A glossary of diagnostic data terms.

A group of patients identified by a confirmed biomarker test result, such as HER2-positive breast cancer or an EGFR mutation in lung cancer. In commercial data, the cohort is summarized to the providers and facilities where those results surfaced, without exposing the patients themselves.

It tells a brand team which providers already have patients who may be eligible for a targeted therapy.

An in vitro diagnostic test that provides information essential for the safe and effective use of a corresponding drug, usually by identifying the patients most likely to benefit. Companion diagnostics are approved or cleared by the FDA.

When a therapy requires a companion diagnostic, testing volume for that assay is a leading indicator of where eligible patients are.

A notification that a specific healthcare provider has had a qualifying testing event, keyed to a biomarker or an ICD-10 diagnosis code. A Hyperdrive HCP alert names the ordering and treating physician and the facility where the result surfaced.

Alerts reach a field team while the treatment decision is still open, rather than months later in claims.

A code from the International Classification of Diseases, 10th Revision. In the US, ICD-10-CM diagnosis codes appear on claims and test requisitions to record why a test was ordered or a service was provided. C50, for example, covers malignant neoplasm of the breast.

ICD-10 codes let alerts and cohorts be defined by diagnosis, not only by test result.

The difference between the patients who should receive a guideline-recommended biomarker test and the patients who actually do. Gaps tend to concentrate in particular practices, regions and care settings.

Closing a testing gap is how more patients reach a precision therapy, and it shows where education and access efforts are needed.

A laboratory accreditation program run by the College of American Pathologists, based on peer inspections against detailed checklists. CMS recognizes CAP accreditation as meeting CLIA requirements.

Accreditation is one signal of a laboratory’s quality program when you evaluate a lab as a customer or partner.

The Clinical Laboratory Improvement Amendments of 1988, the federal standards for US laboratories that test human specimens for diagnosis, prevention or treatment. CMS administers CLIA with the FDA and CDC, and each lab holds a CLIA certificate matched to the complexity of testing it performs.

A lab’s CLIA certificate type indicates which tests it is permitted to run.

The time between removing tissue from the body and placing it in fixative. For breast biomarker testing, ASCO and CAP guidelines recommend keeping it to one hour or less.

Long cold ischemia time can degrade proteins and nucleic acids, which can weaken IHC staining and molecular results.

A five-character code from Current Procedural Terminology, maintained by the American Medical Association, used to report medical procedures and services. Pathology and laboratory services are reported with codes in the 80000 series.

Medicare billing by CPT code shows which tests and methodologies a lab bills for, which lets you infer what it performs.

The practice of scanning glass pathology slides into high-resolution whole slide images so they can be viewed, shared, stored and analyzed on screen instead of under a microscope.

Digitized slides make pathology data usable at scale. The Hyperdrive Lab Data Network includes more than 5 million digital whole slide images.

The standard way to preserve tissue for diagnosis. The sample is fixed in formalin, then embedded in a block of paraffin wax that can be cut into thin sections and stored for years.

Archived FFPE blocks are why tissue can be retested later, for example when a new biomarker becomes relevant to treatment.

A technique that uses fluorescent DNA probes to detect specific gene sequences in cells, such as gene amplifications and rearrangements. HER2 amplification and ALK rearrangement are common oncology examples.

FISH is often used to confirm or clarify an IHC result, so it appears in many biomarker testing pathways.

How long tissue stays in fixative, usually 10 percent neutral buffered formalin, before processing. For breast ER, PR and HER2 testing, ASCO and CAP guidelines call for 6 to 72 hours.

Under- or over-fixation can produce false-negative IHC results, which is why guidelines set limits for key biomarker tests.

The routine stain used in histopathology. Hematoxylin colors cell nuclei blue-purple, and eosin colors cytoplasm and surrounding tissue pink.

H&E is the starting point for a diagnosis and for selecting the tumor area used in follow-on biomarker testing.

The way HER2 IHC results are read, on a scale of 0, 1+, 2+ or 3+ based on the intensity and completeness of membrane staining. A 2+ result is equivocal and is usually followed by in situ hybridization testing.

HER2-low results, 1+ or 2+ without amplification, became clinically relevant once a therapy was approved for them, so how a result is scored now affects treatment options.

