This seems to work given your input and expected output:
Label:[^(]+\([^.]+\.([^)]+)\).*found:[^@]+@\1
Demo[
^]
- Literal text "Label:";
- Any character other than an open bracket, one or more times;
- Literal text "(";
- Any character other than a ".", one or more times;
- Literal text ".";
- Any character other than a closing bracket, one or more times, captured as group 1;
- Literal text ")";
- Anything at all, zero or more times;
- Literal text "found:";
- Any character other than "@", one or more times;
- Literal text "@";
- The same text captured in group 1;
However, your precise requirements aren't entirely clear. For example, does the "found" email always have the domain from the "label" line with just the first segment removed?