Hello, everyone. My name is Harry Jensen. I'm one of the moderators of this talk together with Dr. Bob Gish, who just turned on his camera. And, Bob, I hope you can introduce the webinar and the speakers. Thank you very much. Thank you, Harry. Good day to everyone who's joining us for this important webinar about the use of quantitative S-antigen in clinical practice. As you know, in the U.S., this test has been available through PLEST for four years and, most recently, through LabCorp. I think some of the other reference labs will have this available. It's been available around the world clinically in many countries for up to two decades. There's over 200 peer-reviewed publications about the application of quantitative S-antigen. Many of these are applicable to clinical practice. We're planning to review those today with our wonderful speakers. The conversation about quantitative S-antigen with the FDA started in 2012. Bob Perillo, I, and Tim Block met with the FDA twice and also had dialogue with Roche and Abbott about taking these tests forward for FDA clearance. But it ended up that these are LDTs that are brought forward by reference labs. We can talk about how that availability is in Canada and some of the other countries. We would like to seek updates to the guidelines for ASLB and APOSL and some of the other guidelines that are out globally we think would help guide clinical practice. We're going to review that information today. I'd like to introduce our first speaker and thank ASLB, Romney Clayton, and the team for helping us put this together. This is a brief outline about S-antigen quantification, influencing variables, natural history, antiviral therapy. And at the end, we'll have a Q&A session that Carrie Johnson and I will co-chair to bring your questions to the forefront. Our first speaker, Dr. Marcus Kornberg, is a professor of infectious disease and deputy director in the Department of Gastroenterology, Hepatology, and Endocrinology and works at the Center for Individualized Infection Medicine and a clinical director at the Hemholtz Center for Infection Medicine. Marcus, we look forward to your first presentation. Thank you for joining us, you're a global expert in this area. Thank you. Thank you very much for the introduction. It's a great honor to talk to you about the HPS antigen quantification. And I will also give you some data on also influencing variables, which may be important for the interpretation of the data. This is my disclosure slide. Now first, I would give you some insights about the molecular virology of HPS antigen to understand the meaning of S-antigen and why it will be important for the diagnostics. So first, I also would like to bother you with the life cycle of the HPV virus to understand where HPS antigen is coming from and what is the meaning of this. Now, the template or the template of S-antigen are several RNAs, the so-called subgenomic RNAs, which are coming from the ccDNA. And the subgenomic RNAs are independent from the reverse transcription, so independent also from nuc therapy. So that means that these RNAs are translated in the end to protein to the HPS antigen. And this is independent of the viral replication of the reverse transcription. This is important to understand also the data when Milan Sonnefeld will give his talk on the treatment and the S-antigen data. And a particularly important feature of HPS antigen is that it's produced in such high amount. It exceeds also the viral replication by far. So for example, in a high variant patient, if you analyze one milliliter of blood, you will see 10 to the nine variants. So the variants, HPS antigen containing variants, but 10 to the 10 filaments and 10 to the 13 spheres, which are sub viral particles, empty, not containing virus. This is important and that makes it highly sensitive also for the diagnostics. An important feature, and I will come to this a bit later, is that S-antigen can also be derived from viral integration. And I will also explain this a bit later because it's an influencing variable when interpreting the data. Now, the next issue is, and in a nutshell, to summarize this, we have HPV DNA, what you all know for the diagnostics in HPV infected patients, and what is basically to summarize is the difference. So HPS antigen not only reflects variants, but also defective or sub viral particles, which exceeds the variants by far. So it is also a marker, not only for replication, but also for CCC DNA transcription and mRNA translation, which will be important when it comes to the ultimate goal of HPS antigen loss, and especially in novel therapies. And the first study that really documented this, the correlation of CCC DNA with HPS antigen was a study from Germany, Jörg Petersen was involved here, where there's this correlation of CCC DNA and HPS antigen. However, you will see this, this is not black and white, and I will come to this later on. Now, because HPS antigen is exceeds the variants in such high amount, there's of course an important diagnostic tool, and it has been the mainstay of hepatitis diagnostics, e-diagnostics, and screening since its discovery in sera from patients with acute hepatitis in 1968. And you all know qualitative HPS antigen. Now, why took it so long for the quantification of HPS antigen? And do we have, and since when it can be quantified? And I think, when you look to the data, and I talked to Wolfram Gerlich on this, who distributed also and gave me the slide here, the first report on standardized HPS antigen quantification is really not new. It's in weight per units per volume was almost 50 years ago. Wolfram Gerlich did his first experiments in Frankfurt in 1970. And these are some publications, and this is a quantitative immune electrophoresis of the HPS antigen by that time. But why took it so long? And the importance, of course, is that we have standardized assays. And at the first, in the 70s, we had the Paul Ehrlich Institute units. Then we have the first WHO standards. Then we have also some nanogram per milliliter, some labs use. So it took a while, until 2003 or so, until we had the first WHO standard. And the first assays were developed at that time. So that the first really interesting results from big cohorts came in 2010. There was really a breakthrough in data. You can see some of these publications here. And this was accompanied by a nice review from Maurizio Brunetto, who made really the point, a new role for an old marker 50 years ago. So I think this really was a breakthrough. And you will see some of the data again in the talks. And now we have commercial assays. And also by that time, we had commercial assays. And these are just three examples of these. At least three commercial assays that can quantitate HBS antigen. They are automated, reproducible. They are standardized according to the WHO standard. They have relative low cost for Germany, for example, 5 to 6 Euro per sample, and a short time of evaluation and detection. The dynamic range of these assays are from 0.03 or 0.05, depending on the assay, to 130, 150, or 250. This is not high. And you usually need HBS antigen values up to 20,000 in the high variable carriers. But for this purpose, the automation on board dilution is in place. And typically it will be diluted 1 to 400 or 1 to 500, and thereby increase the upper limit of detection to over 50,000 units. And that is well established and standardized. And therefore also the assays have also in 2011 already been correlated with each other. So this is an example from our center. This is from Milan Sonnefeldt, who is also a speaker here. And you can see the Roche assay, for example, and the Abbott assay correlate very nicely. And these included patients with different phases of the