Natural-killer/T-cell lymphoma (NKTL) is a rare subset of non-Hodgkin lymphoma that demonstrates a unique geographic distribution, with higher prevalence in Asia compared to the West1. While cure remains achievable in early-stage disease, the prognosis for advanced-stage NKTL is dismal2. Recently, some progress has been made in uncovering the molecular pathogenesis of NKTL. In a genome-wide association study, strong correlations between HLA-DPB1 single-nucleotide polymorphisms and NKTL susceptibility were discovered, implicating altered antigen processing and presentation to CD4-positive T-lymphocytes in this Epstein-Barr virus (EBV)-associated malignancy3. Next generation sequencing also revealed recurrent somatic mutations such as TP53, JAK3, STAT3 and DDX3X in NKTL4,5. In this paper, we report a pair of male siblings from a non-consanguineous Chinese family who were diagnosed with NKTL, and provide initial evidence for novel recessive germline mutations in FAM160A1 identified through next-generation sequencing.
The index patient was 35 years old, when he presented with nasal blockage in March 2013. 18-FDG-PET/CT imaging revealed an 18-FDG-avid nasal mass infiltrating into the palate, as well as enlarged cervical lymph nodes. Biopsy of the mass showed abnormal lymphoid cells positive for CD56 by immunohistochemistry (IHC) as well as EBV-encoded RNA (EBER) by in-situ hybridization. He was diagnosed with stage IIA extranodal NKTL, nasal type, and treated with 4 cycles of bortezomib-GIFOX (gemcitabine, ifosfamide and oxaliplatin) as part of a clinical trial followed by radiotherapy to the nasal region. He had primary-refractory disease and was further treated with 4 cycles of SMILE (dexamethasone, methotrexate, ifosfamide, l-asparaginase, and etoposide) followed by high-dose chemotherapy and autologous stem cell transplantation. He progressed and received ruxolitinib off-label, followed by RAD001 (mTOR inhibitor) and LBH589B (histone deacetylase inhibitor) as part of another clinical trial. He then had radiotherapy to an ulcerating penile lesion before he died of progressive disease 27 months after diagnosis (Supplementary Table 1).
His younger brother was 18 years old, when diagnosed with NKTL affecting the nasal floor in 1998 following a bout of epistaxis. He received 6 cycles of CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) with high-dose methotrexate and went into complete remission. Three years later he was diagnosed with chronic myeloid leukemia, for which he received hydroxyurea. In 2004, he had a relapse of NKTL and received ESHAP (etoposide, prednisolone, cytarabine and cisplatin), total body irradiation and allogeneic bone marrow transplant. He died of transplant-related complications 6 years from diagnosis. These siblings have an older brother unaffected by any hematological malignancy. Their father died of a non-malignant condition in his fifties and their mother remains well at the age of 68 years. Among their first-degree relatives, none were known to be inflicted with hematologic malignancies (Fig. 1).
Despite recent improvements in the understanding of familial lymphoma, the specific predisposition genetic factors remain largely unknown6. To the best of our knowledge, familial cases of NKTL are extremely rare occurrences. Previously, a familial occurrence of nasal NKTL in a father and son pair with known heavy pesticide exposure had been reported in 2001, but no specific genetic element was evident7. In another study, biallelic truncating MSH2 mutations had been described in familial T-cell non-Hodgkin lymphoma affecting three siblings8. In order to identify potential pathogenic germline mutations that contribute to NKTL in the affected siblings, we performed whole exome sequencing on DNA samples obtained from both of them as well as their unaffected family members (older brother and mother). As both parents of the affected siblings were unaffected by NKTL, only autosomal recessive and x-linked recessive models were explored. Details regarding sample curation, next generation sequencing, bioinformatics as well as pathological analyzes are described in the Supplementary section. Using the x-linked recessive model, a single nucleotide variant located in the 3’-untranslated region (UTR) of HNRNPH2 (X-100668489-T-G, rs41307260) was initially detected. This particular variant however, had low coverage (14X) and subsequently was not observed on Sanger sequencing. Out of 39 candidate variants identified in the autosomal recessive model, one was in the 5’-UTR (RDH13), three in the 3’-UTR (GP6, NLRP2, LILRB3), and one in the coding exon (FAM160A1 c.2827C > T). These five variants were verified by Sanger sequencing (Supplementary Figure 1). Given the extreme rarity of familial NKTL, it is unlikely that any pathogenic variant would present as a common polymorphism in the population (minor allele frequencies: RDH13, 0.2101; GP6, 0.7535; NLRP2, 0.3729; LILRB3, 0.05749), therefore leaving only FAM160A1 as the remaining candidate (minor allele frequency: 0.00001892). Both affected brothers were FAM160A1 homozygous mutants, their mother was a heterozygous carrier, and their older unaffected brother was homozygous wild-type. In keeping with this result, direct sequencing of peripheral-blood DNA of their paternal aunt revealed heterozygous FAM160A1 c.2827C/T (Fig. 1).
The non-synonymous substitution at FAM160A1 c.2827C > T results in an amino acid alteration from arginine to cysteine (p.R943C), and is predicted to be deleterious based on multiple in-silico algorithms including PROVEAN (prediction score, −4.74), SIFT (prediction score, 0.000), PolyPhen-2 (prediction score, 1.000), MutationAssessor (prediction score, 2.99), MutationTaster, and M-CAP (See Supplementary Tables 2–6 and References).
Somatic variants detected in the proband included 66 non-silent mutations, including those previously described in NKTL such as DDX3X, STAT3, and PRDM14. Specifically, stopgain mutations in DDX3X c.G1639T (p.E547X), missense mutations in STAT3 c.G1981C (p.D661H), and PRDM1 c.C1877T (p.T626M) were observed. Additionally, within limitations of the analysis, an attempt to identify somatic variants in the affected younger brother revealed 192 potential non-silent mutations (Supplementary Table 7).
