Surfing the MASH Tsunami

Send us Fan Mail

THE NASH Tsunami audience came to know Jörn Schattenberg in the Fall of 2020 when he shared a paper he had recently co-authored on why NASH drug trials failed. This week, the same group that discussed that paper – Jörn, Stephen Harrison, Louise Campbell and Roger Green – reflect on what has improved in the intervening time period and what has not. The group suggest that researchers are making progress in reducing screen fail rates, but not in the critical issues of accessibility and equity.

In this excerpt, Jörn asks whether these new, more focused approaches might lead to investigators overfitting patients. The positive of this outcome, he notes, is that sweet spot patients will be in a position to respond to drug. Louise takes on an additional issue: what is the affect of patient lifestyle mitigation on these trials. Her concern: patients who work hard enough to improve their liver condition after learning they have the disease may be so successful that they no longer qualify by the time they would be called for biopsy. 

The conversation strays into discussing the high value of bring artificial intelligence assists into histopathological interpretation before Stephen returns to the question of recruiting more patients. He notes that the goal is not merely to pull more patients into the first stages o recruitment ("the top of the funnel") but to improve our distribution of different kinds of minorities and subpopulations with high rates of NASH. He points out the fallacy of treating only mainstream Caucasian Americans and assuming results of these trials will hold true for everyone given what we now know about the important of genotyping and mutations tied to race or ethnicity. 

As this conversation ends, Stephen outlines potential solutions based on shifting recruitment out of brick-and-mortar trial sites and to recruiting and testing at primary care offices or even patients' homes.

Send us Fan Mail

THE NASH Tsunami audience came to know Jörn Schattenberg in the Fall of 2020 when he shared a paper he had recently co-authored on why NASH drug trials failed. This week, the same group that discussed that paper – Jörn, Stephen Harrison, Louise Campbell and Roger Green – reflect on what has improved in the intervening time period and what has not. Large metropolises are identified as missed opportunities. 

Stephen starts this conversation by shifting focus from issues surrounding trial criteria to the challenge of finding the right patients for these trials. He notes that of the millions of patients with NASH in the U.S. today, only a small fraction might ever realistically find their way to a NASH trial. The biggest factor is that they can only find their way to a trial if they live with  clear access to a trial center. This is not simply a question of rural and exurban residents having too far to travel. He cites Chicago as a city that contributes minimal patient enrollment, despite being a large cites whose population  suffers from high levels of metabolic disease. 

Jörn adds that while we might believe that European countries with more socialized, government-centered systems would not exhibit this maldistribution in where trial patients live, that belief would be wrong. Roger suggests that the U.S. system, with its highest levels of investment in medical technologies and broader distribution of care sites, might actually be better equipped to attack this issue. 

At that point, Roger returns to an earlier comment to ask Stephen addresses how mid-trial reassessment of screening criteria works and why it improves screen fail rates. Stephen's answer takes up the rest of this conversation.

Send us Fan Mail

THE NASH Tsunami audience came to know Jörn Schattenberg in the Fall of 2020 when he shared a paper he had recently co-authored on why NASH drug trials failed. This week, the same group that discussed that paper – Jörn, Stephen Harrison, Louise Campbell and Roger Green – reflect on what has improved in the intervening time period and what has not. The group suggest that while non-invasive techniques (NITs) have made a significant contribution to patient enrolment, there remains reliant on additional screening methods.

This conversation opens with Jörn's observation that NITs are not always part of the protocol when assessing patients for their inclusion or exclusion in clinical trials. However, he states, experienced researchers will rely on these tools along with presence or absence of other metabolic diseases to decide which patients should not progress to biopsy due to high likelihood of screening out of the study anyway. Stephen concurs, adding that by considering these other risk factors, the clinical trials benefit from a more enriched patient population. and fewer unnecessary biopsies. 

From here, Stephen goes on to discuss two other pivotal issues: advance in strategies for reading biopsies and the impact of high screen fail rates on staff morale. On the first point, he states that good drugs have been lost in development because the strategy for reading biopsies, which relied on a single pathologist to read, created interpretative barriers that led to trials failing or being discontinued. This has led to a place today where all trial design incorporate consensus histopathology reads. On the second, he contrast staff performance for two studies on the same drug: an obesity study with a 25% screen fail rate and a NASH study with an 80% rate. He describes the palpable difference how staff feels about working on each of these projects. 

As the conversation ends, Louise comments on changing FibroScan thresholds for admission to some trials and Stephen describes how inclusion thresholds might vary from trial to trial and ways he uses other NITs to complete his assessments.

