And good afternoon, everybody. So I'll just begin by declaring that I do not have any disclosures. And I just want to start also by thanking the organizing committee for the opportunity to present on this case and to discuss the questions that it raises. So this is a case of a 59-year-old woman who had biopsy-proven cirrhosis due to chronic hepatitis C virus infection. She had been treated with direct acting antivirals and has had a sustained virological response. We're told that she does not have any viruses on AGD. Now, many patients with chronic hepatitis C virus infection either have or will undergo treatment with DAAs. And the vast majority of these would have sustained virological response rates that exceed 90%. And so this is really an important question and one that will be frequently encountered in practice. And there's several questions that can be raised. So first of all, what is this patient's risk of developing HCC? What tools are available to predict her risk? And what treatments may be available to reduce the patient's risk of developing HCC? And what I'd like to do is to discuss and address each one of these separately. So we'll start off with the first question. What's her risk of developing HCC? Now, we recognize that HCV-infected patients who have cirrhosis have a higher risk of HCC. We've known for some time that viral eradication by interferon can reduce the risk of HCC substantially. And also that the risk reduction that occurs is similar whether you have cirrhosis or not. We now also recognize that viral eradication, regardless of whether it's induced by interferon or by DAAs, will also reduce the risk of HCC substantially. Now, patients who do not have cirrhosis at the time of antiviral treatment have a very low risk of HCC after SVR. However, amongst patients with established cirrhosis prior to antiviral therapy, there is a substantial risk of HCC. And this risk of HCC will persist after SVR. And these patients present a conundrum because although the SVR has reduced the risk of HCC, there's still a residual absolute risk for HCC. And therefore, we need to consider continued efforts at surveillance for HCC for these patients. So why is this? Now, in the presence of chronic hepatitis C virus infection and cirrhosis, there are really two types of changes that are notable that are relevant to the cancer risk. The first is fibrosis. And the second is epigenetic changes, which I'll refer to as epigenetic scarring. And changes in histone 3K27 acetylation have been observed with chronic hepatitis C virus infection. And these changes have been shown to be associated with the expression of cancer-related genes and subsequently associated also with the onset of HCC. Now, following DAA treatment, the hepatitis C virus infection will likely be cleared. The fibrosis may improve. However, the epigenetic H3K27 acetylation changes persist. And there's continued expression of genes such as SOX9. And as a consequence of these changes, there's a persistent risk of cancer. And given the persistent risk of cancer, then surveillance for hepatocellular cancer is clearly warranted in these patients and those who have cirrhosis. However, in patients who do not have cirrhosis, we generally do not recommend any screening regardless of their risk, which is quite low. So our current guidelines recommend that all persons with hepatitis C virus infection and cirrhosis should undergo surveillance. Moreover, that surveillance should be continued for as long as they remain eligible for curative treatments. And the guidelines also advocate using the same screening strategy for all of these patients, and that is ultrasound with or without AFP every six months. Now, this is really a one-size-fits-all strategy. It's one that raises many questions in the DAA era. But it also provides us with some opportunities for improvement. And in particular, in addition to achieving SVR, we recognize that there might be individual patient-specific characteristics that could also impact on the risk of HCC. Thus, the next question would be, how can we predict her risk of HCC? Now, some people may have favorable characteristics and may have a low risk, whereas others may have adverse characteristics that give them a higher risk. And for these individuals, having more aggressive strategies, such as perhaps annual CT scans, might become more efficacious or more cost-effective compared to ultrasound. Now, to begin with, surveillance is thought to increase survival or become cost-effective in cirrhotic patients when the risk exceeds some 1.5% a year. Surveillance can result in early detection of cancer and provide a benefit. But we should also recognize that surveillance can also confer harm. So, for example, if you're not going to develop HCC or in the timeframe of interest, such harm could include things like unnecessary anxiety, unnecessary biopsies, imaging studies, or even treatments. And therefore, taking a more personalized approach could be useful for patients whose risk exceeds, perhaps exceeds a predetermined risk threshold. And there are several tools that are available by which you could try to predict her risk. These tools may include individual risk factors. They