infection. And also we have here patients on treatment with interferon and lamivudine. So there's a high correlation with these assays so that they also be exchanged if you have patients. There's also a trial here on all the three assays. You can see also the high correlation. And also here where samples included, for example, some mutations, different genotypes, and so forth. So overall, there was a very high correlation. There's maybe a small problem here with the Liaison assay that underestimates HBS antigen levels by some 0.1 log value. This may be not so important, but for example, if you use an assay, maybe stick to the same one in the course of the infection. Now the question I would like to ask is what impacts the level of HBS antigen quantification? This is important to consider if interpreting the results. I've shown you this result of correlation with ccDNA and HBS antigen, but this is not the complete story. Now in 2010 already, the group from Alex Thompson and Stephen Locarnini showed already that there's a correlation of the ccDNA and the S antigen by the E antigen positive patients, but not so much by the E antigen negative patient. And this is important. And the reason for this, and at that time already Stephen Locarnini discussed the possibility of integrated HPV DNA, that this may be a relative higher amount in the E antigen negative patients. And looking at the life cycle, I introduced this already, that integration leads to S antigen production. And also our study and the study from Stephen Locarnini showed that the HBS antigen ratio to HPV DNA replication is higher in the E antigen negative patient, especially those with low replication. So the inactive carriers, meaning that the HBS antigen is not coming from replication and maybe from ccDNA, but rather from integration. So the relative amount is higher in the E antigen negative patients. And this explains the results by siRNAs, by the first generation siRNA that was developed to target the RNAs. And they target really the RNAs. There is a drop in S antigen levels, but this first generation did not target the integration part. So integrated DNA leads to HBS antigen, not to viral replication to S antigen, but as the amount is higher in the E antigen negative patient, the amount of virus suppression or S antigen suppression was more significant in the E antigen positive patient than the E antigen negative patient. And this study from Vaudel really made the point that this is due to integrated DNA. So there's a variable. And there's of course a concern that mutations in the S antigen, so there is an overlapping reading frame, there's a polymerase and so on, there can be mutations. And some of the mutations I made here, two examples have really an impact in vitro for the S antigen and for the different assays. So if you have mutations in the S antigen, for example, this one, the G145K, you have lower amounts in vitro in the assay. And this is more affected by, for example, here the Roche assay. So the Roche assay may have more problems than the Abbott assay. And for this mutation, it's the opposite. Now, is this really so relevant? Are these mutations so relevant? And is this really a major population in the patients? This, of course, needs to be answered. But a study by Maurizio Brunetto in E antigen negative patients with genotype D showed indeed there is pre-S gene mutations and that accumulate over time and is related to progression of the disease. But if you look into the different phases of the infections or in patients with cirrhosis or chronic hepatitis, there was not a difference in the pre-S gene mutated patients, pre-S mutations or vial type. So that in the overall population, there is not such an impact of mutations. But we have this and this patient sometimes where we have really low levels of S antigen. And sometimes I have even a patient with a non-detectable S antigen, but this replication in this patient has multiple of mutations in the S antigen, but these are rare cases and exceptional cases. Co-infections may be more relevant. And I have to emphasize the hepatitis delta, which is overseen in many cases. But here, when you look also to the S antigen HPV DNA ratio, it's higher in the delta patients. Usually they have low HPV DNA levels, but they have similar or even higher S antigen levels compared to mono-infected patients. For a co-infected patient with hepatitis C, usually we have the opposite. We have lower S antigen levels in co-infected patients with hepatitis C because the C virus induces interference-stimulated genes and maybe immunological responses are the key for the lower S antigen levels and usually for lower replication of the B virus. And for HIV, you can also see that the ratio for HPS antigen to HPV DNA is different. So you have higher HPS antigen levels in the co-infected patients despite similar HPV DNA levels. And also this may be due to immune responses because we have shown there's a correlation of the CD4 count in the HIV patients. So you have higher S antigen levels in patients with immunosuppression that have lower CD4 counts. So immune status is an important issue when considering the S antigen levels. Also, the phase, not only the phase of the infection, what you will hear also from Magani, but also the fibrosis stage may correlate with S antigen, but interestingly only in positive patients. So it makes it all a bit, seems to be very complicated, but you need to know your patients. And the fact that here in the cirrhosis patient, you have lower S antigen levels and higher S antigen levels in F0, maybe in fact, due to the fact that we have different phases of the infection, the immune tolerant or HPV high inactive carriers, the ER germ positive infection patients, they are younger, they have less inflammation, maybe less immune responses towards the virus. And therefore they have higher S antigen levels. This may be the explanation for this. And we also know that age is a factor on patients with fibrosis and cirrhosis are older. So with increasing age, we have also less S antigen levels that you need to consider when interpreting the data. I summarized this together with Dieter Glaeber in an editorial here in this journal, that there are many different factors also contributing to different levels of S antigen. I summarized some of them. We have also, I did not mention the genotype, genotype A patients usually have a bit higher of S antigen levels. The reason is not clear, but for example, the thousand cutoff that we established for genotype D patients for inactive carrier, for genotype A, we may or less in our center, we have defined a cutoff of 3,500, but on a small number of patients. Integration I showed you, the immune status may be important, the co-infections, and of course also hepatocytes, cellular carcinoma due to clone expansion. This can be also HPS antigen positive and maybe age, which reflects the duration of infection. There are also some new aspects in the last two minutes. I just want to briefly mention this. There are also new assays, they're called ultra sensitive assays. I mentioned the upper limit of detection is usually 0.05. In these assays, they are 0.005. They are more sensitive, 10 times more sensitive than the old assays. For example, the Lumipulse assay from Fuji Arabia or the Abbott Architect ultra sensitive assays, also they have been correlated, shown excellent correlation with the other assays. So that is not the issue and they can be exchanged and can be used. The advantage is that these assays detect some levels of S antigen and the Abbott Architect assay or the Rush assay show here no detection or low detection. So they seem to be more sensitive. The question, of course, is, and we also discussed it at the ESA, the endpoint