Interview with Prof Dr Sven Francque

Last year we interviewed Prof Dr Sven Francque one of the leading researchers worldwide in the field of NAFLD and NASH.

The FDA recently recognized the severity of fibrotic NASH, allowing for the conditional approval of drugs. What has changed with regard to NAFLD and NASH in the last 10 years from a clinical and epidemiological point of view that ultimately led to this decision?

The disease is not new, but we have better epidemiological data now showing us how big the problem is. The estimates are currently that 25 to 30% of the adult population has NAFLD. Not NASH but NAFLD, with quite important geographic differences. The highest prevalence is probably in the Middle East. Also, Hispanics are more prone to developing NAFLD and NASH. Caucasians are a little bit in the middle and then African Americans tend to have less severe NAFLD. So there is ethnic and geographical distribution but on average it's about 25 to 30% which is considerable. And the estimates are that about 5 to 10% of those people with fatty liver have steatohepatitis, which makes that you have a considerable percentage of the adult population that has steatohepatitis and hence is at risk of developing cirrhosis and hepatocellular carcinoma. What we also have learned is that although the data are not so easy to correctly interpret—because we have the confounding of obesity, diabetes, dyslipidemia, and so on—the arguments are quite solid that NAFLD also contributes independently to the development of cardiovascular disease—and that's also quite recent data—on fatigue of patients and their well-being, and their ability to work and to have social interaction. So the disease burden is probably way more than just the liver condition and end stage liver disease, but even if you restrict it to liver related complications, it's becoming the leading cause of liver transplantation and the leading cause of hepatocellular carcinoma. So because we have better epidemiological data, but also better calculations or projections of what it will be like in the next 2-3 decades, the regulatory authorities realized that this is a serious problem. On the other hand, if you go to clinical trials and you want to see results of clinical trials where you want clinical endpoints, that takes long, 10 or 20 years, and that’s if you only start off with patients that are already very advanced, as in being close to cirrhosis or already having cirrhosis. For the other patients that are not so advanced in their disease, it will take another 10 to years before they get to cirrhosis and decompensation. This means that if you take all the procedures into account you will have a drug probably approved in 30 years. With the disease that’s already now becoming the leading cause of liver transplantation and hepatocellular carcinoma, you can’t wait. And it's that combination that made the regulatory authorities accept conditional approval or accelerated approval—depending on whether you are in the American or the European context—based on surrogate endpoints on which we think will translate into clinical benefit.

With the current knowledge, is early detection of NAFLD and NASH possible?

Regarding the diagnosis of NAFLD, so not NASH, but just the presence in fact of fat in the liver, ultrasound is probably the best balance between being accurate, cheap and easily available. A diagnosis based on the combination of clinical parameters and simple lab tests, such as fatty liver index and some other scores is probably OK for epidemiological purposes, but it's not accurate enough to make a formal diagnosis for an individual patient, whereas ultrasound is also easy to apply, it's harmless, it's cheap, and it’s more accurate than those scores based on biochemical routine lab tests and clinical parameters. The disadvantage of ultrasound is that if you have only very low degrees of liver fat, you will probably not detect it, but it is debatable whether those minor degrees of liver fat are clinically important or not. In conclusion, if in ultrasound you clearly see steatosis then you have a quite reliable diagnosis.

I'm going to combine two questions in one. As a physician, what is the standard of care for NASH or NAFLD patients of your choice at the moment? And, since 9 out of 10 people with obesity suffer from NAFLD, do you reckon that a nutrition and lifestyle-based approach is feasible in practice?

