Researchers have developed a new breast cancer test “that predicts whether or not their breast cancer will return after surgery”, The Daily Telegraph has today reported. The newspaper says the test may mean that thousands of women with a low risk of recurrence could be spared unnecessary chemotherapy.

The story is based on new research that compared an existing method for predicting cancer recurrence, the Oncotype DX recurrence score (RS), and an adapted version that also took other clinical data into account. To test this new method, called the “recurrence score-pathology-clinical assessment” (RSPC), researchers examined long-term study data on 1,444 women with early-stage, hormone-sensitive cancer that had not spread beyond the breast.

The researchers found that under the RSPC model, more patients were classified as being at low risk for disease recurrence compared to the original test. It did not, however, improve the ability to predict which patients would benefit from receiving chemotherapy. As such, the test should be considered “still in development” and not yet ready for use in practice. Its accuracy and ability to guide treatment choices will now need testing prospectively by applying the model to women with breast cancer before treatment and waiting to see if its results later prove to be accurate.

Where did the story come from?

The study was carried out by researchers from the Queen Mary University of London, the Royal Marsden Hospital, the University of Newcastle in Australia, the University of Pittsburgh in the US and the Genomic Health testing company. The research was funded by the US National Institutes of Health, the pharmaceutical company AstraZeneca, Breakthrough Breast Cancer, the Royal Marsden, the UK National Institute for Health Research and Cancer Research UK.

The study was published in the peer-reviewed Journal of Clinical Oncology.

The media generally reported the research accurately. Although the Daily Express reported that the new test could “save thousands of lives”, this is not supported by the research. While the study found an improved ability to classify risk of disease recurrence, it did not report on how this affected patients’ survival rates.

What kind of research was this?

This research compared an existing method for predicting breast cancer recurrence against a new model that takes into account additional factors relating to the disease.

The existing technique expresses the odds of cancer returning in terms of a “recurrence score” (RS), a number between 1 and 100 that classifies patients as having low (

<18), intermediate (18-50) and high risk (>

50) of cancer recurrence. The score is derived by performing gene tests to establish the likelihood of cancer recurrence.

The new model examined in this research combined patients’ RS values with additional clinical data on their age and the size and grade of their tumours. They called the new measure “the recurrence score-pathology-clinical assessment” (RSPC).

The form of cancer examined in the study was “ER-positive” breast cancer (which means the tumour possessed receptors for the hormone oestrogen that had not yet spread to nearby lymph nodes).

What did the research involve?

The researchers developed their new risk-assessment measurement based on the currently used RS measure as well as pathological and clinical factors. They then compared the ability of this new measure to determine the risk of cancer recurrence, as well as the benefit of chemotherapy, compared to RS assessment alone.

To compare the two prediction models, the researchers performed a meta-analysis of data from two previous randomised control trials. Patients from the previous trials were included in the analysis if there were data available on their recurrence score (RS), age and tumour size and grade. The researchers used this data to generate RSPC values for each subject and looked at how accurately they predicted recurrence over 10 years. The researchers also assessed the ability of the new measure to predict the benefit of chemotherapy, compared to the RS alone.

Meta-analysis is a useful method of estimating an overall effect or outcome. By combining studies, this technique increases the number of participants included in the analysis, and thus improves the “power” or ability of the analysis to detect an effect.