The microscopic examination of tissue to diagnose disease, typically by a pathologist reviewing stained sections cut from a biopsy or surgical specimen.

Every tissue biomarker result starts with a histopathology diagnosis that confirms the tumor type.

Software that stores, organizes and displays digital whole slide images, usually connected to the laboratory information system so each image stays linked to its case.

In a digital pathology lab, the IMS holds the slide archive that research and AI programs draw on.

A staining technique that uses antibodies to detect specific proteins in a tissue sample under the microscope. ER, PR, HER2 and PD-L1 are commonly assessed by IHC.

IHC is the first-line test for several major biomarkers, so IHC volume shows where those patients are being identified.

A test designed, validated and performed within a single laboratory, rather than sold as a kit to other labs. LDTs are run in CLIA-certified laboratories that meet high-complexity testing requirements.

Many specialized oncology and genomic tests are LDTs, so a lab’s menu can be broader than its FDA-cleared assays.

A method that sequences many genes, or millions of DNA fragments, in parallel. In oncology it is used for targeted gene panels and comprehensive genomic profiling, on tissue or on blood in a liquid biopsy.

A single NGS result can identify several actionable biomarkers at once.

How PD-L1 IHC results are measured. The tumor proportion score (TPS) is the percentage of viable tumor cells with membrane staining. The combined positive score (CPS) counts staining tumor cells, lymphocytes and macrophages, divided by viable tumor cells, times 100.

The scoring method and cutoff depend on the tumor type and the therapy, so the same stain can be read against different thresholds for different drugs.

A technique that copies a specific DNA sequence millions of times so it can be detected and measured. Variants such as RT-PCR and qPCR extend it to RNA and to quantitative results.

PCR-based assays remain a fast, lower-cost option for single-gene biomarker tests.

A form of PCR that first converts RNA into complementary DNA, then amplifies it. It is used to measure gene expression and to detect gene fusions at the RNA level. Real-time PCR is a different technique, often abbreviated qPCR.

Many actionable fusions are detected most reliably in RNA, so RT-PCR and RNA sequencing can find results a DNA-only test misses.

The percentage of cells in a sample that are tumor cells, estimated by a pathologist from the H&E slide. Molecular assays each set a minimum tumor content for a valid result.

Samples below an assay’s minimum can return a failed or false-negative result, one of the quieter causes of testing gaps.

An organization that combines the purchasing volume of member hospitals and health systems to negotiate contracts with manufacturers, distributors and suppliers.

A hospital’s GPO affiliation helps a sales team understand which contracts may already shape its purchasing.

Whether a provider or facility has a contract with a given health plan to deliver care at negotiated rates. Out-of-network care typically costs the patient more.

Network status affects where patients can be referred and tested. In Voyager it is checked on demand.

An annual report that Medicare-certified institutional providers, including hospitals, file with CMS. It covers facility characteristics, utilization, costs and charges.

Hospital cost reports provide verified bed counts and total inpatient and outpatient charges for each hospital.

The part of Medicare that covers outpatient care, physician services and clinical laboratory tests. CMS publishes Part B utilization and payment data at the provider level.

Part B lab spend by CPT code shows what laboratories bill for. The most recent data in Voyager is from 2024 and is updated annually.

Medicare’s outpatient prescription drug benefit, delivered through private plans. CMS publishes prescriber-level data on the drugs dispensed under Part D.

Part D prescribing history shows which providers already treat with a given class of therapy.

A unique 10-digit number assigned to US healthcare providers through CMS’s National Plan and Provider Enumeration System. Type 1 NPIs identify individuals, and Type 2 NPIs identify organizations such as hospitals, groups and labs.

The NPI is the key that links a provider across claims, enrollment and testing data.

The ordering physician requests a test. The treating physician manages the patient’s care and treatment decisions. They are often different people, for example a surgeon who orders a biopsy and an oncologist who treats.

Reaching only the ordering physician can miss the person who makes the treatment decision, which is why Hyperdrive HCP alerts name both.

CMS’s system of record for Medicare provider enrollment. It holds each enrolled provider’s enrollment status, practice locations and reassignments to groups.

PECOS enrollment and reassignment data show which providers bill through which physician groups.

One of two HIPAA methods for de-identifying health data. A qualified statistical expert analyzes the data set and documents that the risk of identifying an individual is very small.

