Oncotype DX vs MammaPrint vs EndoPredict — Breast Cancer Genomic Test Comparison & Calculator
Compare Oncotype DX, MammaPrint, and EndoPredict breast cancer genomic assays. Understand recurrence scores, clinical utility, and when each test guides chemotherapy decisions.
Oncotype DX is a proprietary 21-gene assay that reports a score from 0 to 100. This site accepts that external result and keeps TAILORx node-negative strata separate from RxPONDER one-to-three-node evidence; it does not calculate the assay or turn trial observations into patient-specific advice.
Oncotype DX — The 21-Gene Recurrence Score
Oncotype DX is the most widely used genomic assay for early-stage breast cancer in the United States and carries an NCCN Category 1 recommendation for chemotherapy decision-making in HR+/HER2- disease. In the UK, NICE diagnostic guidance DG34 recommends Oncotype DX for ER-positive, HER2-negative, node-negative early breast cancer to guide chemotherapy decisions, alongside EndoPredict (DG42) and Prosigna as approved genomic tests. These tests are funded by the NHS for eligible patients.
How It Works
The assay measures expression of 21 genes — 16 cancer-related genes and 5 reference genes — from formalin-fixed paraffin-embedded (FFPE) tumour tissue using reverse transcription polymerase chain reaction (RT-PCR). The result is a continuous Recurrence Score (RS) from 0 to 100.
Key Trial Evidence
TAILORx (2018) — the landmark trial for node-negative disease:
- 10,273 women with HR+/HER2-/N0 breast cancer
- RS 0-10: assigned endocrine-therapy cohort; not randomized
- RS 11-25: randomized comparison; age <=50 exploratory subgroups remain separate
- RS 26-100: assigned chemoendocrine cohort; not randomized
RxPONDER (2021) — extended evidence to node-positive disease:
- 5,083 women with HR+/HER2-, 1–3 positive lymph nodes, RS 0–25
- Postmenopausal women: no chemotherapy benefit demonstrated in the randomized comparison
- Premenopausal women: improved IDFS and distant relapse-free survival observed; no patient-specific benefit percentage is inferred
Key point: Use the Oncotype DX Breast source-specific report interpreter only with a final external report. It does not recommend treatment.
Oncotype DX Score Interpretation
| RS Range | Source population | Required context | Source-specific interpretation | |----------|------------|-----------|---------------------------| | 0–10 | TAILORx N0 | chronological age | assigned endocrine cohort; not randomized | | 11–25 | TAILORx N0 | age 50/51 boundary | randomized; exploratory age subgroups separated | | 26–100 | TAILORx N0 | chronological age | assigned chemoendocrine cohort; not randomized | | 0–25 | RxPONDER N1 | menopausal status | menopause-specific randomized observation | | 26–100 | RxPONDER N1 | any | outside RxPONDER score evidence |
MammaPrint — The 70-Gene Signature
MammaPrint uses a 70-gene expression profile to classify breast cancer into binary risk categories: low risk or high risk. It was the first FDA-cleared genomic test for breast cancer and has broader applicability than Oncotype DX.
How It Works
MammaPrint measures the expression of 70 genes associated with metastatic potential using DNA microarray technology. Unlike Oncotype DX's continuous score, MammaPrint provides a binary result — the tumour is classified as either genomically low risk or high risk based on a predefined signature threshold.
The test can be performed on fresh or FFPE tissue, offering flexibility in tissue handling compared with some competing assays.
Key Trial Evidence
MINDACT (2016) — the pivotal prospective validation trial:
- 6,693 women with early breast cancer (node-negative or 1–3 positive nodes)
- Patients classified as clinical high risk / genomic low risk were randomised to chemotherapy vs no chemotherapy
- 5-year distant metastasis-free survival without chemo: 94.7% — demonstrating that patients classified as genomic low risk can safely avoid chemotherapy even when clinical features suggest higher risk
- Benefit extended to patients with 1–3 positive lymph nodes
Broader Applicability
MammaPrint can be used for:
- HR+/HER2- tumours (like Oncotype DX)
- HR+/HER2+ tumours (unlike Oncotype DX, which is limited to HER2-)
- Node-negative and node-positive disease (up to 3 positive nodes)
This broader indication profile makes MammaPrint a useful option when the clinical scenario falls outside Oncotype DX's validated indications.
EndoPredict — The 12-Gene Score
EndoPredict measures 12 genes (8 cancer-related genes and 3 reference genes, plus a normalisation gene) and produces two distinct scores — a pure molecular score and the EPclin score, which integrates molecular data with tumour size and nodal status.
How It Works
The assay is performed on FFPE tissue using quantitative RT-PCR and produces:
- EP score: The molecular-only component (12-gene expression)
- EPclin score: Combines the EP score with tumour size (pT) and nodal status (pN) for a comprehensive risk assessment
An EPclin score below 3.3 indicates low risk of distant recurrence over 10 years; 3.3 or above indicates high risk.
