I find the whole "wait for the RAVs to become un-measureable" logic flawed.
Accepting the observation that some RAVs become unmeasurable over time, to me the question is "does that matter?"
Go with me on this.
Say you have no measurable RAVs at baseline, and treat, but fail. What does this mean? It means you MUST HAVE HAD RAVs at baseline, despite the fact we could not measure them, and this is why you failed.
Post treatment we may be able to see the RAVs (because we have killed just about everything else) but we need about a 10% population to see them. Now consider this:
Starting viral load 1,000,000
Limit of detection 10
Almost all patients become undetectable (99.5%) but 5-10% fail to SVR. It follows that when they are undetectable they have < 10 RAVs, or to put it another way from that initial population of 1,000,000 there were only 0.001% RAVs
Conjecture - generally RAVs are at such low levels we can't measure them, despite the fact they exist.
Now with biological systems we see decay in terms of 1/2 life - the time taken for something to fall by 1/2. If it takes say 1 month for RAVs to fall by 1/2 it will take 2 to fall to 1/4, 3 to fall to 1/8. Even at this rapid rate it will take 20 months for 1 million RAVs to decay to 1 RAV (2**20 == 1 million).
To me one of the best reasons to delay retreatment is to wait for better, more potent drugs and particularly triple and quadruple regimens.
If this was HIV nobody would suggest waiting to RAVs to decay before changing regimens but prices are different and there are more options about that.
The first of my originator Harvoni failures are coming through at SVR12 and 24 following 24 weeks generic Sof+Dac+[Simeprevir or Asunaprevir]+Ribavirin. The Simeprevir was originator out of Egypt, and the Asunaprevir originator out of Russia
The combinations of Viekira-pak+Sofosbuvir and Zepatier+Sofosbuvir, both +/- Ribavirin are also looking good with patients in the USA using originator V-pak/Zepatier and generic Sofosbuvir.
www.aidsmap.com/Treatment-intensificatio...atment/page/3014995/