Multinodular Goiter Progression Toward Malignancy in a Case of DICER1 Syndrome: Histologic and Molecular Alterations

Abstract

Objectives

Multinodular goiter (MNG) and well-differentiated thyroid carcinoma (WDTC) are emerging phenotypes of DICER1 syndrome.

Methods

Histologic and molecular findings of botryoid-type embryonal rhabdomyosarcoma (bERMS) and thyroid nodules from a 12-year-old DICER1 mutation carrier (p.Arg1060Ilefs*7) were investigated, providing interesting clues for understanding thyroid carcinogenesis.

Results

The patient had bERMS at age 7 years. The thyroid was enlarged and multinodular (61 g). Histologically, some nodules were classified as adenomatous and others as tumors with “intermediate” nuclei. One displayed vascular invasion and was classified as WDTC not otherwise specified (NOS). Somatic DICER1 mutations were identified in bERMS, two tumors with “intermediate” nuclei and WDTC. No somatic DICER1 mutations were found in adenomatous nodules. No molecular alterations were detected in BRAF600, NRAS61, HRAS12/61, KRAS12/61, TERT promoter, RET/PTC1, RET/PTC3, and PAX8/PPARγ.

Conclusions

The findings obtained from this single case support the assumption that DICER1 syndrome–related WDTC NOS may develop on a background of MNG, via a stepwise process, involving DICER1 somatic mutations and additional molecular events, distinct from the classic pathways of papillary/follicular carcinoma.

DICER1 syndrome is an autosomal-dominant tumor susceptibility disorder, caused by DICER1 germline pathogenic mutations. The ribonuclease encoded by DICER1 is required to process precursors of microRNAs (miRNAs) to mature miRNA. As a central regulator of miRNA function, DICER1 can tune gene expression posttranscriptionally and influence cancer-relevant processes.¹

Patients affected by DICER1 syndrome present an increased risk for a variety of benign and malignant tumors. Pleuropulmonary blastoma, ovarian sex cord stromal tumors, and cystic nephroma are common phenotypes.^1,2 Less frequently, botryoid-type embryonal rhabdomyosarcoma (bERMS) of the cervix or of other organs has been reported.²

A large spectrum of thyroid phenotypes may occur. Familial multinodular goiter (MNG) arising in children is the most frequent manifestation of germline DICER1 mutation in the thyroid. In DICER1 mutation carriers, the cumulative incidence of either MNG diagnosis or thyroidectomy has been estimated to be 75% in women and 17% in men by age 40 years.³ Moreover, well-differentiated thyroid carcinoma (WDTC), despite being considered a “rare” occurrence in DICER1 mutation carriers,¹ is an emerging phenotype: individuals with DICER1 syndrome have a 16- to 24-fold increased risk for developing WDTC, encompassing papillary thyroid carcinoma (PTC) or follicular thyroid carcinoma (FTC).³

Case Description

A 12-year-old girl was found to have thyroid nodules in a follow-up consultation at our hospital. She had had bERMS Image 1 of the cervix at age 7 years and was treated by surgery and adjuvant chemotherapy (nine cycles of the IVA regimen: ifosfamide, vincristine, and actinomycin). The patient had a family history of MNG (two aunts, one cousin). She was clinically and biochemically euthyroid. Thyroid ultrasonography showed multiple nodules in both lobes. Fine-needle aspiration biopsy (FNAB) of two nodules led to diagnoses of follicular neoplasm (Bethesda category IV) and colloid nodule (Bethesda category II). A total thyroidectomy was performed.

Image 1

Botryoid-type embryonal rhabdomyosarcoma: polypoid neoplasia, variable in cellularity (A, H&E, ×40), composed of round to spindle undifferentiated cells, crowded underneath the squamous epithelium (cambium layer) (B, H&E, ×100). By immunohistochemistry, the neoplastic cells showed immunoreactivity for vimentin (C, ×100) and, focally, for desmin (D, ×100). Differentiating rhabdomyoblasts (E, H&E, ×630), positive for myogenin (F, ×630), were also observed. A c.5439G>A (p.Glu1813Asp) RNase IIIb mutation was identified (*) (G).

Pathologic Findings

The thyroid was enlarged (61 g) and multinodular. On cut surface, multiple and heterogeneous nodules were identified. Some nodules were solid and had tan areas Image 2. Five nodules were studied individually for molecular alterations (Image 2 and Image 3).

Image 2

Total thyroidectomy specimen (A). B, C, Right lobe. D, E, Left lobe. Most of the lesions displayed follicular cells with nuclei of the “intermediate” type (red), while others looked like adenomatous nodules (blue). A well-differentiated thyroid carcinoma (WDTC) with angioinvasion was identified (yellow). Cardinal numbers indicate nodules and tumors submitted to the molecular study: (I) adenomatous nodule, (II) adenomatous nodule with papillary hyperplasia, (III and IV) tumors with “intermediate” nuclei, and (V) WDTC.

