Please be advised that today's conference is being recorded. If you require any further assistance, please press star zero. I would now like to hand your conference over to your speaker today, Lauren Glaser, Senior Vice President, Investor Relations. Thank you, and please go ahead, ma'am.
Good morning, and thank you for joining us as we discuss today our company's two presentations at The Liver Meeting, the 2019 annual meeting of the American Association for the Study of Liver Diseases, currently being held in Boston. We are pleased to share the encouraging new data for ABI-H0731 and ABI-H2158, our two lead HBV core inhibitor candidates featured today in two late-breaking poster presentations. In a moment, I will turn the call over to Dr. John McHutchison, Assembly CEO and President, to set the stage with some introductory remarks. Then Dr. Richard Colonno, EVP and Chief Scientific Officer of Virology, will review the data that are being presented. Following the prepared remarks, we will open the call for Q&A.
The slides we will refer to during the call, as well as the posters and press release issued this morning, may be accessed in the Events and Presentation section on the Investors page of the assemblybio.com website. As a reminder, today's conference call will include forward-looking statements, including statements regarding our research and development program, evaluation of interim data, the timing of clinical trials, and therapeutic potential of our development programs, including ABI-H0731 and ABI-H2158. These statements are subject to the safe harbor protections provided under the Private Securities Litigation Reform Act of 1995. Actual results may differ materially from these forward-looking statements due to numerous factors, including those discussed in the Risk Factor section of our Form 10-Q for the quarter ended September 30th, 2019, which was recently filed with the SEC. John, I'll now hand the call over to you.
Thanks, Lauren, and welcome to everyone on the call today. Note that our presentations provide us the opportunity to talk with you all about Assembly's vision and our broad portfolio of core inhibitors, which hold great promise as a central part of future potential curative therapeutic regimens for patients with chronic hepatitis B infection. This transformative potential for the many millions of patients globally with chronic hepatitis B and the quality of science and the teams driving it are precisely what attracted me to join Assembly. As you will learn from our presentation today, we are seeing exactly what we would hope to observe at this stage of development, namely favorable tolerability data in combination with NUCs with once daily oral dosing.
Antiviral activity for the 731 NUC combination is superior to standard of care NUC monotherapy in terms of both DNA and pgRNA reductions using sensitive assays. In the absence of ALT elevations, a subsequent decline in viral antigens, which are surrogate markers of the cccDNA pool and its depletion, which will be necessary to cure patients. We have had an exciting few days here at The Liver Meeting. Positive feedback from researchers, clinicians, scientists, and the broad liver disease community, as well as our industry counterparts. I am even more excited about the year ahead as I believe we are strongly positioned heading into 2020.
Before Rich takes you through the data carefully today, I'd like to briefly set the stage with regard to our hepatitis B programs, including what we have achieved to date, where we are heading and what to watch for in the future, and what we are learning about hepatitis B in the field in general. First, on slide four, I would like to outline our vision to bring the quarter of a billion people chronically infected with hepatitis B worldwide closer to a cure. As with hepatitis C, we believe that combination regimens with complementary, non-overlapping mechanisms will be required to completely suppress hepatitis B viral replication, and different treatment regimens may be needed depending on the patient populations and their treatment history.
To date, both 731 and 2158 have shown favorable tolerability with once-daily dosing, as I said, and we believe that the emerging data for 731 support dosing when combined with the standard of care NUC therapy. Our initial approach to cure includes the combination of a core inhibitor and a NUC. Ultimately, a simple all-oral regimen that is well-tolerated will be the ultimate winner for the many millions of infected patients worldwide. Slide five now outlines our development strategy. Today, we are focused on the first and second waves with our first-generation core inhibitor candidate, 731, that has shown faster and deeper declines in HBV DNA and pgRNA when added to the NUC therapy than with NUC.
The 211 extension study of 731 plus NUC therapy continues currently. Over the course of 2020, treatment will be consolidated and eventually withdrawn to see if viral suppression can be sustained in some patients. In parallel, we are continuing regulatory discussions and shaping our clinical trial plan to advance our portfolio into later stage in registrational trials. We are also getting an early look today at initial data from the first cohort of patients in the phase 1-B trial of 2158, our next-generation core inhibitor candidate, shown as our second wave in the middle of the slide of more potent core inhibitors. During Q1 2020, we expect to complete that study and to begin a phase I healthy volunteer study for our third generation compound 3733.
