TOPIC:

Resistance and Treatment Failure are really the same thing. Failure to clear represents the selection of resistance. The way I explain it to patients is like this:

• You have 1000 soldiers and you shoot them all with a machine gun
• 100 were wearing bullet proof vests so you walk along with a sword and chop their heads off
• 10 "idiots" put their vests on wrong, covering their necks so you drop rocks on their heads
• 1 "idiot" put the armour plate meant for the front of the vest on his head....

Log Kill Rates

Each DAA (Direct Acting Antiviral) drug has what is known as a log kill. A log kill of -1 kills 9 out of 10, a log kill of -2 kills 99 out of 100, etc

Sofosbuvir has a log kill of something like -4.5 which means it kills about 31999 out of 32000 viruses

www.medscape.com/viewarticle/811157

Ledipasvir has a log kill of -3 which means it kills 999:1000

www.ncbi.nlm.nih.gov/pubmed/24320933

Daclatasvir has a log kill of -3.3 which means it kills 1999:2000

Ribavirin has a log kill of only -0.5 which means it kills about 2 out of 3 viruses

www.ncbi.nlm.nih.gov/pmc/articles/PMC3002526/

I can't find data for log kill for Simeprevir (pointers welcome).

Log kill reflects the fact that for any one drug there are typically some survivors and the next drug's job is to "take them out". It makes sense that the next drug should use a different mechanism so taking Ledipasvir + Daclatasvir would not make much sense because either one can inhibit NS5A for GT1.

You don't get entirely additive benefits. Some of the viruses present will be killed by more than one method so the extra killing power goes to waste.

The goal of therapy is to reduce the number of viruses to absolute zero or very close. We know in chimpanzees that 1-10 virions (single virus units) is typically not enough to cause infection but > 100 almost always does cause infection.

www.ncbi.nlm.nih.gov/pubmed/14554088

How resistance happens

As mentioned in Understanding Hep C and the new DAAs the Hep C virus is like a mini photocopier. The thing is it's not a very good one. While it does make decent copies we all know a copy of a copy of a copy can end up blurry. With HCV that means it mutates into many "imperfect" forms.

Almost all of these imperfect copies are defective and die. Some however still work, although maybe not as well as "mum" or "dad" or really "the clone master" that originally entered the body.

In viruses we see rapid Darwinian evolution - survival of the fittest. If I can't breed as fast as you, then you have more clones and before we know it I'm a small minority group subject to persecution.

Now enter a DAA. All of a sudden the playing field changes. Whereas before putting on my bullet proof jacket just so worked great, now perhaps it does not and the small minority groups that mutated a little have a great advantage.

Enter a second DAA - it's job is to kill these ones off.

Enter a third DAA - it's job is to pick off any survivors....

Looking to the future

On first principles we observe with HIV there has been a progression from 1 to 2 to 3 (or more) medications and HAART - Highly Active Anti Retroviral Therapy. Given HIV and Hep C are similar there are probably useful lessons but it is still experimental.

If all of Sofosbuvir + Daclatasvir + Simeprevir were available it's hard to argue that might not be gold class.

If you wanted to go totally experimental Ribavirin has been shown to provide impressive results in cirrhotics with Sof + Dac so HAART in Hep C might end up as: Sof+Dac+Sim+Riba.

The fundamental problem with small trials is the margin of error. At n=100 this is +/-10% so the 96.6% you might read should really be 96.6% +/- 10%. In other words the confidence intervals of the various options overlap making it impossible to be definitive.

A word of warning

All drugs have side effects. For all genotypes there are proven combinations with excellent results. Sometimes less is more so my advice is go with the evidence and take the recommended combination(s). If that means you're cirrhotic and Ribavirin is suggested please take that or if you really don't want to make sure you extend to 24 weeks.

Is it your opinion that treatments for HCV will eventually go the same way as Antibiotic treatments for infections Doc? Will the HCV antivirals keep having to be modified and strengthened?

There is no reason to think it won't, but not for old patients who no longer have an exposure risk.

By definition if you fail treatment your viral load is going to contain a lot of relatively resistant clones. Your treatment just selected them.

