"Can you have a interim efficacy analysis on a blinded trial?"
Yes.
Indeed I'm sure all true IPIX aficionados will recall that the object of their affections did precisely that back in 2017 with the double blind study of brilacidin in OM:
"Cellceutix is encouraged by the interim data from the OM trial (a double blind, placebo controlled study), given the fact that there are no FDA-approved drugs for the prevention of OM in the study population. According to the preliminary interim results from the trial, only 2 of 9 patients (22.2%) on Brilacidin-OM therapy developed severe OM, defined as Grade 3 or Grade 4 on the World Health Organization (WHO) OM Grading Scale. Patients in the placebo-treated control arm developed severe OM at a much higher rate, with 7 of 10 patients (70%) progressing to that condition. There are now an additional 11 patients active in the trial and more are added weekly."
DSMB and statistician unblind to look at results, but usually keep these results secret to maintain blinding, unless they are fantastically good or bad.