Invasive follicular variant of papillary thyroid cancer harboring the NRAS mutation Q61K and presenting with bone metastasis-A case report.

INTRODUCTION: The follicular variant of papillary thyroid cancer (FVPTC) can be noninvasive or invasive. The invasive form of FVPTC commonly harbors BRAF mutations whereas RAS mutations are more often associated with noninvasive FVPTC and a favorable clinical outcome. CASE REPORT: A 47-year-old man presented with a metastasis to his right iliac bone as the initial manifestation of a 1.6cm invasive FVPTC. After total thyroidectomy, the patient underwent additional treatment, including thyroid hormone suppressive treatment to non-detectable TSH levels, repeated courses of radioiodine treatment, external beam radiation, and treatment with the tyrosine kinase inhibitor sorafenib. Despite these therapeutic efforts, the disease progressed with growth of the iliac mass and additional metastatic spread to cervical and lumbar vertebrae causing increasing pain and disability. The patient succumbed to the disease four years after presentation. Retrospective next-generation sequencing of the primary tumor using a pan-cancer targeted mutation and gene fusion panel revealed NRAS Q61K mutation and no other oncogenic alterations. DISCUSSION: The study challenges the concept that thyroid neoplasms with isolated RAS mutations are often associated with favorable clinical behavior and may be candidates for conservative management.
CONCLUSION: An isolated RAS mutation in invasive FVPTC may be associated with an aggressive clinical behavior.

Introduction

The follicular variant of papillary thyroid cancer (FVPTC) is characterized by nuclear features consistent with classical PTC (cPTC) but having follicular rather than papillary architecture [1]. Recent reports suggest that the incidence of FVPTC is on the rise, accounting for up to half of all PTCs in some patient series [2], [3].

FVPTC consists of different subgroups with varying pathological and clinical features [4], [5]. Although the classification of distinct types of FVPTC has varied, in recent studies [3], [4], [5], [6], [7], FVPTC was subdivided into noninvasive tumors (encapsulated and/or well-circumscribed tumors without tumor capsule or lymphovascular invasion) versus invasive tumors (encapsulated tumors with tumor capsule or lymphovascular invasion or tumors not being well-circumscribed but showing infiltration into adjacent thyroid tissue).

There is increasing evidence that noninvasive FVPTC has a favorable clinical behavior and may not behave like a cancer but may be an indolent tumor with very low risk of adverse outcome requiring less aggressive treatment than has commonly been recommended [3], [4], [5], [6], [7], [8]. The noninvasive FVPTC, therefore, is presently undergoing reclassification from cancer to a low-risk tumor with benign clinical characteristics and was recently proposed to be labeled “noninvasive follicular thyroid neoplasm with papillary-like nuclear features” (NIFTP) [8]. In contrast to the indolent clinical behavior of NIFTP, invasive FVPTCs can be associated with both regional and distant metastases, recurrent disease, and even death [9], [10], [11].

Among follicular thyroid neoplasms with papillary-like nuclear features, mutations in the RAS oncogene have been associated with encapsulated/well-circumscribed tumors (including NIFTP) whereas invasive FVPTC has more commonly been associated with BRAF mutations [8], [12]. Studies suggest, however, that there is not an absolute correlation between gene mutations and the clinical behavior of thyroid neoplasms [13], [14], [15].

We report a patient who presented with a bone metastasis as the first manifestation of an invasive FVPTC and who developed additional bone metastases during the course of treatment and ultimately died from his disease. Subsequent molecular analysis revealed that the patient’s primary thyroid tumor harbored an isolated NRAS Q61K mutation with no co-existing mutations detected by a broad, clinically utilized next-generation sequencing oncogene/gene fusion panel. The report is important because it suggests that follicular-patterned neoplasms that harbor an apparently isolated RAS mutation may demonstrate an aggressive clinical course. Our study, therefore, challenges recent views that thyroid tumors with apparently isolated RAS mutations may be candidates for conservative management [14], [15]. The report adds information to the existing literature and provides an important learning point significant for the management of patients with invasive FVPTC.