conference, the clinical relevance of low levels of S antigen so far is unclear. And I'm not sure if this is really relevant at this stage. It may be relevant with the new therapies. This has to be debated. And the final issue is, I've shown you that the several RNAs leads to S antigen, but the S antigen contains three proteins. It's the large S protein, the small and the medium S protein. And also these could be quantified in theory. And there are antibodies available in western blot. These antibodies are also from the Gerlich lab. I mentioned him already and Dieter Gleber. And you can look here at the western blot and you can see the composition of these fractions. And we also established an ELISA and also Florian van Bommel from the Baird group established an ELISA and potentially this can be automated in the future. You see also correlation. We see in this case, we see Rush assay, the total S antigen. You can see that the fractions differ and this may have a role, for example, in specific research question, but so far not for clinical practice. I want to summarize. HBS antigen reflects the cDNA transcription activity, the subgenomic RNAs rather than replication, but also integration. Quantitative S antigen assays are commercially available and validated. In addition to replication and integration, several variables affect S antigen levels, such as genotype, S mutations, immune complexes, cirrhosis, age, immune status of the patient, which must be considered when interpreting the data. New developments are ultrasensitive S antigen assays, determination of the three S proteins, but so far the clinical meaning is not established. I thank you very much for your attention. Thank you very much, Markus, for this beautiful talk, giving more the background of the HBS antigen and the biological aspects. We move from the lab, from the virology to the clinic now with two speakers. The first one is Dr. Mark Ghani from the NIH. We moved from Germany to the US. He's a well-known investigator at the liver branch of the National Institute of Health in Bethesda, Maryland. His work has been honored with the Regal Award for Excellence in Gastrointestinal Liver Disease and with several service awards from the NIDDK. Dr. Ghani's focus is in particular on translational research to improve care and treatment and outcomes of patients with viral hepatitis. Mark Ghani will talk in particular on how to utilize quantitative S level in the natural history of the disease to see whether it could predict or rather announce an S loss and towards outcomes of chronic hepatitis B. Dr. Ghani. Well, thank you for that kind introduction, Harry, and I'd like to thank the organizers, Dr. Gish and Dr. Janssen, as well as the ESLD for the kind invitation to participate in this important webinar. I have nothing to disclose. This is an outline of what I plan to cover over the next 15 minutes. I'll discuss the role of surface antigen levels in predicting surface antigen loss and defining the natural history of chronic infection with a particular emphasis on the inactive carrier phase and also the role of surface antigen levels in predicting clinical outcomes, particularly cirrhosis and hepatocellular carcinoma. So you've seen this slide from Dr. Kornberg and just to remind you that surface antigen serum can originate from two sources, from the equivalently closed circular DNA, as well as integrated HPV DNA. And Dr. Kornberg has told you that the surface antigen in circulation circulates in three forms, in the intact virion or Dane particle and in two subviral spherical and filamentous particles that are devoid of DNA and greatly outnumber the actual intact virion. Now, clearance of surface antigen is an important milestone in the natural history of the infection and it's associated with improved outcomes, including lower rates of cirrhosis, lower rates of hepatocellular carcinoma and liver-related death. But unfortunately, it rarely occurs. And in two recent meta-analyses, the rate of spontaneous S antigen clearance, the pooled annual incidence was shown to range from 1 to 1.2% and the rates were similar in both endemic and non-endemic regions. Now, many factors have been identified that influence the rate of S antigen loss and these can be broadly viewed as host factors and viral factors. And the host factors include older age, male sex, the disease phase, a higher BMI. Viral factors include being negative for the E antigen, genotypes A and B, a low HPV DNA level defined as less than 10 to the 3 RU per mil, a low surface antigen level defined as less than 10 to the 2 IU per mL. Now, the most important of these factors are the viral ones, and in particular, the DNA and surface antigen levels, and I'd like to share with you some data looking at the role of quantification of surface antigen in predicting surface antigen loss. So low surface antigen levels have been associated with S antigen loss in an Asian cohort of E negative patients followed for a mean of seven years. As you can appreciate in this figure, surface antigen levels were almost 50% lower in those individuals who achieved surface antigen loss compared to those who did not. Both a cutoff of less than 100 as well as a cutoff of less than 1,000 were predictors of surface antigen loss, but a surface antigen level of less than 100 was more specific but less sensitive compared to a cutoff of less than 1,000. Thus, the lower level had better negative predictive value, whereas the higher level had better positive predictive value. Notably, I'd like to highlight that even using a cutoff of less than 100, it can take upwards of eight years to finally achieve S antigen clearance. Surface antigen levels have also shown to be predictive in a multiracial cohort, and in this study of almost 300 patients with chronic hepatitis B with low DNA levels, you can appreciate here that a level of less than 100 was almost 20 times…there was almost a 20-fold higher rate of S antigen clearance compared to those individuals who had a surface antigen level greater than 1,000. It's also been shown that surface antigen levels decline more rapidly closer to the time of S antigen loss, and this is not surprising. In this study, surface antigen levels were measured at five, three, and one years prior to S antigen loss among 46 patients with and 46 patients without S loss that were matched for age, sex, genotype, and E antigen status. You can see on the left figure that between years three and one prior to S antigen loss, there was a steeper decline or drop in surface antigen levels in those individuals who ultimately cleared S compared to those who did not. Moreover, using cutoffs of less than 200 or 100, the time point one year prior to S antigen clearance had the best positive predictive value and negative predictive value for S antigen clearance. Low surface antigen levels have also been…after E seroconversion have also been shown to predict future surface antigen loss, and in this study of almost 400 patients, compared to a level of more than 1,000, individuals with surface antigen levels between 100 and less than 10,000 had a fourfold higher rate of surface antigen loss, and those with under 100 had a 24-fold higher rate of S antigen loss compared to those individuals with levels greater than 1,000. It was also shown in this study that surface antigen levels were better predictor of surface antigen loss compared to using HPV DNA levels. In a subanalysis of patients with DNA levels less than 200 IU per ml, surface antigen levels were also significantly predictive of S antigen loss within a period of 6 years, and you can see here that those individuals having a level less than 100 had a 14-fold higher rate of S antigen clearance compared to those with a level greater than 