Good afternoon, welcome to the joint hepatobiliary neoplasia SIG and liver transplantation and surgery HAC symposium. We're giving you the welcome and introductory remarks. I am Dr. Jean-Marie Llebet from Mount Sinai, and I'm co-chairing these introductory remarks with Dr. Michael Shielsky. The topic of the session is on hot topics in the pathogenesis and treatment of liver cancer, both hepatocellular carcinoma and intrapartic carcinoma. Here you have the outline of the session, we'll have two parts. Part one will be addressing the concept of pathogenesis and treatment for early tumors, and we will have two moderators, Professor Jessica Zuckman-Rossi and Professor Michael Shielsky, and three presentations. First one, pathogenesis of liver cancer, targeting the FGF map kinase axis in the basal carcinoma and intraparticular angiocarcinoma will be addressed by Sandra Rebusio. Second presentation on resection versus transplant for the management of intraparticular angiocarcinoma by Julie Heimbach, and the third presentation addressing the question on are there any improvements in the management of HCC with local-regional therapies by Professor Laura Cooley. Second session will be devoted to trial design and management of advanced tumors. The moderators will be myself and Professor James Trotter. The first presentation will be focused on trial design and endpoints in advanced HCC, overall survival and progression-free survival, and I will address this topic. Second presentation will be on immunotherapy for HCC, current results and need of combinations by Professor Richard Finn. Third presentation on advances in precision oncology in intraparticular angiocarcinoma by Professor Greg Gorse, and finally, last presentation on the role of molecular tumor boards for the management of liver cancer by Professor Robin Kelly. Finally, we'll have a wrap-up and closing remarks that will be addressed by co-chairs Professor Jessica Zucman-Rossi and Professor James Trotter. I'd like to thank all of you for attending this session, and I am hoping that you will enjoy all these hot topics. Thank you very much. So good morning, everyone. I would like to thank the organizers for the invitation to talk about the role of fibroblast growth factor signaling in liver tumors and its therapeutic targeting. I have nothing to disclose. First, I will briefly introduce the FGF-FGFR signaling pathway. So this pathway is composed of three main components, the ligands, the receptors, and the co-receptors. FGF ligands are secreted growth factors and include 22 members. They are subdivided into seven subfamilies and are classified in three main categories according to their mode of secretion, paracrine, endocrine, or intracrine, and they will interact with FGF receptors that are tyrosine-carnate receptors and that are encoded by four different genes. So to be fully active, FGF receptors need a co-receptor. Heparin-sulfate proteoglycans interact mainly with FGFR1, 2, and 3, and paracrine FGF, while cloto-, alpha-, or beta-co-receptors interact with FGFR1 and 4, and endocrine FGF. So the activation of FGF receptors will induce several downstream signaling pathways that are involved in sense survival, migration, and angiogenesis. So an aberrant activation of this pathway has been involved in many human cancers, and this occurs through different mechanisms that are summarized on this slide. First, gene amplification can lead to an abnormal accumulation of the receptor or the ligands. Or activating mutations can also occur in the extracellular domain of FGF receptor, leading to its constitutive dimerization or an increased ligand affinity. But mutation can also occur in the tyrosine kinase domain and induce constitutive dimerization and activation of the receptor. And finally, the third mechanism involves chromosomal translocation, resulting in the fusion of the tyrosine kinase domain of FGF receptor with another signaling protein containing dimerization domain that will induce ligand-independent receptor dimerization and activation. So in human cancers, aberrant activation of the FGFR pathway occurs mainly through genetic alterations of FGFR receptor that were found in 7% of the cases with FGFR1 as the main alter gene. And these alterations represent the main mechanism of activation, followed by activating mutations and translocations that are less frequent. So FGF-FGFR alterations have been identified in almost every type of malignancy, and liver cancer is among the most commonly affected, with FGFR2 fusions as the main alterations identified in intrahepatic cholangiocarcinoma and FGF19 amplifications that were mainly identified in around 5% of hepatocellular carcinoma. So the oncogenic role of FGF19 in HEC was initially suggested by the analysis of regions of recurrent amplification in tumors, and this analysis identified a minimal common region of focal amplification on chromosome 11q, including FGF19, but also another well-known oncogene that is cyclin D1. And as you can see, this amplification was associated with FGF19 overexpression. So FGF19 amplifications have been identified in a small fraction of HEC that belong to the proliferation molecular subclass, and this genetic alteration is associated with more aggressive tumors. So what is the functionality of FGF19 in the liver? So FGF19 is the ligand for FGFR4 and the co-receptor cloto-beta, and this pathway is critical for bile acid homeostasis. So in physiological conditions, FGF19 from the gut, that is under the transcriptional control of FXR, activates FGFR4 in hepatocytes to inhibit bile acid synthesis. So in normal conditions, FGF19 is not expressed in hepatocytes. By contrast, in hepatocellular carcinoma, overexpression of FGF19 through gene amplification or unknown mechanism leads to autocrine and paracrine hyperactivation of tumor cells and dependence on FGFR4. So the oncogenic effect of FGF19 