Hep C Discussion Forum

Hi all,

To eliminate HCV from the world, we need an affordable drug or drug combination with a 100% SVR rate with minimal side effects. We are not there yet, but 80-90% SVR rates are now possible. NS-5A RAVs seem to be the holdup.

I've had some questions about RAV's so I'll do a short post with the latest information. It's a bit technical.

RAVs from the early antiproteases(Victrelis and Incivek) were well documented, with the specific RAVs for each drug and genotype named. These are now not regarded as very significant. Most were not very fit and had a limited survival time.

RAVs from the NS-5A site are very different. We now have a list of these RAVs, for each DAA and for each genotype. There is considerable overlap- for genotype 1a,1b, 2 or 3, many of the same RAVs are found after treatment-failure with either Ledipasvir, Ombitasvir or Daclatasvir. Among the worst of these are RAVs Y93H and L31V/M.

Explanation: The NS-5A complex is an unstable, 447 aminoacid long phospho-protein. It has a terminal helix with three structurally distinct domains. We are only interested in the first 100 aminoacids in Domain 1. This is remarkably preserved across all genotypes. NS-5A blockers bind closely with Domain 1, preventing polymerase activity.  The number for each RAV is the position in the domain and the letters indicate the substitution (caused by the faulty viral replication). Thus Y93H means the amino acid at site 93 has been changed from tyrosine to histidine. For L31V, leucine has been changed to valine.(Sorry, someone asked how these weird names came about). An amino acid substitution as virtually any site can produce a RAV, so there are hundreds or thousands of these RAVs produced. Lets just concentrate on RAV Y93H.

Y93H can be found in Rx-naive patients, and can make up 10-20% of the viral population. This is a very fit mutation, able to enter cells and replicate. It is found in every patient who fails treatment with an NS-5A blocker, irrespective of genotype (some say Geno 1b is less of a problem).  It persists, and can be found 48 months after relapse. It reduces the effectiveness of the DAA by a factor of between 10 and 1000 (or more in Geno 3's). Any monotherapy with an NS-5A blocker has shown a massive selection of Y93H in a matter of days.

The obvious answer is to use a DAA that is effective against Y93H. The newer NS-5A inhibitors, Velpatasvir and Elbasvir seem to have better activity against this RAV, and to decrease the chance of it's selection.  However, if Y93H is found at baseline, Merck suggest treatment should be extended from 12 to 16 weeks and Ribavirin added with it's new drug , Zepatier. The other answer is to use a DAA that blocks a different site and is effective against Y93H.  Sovaldi can be that drug and will work, provided Sovaldi is not blocked by NS-5B RAVs. Merck hope Grazoprevir will do the same, again hoping NS-3 RAVs don't interfere. It is interesting that Ribavirin and Interferon are both highly effective against Y93H, but we don't want to go there.

Essentially, NS-5A RAVs are the reason we cannot achieve a 100% SVR rate. Pre-treatment testing for RAVs has been suggested. Population Sequencing is now pretty cheap, and may be preferable to expensive deep sequencing assays. IL-28B gene testing has been overlooked lately, but it has been shown that those with the TT allele are much more likely to have these RAVs.

If Y93H and the L31 RAVs are present at baseline, treatment may have to be extended and/or Ribavirin added. A nucleoside NS-5B blocker may need to be part of the DAA combo. I'm really looking forward to the new Sovaldi, MK-3682. It should be approved later in 2016.

Does it all matter? Patients may be happy to get their Harvoni/V-Pak or Zetapier and expect an SVR rate of about 90%. It's only when they relapse that questions are asked. This whole article is supposed to explain why that can happen. Hope it makes sense. Cheers.