We are also in parallel looking at the potential to pair the core inhibitor NUC combination with other complementary mechanisms through carefully executed scientific cross-company collaborations. In this fashion, we believe our core inhibitor programs could be central to multiple regimens and programs aimed at finite curative potential strategies. In terms of our clinical objectives, we have always believed that the cure for chronic hepatitis B infection would need to be pursued in a stepwise fashion. Slide six gives a sense of Assembly's approach. The data we have generated from our phase I and II studies to date with ABI-H0731 suggest we have achieved the first three of the objectives as shown on the slide. We will be observing the Study 211 subjects for further viral antigens declines with continued consolidation of therapy, during which we hope to see sustained DNA and RNA suppression.
Finally, treatment withdrawal with both of those events anticipated during 2020. As we work on finalizing our criteria for withdrawal of treatment for the patients currently enrolled in Study 211, we continue to evaluate which biomarkers may be best to predict potential cure. Assembly is focusing on sensitive assays using HBV DNA as well as pregenomic RNA , since these are the primary surrogate markers for cccDNA transcription. Rich will also outline that we are now observing associations with pgRNA declines and subsequent declines in other transcripts of cccDNA like HBeAg and HBcAg. We believe this is novel, only observed with core mechanism to date, and therefore reflects the decreasing pool of cccDNA, a prerequisite for eventual curative regimen.
With that broad and high-level introduction to our approach and vision for what we hope to achieve for patients with hepatitis B, I'll now turn the call over to Rich to review the AASLD presentations in detail.
Thanks, John. Good morning, everyone. I certainly share John's enthusiasm for the data we are presenting today. I'll start on slide seven, which provides an overview of patient metrics, excuse me, for our phase II studies on 731. Today, we are reporting final week 24 data for the NUC-suppressed and treatment-naive patients in Study 201 and Study 202 respectively, as well as interim data for Study 211, the ongoing long-term extension study where all patients received combination therapy. Of the 97 patients completing Study 201 and Study 202, 87 have currently received combination therapy in Study 211 for a minimum of 16 weeks, with cumulative treatment times of greater than 40 weeks for those who have received the combination from day one in Study 201 and Study 202.
To increase our ability to detect very low levels of viral DNA and pgRNA, we developed and utilize highly sensitive assays at Assembly to supplement the commercially available cobas DNA assay. These new assays enabled us to quantitate viral DNA down to five international units per ml and pgRNA down to 35 units per ml. Slide eight lists the demographics and baseline characteristics of the patients enrolled in the original Study 201 and Study 202 studies. Patients were predominantly Asian, and infection was mostly with viral genotypes B and C. Of particular note was the relatively normal range of ALT levels and the presence of high levels of pgRNA in all patient populations. Now turning to the results on slide nine.
In the left graph, treatment-naive patients in Study 202 who received 731 plus standard of care entecavir therapy resulted in a faster and more significant decline in DNA levels in the first 24 weeks of treatment than entecavir alone. In addition, on the right graph, the combination of 731 plus entecavir exhibited a rapid 2-log decline in pgRNA levels, something not observed with standard of care in NUC therapy. We believe this initial rapid phase decline of pgRNA is mechanism-based inhibition, meaning that packaging of the pgRNA is blocked, and pgRNA is simply trapped in the infected cell, unable to be secreted and subsequently detected in the plasma. Importantly, we believe the second slower phase decline in pgRNA over time reflects the decay and actual reduction in cccDNA levels. More on this later. Let's now turn to the treatment of NUC suppressed patients in Study 201 on slide 10.
The objective of this study was to see if the addition of 731 to ongoing NUC therapy could drive the persistent residual viral levels in these patients to undetectable levels or to target not detected. Most of these long-term patients harbor low level viremia at study entry. Using a highly sensitive PCR gel assay, we were able to confirm that a high percentage of the HBeAg-positive patients examined did indeed have detectable HBV DNA at the time of study entry, despite years of prior NUC therapy. The addition of 731 to their ongoing NUC therapy caused 81% of those treated with the combination to achieve target not detected by week 24 versus 0 of 12 NUC patients who received NUC alone.
We view this as an important result as we believe the inability of NUC therapy to eliminate this residual level of infectious virus likely accounts for the inability to cure patients with those therapies. These patients also harbor significant levels of pgRNA, as NUC therapy has little effect on reducing cccDNA levels despite years of treatment. As with treatment-naive patients, the addition of 731 caused a rapid reduction in pgRNA levels, while patients continuing on NUC therapy alone showed no change in their pgRNA levels over 24 weeks. The last two slides represent the final viral nucleic acid results for the initial six-month studies-201 and 202, showing faster and deeper declines in HBV DNA compared to NUC therapy alone, down to target not detected levels, and significant reductions in the levels of pgRNA, the primary surrogate marker of cccDNA, to the lower levels of quantitation.