Image 3

Study of DICER1 somatic mutations in four thyroid nodules: nodules circled in green and blue correspond to adenomatous nodules without papillary hyperplasia (green, I in Image 2 and Image 4B) and with papillary hyperplasia (blue, II in Image 2 and Image 4D). No DICER1 somatic mutations were detected in any of them. WT, wild type. The nodule circled in red corresponds to a tumor with nuclei of the “intermediate” type (III in Image 2 and Image 4F). A c.5438A>G (p.Glu1813Gly) RNase IIIb mutation was identified (*). The nodule circled in black corresponds to well-differentiated thyroid carcinoma (V in Image 2 and Image 4H). A c.5438A>G (p.Glu1813Gly) RNase IIIb mutation was identified (*).

Histologically, the nodules displayed a heterogeneous morphology. Some nodules looked like adenomatous nodules, displaying follicular cell nuclei (Image 3 and Image 4 ) and frequently exhibiting papillary hyperplasia (Image 3 and Image 4). Most of the nodules had a microfollicular, papillary, trabecular, or solid growth pattern and displayed follicular cells with nuclei of the “intermediate” type, larger than common follicular cell nuclei, with chromatin clearing, slightly irregular contour, and moderate crowding (Image 4). The distinction between adenomatous nodules with follicular cell nuclei and tumors with “intermediate” nuclei was not easy because intranodular heterogeneity determines subtle morphologic differences rather than a clear-cut separation. In an oversimplified way, the two types of lesions were separated as shown in Image 2. Brisk mitotic activity (up to 18 mitoses per 10 high-power fields) was found in tumors with “intermediate” nuclei, and in one (1.4 cm), a focus of vascular invasion was identified (Image 4). The aforementioned nuclear features of the invasive tumor did not fit with either FTC or PTC, and therefore, a diagnosis of well-circumscribed, partly encapsulated, angio-invasive WDTC not otherwise specified (NOS) was made.

Image 4

A, B, Adenomatous nodule without papillary hyperplasia (green area in Image 3). C, D, Adenomatous nodule with papillary hyperplasia. Although some nuclei show slight irregular contour, chromatin clearing, and some crowding, most show common follicular cell nuclei, and the nodule was classified as adenomatous (blue area in Image 3). E, F, Tumor with follicular, papillary, and trabecular pattern, with nuclei of the “intermediate” type. There were numerous mitotic figures (arrows) (red area in Image 3). G, H, WDTC with a focus of vascular invasion (black area in Image 3). (A, C, E, G, H&E, ×100; B, D, F, H, H&E, ×400)

Genetic and Molecular Findings

A germline pathogenic, truncating DICER1 mutation was detected in a peripheral blood sample from the patient (p.Arg1060Ilefs*7). DICER1 somatic mutations were searched by Sanger sequencing in formalin-fixed, paraffin-embedded tissue from the bERMS, thyroid nonnodular parenchyma, and five thyroid nodules/tumors (Image 2): one adenomatous nodule (Image 3), one adenomatous nodule with papillary hyperplasia (Image 3), two tumors with “intermediate” nuclei (Image 3), and the WDTC (Image 3). The presence of the germline mutation was confirmed in all tissues. No DICER1 somatic mutations were detected in the nonnodular thyroid parenchyma or in the adenomatous nodules, with or without papillary hyperplasia. Distinct somatic DICER1 missense mutations in the RNase IIIb hotspot domain were identified in the two tumors with “intermediate” nuclei (p.Glu1813Gly and p.Asp1810Asn, respectively) in the WDTC (p.Glu1813Gly) and in the bERMS (p.Glu1813Asp) (Image 2). No mutations of TERT promoter, BRAF600, NRAS61, HRAS12/61, and KRAS12/61 or rearrangements of RET/PTC1, RET/PTC3, and PAX8/PPARγ were detected in any sample.

Discussion

The occurrence of WDTC in the context of MNG rarely has been reported in DICER1 syndrome.³ Pleuropulmonary blastoma, cystic nephroma, Sertoli-Leydig cell tumor, and, like in the present case, bERMS have been described to occur together with WDTC in DICER1 mutation carriers.⁴

The histologic description and diagnosis of DICER1 syndrome–associated thyroid malignancy vary from publication to publication. Both PTC (n = 20) and FTC (n = 6) have been reported Table 1.³ Demonstrating the difficulty of making the differential diagnosis between FTC and PTC in this setting, Shin et al¹⁴ reported one FTC that, when revisited by other authors,⁷ was reclassified as PTC. We think the reason for such diagnostic difficulty stems from the fact that, like in the present case, the morphologic features of the tumors do not fit with either PTC or FTC, thus supporting the diagnosis of WDTC NOS.

