Intrinsic breast-cancer subtype
In one sentence
An intrinsic breast-cancer subtype is a category assigned by comparing a tumor sample's gene-expression pattern with reference patterns.
The intuition
One description of a song names its instruments; another describes its musical style. Both can be informative, but they answer different questions. Clinical receptor tests describe selected proteins and gene amplification. An intrinsic subtype summarizes a broader pattern of gene activity in the sample.
The analogy stops there: a tumor is a changing mixture of cells, and a classifier's label depends on its measurements and rules. “Intrinsic” in this name does not mean inherited from a parent or impossible to change.
How it works
Gene expression is the production of RNA (ribonucleic acid) from genes. Expression profiling asks which transcripts are relatively abundant. The original intrinsic-subtype research identified recurring breast-tumor patterns, rather than defining groups solely by estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2) test results. Perou and colleagues' original study established this expression-based approach.
PAM50 is a classifier built around a 50-gene expression panel. It compares a sample's pattern with reference patterns; one high gene does not determine the label. In the classic research framework, the groups include the following. These are relative expression patterns, not treatment guarantees.
| Expression label | The pattern it summarizes |
|---|---|
| Luminal A | Luminal and hormone-related expression, usually with less proliferation-related expression than luminal B |
| Luminal B | Luminal expression with more proliferation-related expression than luminal A |
| HER2-enriched | A reference pattern involving HER2-related and other growth-related expression |
| Basal-like | Expression resembling aspects of basal breast epithelial programs |
| Normal-like | Similarity to a normal-breast reference pattern; sample composition is an important concern |
The PAM50 development study and a direct clinical-marker comparison support this distinction between expression classes and receptor categories.
Clinical assays and research implementations differ in preprocessing, permitted outputs and intended use. A particular assay need not report every research category; check its permitted outputs and intended use. A subtype label is also different from a risk-of-recurrence score, which may combine expression features and clinical inputs under a separate validated model.
Why it matters in cancer
Intrinsic subtypes help researchers organize biologically different tumors that share a clinical receptor category. They can support defined prognostic applications when the assay, population and intended use have been validated. A research label alone does not establish treatment eligibility or benefit.
The names can mislead. HER2-enriched does not automatically mean HER2-positive pathology; basal-like does not automatically mean TNBC. “Luminal” describes an expression program rather than proving a tumor's exact cell of origin. A normal-like label does not mean the tissue is benign: normal cells in a bulk specimen can influence the measured pattern, a concern raised by the PAM50 investigators.
Worked example
A fictional tumor is ER negative, PR negative and HER2 negative on validated tissue tests. A separate, quality-controlled RNA analysis classifies it as HER2-enriched. These results can coexist. The RNA label does not create HER2 amplification or replace the tissue tests. The next interpretation depends on the classifier, specimen composition and clinical question.
Common confusions
- A receptor-based “luminal-like” surrogate is not the same measurement as a PAM50 intrinsic subtype.
- “Molecular subtype” is a broad phrase; it can describe other frameworks besides intrinsic subtyping.
- A subtype is a pattern across genes, not necessarily a single actionable mutation or pathway dependency.
- Basal-like does not prove that the cancer began in a mature basal cell.
How it is measured
Record the tissue sample, tumor content, platform, classifier version, preprocessing and allowed output labels. Bulk RNA measurements mix transcripts from malignant cells and their neighbors. A label from one platform should not be treated as interchangeable with every expression label produced elsewhere.
Explain it back
Why should a receptor category and an intrinsic subtype stay as two descriptions? They measure different features with different methods, and their overlap is substantial but incomplete.
Related concepts
Sources and scope
General education; fictional example. Source-checked October 9, 2026. Expert and learner review remain pending. The subtype list describes the classic research framework, not every clinical assay's output or indication.
- Perou et al., Molecular portraits of human breast tumours (2000): original expression-based subtype framework.
- Parker et al., Supervised risk predictor of breast cancer based on intrinsic subtypes (2009): PAM50 development, subtype versus risk models and normal-like sample-composition concerns.
- Bastien et al., PAM50 subtyping and concordance with standard clinical molecular markers (2012): luminal proliferation patterns and incomplete correspondence with receptor results.