on this answer the trace flag 1448 is advised:
Also, since you are using always-ON with T-Rep, I would suggest you to enable TF 1448.
Trace flag 1448 enables the replication log reader to move forward even if the asynchronous secondary replicas have not acknowledged the reception of a change. Even with this trace flag enabled,, the log reader always waits for the synchronous secondary replicas. The log reader will not go beyond the min ack of the synchronous secondary replicas. This trace flag applies to the instance of SQL Server, not just to an availability group, an availability database, or a log reader instance. This trace flag takes effect immediately without a restart. It can be activated ahead of time or when an asynchronous secondary replica fails.
Reference : Configure Replication for AlwaysOn Availability Groups
However,
on this other answer here, it is not:
I would absolutely not enable this if you care about your replication
setup and data integrity. It allows for uncommitted transactions to be
sent to subscribers which can lead to data incorrectness in the event
of a failure. You'll have to go manually fix your data in the
distributor/subscriber at that point.