has been largely demonstrated in different experimental models. The first study showed that ectopic overexpression of FGF19 in mouse skeletal muscle was able to induce the development of HEC in around 50% of mice by 10 months of age. More later, another study showed that FGFR4 inactivation in the same transgenic mouse model was able to inhibit HEC development, providing a validation of FGFR4 dependence in FGF19-driven HEC. But also another elegant study from the group of Scott Law also showed the oncogenic potential of FGF19 using an original mouse hepatoblast model. So in this model, an immortalized line of murine hepatoblasts, lacking p53 and overexpressing MYC, was produced with a vector containing FGF19 and a GFP tag. And hepatocytes overexpressing FGF19 were then transplanted into the liver of mice by intrasplenic injection. And as you can see, following transplantation, GFP-positive tumors developed within 8 weeks, showing that they arose from the FGF19-transfected hepatoblasts. Moreover, other experiments in HEC cell line amplified for FGF19 also showed that inhibition of FGF19 in vitro by shRNA or in vivo using a neutralizing antibody was able to block very efficiently tumor cell growth. So now I will come back to the genomic landscape of FGF, FGFR, and biliary cancers. Many studies have reported genetic alterations in this pathway, but here I will present only one of the most extensive studies that has been done to date. So this study includes a large series of 377 tumors representing the spectrum of biliary cancers. FGF, FGFR genomic alterations were analyzed on the whole series, and the results showed that these genomic alterations are enriched in intrahepatic cholangiocarcinoma and were identified in 30% of this tumor type. FGFR2 was the most frequently altered gene, mostly by fusions in 85% of the cases. FGF19 amplifications were also detected in 11% of the cases. As you can see, FGFR2 fusions were identified quasi-exclusively in intrahepatic cholangiocarcinoma and represent a null mark of this tumor type, so this association was also confirmed by multiple other studies. So at the clinical level, FGF, FGFR genomic alterations in biliary cancers are associated with more indolent disease scores. Indeed, as you can see, they occur in younger patients with an earlier tumor stage and showed also a longer overall survival. In terms of molecular surpluses, FGFR2 fusions are found in both the proliferation and the inflammation class of intrahepatic cholangiocarcinoma, while FGF19 amplifications that are less frequent are preferentially associated with the proliferation class. So now I will come back to the functionality of FGFR2 fusions that is the most frequent FGFR2 alteration identified in intrahepatic cholangiocarcinoma. So these fusions were firstly identified 7 years ago and involved BQAN as a fusion partner. So even if BQAN is one of the most frequent fusion companions for FGFR2, multiple other partners have been also identified, with more than 30 reported to date. So the genomic breakpoints are always located within the same intronic region. All of these fusions contain the same portion of the N-terminus of the FGFR2 receptor, including exon 1 to 19, with an intact tyrosine kinase domain, and are fused to different C-terminal partners containing a dimerization domain, and the consequence is the autodimerization of the receptor and its constitutive activation. The oncogenic potential of FGFR2 fusions was proven in several experimental models. Here two frequent FGFR2 fusions were analyzed, and as you can see, overexpression of these two fusions are able to transform NIH333 cells, both in vitro in soft agar assay, but also in vivo in subcutaneous xenograft models. And interestingly, in these models, inactivation of the kinase activity of FGFR2 is able to reverse the oncogenic potential of the fusions. And finally, in terms of signaling, Western blot analysis showed that fusions induce constitutive phosphorylation of FGFR2 and activate selectively the MAP kinase pathway that is downstream of the receptor. And interestingly, FGFR2 fusions can be detected by fluorescent in-situ hybridization, and as you can see, the fusion can be visualized by the co-localization of the probe that recognize FGFR2 and the probe that recognize the fusion partner that is here, PP-HLN1. So as I showed you, the oncogenic activation of FGF-FGFR plays a crucial role in a subset of liver cancer. So now, how to translate these findings into therapeutic targeting? So several preclinical studies showed that targeting FGFR4 could be efficient in HEC with FGF19 amplification. So first, in a transgenic mouse model of HEC overexpressing FGF19, as you can see, inhibition of FGFR4 with a neutralizing antibody is able to inhibit tumor growth. And the first selective irreversible inhibitor of FGFR4, the BLU9931, was developed 5 years ago, and its evaluation in a panel of 10 HEC cell lines showed an efficient inhibition of cell viability, but only in cell lines overexpressing FGF19, but also keeping a high level of expression of FGFR4 and Cloto-beta, meaning that a full functional pathway is required for sensitivity. It was then confirmed in vivo in tumor xenograft models that inhibition of FGFR4 could block tumor growth in tumor cells overexpressing FGF19 with or without gene amplification. So as you can see, in this study, there is not a good correlation between the copy number of FGF19 and its expression. And recently, by analyzing a large panel of 34 liver cancer cell lines, we confirmed that amplification of FGF19 leads to its overexpression only in cell lines that retain an hepatocyte differentiation. And similarly, the other partners of the pathway are more expressed in more differentiated cell lines. In this previous work, we confirmed in our study that FGFR4 inhibition was only effective in tumor cells amplifying and overexpressing