Are there any best practices when the documentDB cluster storage limit is exceeded?

You cannot increase storage capacity beyond the documented 64 TiB limit (note this is tebibyte, not terabyte).

As a managed service, DocumentDB automatically scales and handles your storage layer for you; this is part of it's purpose built design and there is no API, SDK, or console command available or needed to interact with the storage layer. Your storage volume is automatically divided into 10 GB segments spread across many disks and is replicated six ways, across three Availability Zones (AZs). This allows DocumentDB to transparently handle the loss of up to two copies of data without affecting write availability and up to three copies without affecting read availability. And since this storage volume is self-healing, data blocks and disks are continuously scanned for errors and repaired automatically.

If you are approaching the 64 TiB limit, consider moving archived, or older, data to another DocumentDB cluster. Alternatively, if you have data retention requirements but are not querying or interacting with these older documents, you could export them to S3, for example, then delete those documents that are no longer needed.