1,000. So, turning now to the natural history, this is data from the hepatitis B research network of over 1,400 patients, showing that surface antigen levels vary across the different phases of hepatitis B. Levels are highest among those individuals who are E antigen positive and lowest in the inactive carrier phase. Interestingly, even after S antigen seroconversion, surface antigen levels can remain in the three to four log range and really do not begin to significantly decline until HPV DNA levels become low or undetectable, as you can see here. So, these observations suggest that measuring hepatitis B surface antigen levels may be useful in distinguishing the phases of disease, and in particular, among E negative patients. And in this study, surface antigen levels were shown to be lowest in the inactive carrier phase. These were over 200 E negative Italian patients, a geotype D, highlighting that those individuals who were classified as inactive carriers had the lowest surface antigen levels in the inactive carrier had the lowest surface antigen levels compared to those with an indeterminate phase or those with E antigen negative chronic hepatitis B. Moreover, this analysis showed that a single point determination of surface antigen and HPV DNA can accurately predict the inactive carrier phase over a three-year observation period. Using a cutoff of surface antigen of less than 1,000 and a DNA less than 2,000, these two parameters had were 94% accuracy in predicting S antigen loss. Therefore, a single point determination of these two parameters were comparable to monitoring patients carefully over a one-year period. So, there was some, there was some debate when this paper came out, whether these, this prediction or value of measuring surface antigen was restricted to geotype D patients, or did it apply to patients with all genotypes? And this study here addressed that question, showing among close to 300 E negative patients with low DNA, that indeed a low surface antigen level was predictive of persistent inactive carrier state. However, as compared to the prior study here, a cutoff of less than 100 IU per mil was the most accurate predictor of the persistent IC phase when looking at patients of all genotypes, not just a genotype D. So, turning now to the role in clinical outcomes, and in particular cirrhosis risk, and this is data from the reveal study showing that higher surface antigen levels are associated with an increase in cirrhosis risk, and you can see that the highest risk was associated in those individuals with a level greater than 1,000. However, in this analysis, it was shown that HPV DNA was actually a stronger predictor of hepatocellular carcinoma, sorry, of cirrhosis, sorry. Turning now to hepatocellular carcinoma, a recent study to hepatocellular carcinoma, a recent meta-analysis of eight studies, including almost 12,000 patients, highlighted that a surface antigen level greater than 1,000 IU had a 2.5-fold increased HCC risk compared to a cutoff of less than 1,000. And back to the reveal study again, here is data again showing that the highest surface antigen level is associated with an increase in HCC risk, and the cutoff of more than 1,000 IU per mil was associated with the highest risk of hepatocellular carcinoma. But again, as in the analysis for cirrhosis, HPV DNA was a stronger predictor of development of HCC. Here's data from another population-based cohort study showing that both surface antigen and HPV DNA were predictive of development of hepatocellular carcinoma. In the case of surface antigen level, the highest risk were in those individuals who had surface antigen levels between 1,000 and less than 10,000, presumably indicating that these are immune-active patients. And in a receiver operating curve characteristic analysis, it was shown that actually HPV DNA was the best predictor and a significantly better predictor than surface antigen level. It was also shown in this paper that the change in surface antigen level was associated with HCC risk, and that the lowest risk was observed in those individuals who had a persistently low surface antigen level, less than 1,000 for up to three years, and the highest risk was observed in those patients with a persistently high surface antigen level greater than 1,000. In addition, it was also shown that those individuals who transitioned from a low to a high surface antigen level, that their risk of HCC increased, and conversely, those individuals who transitioned from a high to a low level, that their risk of HCC declined. Surface antigen, so where can surface antigen level, where surface antigen levels can actually be a better predictor than HPV DNA is in the population of patients with low DNA levels. And if I can ask you to focus on the rightward panel here, you can see that those individual among patients with a low DNA level, those individuals with a cutoff less than 1,000 had a significantly lower risk of developing HCC compared to those individuals whose surface antigen levels were greater than 1,000. Of note, using a lower cutoff of 10 or 100 was not significantly associated with a better stratification of HCC risk. And similar findings were shown in an analysis of the REVEAL cohort, showing that among patients, showing that a surface antigen level greater than 1,000 had a 4.4-fold and a 2.2-fold higher risk of HCC compared to less than 1,000 in patients with undetectable or HPV DNA levels less than 2,000, but not in those individuals with DNA greater than 2,000. So how might we use surface antigen levels for monitoring untreated patients in our clinical practice? Well, for those individuals who are, it may be helpful in predicting the phase of infection. And for those individuals who are E antigen positive, a level greater than 10 to the 5 together with a high DNA level and normal ALT is more suggestive of an immune-tolerant patient. Similarly, those individuals who are E antigen negative, a level below 1,000 together with a DNA less than 2,000 is quite predictive of a persistent inactive carrier state. Furthermore, among E negative patients, a level of surface antigen below 100 is associated with a higher chance or opportunity of clearing surface antigen within the next six to eight years. But where I think surface antigen levels may have the most value are in individuals who are E negative with low viral levels defined less than 2,000, where a surface antigen level greater than 1,000 was shown to be significantly associated with a higher risk of hepatitis C. And suggests that these patients do require more closer monitoring. So in summary, surface antigen provides different but complementary information to HPV DNA. There's a limited role for using surface antigen levels for monitoring E positive and negative patients with high DNA. However, among E negative patients with low HPV DNA levels, surface antigen cutoff of less than 100 is predictive of future S antigen loss. And a level less than 1,000 is associated with persistence of the IC phase, presumably lower risk of disease progression and lower HCC risk. And thank you for your attention. Thank you very much, Mark, for this very nice and eloquent talk on the natural history in HBS antigen quantification. And we move then from the natural history to treatment. And let me introduce the next speaker, who is Milan Sonneveld. He's a consultant in gastroenterologist and hepatologist in the Rasmus University in Rotterdam, where he was also trained, as well as in the TH Chan School of Public Health in Harvard, United States. His research is really focused on optimizing treatment for chronic hepatitis B. And he's the chairman of the Dutch HPV Guideline Committee. And Milan will teach us on how to use HBS antigen, particularly in