In that case the cell size is the biologically meaningful part. We could pay more attention to size before making interpretations.
I like the idea of looking at relative upregulation of transcripts since it fits well with gene signatures and cell annotations derived from scRNA seq
