Forum:Why did expression based subtypng of breast cancer gain much more acceptance than others

Question

[CY]

This is may not be entirely technical question but rather a academic question. But the technique behind the application is within the scope of bioinformatics. So I would try to ask here that:

In each cancer type, there have been tons of papers that describe RNA-Seq based subtyping resulting in molecularly distinct subgroups as well as survival difference. However, only very few of them gained acceptance at clinical setting, such as breast cancer (luminal subtype A, luminal subtype B, normal breast-like subtype, HER-2 overexpression subtype and basal-like subtype).

My question is: why did this molecular subtyping of breast cancer gain much acceptance but not others? Did this one own some advantage that other did not?

Answer 1

[Kevin Blighe]

Hi, I moved this to a Forum post.

Breast cancer, being a leading cancer in women in terms of mortality (behind lung), is obviously going to have a strong research focus as a result of this. I am not sure that there are many adults who do not know a person who has died as a result of complications arising from breast cancer.

The molecular sub-types that you mentioned were originally defined all the way back in 2000, I believe, by Perou's landmark paper, closely followed by the work of Sørlie:

• Molecular portraits of human breast tumours
• Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications

These molecular sub-types have been consistently verified by multiple other studies and they align well with immunohistochemistry; so, they bring with them a very relevant clinical angle. Others have since attempted to define copy number sub-types of breast cancer (e.g. Curtis / Caldas), but these never quite stuck, possibly due to a lack of concordance with clinical info.

Each cancer is different and, in breast, we can say that these are the true sub-types and that they have been identified and verified by much research. In other cancers, molecular expression sub-types are actually less relevant, for example lung and endometrial.. In lung, we could say that mutation signatures are more relevant as these align better with the clinical data; whereas, in endometrial, somatic copy number and mutation signatures align better [with clinical information].

You also may be surprised to learn, however, that clinical diagnostic / pathology labs are not regularly screening for these molecular sub-types. IHC is still very much the de facto way to screen for a breast tumour sub-type.

Kevin

Answer 2

[GenoMax]

In addition to years of scientific support noted by Kevin it may be due to Food and Drug Administration (in US, equivalent agencies elsewhere) approval for these tests (also insurer support for reimbursement). Your question led me to this page at NCBI which summarizes genetic tests for various cancers. While there are exceptions most other cancer's have FISH/cytogenetic tests compared to breast cancer.