FGF19, but also retaining expression of FGFR4 and Cloto-beta. So one of the main conclusions of these two works is that FGF19 amplification alone is not a good predictor of sensitivity to FGFR4 inhibitors. Several preclinical studies have also evaluated the impact of FGFR inhibition on the oncogenic activity of FGFR2 fusion proteins. So these studies in NIH3C3 cells showed that FGFR inhibitors block efficiently phosphorylation of the fusion protein and the downstream MAP kinase activity and are able also to reverse oncogenic properties of FGFR2 fusion proteins. So based on these preclinical findings, multiple FGFR inhibitors have been developed and are in clinical trial in hepatobiliary cancers, and I will discuss some of them. Last year, the first phase 1 clinical trial in patients with advanced HEC shows clinical benefit of the FGFR4 inhibitor Fisogatinib in patients with aberrant FGF19 signaling with an overall response rate of 17% in FGF19 positive patients. And as you can see, no response was observed in FGF19 negative patients. However, acquired resistance was observed in some HEC patients with an initial objective response, as you can see here, in these two patients who progressed under treatment. And in these two patients, FGFR4 mutations were detected in the circulating tumor DNA following Fisogatinib treatment, and these mutations have been shown to confer resistance by reducing the binding of the inhibitor to the receptor. So now in patients with cholangiocarcinoma, a recent phase 2 trial evaluating Pemegatinib, that is, the FGFR1,2, 3 inhibitor shows clinical benefit, specifically in patients with FGFR2 fusion with an objective response rate of 35%. And as you can see, no response was observed in patients with other FGFR4 alteration or no alterations, which show a poor progression-free survival. However, again, some patients with an initial response develop later resistance, as you can see in these two patients. And this was due to the occurrence of multiple independent point mutations in FGFR2 that were not detected at baseline, suggesting that high molecular heterogeneity is associated with acquired resistance. And this was confirmed in this patient with different FGFR2 mutations identified in three independent models, and intratumoral heterogeneity was also observed in the tumor 2 that show FGFR2 mutation only in one part of the tumor. As previously described for FGFR4, mutations in FGFR2 confer resistance by reducing the binding of the inhibitor to the receptor. So, how to overcome resistance? In intrapathic cholangiocarcinoma, or Boring FGFR2 fusion, a recent study showed that the treatment with a second FGFR inhibitor that fits covalently to the receptor provides clinical benefit after resistance to the first ATP-competitive FGFR inhibitors. As you can see here, in these three patients showing a partial response or stable disease with a second inhibitor, but still all the patients progressed. And again, multiple FGFR2 mutations were detectable in their plasma, suggesting that the second inhibitor is not completely able to overcome on-target resistance. But interestingly, in proclinical models of intrahepatic cholangiocarcinoma, oncogenic activity of FGFR2 fusion seems to be specifically supported by the activation of the MAP-KNS pathway, as you can see here, with a strong inhibition of ERG phosphorylation following inhibitions of FGFR2. So, MEK inhibitors such as trametinib that is already approved in melanoma could be an interesting alternative therapy to overcome resistance. And accordingly, in HEC, using our preclinical models of 34 liver cancer cell lines, we showed that cell lines who responded to FGFR4 inhibitors were also highly sensitive to the MEK inhibitor trametinib, suggesting that this therapy may overcome resistance. So, in conclusion, the FGFR axis plays a critical role in a fraction of hepatobiliary cancer. Targeting this pathway is promising, both for the treatment of hepatocellular carcinoma and intrahepatic cholangiocarcinoma. However, currently, these therapies are still limited by acquired resistance that is related to tumor molecular heterogeneity, and MEK inhibitors may be an alternative option to overcome resistance. Thank you for your attention. Hello. I'm Julie Heimbach. And today, I would like to speak to you about resection versus transplant for the management of intrahepatic cholangiocarcinoma. I'm currently Professor of Surgery and Chair of the Division of Transplant Surgery at the Mayo Clinic in Rochester, Minnesota. And I have nothing to disclose. What I'm going to speak to you about today is first some background and outcomes of resection for intrahepatic cholangiocarcinoma. Then, we're going to speak about liver transplantation and the setting of small, unresectable intrahepatic cholangiocarcinoma. And finally, we're closing with the novel use of neoadjuvant chemotherapy followed by liver transplant in patients with large, unresectable intrahepatic cholangiocarcinoma. So first, the incidence of intrahepatic cholangiocarcinoma worldwide for the countries where it is known, you can really see a sort of a striking difference highlighted with the sort of three different colors with the light blue being the infrequency. And really, particularly, this is the most prevalent in Southeast Asia, particularly remarkable in Thailand and remarkable in this specific area of Thailand in the Northeast where the incidence is more than 85 per 100,000 compared to the low incidence where it's less than 1 per 100,000. The diagnosis of intrahepatic cholangiocarcinoma, of course, underlying risk factors such as clonarchus, which is the issue in Southeast Asia, presence of colodocal cysts, PSC, hepaticolithiasis, as well as actually the same risk factors for the development of hepatocellular carcinoma such as cirrhosis, viral hepatitis, obesity, and alcohol, although hepatocellular