licensed treatment with nucleoside analogs, as well as Pag-interferon. And definitely there's much more to see of quantum S levels with the new treatments that are coming, aiming for HBS antigen loss and functional cure. But Milan will mainly focus on the drugs that you have in your hands, in your practice, the licensed drugs. Dr. Sonneveld. Yes, thank you, Erie, for the kind introduction. And indeed, I'll mainly focus on the use of self-centred monitoring during treatment with the currently available antiviral agents. But of course, I'll also touch upon some data with the new antivirals. And these are my disclosures. And these are some of the things that I want to discuss with you, namely the things that I can think about where you could find practical applications of surface antigen monitoring during antiviral therapy for chronic hepatitis B. I think there are several potential indications to test this in patients treated with nukes. It has to be tested in patients treated with interferon. And I'll end up with discussing some data for the novel treatments. So let's start off with looking at some of the data for the nucleoside analogs. First of all, we should note, and it's also already stated by Professor Kornberg, is that patients treated with nukes tend to have very rapid declines in HPV DNA, but the decline of surface antigen is generally very slow, somewhere in the neighbourhood of 0.04 to 0.16 logs per year. Now, in some patients, this may be more rapid, and that's mostly during the first year of treatment, in patients with higher pretreatment levels, in patients with higher pretreatment ALT, and in patients who go on to achieve e-antigen clearance. And of course, patients with more rapid on-treatment decline will achieve surface antigen clearance faster. However, for the majority of patients on antiviral therapy, the estimated duration that you need to treat your patient until surface antigen clearance is somewhere between 20 to 40 years, depending on the number of favourable characteristics that may be present in your patient. So for most of you, if you test it once a year, you will not see such a great difference compared to the year before, and for the individual patient, that may not mean so much. But on group levels, this is certainly interesting. Now, another potential application would be to identify patients at increased risk of HCC. We've already seen the data for untreated patients from Margani, but what is important to note is that there is actually no well-performed large study looking into the association between on-treatment surface antigen levels and long-term risks of HCC. However, if we take some of the available data, such as the knowledge that a high surface antigen level of above 1,000 is associated with incident HCC in untreated patients with low DNA levels, and we combine that with the knowledge that persistence of surface antigen positivity is also an independent risk factor of HCC in those with suppressed HPV DNA, then we can already consider surface antigen a risk factor. And this is actually confirmed by data from patients who already have developed HCC, where we see that an elevated surface antigen level above 1,000 is actually an independent risk factor for HCC recurrence after curative treatment by the surgeon. So in the end, what I think is that a high surface antigen level above 1,000 should be considered a main risk factor for HCC, although how we should implement this in clinical practice is still uncertain. Now, another important aspect of surface antigen levels is that they could potentially be used to identify patients who are more likely to achieve a response to pegylated interferon add-on. As you all know, the majority of patients who we treat with nukes will not achieve e-antigen clearance, if they are e-positive at baseline, or surface antigen clearance during long-term therapy. And therefore, various randomized studies have in fact tried to increase response rates by adding interferon to the nuke backbone. However, the majority of patients does not achieve a substantial response to this add-on therapy, and therefore we truly need selection tools. And it turns out that if you look at larger data sets of patients treated with add-on therapy, then actually a low surface antigen level, here defined as less than 4,000 in this group of e-positive patients, is actually the main determinant of a benefit to antiviral therapy with peg interferon add-on. So if you want to select e-antigen positive patients for peg add-on therapy, then I think assessing surface antigen is an important factor to consider. Now, where surface antigen monitoring is arguably the most important is in the context of identifying patients who could be eligible for therapy withdrawal. And the question would be, why would you want to stop antiviral treatment with nukes? Because they are relatively well tolerated. Well, obviously, if you stop treatment, then you have a reduction in treatment-related side effects, which may be minimal but may still be there, particularly during long-term treatment. Now, you have a reduction in treatment-related costs. And you will note that about 30% to 50% of patients will achieve a sustained disease remission after discontinuation of new treatment. And in fact, a recent randomized study, which is not published yet, has shown that discontinuing antiviral therapy in e-negative patients is actually associated with an increased probability of surface antigen clearance, a so-called stop-the-cure strategy. Now, there are also major downsides to therapy withdrawal. First of all, you have to monitor your patients very intensively after therapy withdrawal. Most of the patients will develop HPV DNA positivity, which may be associated with an increased risk of HCC in the long run. We don't know. And of course, an HPV DNA rebound may result in severe ALT flares, which have been reported to result in decompensation and even death in patients with advanced liver disease. So careful selection of patients is essential to optimize outcomes and limit risks. Now, first of all, it's important to note that patient ethnicity is actually the main factor determining the outcomes after nuke cessation. Non-Asian patients do a lot better than do Asian patients. Non-Asians have both higher rates of virological responses and surface antigen clearance than do Asians after nuke withdrawal. However, the other important predictor is surface antigen. The lower the surface antigen at the end of nuke treatment, the higher the rates of virological response, surface antigen clearance, and perhaps most importantly, the lower the chance of ALT flares. Patients with the lowest surface antigen levels, a cutoff of 50 in this particular study, had excellent chances of sustained response, excellent chances of surface antigen clearance, but virtually no ALT flares after discontinuation. And importantly, these findings were consistent for Asians and non-Asians. And as you can see here, the Asians generally had a lower chance of treatment of success after therapy withdrawal, but still those with the lowest surface antigen levels did best. And importantly, those with the lowest surface antigen levels, regardless of ethnicity, had virtually no risk of ALT flares after therapy withdrawal. So really here, you can use surface antigens to select your patients for therapy cessation. And indeed, these data from the study that I've just shown you, they combine well with previous smaller studies, where we can conclude that lower surface antigen levels at the end of nuke treatment are associated with a lower risk of relapse, a lower chance of ALT flares, and a higher chance of surface antigen clearance. And the only question we have to ask ourselves is what cutoff to use. If you