may include assessments of fibrosis. They may include circulating biomarkers. And they may include risk prediction models and risk scores. And what I'd like to do is examine the data behind some of these. So, in a recent single-center study from Milan, they looked at a cohort of patients with cirrhosis that were treated with DAs for hepatitis C. And in this cohort of 565 patients, 28 developed de novo HCC. And the study identified several factors, factors such as being male or having diabetes as having very high odds ratios for HCC. Likewise, they also found that the presence of fibrosis determined either using liver stiffness measurements or using the fibrosis force score were also predictors of de novo HCC. And indeed, several other studies have shown an increased risk of HCC with increased liver fibrosis and stiffness. So, a study from the interferon era showed a dramatic increase in HCC with increasing fibrosis assessed using a elastography. Likewise, a more recent study of 432 patients with cirrhosis also showed a significant increased risk of HCC with increasing fibrosis. And then, more recently, a multi-center perspective study of 258 patients treated with DAs, a liver stiffness measurement of greater than 27.8 kilopascals had an area under the ROC curve of 0.69. So, taken together, these studies indicate that the extent of fibrosis correlates with the risk of HCC. So, other than FibroScan, there are other ways that one can assess the extent of fibrosis. For example, the Fibrosis-4 score is a simple non-invasive marker of advanced fibrosis, and it could be used to predict risk. This score is derived from age, ALT, AST, and the platelet count, and has been validated as a measure of fibrosis. Now, of note, the individual components of this score, such as age and ALT, themselves have also been strongly associated with risk of HCC. Now, as shown here in the orange line, a score of greater than 3.25 is strongly associated with the risk of HCC. So, one could ask, what happens over time? Does the risk of HCC vary with time? And if so, how would that impact on your recommendations for surveillance? Now, the duration of time for which patients would remain at higher risk is really uncertain. A recent study, however, looked at this. They looked at the annual incidence of HCC after SVR, and they found that patients with cirrhosis continue to have a high risk for HCC for many years, and thus should continue surveillance efforts. Although there was a tendency towards a reduction in the fibrosis score with DAA use over a short duration of follow-up, there's several factors to consider here. So, for example, resolution of fibrosis could potentially reduce the risk of HCC, but this may not happen in some patients, particularly those who have advanced cirrhosis. And other factors, such as diabetes or obesity or alcohol use, could also mitigate any impact of the decreases in fibrosis. And moreover, it's also possible that some of the preneoplastic changes may persist indefinitely after SVR. So, even though you see some regression of fibrosis, it may not necessarily correlate with an absolute reduction in risk of HCC, and the risk is believed to persist over many years. Now, this particular study also looked at the effect of changes in the fibrosis-4 score. So, if you had a high score prior to treatment, and the score went down, then the risk was reduced. Whereas, if you had a low score to start with, and the score went up, the risk of HCC increased. And thus, the change in the score here after treatment may also be quite helpful. So, what about liver stiffness? Does the measurement of liver stiffness help us to predict the risk of developing HCC? Now, in a recent study, it suggested that a change in the liver stiffness after DAA therapy could be helpful in identifying and predicting the development of HCC. Now, this was a single-center, retrospective analysis of some 139 HCV cirrhotic patients. 14%, a high number, developed de novo, and some had recurrent HCC after treatment. The patients who developed de novo HCC in this study had a significantly lower reduction in their liver stiffness of more than 30% from baseline compared to those that did not develop HCC. So, the bottom line is that if you lower your liver stiffness, it might reduce your risk for HCC, but you may not completely eliminate it. Now, some studies have looked at developing models to predict the risk of HCC after SVR. And in a recent study from the VA, looked at more than 45,000 patients who received treatment for hepatitis C, of which 64% were DAA-only regimens, they identified 1,412 patients with HCC. And they identified several predictors for HCC, such as age, platelet, AST, and albumin. And based on some of these identified predictors, they developed risk stratification models to help to estimate the risk of HCC. And this graph shows the survival free from HCC that was either predicted from these models, as indicated in the dashed lines, or that was actually observed. And there's just a couple of things to highlight here. One is that there is a dramatic variation in the risk of developing HCC in these patients. And the second is that the models that have been generated could be