Hi Lev, check out this doc page ... it's a relatively simple process but you have to move both S3 segment files and RDS druid_segment records.
There may also be complications if the datasource already exists on the destination cluster, e.g. overlapping time intervals. So it may be safer to first copy the table to one with a new name that doesn't conflict with anything on the destination cluster, then move that "new" table over and insert/select it's data into the final destination datasource, removing duplicates in the process.