Establishment and characterization of patient-derived xenografts and cell lines from nasopharyngeal carcinoma

Abstract

Nasopharyngeal carcinoma (NPC) is a common cancer in Southern China, including Hong Kong. Epstein-Barr virus (EBV) infection is detected in almost all undifferentiated NPC which is the predominant histopathological type in this endemic region. The research into the roles of EBV infection in NPC pathogenesis is greatly hampered by the lack of representative, authenticated and well-characterized NPC xenograft and cell line models. The lack of these pre-clinical models also hinders the investigation of novel therapeutic approaches to druggable targets of NPC. Extensive efforts to establish NPC patient-derived xenografts (PDXs) and cell lines were carried out in this study. Fifty-eight clinical NPC samples were transplanted under kidney capsules in NOD/SCID mice, and four new NPC PDXs were successful established. Among these four PDXs, Xeno23 and Xeno47 were established from recurrent NPC, while Xeno32 and Xeno76 were from primary NPC, representing a take rate of 17.6% and 4.9% respectively for recurrent and primary NPC. NPC PDXs and cell lines are known to be difficult to be established for unclear reasons. One potential limiting factor leading to the difficulties in establishing NPC PDXs or cell lines may be the lytic reactivation of EBV, which may be induced by the sub-optimal growth environment in immunodeficient mice and in vitro culture conditions. Whole genome sequencing (WGS) was performed on these newly established PDXs. The results revealed genetic alterations including CYLD, TRAF3, TP53, CDKN2A, and NRAS mutations, which were consistent with those reported in patient tumors. The resemblance of these newly established PDXs to patient tumors with respect to genomic alterations supports their potential applications in preclinical drug trials of NPC. Some of the potential driver mutations for NPC progression, e.g. TP53 and NRAS, are present in the NPC PDXs. A new EBV-positive (EBV+ve) NPC cell line, NPC43, was successfully established from an NPC sample which failed to be propagated as PDX. The establishment of NPC43 required the inclusion of Rho kinase inhibitor, Y27632, in the medium. The Y27632 was shown to exhibit a significant inhibitory effect on lytic EBV reactivation in NPC43 cells. The EBV episomes are retained in NPC43 cell line which represents the second NPC cell line after C666-1 harboring EBV. Detailed characterization of NPC43 was performed including STR profiling, cytogenetic analysis, quantification of EBV copy number, expression profiling of EBV proteins and microRNAs and WGS, etc. Using this approach, two more EBV+ve cell lines (C17 and NPC47) were also established from established NPC PDXs by my colleagues. This may support the hypothesis that lytic reactivation of EBV is a key factor preventing the establishment of EBV+ve NPC cells in vitro and possibly PDX in vivo. Furthermore, NPC43 could be induced to produce infectious EBV, making NPC43 a unique and essential tool for the study of the regulation of latent and lytic infection of EBV in NPC. A druggable target, amplification of CCND1, was also revealed by WGS analysis of NPC43. NPC43 was sensitive to treatment with CDK4/6 inhibitor, supporting its application in preclinical evaluation of potential targets in NPC.