Now let's turn and take a look at the interim results for extension Study 211. Slide 11 shows that the original 202 treatment-naive patients on entecavir who switched to combination therapy saw immediate and enhanced declines in both their HBV DNA and pgRNA levels. Those treated throughout with 731 plus entecavir combination achieved mean DNA and pgRNA declines of 6.3 logs and three logs, respectively. Most encouraging is the observed acceleration in the second phase decline of pgRNA levels, which we believe reflects reductions of cccDNA pools. When we look at NUCs-suppressed patients in Study 201 who are now in Study 211, this is on slide 12, the switch from NUCs alone to the combination showed an increase in the number of patients achieving both DNA target not detected and in reductions of pgRNA levels to less than 35 units per ml.
Both of these thresholds were only achieved when 731 is part of the treatment regimen. As I mentioned earlier, we believe that the secondary decline phase of pgRNA reflects the decay of cccDNA pools. If true, then the magnitude of that decline should correlate with the decline of other surrogate markers of cccDNA, namely e-antigen and core antigen, and to a lesser degree S antigen, as the latter can also be produced from integrants, which a core inhibitor cannot inhibit. As a reminder, while integrated HBV DNA can produce S antigen, it cannot produce pgRNA or infectious virus.
Stratifying the pgRNA declines achieved by the original 21 patients from Study 202, who are now in Study 211 into 3 categories, greater than 3 logs, 2-3 logs, and less than 2 logs, in the table at the top of slide 13, showed pgRNA reductions of 3 logs or more was highly correlated with higher mean declines in viral antigens. Specifically, 11 or 52% of these 21 patients who exhibited greater than 3 log declines in pgRNA had mean declines of 1 log of HBeAg, 1.4 logs of HBcAg, and nearly 1 log of HBsAg. This was not driven by just a couple of patients, as 82% and 91% of these 11 patients exhibited HBeAg and HBcAg declines of greater than a 0.5 log, respectively. 55% showed greater than a 0.5 log decline in HBsAg.
Lower mean declines in percentages of patients showing a half a log antigen declines were observed in those patients failing to achieve a three-log decline in pgRNA levels. It is important to emphasize that 20 of the 21 patients in this overall group had normal ALT levels less than 40 units per liter, indicating that these declines were due to the activity of ABI-H0731 plus entecavir and not immunologic flares. The graphs at the bottom of the slide show the individual antigen declines for the 21 patients summarized in the table above. The x-axis shows the log decline in pgRNA, and the y-axis shows the log decline in either e, core, or S antigen.
The results shown in the left and middle graph show a clear correlation with pgRNA declines, while the plot of S antigen on the right shows a similar correlation but lower overall declines in S antigen due to the fact that S antigen is partially derived from integrants and not cccDNA. The correlations illustrated in these graphs have R values that are statistically significant. These data offer the strongest evidence to date that direct-acting antivirals can reduce cccDNA levels in the absence of ALT flares, a result we are truly excited about and we think a breakthrough in the field. Moving on to slide 14. For NUC-suppressed patients originating from Study 201, surrogate markers of cccDNA started at much lower concentrations, the dynamic range of decline for analysis is much more limited.
Nonetheless, among the 27 HBeAg positive patients receiving 731 plus NUC for greater than 40 weeks, 18 or 67% were both HBV DNA target not detected and had pgRNA levels less than 35 units per ml. With 10 of these 18 patients also having HBeAg levels less than one international unit per ml. Interestingly, while drops of greater than a half log in both HBeAg and/or HBcAg were observed in several patients, only one patient had that magnitude of decline in HBsAg. This is consistent with recent studies indicating that a high percentage of HBsAg in these long-term infected patients can come from integrants and not cccDNA. Again, these changes occurred without flares as ALT levels in 25 of these 27 patients was again below 40 units per liter. Turning now to safety and tolerability data.