look in the literature, then a cutoff of 100 is often applied, but this is more of a choice of convenience because most studies included very little patients with surface antigen levels below the 100 cutoff. But larger studies, so that's the one I've just shown you, show us that those with even lower levels, so below 50 or even below 10, they do even better. So what constitutes the optimal cutoff is actually a trade-off between benefits and risks, and of course, patient preferences. Now, I expect in the future that we will be combining surface antigen with novel biomarkers such as HPV RNA or correlated antigen, because risk scores, such as the scale B score, are actually even superior in predicting outcomes after therapy cessation. But of course, most of these tests are not widely available yet, and until that is the case, we should be using surface antigen because it is the best we've got, and it's actually already pretty good. If we then move over to the use of surface antigen monitoring in PEG-related interferon, then it's important to note that PEG-interferon is by definition a finite treatment, so end-of-treatment monitoring of surface antigen is not really helpful. Now, what we want to achieve is that we want to identify patients as early as possible in order to save them from long-term interferon treatment and all the interferon-related side effects. Now, the problem with HPV DNA during PEG-interferon therapy is that we see a rapid decline in responders, but also in patients with a relapse, and even in patients with a non-response, we see an early decline during the first 12 to 24 weeks. And therefore using HPV DNA during those early phases to select patients who are unlikely to achieve a response is not really effective. If we then look at surface antigen kinetics on the right, then we see a very rapid decline in patients with a sustained response, whereas there's virtually no change in the S antigen levels in those with a relapse and in those with a total non-response. And this is really where surface antigen is at its most powerful. I would also like to point your attention to the extreme declines observed in these PEG interferon responders, which compares very favorably to new treated patients. And here you can achieve logs of decline over a year of treatment in those with a favorable response. However, in order to identify patients most likely to benefit, we have to measure surface antigen. And in clinical practice, what we do is we assess surface antigen in our E-positive patients treated with interferon at week 24, because we know based on meta-analyses that if surface antigen is still above 20,000 at week 24, then whatever the genotype, patients will not respond. And therefore you have to, and you can use this as a very reliable stopping rule at week 24 of PEG interferon therapy in E-antigen positive patients. Now for E-negative patients, it's a bit more complicated because you have to use a combination of surface antigen and DNA. But the upside of this is that this stopping rule is actually effective already at week 12. And therefore also for E-negative patients, surface antigen is really pivotal in the effective use of PEG interferon therapy in your patients. And chances are that in the long run, this will also be the case with novel immunomodulators. Also for PEG interferon-like therapies, I expect that surface antigen will remain important even in the face of the availability of novel biomarkers such as HPV RNA. In a recent study of about 300 patients treated with nukes and interferon, a HPV RNA response during treatment, which was defined as a two log decline or a decline to undetectable levels, was actually shown to be a good predictor of sustained response and surface antigen clearance. But if you look more closely in this group with the HPV RNA response, then it turned out that a significant proportion, actually almost half of the patients, did not achieve a concomitant decline in surface antigen. So there was an HPV RNA response, but no surface antigen response. And these patients actually did very poorly. They had low rates of sustained response and none of them or only 1% clear surface antigen. Whereas the patients who also had a strong decline in surface antigen did very well. So in the future, again, I expect also for these types of therapies, we will need combinations of biomarkers and surface antigen will remain very important. Now, another interesting thing to do with surface antigen is to use it to classify flares surrounding antiviral therapy. And most of these data is from interferon treatment because we know that flares are an important and integral part of treatment, both during and after therapy. And what you see here is two patients who were treated with interferon who experienced a strong ALT flare of about 30 times the ULM. And the ALT is shown in red. And what you see on the left part is that in this case, the flare was followed by a rapid decline in surface antigen, E antigen and DNA. In the other case, that was not seen. And what we noticed when we collected data on more of these patients with flares is that an early decline of surface antigen after a flare, so within four weeks, but often already within two weeks after the flare is highly predictive of surface antigen clearance. And it's actually associated with 64% surface antigen clearance in that treatment episode. So you can use surface antigen here as a truly good predictor of outcome. The last thing I would like to touch upon is the use of surface antigen monitoring in patients treated with novel antiviral agents. As you know, there are many new agents being developed for chronic hepatitis B, and all of these agents aim to increase the number of patients achieving surface antigen clearance. What we know from data and meta-analyses of patients treated with interferon or mucs is that if you look at the early surface antigen declines, so at weeks 12 or 24, and you plot the mean decline for a particular cohort against the rate of surface antigen clearance during the long-term, then there's an almost linear association there. And therefore we can note that on a group level, you can actually also use your mean surface antigen decline during weeks 12 or 24 of treatment to predict the long-term outcomes with that antiviral regimen. So on-treatment surface antigen kinetics with novel agents are very much of interest also with novel drugs. And in this slide, we see some of the data with the novel agents. On the left, you see data for short-term treatment with a capsid assembly modulator, in the middle data for nucleic acid polymers, and on the right for a RNA-interfering agent. And indeed, particular for these last two groups, we have seen very impressive surface antigen declines with rates of surface antigen clearance of up to 50%, which is of course extremely impressive compared to what we have in our hands at this time. But it should be noted that surface antigen declines and clearance with these agents may not be exactly the same as what we know from the conventional agents, because some of these drugs directly inhibit surface antigen production or indeed a release from the host cell. And we do not really know whether that is exactly the same as the surface antigen clearance that we know from the natural history or indeed Nucor-PEG interferon treatment. This remains to be determined. So that was really the points that I was trying to make. Some of these conclusions are shown on this slide, which you can read back if you want. And I would like to thank all of you for your attention. Thank you very much, Milan, for this beautiful talk. Bob, would you like to start the discussion? Maybe everyone turn on their camera. Thank you, Harry. Thank you to all the wonderful