Table 1

Reference	No. of Cases	Histopathologic Classification
Rome et al4	1	FTC
Oue5	1a	FTC
Poiana6	1a	PTC
de Kock et al7	3a	fvPTCb; fvPTCc; PTC within FA
Schultz8	1a	fvPTC
Puckett9	1	PTC
Durieux et al10	2a	fvPTC; FTC
Rutter et al11	2a	PTC foci within adenomatous nodules
de Kock et al12	3	PTC
Khan et al3	11	PTC (n = 2); fvPTC (n = 6); FTC (n = 3)

FA, follicular adenoma; FTC, follicular thyroid carcinoma; (fv)PTC, (follicular variant) of papillary thyroid carcinoma; PTC, papillary thyroid carcinoma.

^aCases in which pictures of the well-differentiated thyroid carcinomas were published.

^bThis case was also described by Slade et al.¹³

^cThis case was also described Shin et al¹⁴ and was first classified as FTC.

The “intermediate” appearance of the nuclei of the neoplastic cells in DICER1-related thyroid carcinomas fits with the molecular profile of such carcinomas: none of the molecular alterations characteristically associated with PCT or FTC (BRAF600, NRAS61, HRAS12/61, and KRAS12/61 or TERT promoter mutations and RET/PTC1, RET/PTC3, and PAX8/PPARγ rearrangements) was detected in the present case, and this finding is consistent with the findings in previous reports.^11,14 Yoo et al¹⁵ have recently proposed a three-tiered molecular classification of WDTC that includes, besides BRAF-like and RAS-like, a non-BRAF, non-RAS subgroup of tumors. Noteworthy, DICER1 is one of the driver genes defining the latter subgroup.¹⁵ Taking all this together, one may conclude that DICER1 syndrome–related WDTC reflects a carcinogenic process that is distinct from the classical pathways toward PTC or FTC.

It has been advanced that previous chemotherapy, especially with alkylating agents, could be, in some cases, the trigger of additional mutations and might accelerate the carcinogenic process.^3,7 However, the occurrence of WDTC in DICER 1 syndromic patients has been reported independently from previous treatment.^1,3,10,11 In the present case, WDTC developed after an intensive chemotherapy protocol, and we cannot exclude that such treatment predisposed the patient to the development of thyroid carcinoma.

The molecular findings, presented in this case report, are limited to a single case study. However, our data support previous findings in similar cases,^3,11,12 and thus we think they may contribute to clarify the peculiar carcinogenic process of DICER1 syndrome–related thyroid tumors and their morphologic/matched molecular characteristics.

Somatic mutations in the characteristic hotspot RNAase III domain have been detected in both benign and malignant tumors. These mutations are predicted to affect the metal-ion binding function of the catalytic core of the protein.¹² In several studies,^3,11,12 distinct mutations were detected in different tumors from the same individual, as also occurred in the present case. The two most important findings refer to the absence of mutations in adenomatous nodules and the detection of the same mutation in one of the noninvasive tumors with “intermediate” nuclei and the WDTC. Two studies, by de Kock et al¹² and Khan et al,³ have recently proposed a stepwise model to explain the development of DICER1 syndrome–related MNG/WDTC: DICER1 biallelic mutations may offer a proliferative advantage (as demonstrated, in the present case, by the high mitotic activity in tumors with “intermediate” nuclei), thus creating a favorable milieu for the development of cancer. Following such an initiation event, additional molecular alterations, not yet described, are probably required for malignant transformation in terms of invasiveness.^3,12

The evidence we have obtained confirms that DICER1 somatic mutations may determine the progression from adenomatous nodules to noninvasive tumors and that additional events are probably necessary to drive invasion. It would be interesting to confirm this hypothesis by microdissecting the cells with “intermediate nuclei” within an adenomatous nodule/tumor and finding whether DICER1 somatic mutations are confined to the neoplastic cells with “intermediate” nuclei.

This study was supported by the Portuguese Foundation for Science and Technology through a PhD grant to R.B. (ref.: SFRH/BD/111321/2015). IPATIMUP integrates the i3S Research Unit, which is partially supported by FCT. This work was financed by FEDER—Fundo Europeu de Desenvolvimento Regional funds through the COMPETE 2020—Operacional Programme for Competitiveness and Internationalisation (POCI), Portugal 2020, and by Portuguese funds through FCT—Fundação para a Ciência e a Tecnologia/Ministério da Ciência, Tecnologia e Inovação in the framework of the project “Institute for Research and Innovation in Health Sciences” (POCI-01-0145-FEDER-007274). Further funding was obtained from the project “Advancing Cancer Research: From Basic Knowledgment to Application” (NORTE-01-0145-FEDER-000029) and “Projetos Estruturados de I&D&I,” funded by Norte 2020—Programa Operacional Regional do Norte.

Author notes

" Primary authors.