We are pleased that the 731 nuc combination has shown favorable tolerability data with prolonged therapy. Overall, 731 has been administered to 276 subjects in several clinical studies, with 126 patients treated for at least four weeks and 87 patients treated for 12 to 60 weeks. These data further differentiate 731 from other core inhibitors that have faced challenges and limitations in dosing to optimal therapeutic levels. Slide 15 shows final tolerability data over 24 weeks in Study 201 and Study 202, with similar findings observed thus far in Study 211. 731 was well-tolerated when administered with a nuc for 24 weeks. 26 of the 58 subjects reported no adverse events. Of the 32 reporting one or more AEs, all were Grade 1 or 2. No serious AEs were reported. Five of 58 patients receiving the combination reported a rash, 4 were Grade 1, and 1 was Grade 2.
No systemic signs of laboratory abnormalities were observed, and all patients continued therapy through week 24. Slide 16 outlines the laboratory findings for Study 201 and Study 202. Overall, the abnormalities observed were Grade 1 or 2 in severity and occurred in similar proportions of patients across the two treatment groups. With longer-term treatment in Study 211, the nature, frequency, and severity of laboratory abnormalities have been similar thus far to those observed at week 24. Grade 3 elevations in ALT and/or AST have been observed in only three patients treated with the combination beyond week 24. It is important to note that in two patients, the elevations were transient and normalized within four to eight weeks while continuing on treatment.
In the third patient, ALT and AST levels fluctuated during treatment between Grade 3 and Grade 2. At week 52 of the combination therapy, there were no signs of hepatic decompensation, increase in direct bilirubin, or decrease in albumin. Most importantly, all three of these patients remain on treatment. In summary, on slide 17, we are really encouraged by the potent antiviral response, the emerging signs of the cccDNA pool diminishing, and the tolerability data we have seen and are continuing to see for the 731 plus nuc combination in patients treated for 40 weeks or more. The combination demonstrated faster and greater reductions in DNA and, most importantly, pgRNA than nuc therapy alone, with DNA target not detected and pgRNA below 35 units per ml thresholds only achieved in patients receiving the 731-containing combination.
Second-phase declines in pgRNA, the primary surrogate marker of cccDNA, of greater than three logs were strongly associated with reductions in viral antigens, suggesting declining cccDNA pools. 731 has been very well-tolerated in combination with NUC therapy, with no serious adverse events reported to date. In the last few minutes, let me turn briefly to our other poster presentation at AASLD, highlighting early data for our second core inhibitor, 2158, shown on slides 18 and 19. In vitro, 2158 exhibits 10-fold enhanced activity in blocking the virus and also blocking cccDNA formation. In our ongoing phase I-B monotherapy studies in HBV patients, potent antiviral activity has already been observed now in the initial low-dose cohort of 100 milligrams in HBeAg-positive patients treated for just 14 days. HBV DNA and pgRNA declines of 2.3 logs and 2.1 logs, respectively, have been observed.
We believe that the PK profile observed also supports once-daily dosing. 2158 has been well-tolerated, with no AEs or lab abnormalities beyond Grade 1, except for one patient with Grade 3 cholesterol and Grade 2 triglyceride levels that were already Grade 2 and Grade 1, respectively, at baseline. A Grade 1 rash was reported in one patient with no accompanying systematic signs or lab abnormalities. It was transient and resolved within 24 hours without treatment. The trial is ongoing, with the second dose cohort enrolling at 300 milligrams, and we anticipate the study to be completed during the first quarter of 2020. I've spent much of my career working on antivirals for hepatitis B, hepatitis C, and HIV, but curing chronic hepatitis B patients has proven to be the most challenging.
The field is learning more about what to look for on the path to cure, and these latest data give us confidence that we're making solid progress on that path. We are looking forward to seeing further data from the Study 211 in 2020, particularly as patients come off therapy, and to advancing our core inhibitors toward registrational studies and eventual regulatory submissions. With that, John and I are now happy to take your questions. Operator, would you please open the Q&A session?
Thank you. As a reminder, to ask a question, you will need to press star one on your telephone. To withdraw your question, please press the pound key. Please stand by while we compile the Q&A roster. Our first question comes from the line of Geoffrey Porges with SVB Leerink. Your line is now open.
Thank you very much for taking the questions and for the overview presentation. Rich and John, just a couple of questions on your slides. First, on page 11, the second chart of pgRNA levels, it shows a drop-off between week 50 and week 60. Could you confirm that that's just very small patient numbers, or could you give us a sense of how many patients there are in the 211 cohort out at that 50 to 60 week duration? Secondly, in the regression analyses, particularly on the surface antigen, it looks as though that graph is very much driven by the four outliers. Could you talk about the outliers?