Expert determination can support richer data fields than Safe Harbor while still meeting HIPAA.

Individually identifiable health information held or transmitted by a HIPAA covered entity or its business associate, such as a name, address or medical record number tied to a patient’s care.

De-identified data contains no PHI, and Hyperdrive shares no PHI with biopharma.

The other HIPAA de-identification method. It removes 18 specified identifiers, including names, most geographic detail smaller than a state and most dates, and requires no actual knowledge that the remaining data could identify someone.

Safe Harbor is the simpler standard, but removing dates and fine geography limits what the data can support.

Combining records from two or more sources that refer to the same patient, provider or organization, such as joining lab results to claims, or a provider’s NPI to their group and hospital affiliations.

Linkage at the provider level, rather than the patient level, ties lab activity to the physicians and organizations behind it.

The order of treatments a patient receives for a disease. First-line therapy is the initial treatment, and second-line and later therapies follow if the disease progresses or a treatment is not tolerated.

Many targeted therapies are approved for a specific line of therapy, which affects which patients and providers matter to a brand.

Closed claims come directly from a payer and cover a defined set of members while they are enrolled, so each record is complete within that window. Open claims come from clearinghouses and other points in the billing stream, covering far more patients across payers but with gaps in any one patient’s history.

The trade-off is completeness against coverage. Either way, a claim arrives only after the service has been billed.

The sequence of events a patient moves through for a condition, from first symptoms and diagnosis through testing, treatment decisions, therapy and follow-up.

In precision medicine, the biomarker test is the point in the journey where a targeted therapy becomes possible.

Health data collected outside of controlled clinical trials, from sources such as electronic health records, insurance claims, laboratory results, registries and pharmacy records.

Laboratory results are one of the few RWD sources that capture the biomarker status behind a treatment decision.

Clinical evidence about how a medical product is used, and about its benefits or risks, that comes from analyzing real-world data. The FDA considers RWE in some regulatory decisions, including support for new indications.

RWE studies depend on RWD sources with the right clinical detail, such as confirmed biomarker results.

The setting where a service is delivered, such as a hospital inpatient unit, hospital outpatient department, physician office, ambulatory infusion center or the patient’s home.

Site of care affects who orders and bills a test or therapy, and what it costs.

A privacy technique that replaces patient identifiers with an encrypted token, so records about the same person from different sources can be matched without revealing who that person is.

Tokenization is a common way to link de-identified data sets. The linked result is typically reviewed again for re-identification risk.

An entity such as a hospital, health system, physician group, laboratory or clinic, typically identified by a Type 2 NPI.

Selling often happens at the HCO level, so knowing which HCPs belong to which HCO shapes account planning.

An individual clinician, such as a physician, nurse practitioner or physician assistant. In commercial teams, HCP usually means a provider a sales or medical team engages.

HCP-level data, identified by NPI, is the unit most field teams work from.

Choosing which healthcare providers a commercial team should engage, and in what order, based on factors such as specialty, patient volume, testing behavior and access.

Targeting is only as good as the data behind it, which is why current, verified provider data matters.

A clinician recognized by peers as an expert in a therapeutic area, often through research, publications, guideline work or speaking. Medical affairs teams typically lead KOL engagement.

KOLs shape how a field adopts new tests and therapies, including biomarker testing practices.

The work of securing coverage, reimbursement and favorable placement for a drug or test with payers, so the patients who need it can get it.

Without coverage, a test or therapy may not reach patients even when providers want to use it.

Identifying where patients with a specific condition or biomarker are being diagnosed and treated, summarized to the providers and sites involved rather than to individual patients.

For rare biomarkers, patient finding shows which few providers actually see eligible patients.

The share of a provider’s or organization’s patients covered by each type of payer, such as Medicare, Medicaid, commercial plans and self-pay.

Payer mix affects reimbursement and access, and helps explain why testing or prescribing patterns differ between sites.

The pattern of referrals between providers and organizations, such as which primary care physicians send patients to which specialists, and which practices send specimens to which labs.

Referral networks show influence and patient flow that a list of providers alone does not.

Dividing a market into sales territories and assigning accounts and providers to reps, usually to balance workload and opportunity across the team.

Accurate provider counts, locations and affiliations are the inputs that make territories balanced.