Send us Fan Mail

THE NASH Tsunami audience came to know Jörn Schattenberg in the Fall of 2020 when he shared a paper he had recently co-authored on why NASH drug trials failed. This week, the same group that discussed that paper – Jörn, Stephen Harrison, Louise Campbell and Roger Green – reflect on what has improved in the intervening time period and what has not. 

After Roger Green sets the stage for this conversation, Jörn reflects back on the 2020 paper he co-authored with Joost Drenth and considers important changes in patient screening since then.  In his own words, “The way we use biomarkers and AI to augment outcomes has changed the field dramatically.” Whereas biopsy used to be the core method used to recruit patients, Jörn notes that many practitioners now rely on NITs to assess the patient's likelihood of screening into a trial before conducting a biopsy. The result is a significant reduction in non-essential biopsies. 

Stephen follows Jörn by discussing ways his own practice has benefitted greatly from the developing a prescreen strategy that relies heavily on NITs. Instead of doing "five biopsies a day, maybe 25 or 30 a week," he now starts with FAST scores and cT1 and, if these seem promising, moves to multiparametric MRI before deciding who to biopsy. Specifically, his practice is now able to utilize VCTE (FibroScan) and other NITs to mine data in real time and refer patients to the proper clinical trial as a result. 

As the episode winds down, Stephen alludes to the idea of "mid-trial corrections" in pre-biopsy NIT screening criteria and advances in how biopsies are read in trials today as other areas where procedure has improved markedly.

Send us Fan Mail

THE NASH Tsunami audience came to know Jörn Schattenberg in the Fall of 2020 when he shared a paper he had recently co-authored on why NASH drug trials failed. This week, the same group that discussed that paper – Jörn, Stephen Harrison, Louise Campbell and Roger Green – reflect on what has improved in the intervening time period and what has not. The group suggest that researchers are making progress in reducing screen fail rates, but not in the critical issues of accessibility and equity. 

As the conversation begins, Jörn harkens back to the 2020 paper and reflects on improvements in cost-effectiveness and duration of the trial process since. “The way we use biomarkers and AI to augment outcomes has changed the field dramatically.” Yet clinical trials are still reliant on the same endpoint. A surrogate is required for conditional drug approval. Roger asks for the most important improvements that have resulted from conversations around patient selection for a trial. Jörn recalls the past reliance on biopsies and compares today’s focus on non-invasive techniques (NITs). 

Stephen joins to highlight improvements in the way practices utilize VCTE (FibroScan) and other NITs to mine data in real time and refer patients to the proper clinical trial. One idea Stephen mentions: evaluating NIT values during the recruiting process to reset cutoff levels. He also discusses expanding screening criteria to include non-hepatic metabolic risk factors such as obesity and diabetes. Both approaches to patient refinement are contributing to more successful studies and limiting unnecessary biopsies. With fewer screen failures, sites remain engaged. Louise Campbell agrees that mitigating screen fail rate is key to fighting study fatigue. 

The conversation shifts to finding patients for these trials. Stephen notes that major metropolises still do not have access to NASH clinical trials. "We’re putting almost zero NASH patients in clinical trials out of a city as large as Chicago that has high rates of obesity, diabetes and metabolic syndrome.” NITs like the FibroScan are unavailable in large swaths of the country. Even where they are available, a lack of education around NASH diagnostics tests remains a barrier. Jörn adds that multiple hurdles prevent patients from accessing NASH trials in socialized healthcare systems. Roger believes the US system may have a better shot than Europe at broadening recruitment processes because the technology developed under larger U.S. budgets means that some pivotal technologies are more widely available.

Next, Roger comments that for as long as screen fail rate has been discussed on this podcast, we had never discussed the idea of mid-trial adjustments. Stephen gives an example of adjusting FibroScan thresholds mid-trial to boost the number of eligible patients. Jörn shares his belief that the use of NITs outside the formal qualification criteria has helped reduce screen fail rates.

Stephen shifts focus to a bigger issue: equity in the representation of underserved patients in these trials. There is an issue of decentralized trials, whereby at-risk groups are under-referred due to several factors. Stephen outlines potential solutions. Roger suggests tertiary care research universities that have their own primary practice as suitable to become an accessible point of referral. He notes this may not translate to the European system and Louise agrees. 

What is there to look forward to in this space in 6 months to a year? Louise is looking for the approval of FibroScan for primary care in the UK. Jörn is hoping for an opportunity to offer health care services through referral systems. Stephen says we’ll hopefully be a step closer to getting a drug approved. Roger finishes with a Stephen quote: is the juice worth the squeeze? This "juice" takes forms of cash and optimism, and it will be worth the squeeze if a drug sees successful development.