Lastly, I don't really know what to make about surface antigen, because you're saying here that if you focus on the sort of 10 patients that have the higher reduction in pgRNA, you're getting a significant reduction in surface antigen. Should we be expecting you to engineer your development strategy and your combinations towards reductions in surface antigen, or should we be sort of ignoring surface antigen? Sorry for the long list of questions. Lastly, could you perhaps comment on the competitive data, which I'm sure you've seen, suggesting that the addition of a core inhibitor to the combination of an oligo and a NUC didn't add anything in terms of benefit?
Okay, thanks for the long list of questions. If I could just start with the first one, the easiest one. Yes, the last time points on those graphs, slide 11, clearly just sort of one or two patients, one, two or three, depending how far back you go, especially the last time point. I would just ignore that. It's really just the decline from two logs to over three logs that we're really concentrating on, and that's the correlation with the data that we have.
Great.
Okay, the second question about This one is a correlation.
Regression.
Yeah, regression data. Okay. Again, these are the 21 patients, and each dot represents the actual antigen level for each of those patients. It's just really to point out the correlation. Yes, you only get these large declines in those patients that have significant pgRNA decline. That's the only statement we're trying to make here, which no one has ever shown before. We've always predicted that pgRNA is the primary marker of cccDNA because there's no other source of pgRNA. It only comes from cccDNA. The fact that when you do demonstrate that you can drop pgRNA levels significantly, we found that significant being beyond the 3 logs, you start seeing these more dramatic effects.
We would anticipate that as patients continue to be treated, and we drive the RNA levels down in more patients, we're going to see these levels of antigens continue to go up. As you can see, the P values there are very strong, you know, 0.001. Clearly, these are highly significant data, even though it's a smaller data set. The obvious correlation, even with the table, in terms of how you see the means and the number of patients that actually even have any kind of significant decrease go down as the pgRNA does not go down, also gives us a lot of encouragement that we are on the right track with this as the primary marker of cccDNA.
Jeff, it's John. Your last two questions of the four you asked.
Yeah.
S comes from two sources. It comes from cccDNA, and it comes from integrated virus, as you know. That explains the difference in the correlations. Our focus will be on antigens that aren't integrated. If we see depletion of the cccDNA core, we'll see pgRNA go down, we'll see E and core go down, and that will be the focus of the way we attempt to predict what will happen and how we will stop therapy in the future. Regarding the competitive data, we were not surprised at all that a short four-week duration of study would show any difference. We didn't see all of the pgRNA data, of course. Given what we have found and understanding the half-life of cccDNA, the need to prevent replenishment, you wouldn't expect to see these things change in such a short period of time.
We were not surprised by that. Thanks for your questions, Jeff.
Great. Thanks for the answers.
Thank you. Our next question comes from the line of Raju Prasad with William Blair. Your line is now open.
Great. Thanks for taking the question and congrats on the data. One question on maybe kinetics of the decline in antigens as it relates to pgRNA. Could you just comment on when you started seeing kind of that depth of response in the patients with the 3 log decreases in pgRNA? Was it 12 weeks afterwards, or was it kind of concurrently with the pgRNA?
Yeah, no, that's a great question. If you actually go to the poster, we didn't go over that data here, and look at individual patients, you can get a sense of that. We have what I call our silver step patient that literally was on entecavir for six months, pretty much flatlined everything for six months, and then as soon as that patient went on to Study 211 in the combination therapy, all the antigen just went over a waterfall and went down very dramatically. That indicates that effect, if the drug is truly effective in that person, okay, whether absorption or just sensitivity that patient has, you can see these dramatic effects in as short as 12 weeks on combination, whereas others would take much longer.
It's got to be an individual patient kind of basis thing as we develop this data, but we have a whole range. The relationship is not time-based, saying that only the patients that are out 40 plus weeks or 50 weeks have these effects. It can actually happen sooner. There's a range of activities, but it's patient specific.
Great. Two more, if I may. Does this pgRNA leading indicator, does that change your thoughts on what you're going to do with ABI-H2158, just given the increased potency there? It'd just be interesting to hear your thoughts on clinical development moving forward with these next-gen candidates. Although your lead program is obviously far ahead.
This is Rich. Yeah, no, clearly our thinking hasn't changed. If you remember way back, if you followed our presentation, we've always believed pgRNA would be the primary predictor of cccDNA. We also, based on also what John said and what I said earlier, the turnover of cccDNA is three to four months. Okay? There's no immediate effect you're going to get by, even if you started to inhibit pgRNA today. It takes a while to decay away what's there. It's not going to actually change how we develop ABI-H2158. It's clearly our primary marker, and we want to be able to drive that RNA down to really undetectable levels.
