Thank you for your input. The MUM-1, used in a panel can deliver valuable extra information. The background of my question was to see which markers for plasma cells are preferred for most specific detection of plasma cells (i.e. in multiple myeloma).
I agree with the other colleages. MUM-1, specially when we are dealing with the diagnosis of plasma cell myeloma is probably the best marker. After optimization, the use of CD138 has some advantage, but you have to take in account some other neoplasmas (e.g. plasmacytoid urothelial carcinoma) can express this marker too
I expect that CD138 is one of the most widely used in routine practice. As mentioned I believe it is not only restricted to plasmacytoid urothelial carcinoma, but even urothelial carcinoma in general and perhaps carcinomas of other sites.
CD138 is the plasma cell marker most commonly used in human tissue diagnosis in surgical pathology. Some cross reactivity is seen with a few other cell types.
I use CD138 too, and you can see this beautiful example in Qiao's Pathology (http://www.flickr.com/photos/jian-hua_qiao_md/8182841171/in/set-72157631991627598).
CA125 for ovarian cancer & alfa-fetoprotein for hepatocellular carcinoma - especially when they are high but PSA , although commonly used, is very tricky and unreliable.
thank you very much for the feedback. In summary it seems CD138 (clone: MI15) is the most frequently used and best documented marker for visualizing plasma cells on human FFPE tissue. The pathologists I have spoken to in the meantime mostly agree, some of them though are still using the Plasmacell Marker (clone: VS38c). This question can now be closed.