Search

Fibrolamellar Hepatocellular Carcinoma (FLC) is a rare liver cancer affecting adolescents and young adults, with no gender or ethnicity predilection and without history of underlying viral hepatitis, cirrhosis, or other known risk factors. Almost all FLC patients present a somatic heterozygous deletion in chromosome 19p13.12, DNAJB1::PRKACA, which is sufficient to drive FLC in mice. A few studies comparing FLC tumors with adjacent non-transformed liver (normal) samples revealed many transcriptional differences. However, there were done in very small datasets and analyzed using different bioinformatic methods, resulting in just 18-47% agreement between them. This study aims to comprehensively characterize the transcriptome of FLC at bulk and spatial single-cell resolution. The whole transcriptome of 109 FLC frozen patient samples, the largest RNA-seq dataset of FLC to date, was sequenced using different library preparation and ribosomal depletion methods. Only paired tumor and normal tissue samples resected from the same patient were used and divided into two groups: exploration (3 datasets, 67 samples) and testing (2 datasets, 17 samples). Additionally, as external validation datasets, RNA-seq samples from previously published studies were collected, including Sorenson et al. (FLC: 26, normal: 9, paired: 8), the TCGA-LIHC study (FLC: 6, normal: 1, paired: 1), Hirsch et al. (FLC: 15, normal: 3, paired: 0) and Francisco et al. (FLC: 27, normal: 10, paired: 9). All were reanalyzed using state-of-the-art bioinformatic methods: mapped to the Human Genome GRCh38.103 and transcripts quantified using Salmon 1.6.0, unsupervised clustering exploration using PCA, tSNE and UMAP, differential expression calculated using DESeq2 1.28.1, and checking for detectability and consistency among datasets. We found 857 up- and 988 down-regulated genes presenting the same dysregulation in the exploration datasets and confirmed in the testing and external datasets. We call these genes the transcriptional FLC signature. The FLC signature was further characterized by comparing it with the genes differentially expressed in other liver cancers: hepatocellular carcinoma (41 paired samples), hepatoblastoma (22 paired samples), and cholangiocarcinoma (27 paired samples). We found 276 up- and 352 down-regulated genes altered in other liver cancers as well as FLC, but 156 up- and 68 down-regulated only in FLC. The 112 genes with the strongest dysregulation (56 up and 56 down) were used for a MERFISH screening, providing for the first time a single-cell spatial transcriptomic characterization of FLC. This showed clear differential expression patterns in tumor, normal, stromal, and infiltrating immune cells, allowing the identification of how different cell types contribute to the transcriptional FLC signature. Citation Format: David Requena, Aldhair Medico, Luis F. Soto, Mahsa Shirani, James A. Saltsman, Gadi Lalazar, Michael P. LaQuaglia, Sanford M. Simon. Bulk and spatial single-cell transcriptomic characterization of fibrolamellar hepatocellular carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 1516.

