Do you guys think a cure for Hep B may eventually be achieved with Gene editing? I have watched many potential drugs and the last hopeful one for a functional cure was a NAP by Replicor. Replicor has been quite quiet for quite long now, dont know what’s going on with them. Advances in Base and Prime editing in many areas is quite hopeful. Beam therapeutics has one in the pipeline for hbv. But they seem far off from it.
Welcome to the Community!!!
Replicor is indeed very busy with a number of clinical trials. Its leader, Andrew Vaillant, is an active member of this community and I’m sure you’ll learn a lot about it from him.
As to gene editing. There is a lot of interest in that possibility as conceptually it would work. However, both the technical and logistical/safety barriers to using it to cure HBV are truly enormous. Technical issues include problems getting good enough efficacy and specificity. These are huge issues for an infectious disease that can spread from cell to cell. Logistical/safety barriers are probably even higher. How would we deliver the therapy to every infected cell in the liver? How would we avoid complications of a possible immune response to the foreign editing proteins that might cause excessive immune attack on the infected liver? How could we make it cheap enough to be deployed to people world-wide? Perhaps these problems can be overcome, but certainly not any time soon.
I wish you the best.
You may not be aware that Replicor has been engaged in compassionate use activity with REP 2139-Mg since 2021 with a goal to proving out the safety and efficacy of REP 2139-Mg in very sick patients with HBV / HDV decompensated cirrhosis (no treatment options) and who have failed other therapies for HDV (bulevirtide and lonafarnib). The latest published data can be found here. You can get regular updates by asking to join our newsletter at email@example.com.
We are demonstrating that the ability of NAPs rapidly achieve HBsAg and HDV RNA loss seen in our previous trials is easily and safely achieved in these very sick patients with a weekly and convenient SC administration. Stay tuned for many more patients from many other countries in the coming months. This will be the final exploratory work before we initiate the clinical trials requires for registration.
John has very nicely identified the substantial challenges for a gene editing approach but there is one other important one which is the issue of HBV quasispecies.
HBV replication has no error correcting function and rapidly generates thousands of mutations in progeny virus. These viral quasispecies become established in the genomic reservoirs of the virus (cccDNA and integrated HBV DNA) and their enrichment in individual patients can evolve rapidly with selection pressure. In natural infection, this selection pressure is for immune escape variants and during therapy with ASOs, siRNA, CRISPR Cas9 or ARCUS endonucleases, this selection is for escape variants harbouring 1 or 2 nucleotide mismatches to trigger sequences. In many patients, ASO/siRNA escape mutants have now been shown to be present prior to therapy.
The expected rapid rebound of viremia during treatment or lack of antiviral response with ASOs and siRNA was conclusively demonstrated in animal models and in human trials with ASO / siRNA agents which have no immunostimulatory activity and or are efficiently targeted to hepatocytes (ARB-1467, RO7062931).
Current antiviral responses to siRNA (JNJ-3989 / VIR-2218 / RG6346 / AB-729) are all comparable to each other and occur concomitantly with delivery of dsRNA (a TLR3 agonist) to immunoreactive cells of the liver via their GalNAc conjugation.
Current antiviral responses to bepirovirsen are driven by the delivery of a TLR9 stimulatory motif (present in bepirovirsen) to immunoreactive cells of the liver. In this case, the unconjugated nature of bepirovirsen drives the bulk of accumulation in these immunoreactive cells. Transition of bepirovirsen accumulation to mostly hepatocytes via GalNAc conjugation (the now abandoned GSK3389404) eliminates its antiviral effects, confirming the absence of an ASO effect and the presence of an immunostimulatory effect.
Gene editing approaches (like ASO and siRNA and CRISPR Cas9) all require a sequence specific targeting mechanism which requires 100% fidelity between the guide RNA and the target RNA / DNA in order to have specificity and mechanistic engagement. So while the initial preclincal data from Beam (see here) look interesting, you have to understand these these studies were done in model systems which do not model the genetic diversity present in HBV. These results have to be interpreted with caution.
