Extend existing cube while still making use of pre-aggregation

From a organisation/readability perspective, it would be useful for us to extend an existing cube to separate out measure/dimensions specific to that subset of data.

Problem is, when using the new cube (which makes use of extend), we can no longer rely on the pre-aggregations created by the cube which is being extended. Even though 90% of the data is similar.

Is there a way that we could separate this data for readability while still being able to make use of a single pre-aggregation?