Non-Hodgkin lymphomas (NHL) are a diverse group of blood cancers on the rising trend both worldwide and in Singapore. Better understanding of the pathogenesis and biology of these tumors is urgently required to improve the diagnosis and prognostication. Moreover, the outcome using currently available therapies is poor and there is an immediate need for better treatment options. With an increasing number of genomic studies published, there is a growing demand to bring these discoveries into in vitro and in vivo models. However, the lack of publicly available cell lines and/or animal models for the majority of the NHL subtypes makes these studies difficult to proceed.With that objective, we have started collecting fresh NHL tumor samples and implanting them into NOD scid gamma (NSG) mice to generate patient-derived xenograft (PDX) models. After tumor dissemination, the cells are injected either subcutaneously or intraperitoneally into 6-10 week old mice. Each passage is thoroughly immunohistochemically characterized by a senior hematopathologist and contrasted to the primary patient specimen. Whole exome sequencing (WES) is performed and compared to the primary patient data where possible. A model is considered stable if it has not changed its phenotype for at least 3 passages. A full list of currently available stable models is listed in Table 1. Efforts are also under way to generate stable cell lines based on these models. Strategies tested are spontaneous immortalization, human telomerase reverse transcriptase (TERT), Herpesvirus saimiri (HVS) or Epstein-Barr virus (EBV) transduction and feeder culture.We have recently reported a comprehensive molecular profiling of monomorphic epitheliotropic intestinal T-cell lymphoma (MEITL; previously also known as type II enteropathy-associated T-cell lymphoma; Nairismagi et al., Leukemia, 2016). MEITL, a newly recognized WHO entity, is a rare aggressive primary intestinal disease with poor prognosis and a median overall survival of only 7 months. Frequent alterations were identified in the JAK-STAT and G-protein-coupled receptor signaling pathways and a number of epigenetic regulators. Specifically, 60-30% mutation frequencies were found in the STAT5B, JAK3, SETD2, GNAI2 and CREBBP genes.Tumor tissue was collected from a 55 year old Chinese male with relapsed MEITL. The cells have been propagated in the NSG mouse by subcutaneous or intraperitoneal injection up to 6 and 3 passages, respectively. All passages have been histologically characterized and maintained their CD3+ CD8alpha-alpha+ cytotoxic granules+ MATK+ and EBER- phenotype with aberrant expression of both TCR beta and gamma. WES was performed both in the patient primary sample and in passage 1 (P1) PDX sample. There were 83 somatic mutations present in the primary sample, all of which were also identified in the P1 PDX sample. Furthermore, Sanger sequencing was used to validate the presence of all 83 mutations in the remaining passages. There was only 1 mutation lost in the P6 subcutaneous PDX model compared to the primary patient specimen demonstrating the outstanding stability of this model. Using peritoneal fluid cells from P2 we have previously performed ex vivo studies and identified novel treatment modalities for this deadly disease (Nairismagi et al., Leukemia, 2016). Efforts are currently under way to test these regimens also in in vivo setting.In summary, we have generated a number of NHL PDX models with the objective to enhance drug testing. MEITL presents a real life situation where clinical trials are not possible due to the rarity of the disease; however, the availability of PDX models allows for efficient testing of novel therapies and can lead to improved patient outcome. Efforts are under way to further characterize the existing models and tissue collection is ongoing to establish more models. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.
Read full abstract