Power BI Filter/Restrict Load Based on Existing Values

I guess this could be categorized as a cross-post to this, but I'm framing it with different context due to potential confusion with another product being mentioned (in the other thread).

Hoping someone can help.

I have a relatively small SQL dataset that contains keys.  On another server, the same key exists, but the table is huge (and the server is slow).  I need Power BI to use the key in the 1st table and only load records from the 2nd table that contain the key from the first table.

Here is an example of what we need.  It is too slow and often times out (again, the table is huge).

SELECT
  a.uniquekey
, b.metric1
, b.metric2
, b.metric3
FROM ININ.dbo.table1 a
LEFT JOIN LinkedServer.ININ.dbo.table2 b ON a.uniquekey = b.uniquekey
WHERE b.uniquekey IN (SELECT uniquekey FROM ININ.dbo.table1);

I also loaded both tables and did a Merge query w/an Inner Join to a new table.  I can't find a way to systematically drop the unneeded when it's finished.  It's very slow, too.

Is there a way to filter the PBI load based on a field/values in a table earlier in the (same) load?  If not, is there some M code or something to filter the huge table down and drop the prefiltered table?

Thanks in advance for any/all  help.