If we can get it down to the lowest levels possible, which is why we develop these more sensitive assays, then that'll be a key driver going forward in terms of when the next phase, of when we actually take patients off, or we put them into the consolidation period before taking them off. The pgRNA is a primary driver for us.
Raju, it's John here, just to add on. It's important for everybody listening to understand that these second and third-generation compounds have been designed to be more potent in terms of their ability to prevent the generation of cccDNA in a cell-based assay. You won't see that in the 14, 28-day viral decline curves, which is what we're showing partly in the first cohort in our posters. Next year, we've got to dose for longer, we've got to look at the slopes of the RNA declines to make sure we're there, and then we need to look at antigen declines. If our drugs are more potent, those antigen declines should occur earlier if we're shutting off the tap to cccDNA and replenishing it more effectively. Please be aware of all those things that we need to look for next year.
Great. Thanks, John.
Thank you. Our next question comes from the line of Salim Syed with Mizuho Securities. Your line is now open.
Hey, guys. Congrats on the data and also for fitting so much data onto one poster. Kudos whoever committed the poster. Three from me, if I may. One on when you're thinking about the relationship between baseline HBsAg levels and then how many of those patients actually went to E negative or became HBsAg under 100 units, could you maybe describe that relationship, what we know about that relationship so far? Two, on genotyping, it looks like in the poster, genotype C patients have slightly flatter slope in pgRNA. Is that something that you can draw a conclusion from, or are we still looking at genotype C? Lastly, just on safety. It looks like the three patients had ALT/AST elevations. It looks like they're uncommon. Was that due to the core inhibitor or the NUC, do we know?
Also, what does that say about safeties for the core inhibitors, given you guys have used novel chemistry? Thanks so much.
Okay. Salim, why don't I start with the safety first, then I'll let Rich take the other two. We were careful describing the three Grade 3 ALT elevations as Rich described to you today. We thought they should be included. They're in the open label extension study of the trial. Just recall from the graph we showed you, in the first 24 weeks, the rates of Grade 1 and Grade 2 elevations of ALT were actually than the Nuc arm, and there were no Grade 3 elevations.
Hello?
induced liver injury, unless it's rare and idiosyncratic, occur within the first 8-12 weeks. These observations beyond 24 weeks are isolated. One was ALT only, the other was AST only, the other was fluctuation. I believe they represent the natural history of fluctuating disease as we continue to observe these patients continually. The patients all continue on treatment. There's no bilirubin elevation or evidence of any hepatic impairment. This is highly un- drug-induced liver injury occurring late in the development of the program. Rich, I'll let you answer the other question.
Sure
about baseline surface and genotypes.
Right. In terms of the genotype, I can go backwards. I'll do reverse order. Again, genotypes, these compounds are. They will work against all the genotypes. We've never found any differentiation between any particular genotype in terms of its ability to inhibit. I think you're just looking at a very limited data set, and so I wouldn't isolate one patient and feel that that's representative of some genotype of HBV. In terms of S antigen baseline levels correlating with anything, no, there is no correlation whatsoever with baseline. Pretty much all the patients have a fairly tight range, unlike the other antigens, a pretty tight range of between 1,000 plus upwards to almost 50,000 to 100,000. There's that couple of wild. There's no real relationship with which patients drop the most, et cetera, based on their baseline.
Small numbers of patients here. Making inferences about different genotypes and different responses and everything is really premature, I think. We need larger data sets and more information, as is usually the case with early development.
Yeah. It's really about that group of 21 patients, which is pretty dramatic, okay, when half of them have such a dramatic response. The correlations are so strong. Again, it's really the greatest example we have of this correlation between pgRNA really being reflective of cccDNA pools and having the confirmation that the antigens go down in parallel with the pg. We always said that there are three or four surrogate markers of cccDNA, and that they should all go down in some kind of parallel fashion. If you had one go down and not the others, it would make little sense. The only exceptions to that and the complexity are depending on the patient, and we have different examples in the poster.
Even one of the individual patients we put up there had quick drops in all the other markers, S-antigen actually didn't move in 24 weeks in that particular patient. Versus another patient, we can get 3.5 log drops in S-antigen because that patient probably has most of their S-antigen from cccDNA. I think as we continue to treat patients, we're going to see this variety of responses on S. That's why we feel S is less predictable of what we're really trying to achieve, and that is to try to exhaust the cccDNA pool.
