Uncommon diagnostic pitfalls of mucoepidermoid carcinoma on cytology: A tertiary care center, retrospective descriptive study

Divya Shelly; PS Mishra; Prachi Nichat; AK Das; Pratibha Chandra

doi:10.4103/jmms.jmms_43_21

ORIGINAL ARTICLE

Year : 2022 | Volume : 24 | Issue : 1 | Page : 37-41

Uncommon diagnostic pitfalls of mucoepidermoid carcinoma on cytology: A tertiary care center, retrospective descriptive study

Divya Shelly¹, PS Mishra², Prachi Nichat³, AK Das¹, Pratibha Chandra¹
¹ Department of Pathology, INHS Asvini, Mumbai, Maharashtra, India
² Department of Pathology, Army Hospital Research and Referral, Delhi, India
³ Department of Pathology, MH Secunderabad, Andhra Pradesh, India

Date of Submission	16-Mar-2021
Date of Decision	19-Jun-2021
Date of Acceptance	17-Jul-2021
Date of Web Publication	21-Jan-2022

Correspondence Address:
Dr. Pratibha Chandra
Department of Pathology, INHS Asvini, Near RC Church, Colaba, Mumbai-400005
India

Source of Support: None, Conflict of Interest: None

Check

DOI: 10.4103/jmms.jmms_43_21

Abstract

Introduction: Diagnosis of mucoepidermoid carcinoma (MEC) on fine-needle aspiration cytology (FNAC) is grim with important diagnostic pitfalls, leading to wrong treatment decisions. This study highlights uncommon mimics of MEC on FNAC smears of major salivary glands and compares the cytologic findings with definitive histopathology diagnosis for identification of potential diagnostic pitfalls. Methods: This is a retrospective descriptive study of MEC cases diagnosed over a duration of 5 years (April 2015–April 2020) at a tertiary care center with available preoperative FNAC and postoperative histopathology resection specimens. Results: Out of a total of 18 MEC cases diagnosed by histopathologic examination, 8 (44%) were wrongly diagnosed on preoperative FNAC as a different benign or malignant entity. Further details of these cases are shared in the text. Discussion: Although FNAC remains an important preoperative diagnostic tool in salivary gland lesions, utmost care is required in the cases of MEC which are notorious for misinterpretation on cytology. A number of uncommon mimics, both benign and malignant, need to be considered and carefully excluded to spare the patient of avoidable miseries of misdiagnosis. Conclusion: MECs of salivary glands can mimic the morphology of a variety of benign as well as malignant lesions on cytology with low cyto-histologic concordance especially in cystic lesions.

Keywords: Fine-needle aspiration cytology, mucoepidermoid carcinoma, salivary gland lesion, uncommon diagnostic pitfalls

How to cite this article:
Shelly D, Mishra P S, Nichat P, Das A K, Chandra P. Uncommon diagnostic pitfalls of mucoepidermoid carcinoma on cytology: A tertiary care center, retrospective descriptive study. J Mar Med Soc 2022;24:37-41

How to cite this URL:
Shelly D, Mishra P S, Nichat P, Das A K, Chandra P. Uncommon diagnostic pitfalls of mucoepidermoid carcinoma on cytology: A tertiary care center, retrospective descriptive study. J Mar Med Soc [serial online] 2022 [cited 2023 Mar 23];24:37-41. Available from: https://www.marinemedicalsociety.in/text.asp?2022/24/1/37/336194

Introduction

Fine-needle aspiration cytology (FNAC) is a clinically useful preoperative diagnostic tool in salivary gland lesions with acceptable accuracy. It is often used for guiding the clinician in assigning the risk of malignancy and in instituting definitive treatment for neoplastic cases.^[1] The six-tier Milan System for Reporting Salivary Gland Cytopathology (MSRSGC) 2019 has proved invaluable in achieving these objectives as also in establishing a common mode of communication between the clinician and the cytopathologist.^[1] FNAC diagnosis has a remarkable sensitivity and specificity for salivary gland lesions, approaching 80% and 88%, respectively.^[1] However, these figures are not uniform for all types of salivary lesions, with mucoepidermoid carcinoma (MEC) being a prominent example.^[2] Therefore, the emphasis on strictly adhering to diagnostic guidelines as elucidated in MSRSGC cannot be overemphasized in order to minimize wrong preoperative diagnosis. MEC is known to pose an array of diagnostic challenges even at the hands of best and most experienced cyto-pathologists possibly due to its inherent clinical and morphologic heterogeneity. Presence of mucus-producing cells, intermediate cells, as well as epidermoid cells in the aspirate is warranted for its accurate diagnosis. However, predominance of only one type of cells along with some other cytologic red herrings can lead to important diagnostic pitfalls.^[3] MEC is the most common primary salivary gland malignancy accounting for up to 30% of cases with frequent involvement of the major salivary glands. It is attributable for significant morbidity and mortality in patients depending on the stage and grade of the disease.^[2],[3] A number of well-known cytologic mimics of MEC are described in the literature, and they include a heterogeneous group of benign and malignant lesions such as sialadenitis, sialometaplasia, pleomorphic adenoma (PA), adenosquamous carcinoma, and metastatic malignancies. Though these lesions can cause diagnostic misperception during FNAC interpretation, often, they can be identified and a correct diagnosis of MEC can be rendered by expert cyto-pathologists. On the other hand, some lesser-known uncommon entities can still lead to a wrong diagnosis, causing management errors in the cases of MEC. In this study, we have highlighted the challenges faced while diagnosing MEC of salivary glands on FNAC with special emphasis on pitfalls resulting due to uncommon cytologic mimics.