For example,initial development of many siRNA and ASO approaches for HBV used models with the same lack of genetic plasticity normally found in natural infection. In these models, antiviral responses were profound. However, evaluating these technologies in a model which appropriately models the genetic plasticity of HBV infection (WHBV infected woodchucks), antiviral activity was either absent at the baseline or rapidly disappeared with continued dosing.
Thank you John for the reply.
Thank you Andrew. The data for the compassionate use of the molecule looks promising given the fact these patients are at this stage of illness. I hope your team succeeds in bringing this to the masses and make it even better. Looking forward.
We have been hearing about a new clinical trial for Replicor for several years now. There was one announced by Replicor itself to be held in conjunction with another HIV related scientific organization, trialing the injected form of NAP. Dr Vailant cited that the Covid pandemic caused delays, but I wonder if preparations are now on again and if we can expect a firm date? This should be better than compassionate use, as we do want to see safety and efficacy data from a larger pool of patients. We have known about HBV quasispecies, yet we have successful vaccines, anti-viral drugs, and even monoclonal HBsAg antibodies. I think this speaks to conserved regions of the HBV genome that can be targeted
Indeed there has been a delay in the initiation of new phase II trials and we hope to have these underway soon. Issues in the field surrounding the NAP technology have been driven by a lack of understanding of phosphorothioate oligonucleotide chemistry, how NAPs work, their biochemical effects in the liver and how these impact safety, especially in view of the high prevalence of host mediated flares which routinely accompany NAP therapy (and the establishment of functional cure with other approved agents). These safety concerns in the community are best addressed by deploying REP 2139-Mg in very sick patients with very advanced liver disease (something which is very difficult to do in a phase II trial). The compassionate use of REP 2139-Mg in patients with compensated and decompensated cirrhosis provides a unique safety assessment which more effectively expands the safety envelope and addresses these safety concerns than additional assessments in patient populations we have already examined where liver disease is less advanced. REP 2139-Mg to date is the only antiviral agent other than NUCs which is proving to be safe and effective in patients with decompensated cirrhosis. These compassionate use cases are also quickly demonstrating the efficacy of once weekly SC administration of REP 2139-Mg which will be the mode of administration moving forward. You should look for the data from the compassionate access program to expand to a MUCH greater number of patients from many different countries and well known investigators during the spring / summer cycle of meetings. Of course phase II trials are on the way and if you are not already subscribed to our newsletter, please ask to be subscribed at firstname.lastname@example.org to get the latest news.
With regards to HBV quasispecies, some important corrections are needed to your comments. While prophyactic vaccination against horizontal transfer is highly effective, there is a significant percentage of vaccine breakthrough following birth dose vaccination in vertical (maternal) transfer (up to 30% in some populations) see here, here, here and here. The rapid evolution in HBV quasispecies is the reason for this breakthrough see here and here.
It is important to note that the most recent therapeutic vaccine trial with VBI-2601 / BRI-179 (see here) failed to achieve any reduction of HBsAg despite the induction of anti-HBsAg titers > 1000 mIU/mL, again due to quasispecies escape. This will be the fundamental challenge of all therapeutic vaccines in development for chronic HBV.
It is also important to understand that later generation NUCs such as ETV, TDF and TAF have well characterized ability to stimulate the innate immune response by stimulation of TLR7 by guanosine (ETV) or stimulation of the purine P1 receptor by tenofovir (TDF / TAF) which are well characterized and the effects of which have been observed in clinical trials here, here and here which drive declines in HBV RNA and HBcrAg see here, here and here. This additional effect of ETV and TDF/TAF is the underlying mechnism for the lack of vifral breakthrough which occurred with earlier generation NUCs (ADV and 3TC) which have far weaker immunostimulatory properties.
Recombinant monoclonal antibodies (like VIR-3434) do have immediate reductions of HBsAg but with finite effect, leaving significant amounts of HBsAg present despite the presence of uncomplexed free VIR-3434, even though they are designed to be “broadly reactive”. This VIR-3434 escape HBsAg is certainly a function of HBsAg quasispecies. In addition, IV infusion of IgG is known to alter immune function and this has not yet been considered in the antiviral responses to VIR-3434.
