Can we get more details? What instrument is this? What rough pump are you running?
~40-50mTorr(with He) is about what I get using an Edwards E2M1.5 on an Agilent 5975. That's consistent with what the user manual says. I've run it on a Varian DS 102 before and could get ~20mTorr foreline pressures(about what I get now on the 5971 using that exact same pump) at .5mL/min.
Personally, I'd try fitting a blank ferrule in the transfer line line nut and then pumping down overnight to see what you get. This will let you isolate the problem to either the vacuum chamber or the transfer line.
If it pumps down okay with a blank ferrule, consider these points:
1. Are you using the appropriate ferrule material(I like graphite/vespal) with the correct ID for your column and OD/profile for the transfer line? Also, if using vespal, be sure you tighten it slightly after heat cycling it.
2. Is the column broken or otherwise damaged somewhere close to transfer line, you can get a pesky vacuum leak. The severity depends on the column ID and distance from the transfer line nut.
3. Make sure you're not using a column outside the specs for what the vac system on your MS can support. As an example(and I only know this off the top of my head because I was in a position of needing to do just this) the 5971 specifies that .530mm columns must be 50m or longer.
4. This one is probably rare, but I got a call a few weeks ago to look at a system at work that I don't normally run/maintain. Someone had gotten heavy handed on installing the column, and the transfer line nut inside threads were stripped(fortunately the line itself was fine-that would have been a $$$ repair). They couldn't get it to get to the vacs they were use to. Fortunately, a new nut got it going again.
If you can't pump down with the blank ferrule, you'll need to look at all the seals in the vacuum chamber.