Methods

This was a retrospective descriptive study carried out at a tertiary care military hospital in the western India. Inclusion criteria for this study were as follows: cases of MEC diagnosed between April 2015 and April 2020 at our tertiary care hospital along with cases of MEC diagnosed on histopathology, involving major salivary glands, with a preoperative FNAC diagnosis. Exclusion criteria for this study were as follows: cases involving minor salivary glands were excluded from this study, primarily due to the lack of a preoperative FNAC diagnosis along with cases with incomplete or unavailable clinical or radiologic data.

The relevant clinical data of all cases in this study, including radiology findings, were retrieved from hospital records and noted. FNAC and histopathology findings of these cases were reviewed independently by two experienced onco-pathologists (PSM and DSH). In cases with differences of opinion, a diagnosis was reached by consensus. In few cases, immunohistochemistry (IHC) was performed with selected panel of antibodies, as guided by standard current literature, on archived tissue blocks to rule in or rule out the closest differential diagnosis in respective cases. IHC was performed by manual method following the standard institutional protocols.

All cytology and histopathology slides were viewed under Olympus Semi-motorised fluorescence microscope Model BX53, Japan.

A final diagnosis in each case was given after comprehensive evaluation of all relevant clinical, radiological, cytological, and histopathology findings. Histopathologic examination (HPE) proved to be diagnostic in all cases, and no molecular studies were performed in this study. Grading of MEC was done according to the Brandwein system.

Frequency calculation was used as the method of statistical analysis.

Results

A total of eighteen cases with histopathological diagnosis of MEC, involving the parotid (16) and the submandibular glands (2), were selected for this study. Ages of the cases ranged from 14 to 78 years with a mean of 54.8 years. Eleven patients were males and seven were females with a M: F ratio of 1.6:1. Out of a total of 18 cases, 10 (56%) were diagnosed as MEC on both preoperative FNAC and final HPE, whereas 8 (44%) had a discordant diagnosis as an entity other than MEC on preoperative FNAC [Table 1]. All correctly diagnosed MEC cases were of the parotid gland, whereas 2 out of 8 wrongly diagnosed cases were from the submandibular gland and the rest 6 were from the parotid gland. Five of 8 (62.5%) discordant cases had cystic areas within the tumor; Out of the 8 cases wrongly diagnosed by FNAC; 2 (25%) were labeled nonneoplastic, with 1 each diagnosed as a mucous retention and an inflammatory cyst. Three of the 8 (37.5%) cases were put in benign neoplastic category: one each as a Warthin's tumor, an oncocytoma, and a PA. Rest 3 of the 8 (37.5%) cases were diagnosed as acinic cell carcinoma. [Table 2] shows the clinical findings, radiology, pre-operative FNAC, and HPE diagnosis on resection specimens of these 8 MEC cases.

