"drop the loser" or futility in crossover trials

#1
0 down vote favorite

I wonder if anyone can advise me if there are methods (or even a concept) of dropping losers for futility in crossover trials For exampe if I have 4 treatments (A,B,C and D - and maybe all 24 possible sequences in my crossover) then if I did an interim that said "drop B" what would I do to patients who had been randomised to an sequence but who had yet to received their "B" period eg a patients on CADB who was halfway through the "A" period when it was decided that B be dropped. many thanks