Thank you for your links to articles regarding failures of vaccines against vertical transmission. From my own reading, I think this problem has largely been solved by using anti-viral to reduce the viral load of mothers during pregnancy and timely injection of vaccine and HBIG. I am not aware that failures in preventing vertical transmission were due to HBV quasispecies.
With regard to HBsAg monoclonal antibodies, I quote from a 2022 review co-authored by Prof Lampertico:
“The Phase 1 study VIR-3434-1002 investigated the safety, tolerability, and antiviral
activity of the neutralizing vaccinal monoclonal antibody VIR-3434. VIR-3434 is an en-
gineered human monoclonal antibody targeting the conserved antigenic loop of HBsAg:
its mechanisms of actions include the inhibition of viral entry, antigen presentation, and
stimulation of T cell, HBsAg clearance, and delivery to dendritic cells. VIR-3434 was
administered to non-cirrhotic HBV patients at single ascending doses (6 mg, 18 mg, 75 mg,
and 300 mg): treatment was well tolerated and safe, and induced a rapid HBsAg decline
(ranging from 1 to 2 LogIU/mL) with a nadir between days 8 and 15, with the 300 mg
dose showing the most profound and durable response. At nadir, 23/24 (96%) patients
achieved HBsAg <100 IU/mL, while 5/6 (83%) patients receiving the 300 mg dose achieved
HBsAg < 10 IU/mL .”
As the case of Bepirovirsen demonstrates, efficacy can vary significantly with a larger pool of
patients. We certainly welcome results from large clinical trials.
Thanks, I am already a subscriber to Replicor’s newsletter.
Exactly. The conserved regions of the hbv genome. However i do believe that unlike genetic disorders where the genetic defects occur at only a few known regions(or even one) where the molecules delivered are sure to do the editing, in infectious diseases like hbv or even hiv the initial approach should be to bring down the viral load or even the infected cells by combined therapy like the one by Replicor with or without seroconversion then do a final clean up by gene editing. Even with this a sterilizing cure may not be achieved 100%.
Thank you Andrew for the deep and broad knowledge you bring here. While the delay in initiating the next step is unfortunate, i hope you overcome these challenges. I do believe that people do look forward to more data from your team.
There are many ways to reduce the cccDNA pool. Many recent papers mentioned the fact that liver cells turnover reduced cccDNA. An example is a paper by Thomas Tu: " Mitosis of hepatitis B virus-infected cells in vitro results in uninfected daughter cells"
So in many cases, long time use of NUC can reduce the pool of cccDNA to very, very low or zero. However, the very low cccDNA still may not reduce serum viral antigen loads, such as serum HBsAg, due to integrated hbvdna - so our immune system is still compromised. Here, monoclonal HBsAg antibodies may help to reduce the serum HBsAg load. This then provides a window to apply immuno-modulatory stimulus, such as prophylactic or therapeutic vaccines and interferon, to clear or control the viral infection. I understand that clearing cells with integrated hbvdna may involve killing the infected cells(non-cytolytic removal of integrated hbvdna is not available as it is now part of the human genome?) . So the expected rise in ALT and risks of decompensation must be carefully considered and managed. Gene editing is excellent, but it will take further research to reduce the very risks and improve efficacy.
Yes it is true that long term NUC therapy can result in the elimination of detectable cccDNA while HBsAg (and its immunosuppression) remains persistent from integrated HBV DNA. However, persistent and inactive cccDNA is still present to support the observed rapid rebound of viral infection becuase HBsAg allows the reactivation of inactive cccDNA (see here).
