Cytologic risk stratification of medullary thyroid carcinoma: Does it make the grade?

Abstract

Medullary thyroid carcinoma (MTC) is a neuroendocrine malignancy arising from calcitonin-producing parafollicular/C cells of the thyroid. For patients with MTC, risk stratification relies on clinical, radiologic, pathologic, molecular, and serologic results. These data points collectively guide decisions regarding the extent of primary treatment, intensity of disease monitoring, and management of recurrent, persistent, or metastatic disease.¹ Until recently, histologic grading had not been a widely accepted part of the clinicopathologic risk stratification of patients with MTC.

In 2020, teams from the Memorial Sloan Kettering Cancer Center (MSKCC) and Royal North Shore Hospital in Sydney, Australia, independently published their respective histologic grading systems for MTC. The MSKCC system proposed a binary grading system based on tumor necrosis and mitotic activity,² whereas the Sydney system proposed a three-tier grading system based on mitotic count, Ki-67 labeling index, and tumor necrosis.³ A harmonized, two-tier grading scheme based on these two systems was developed in 2022.⁴ This consensus International Medullary Thyroid Carcinoma Grading System (IMTCGS) requires the assessment of tumor necrosis, mitotic count (≥5 mitoses per 2 mm²), and Ki-67 proliferative index (≥5%) on resected MTCs. Tumors with one or more of the above three features are classified as high-grade MTC, whereas those that are negative for all three parameters are low grade (Figure 1). The prognostic potential of the IMTCGS was demonstrated on a retrospective, multi-institutional cohort of 327 MTC cases⁴ and further validated in additional retrospective studies.^{5, 6} The fifth edition of the World Health Organization classification of endocrine and neuroendocrine tumors endorses the IMTCGS,⁷ and the College of American Pathologists includes this grading scheme in the synoptic reporting of MTC.⁸

In this issue of Cancer Cytopathology, Viswanathan et al. explore whether the practice of MTC grading can be extended to fine-needle aspiration (FNA) cytology samples as well.⁹ Their study involved 25 surgically resected MTCs (23 primary thyroid tumors and two lymph node metastases), which were stratified into 16 low-grade and nine high-grade tumors per the IMTCGS. FNA samples associated with these resected tumors were evaluated via the same three IMTCGS criteria of necrosis, mitotic count, and Ki-67 labeling. The authors examined the degree to which FNA cytology-based grade correlated with the reference histologic grade.

Among the three histologic grading criteria in the IMTCGS, assessment of the Ki-67 proliferation index emerged as the most straightforward to evaluate in FNA cytology specimens. Via the same ≥5% Ki-67 labeling cutoff established for histologic grading, cytologic Ki-67 labeling demonstrated high specificity, albeit low sensitivity, for high-grade MTC. In comparison, identification of tumor necrosis in FNA cytology specimens appeared to be more difficult. None of the FNA samples associated with high-grade MTC in their study showed conclusive evidence of the cellular debris characteristic of coagulative necrosis. Factors contributing to the low sensitivity of FNA for necrosis likely included sampling error, difficulty distinguishing necrotic from fibrinous debris, and liquid-based preparations that reduce background material on slides. However, the authors do point out that in FNA cytology samples, “apoptotic bodies” (perhaps reminiscent of the “punctate karyorrhectic nuclear debris” described by the IMTCGS⁴) could serve as a proxy for the tumor necrosis seen in resections of high-grade MTC.

For histologic grading of MTCs, mitotic count is defined as the number of mitotic figures per 2 mm² of tumor (approximately 10 continuous high-power fields). Understandably, this method of measuring mitotic activity proved to be impractical on FNA cytology specimens, given the tendency for aspirated tumor cells to be unevenly distributed on cytology preparations. The authors explored different ways of reporting mitotic counts on cytology slides (e.g., maximum number of mitotic figures per slide and mitotic count per 50 high-power fields), and noted that multiple mitotic figures per slide were seen only in FNAs of high-grade tumors. Although not used in their study, the mitotic index (percentage of tumor cells in mitoses) could be a relatively simple alternative that is more straightforward to standardize on cytology specimens. Manual counting of mitoses and tumor cells may be time consuming for routine practice, but it is similar to the recommended method of calculating Ki-67 labeling and potentially amenable to digital cytology and automated counting methods.

Viswanathan et al.’s study highlights a few caveats regarding tumor grading on cytology specimens, first and foremost of which is sampling adequacy. The IMTCGS recommends “generous sampling of tumors in resection specimens” to ensure accurate histologic grading.⁴ In MTC resections, mitotic counts and Ki-67 labeling are evaluated at hotspots of proliferative activity, and even a limited focus of tumor necrosis is considered sufficient for classification as high grade. In this light, the low sensitivity of FNA cytology for proliferative grade and necrosis is not surprising. Undersampling and potentially undergrading tumors remain key concerns when adapting resection-based grading schemes to small biopsy specimens. Another question raised by this study involves the clinical relevance of grading metastatic (as opposed to primary) tumors. The retrospective validation study of the IMTCGS was performed on primary thyroid tumors only.⁴ Yet, in the current study, histologic and cytologic grading were performed in metastatic tumors (paratracheal lymph node and cervical lymph node) for two cases. Furthermore, in two additional cases, cytologic grade was performed on FNA samples of metastatic sites (lymph node and liver), whereas the histologic grade was determined on the primary thyroid tumor.

Perhaps a more fundamental issue regarding cytologic grading of MTC is one of therapeutic utility. To paraphrase one of Dr Ian Malcolm’s iconic lines in Jurassic Park,¹⁰ just because we can grade MTC on cytology specimens, should we be doing this in routine clinical practice? An examination of current MTC management guidelines sheds some light on this question. Total thyroidectomy is the standard of care for MTC, with consideration of central and/or lateral neck lymph node dissection based on radiologic assessment of tumor size and extent, serum calcitonin and carcinoembryonic antigen levels, and intraoperative findings.^{1, 11} Given that IMTCGS histologic grade is an independent predictor of locoregional recurrence,⁴ cytologic grade on preoperative FNA samples could theoretically be added to this list of features that guide recommendations regarding aggressive prophylactic neck dissection. A potential prognostic and/or theranostic role for integrating cytologic grade with FNA-based molecular testing can also be considered,^12-14 in view of the emerging insights into the relationship between histologic grade, molecular profile, and outcomes of MTCs.¹⁵ As we gain practical experience with both histologic and cytologic grading of MTC, studies that help define the role and priority of grading in clinical decision-making will be valuable. The proof-of-concept study by Viswanathan et al. is an important first step toward this effort.

Michiya Nishino: Conceptualization, writing–original draft, and writing–review and editing.

Michiya Nishino has received contractor fees from Ebix/Oakstone Continuing Medical Education, unrelated to the submitted work.