That’s three questions in one [laughs]. The first point is that it’s important to make a distinction between what currently we call NAFLD/NASH. With NASH we mean steatohepatitis, which means that the liver is suffering from the accumulation of fat, or in relation to the accumulation of fat, so that you have inflammation and some degree of liver cell damage; that's the concept of steatohepatitis. But in daily practice we cannot work with concepts, we need a concrete diagnosis and so we have criteria to diagnose NASH and those criteria for the time being are based on what we see on the liver biopsy, where we need a certain number of lesions to make a diagnosis. If you have a patient with the diagnosis of fatty liver disease, first of all you have to make sure that it’s NAFLD and not something else, because there are other reasons to have fatty liver: alcohol, some medications, and a whole list of other causes. Second question is: do I have any indication about the severity? Is this just fat in the liver or is this a patient where there is an indication that there is steatohepatitis? We have scores for the diagnosis of fatty liver, but also scores to have an idea about the likelihood of having fibrosis in the liver. These last ones are far from being perfect, but they have a high negative predictive value, which means that if the result is reassuring, there is a high likelihood that there is nothing seriously going wrong with the liver. This was about the diagnosis. Disease management, of course, includes trying to get rid of the causes of steatohepatitis and steatosis, which is mainly overweight, the metabolic context. Then you incur the problem of lifestyle modification and trying to improve the metabolic context. We should acknowledge that our healthcare system is not so well equipped. I do not have the time in my consultation and it's also not, I think, my job to have a comprehensive treatment of all metabolic risk factors in patients so I need a team with a dietician, somebody who can give some counselling on physical activity, ideally somebody that can also work on the motivation of the patients to help them in achieving that, and that's the ideal world. Those elements are there but a holistic management system that helps patients doing that in a coordinated way may exist, but it's mostly not reimbursed by the healthcare system, so patients have to pay for that, which is then of course a barrier. It is sometimes done in obesity clinics in relation sometimes to bariatric surgery but it's mostly not supported by the Social Security system. And as such in only in place for those who can afford to pay and who are willing to pay. This is a highly preventable disease, and if it's there, it's for many patients in principle a treatable disease because if you are able to restore a normal metabolic situation, the liver will recover. That's the big advantage of the liver, which is a very robust organ with a very important restorative capacity, with a lot of repair mechanisms. If you have already atherosclerosis it can indeed improve if you improve the metabolic context, but it will not completely go away. That's not the case for the liver. The liver can really recover almost up to a normal liver if it's not too advanced already. That's in a way also a frustration, because there are many patients that could have been less sick if they would have had the support to change the metabolic context of their condition. We as physicians have to take care of the medical problems, but we would like to do much more preventive and early curative support rather than treating the patients when they are already quite down the road.

It is very much also a societal problem, in a way.

Absolutely, absolutely.

Going a bit more specific into the chemistry: you already wrote detailed reviews about this topic, but could you explain again briefly what makes PPARs attractive molecular targets for the treatment of NASH?

There are several reasons. First of all: these are nuclear receptors that are key in the regulation of a lot of metabolic processes, but also inflammation and also fibrogenesis so they are really at the crossroads and the link between all those different pathways that are important in the pathophysiology of this disease. You have almost to look at every individual cell type to have really a look at what the PPAR does in that particular cell, so it's very differentiated across organs and within the organs. It's even different according to the different cell types, as with many of those nuclear receptors. That's not only true for PPARs, but they have a wide range of effects that needs to be taken into account. What is important though is that they are quite key in metabolism, and they are one of the most important nuclear receptors in the function of the adipose tissue, and the adipose tissue is a very important driver of this disease. They are also very important in hepatocyte metabolism and in stellate cell metabolism, so they play a key role in several of the key cells involved in the pathophysiology of this disease. That's why their modulation is potentially quite powerful in treating this disease. However, as we have seen also with other drugs, since this is a very complex disease it takes sometimes a combined effect at different sites. I mean, you cannot say that a viral disease is simple, because you have all the host’s immunity complexity of the host-virus interaction, but what is easy in a viral disease is that your enemy is identifiable: you have a single well identified enemy. With alcoholic liver disease, you know the culprit: it's the alcohol. There's also a plethora of mechanisms that explain why alcohol is toxic and why one patient has a liver disease and the other one has neurological lesions, but at least you have an identifiable cause. In this disease there is not such a simple, easily identifiable enemy, but one of the main drivers is the dysfunction of the adipose tissue, and one of the key regulators of that, or one of the key factors that can be modulated to improve that is PPARs.

Imagine that we will arrive in a situation in hopefully 5-6 years where we have the approval for the first drug to treat NAFLD. What do you think the future generation of healthcare researchers and medicinal chemists should then focus their research on?