Table 1: Cases of mucoepidermoid carcinoma with concordant and discordant fine-needle aspiration cytology and histopathological examination diagnosis

Click here to view

Table 2: Clinical and radiological features along with fine-needle aspiration cytology diagnosis and final histopathological examination grading of all the eight discordant cases of the study

Click here to view

Discussion

Current diagnosis of salivary gland tumors is based on a comprehensive analysis of clinical findings, radiologic features, cytologic findings, and HPE findings with an ever-evolving role of molecular diagnostics. Even so, the role of FNAC as a minimally invasive, cost-effective, and rapid initial diagnostic modality remains clinically relevant. Radiology is pivotal in exploring the site and extent of involvement, while FNAC is required in diagnosing suspicious neoplastic lesions. There are four questions to be answered on FNAC: is the origin of mass salivary gland or not, is the lesion neoplastic or nonneoplastic, if neoplastic – is it benign or malignant and finally, is it a primary or a metastatic lesion. The answers influence the management and course of treatment. If proven malignant, the decision for surgery and management of facial nerve can be taken in advance with informed consent from the patient for possible adverse outcomes. In cases of lymphomas, surgery can be altogether avoided. However, like all diagnostic investigations, there are some pitfalls while interpreting FNAC of salivary glands, particularly with respect to MEC. We, in this study, discuss uncommon FNAC mimics of MEC of major salivary glands.

The age and sex profiles of MEC cases in this study were at variance with other similar studies, with the findings of more common occurrence in late adulthood (mean 54.8 years) and a slight male preponderance (M: F = 1.6:1). MEC is the most common malignant salivary gland tumor in children and young adults. In adults, it affects the age group from third to sixth decade with a female predilection.^[3],[4] This age- and sex-related variance in our study does not have clear explanations; however, it may be partly due to the patient cohort in our study which is representative of pan-India owing to their affiliation to the Indian Armed Forces. The details of ethnicities and geographic distribution of cases have not been included in this study.

All the cases presented with palpable painless mass in the parotid or the submandibular gland, with one case having features of facial nerve palsy. Radiology of salivary glands is challenging due to occurrence of a wide variety of lesions in this organ and a considerable overlap of radio-imaging features. Ultrasonography and/or magnetic resonance imaging (MRI) were performed preoperatively in all cases, which remain the radiologic investigations of choice in salivary glands. In our study, ultrasonography in five out of seven cases showed a hypoechoic lesion. However, the accuracy for diagnosis of malignant salivary gland lesions on ultrasonography is low at approximately 20%.^[5] Ultrasound can be used to assess the cystic nature of the lesion and may be useful in guiding FNAC. Seven cases in our study with some cystic lesional components were examined by ultrasound-guided FNAC. MRI is the radiologic investigation of choice in cases suspicious for neoplasms.^[6],[7] Fourteen cases were undertaken for MRI which described the extent of these salivary gland lesions. Out of the total 18 cases studied, an overwhelming number, i.e. 16 (90%) were seen in one of the parotids, which is consistent with the far more common occurrence of MEC in this location.

Eight of the 18 (44%) cases in our study were wrongly diagnosed on preoperative FNAC as a condition other than MEC. Other studies on this topic put this figure at 15%–50%, thereby placing our finding in the standard range of cytopathological discordance observed in MEC.^[8] The presence of epidermoid, intermediate, and mucus-producing cells in cytology smears from a salivary gland is diagnostic of MEC.^[9] Thick mucoid background is another consistent finding. The grading can be low, intermediate, or high depending on the cellularity of smears, proportion of intermediate- to mucus-producing cells, and nuclear features. However, it is often difficult to grade MEC on cytology alone and the same has not been attempted in this study. Partly, cystic nature of 5 out of 8 (62%) lesions may provide an explanation for a high normal percentage of cyto-histopathological discordance in our study. We know that FNA yield in cystic lesions of the salivary gland may be inadequate or unrepresentative of the disease and the pathologist is likely to commit a mistake in the enthusiasm of making a timely diagnosis. A repeat of FNA or a guided FNA and taking material from multiple passes in many of these cases may be the key to a correct diagnosis.^[10]

An interesting and the most important finding of our case series has been the variety and types of preoperative FNAC diagnoses offered in the eight discordant cases. The benign nonneoplastic conditions such as mucus retention cysts and inflammatory cysts are well-known differential diagnosis of MEC on FNAC and were labeled as such in 2 of these 8 (25%) cases. Each of these two cases in this study [Sr nos. 4 and 7, [Table 2]] comprised cystic components, emphasizing the need for careful interpretation of FNAC in such cases and asking for ultrasound-guided FNA with preferably multiple passes in the lesion. Low- and intermediate-grade MECs are particularly prone to this kind of misdiagnosis as they may yield pauci-cellular aspirates owing to their cystic nature and sometimes may even be acellular with only thick viscid mucus. It is deleterious to diagnose such cases as mucocele or retention cyst under the ever-increasing pressure of offering timely preoperative diagnosis. Similar finding of pools of mucin with few foamy cells was noted in one case of our study despite aspirating it on two occasions and was wrongly diagnosed as an inflammatory cyst. It turned out to be an intermediate-grade MEC on histology. Some cases may show inflammatory cells as seen in the other case in this category. Scattered mucinous cells may also be mistaken for foamy histiocytes, in which case, clustering of such cells must be sought for, because it is indicative of epithelial origin.^[11]