speakers. This was a very, very thorough review and refined review. And yes, we could start some discussions. One of the questions came through multiple times in a little bit different format is you test baseline in all patients, and then how often are you testing QANTAS in the clinic? How frequently? Do you want to start, Milan, with your experience? Yeah, it's a good question, actually, and I'm sure that will vary across the globe. But in the end, for me, we do it in every patient, actually at every visit, because for us, the test is so easily available, and it has actually supplanted the qualitative test. So, and also there's another reason to do it. We discussed this yesterday also, is that when we screen our patients for trials with novel drugs, we need these data as well. So there's multiple reasons why we do it, but they're not always only for the patient's clinical management. So if we discard the other reasons, I would say a yearly assessment would be the consistent approach that we apply. Harry, are you on board with the annual assessment? Yeah, well, yeah, I do it as well. It's a very cheap and easy test. It predicts several outcomes. And in particular, as Mark mentioned, I think it's very helpful to distinguish E antigen negative patients who are truly inactive versus E antigen negative patients who still have flares. There's a lot of these patients we see that are really in the gray zone with HPV DNA levels of, let's say, three logs, four logs. The ALT is bouncing around to the upper limit of normal. And their S antigen really can help you to predict whether these patients will do well at the end of the day and whether they will move on to become inactive and become as negative. Great, and Mark, could you comment on the Hepatitis B Research Network? Didn't they really support what Harry just mentioned about refining the indeterminate patients? Yeah, so actually we looked at surface antigen as well as two of the other more novel markers, HPV RNA, as well as hepatitis B correlated antigen. And particularly this group of indeterminate HPV DNA, we focused really on the E negative indeterminates and we actually subdivided them into those with high DNA versus low DNA. And actually, to our surprise, the levels of both quantitative surface as well as HPV RNA and correlated antigen were not that helpful in distinguishing those patients from E negative chronic hepatitis B. They were from inactive carriers, but not so much from E negative. And it's unclear whether either marker alone or in combination is actually better than just the HPV DNA level. So I think these are questions that still need to be addressed. Marcus, in your clinic, are you baselining annual? Yeah, well, no, we do it the same way. Milan is doing it at every visit. It's cheap and we get it always quantified. And I think it's sometimes very helpful because you better know your patients. And I think it's good to have as much information in your patient as possible. And it's a cheap marker. Therefore, I can only encourage everyone if the test is available to do it when the patient is coming. And I mean, it's much cheaper than HPV DNA. You get some interesting information. And once we have also the other markers on board, I think the way for personalized medicine in hep B is the way to go in the future with the novel drugs. And we will get also HPV RNA correlated antigen in the future. And as much information we get, the better we manage our patient. Maybe we get even host markers in the future. We have some data and published this a couple of years ago, the intermediate and the high replicative E negative patients. We could also not distinguish them by S antigen, but interestingly by cytokines. So there is something on the immune response. Harry, back to you. Do you see some other questions in the QA boxes which you covered? Yeah, so I had a question for Marcus actually. So you mentioned the small, middle and large HBS antigen protein, right? And the trouble that we would have particularly negative patients is that actually the viral replication can be completely shut off, but due to the integrated HPV DNA, there is still S production. So is there any role of any of these sub-fractions of HBS antigen to disentangle really whether this is coming from integrated or S antigen for viral replication, really? That's an excellent question. And there is some data that hints towards this, the data from Florian from BAML. We have also similar data. He published in gutter paper that he could better distinguish the E negative hepatitis from the inactive carriers with the fractions, with the L and the M. So this is the idea that maybe the integrates do more of the small of the S and the spheres and the filaments, but this is not completely resolved. This is not completely clear. There can be also some modification of the RNAs. For example, the Delta patients, they have much higher L and everyone, and also Stefan Orban thinks always that the high amount of S antigen comes from integration in the Delta patients. But this is not completely clear. It could also be that there's some post-translational modification of the RNAs. So HPV is very complex, but it can help maybe. So the L is some, I think it's an interesting marker that maybe could help to better discriminate. But at this stage, I think we don't need it. Okay. I had another question, Bob. Can I raise that or not yet? Go ahead. Yeah. So this is the point that Milan brought up because we're moving into a new era of new treatments trying to cure HPV or functionally cure HPV, which is defined actually as S loss. The question is with this new compound where we might have potentially a rather artificial S loss, not really moving along with an effective immune response, what will be the sustainability, right? So I would like to ask the panelists maybe to start with Milan and the others as well, to what effect this is sufficient S antigen loss by itself. Hi, Harry. It's a good question. Of course, we raised this issue recently in a discussion paper in gastroenterology. And to be honest, I don't know. What I'm afraid of is that we will have to redefine these types of responses and that we have to redo all these natural history type studies by having to show that a surface antigen loss achieved with agents that, for example, prohibit surface antigen release or only inhibit surface antigen production, but not other viral replication parts, that we will have to redefine what constitutes clearance and response. And perhaps also even have to show again whether these outcomes are still as robust with regards to clinical outcomes as far as we know now. So that's a difficult question to answer. I am afraid that if you apply these new agents, then we will have to study again whether a surface antigen loss achieved with these agents is in fact associated with an excellent long-term outcome. We would have to look at that, yeah, for sure. Definitely, I think if there was an effective immune response initiated, I think it's likely to be sustained. Yeah, I mean, another way would be you have to show that at least there's also some histological improvement, for example, that would be another way around this. But of course, that also introduces a novel problem because then you would have to do a new liver biopsies or whatever. Yeah. Bob, can I ask you one more question? Or Markus and Mark, you wanna comment on that? No, maybe I also want to add something on this issue. I mean, this is of course a $100 million question. And basically if you have the HPS antigen loss, you basically have the same as occult hepatitis B, if you be accurate. It's the same definition. But for this, maybe the ultrasensitive as antigen quantification or assays may help because there are some data from the occult