carcinoma is more common than intrahepatic cholangiocarcinoma. The use of cross-sectional imaging is really essential in establishing the diagnosis. And the difference for this is what we compared to hepatocellular carcinoma is in this case of intrahepatic cholangiocarcinoma. We see peripheral rim arterial phase enhancement, as is shown here in this image, which is then followed by centripetal hyperenhancement, which comes in slowly in the venous and with blade phase images as opposed to the washout that we typically see in these later phases in hepatocellular carcinoma. We also may see elevated CA99 in some patients. If it is important to differentiate between hepatocellular carcinoma and cholangiocarcinoma, then we would need to perform a biopsy, as the radiographic imaging is not satisfactory to clearly differentiate between these two tumors. And even in a biopsy, we have to be aware of the possibility of sampling error for those tumor types. So once we've established the diagnosis, how can we treat it? Well, the gold standard would be resection. But unfortunately, this is possible in only a minority of cases, about 30 to 40%. And those that are able to have a curative resection, their five-year survival can be expected to be around 40%. This is impacted by the presence of nodes, which can reduce the five-year survival significantly. So knowing that surgical resection is the mainstay, with the predictors of outcome being the same as they are in nearly all other surgical therapies, such as the ability to complete a R0 resection, whether there are multiple lesions and whether there are nodal metastases, this leaves us with the situation that resection may not always be possible either, because of the very large size of the tumor and the setting of the neurons patient. Or in the patient with cirrhosis, they can have the underlying cirrhosis, which may prohibit resection. And unfortunately, recurrence is common. So leading to the question of whether transplant can be considered in early-stage enteropathic chondrocarcinoma. So this series from Sap Sochan et al in American Journal of Transplantation in 2014 is the first really report to pose this question of whether the transplant should be reconsidered in these very early patients. This is from Spain. They had a multi-center retrospective series looking at a large number of patients who are undergoing transplant for HCC. And of these, there was a small number of patients, actually just eight patients out of more than 2,000, who were actually found to, instead of having HCC, having enteropathic chondrocarcinoma. And for this small subgroup, they had a 73% five-year survival versus 45% overall. So the 29 overall enteropathic clanges that were thought to be HCC at transplant, the full group had a worse survival when it was more than two centimeters, but that subgroup had acceptable survival. A larger international series, so not just Spain, they were able to gather 48 intra-opathic and 33 mixed types. And of that cohort, which had 81 total patients, there were 15 with these very small, less than two centimeter intra-opathic chondrocarcinoma, and then there were 32 with the larger tumors. So the very early group, again, had satisfactory survival at 65% for five years. And so this has led to the development of a prospective trial. And this is ongoing, although there has been some challenges in identifying and enrolling these. So this is another paper, again, actually coming to that question of liver transplant or resection for small enteropathic chondrocarcinoma. And this is from France, retrospective again, multi-center, looking at intra-opathic clangio with mixed type or pure intra-opathic clangio or resection. And what you can see there, the light bar is the liver transplant group showing superior outcome compared to those, the dark bar, which underwent resection. And looking at outcomes for both less than two centimeters and the subgroup of two to five centimeters. And this is just highlighting the experience for the liver transplant alone, both observed and recurrence-free survival, which is shown on the right. And the dark bar is the smallest, less than two centimeter, whereas the lighter is the five centimeter. Again, in both of these groups actually showing acceptable survival. This is a center from New York, single center, looking at 32 patients. There were nine with separate intra-opathic clangio as well as another tumor that was HCC, seven with just intra-opathic clangio and 12 with a mixed subtype. Nearly all of them had thought to be HCC with 28 having a preoperative established diagnosis of HCC and just four with a known intra-opathic. Many of these patients received the same treatments that we offer to patients with HCC and the overall survival was 71 and 57%. Those with the smaller lesions had a 78% five-year survival. So this is breaking it down using the Milan criteria that they technically don't apply to intra-opathic, but those with the smaller lesions shown on top with the five-year survival above 75% in this group of patients undergoing transplant for what was thought to be HCC but in retrospect was intra-opathic. This paper actually comes from Mayo Clinic in Jacksonville. Again, a retrospective analysis of patients transplanted for what was thought to be a primary liver tumor thought to be HCC. So it was 618 patients including those with known tumors and those with incidental tumors. There were 12 in this series with very early intra-opathic phalangeal carcinoma compared to 314 with HCC that was within Milan. All of these were established based on imaging. The five-year survival in this series was 64% for those with intra-opathic phalangeal versus 70% for those with hepatocellular carcinoma. Recurrence was more common in the intra-opathic group compared to the hepatocellular carcinoma group. So overall, multiple smaller series is satisfactory. What