Fibrolamellar hepatocellular carcinoma (FLC) is a lethal pediatric cancer affecting adolescents and young adults. Besides surgical resection, there is no approved and effective therapy. FLC is driven by a fusion oncokinase, DNAJB1-PRKACA, that arises from a deletion fusing exon 1 of DNAJB1 and exons 2-10 of PRKACA, the catalytic subunit of protein kinase A. Expression of this chimeric oncokinase is capable of recapitulating the key histologic and transcriptomic features of FLC. Using PDXs, we tested therapeutics that have been used clinically for FLC or that target pathways identified via the transcriptomic signature of FLC, in addition to an agnostic library of clinical-stage drugs. Several compounds were identified with a varying degree of efficacy in different PDX lines as compared to our control line, primary human hepatocytes. One of the top hits was irinotecan, and its active metabolite SN38. Irinotecan is a topoisomerase I inhibitor and is clinically utilized in a wide variety of adult and pediatric cancers. A second top hit was Navitoclax, an inhibitor of the anti-apoptotic proteins Bcl-2 and Bcl-xL. Tool compounds that targeted Bcl-xL, but not bcl-2 were efficacious. This suggested a possible link between Bcl-xL, which is overexpressed in some PDX lines, and the resistance profile they showed in our screen. Unfortunately, Navitoclax has an on-target and dose-limiting toxicity of thrombocytopenia, which limits its clinical application. DT2216 is a VHL-based PROTAC that degrades Bcl-xL. DT2216 has two potential benefits over Navitoclax: 1) DT2216 is a more potent than Navitoclax against various Bcl-xL dependent tumor cells; and 2) it is less toxic to platelets than Navitoclax by targeting Bcl-xl to VHL that is poorly expressed in platelets to avoid induction of severe thrombocytopenia. A phase 1 clinical trial of DT2216 in relapsed/refractory malignancies is now ongoing (NCT04886622). We tested if the combination of DT2216 and irinotecan might have a therapeutic potential in FLC. Indeed, in vitro screening of the proposed combo, DT2216 and SN38, showed an additive or a synergistical effect in 4 validated FLC cell lines. We tested both the efficacy of DT2216 on Bcl-xL levels in FLC as well as platelet counts to identify any developing thrombocytopenia. DT2216 was less disruptive to platelets than Navitoclax at therapeutically relevant pre-clinical doses. Next, we tested 4 of our PDX models for FLC using various dosing regimens to identify the most efficacious regimen with the minimum side effects. Mice cohorts were evaluated for weight loss and tumor volume. Mice treated with DT2216 and SN38 showed consistent therapeutic benefit which was not evident in controls and less notable as monotherapy. In conclusion, TOPO1 and Bcl-xL prove to be promising targets of interest in FLC. Targeting both with DT2216 and irinotecan leads to tumor shrinkage in pre-clinical models and offer a potential therapeutic/pharmacological intervention for FLC. Citation Format: Bassem Shebl, Gadi Lalazar, Denise Ng, Tova Finkelstein, Guangrong Zheng, Peiyi Zhang, Carly Rosemore, Michael Ortiz, Daohong Zhou, Sanford Simon, Philip Coffino. Targeting Bcl-xL in fibrolamellar hepatocellular carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3888.

Fibrolamellar Hepatocellular Carcinoma (FLC) is a usually lethal liver cancer affecting adolescents and young adults without previous liver disease. FLC is characterized by a ~400kb nucleotide deletion resulting in a fusion of a heatshock protein cofactor, DNAJB1, with the catalytic subunit of protein kinase A (PRKACA). One limitation to the study of this disease is the lack of in vitro models. Organoids have shown utility as a model system for studying many diseases, including cancer. This study describes the development and characterization of FLC tumor organoids. With IRB approval, normal liver and FLC tumor tissue was obtained from patients undergoing surgical resection. Liver cells were isolated from the tissue samples and placed into a 3D basement membrane matrix and the media supplemented with growth factors specific to promote liver organoid expansion. Normal and tumor organoids were analyzed by microscopy for morphology, by PCR for presence of the DNAJB1-PRKACA chimeric transcript, by Western blot for chimeric protein expression, and by RNA sequencing for transcriptomic changes. Eight organoid lines have been developed from normal liver and eight from FLC tumor tissue, including multiple independent organoids developed from different metastases from the same patient. A distinct morphological difference between normal liver organoids and FLC tumor organoids can be seen with either brightfield microscopy or with H&E. The DNAJB1-PRKACA chimeric transcript was detected in all of the tumor tissue and tumor organoids by RT-PCR and not detected in any of the normal tissue or organoids. The presence of the DNAJB1-PRKACA fusion protein was verified by Western blot in tumor organoids and tissue but not detectable in the normal tissue. The RNA sequencing analysis shows clustering of tumor organoids with tumor tissue. Tumor organoids injected subcutaneously in mice grew as tumors. Now that the normal and tumor organoids have been developed and validated, they are being used for screening for potential therapeutics. Citation Format: Nicole J. Croteau, David Requena, Gadi Lalazar, Denise Ng, Solomon Levin, Bassem Shebl, Ruisi Wang, William J. Hammond, James A. Saltsman, Helmuth Gehart, Hans Clevers, Michael P. LaQuaglia, Sanford Simon. Human liver organoids for disease modeling of fibrolamellar hepatocellular carcinoma [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3905.

