Meta-analysis with 2 GEO transcriptomic datasets possible?

Question

[Dr.]

I could extract only two NCBI Gene Expression Omnibus (GEO)1` Data Series (GSE IDs) for my specific research query on stroke.

Can I pool those 2 datasets for meta-analysis?

As no more datasets are relevant for consideration, is a meta-analysis with 2 datasets valid in this scenario?

[ Please mention any References (book, article) that I can cite to support meta-analysis with 2 datasets.]