Patient-derived tumor organoids for personalized medicine in a patient with rare hepatocellular carcinoma with neuroendocrine differentiation: a case report.

Background: Hepatocellular carcinoma with neuroendocrine differentiation (HCC-NED) is a very rare subtype of primary liver cancer. Treatment allocation in these patients therefore remains a challenge.
Methods: We report the case of a 74-year-old man with a HCC-NED. The tumor was surgically removed in curative intent. Histopathological work-up revealed poorly differentiated hepatocellular carcinoma (Edmondson-Steiner grade IV) with diffuse expression of neuroendocrine markers synaptophysin and chromogranin. Three months after resection, multifocal recurrence of the HCC-NED was observed. In the meantime, tumor organoids have been generated from the resected HCC-NED and extensively characterized. Sensitivity to a number of drugs approved for the treatment of HCC or neuroendocrine carcinomas was tested in vitro.
Results: Based on the results of the in vitro drug screening, etoposide and carboplatin are used as first line palliative combination treatment. With genomic analysis revealing a NTRK1-mutation of unknown significance (kinase domain) and tumor organoids found to be sensitive to entrectinib, a pan-TRK inhibitor, the patient was treated with entrectinib as second line therapy. After only two weeks, treatment is discontinued due to deterioration of the patient's general condition.
Conclusion: The rapid establishment of patient-derived tumor organoids allows in vitro drug testing and thereby personalized treatment choices, however clinical translation remains a challenge. To the best of our knowledge, this report provides a first proof-of-principle for using organoids for personalized medicine in this rare subtype of primary liver cancer.

Introduction

Primary liver carcinomas with concurrent hepatocellular and neuroendocrine tumor components in the same liver lesion are very rare[1]. They consist of two morphologically distinct cell populations that express hepatocellular or neuroendocrine markers and are classified as Hepatocellular Carcinoma-Neuroendocrine Carcinoma (HCC-NEC)[2] or liver mixed neuroendocrine non-neuroendocrine neoplasms (MiNEN)[3]. The published case reports describe an aggressive tumor phenotype and poor overall prognosis[3,4]. Even rarer are HCCs with neuroendocrine differentiation (HCC-NED)[5]. HCC-NEDs are comprised of morphologically uniform cells that stain positively for both hepatocellular and neuroendocrine markers. Patients are usually treated by means of surgical resection, transarterial chemoembolization or systemic (chemo)therapy for liver cancer or neuroendocrine malignancies. Because HCC with neuroendocrine differentiation is a very rare tumor entity, therapy in these patients remains ill-defined[4,6,7]. Here, we report a case history of a 74-year-old man with HCC-NED. We provide a comprehensive histopathological characterization and a genomic analysis of this rare tumor. Furthermore, we describe the generation of tumor organoids that retain the key characteristics of the originating tumor. The organoids were used in drug screens to identify the most promising treatment options.

Methods

Patient information and biological material

Human biopsy and resection tissue was collected from patients undergoing diagnostic liver biopsy or liver surgery at the University Hospital of Basel. Written informed consent was obtained from all patients. The study was approved by the local ethics committee (protocol numbers EKNZ 2014-099 as well as BASEC 2019-02118). For the HCC-NED patient described in this study, written informed consent to publish the case details was obtained from the family.

Liver cancer organoid culture

Tumor organoid lines were generated from liver biopsy or resection tissue according to published protocols[8,9]. Briefly, tumor tissues were dissociated to small-cell clusters and seeded in domes of basement membrane extract type 2 (BME2, R&Dsystems, Cat. No. 3533-005-02). Polymerized BME2 domes were overlaid with expansion medium (EM): advanced DMEM/F-12 (Gibco, Cat. No. 12634010) supplemented with 1× B-27 (Gibco, Cat. No. 17504001), 1× N-2 (Gibco, Cat. No. 17502001), 10 mM Nicotinamide (Sigma, Cat. No. N0636), 1.25 mM N-Acetyl-L-cysteine (Sigma, Cat. No. A9165), 10 nM [Leu15]-Gastrin (Sigma, Cat. No. G9145), 10 μM Forskolin (Tocris, Cat. No. 1099), 5 μM A83-01 (Tocris, Cat. No. 2939), 50 ng/ml EGF (Peprotech, Cat. No. AF-100-15), 100 ng/ml FGF10 (Peprotech, Cat. No. 100-26), 25 ng/ml HGF (Peprotech, Cat. No. 100-39), 10% RSpo1-conditioned medium (v/v, homemade). Cultures were kept at 37 °C in a humidified 5% CO2 incubator. Organoids were passaged weekly at 1:4–1:6 split ratios using 0.25% Trypsin-EDTA (Gibco, Cat. No. 25200056). Frozen stocks were prepared at regular intervals. All organoid cultures were regularly tested for Mycoplasma contamination using the MycoAlert™ Mycoplasma detection kit (Lonza, Cat. No. LT07-118).

Histology and immunohistochemistry

Tumor and liver tissues were fixed in 4% phosphate-buffered formalin and embedded in paraffin using standard procedures. Tumor organoids were released from BME2 by incubation in Dispase II (Sigma-Aldrich, Cat. No. D4693). Organoids were fixed in 4% phosphate-buffered formalin in PBS for 30 min at room temperature following encapsulation in HistoGel (Thermo Fisher Scientific, Cat. No. HG-4000-012) and subsequent dehydration and paraffin embedding.