about the larger intra-opathic phalangeal carcinoma patients? Those with liver transplant, can it be used for locally advanced intra-opathic phalangeal carcinoma that is not resectable? And this is a really provocative series published in Lancet Oncology in 2018 from the group in Methodist, Houston. This is a single center prospective analysis of six patients with large intra-opathic phalangeal carcinoma treated with an aggressive regimen of neoadjuvant chemotherapy followed by liver transplant. The average tumor burden was 10 centimeters with patients having four lesions on average. They were treated primarily with gemcitabine and cisplatinin, although there are also other agents used. Really this can show at five year survival, really a compelling analysis. So again, this is a very small number of patients and only at one center with data available thus far. I do know that there are ongoing efforts to develop a multi-center prospective trial with this same applied inclusion criteria and therapy across multiple centers. So we'll need to await these data to have more information about the efficacy of this treatment strategy. However, there are challenges for applying a transplant to intra-pathic phalangeal carcinoma. One, of course, being allocation questions. So for hepatocellular carcinoma, there is a standard malexception score for patients. Now that is restricted to those with an AFP less than a thousand, although you can treat it and get it down under that. The HCC needs to be within Milan or beyond Milan with downstaging criteria. If they're downstaged successfully, they would be able to receive a mild exception after a six-month period of observation. However, their score is now fixed at maiden mild at transplant minus three. Similarly, for perihilarclangial carcinoma, there's also a standard mild exception for early unresectable perihilarclangial carcinoma, those with no metastases and a radial diameter less than three centimeters treated with neoadjuvant chemo radiotherapy. And they also would receive a fixed score, medium mild at transplant minus three. However, for intra-pathic phalangeal carcinoma, there is currently no standard mild exception for this group. There could be an option to use transplant based on their calculated mild score, or potentially you can use a living donor liver transplant for this patient population. It would be possible to appeal to the National Liver Review Board to request a score. However, there is no current policy or guidance regarding the use of transplant for intra-pathic phalangeal. So this would really be a challenge for the NLRB to approve given that lack of really any guidance on this topic. And this is an excellent paper from Mazzuferro et al, just published in Journal of Hepatology, that really kind of walks through all these issues of transplant and resection for intra-pathic phalangeal carcinoma. So this would be looking at the various factors such as eligibility. So for patients who are being considered for transplant, it's really essential that we would establish the diagnosis, whereas those that are having a resection, sometimes it is useful to have established the diagnosis, but it's certainly not mandatory if you've already made the determination that the patient needs and is eligible to undergo resection. CA-19-9 and the role really still remain to be defined in the setting of transplant for intra-pathic phalangeal carcinoma, but in this paper in particular, highlighting the consideration of less than 500 in the absence of cholangitis being an important factor. The size of the tumor for transplant, what has been established thus far, is really a potential role of transplant in the unresectable lesion less than two centimeters as really just being a possibility. So very early tumor in the setting of cirrhosis that cannot be resected, whereas in resection, the size really is not so much of a factor as long as it's able to be resected with a negative margin and there's adequate liver remnant. So in the setting of transplant, we really haven't figured out what multifocality means. The data so far is just for a single less than two centimeter tumor, and for resection, this would be a relative contraindication. Lymph node assessment, it's essential in transplant to be evaluated and certainly also recommended in the setting of resection given the impact on outcome. Patients with chronic liver disease would not be restricted in any way from transplant, whereas this can be a significant factor that can prevent safe resection. The role of neoadjuvant therapy for very early tumors is experimental. It is certainly also experimental for those with large enteropathic lesions who are also being considered for transplant, as in the Methodist experience. It is also actually being used in an experimental fashion in the setting of resection. So the key questions at this point in time, I think, are that most of these series have identified patients based on explant findings. These are patients that they thought had a postular carcinoma and actually they had enteropathic glandular carcinoma. What was the pre-transplant tumor stage based on the imaging, and in particular, were there any differences on imaging characteristics for these enteropathic glandular carcinomas that were thought to be HCC? Also, we can see that the reported series really are including only a very small proportion of patients. Usually, this is just a handful of patients out of a very large number transplanted for HCC. Is this because most of the enteropathic glandular carcinomas are actually identified but not referred because they are known to be enteropathic glandular carcinoma and thought to be not eligible for transplant? Or is it actually rare that this tumor is diagnosed at an early stage, early enough that could be considered for transplant? I think this is an open question. The key