Understanding Oncotype DX & Recurrence Scores

Tests such as Oncotype DX and MammaPrint are run on your tumor tissue, not your blood, and not your inherited DNA. They measure how active certain genes are inside the cancer cells that were already removed. Oncotype DX reads 21 genes and reports a Recurrence Score from 0 to 100. MammaPrint reads 70 genes and reports a binary result, low risk or high risk. Both are used mainly in hormone receptor-positive, HER2-negative early breast cancer.

These tests are not a scan. Nothing about the score tells you whether cancer is present in your body now.

This is the part most often misread. The score does estimate the chance of the cancer returning over roughly the next decade with hormone therapy alone. But its practical purpose is different: it estimates whether adding chemotherapy would meaningfully change that chance for you.

Those are separate questions. Two people can have similar recurrence risk and very different expected chemotherapy benefit, because benefit depends on how the tumor's biology responds to chemotherapy rather than simply on how likely it is to return. A tumor can be relatively likely to recur and still gain very little from chemotherapy. That is why a score can be reassuring about treatment intensity without claiming your risk is zero.

TAILORx enrolled 10,273 women with hormone receptor-positive, HER2-negative, node-negative breast cancer. Scores of 0 to 10 were treated as low risk, 11 to 25 as intermediate, and 26 or above as high. The intermediate group, where practice had genuinely been uncertain, was randomly assigned to hormone therapy alone or hormone therapy plus chemotherapy.

Adding chemotherapy made no meaningful difference for most of that group. Five-year invasive disease-free survival was 92.8% with hormone therapy alone and 93.1% with chemotherapy added, a gap too small to be considered real. The exception was women aged 50 or younger with scores in the upper part of that band, 16 to 25, where a small benefit appeared, possibly because chemotherapy suppressed ovarian function rather than because it acted directly on cancer cells.

Taken together, the results indicated that roughly 70% of women with this type of breast cancer could safely skip chemotherapy: anyone with a score of 0 to 10, women over 50 with scores of 11 to 25, and women 50 or under with scores of 11 to 15.

MINDACT tested the 70-gene signature in 6,693 patients. Its central group was women judged high risk by traditional clinical features but low risk by the genomic test. At long-term follow-up, chemotherapy improved eight-year distant metastasis-free survival by about 2.6 percentage points overall, but the pattern split sharply by age: women over 50 saw almost nothing, around 0.2 points, while women under 50 saw closer to 5 points. The conclusion for postmenopausal women in that group was that chemotherapy could reasonably be omitted.

The score is one input. Your oncologist reads it alongside tumor size, grade, node status, menopausal status, age, other health conditions, and what matters to you about side effects. A score falling inside a published trial band is not automatically the recommendation for you, because those trials had specific entry criteria and your situation may sit outside them. Node-positive disease, for instance, was studied separately and later.

Results usually take one to three weeks after surgery and often appear in your portal before anyone has discussed them with you. A number with no interpretation attached is not yet a decision, and treatment planning in this setting is rarely urgent to the day. Waiting for the conversation where the score is placed next to everything else is reasonable, and writing down your questions in the meantime tends to make that conversation more useful.

• NCI Dictionary - Recurrence Score
• NCI Dictionary - Oncotype DX
• NCI - Breast Cancer Treatment (PDQ) Patient Version