review Board (USA) and the Medical Association in Berlin (Germany). DISCUSSION: The feasibility of the study is considered to be very high because of its similar design to the EURAS-OC study. All relevant methodological and logistical features of the study were successfully tested in the EURAS study. The chosen design minimizes the impact of referral and misclassification bias, healthy user effect and loss to follow-up. Overall, it is expected that the study design is robust enough to interpret hazard ratios of 1.5 or higher.