There is misconception about the extent of HBsAg loss required to achieve functional cure. Significant data from NUCs, pegIFN and NAPs show that HBsAg has to be brought below 1 IU/mL to during therapy to have a significant impact on the likely hood to achieve functional cure. HBsAg loss (< 0.05 IU/mL) greatly increases this likelihood. This is confirmed by the failure of combination trials with NUCs + CAM + siRNA to establish functional cure even with HBsAg reduced to 10 IU/mL. Here the limitations of monocloncal antibodies and therapeutic vaccines (due to HBsAg quasispecies) are unlikely to achieve these levels, we will need to see more clinical data to know for sure.
Certainly there is a problem with the way HBsAg responses are presented in clinical trials. Showing log decline from baseline hides the baseline HBsAg levels and the absolute lower level of HBsAg achieved during therapy. If we were to resort to this approach with NAPs, we would demonstrate 5-7 log decline in HBsAg. The average baseline level of HBsAg in chronic HBV infection is ~10,000 IU/mL with only a small fraction (~5%) having HBsAg < 1000 IU/mL. In addition we know that patients with these lower levels of HBsAg respond much more strongly to immunotherapy (indeed the HBsAg loss rate with pegIFN with HBsAg < 500 IU/mL is ~30%). So we need be to careful about the interpretation of data when HBsAg is only presented as log decline from baseline. Good examples of how enrolling patients with low baseline HBsAg can confound clinical trial results are in the bepirovirsen phase II study, where HBsAg loss is confined to patients with baseline HBsAg < 1000 IU/mL and VIR-2218 + pegIFN, where again HBsAg loss is confined patients with low baseline HBsAg. A more significant issue with this latest trial (especially with the use of pegIFN in patients with low baseline HBsAg) is the absence of a pegIFN control. We cannot know for sure but it is likely that the bulk of HBsAg loss was driven by pegIFN and not VIR-2218. We can also not extrapolate the outcomes from these trials to the patient population at large. This is why clinical trials with NAPs have always excluded patients with baseline HBsAg < 1000 IU/mL.
Indeed integrated HBV DNA must be removed from the liver for functional cure, which is why liver enzyme flares are an important milestone for functional cure. In fact, the stronger the flare, the more likely the HBsAg loss (see here). Gene therapy approachs are being developed by some even if the significant safety and efficacy issues can be worked out, we have again the problem of variable integration breakpoints and quaspsieces integration which will always escape targeted gene therapy. W
There is a misconception about the “dangers” of liver decompensation with liver enzyme flares. While this is certainly the case for other liver diseases such as fatty liver disease and alcohol mediated cirrhosis (where these flares are a sign of generalized liver damage) this is not the case for chronic HBV (where these flares are a sign of immune mediated clearance of infected cells). The main exception to this is acute HBV infection or reactivation of HBV infection from breakthrough on older NUCs (which does cause liver decompensation). There is quite a large body of literature on this demonstrating that liver enzyme flares during the natural history or treatment of chronic HBV infection are always associated with improved virologic status and are a critical milestone for true HBsAg loss and functional cure (for a summary see here). This is why HBsAg loss, either historically or with current therapies in development, is always associated with liver enzyme flares.
Thanks!! Very informative!! Appreciate your time here
Thank you for getting back to me.
We are very aware that good results in new treatment methods, including stopping, seem to occur in patients with low baseline serum HBsAg.
That is why everyone is striving to lower serum HbsAg first. Most of the reviews by experts stress this combination approach of NUC and viral antigens lowering as a prerequisite to a functional cure.
I agree ALT flares and decompensation are problematic. Even in a NAP trial, one participant had a viral rebound during follow-up with hepatic decompensation and was placed on TDF therapy. This is a risk that we may have to take after careful consideration in the hand of an expert.
Removing integrated hbvdna is necessary for a functional cure. This is harder than removing cccDNA. But integrated hbvdna cannot produce an HBV virion and can only be removed, at present, by killing the infected cells, with a resulting ALT flare. I hope there is another way to stimulate immune control of HBV infection without first reducing serum HBsAg to almost zero.