quite useful to predict the risk of developing HCC. And being able to estimate the risk may be quite useful in making decisions regarding screening strategies. So let's take a look at how some of these predictive models could be used in practice. And here are some potential scenarios. These are three individuals, all of whom have had treatment for hepatitis C virus infection. The pretreatment information is listed here. And as you can see, it's quite varied. But based on the models, we could try to generate an estimate of the three-year risk of HCC. And the first patient, who had cirrhosis but did not have an SVR, has an extremely high predicted three-year HCC risk. And such patients might consider screening by CT or MRI or abbreviated MRI, a more aggressive screening and surveillance approaches. Patients with cirrhosis who achieve SVR could have a relatively higher three-year risk, or they could have, as in the second patient, or they could have a relatively lower risk, as in the third patient. And thus, the use of such tools may actually allow clinicians to be able to better estimate the risk of HCC in individual patients. So having consulted on the risk of cancer and advised on the appropriate surveillance approaches, the next question is, how can we reduce the risk of HCC in the future? And there's several ways that this could be approached. For example, we know that the risk of HCC is increased by being overweight or obese or by consuming alcoholic drinks or by consuming foods that are contaminated with aflatoxins. So first, we would start with modifying any risk factors for HCC. Now, modifiable risk factors here could include other viral infections, such as hepatitis B virus infection. They could also include other lifestyle factors, such as alcohol and tobacco. And then if there's associated diabetes or obesity or hypertension, then lifestyle changes or even medications to target metabolic syndrome could be considered. Next, we can consider preventive approaches. So the risk of liver cancer can be decreased by drinking coffee, by being physically active, and possibly by some dietary changes as well. And finally, it would be ideal, of course, if we were able to treat or reverse the underlying processes, such as fibrosis or even reverting the epigenetic changes that confer a higher risk of cancer. But we're not there yet with therapeutic options for that. So let's look at coffee. There have been several studies that have looked at coffee drinking and the risk of HCC. In a meta-analysis of some 12 prospective studies that were done with more than 3,400 cases of HCC, the consumption of coffee was found to reduce the risk of HCC, even with low amounts of consumption. And thus, coffee drinking to reduce the risk of HCC would seem to be a very reasonable recommendation to make based on these data. Another area to consider is diet. Many studies have looked at dietary factors and their association with HCC incidents. And in a meta-analysis of 19 such studies that involved more than 1.2 million individuals and more than 3,900 cases of HCC, it was found that an increased intake of vegetables but not fruit was associated with a lower risk for HCC. And other studies have noted some similar results. For example, an inverse association of vegetables with HCC risk was reported in a prospective cohort study from a multi-centered Western population as well. And then recently, results from the multi-ethnic cohort study have also been reported that identify a reduced risk of HCC with high-quality diets. Now, there's also been a lot of interest in specific diets, such as the Mediterranean diet. An analysis of two case-control studies from patients in Italy and Greece looked at this. They examined 518 cases of HCC and had 772 controls. But what they observed was a protective effect against HCC with a higher adherence to the traditional Mediterranean diet, which was assessed in these studies using a diet score ranging from zero to nine with the highest adherence. And so these results really suggest that there might be some potential benefits from adhering to a Mediterranean dietary pattern for patients who have chronic viral hepatitis. Let's move on to physical activity. And while several studies have suggested an association between physical activity and the risk of HCC, the evidence in the past has really not been very conclusive. However, recent analysis of the European Prospective Investigation into Cancer and Nutrition, or EPIC, cohort, looked at more than 460,000 persons, and they found that physical activity was associated with a reduced risk of developing cancers with a 45% lower risk of HCC in those who had high physical activity compared to low physical activity. And moreover, they noted that this risk was really independent of other risk factors such as age or gender or smoking status, body weight, et cetera. Similar observations have been made in other studies. So in the NIH AARP Diet and Health Study, an association between the frequency of vigorous physical activity and HCC was examined in more than 500,000 participants, of whom 628 developed HCC over a 10-year period. And again, in those with the highest versus those with no vigorous physical