So if the F5 spans to both Data Centers, then you are in good shape.
So if the Meraki is down to DC1, but still active to DC2, then the clients will never notice. They will simply connect to the F5 in DC2 since its all part of the same namespace, the F5 will then route to an available Client Access Service running on any Exchange Server it can connect to and that will be proxied to the mail server where the mailbox lives.
https://techcommunity.microsoft.com/t5/exchange-team-blog/client-connectivity-in-an-exchange-2016-coexistence-environment/ba-p/603945
You just need ensure you have health checking enabled on the F5 to mark any server down in the pool:
https://video2.skills-academy.com/en-us/exchange/architecture/client-access/load-balancing?view=exchserver-2019
To ensure that load balancers do not route traffic to a Mailbox server that Managed Availability has marked as offline, load balancer health probes must be configured to check <virtualdirectory>/healthcheck.htm , for example, https://mail.contoso.com/owa/healthcheck.htm.
I would also review F5 documentation. They have alot in regards to Exchange