Case report

A 47-year-old man with no family history of thyroid cancer and who was previously healthy was seen in the emergency room because of progressive right hip and groin pain for approximately one year. Abdominal and pelvic CT revealed a 13 cm hypervascular mass arising from the right iliac bone with extensive bone destruction and involvement of the right psoas, iliopsoas, and gluteal muscles (Fig. 1). A subsequent core biopsy of the mass showed evidence of metastatic thyroid cancer with histopathology revealing follicular architecture and some nuclear features consistent with PTC (Fig. 2). Further metastatic workup, including chest CT and brain MRI, revealed no evidence of additional metastases. A thyroid ultrasound showed a 2 cm nodule with internal calcifications in the right thyroid lobe (Fig. 3). The patient underwent a total thyroidectomy to facilitate radioiodine treatment as part of the management of the iliac metastasis. At the time of the thyroidectomy, no central compartment lymphadenopathy was appreciated by inspection or palpation. Histopathological examination of the thyroid gland showed a 1.6 cm infiltrative FVPTC with focal extrathyroidal extension and vascular invasion (Fig. 4). There were no regional lymph node metastases. Despite aggressive treatment that included suppressive thyroid hormone treatment to non-detectable TSH levels, repeated courses of radioiodine treatment, external beam radiation of the iliac bone metastasis, and treatment with the kinase inhibitor sorafenib, the disease progressed with growth of the iliac/pelvic mass and additional metastatic spread to cervical and lumbar vertebrae causing worsening pain and disability. Four years after his presentation, the patient, now terminally ill, returned to his homeland to spend time with his family. It should be noted that the patient described here was among a group of 179 patients with FVPTC that we reported recently [3]. Although the patient was mentioned in that report, detailed descriptions of histopathological and imaging findings and gene sequencing were not included in our previous study.

Fig. 1

Abdominal/pelvic CT obtained when the patient presented to the ER with right hip pain.

Fig. 2

Core biopsy of right iliac lesion. (A) Follicular cells with ovoid nuclei, chromatin pallor, and mild nuclear contour irregularity (hematoxylin & eosin, 400× magnification). Tumor cells are positive for thyroglobulin (B) and TTF-1 (C) by immunohistochemistry (400× magnification).

Fig. 3

Thyroid ultrasound demonstrating a 1.4 × 1.5 × 2 cm heterogenous nodule with internal calcifications in the right thyroid lobe.

Fig. 4

Total thyroidectomy. (A) Thyroid tumor partially surrounded by fibrous capsule (dotted line), with infiltrative growth into the surrounding thyroid parenchyma (20× magnification). (B) Tumor cells show follicular architecture and nuclear atypia characteristic of papillary thyroid carcinoma, including nuclear crowding, mild nuclear contour irregularity, nuclear grooves (arrows), and chromatin pallor (400× magnification). (C) Tumor shows vascular invasion (arrow; 100× magnification). Hematoxylin & eosin.

The case report adheres to the Surgical CAse REport (SCARE) guidelines [16].

Targeted next-generation sequencing (NGS)

Analysis of oncogenic alterations was performed retrospectively from the patient’s archival formalin-fixed paraffin-embedded thyroidectomy specimen using targeted next-generation sequencing (NGS) technology as described recently [17]. The NGS mutation panel assayed for hotspot mutations in 91 genes (including BRAF, NRAS, HRAS, KRAS, TERT, P53) and the gene fusion panel targeted rearrangements involving 52 genes, including RET and PPARG. Results from the molecular analysis revealed that the patient’s thyroid tumor harbored an NRAS Q61K mutation. Details regarding the NGS methodology and the list of targeted genes are provided in the Supplement.

Discussion

In earlier reports, tumors diagnosed as FVPTC were commonly described and managed as one entity. More recently, the importance of subdividing these tumors into noninvasive and invasive neoplasms has become clear [4], [5]. In particular, the recognition that NIFTP may be sufficiently treated by lobectomy alone without requirement for subsequent radioiodine treatment has important clinical implications [8]. In contrast, invasive FVPTC should be recognized as malignant with potential for distant metastases and even death from disease [9], [10], [11]. Invasive FVPTCs should be managed by total thyroidectomy and, if indicated, postoperative radioiodine treatment. Therefore, if a hemithyroidectomy is performed as initial procedure, no further surgery may be needed if pathology is consistent with NIFTP, whereas a completion thyroidectomy should be performed if pathology reveals invasive FVPTC.