Histopathological evaluation was assessed by three board-certified pathologists (MSM, JV and LMT). Tumors were classified based on architecture and cytological features, and graded according to the Edmondson grading system[10,11]. The following primary antibodies were used for automated diagnostic immunohistochemical staining on a Benchmark XT device (Ventana Medical Systems) at the Institute of Pathology of the University of Basel: AFP (Ventana, Ref-Nr. 760-2603), ARG1 (Ventana, Ref-Nr. 760-4801), CD10 (Ventana, Ref-Nr. 790-4506), CD56 (Ventana, Ref-Nr. 790-4465), CHGA (Ventana, Ref-Nr. 760-2519), GPC3 (Ventana, Ref-Nr. 790-4564), HLA-ABC (Abcam, Cat. No. ab70328), Hep Par-1 (Ventana, Ref-Nr. 760-4350), KRT19 (Ventana, Ref-Nr. 760-4281), Ki-67 (Dako, Cat. No. IR626), Pan-TRK (Abcam, Cat. No. ab181560), PD-L1 (Ventana, Ref-Nr. 740-4907), SYP (Ventana, Ref-Nr. 790-4407), and SSTR2 (Abcam, Cat. No. ab134152).

Xenograft mouse model

Experiments involving animals were performed in strict accordance with Swiss law and were previously approved by the Animal Care Committee of the Canton Basel-Stadt, Switzerland. Tumor organoids, corresponding to 2 × 106 cells, were released from BME2, resuspended in 100 μl 50:50 (v/v) BME2:expansion medium, and injected subcutaneously into the flank of one male NSG (Non-obese diabetic, Severe combined immunodeficiency, Gamma) mouse (The Jackson Laboratory) at 8 weeks of age. The mouse was housed in an individually ventilated cage (Tecniplast Green Line) at 22 °C, 55% humidity and a light cycle of 12:12 h. Tumor growth was assessed weekly by caliper measurement. The tumor was harvested when it reached 1000 mm3 in size, fixed in 4% phosphate-buffered formalin and processed for paraffin embedding and immunohistochemistry as described above.

Drug screenings

All compounds were dissolved in DMSO at 10 mM (except for cisplatin and carboplatin) and aliquots were stored at −20 °C, 4 °C or room temperature according to the manufacturer’s recommendations. Sorafenib tosylate, lenvatinib mesylate, cabozantinib mesylate, regorafenib, octreotide acetate, lanreotide acetate, etoposide, sunitinib malate, everolimus, entrectinib, larotrectinib: all from Selleckchem; pasireotide ditrifuloroacetate (MedChem Express); cisplatin (Sandoz); carboplatin (Labatec). For drug screenings, organoids were dissociated with 0.25% Trypsin-EDTA (Gibco) and seeded at 1000 cells/well in 384-well plates in organoid expansion medium supplemented with 10% BME2. Two days later, compounds were added in a 2-fold dilution series ranging from 0.02 nM to 10 μM. After 6 days of treatment, cell viability was measured using CellTiter-Glo 3D (Promega). Luminescence was measured on a Synergy H1 Multi-Mode Reader (BioTek Instruments). Results were normalized to vehicle control (100% DMSO or 100% water). All experiments were performed twice. Dose-response curves were calculated using Prism 9.3.1 (GraphPad), nonlinear regression algorithm was used with a constrain of 0 for the bottom and 100 for the top.

DNA extraction and whole-exome sequencing

DNA from the tumor, adjacent non-tumoral liver tissue and organoid was extracted using the Qiagen DNeasy Blood & Tissue kit (Qiagen, Cat. No. 69504) following the manufacturer’s instructions. Extracted DNA was subjected to whole-exome sequencing. The Twist Human Core Exome kit was used for whole exome capture according to the manufacturer’s guidelines. Sequencing was performed on Illumina NovaSeq 6000 using paired-end 100-bp (mean sequencing depth 135× for HCC, 153× for the organoid and 129× for germline (adjacent non-tumoral liver tissue)). Sequencing was performed by CeGaT (Tübingen, Germany).

Reads obtained were aligned to the reference human genome GRCh38 using Burrows-Wheeler Aligner (BWA, v0.7.12)[12]. Local realignment, duplicate removal, and base quality adjustment were performed using the Genome Analysis Toolkit (GATK, v4.1 and Picard (http://broadinstitute.github.io/picard/)). Somatic single nucleotide variants (SNVs) and small insertions and deletions (indels) were detected using MuTect2 (GATK 4.1.4.1)[13] and Strelka (v.2.9.10)[14]. Only variants detected by both callers were kept. We filtered out SNVs and indels outside of the target regions (i.e., exons), those with a variant allelic fraction (VAF) of <5 % and/or those supported by <3 reads. We excluded variants for which the tumor VAF was <5 times that of the paired non-tumor VAF. We further excluded variants identified in at least two of a panel of 123 non-tumor samples, captured and sequenced using the same protocols using the artifact detection mode of MuTect2 implemented in GATK. All indels were manually inspected using the Integrative Genomics Viewer[15]. FACETS (v.0.5.14)[16] was used to identify allele-specific copy number alterations (CNAs). Genes with total copy number greater than gene-level median ploidy were considered gains; greater than ploidy + 4, amplifications; less than ploidy, losses; and total copy number of 0, homozygous deletions. Somatic mutations associated with the loss of the wild-type allele (i.e., loss of heterozygosity [LOH]) were identified as those where the lesser (minor) copy number state at the locus was 0. For chromosome X, the log ratio relative to ploidy was used to call deletions, loss, gains and amplifications. All mutations on chromosome X in male patients were considered to be associated with LOH. Comparison of copy number between organoids and tumor were performed at gene level.