activity, there was a significantly lowered risk of HCC. And so these data do support the hypothesis that physical activity may decrease the incidence of HCC as well. And while these studies have demonstrated an association, they really have looked at physical activity at a single point in time. And in a recent study, they looked at temporal changes. This was also from the NIH AARP Diet and Health Study cohort. And they noted that moderate to vigorous intensity physical activity was correlated with a reduced risk of HCC similar to the other studies. But what they did was they also examined the patterns of change in physical activity, and they identified seven different trajectories using four different time points in the course at different age ranges, and then assessed the association between each of the trajectories and the risk of HCC. So compared to those individuals who had consistently low physical activity patterns, as shown there in the red lines, those who maintained activity over time had a 26 to 36% lower risk of cancer, whereas those who increased it over time had no associations. But those who decreased activity over time had a higher risk, although it was not significant. These results do suggest that sustained physical activity can be associated with a lower risk of HCC, while increasing physical activity later in life may not quite yield the same benefit. But clearly, further research is needed to understand the associations here between physical activity and the risk of cancer. So let's move on to medications. One of the medications that's frequently used in patients with diabetes is metformin, and several studies have examined the use of metformin. In a meta-analysis of several of these studies, there was an overall reduction in the risk of HCC noted with the use of metformin. I'll note that a similar reduction has really not been seen with the use of insulin, and thus these effects could be related to the drug, as opposed to underlying effects on glycemic control. Another medication that's frequently used are statins, particularly in the context of underlying metabolic syndrome, and the effects of statin use on HCC risk has also been extensively studied. In a meta-analysis that was performed a few years ago of several such studies, the use of statins was reported as reducing the risk of HCC with an overall odds ratio of 0.63. But the association between statin use and HCC has also been more recently examined in larger studies, in a study from Asia that involved a nationwide population-based study. A case-control analysis revealed a beneficial effect of statins in reducing the risk of HCC in persons with cirrhosis. A more recent study has also looked at the type of statin and has examined the influence of lipophilic versus hydrophilic statins on the risk of incident HCC and death. So this was a study in a nationwide cohort with a confirmed viral infection from Sweden. They used a prospective matched cohort to assess the relationship between statin use and HCC incidence or mortality. And compared with the matched non-users, the 10-year HCC risk was significantly lower amongst those who used lipophilic statins, but was not altered in those using hydrophilic statins. And the analysis from this study, when they looked at the odds ratio, they had an odds ratio of 0.54 with the users of lipophilic statins, such as atorvastatin and simvastatin. In contrast, there was no significant reduction in those with the use of hydrophilic statins, such as prevastatin. Now, the inverse relationship here between lipophilic statins and HCC seemed to be dose-dependent, but there were some limitations to this study. For example, it was a population-based study. There's no data provided on the extent of fibrosis, lipid levels, or the extent of HCC surveillance. But nevertheless, these are intriguing findings. So, conceptually, then, the risk of cancer could be reduced in several ways. This can include lifestyle changes, attention to diet, changes like physical activity, and potential use of medications that might be used if there's a concomitant metabolic syndrome. They can also include modifications of other risks for liver cancers, such as chronic viral hepatitis B and alcohol. And finally, hopefully, in the future, we'll have some ways to revert the epigenetic scarring to reduce fibrosis. And so, just in summary, then, key takeaway points are as follows. First of all, with regards to the risk of HCC, recognize that DEA treatment is associated with the risk of incident HCC, but that this risk will vary from one person to another. Next, risk assessment can be based on determining the extent of fibrosis or changes in fibrosis over time and possibly also using predictive models, such as the ones that have been reported. And based on these, the surveillance strategies could be personalized to an individual's predicted risk. And then finally, for risk reduction, preventive approaches to reduce the risk could include attention to lifestyle, viral, or metabolic factors. So, thank you very much for your attention. Thank you.

cirrhosis

hepatitis C virus

sustained virological response

hepatocellular cancer

fibrosis scoring

liver stiffness measurements