Key Takeaways
- Registry-confirmed diagnoses at scale: Clinical Specialty (CS) Oncology includes more than 150,000 patients with registry-confirmed cancer diagnoses sourced from a hospital registry that contributes to an NAACCR Gold-certified central cancer registry.
- A regularly updated resource: Semiannual refreshes inclusive of the cancer registry and biomarkers have added as many as 19,000 new patients annually, allowing the dataset to reflect evolving therapeutic patterns and newly approved regimens over time.
- Longitudinal coverage across the patient timeline: Integration with NashBio’s Structured Clinical Data (SCD) environment provides a median of 7.2 years of continuous electronic health record (EHR) history, allowing researchers to reconstruct treatment sequences across lines of therapy and track serial tumor biomarker measurements.
- Structured, analysis-ready format: CS Oncology pairs registry-grade data with structured EHR records in a long-format schema, supporting cross-domain linkage for safety, efficacy, and outcomes research.
The Convergence of the Patient Journey
As oncology moves toward precision medicine, researchers recognize the value of linking verified disease phenotypes with therapeutic and outcomes data over multi-year timelines (Booth et al., 2019). However, the oncology real-world data (RWD) landscape remains fragmented: claims-based datasets often lack clinical granularity, specialty networks may miss pre-diagnosis context, and registries are frequently disconnected from longitudinal treatment records (Penberthy et al., 2022). Researchers need an environment in which registry-confirmed diagnoses are linked to longitudinal clinical data, capturing the patient timeline both before and after the index diagnosis.
CS Oncology: Structure and Scope
NashBio is formally launching CS Oncology, an oncology resource that links longitudinal electronic health records (EHRs) with registry data across cancer types in one unified table. CS Oncology combines registry-confirmed diagnoses with longitudinal clinical information that is often distributed across separate oncology data sources. Five characteristics define the dataset:
- Diagnostic certainty: CS Oncology features biopsy-proven, registry-confirmed cancer diagnoses for more than 150,000 patients. In addition, the cohort continues to expand through semiannual data refreshes, with annual additions historically reaching as many as 19,000 patients across cancer registry and biomarker data.
- Registry-standard data quality: Source data are manually abstracted by Oncology Data Specialists following standardized oncology coding and abstracting practices consistent with National Cancer Institute Surveillance, Epidemiology, and End Results (NCI SEER) Program guidelines. These data contribute to a central cancer registry that meets the highest level of certification from the North American Association of Central Cancer Registries (NAACCR) (Menck and Smart, 1994).
- Granular clinical variables: Available fields include microscopic histology, tumor grading, pathological and clinical AJCC TNM staging, and progression markers such as metastatic sites at diagnosis.
- Serial biomarker surveillance: Six longitudinal tumor markers (PSA, CEA, CA125, CA 19-9, AFP, and beta-hCG) are captured in CS Oncology. These measurements can support evaluation of screening, therapeutic response, biochemical recurrence, and disease trajectory across many cancer types.
- Analysis-ready structure: The dataset is delivered in an analysis-ready long format, reducing the data transformation steps typically required before analysis.
EHR Integration and Longitudinal Clinical Data
Unlike specialty oncology networks that capture a limited window of the treatment timeline, CS Oncology is linked to the NashBio Structured Clinical Data (SCD). This provides a median of 7.2 years of continuous EHR history per patient, including pre-diagnosis baseline context and post-diagnosis follow-up. Linkage to the SCD environment makes several types of longitudinal analyses possible:
- Treatment sequence across lines of therapy: Researchers can reconstruct treatment sequences after diagnosis using structured medication, procedure, and encounter data to evaluate regimen changes and longitudinal treatment patterns.
- Comorbidity risk factor data: The SCD captures the comorbidity profile of each patient, including autoimmune conditions and environmental and lifestyle risk factors. These data are relevant to modeling therapeutic eligibility, adverse event risk, and survival outcomes (Martins et al., 2019).
- Cross-domain connectivity: Linkage at the patient, encounter, and measurement levels connects registry, laboratory, medication, procedure, and imaging domains. For example, raw longitudinal imaging data (CT, MRI, PET/CT) are available as an add-on, allowing researchers to correlate radiographic disease progression with clinical and biomarker endpoints (Shur et al., 2021).
Research and Clinical Applications
The value of linking registry diagnoses with longitudinal clinical records is illustrated by studies such as Johnson et al. (2016) from the Vanderbilt-Ingram Cancer Center (VICC). This study identified fulminant myocarditis as a severe immune-mediated adverse event in patients receiving combination immune checkpoint blockade with ipilimumab and nivolumab. Such analyses require access to longitudinal clinical information that captures therapeutic intervention and patient baseline clinical characteristics.
CS Oncology is structured to support this type of integrated analysis across large patient populations. By pairing registry-grade diagnostic data with years of structured EHR records, NashBio provides a resource that supports a range of research applications, including health economics and outcomes research and real-world evidence studies.
Ready to explore CS Oncology? Contact NashBio to request a data snapshot or schedule a technical demo.