It's John again. I think it's complicated, the data sets for everybody. I think the important thing Half of them have a three log or greater decline in pgRNA. These declines in pgRNA we've shown you in the randomized 24-week first parts of the study, don't occur with the nuke. This isn't observations that you would expect to see in any other sense. The only source of pgRNA is cccDNA. It's actually a very beautiful scientific story that's developing here so far. We're excited about that. We're going to play it out next year. We're going to consolidate. We're going to take people off therapy after a period of consolidation and see what happens, see if we can prevent relapse of HBV DNA.
Okay, excellent. Thanks so much, guys, and congrats again on the data.
Thank you.
Thank you. Our next question comes from the line of Michael Yee with Jefferies. Your line is now open.
Hey, guys. Thanks. Congrats, John and Rich, on the updated data. Two questions. On Study 201, can you just maybe characterize how you're thinking about the use of this data as the development for the next stage? I know you talked about consolidation and all that for next year. What in your opinion would be the perhaps response criteria to take people off? Is that patients under pgRNA less than 35 units? Maybe just talk about how you're thinking about what the response criteria would be or be a time-based criteria. That's question one. Question two is for Study 202. Same situation. How are you thinking about what you see between the difference in combination versus monotherapy and then what happens when you switch people onto combo? Again, is that some sort of time-based therapy?
I know there was discussions with FDA about developing something there and using DNA, et cetera, as an endpoint. Maybe just talk to what that result means for the next steps in that type of population. Thanks.
First, Rich?
Okay, sure. This is Rich. I think, again, the objective is to get into the next stage, the 4th stage, if you will, of this consolidation. What do the markers have to get to before we decide we're going to put them into consolidation mode? Again, consolidation mode being we really can't see anything else happening. Why we continue to treat to make sure we get down to even lower, get rid of those last kind of molecules. The criteria for both patients or all patients for the most part, is pretty much the same. We expect DNA to be target not detected. We have to show, at least convince ourselves that the patient has reached a point where there's no more active virus that we could detect.
Target to have the RNA, and we're actually working on another assay, and we'll say target not detected in terms of RNA. If you can truly show that you have no detectable DNA, so no viral replication, no detectable RNA, pgRNA, so therefore, no evidence that cccDNA is still there, I think that would be a baseline sort of trigger for a consolidation period. We are having internal discussions on exactly whether we should add something else to that, what we call double negative, whether we throw in as an E-antigen level or whether there's some other antigen-related level. Those are still for us. Those ground rules will apply to all patients. A patient antigen negatives in 201 are actually most advanced in that entire process. We don't talk about them because there's really nothing to talk about with those patients.
They are HBeAg negative. Their HBcAg is basically barely detectable. Their pgRNA and HBV DNA in most of those patients are already at this very, very low level of even our assays. For the most part, one can think mentally, those are actually already in a consolidation mode, and those would probably be the first ones that we would take off and see if there's a virologic rebound. Working backwards, we have the HBsAg negatives, highly suppressed. We have the Study 201 patients which were suppressed coming in. Again, everything as I showed you, much lower, and a lot of those patients are already triple negative. Behind that is the patients that are further behind but are probably the most exciting for us because of the dynamic range of all the markers.
It's the original patients that were naive that we've already brought down now to where patients in 201 started. All of these patients, and that's the beauty of our phase II design, we have a variety of patients. They're all at different stages. They're antigen positive, negative, suppressed, unsuppressed. They just give us a wealth of data in terms of following the progression of all these patients, but they all get to the same point at the end of the day.
Hey, Mike, it's John. Can you hear our response? I heard that the line's cutting in and out. Can you hear what Rich said?
We hear about 80% of it. It does kind of break up a little bit, but we hear about 80% of it. For my follow-up question for both.
Sorry about that.
That's okay. My follow-up question is that based on all the data you just said and all the criteria, does the second-generation drug for you just philosophically get more patients or a higher % of patients to this criteria in the same amount of time and/or just faster? That's the point of that molecule.
That would be the hypothesis, that we can do this more efficiently in more patients more rapidly.
Yes
by shutting down cccDNA generation more effectively. Correct, Rich?
Yes, correct.
Yes, that's right, Mike. Look, stay tuned. When we have the stopping criteria approved and agreed upon, we can talk about them, but nobody's stopped yet, and we're in the process of actively working and thinking about this right now. It's fairly advanced.