Three of the 8 (37.5%) MECs were diagnosed as benign neoplasms in this study, and barring from PA they are sparingly mentioned in the literature as a potential pitfall in FNAC diagnosis of salivary lesions. This has led the authors to believe that there may be a need for further similar studies to find out their true frequency as common red herring in such cases. One of these two cases was of a cystic-solid lesion of parotid in a young male [case 5, [Table 2]] which was diagnosed as Warthin's tumor on FNAC. The smears showed clusters of polygonal cells with abundant oxyphilic cytoplasm [Figure 1] and round regular nuclei in a background of granular debris, scattered lymphocytes, cyst macrophages, and acinar cells. Intermediate cells of MEC may be confused with oncocytes, both being polygonal with abundant cytoplasm. The distinctive features are presence of sharp cytoplasmic borders in oncocytes and homogenous cytoplasm in intermediate cells as opposed to granular in the former.^[12] Interpretation of these findings requires meticulous clinicopathological correlation as most of the cases of Warthin's tumor are seen in male smokers and the radiologic findings may be of a well-delimited, solid, cystic lesion, invariably in the parotid. At times, mucus cells may also resemble oncocytes. The scattered lymphoid cells, a prominent feature in Warthin's tumor on FNAC, are seen in up to 30%–40% of MECs and are derived from the inflammatory host response noted in the stroma.^[13] The incorrect FNAC interpretation of the other case [case 8, [Table 2]] diagnosed as oncocytoma in a young female preoperatively, can also be explained based on the presence of a predominant population of intermediate cells which may be misinterpreted as oncocytes. Both the cases were finally diagnosed as intermediate-grade MEC on HPE. The most plausible explanation for wrong preoperative diagnosis in these two cases comes from the fact that majority of intermediate-grade MECs have abundant population of intermediate cells and their predominance in cytology smears can mislead the pathologist to a diagnosis of oncocytic neoplasms. However, this may be avoided to a large extent by holistic interpretation of clinico-radiologic inputs along with the cytology findings.

Figure 1: Photomicrographs of a case of mucoepidermoid carcinoma showing × 10 (a and b) and × 20 (c) MGG-stained smears with numerous intermediate cells, scattered mucinous cells, and epidermoid cells in a mucinous background. Lower panel (d and e) shows a case of mucoepidermoid carcinoma, diagnosed as acinic cell carcinoma, with acinar-like clusters of cells with abundant finely vacuolated cytoplasm; nuclei exhibit mild-to-moderate pleomorphism; (f) shows histopathology of the case. Note multiple variably sized cystic spaces

Click here to view

PA is the most common benign tumor of salivary glands and as the name suggests, it is notorious for architectural as well as cytologic diversity even in the same tumor. The intermediate cells of MEC can be mistaken for myoepithelial cells of PA. PAs are prone to squamous metaplasia, which may also add to the confusion.^[13] At times, extensive myxoid degeneration in a PA can lead to aspiration of pauci-cellular material, which might be interpreted as epithelial mucin.^[5],[13] Thus, as in one of our cases, MEC and PA can be mistaken for each other on cytology.

Three of the 8 discordant (37.5%) cases were diagnosed as acinic cell carcinoma, one with a remark of high-grade transformation. On final HPE, 2 cases were of high-grade and one of low-grade MEC. On extensive search of literature, we found only one study which also reported discrepant FNAC diagnosis of acinic cell carcinoma in a case of MEC.^[11] Neoplastic cells in acinic cell carcinoma morphologically emulate serous acinar cells. In our cases, cells were seen haphazardly scattered, with occasional acinar formations, in a cystic necrotic background. Most of these cells showed abundant finely vacuolated cytoplasm and were possibly misinterpreted as acinic epithelial cells [Figure 1]. Nuclear features in one of these three cases were of high grade, exhibiting at least moderate degree of pleomorphism, and scattered mitotic figures, prompting a remark of high-grade transformation by the pathologist. This is a reminder of the diagnostic wisdom to refrain from grading of salivary tumors on cytology, unless features are easily evident. On review, these “acinar” cells were, indeed, intermediate cells as well as mucus-producing cells, with many showing histiocyte-like morphology. We strived hard to find the differentiating features missed in the original report. A search for epidermoid cells and periodic acidic–Schiff staining to highlight mucus cells can be done; however, it was in vain in our case. A subtle feature noted was that the nuclear orientation in mucus cells is central as opposed to paracentric in acinar cells. However, the diagnostic dilemma was settled by IHC staining of tissue from all the three cases with a panel of CK7, CK20, C-KIT, DOG1, and MIB1 antibodies. All cases were negative for DOG1 which is a robust marker seen to be positive in most cases of acinic cell carcinoma.^[14] Nuclear proliferation index as highlighted by MIB1 helped in labeling two of these cases as high grade; however, the same is not yet a standard recommendation.

