Best practice to have different sets of test cases for different environments in Azure DevOps

I will try to explain the setup the procedure and hopefully someone can tell me what is the best approach to achieve my question at the end.

Setup:

We have the following environments for our application: QA UAT and Prod.

QA is our internal test environment whereas UAT and Prod are on customer side.

Procedure:

1. We prepare test cases in DevOps, and then run tests in the QA environment.
2. If all is ok and we are ready for UAT, application is deployed to UAT in order to have a test session (UAT session) with the customer, The test session will run test cases which are a subset of the original test cases which were used during the QA phase
3. Similar is repeated for Prod. except that usually the customer is not involved.

Question:

What is the best practice to have these different test cases sets (QA, UAT and Prod), in order to keep a record of the test runs on each of the environments.

I can think of:

Creating 3 test plans which reference the main set?

or creating 3 test suites?

or creating 3 configurations?

Your help is appreciated.

P.S. Mostly we do manual testing

different sets of test cases for different environments in Azure DevOps

We recommend you create 3 test plan for different environments, for each environments, you can create a test plan and import the existing test cases into that plan. You can also, if you wish, divide the test cases into separate test suites within the plan to enable easier management and monitoring of these separate sets of test cases.

Note: If you copy or clone the test cases. A copy creates a new baseline. Changes to these new test cases don't affect your previous test plans.