Thymus neuroendocrine tumors with CTNNB1 gene mutations, disarrayed ß-catenin expression, and dual intra-tumor Ki-67 labeling index compartmentalization challenge the concept of secondary high-grade neuroendocrine tumor: a paradigm shift

Abstract

We herein report an uncommon association of intimately admixed atypical carcinoid (AC) and large cell neuroendocrine (NE) carcinoma (LCNEC) of the thymus, occurring in two 20- and 39-year-old Caucasian males. Both tumors were treated by maximal thymectomy. The younger patient presented with a synchronous lesion and died of disease after 9 months, while the other patient was associated with a recurrent ectopic adrenocorticotropic hormone Cushing's syndrome and is alive with disease at the 2-year follow-up. MEN1 syndrome was excluded in either case. Immunohistochemically, disarrayed cytoplasmic and nuclear ß-catenin expression was seen alongside an intra-tumor Ki-67 antigen labeling index (LI) ranging from 2 to 80% in the younger patient's tumor and from 3 to 45% in the other. Both exhibited upregulated cyclin D1 and retinoblastoma, while vimentin was overexpressed in the recurrent LCNEC only. Next-generation sequencing revealed CTNNB1, TP53, and JAK3 mutations in the synchronous tumor and CTNNB1 mutation alone in the metachronous tumor (the latter with the same mutation as the first tumor of 17 years prior). None of the 23 T-NET controls exhibited this hallmarking triple alteration (p = 0.003). These findings suggested that LCNEC components developed from pre-existing CTNNB1-mutated AC upon loss-of-function TP53 and gain-of-function JAK3 mutations in one case and an epithelial-mesenchymal transition upon vimentin overexpression in the other case. Both tumors maintained intact cyclin D1-retinoblastoma machinery. Our report challenges the concept of secondary LCNEC as an entity that develops from pre-existing AC as a result of tumor progression, suggesting a paradigm shift to the current pathogenesis of NET.

Molecular Classification of Neuroendocrine Tumors of the Thymus

“…Cases with metachronous metastases showed evidence of morphologic (e.g., switch from AC to NET G3 and increase of mitotic rate and/or mitotic index) and genomic progression (switch from one CNI cluster to another), confirming the previous notion that some NET can either coexist or progress from low-to higher-grade tumors (nominally classified as LCNEC according to current terminology) through acquisition of genomic alterations. 17 Further studies are required to establish whether progression from low-/ intermediate-to high-grade tumors, which has been postulated in the lung, also occurs in the thymus. 32 Another major finding of this study was the identification of a new subgroup of TNET of intermediate grade within the group of LCNEC that we provisionally termed NET G3 in analogy to the terminology in GEP-NET.…”

“…Other genes with reported mutations in NET, EZH2, DNA methyltransferase 3 alpha (DNMT3A), RB1, KIT proto-oncogene, receptor tyrosine kinase (KIT), lysine methyltransferase 2A (KMT2A),Janus kinase 2 (Jak2), and Jak3 showed wild-type sequence. 17,30,31 Proposal for an Integrated Morphomolecular TNET Classification…”

“…A recent small case series even provides genetic evidence that some thymic LCNEC may develop from AC. 17 Based on retrospective clinical data, we previously reported that a cutoff at 15 mitoses separated TNET with an increased risk of recurrence. 16 At the other end of the spectrum and regarding the distinction from SCC, we and others found a better prognosis for LCNEC than SCC, which may again be due to the fact that the current WHO definition of LCNEC embraces a spectrum of heterogenous tumors with very different prognoses.…”

Clinicopathological Profiling of Lung Carcinoids with a Ki67 Index > 20%

“…Table 1). Future studies including genetic profiling are needed to either support or reject the WD-NEN to PD-NEN transition hypothesis recently proposed [5, 25]. …”

Pulmonary neuroendocrine carcinomas—evidence for a spectrum of differentiation

“…This finding elegantly underscores a number of important prior observations that would support this notion. For instance, tumors that by current diagnostic criteria are composed of more than one pattern or that show transitional forms within the same tumor mass have long been recognized in the literature [7,8]. In addition, neoplasms that show low-grade morphologic features but have mitotic counts exceeding the specified cutoff have not only been described in other organ systems but are increasingly also observed in the lung [9].…”