All important points but I would add the following:
Outcomes (i.e. HBsAg loss) in trials where these are restricted (or largely restricted) to patients with low baseline HBsAg (e.g. bepirovirsen and VIR-2218 + pegIFN) cannot be extrapolated to the patient population at large. The average baseline HBsAg in >1500 IU/mL in > 90% of patients and HBsAg loss in this group is very rare or absent with these approaches. Currently only NAPs achieve HBsAg loss regardless of baseline HBsAg (even when > 125,000 IU/mL at the baseline).
Decompensation with ALT flares does not occur during therapy with well suppressed viremia, even in patients with decompensated cirrhosis. This is an important distinction to me made: flares during therapy versus flares in the absence of therapy. Hepatic decompensation in the single REP 401 participant you are referring to occurred between 3-6 months after removal of TDF + pegIFN + REP 2165-Mg in a individual which did not achieve HBsAg loss (although his HBsAg did decline from 11065 to 238 IU/mL during therapy). Thus, this participants’ decompensation was unremarkable and responded rapidly to ETV (not TDF). Its is worth noting that 4 other patients experienced ALT flares during the follow-up in the REP 401 study, none of which were accompanied by decompensation and 3 of which self resolved and were followed by partial or functional cure.
I would suggest reconsidering your last statement. Removal of latent (inactive) cccDNA will be the most difficult genetic HBV reservoir of all to target. Elimination of integrated HBV DNA is not achievable by any of the methods currently in development. Latent cccDNA behaves like normal condensed chromatin (DNA) inside the nucleus. It is virtually insoluble and transcriptionally inactive and thus also immunologically silent (unlike integrated HBV DNA, which continually expresses HBsAg as subviral particles). There is no current way to target latent cccDNA (even in the event of HBsAg loss), and this latent cccDNA can reactivate and support rebound of infection. Every patient with functional cure likely has a small reservoir of latent cccDNA. This persistent latent cccDNA is the basis for HBV viral reactivation in patients receiving immunosuppressive therapies like rituximab, even if HBsAg negative for years.
On the other hand, restoration of immune function during HBV infection with HBsAg loss leads to efficient targeting of integrated HBV DNA (in this scenario, HBsAg specific immune function can now locate and remove cells with integrated HBV DNA - which only efficiently express HBsAg). This is why increased likelihood of HBsAg loss on therapy (and functional cure) is higher as ALT flares become stronger during pegIFN and or NUC therapy and why the rate of functional cure with NAPs (when combined with pegIFN) is so high - the high incidence of on-therapy flares. This is also why HBsAg loss has not occurred with out ALT fares in any of the other investigational agents evaluated to date.
Inactivation of cccDNA is the easiest to achieve. It occurs universally with pegIFN and in the the majority of patients with HBeAg seroconversion on NUC therapy (but unfortunately not with any CAMs to date).
Do we know how/when cccDNA “wakes up” and starts infecting other cells? Is there any triggering event?
During the natural history of HBV infection, a resurgence in cccDNA activity is usually accompanied by a flare of virus in the blood (HBV DNA) and a flare of liver enzymes (e.g. ALT) indicating the targeting of infected (and likely uninfected) cells of the liver and reduction in liver metabolic function (decompensation).
Because cccDNA is decorated with the same host proteins as the host DNA (chromosomes) inside the nucleus of the cells, the transition of cccDNA between uncondensed (active) and condensed (inactive) forms is regulated by same host processes with regulate the condensation state of chromosomes. cccDNA can thus switch between active and inactive states; we do not know the frequency with which this occurs but it certainly is a continual process.
The virus has evolved two mechanisms to keep cccDNA as active as possible, these are:
- Production of the HBx protein which decorates cccDNA and acts to promote decondensation (activation) of cccDNA.
- Production of large excess of HBsAg (as subviral particles) which act to inhibit immune processes normally involved in inactivating or degrading cccDNA.
So the triggering event for “activation” of cccDNA is mostly a the absence of a properly functioning immune system capable of controlling decondensation (re-activation) of cccDNA when it occurs.
This is why HBsAg loss is an important feature of functional cure, its absence ensures that the host immune function maintains efficient silencing of cccDNA.