Now these are probably going to be less functional but, if you were treated, treatment failed, then you passed that virus on, there is 1 extra person with a relatively Sof/Dac or Sof/Led resistant version of HCV.

This problem will occur in our goals (jails) where carriage rates of 30%+ are common. If say we treated that population for OH&S reasons we know that a number of treatment success stories (from the PEG/Riba days) will come back into jail reinfected. So they are still using unsafe IVDU practices.

If those felons got the virus from the wild it should be the usual sort, but if they were in jail treatment failures and passed it on, now we have resistant stuff floating around in the community in a group that are not being careful.

In Tasmania we incarcerate 1500 people a year. 30% (600) have HCV. Of the 900 who enter jail HCV negative about 12.5% exit HCV positive so about 8% of people passing through our penal system get HCV while doing time.

What we need to do is wipe this thing out now while we have the upper hand. We did it to smallpox but are, for example, failing with Whooping Cough due to falling vaccination rates where once we are winning.

The time to treat this on a mass scale is right now.

Dr Freeman,
I have a question concerning failed treatment & what would be your advise to someone in my situation?

10 years ago I went through the old Peg Interferon/ Riba tx for 24 weeks, as Geno 3. I cleared the virus after first 4 weeks & was cleared all during the tx. 1 month after I finish the tx virus returned/ relapse.

I did not bother to treat since as was waiting for the new DAA drugs to come along, as the old tx was so horrendous & with so many side effects I'd never want to repeat it.

Fast forward to now & am contemplating the generic DAA tx of Sofo/ Dac for 24 weeks & not include the Riba after listening to Prof. G. Foster video concerning the possible "resistance" for people who had Riba in the past, he suggested not to include it with the new DAA.

My fibroscan was 8.7 PKA in July 2015 ( 8.1 in 2005 - so not much damage in 10 years, I expected more), so probably borderline F2/F3 ?? - do you think I could get away with 12 weeks tx Sof/ Dac or better 24 weeks ? without the Riba? - due to possible "Riba-resistance" ?
Other bloods again reasonable not showing any signs of advanced liver disease - so am not able to get any tx in the UK.

Here in the UK patients are still being offered Peg & Riba for Geno 3, some get Peg/ Riba/Sof for 12 weeks if lucky, but waiting lists are very long.
What would be your advise for someone in my situation?

Many thanks in advance,

Jolie

Hello Jolie,

I will start this with a disclaimer - my experience with Hep C the new DAAs only goes back to March, and the entire world's experience is really only a couple of years. Data is lacking and the margins of error in the encouraging trial data is large.

First the guidelines say 12 weeks Sof+Dac is appropriate for you.

Now let's consider some hypotheticals.

Fibrosis is not "you have cirrhosis" or "you don't" but our guidelines deal with the issue of fibrosis in that way because of the need to group people to get statistically meaningful numbers. We know that longer 24 week treatment is often recommended in the context of a high level of fibrosis, but why is this?

My working model goes like this. Fibrotic tissue has a very poor blood supply, therefore getting medications into fibrotic areas is harder. You might imagine fibrosis as like the walls of a castle protecting the areas inside. Fibrosis is a spectrum. You might also imagine that F4 represents strong castle walls made of bricks and F1 represents weak walls, like the straw house the little pig made to hide from the wolf.

Now consider that for an antibiotic/antiviral to work we need to reach what's called the MIC (Minimum Inhibitory Concentration) - below this level it does not work, above this level it works. Like fibrosis this is a spectrum.

Now imagine to kill you I have to make you wet with rain. If I send you out in the rain stark naked it will not take long (low fibrosis), and if it's raining heavily (high doses) it won't take long either.

Now imagine I put some clothes on you. Now it's going to take longer to get you wringing wet. This is like mild fibrosis.

Now imagine I put a cheap spray jacket on you. Now it's going to take longer again and this is like having higher fibrosis.

Now imagine you have a Gortex Jacket. This keeps you dry, but over time water leaks in through the cuffs, around the neck. This is like cirrhosis.

So you could intuit that the optimal treatment times might really look like (total speculation)

F0 8 weeks
F1 12 weeks
F2 16 weeks
F3 20 weeks
F4 24 weeks

Now I just pulled those numbers out of my ear, but I hope you can see what I'm saying. More fibrosis probably needs more treatment and a binary division < F4 = 12 weeks and F4 = 24 weeks does not make a huge degree of logical sense.