The majority of thyroid cancers harbor single gene mutations, most commonly genes in the BRAF or RAS signaling pathways [18], [19]. The relationship between RAS mutations and the clinical behavior in thyroid tumors are controversial. Recent reports suggest that thyroid nodules with RAS mutations tend to be benign or indolent and that thyroid cancers harboring isolated RAS mutations are commonly associated with an excellent prognosis [14], [15].

In contrast, other studies have suggested that RAS mutations are markers for aggressive behavior in thyroid cancers [13], [20]. Data from NGS analysis of poorly differentiated and anaplastic thyroid cancers provide a possible explanation for the disparate biological behavior reported for RAS-mutated tumors. Thus, RAS mutations may represent an early step in tumorigenesis with the accumulation of subsequent genetic changes promoting cancer progression and biologically aggressive phenotypes. In particular, broad genotyping studies have identified coexistence of RAS mutations with mutations of the TERT promoter and/or EIF1AX as possible signatures of aggressive tumor behavior [20].

Most previous reports of aggressive invasive FVPTCs with distant metastases [9], [10], [11] predate the widespread use of NGS technology why their molecular composition is unclear. Our report represents an unusual example of an aggressive invasive FVPTC harboring an isolated RAS mutation, contradicting recent reports in which an isolated RAS mutation was considered indicative of a low-risk tumor [14], [15].

A limitation of our report is that the molecular analysis does not rule out that genes, other than RAS, may have been mutated in the patient’s tumor and could have escaped detection by the NGS cancer panel used here. While mutations in TERT promoter, TP53, and other genes were not identified in our case, the NGS panel did not test for mutations in EIF1AX or other molecular changes that may provide prognostic information, such as microRNA expression profiles and somatic copy number alterations. The ability to rule out co-existing mutations are directly related to the breadth and relevance of the molecular alterations that are included in a test panel. Because most general NGS cancer panels target the most frequently altered oncogenes and tumor suppressors, less common molecular changes may have escaped detection. As NGS tests, particularly those tailored for thyroid cancer, strive towards a more comprehensive panel of molecular alterations, prognostication of RAS-mutated thyroid cancers would be expected to improve in the future.

Conclusions

Our report provides novel information about invasive FVPTC and suggests that apparently isolated RAS mutations can be found in invasive FVPTC and may be associated with aggressive clinical behavior. The observations provide an important learning point and have implications with regards to surgical approach, adjuvant treatment, and patient prognosis.

Conflicts of interest

None of the authors has any conflict of interest to declare.

Funding

CN was supported by grants from the National Cancer Institute/National Institutes of Health (1R21CA165039-01A1 and 1R01CA181183-01A1), the American Thyroid Association (ATA), the ThyCa: Thyroid Cancer Survivors Association Inc. for Thyroid Cancer Research, the Guido Berlucchi “Young Investigator” research award 2013 (Brescia, Italy), and the BIDMC/CAO, Boston, MA.

Ethical approval

N/A (All images are completely anonymised, the individual cannot be identified by any of the images or through any part of the text).

Consent

N/A (All images are completely anonymised, the individual cannot be identified by any of the images or through any part of the text).

Author contribution

Study concept and design: Raman Mehrzad, Per-Olof Hasselgren.

Data collection: Michiya Nishino, Carmelo Nuccera, Dora Dias-Santagata.

Interpretation of data: Raman Mehrzad, Michiya Nishino, Carmelo Nucera, DoraDias-Santagata, James V. Hennessey, Per-Olof Hasselgren.

Initial drafting of paper: Raman Mehrzad, Per-Olof Hasselgren.

Registration of research studies

N/A.

Guarantor

Per-Olof Hasselgren.