Breast cancer relapse tests vary in accuracy

Cancer Research UK

Branded tests designed to predict if breast cancer will come back vary in accuracy, according to a new study. 

Four tests differed in their ability to predict if breast cancer would return 0-10 or 5-10 years after diagnosis. 

“This could help patients and their doctors decide on the best course of treatment."  Professor Arnie Purushotham, Cancer Research UK

Lead researcher Dr Ivana Sestak said the results could help doctors decide on the best test for their patients.

“If a woman is deemed high risk of recurrence by any test, the benefit of chemotherapy is greatest. In contrast, if a woman is deemed low risk then endocrine therapy alone will be sufficient and patients could forego chemotherapy and its side effects,” she said.

Researchers at Queen Mary University of London used each test on tumour samples from 774 women, and looked at whether the patients’ cancers returned after treatment.

The women all had the most common type of breast cancer, called oestrogen-receptor positive (ER+), HER2-negative breast cancer. The cancer was at an early stage and had been treated with hormone therapy. 

Which tests were compared?

  • Oncotype Dx Recurrence Score
  • PAM50-based Prosigna Risk of Recurrence Score (ROR)
  • Breast Cancer Index (BCI)
  • EndoPredict (EPclin) 

Each test looks for different genetic ‘signatures’ to predict the chance of the cancer coming back. This could help decide who needs extra treatment to lower the risk of recurrence.

National Institute for Health and Care Excellence (NICE) guidelines had recommended Oncotype DX in the UK, but a reversal of this decision is under consultation. None of the other tests are recommended by NICE or routinely used in the UK. 

The study, published in JAMA Oncology, found that in women whose breast cancer hadn’t spread to their lymph nodes the older Oncotype DX test was the worst at predicting return of the cancer. 

All four tests were useful in predicting recurrence 0-10 years after diagnosis. But only the ROR test, followed by the BCI and EPclin tests, were useful for predicting later stage recurrence between 5-10 years after diagnosis. 

Deciding which patients to treat, and how to treat them

Sestak said that being able to accurately predict the risk of breast cancer returning is especially important as women may be prescribed hormone therapy to stop the disease returning for many years. 

“If we were better able to accurately assess a woman’s long-term risk, then some women may be able to end their endocrine therapy after five years,” she said. “But for those deemed high risk of a late recurrence, continuing their endocrine therapy would be a valuable option.”

All four tests gave better predictions than other tests that don’t take genetics into account. But all tests performed poorly when it came to women whose breast cancer had spread to their lymph nodes. 

Professor Arnie Purushotham, Cancer Research UK’s senior clinical adviser, said the study shows which tests might be most useful in helping to predict which cases of certain types of breast cancer are likely to return, and within what timeframe. 

“This could help patients and their doctors decide on the best course of treatment and whether to extend the use of hormone therapies which are given to try and prevent this from happening,” he said.

Martin Ledwick, Cancer Research UK’s head information nurse, agreed and said the next step will be to work out if we can use these tests to accurately predict who benefits most from ongoing treatment.