Now when we look at treatment duration we know that it's too long.

It's supposed to be too long because we can't know exactly when to stop.

All we can ever know is that we probably stopped a little early if it comes back. Unfortunately this knowledge is only available in retrospect.

So moving on to Government economics. The goal is not maximal cure rates, it is maximal cure rate PER DOLLAR SPENT.

Consider

12 weeks treatment $60,000 (so 8 weeks is $40,000)
Cure rate @ 12 weeks 95%
Cure rate @ 8 weeks 90%

Take a group of 100 patients.

8 weeks treatment will cost $4,000,000 for 90 cures = $44,444 per cure

12 weeks treatment will cost $6,000,000 for 95 cures = $63,157 per cure

Let's assume that we can retreat with 24 weeks @120,000

In the 8 week group that costs another $1,200,000 and sees only one patient not at cure. $5,200,000 for 99 cures = $52,525 per cure

In the 12 week group that costs another $600,000 and sees only 1/2 a patient not at cure. $6,600,000 for 99.5 cures = $66,331 per cure

So the shorter treatment time, although it does not provide the best cure rate is 25% cheaper per cure, even with retreatment.

If you were planning to spend billions on treatment then 25% is a huge number.

When politicians talk about making hard decisions these are the sort of decisions they are talking about.

Hi, I had a splenectomy when I was young (48 now), would that adversely affect treatment using sofosbuvir + daclatasvir?

No, it should not. You do need pneumococcal vaccination, but I'm sure you already know that.

And this is precisley why I have decided to add another 6 weeks to my 12 week sofos/daclats tx. I am geno 1b with a query on my fibrosis but in 2008 it was an F2 so probably progressed to F2/F3 at the very least. Thinking 18 weeks would be spot on for me. I just can't see any use for ribavirin for me and given that I eat plenty last time around (74kg to 1400mg daily) there has to be some resistence. James I will make an appointment with you to get this happening. Em

James-Freeman-facebook wrote: Hello Jolie,

I will start this with a disclaimer - my experience with Hep C the new DAAs only goes back to March, and the entire world's experience is really only a couple of years. Data is lacking and the margins of error in the encouraging trial data is large.

First the guidelines say 12 weeks Sof+Dac is appropriate for you.

Now let's consider some hypotheticals.

Fibrosis is not "you have cirrhosis" or "you don't" but our guidelines deal with the issue of fibrosis in that way because of the need to group people to get statistically meaningful numbers. We know that longer 24 week treatment is often recommended in the context of a high level of fibrosis, but why is this?

My working model goes like this. Fibrotic tissue has a very poor blood supply, therefore getting medications into fibrotic areas is harder. You might imagine fibrosis as like the walls of a castle protecting the areas inside. Fibrosis is a spectrum. You might also imagine that F4 represents strong castle walls made of bricks and F1 represents weak walls, like the straw house the little pig made to hide from the wolf.

Now consider that for an antibiotic/antiviral to work we need to reach what's called the MIC (Minimum Inhibitory Concentration) - below this level it does not work, above this level it works. Like fibrosis this is a spectrum.

Now imagine to kill you I have to make you wet with rain. If I send you out in the rain stark naked it will not take long (low fibrosis), and if it's raining heavily (high doses) it won't take long either.

Now imagine I put some clothes on you. Now it's going to take longer to get you wringing wet. This is like mild fibrosis.

Now imagine I put a cheap spray jacket on you. Now it's going to take longer again and this is like having higher fibrosis.

Now imagine you have a Gortex Jacket. This keeps you dry, but over time water leaks in through the cuffs, around the neck. This is like cirrhosis.

So you could intuit that the optimal treatment times might really look like (total speculation)

F0 8 weeks
F1 12 weeks
F2 16 weeks
F3 20 weeks
F4 24 weeks

Now I just pulled those numbers out of my ear, but I hope you can see what I'm saying. More fibrosis probably needs more treatment and a binary division < F4 = 12 weeks and F4 = 24 weeks does not make a huge degree of logical sense.