Perfect. Thanks so much.
Thank you. Our last question comes from the line of Brian Skorney with Baird. Your line is now open.
Hey, thanks, guys. Porges asked most of my questions. Maybe help throw a couple final ones in there. Just in terms of when I look at the individual patients in the poster and you kind of brought up that one patient who saw a decline in everything other than HBsAg. How do you think about if you're shutting down all the viral markers and you are seeing hepatocellular turnover, why wouldn't HBsAg at least start declining in that patient? Just out of curiosity, is there any marker you could use for hepatocellular turnover that's non-virally related, just to kind of try to observe and measure what the timeframe would be to ultimate turnover of hepatocytes here and if it varies patient to patient?
Well, S antigen becomes the marker for that turnover because if, like I think we believe in that one patient, because it's a complete disconnect with the other markers, if that antigen is just not going down very rapidly or if at all, then I think one can imagine that in that particular patient, there's a high percentage of integrants providing that S antigen. As we continue to treat, you can remember, as John said, we haven't taken patients off therapy yet. If therapy goes another year overall, by the time we get down to the levels we want to get to and then consolidate for at least six months, then by that time, we would expect that actually turnover of hepatocytes is also taking place. If we truly have prevented new viral replication and new cccDNA synthesis, we should also prevent new integrants.
Okay.
The existing integrants should die off over time. That's why I think in that particular patient, for instance, and there are patients like that, had no effect on S. That presence of S is actually our surrogate marker for cell turnover, because the only way core inhibitors can turn that over is by the cell dying. We stop new cells having integrants, and the cells that have the integrants have to die out. We just have to follow that over time. We can't answer your question directly. We don't have any marker that we follow specifically. One would anticipate that S antigen will eventually decline, but through a different mechanism, and that mechanism could catch up in terms of time, but we just have to see.
One could envision, okay, at the end of the day, that you could literally get to a person who has no more viral replication going on, cccDNA levels, if it's possible, have disappeared, and they could still be expressing S. That patient is not going to be infected by the S or influenced at all. It's going to be a scar that you were infected with hepatitis B. S is not the driver of chronic infection. It is a symptom. It is a sign and a diagnostic tool, but it certainly isn't the driver of infection. The virus is. Our target still remains on the virus, on the cccDNA, and we'll just have to follow and see what happens with S antigen.
Brian, it's John. Just to add to what Rich said, the half-life of an infected hepatocyte is unknown. I've actually been asking a lot of people for their answer to that question, and I get dumbfounded looks.
Yes.
It's days to weeks. You would expect it. To answer your question, as Rich said, if you completely shut down viral replication, over time, surface antigen should decline through loss of infected hepatocytes. How long that takes, nobody knows. It depends on the half-life. There we are. Just recall, Brian, that you don't need to show clearance of surface antigen to reach the regulatory endpoint of a finite therapeutic cure. Absence of DNA 6 months after cessation of therapy, SVR24 for HBV DNA is an approvable endpoint without change in surface antigen for a hepatitis B therapeutic regimen. That's an important point of distinction.
Great. Thanks, guys.
Thank you. That concludes today's question and answer session. I would now like to turn the call back to John McHutchison, CEO and President, for any closing remarks.
Thank you to you all for your time and attention today. We're very encouraged by these latest data for our lead core inhibitors. This is particularly exciting for me because I chose to join Assembly to tackle the challenges of hepatitis B and to move the field forward. I think we're doing that now, and we're showing you some evidence of that. We are now seeing that evidence that we are heading in the right direction, and the combination of ABI-H0731 plus the NUC has been well-tolerated and has shown the ability to eliminate residual viral replication. Some patients achieving DNA target not detected with sensitive assays, more than 3 log or multi-log reductions in pregenomic RNA , and antigen decline suggestive of a diminishing cccDNA pool. These are significant first steps, and obviously there's much more we need to do.
The extended treatment duration study 211 of ABI-H0731 plus the nuc continues. During 2020, as we've said today, we will consider patients for consolidation, and then they'll eventually come off therapy so we can assess whether viral suppression is sustained. We expect to provide additional updates from this study next year. During the first quarter, we also anticipate completion of the phase I-B trial of ABI-H2158 and the start of a phase I for ABI-H3733, our third core inhibitor candidate. We believe Assembly is in a strong position as we head into 2020. We're excited about the prospect of advancing our core inhibitor portfolio into later stage trials. Thank you for joining today's call. We look forward to communicating our progress in the new year.
Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.