HPV fields. Those who have really low, not detectable with the ultrasensitive assays, they are less likely to reactivate. And maybe in this with the new compounds, maybe in that case, the new assays with the ultrasensitive detection could help and maybe a better way for these trials and so on to characterize the patients. Mark? Yeah, I was just gonna comment that what we really need, Harry, is a test that can distinguish the source of the HPS antigen, whether it's coming from the covalently closed circular DNA or whether it's coming from integrated HPV DNA. Because I suspect that there are patients who have been on treatment for many, many years who potentially could stop if most of that surface antigen is coming from integrated HPV DNA. But it's something we don't know because currently we don't have an assay to tell us where that surface antigen is coming. So for these patients, we're obligated to continue therapy long-term. And then to your question about the newer therapies, I mean, I sort of absolutely agree with Milan's comment that particularly using the sRNAs that target the transcript for the surface antigen, it'll target both CCC DNA derived transcripts as well as the integrated transcripts. But as you can see from the studies that have been published so far, once you stop the agent because it doesn't actually target the CCC DNA itself, you're gonna get a rebound in S antigen levels. And as you correctly said, unless that leads to immune control, we're gonna have to probably treat these people long-term. Bob, back to you. Great. So a number of new questions have popped up in the QA. And one of those has to do with how do you use this to change management? And that also has to do with the dynamics. And I think QANT S helps me take care of every patient and that means can change management. First is the discussion with the patient about what phase of infection they're in. I do bring that in in terms of liver cancer risk. So it could potentially alter my liver cancer surveillance strategy. And then really getting a baseline. What's that S antigen dynamic looking like over time? Because we are gonna have five or 10 new therapies within a few years. It will probably be combination therapy. That's just my proposal of the four combined medications. And you wanna know what that dynamic is for the S antigen as you're going into these new therapies. They won't be cheap, they'll be complex to manage. If someone's got an S antigen that's going from 1,000 to 400 to 100, I may just watch them on their new therapy and see if they clear, as opposed to somebody who's at 5,000, every year for five years, on new, they're not gonna clear in a natural history setting either. So I really think S antigen has been useful in all my patients, not just baseline, but the dynamic that's taking place. Would you all agree with that? Yeah. So Bob, this is important. And as I mentioned, so more information, the better you manage your patients. And I think S antigen is also helpful because if something is not fitting, you have a patient with low viral load, high S antigen or the opposite, then you have to think, what is the reason for this? Yeah. Maybe co-infection. Yeah. Oh, did I check Delta? Because I have high S antigen, low replication. Or is it maybe HIV? Did I forget about HIV? So think about if something is not perfectly fitting, something is, you know, and then you think about this. I think therefore S antigen makes a lot of sense to test. And what about surface antibody? There was a question that came in about looking at surface antibody levels or dynamics, you know, the titers. Does that surface antibody come into this in any way? Can I answer that? So for, you know, if a patient achieves S antigen loss, of course you measure this, but it's not completely mandatory to stop your antiviral therapy that has been shown. I mean, two values of S antigen loss or one confirmation six months apart is sufficient that has been shown. But I would, you know, I would make an exception for maybe in a patient with cirrhosis or something, you know, with immunosuppressed patients or some other vulnerable patients that I would consider to stop only if there's anti-HBs. There are always complexes. So anti-HBs is always present, but you do not detect it, you know, because there are complexes. Maybe it's better in the future to test anti-HBc quantitation. Now, that could be also an interesting topic for the future. Next webinar, I would say, right? Antibody, love it. Yeah. I think we're near our time limits. I've emphasized- Can I ask you one question, Bob? So, because this test is used around the globe pretty extensively. I used it while practicing in Europe. I'm using it while practicing now in Canada. And there were some questions in the chat box, but could you once more explain what the availability of this test is and the reimbursement in the United States? Right, so there's a global review that I did at the Singapore Hepatology Conference a few months ago. If you can go onto that website, you'll see a very nice presentation and debate. And I surveyed about 50 different providers around the world on did they have the assay available from Azerbaijan, to Spain, to Italy, to various parts of Asia. And it's widely available, but intermittently used. And they said it's intermittently used because either national guidelines or international guidelines don't define how to use it. So I'm gonna say most countries had it available either at a reference lab or an internal lab at their medical center at low cost. Now in the US, cost is always a different picture. When you send it to Quest or LabCorp, they bill out at about $180 for this test, which is very disproportionate to the assay costs. But when Quest and LabCorp develop it, they have costs to develop the test and they have costs to maintain the test. It has to be amortized across their tests. So even at that value, at that dollar amount, I think it's good. It's reimbursed in the US, no problem. I've ordered far over a hundred tests on patients in the last four years, never had an insurance company have a problem with reimbursement. So I think it's available. I think people need to be using it. I think it needs to be in the guidelines. That's what everybody's asking for is for the guidelines to really cover this in detail. I had this debate with Pietro at the Singapore meeting and he woke up and said, oh, I think we really need to get more detail, really help guide people. So Pietro was fantastic to collaborate with on this. I hope that answered your question, Harry. Did we give enough detail? Thanks. Any follow or final comments from anybody as we wrap up our presentation? Harry, do you want to get some closing comments? Sorry, I didn't, you were asking me some closing comments? Yeah, if you want to do closing comments. No, I just, I found the presentation outstanding, really moving from the basic virology through a natural history to the treatment. It shows the ability of this test to be used. And I think even in the setting of new treatments, which are coming along in for S antigen loss and functional cure, we cannot do without quantitative S level. So it's really, I think a very important test to be used. I hope we made that clear. I'd like to thank everyone for their attention and being around in this meeting. I'd like to thank Dominique Clayton and Brian Kirkland for the support from ASLB, speakers for their dedication to present, to prepare the talk and to answer questions. So thank you very much all for being here and see you at the next webinar for HBBC.

webinar

quantitative S-antigen

clinical practice

chronic hepatitis B

molecular virology

HPV DNA

S-antigen testing

hepatocellular carcinoma

antiviral therapy

patient management