Tumor Gene Testing: Triumphs And Deficiencies

A post on this website several weeks ago discussed the impact of examining breast cancer genes and how it has impacted treatment. It is important to remember that we are not talking about the presence of gene mutations that one inherits from a parent which are called germline mutations. In this case, we are discussing mutations (genetic abnormalities) that are present within the cancer cell, which are called somatic mutations.

As noted in the prior post, several companies have developed cancer genetic assays. Some evaluate under ten genes while others as high as 70. While there may be multiple tests available, the most used by far are the Oncotype Dx and the MammaPrint. The largest study to date from the University of Chicago compared these two tumor genetic assays.

In this study, the researchers used the National Cancer Database (NCDB). This treasure trove of patients’ cancer information was founded in 1988 and now has over 34 million records of patients with cancer. The NCDB currently captures cancer data from approximately 80% of all new breast cancers treated in the United States.

The authors of the study identified almost half a million women who had estrogen receptor-positive breast cancer. It is these women with estrogen receptor-positive cancers who are candidates for this test. One-third of these women had a genomic assay performed on their cancer. Amazingly, about 99% (approximately 144,000) had the Oncotype DX, whereas only 1% had MammaPrint (approximately 5000). There was a relatively short follow-up, nearly three years. After applying various statistical methods, the researchers found that the mortality predicted by Oncotype DX and MammaPrint was approximately the same. The five-year mortality in the low-risk group for both assays was 5%, whereas, for the high-risk group, it was 10%. Both were equally able to predict which patients could avoid chemotherapy.

Because of the large number of patients in the Oncotype DX group, the authors were able to assess how the test performed in racial and ethnic minorities. They noted that the accuracy of the Oncotype DX test was lower in African Americans compared to non-Hispanic whites. There was a similar trend in Hispanic individuals but not as dramatic as that seen in African Americans. It should be noted that this study was the largest study of Oncotype DX in racial or ethnic minorities.

One should also be aware that the short three-year follow-up is a severe limitation of this study. Whereas the ability of MammaPrint and Oncotype DX to predict five-year survival is unquestioned, it may not be as accurate in predicting long-term survival as other models that exist today or may be developed in the future.

In summary, this is a most extensive study to date of the two most ordered multi-gene tests for early breast cancer. There is no clear winner. This large study confirms the findings of other clinical trials and should reassure that both tumor gene profiles perform well in identifying low-risk patients who are unlikely to benefit from chemotherapy. We have known that tumors behave differently in African American and Hispanic ethnic minorities compared to non-Hispanic whites. Neither test was able to predict or account for this difference. In conclusion, Oncotype DX and MammaPrint have been shown to accurately predict breast cancer behavior in the first five years after diagnosis, but there is still room for improvement.