Potentially wrong 'Maps to' ER-positive Breast cancer

[pxp9]

Describe the problem in content or desired feature

The meaning of ER-positive Breast cancer concept which have the code is Estrogen receptor positive in a Breast cancer while HER2-positive means that your cancer is positive in the her2 protein.

I was expecting a map to something like this concept

How to find it

ER-positive Breast cancer: Athena link HER2-positive carcinoma of breast: Athena link

Just query

SELECT * FROM omop_cdm.concept_relationship WHERE concept_id_1 = 42542266 AND relationship_id = 'Maps to'

Note that my cdm instance is in omop_cdm postgres schema.

Expected adjustments Change Maps to relationship to this concept

Screenshots

[vladkorsik]

Hello, @pxp9 thanks for your interest.
Yes, it's a bug to be fixed, for sure. Currently, the HemOnc vocabulary is not supervised by Vocabulary Team but it's likely to be managed by OncoWG members, I know that @mgurley is working on it. As HemOnc is outside the General RoadMap Vocabulary Team will likely be involved in the final integration and curation steps.

However, there are some things to be discussed:

1. Do you really have HemOnc as a source terminology in your data? I just want to understand how the community uses HemOnc.
2. The mapping you've chosen as a target is lexically matching but only partially semantic. Loss of cancer topology (Breast) will occur as neither the name/synonym nor the hierarchy of this concept provides relevant information on cancer topology. In general, we expect to have these entries post-coordinated in vocabularies represented as pairs of Cancer Diagnosis(Condition) + Biomarker modifiers (Measurement + Value). But if we look at 'ER' another way, we may find that the acronym may not only stand for estrogens but for all endocrine receptors (including Progesterone) - the issue of logical OR we can not resolve with mappings that easy.

Feel free to continue the discussion either here or at the OncoWG Vocabulary or at regular Vocabulary WG meetings. The WG sessions with properly announced agendas are the best places to keep most of the community members informed about decisions

[pxp9]

Hello, @vladkorsik

My source data is structured with NLP notes, so the fields in the tables are a convention that we have done. I am not using HemOnc as source terminology.

In my case, er_positive is a field in my tumor diagnostics table. The field means if the tumor was marked as estrogen receptor positive and when I was looking for a proper mapping in Athena, it suggested me the HemOnc concept then I saw that it was not standard and I looked at the Maps to relationship.

I think the meaning of HemOnc concept is a little bit ambiguous (I supposed that ER means Estrogen Receptor but maybe it means endocrine receptor as you said) so that is why I suggested that mapping but maybe there should be a better one.