Send us Fan Mail

Episode 40 focuses on lean NASH, and Michelle Long notes that many patients with lean NASH are diagnosed in the Emergency Department when they present with symptoms of decompensating cirrhosis. This conversation from November 2021 considers cirrhotic patients from a different perspective: their uniquely valuable role in clinical trial strategy as the world evolves beyond biopsy as gold standard.

Mazen Noureddin notes during this week's episode that when he identifies lean NASH patients, he encourages them strongly to participate in clinical trials. This conversation also looks at the value of cirrhosis studies in clinical trial design, although from a quite different perspective.

From the initial description of this conversation: In this conversation, Stephen Harrison starts by pointing out that non-cirrhotic NASH trials and NASH cirrhosis trials differ significantly in goals, endpoints and patient severity. From there, he dives into the NASH cirrhosis trial group with his review of REVERSE, which Vlad Ratziu presented at AASLD. Jörn Schattenberg and Mazen Noureddin also comment on design of this trial. The conversation focuses on two issues: who is the optimal patient for a NASH cirrhosis trial and what is the most appropriate attainable endpoint and, therefore, clinical design? On the optimal patient issue, the group agrees the ideal patient is a well-compensated cirrhotic, because the presence of portal hypertension makes the entire healing issue so much more complex. There was less agreement on the optimal design question.

Send us Fan Mail

As the NASH pandemic grows in the number and diversity of patient cases, one patient group receiving increased notice includes patients with "lean NASH," those with "normal" BMI levels (BMI<23 for Asians; BMI<25 for other racial groups). Last month, Gastroenterology published Best Practice recommendations for diagnosing and treating lean NASH. In this conversation, co-authors, Drs. Michelle Long and Mazen Noureddin join Louise Campbell and Roger Green to discuss Best Practice #6, which focuses on alcohol screening with lean NASH and really ANY patient who might have Fatty Liver disease.

This conversation starts with Louise Campbell pointing to Best Practice #6, querying patients routinely regarding levels of alcohol consumption. She points out that many patients will not provide an honest answer. Mazen points out that, unlike obese patients, with lean patients, the physician has to check every realistic alternative to lean NASH, and alcohol is certainly one of these. He agrees that patients might not always provide accurate answers. The conversation follows along on the challenges of alcohol questioning until Roger asks the final question. You need to listen to this conversation to hear what the various panelist had to say.

Send us Fan Mail

As the NASH pandemic grows in the number and diversity of patient cases, one patient group receiving increased notice includes patients with "lean NASH," those with "normal" BMI levels (BMI<23 for Asians; BMI<25 for other racial groups). Last month, Gastroenterology published Best Practice recommendations for diagnosing and treating lean NASH. In this conversation, co-authors, Drs. Michelle Long and Mazen Noureddin join Louise Campbell and Roger Green to discuss why early diagnosis is so important for patients with lean NASH, which has a subtly different focus than other forms of NASH.

This conversation harkens back to the previous week (Season 3, Episode 39) and Ian Rowe's analysis of different early diagnostic techniques and assessment that "Fibrosis First" was the most cost-effective and highly effective way to identify patients at risk. Roger points out that Fibrosis First looks very similar to what Mazen and Michelle recommend here. For this population, he asks how to identify lean patients that need this kind of workup.

Mazen states that the key marker is an extremely high ALT level (40 or 50 would not create this level of scrutiny, but 80 or 90 would.) Michelle agrees, but adds that most patients are diagnosed when they appear with some level of decompensating cirrhosis. As a result, she suggests that a primary care physician probably does not need to evaluate any patient they would not already consider at high risk: Type 2 diabetes or very high ALT levels. She also indicates that "Fibrosis First" sounds like an intriguing approach and that she needs to find Ian's poster or hear the podcast.

Michelle also comments that primary care physicians might be uncomfortable or intimidated screening for cirrhosis, which they do not feel qualified to treat. Louise asks whether they might be more comfortable assessing liver health vs. screening for cirrhosis.

Roger asks what else a hepatologist might take out of this paper. Michelle repeats their strong call for more research with this population, while Mazen points to Table Two in the publication as a comprehensive list of the various other conditions physicians need to screen for. Further, he notes that by placing the entire diagnostic pathway in a single chart, he empowers a hepatologist or gastroenterologist to approach primary care physicians and delineate which parts of the algorithm are the domain of which specialty. (The chart color codes this issue, with the pre-specialist elements in blue and hepatgologist elements in green.)