It is well known that squamous cell carcinoma is a favorable differential when epidermoid cells are numerous in the aspirate. However, when mucus cells predominate along with a fair number of intermediate cells, they can mimic a number of cells from histiocytes, metaplastic cells, to oncocytes. From extensive deliberations in our cases, the authors would like to include “acinar cells” to this list with acinic cell carcinoma as an important although uncommon diagnostic pitfall.

Conclusion

MECs of salivary glands can mimic the morphology of a variety of benign as well as malignant lesions on cytology with low cyto-histologic concordance (56% in our study). FNAC interpretation of cystic lesion of salivary glands should always be guarded with a differential for MEC in a correct clinical context. Though rarely described in literature, acinic cell carcinoma should be kept in mind as a diagnostic pitfall and further studies may be required to accurately find out its frequency as a cytologic red herring in cases of MEC.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Rossi ED, Faquin WC, Baloch Z, Barkan GA, Foschini MP, Pusztaszeri M, et al. The Milan system for reporting salivary gland cytopathology: Analysis and suggestions of initial survey: Survey on salivary gland cytopathology. Cancer Cytopathol 2017;125 Suppl 1:757-66.
2.	Vasudevan G, Bishnu A, Singh BM, Singh VK. Mucoepidermoid carcinoma of salivary gland: Limitations and pitfalls on FNA. J Clin Diagn Res 2017;11:R04-6.
3.	Joseph TP, Joseph CP, Jayalakshmy PS, Poothiode U. Diagnostic challenges in cytology of mucoepidermoid carcinoma: Report of 6 cases with histopathological correlation. J Cytol 2015;32:21-4. [PUBMED] [Full text]
4.	Diwakar JK, Agarwal A, Garg C, Giri KY, Dandriyal R, Kumar G. A rare case of mucoepidermoid carcinoma of parotid with mandibular metastasis. Ann Maxillofac Surg 2019;9:205-7. [PUBMED] [Full text]
5.	Laishram RS, Kumar KA, Pukhrambam GD, Laishram S, Debnath K. Pattern of salivary gland tumors in Manipur, India: A 10 year study. South Asian J Cancer 2013;2:250-3. [PUBMED] [Full text]
6.	Subrata P, Mondal S, Bose K, Pradhan R, Bandyapadhyay A, Bhattacharyya D. Fine needle aspiration cytology of minor salivary gland tumors: A retrospective 5-year study of 42 cases in a tertiary care hospital. J Cancer Res Ther 2019;15:686-9.
7.	Son E, Panwar A, Mosher CH, Lydiatt D. Cancers of the major salivary gland. J Oncol Pract 2018;14:99-108.
8.	Wu S, Liu G, Chen R, Guan Y. Role of ultrasound in the assessment of benignity and malignancy of parotid masses. Dentomaxillofac Radiol 2012;41:131-5.
9.	Abdel Razek AA, Mukherji SK. Imaging of minor salivary glands. Neuroimaging Clin N Am 2018;28:295-302.
10.	David D, Clayman L, Saleh H. Value of fine-needle aspiration biopsy in initial evaluation of floor of the mouth masses: Report of a case of low-grade mucoepidermoid carcinoma. Diagn Cytopathol 2010;38:81-4.
11.	Parmar MD, Suri SK, Ghoghari DV. Role of fine needle aspiration cytology in salivary glands lesions. J Res Med Dent Sci 2014;2:23-7.
12.	Salehi S, Maleki Z. Diagnostic challenges and problem cases in salivary gland cytology: A 20-year experience. Cancer Cytopathol 2018;126:101-11.
13.	Kocjan G, Shah KA. Salivary glands. In: Gray W, Kocjan G, editors. Diagnostic Cytopathology. 3^rd ed. New York: Churchill Livingstone; 2010. p. 231-49.
14.	Wade TV, Livolsi VA, Montone KT, Baloch ZW. A cytohistologic correlation of mucoepidermoid carcinoma: Emphasizing the rare oncocytic variant. Patholog Res Int 2011;2011:135796.