Introduction: Fibrolamellar hepatocellular carcinoma (FLC) is a liver cancer affecting adolescents and young adults without underlying liver disease. Patients with FLC are often diagnosed with advanced-stage disease and have a poor 5-year survival. Recently we have made advances in both understanding the etiology of FLC as well as developing therapeutics. We have found that the bulk of the genome lacks recurrent single-nucleotide or structural variants (inversions, amplifications) with the exception of a deletion in one copy of chromosome 19. This results in a fusion of the first exon of the heat shock protein, DNAJB1, with the 2nd through 10th exons of PRKACA, the catalytic subunit of protein kinase A. Expression of this fusion in the livers of mice, either by CRISPR or transposon, is sufficient to phenocopy the disease. The kinase activity of the fusion protein is essential for transformation, but equivalent expression of the native kinase is not sufficient. Results: Etiology: We have tested the ability of the native kinase and fusion oncokinase to phosphorylate potential substrates in human liver. We have found a discrete set of substrates that are phosphorylated in human liver for which we are now testing, which are critical for the pathogenesis of this cancer. Therapeutics: We have purified the fusion oncokinase without a tag and used a high-throughput screen to find inhibitors. We have found inhibitors in the nanomolar range for which we are using structural analysis to inform our medicinal chemistry. Additionally, we have developed several “never-in-plastic” patient-derived xenograft (PDX) models that retain expression of the driver, histology, transcriptome, and proteome of the original human tumor. We have also screened for compounds that induce tumor cell death and tumor regression in FLC PDXs. Conclusion: Subtle changes in the kinase substrate profile of the fusion kinase provide new leads for therapeutics for FLC. The ability to find inhibitors of the kinase in the nanomolar range, as well as the ability to screen for inhibitors of the PDX, offers the first hope for treatment against this lethal tumor. Citation Format: Gadi Lalazar, Melissa Jarmel, Ruisi Wang, Michael Tomasini, David Requena, Denise Ng, Sanford M. Simon. Advances in the etiology and therapeutics of a lethal childhood cancer, fibrolamellar hepatocellular carcinoma [abstract]. In: Proceedings of the AACR Special Conference on the Advances in Pediatric Cancer Research; 2019 Sep 17-20; Montreal, QC, Canada. Philadelphia (PA): AACR; Cancer Res 2020;80(14 Suppl):Abstract nr A51.

Genomic analysis of the pediatric cancer fibrolamellar hepatocellular carcinoma Fibrolamellar hepatocellular carcinoma (FLHCC) is a rare liver tumor that usually occurs in adolescents and young adults. Originally considered a variant of hepatocellular carcinoma (HCC), FLHCC is characterized by hepatocytes with deeply eosinophilic, granular cytoplasm interspersed with fibrous bands, without signs of cirrhosis as in HCC. Since little is known of its molecular pathogenesis, we performed RNA-seq and whole genome sequencing of FLHCC paired with normal liver from the same patient. Our results demonstrated that the majority of the DNA is unremarkable with few recurrent mutations or structural variants such as amplifications or inversions The major finding is a heterogeneous deletion of ∼400 kB in chromosome 19 which produces a chimeric transcript and a fully functional protein, a fusion between the heat shock protein DNAJB1 and the catalytic subunit of protein kinase A (PRKACA). Our initial observation of this chimera in 15 out of 15 patients has now been expanded to 116 out of 116 patients. Differential expression analysis of RNA-seq revealed many changes in the gene expression in FLHCC as compared to its paired normal liver. These changes were highly consistent from patient to patient, suggesting a unifying pathogenesis. The RNA-seq from our FLHCC samples had little in common with HCC samples from the TCGA or published reports of other cancers. An analysis of protein expression correlated well with the RNA-seq results. Some observed changes can directly be linked to increased activity of PRKACA. Other changes involve pathways known to participate in the initiation and progression of other cancers, which, along with an increase of PRKACA activity, represent targets for therapeutic intervention. Several of these targets are now being explored in cellular models, genetic, and PDX mouse models, as well as clinical trials. Citation Format: Elana P. Simon, Joshua N. Honeyman, David G. Darcy, Brad R. Rosenberg, Iris I. Lim, Jennifer M. Murphy, Ben Farber, Gadi Lalazar, Catherine Freije, Michael P. La Quaglia, Sanford M. Simon. Molecular analysis of the pediatric cancer fibrolamellar hepatocellular carcinoma. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4730. doi:10.1158/1538-7445.AM2015-4730