Finally, Mazen points out that when he sees lean NASH patients, he tries to place them in clinical trials, which can benefit the individual patient and generally advance our understanding of how to treat this condition.

Send us Fan Mail

As the NASH pandemic grows in the number and diversity of patient cases, one patient group receiving increased notice includes patients with "lean NASH," those with "normal" BMI levels (BMI<23 for Asians; BMI<25 for other racial groups). Last month, Gastroenterology published Best Practice recommendations for diagnosing and treating lean NASH. In this conversation, co-authors, Drs. Michelle Long and Mazen Noureddin join Louise Campbell and Roger Green to discuss the development of the one-page algorithm that lies at the heart of this publication.

This conversation starts with Mazen Noureddin pointing out specific elements of the algorithm "that we're proud of." He notes that they start by identifying two specific patient groups they chose to focus on based in part on cost-effectiveness analysis: patients with Type 2 Diabetes and patients over age 40. They rely on the dominant diagnostic paths, which they recommend following with FibroScan. They recommend confirming the diagnosis with MRI or biopsy if there is any doubt. Most important, they pull all this information into a one-page algorithm that incorporates recommendations for primary care or endocrinology (pre-diagnosis) with hepatologists (post-diagnosis for patients with NASH and Fibrosis Level 2 or higher). They also discuss other tests of more recent development: MAST, FAST, MEFIB, cT1 and other multi-parametric measures.

Louise compliments the publication for its simplicity and clarity. She also notes approvingly that it recommends repeat testing every 6 months to two years, depending on the level of fibrosis. This is a more frequent schedule than the 3-5 years most commonly recommended by other algorithms. She likes not only the increased frequency but also the stratification of test frequency based on the level of risk.

Michelle agrees strongly. She notes that with a 3-5 year follow-up recommendation, many patients will be lost to follow-up. Mazen goes on to note that with T2D patients, physicians check annually for eye, kidney or neurological complications. He asks why FIB-4 should not be added to that list. Louise identifies another benefit: patients are more likely to realize when they have not had their annual tests, whereas they are less likely to recall after a three-year gap. The rest of the conversation centers around reasons that a 6-12 month recommendation makes more sense while requiring little additional time or spending. When Roger suggests that a rationale for 3-5 year testing might be so that payers are not concerned about the likely downstream costs, Mazen notes that based on his research and assessment, annual will be cost effective.

Send us Fan Mail

As the NASH pandemic grows in the number and diversity of patient cases, one patient group receiving increased notice includes patients with "lean NASH," those with "normal" BMI levels (BMI<23 for Asians; BMI<25 for other racial groups). Last month, Gastroenterology published Best Practice recommendations for diagnosing and treating lean NASH. In this conversation, co-authors, Drs. Michelle Long and Mazen Noureddin join Louise Campbell and Roger Green to discuss how they came to undertake this assignment and to organize their thoughts.

The conversation starts with Mazen Noureddin discussing why he, Michelle and Joseph Lim wrote this paper (in his simple description, "we are asked for it," which, he notes, "we were quite excited about.") From there, Mazen and Michelle describe the actual process of organizing the article. First, they needed to align on a single definition of "lean" (they settled on BMI<23 for Asians and BMI<25 for other racial groups.) From there, they relied on clinical experience to develop a basic list of 15 best practice elements. These served as a foundation for their work. From there, they conducted an extensive literature review to clarify each element.

In response to a question from Roger, Mazen comments that there were few surprises in what they found, if any, largely because the work came from the co-authors' extensive clinical experience. Mazen talks about an experience with two Indian brothers to explain how straightforward the process can be, and also how detailed. They note that lean NASH is harder to diagnose and requires painstaking efforts to rule out more obscure diagnoses that might not come into play with overweight or obese steatohepatitis patients. As Mazen notes, "there is some art" in treating these patients.

Michelle shifts focus to the tables and charts they put together for the article. Her point: physicians are too busy to dive into the issues and possibilities the way the co-authors did. This made it important to provide clear, concise guidance with relevant data in a small number of charts and tables and a stepwise approach.

Louise Campbell joins the conversation to compliment the authors and comment on the clear synergy between these best practices and approaches for diagnosing non-lean patients. She then asks whether there are differences between different Asian populations. The conversation ends with Michelle stating that insufficient research on this issue exists and Mazen, while agreeing, noting that the relied on publications from India and other parts of Asia in building their practices.