Can I use a CSV file in a collection runner and only run specific requests multiple times?

I’m writing some tests to set up in a pipeline.

For my tests I would have to upload a number of documents. Then I’d want to see if they can be downloaded and get the correct status (isDownloaded = true/false). And then delete them at the end. I could use a CSV file to loop through the different types of documents I want to upload.
But the microservices I’m testing also have some other functionalities I’d want to test.

When I put all my tests in one collection and add the CSV file, all requests are run for every iteration, but I don’t need the data from the CSV file for every request. So I want the CSV file to be used for a part of the requests and I don’t want all the requests to be run multiple times, since for some requests nothing will change.

Is there a way to define the requests so that they are only executed during the first iteration? Or do I have to create multiple collections? Or is there another solution?