Flatiron Health


Flatiron Health is a health technology company founded in 2012 and an independent affiliate of the Roche Group, partnering with hundreds of cancer centers, life science companies, and regulators worldwide. Their longitudinal, regulatory-grade database contains over 5 million de-identified patient records and 1.5 billion data points across 22 cancer-specific cohorts, spanning solid tumors and hematologic malignancies, incorporating structured, abstracted, derived, and ML-extracted variables with scientifically validated outcome measures 1,2.

Data offerings include non-sampled patient cohorts, multinational datasets from the UK, Germany, and Japan, linked datasets with third-party sources, and customizable options. Their multidisciplinary team of oncology researchers, biostatisticians, epidemiologists, and regulatory specialists offers end-to-end analytical services and project support, with over a decade of expertise and a proven track record collaborating with the Food and Drug Administration (FDA), National Comprehensive Cancer Network (NCCN), and National Institute for Health and Care Excellence (NICE) 3,4.

Updated: April 28th, 2026

Overview

Flatiron Health is a health technology company that provides technology and services to support patient care through Electronic Medical Record (EMR) software used by oncologists, while generating high-quality data for researchers, regulators, and life science companies. Founded in 2012, it became an independent affiliate of the Roche Group in 2018. The company partners with hundreds of cancer centers, leading global developers of oncology therapeutics, and supporting researchers and regulators worldwide 1. Its Flatiron International subsidiaries include dedicated local partnerships in Germany, the UK, and Japan.

Flatiron Health’s longitudinal, regulatory-grade data is designed to be accessible and usable across a broad range of users, including academics, regulators, pharmacovigilance and epidemiology professionals, and research, commercial, and marketing teams 1,2. Their database contains over 5 million de-identified patient records and 1.5 billion data points available for research. Their in-house team also brings over 5 years of direct collaboration with the Food and Drug Administration (FDA), National Institute for Health and Care Excellence (NICE), and other Health Technology Assessment (HTA) bodies 3.

Longitudinal, de-identified patient-level real-world data across 22 cancer-specific cohorts spanning solid tumors and hematologic malignancies are accessible through Flatiron Health’s Horizon Datascapes. Users can access the full non-sampled patient dataset, link to external Electronic Health Record (EHR) or insurance datasets, apply add-on enhancements, or customize disease settings to focus on specific outcomes. Data elements include structured, abstracted, derived, and Machine Learning (ML)-extracted variables, with scientifically validated outcome measures such as real-world response and progression 3,4.

Flatiron Health curates and compiles a range of data offerings to meet the research and data insight needs of diverse teams:

  • Panoramic: Longitudinal, non-sampled patient cohorts across disease settings, capturing every applicable patient in the Flatiron Health network. Biomarker data are also available at scale for both emerging and standard-of-care biomarkers 5.

  • Multinational: Global oncology real-world data sourced from local partnerships in the UK, Germany, and Japan. Data are standardized to US data models, enabling cross-market comparison and analysis, while remaining fully compliant with local regulatory requirements and secondary health data use norms. Researchers can access patient-level data and custom datasets, supported by in-country technology infrastructure and analytical services 6.

  • Multimodal: The multimodel links Flatiron Health real-world data with third-party insurance claims data or pooled with other EHR-derived data to deepen insights into patient outcomes, cost of care, rare oncology diseases, and subgroup analyses without the risk of duplicate patients 7.

  • Custom: Data curated with additional inclusion/exclusion criteria for more precise cohorts and additional variables to answer novel research questions. Flatiron supports custom data projects end-to-end, from planning and scoping through execution, delivery, and post-delivery support, in close collaboration with biopharma partners 8.

  • Enhanced Datamarts: Enhanced Datamarts (EDMs) are longitudinal, de-identified real-world dataset subscriptions featuring deep, disease-curated clinical data models across 22 cancer types spanning solid tumors and hematologic malignancies. EDMs include histologic information, detailed treatment and testing data, and scientifically validated endpoints such as real-world progression and mortality, combining structured, abstracted, derived, and extracted data elements. Records are refreshed monthly with 30-day recency 9.

  • Rapid Enhancements: Optional add-on variables for EDMs, including select outcome variables, covering: reasons for therapy discontinuation, detailed oral therapy data, Charlson comorbidities, additional malignancies, sites of metastases, real-world progression, and real-world response 10.

Beyond data, Flatiron Health offers end-to-end project support and analytic services through flexible engagement models, delivering tailored oncology evidence solutions from research and evidence generation to rapid analytical insights. Their multidisciplinary team, comprising oncology researchers, biostatisticians, epidemiologists, and regulatory specialists, brings over a decade of expertise, direct access to the data, and a proven track record of collaboration with the FDA, National Comprehensive Cancer Network (NCCN), and NICE 2.

Gaining Access

Do I Qualify?

Applications are assessed on an individual basis by Flatiron Health experts.

Typical Timeline

Time constraints may vary for applicants.

Step-by-Step Guide

Researchers can initiate the application process by completing this form.

Publications

This section presents a selection of PubMed articles that utilize the dataset and are authored by individuals affiliated with the Yale University. These articles are provided to inspire researchers and students to use the data in their own work.

Back to top

References

1.
Flatiron Health. Who we are. https://flatiron.com/about-us.
2.
Flatiron Health. Real-world evidence services. https://flatiron.com/real-world-evidence/services.
3.
Flatiron Health. Real-world evidence. https://flatiron.com/real-world-evidence.
4.
Flatiron Health. Horizon datascapes. https://flatiron.com/real-world-evidence/real-world-data.
5.
Flatiron Health. Panoramic data: Maximize use cases across your oncology portfolio with a comprehensive, disease-specific solution.
6.
Flatiron Health. Multinational data. https://flatiron.com/real-world-evidence/multinational-data.
7.
Flatiron Health. Multimodal data genomics database. https://flatiron.com/real-world-evidence/multimodal-data.
8.
Flatiron Health. Custom data solutions. https://flatiron.com/real-world-evidence/custom-data-solutions.
9.
Flatiron Health. Enhanced datamart. https://flatiron.com/real-world-evidence/enhanced-datamart-edm.
10.
Flatiron Health. Rapid enhancements and outcome variables.