Now when we look at treatment duration we know that it's too long.

It's supposed to be too long because we can't know exactly when to stop.

All we can ever know is that we probably stopped a little early if it comes back. Unfortunately this knowledge is only available in retrospect.

So moving on to Government economics. The goal is not maximal cure rates, it is maximal cure rate PER DOLLAR SPENT.

Consider

12 weeks treatment $60,000 (so 8 weeks is $40,000)
Cure rate @ 12 weeks 95%
Cure rate @ 8 weeks 90%

Take a group of 100 patients.

8 weeks treatment will cost $4,000,000 for 90 cures = $44,444 per cure

12 weeks treatment will cost $6,000,000 for 95 cures = $63,157 per cure

Let's assume that we can retreat with 24 weeks @120,000

In the 8 week group that costs another $1,200,000 and sees only one patient not at cure. $5,200,000 for 99 cures = $52,525 per cure

In the 12 week group that costs another $600,000 and sees only 1/2 a patient not at cure. $6,600,000 for 99.5 cures = $66,331 per cure

So the shorter treatment time, although it does not provide the best cure rate is 25% cheaper per cure, even with retreatment.

If you were planning to spend billions on treatment then 25% is a huge number.

When politicians talk about making hard decisions these are the sort of decisions they are talking about.

Thank you for this,
I always had a question about the fixed duration of the tx being either 12 weeks or 24 weeks, why not 16 weeks ? or 18? or 21? so your explanation makes a great sense to me, although I too am not a Hepatologist.

Maybe it has to do with the size of packagings? that they are in packets of 12??? so you can have either 1 or 2 packets ?
But how a size of packaging can determine the length of treatment ?? that's silly, still why 12 or 24 but not 18-21 weeks??

I think adding Riba to the mix would not add anything from what I researched so far including Dr Foster's theory that Riba in tx experienced patients might actually be contra indicated due to possible resistance.

Yes, I get that economic perspective for rationing the tx, piety that human beings, their health & well being is so low on the list of priorities for our Western Gov. political agenda's & very low on the list of priorities for Western Pharma's, where only the highest profit & keeping the shareholders happy is the no1 priority.

The reality is simple.

We know we are giving too much treatment, but we must have a margin of safety.

We can only ever know we did not give enough treatment.

Why not 81 days? I like that number.

I actually give patients 90 days. Why - because the manufacturer had 36 g units of Sofosbuvir and 400 x 90 = 36 g.

Will those extra 6 days hurt - can't see how. Will they help? Well I'll tell you when the SVR12 data is available just after we present it at EASL in Barcelona next year.

Hi regarding the economic perspective for rationing tx, I was wondering how they come up with the milligrams to be taken. Eg 400mg of Sofosbuvir . Is it that this mg is the minimum needed to eliminate the virus or it is no more efficient taking more or is it that its the amount where one has the least side effects and can eliminate the virus? kindly

In the phase 1 trials a pharmaceutical company does the first human testing to establish dose. Typical trial size is 50 people. Hypothetical doses might have been

50
100
200
400
800
1600
3200

And results might have been

50 Fail
100 Partial
200 Works
400 Works
800 Works Mild Sides
1600 Works More Sides
3200 Works Major Sides

So we pick 400 mg as a good dose to get effect, but no sides

Now we move on to Phase 2 trials in 200 people to prove effect at selected dose in ~ 200 people.

Then if that works we do Phase 3 in ~ 3000 people (we know it works) to harden up the numbers and look for unexpected sides.

Then the cash register starts ringing with FDA/TGA approval based on Phase 3 results.

What are the options for a post transplant patient who failed 24 weeks of Harvoni+Ribavirin?

Thank you,
P

Data on this is going to be largely absent but if it was me I would be chasing 24 weeks Sofosbuvir + Daclatasvir + Simeprevir +/- Riba (maybe)

Simeprevir is available on the PBS but only in combination with Interferon/Riba, however if you have a helpful consultant they could probably prescribe it to you and you might choose to not self inject the Interferon and perhaps not take the Riba.

At the worst with S+D+S you would get 24 weeks viral suppression and a good quality of life.

Thank you. The viral load test was repeated and It turned out to be a false positive!

P