takeaways for the consideration of transplant versus resection for intra-epatic cholangiocarcinoma are that resection is currently the standard of care. Patients with small intra-epatic cholangiocarcinoma less than two centimeters may have satisfactory outcomes following liver transplant. This is based thus far on retrospective data. The role of neoadjuvant chemotherapy and liver transplant for large intra-epatic cholangiocarcinomas in non-sorbetic patients really remains to be defined. Access to transplant in the United States would generally require adequate calculated MELD score or living donor liver transplant until it can be established as a suitable indication for a MELD score exception. Thank you very much for the opportunity to speak. I would like to thank the organizers for the opportunity to speak on improvements in local regional therapy and HCC. My name is Laura Kulik and I am a transplant hepatologist at Northwestern. My area of interest has been in the therapy for HCC, particularly the role of local regional therapies as a bridge or downstaging to transplant. These are my disclosures. In the words of a well-known educator, innovation is applied creativity. By definition, innovation is always about introducing something new or improved or both. And it is usually assumed to be a positive thing. This is a recent editorial on the milestones of HCC that highlights the multiple systemic agents that have been approved for HCC in the last few years. The authors recognize that other important advances have not been noted in this figure. I will focus my talk on the use of radiation and the treatment of HCC with radioembolization and SBRT across the BCLC staging system. The premise of radiation segmentectomy is the delivery of high doses of radiation to one or two segments of liver with the goal to induce a sustained local response and atrophy of the treated segment. Lewandowski et al. reported the results of 70 well-compensated patients who did not undergo subsequent OLT or resection treated in this manner. All patients had a single lesion less than 5 centimeters and a target dose of greater than 190 gray, which has been previously described as an ablative dose. The overall survival is based on the initial size of the untreated tumor with a five-year survival of 75% in lesions up to 3 centimeters. The results of this cohort were then compared to other accepted curative therapies, such as RFA, resection, and transplantation. The authors concluded that radiation segmentectomy leads to a response tumor control and survival comparable to other curative options in early HCC with preserved liver function. This is an ongoing proof of concept study from Indiana University comparing SBRT to radiation segmentectomy with Y90 in previously untreated patients with a single lesion up to 3 centimeters without vascular invasion or metastatic disease. In addition to a lack of clinically significant portal hypertension and preserved liver function, patients with cirrhosis require an adequate future liver remnant. This is generally accepted to be at least 40% in order to undergo resection in patients with cirrhosis. Y90 has been used in a lobar fashion to treat the tumor while simultaneously inducing a hypertrophy in the contralateral lobe, which leads to an increase in the future liver remnant. Hypertrophy has been reported to occur as early as one month post Y90 and the degree of this hypertrophy increases over time with immediate increase from baseline to 45% at nine months. This has subsequently been supported by other studies, which is described in a systemic review. This study reports the outcomes in patients treated with Y90 for HCC and subsequently resected. This is a single center retrospective trial that consists of 31 patients. The majority of these patients were BCLC-A and a greater degree of hypertrophy was seen leading to an increase in FLR and those who received radiation lobectomy compared to radiation segmentectomy. The mean time from receipt of Y90 and resection was 2.9 months and 29% of patients recurred after a median time of 34 months, which was significantly higher in patients who were non-responders as well as those who had less than 50% of necrosis on X-plant. The overall survival post resection at three years is 86%. A limitation of the study is that it only includes patients who had been resected and therefore is not an attempt to treat study. This is an ongoing trial that looks at patients intended to be resected after Y90 therapy. Cohort one will be those treated with standard of care Y90 and will be used to inform the optimal dose that will be used in cohort two utilizing PET-CT. This will give some insight into the rate of success of getting patients to transplant after Y90. There are currently no randomized controlled trials that compare SBRT to other forms of local regional therapy. Most of the reported data on SBRT and HCC is focused on local control and generally in patients unsuitable for a refractory to standard local treatments. This is a pooled analysis of 297 patients from University of Toronto and University of Michigan that were treated with a definitive intent over a period of 13 years. These patients had no evidence of vascular invasion. They were not enrolled if there was an intent of SBRT to bridge to transplant. However, 25 patients were ultimately transplanted. The primary endpoint was overall survival from the day of SBRT. This highlights some of the baseline characteristics. There was a small percentage of patients who were decompensated. The majority of patients had undergone liver local regional therapy before SBRT and were intermediate or advanced according to BCLC due to the presence of being performance status one or greater. The median size of the treated tumor was 2.7 months for the entire cohort. The overall survival in this heterogeneous cohort was 25.6 months. There was