Especially in this complex disease, it's likely that there are different phenotypes of patients. Probably we are not able to identify them for the moment. We know that there are ethnic differences, and we know some genetics play a role but that's more or less it, still there are differences between people. You have people that are morbidly obese and still have only a tiny amount of fat in the liver and no steatohepatitis at all, and there are people that are almost as slim as you are and do have steatohepatitis, and the whole spectrum in between, so there must be differences. We know that there are altered pathways in some patients. For some, the main problem will be that the liver is not capable of coping with metabolic stress, in other patients it will be more the metabolic stress itself which is so important that the liver cannot cope with it, but for the moment we do not understand that heterogeneity of patients, and what we really need is to have a better understanding of that differentiation, which will help us also to target the treatment because some treatments will—by the background of underlying mechanisms—be more appropriate for some patients and less appropriate for other patients. For example, a patient that doesn't have that much adipose tissue dysfunction, but rather has a problem of some direct mechanisms such as metabolic pathways in the liver that are deranged, they are probably not going to benefit much from a PPARγ drug, or GLP1 that makes you lose weight, because the problem is not primarily there. That patient will probably benefit more from a drug that tackles intrahepatic mechanisms like the PPARα and also the PPARδ and γ mechanisms inside the liver, or probably that what is currently in the diet hormone receptor related mechanisms, for example. That differentiation and that heterogeneity mainly in underlying mechanisms is something that we do not understand for the moment, and that is something that we need to learn because that will help target treatment. Now we are developing treatments in a heterogeneous group of patients, which probably explains why you have responders and non responders and not always overwhelming results. That's for sure something where we should improve.

In clinical trials you mentioned two main problems: one is the endpoint and the other is this heterogeneity of patients that don't respond to the same treatment. Do you see other bottlenecks for the development of a new drug during clinical trials, in your experience?

Well, as you say, one of the one of the issues is of course the definition of the endpoints, which is currently based on the liver biopsy. The problem, of course, is that the liver biopsy is an invasive procedure, but on the other hand it gives you very detailed information about the liver, so it's a very powerful tool. However, by having the need of such a tool, you are restricted in the numbers of patients. You need specific expertise to perform the biopsies, to read and analyze them and so on. I think the liver biopsy will still be an important tool in early studies to have an idea how a drug works and how efficacious it can be, but especially in the phase III trials, I think we will need to work with other tools as surrogate endpoints. The clinical endpoints will always be a problem in this disease, so we need something to replace the biopsy or that can help us reduce the need thereof in order to be able to recruit and test larger groups of patients and have faster progress than we are seeing nowadays. The problem with the tools we are working with, is that their correlation with the liver biopsy is not perfect. You can measure liver fat with MRI for example and look at the decrease of liver fat. That will tell you something about the fat in the liver, but does that necessarily mean that the underlying inflammation is less? There’s plausibility for that, but as scientists, you know that even if it's plausible, it's not sure that it's really the case. We also know from trials that sometimes you see a lot of interindividual variation, and so if you have to take into account that margin of inaccuracy because of interindividual or intra individual variation, defining your endpoint is a potential problem. To sum it up, we have not yet discovered how exactly we can reliably assess treatment efficacy with something else than a liver biopsy. We do have some indications: measuring liver fat, measuring liver enzymes, measuring those fibrosis indicators, measuring the stiffness of the liver tissue which is a physical characteristic of the tissue, but to what extent they tell you the same as what the liver biopsy does and to what extent this will translate into clinical benefit, to be honest, we don't know.

One final very generic question: 10-20 years from now what do you think will emerge as one of the new main causes of mortality for European patients, assuming that we manage to solve one of the others?

That's a difficult one. If we can convince people to go to a healthier lifestyle, I guess that the impact of cancer will still be the most important reason why people will die. We are exposed to a huge amount of chemical substances in whatever we eat, drink, the air we breathe. Nobody sees it, but it's full of all kinds of particles that might have an impact on a lot of different mechanisms. Everything is so polluted, in fact, that even if we return back to the Garden of Eden I think the consequences of that will be with us for decades. and So I guess it will be... either cancers or other things that emerge from those chemical substances that we all massively absorb every day.

The interview was recorded in the office of Prof Dr Sven Francque and the text was edited and adapted for clarity by Lorenzo Cianni and Nicolò Filippi.