Key Trial Evidence
EndoPredict has been validated in multiple retrospective-prospective analyses:
- ABCSG-6 and ABCSG-8 trials: Demonstrated strong prognostic performance in postmenopausal women with HR+/HER2-, node-negative or node-positive (1–3 nodes) breast cancer
- 10-year and 15-year follow-up data: EndoPredict showed particularly strong prediction of late distant recurrence (years 5–15), outperforming other assays in this time window
EndoPredict's Unique Strength: Late Recurrence Prediction
HR+/HER2- breast cancer has a well-recognised pattern of late recurrence — distant metastases can occur 10, 15, or even 20 years after diagnosis. While Oncotype DX primarily predicts 10-year recurrence risk, EndoPredict's EPclin score provides robust discrimination between patients at low vs high risk of late events beyond 5 years. This information is particularly valuable for decisions about extended endocrine therapy (continuing beyond the standard 5 years).
Clinical recommendation: Consider EndoPredict when the primary clinical question is whether to extend endocrine therapy beyond 5 years, as it provides the strongest evidence for late recurrence prediction.
Head-to-Head Comparison
| Feature | Oncotype DX | MammaPrint | EndoPredict | |---------|------------|------------|-------------| | Genes tested | 21 | 70 | 12 | | Score range | 0–100 (continuous) | Low / High (binary) | EPclin (continuous) | | Key trial | TAILORx, RxPONDER | MINDACT | ABCSG-6/8 | | Node status | N0 (+ N1 with RxPONDER) | N0–N1 (up to 3 nodes) | N0–N1 | | Late recurrence | Moderate | Moderate | Strong | | HER2 status | HER2- only | HER2- and HER2+ | HER2- only | | Tissue required | FFPE (paraffin) | Fresh or FFPE | FFPE | | Turnaround | ~10–14 days | ~10 days | ~5–7 days | | NCCN recommended | Category 1 | Category 1 | Category 2A |
Key Takeaways from the Comparison
- Oncotype DX has the most extensive prospective evidence base, with both TAILORx and RxPONDER providing Level 1 evidence for chemotherapy benefit prediction
- MammaPrint offers the broadest applicability (including HER2+ tumours) and robust prospective data from MINDACT
- EndoPredict provides the fastest turnaround and strongest late recurrence prediction, with the added value of integrating clinical factors into the EPclin score
When to Use Each Test
The choice of genomic test depends on the clinical scenario, patient characteristics, and the specific clinical question being addressed.
HR+/HER2-/Node-Negative (N0)
Oncotype DX is the first-line choice in this scenario, supported by the largest body of prospective evidence (TAILORx). MammaPrint and EndoPredict are acceptable alternatives with strong supporting data.
HR+/HER2-/Node-Positive (1–3 nodes)
Oncotype DX (RxPONDER data) or MammaPrint (MINDACT data) — both have prospective evidence in node-positive disease. The choice may depend on menopausal status, as RxPONDER demonstrated a critical interaction between RS and menopausal status in node-positive patients.
Concern About Late Recurrence
EndoPredict provides the strongest evidence for predicting late distant recurrence (years 5–15). This is particularly relevant for patients considering extended endocrine therapy duration.
Need for Fast Results
EndoPredict has the fastest turnaround (~5–7 days), which can be advantageous when timely treatment decisions are needed.
HR+/HER2+ Tumours
MammaPrint is the only assay validated for use in HER2-positive tumours. Oncotype DX and EndoPredict are restricted to HER2-negative disease.
Using Genomic Score Calculators at Point of Care
Interpreting genomic test results requires integration with clinical factors — particularly age, menopausal status, and nodal status. OncoToolkit provides calculators to assist with interpretation:
- Oncotype DX Breast report interpreter — Match a final external score to a source-specific TAILORx or RxPONDER stratum
- PREDICT Breast official tool guide — Understand how PREDICT population estimates differ from genomic assay reports
These calculators help clinicians contextualise genomic test results within the broader clinical picture, supporting informed shared decision-making with patients about chemotherapy benefit. OncoToolkit hosts over 130 oncology calculators spanning staging, prognosis, treatment toxicity, and supportive care.
Frequently Asked Questions
Is there one “good” Oncotype DX score?
No. TAILORx and RxPONDER used different populations and score ranges. The interpreter returns the applicable source stratum rather than one Low/Intermediate/High or treatment category.
What is the difference between Oncotype DX and MammaPrint?
Oncotype DX analyzes 21 genes and produces a continuous score (0–100), validated primarily by TAILORx and RxPONDER. MammaPrint analyzes 70 genes and gives a binary result (low/high risk), validated by MINDACT. Oncotype DX is the most widely used in the US with NCCN Category 1 recommendation. MammaPrint has broader applicability including HER2-positive tumours and validation in node-positive disease up to 3 nodes.
Does this route decide whether chemotherapy is needed?
No. It reports source-specific trial context and leaves treatment decisions outside automated computation.
Which breast cancer genomic test is most accurate?
Each test excels in different areas. Oncotype DX has the strongest prospective evidence for chemotherapy benefit prediction in HR+/HER2- disease. MammaPrint has the broadest applicability across tumour subtypes and nodal status. EndoPredict provides the best late recurrence prediction (years 5–15). No single test is universally superior — the choice should match the clinical question.
What is the EndoPredict EPclin score?
The EPclin score combines EndoPredict's 12-gene molecular score with clinical factors (tumour size and nodal status) into a single continuous risk score. EPclin <3.3 indicates low risk and ≥3.3 indicates high risk of distant recurrence. Its unique strength is robust prediction of late recurrence (years 5–15), making it particularly useful for decisions about extended endocrine therapy.
This article is intended for healthcare professionals. Genomic test results should be interpreted within the context of individual patient characteristics, clinical stage, tumour biology, and multidisciplinary team discussion.