Gene expression profiling revealed that FAM160A1 was significantly overexpressed in the index patient’s tumor tissue compared to sporadic NKTL (n = 12) (Fig. 2). In keeping with this finding, strong cytoplasmic FAM160A1 staining by IHC was observed in scattered tumor-infiltrating cells, which corresponded to CD68-positive histiocytes. Lymphomatous cells, which were EBER-positive, did not stain positive for FAM160A1. This staining pattern was consistently observed in independent samples of metachronous skin metastases over the clavicle and foreskin (Supplementary Figure 2). Among the sporadic NKTLs with available tissue for IHC (n = 14), two displayed weak-cytoplasmic staining for FAM160A1, while the rest were negative. Examination of mutation frequencies in sporadic NKTL cases, derived from Singapore (n = 33) and Chinese cohorts (n = 189) in a previously published dataset3, did not reveal any FAM160A1 c.2827C > T variants. Furthermore, a preliminary analysis of our in-house clinical dataset of sporadic NKTL suggests that higher expression of FAM160A1 did not lead to differences in survival outcomes (data not shown). Taken together, this indicates that alterations in FAM160A1 may be unique to familial rather than sporadic NKTL.
FAM160A1 is a member of the UPF0518 family of proteins, each containing a conserved retinoic acid induced 16 (RAI16)-like domain with unknown biological function. Other members of this family include FAM160A2, FAM160B1, and FAM160B2. In particular, FAM160A2 (also known as FHIP), is a component of the FTS/Hook/FHIP complex (FHF complex), which functions to promote vesicle trafficking and/or fusion via the homotypic-vesicular protein sorting complex (the HOPS complex)9. In terms of its mutation frequency in cancer tissues, querying 35976 cancer samples across 150 studies revealed only 52 somatic point mutations (44 missense, 7 truncating, and 1 non-start) scattered across the gene10. Analysis of the human interactome11 and gene ontology12 revealed significant interactions of FAM160A1 with AKT-interacting protein (AKTIP, also known as FTS) and Protein Hook homolog 3 (HOOK3) (adjusted p-value 9.39e−7)—both components of the FTS/Hook/FHIP (FHF) complex9, suggesting that FAM160A1 may be part of this complex as well. A relative increase in expression of genes encoding for members of the FHF complex, including AKTIP (1.19 fold), HOOK3 (1.19 fold), HOOK2 (1.14 fold), HOOK1 (2.95 fold), but not FAM160A2 (also known as FHIP) (0.94 fold), was observed.
Our current findings not only reinforce the existence of familial T and NK-cell lymphomas as per previous reports7,8, but, in addition, provide evidence for a novel genetic basis that may explain this rare phenomenon. The importance of delineating a detailed family history, in addition to discriminating usage of next-generation sequencing cannot be further emphasized, in order for our current findings to be confirmed in future cohorts of patients.
In summary, this report implicates a novel recessive germline mutation in FAM160A1 with familial NKTL.
Kwong, Y. L. et al. Asian Oncology Summit. Management of T-cell and natural-killer-cell neoplasms in Asia: consensus statement from the Asian Oncology Summit 2009. Lancet Oncol. 10, 1093–1101 (2009).
Yamaguchi, M. & Miyazaki, K. Current treatment approaches for NK/T-cell lymphoma. J. Clin. Exp. Hematop. 57, 98–108 (2017).
Li, Z. et al. Genetic risk of extranodal natural killer T-cell lymphoma: a genome-wide association study. Lancet Oncol. 17, 1240–1247 (2016).
Jiang, L. et al. Exome sequencing identifies somatic mutations of DDX3X in natural killer/T-cell lymphoma. Nat. Genet. 47, 1061–1066 (2015).
Koo, G. C. et al. Janus kinase 3-activating mutations identified in natural killer/T-cell lymphoma. Cancer Discov. 2, 591–597 (2012).
Cerhan, J. R. & Slager, S. L. Familial predisposition and genetic risk factors for lymphoma. Blood 126, 2265–2273 (2015).
Kojya, S. et al. Familial nasal NK/T-cell lymphoma and pesticide use. Am. J. Hematol. 66, 145–147 (2001).
Scott, R. H. et al. Familial T-cell non-Hodgkin lymphoma caused by biallelic MSH2 mutations. J. Med. Genet. 44, e83 (2007).
Xu, L. et al. An FTS/Hook/p107(FHIP) complex interacts with and promotes endosomal clustering by the homotypic vacuolar protein sorting complex. Mol. Biol. Cell 19, 5059–5071 (2008).
Cerami, E. et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2, 401–404 (2012).
Huttlin, E. L. et al. The BioPlex network: a systematic exploration of the human interactome. Cell 162, 425–440 (2015).
Ashburner, M. et al. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat. Genet. 25, 25–29 (2000).
This work was supported by the National Medical Research Council of Singapore (TCR12DEC005), Tanoto Foundation Professorship in Medical Oncology, New Century Foundation Limited, Ling Foundation, Singapore National Cancer Centre Research Fund, Biomedical Research Council of A*STAR as well as the SingHealth Duke-NUS Academic Medical Centre and Oncology ACP.
Conflict of interest
The authors declare that they have no conflict of interest.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Chan, J.Y., Ng, A.Y.J., Cheng, C.L. et al. Whole exome sequencing identifies recessive germline mutations in FAM160A1 in familial NK/T cell lymphoma. Blood Cancer Journal 8, 111 (2018). https://doi.org/10.1038/s41408-018-0149-5
This article is cited by
Journal of Hematology & Oncology (2019)
Nature Communications (2019)