I saw a hypothetical derivation where the 20 became 18-19 due to under-accrual, the 18-19 became 17 or so due to not achieving viral response at 4 weeks and being randomized to one of the other treatment arms, the 17 became 15 due to side effects, and the 15 became 13-14 due to other discontinuations (could be believed lack of efficacy or it worked so well they wanted out of the side effects). According to the Bernstein report, the market is then expecting 6-8 undetectables, yielding an ITT SVR of 30-40% and an SVR of completers of 45-55%. I think the numbers in the IHub surver look too high for either measure, but especially (clearly) on an ITT basis. Remember, these are people with only 12 weeks of treatment so significantly shorter protocol than SOC. Also, some of the non undetectables may become undetectables in one of the other arms.
But apparently even this presentation could be muddied. The 12 week responders could include just those who completed the 12 weeks and had the RVR at 4 weeks , it could also include slow responders who got a response only at 12 weeks (but not at 4 - unlikely though as they will still be in treatment in 1 of other 2 treatment arms), or it could also include those who discontinued and went over to SOC and either responded (or continued to respond - again, though, these folks will still be being treated so SVR12 unlikely to be reported). So, a variety of presentation of numbers is possible.
Jon