It would be very handy if it could be implemented.

From the PCRE man pages:
"If you compile PCRE with UTF-8 support, but do not use it at run time, the library will be a bit bigger, but the additional run time overhead is limited to testing the PCRE_UTF8 flag in several places, so should not be very large."

The u modifier to pass the PCRE_UTF8 flag could be added (same modifier used in PHP).