Molecular Methods for Classifying Breast Cancer Patients Produce Inaccurate and Inconsistent Results
Tuesday 23 February 2010 - New research suggests that recently developed molecular methods of classifying breast cancer subtypes may be unsuitable for selecting patients for clinical trials.
Much research has been carried out over the past decade into tests for molecular subtypes of breast cancer – basal-like, HER2, luminal A, luminal B and normal breast-like – based on expression of genes. It is hoped this system of classification will be used to help select the best treatment for patients based on the specific molecular characteristics of their tumour, so only those who would benefit receive the drug and others avoid unnecessary side effects.
Three molecular tests for the identification of these subtypes have already been developed. It was assumed that these different methods would produce similar results, but the new research shows that different tests to identify the same molecular subtypes of breast cancer produce inconsistent and widely variable results.
For example, the researchers looked at samples from patients known to be HER2 positive by current methods. One of the molecular tests found half the patients not to be of HER2 subtype, while two others classified all the patients into different molecular subtypes. Therefore, if these new molecular tests were used on patients, there would be a significant risk they would not receive the treatment they need, such as Herceptin (trastuzumab).
Dr Jorge Reis-Filho, from the Breakthrough Breast Cancer Research Centre at The Institute of Cancer Research (ICR), says doctors should continue to use the current classification system, which is based on oestrogen receptor and HER2, while the tests are further refined to ensure they are accurate and consistent.
“Until tests for molecular subtypes of breast cancer are standardised they should not be used in a clinical setting,” he says.