an improved overall survival of 37 months in those who have preserved liver function and performance status compared to 23 months in the patients who did not have preserved performance status and liver function. The significant variables that impacted overall survival were subsequent receipt of transplant, AFP less than 10, ECOG of zero, and child's pew classification of A. This study also reported all forms of recurrence at one in five years. The highest rate of recurrence was related to the development of new lesions outside the radiated area. It is also important to note that in the follow-up of two to three years, there was no significant dose effect of radiation that impacted local recurrence, which highlights the radiation responsiveness of HCC to SBRT. SBRT is used as a bridge transplant. However, there has been limited data comparing SBRT to other forms of local regional therapy. This is an intent to treat analysis from a single center in Toronto where there is no upper limit of tumor size or number for OLT candidacy. Patients receive SBRT, TACE, or RFA based on the decision of the tumor board. The dropout rate was not significantly different between the two groups. And of note, patients that receive SBRT were deemed ineligible for TACE or RFA or had progressed after receiving either of these treatments. HCC recurrence was highest in the TACE group followed by SBRT and lowest in the RFA group. In the intent to treat analysis, overall survival from listing was not significantly different in three groups, nor was overall survival from the time of transplant. The authors concluded that SBRT can be used safely as a bridge to transplant in HCC when other conventional therapies are not applicable. And it may offer an advantage in patients who have liver dysfunction. This is a small prospective study of 23 patients using SBRT and decompensated cirrhosis up to Child's Pugh C10. The median tumor size among Child's Pugh B and C patients range from 2.6 to 4 centimeters. The majority of patients have received chemo embolization prior to SBRT or in combination with chemo embolization. In the middle panel, we see there was no detectable difference between local control and overall survival based on baseline Child's Pugh score. Child's Pugh class progressed and 43% of patients remained stable in 30% and improved in 22% of patients, which was associated with an improvement in overall survival. This chart highlights the comparison of SBRT to other therapies, most being retrospective in nature and showing comparable outcomes in terms of local control. The bottom study is a randomized controlled trial of SBRT plus TACE versus monotherapy serafinib, which showed superior time to progression and overall survival in the combination group with SBRT. A prior randomized control trial called PREMIER reported a significant improvement in time for progression in patients treated with Y90 compared to TACE. More recently, the trace trial results were reported at SIR. This is a randomized control trial of Y90 versus drug eluding beads in patients who are BCLC A or B, but they also allowed for ECOG1 and segmental portal vein thrombosis. Primary endpoint was time to progression at two years, which was assessed in a blind fashion. The time to progression as well as time to local progression overall survival was significantly improved in the Y90 group. Additionally, the safety between these two therapies was comparable. This study shows that Y90 was superior to drug eluding beads in this cohort. Highlighted in the red box is the expansion of agents available for treatment in and via stage of HTC. While the natural history of HTC without a curative option is to progress to the right, there are patients, even those with advanced stage with portal vein thrombosis who can transition to an earlier stage that may lead to a curative therapy. There have been two head to head randomized control trials of serafinib versus resin Y90 that did not meet the primary endpoint of improved overall survival and therefore are negative studies. On the right are proposed shortcomings of these studies that may have led to a negative trial result. Some of these are the inclusion of patients with main portal vein thrombosis, which portends a dismal prognosis as well as a lack of experience in the use of Y90 and no use of boosted Y90. The greatest transplant benefit can be seen in patients with more advanced HTC after proper patient selection based on radiographic and AFP response. This can even be seen among patients who are advanced with portal vein thrombosis. Mazzaferra and colleagues have devised a prognostic scale that can be utilized to predict survival as well as risk of tumor progression and liver decompensation within three months of Y90. The three factors that are assessed in assigned points include location of portal vein thrombosis with exclusion of main portal vein, tumor burden, and bilirubin level. Based on this model, median overall survival ranges from 32.2 to 7.8 months in patients with PBT. This can be used to appropriately select patients with the best chance of a favorable outcome when treated with radioembolization. Improvement in overall survival, particularly among patients with portal vein invasion, may be feasible when utilizing a personalized approach. A tumor threshold greater than 205 gray has been shown to result in improvement in overall survival as well as response rates. The MAA scan prior to Y90 is used to verify the distribution of glass beads to determine if there is a predicted off-target delivery of radiation as well as to be used to boost the radiation dose within the tumor to a goal of greater than 205 gray. In this study, 85 patients showed an improvement in response rate when they received a boosted dose of radiation. However, there are patients that a boosted dose of radiation should not be considered due to increase of