@Zabobon is on the right track, but in my experience it's really common for something like this:
Profile ID:
31340425
To actually be:
Profile ID:<one or more whitespace characters>
31340425
So I tend to use a regular expression instead of literal text replacement:
Search:
(?m)(?<=:)\h*$\R(?=\d+)
Replace:
%Space%
Depending upon the task I frequently preprocess the data to remove trailing whitespace characters, non-breaking-spaces, and other issues.
This practice has saved me more than a few headaches.
When pulling data from unknown sources you also have to watch out for unusual invisible characters – those can really get you chasing your tail trying to figure out why your search/replace patterns aren't working...