liver toxicity, which is highlighted here. DOSIFER was a randomized controlled trial performed in France that demonstrated that the personalization of treatment with radioembolization matters in advanced HCC patients. Patients were randomized to standard dosimetry versus personalized dosimetry, which targeted a minimum of 205 gray delivered to the index lesion while keeping the dose delivered to non-tumorous tissue under 120 gray. Patients had to have at least one tumor greater than 7 centimeters or larger, and one of the exclusion criteria was poor tumor or PV uptake on the MAA scan. Patients were evaluated in an intent to treat fashion. The baseline characteristics were similar between the two arms with approximately three quarters having PVT and the mean index lesion being 10 to 11 centimeters. The primary endpoint was response by easel criteria, which was met. Response rate was 77% in the personalized arm versus 22% in the standard arm. Furthermore, overall survival was significantly longer at 26.6 months versus 10.7 months respectively. Of note, the 10.7 months in the standard arm is in line with the median overall survival in the Y90 arms of SERA and CEROBINIB. These trials bring up the question of the interpretation of negative phase 3 trials due to a lack of personalized dosimetry. Lastly, the proportion of patients that were able to undergo surgery after Y90 was significantly higher in those who received personalized dosimetry, which included eight patients with portal vein thrombosis. A retrospective trial from 11 centers in the United States and Europe have reported the results of patients with PVT excluding main portal vein. They were treated with local regional therapy and subsequently underwent transplant. The median time from diagnosis to successful downstaging was 6.1 months, and the median time after downstaging to actual transplant was 11.4 months. Predictors of HCC recurrence are listed here specifically with an AFP greater than 10. Recurrence was 11% in those who had an AFP prior transplant less than 10 compared to 50% in those who had an AFP greater than 10. There was no significant difference in recurrence based on the type of local regional therapy that was used to sound downstage, which is likely due to small numbers. However, it must be pointed out that 75% of patients that recurred were treated with TACE versus 12.5% in those treated with radioembolization. The largest report of Y90 prior to transplant has been recently published. A total of 207 patients at a single center were treated with radioembolization prior to transplantation, of which 82% were bridged and 18 were intended to be downstaged. Over 50% were child's pub B or C, 39% were BCLC C or D, and 19% were beyond the malign criteria. A total of 80% of patients had no prior therapy. The majority of patients were treated with one session of Y90 in a segmental fashion. The authors also performed an intended treat analysis of all T2 patients, which included 362 patients treated with radioembolization. 150 of these patients were not listed for various reasons. Among the 212 patients with T2 disease that were listed, a total of 19% dropped out, of which 5.2% was related to progression. This slide shows in detail stage at Y90 therapy in a transplant. And those who were T2 at the time of Y90, 98% remained T2 with only 2% progressing beyond T2. And the group intended for downstaging, 47% were downstaged to transplant, including 2 patients with portal vein thrombosis, whereas 53% remained greater than T2 at the time of transplant. The median overall survival post-transplant was 12.5 months with a 5-year overall survival rate of 77%, which is comparable to patients transplanted without HTC. A multivariate analysis showed that age greater than 65, lack of radiographic response, and the degree of necrosis on explant were significantly associated with overall survival. Of note, 17 patients who were child's PUD were treated in a segmental fashion with Y90 and achieved a 5-year overall survival rate approaching 92%. Ultimately, 24 patients recurred. The risk of recurrence was significantly associated with the degree of necrosis on explant with the highest rate of recurrence in those with partial necrosis. This is the intent to treat analysis of T2 patients treated with Y90 showing improvement in overall survival with transplant. And the authors concluded that Y90 is an effective treatment for HTC in the setting of bridging and downstaging to transplant. This is the first study that demonstrated proof of concept of combination therapy with local regional therapy and immunotherapy. Patients were treated with sublethal TACE or RFA followed by CTL-4 administration. This study showed that biopsies of untreated lesions had increased infiltration of T-cells post-immunotherapy compared to pre- immunotherapy consistent within a scopal effect. However, there are still many unanswered questions regarding this combination of immunotherapy and local regional therapy. To this end, there are several ongoing trials examining the safety and efficacy of combination therapy. A multidisciplinary approach has been shown to improve overall survival in the care of patients. It is therefore critical that all disciplines work as a team. My takeaway points are that HTC is responsive to radiation and in Y90, the dose of radiation matters. Innovations in Y90 and SBRT have led to promising results in patients with HTC. Combination of local regional therapy with systemic agents, particularly immunotherapy, is an exciting area of research. And a multidisciplinary approach is more critical as additional therapies, particularly combinations, become available to our patients. Thank you.

liver cancer

hepatocellular carcinoma

intrahepatic cholangiocarcinoma

FGF map kinase axis

resection versus transplant

local-regional therapies

Y90 radioembolization

SBRT

immunotherapy