Noninvasive carcinoma ex pleomorphic adenoma of the parotid gland: A difficult diagnosis on fine needle aspiration

INTRODUCTION

Pleomorphic adenoma (PA) is the most common benign parotid gland neoplasm comprising 50% of all parotid tumors. Primary parotid carcinomas, on the other hand, are rare accounting for 1–3% of all head and neck malignancies. Carcinoma expleomorphic adenoma (CXPA) is a rare epithelial malignancy that arises from a primary or recurrent PA. Depending on the extent of invasion of the carcinomatous component, CXPA can be divided into noninvasive (NI) or invasive.[1234]

Most often, the carcinomatous component of CXPA is adenocarcinoma not otherwise specified (NOS). But any other histological sub-types of carcinomas may be seen with similar findings on fine needle aspiration (FNA).[125]

The specificity and diagnostic accuracy of salivary gland FNA are high (86–98%) and the sensitivity varies in the range from 64% to 94%.[678] However, the sensitivity in detecting CXPA on FNA is lower and has been reported to be as low as 29%.[9] The low sensitivity is mostly attributed to sampling error and may subsequently prevent the appropriate surgical planning.

Literature on FNA findings of CXPA including NI-CXPA of the parotid gland is scarce. Herein, we present a case of NI-CXPA, which was interpreted as high-grade adenocarcinoma, NOS on FNA.

CASE REPORT

Cytological findings

The FNA was performed using a 25-gauge needle. Four air-dried Diff-Quik direct smears were examined. The cellular specimen showed a high-grade tumor arranged in cohesive and loosely-cohesive clusters, papillary-like groups and as single cells. Small fragments of fibromyxoid matrix, some of which were closely associated with tumor cells, were present in the background. The matrix was densely homogeneous with focal vague fibrillary quality. The tumor cells were pleomorphic, round, polygonal to oval and displayed a high nuclear-to-cytoplasmic ratio. The nuclei were large, pleomorphic, with relatively smooth membranes, irregularly-clumped chromatin and small to inconspicuous nucleoli. The cytoplasm was scant-moderate, ill-defined and had a finely granular texture. Mitoses or necrosis were not identified. The cytological findings were not characteristic of any specific parotid gland malignancy, and a diagnosis of primary high-grade adenocarcinoma was rendered [Figure 1a–d].

Close

Figure 1

(a) Group of cohesive and pleomorphic tumor cells associated with a small amount of metachromatic fibromyxoid stroma with a vague fibrillary quality. This was a focal finding (Diff-Quik [DQ] stain, ×100). (b) A cluster of cohesive and three-dimensional pleomorphic tumor cells showing small fragments of homogeneous metachromatic fibromyxoid stroma in the background. The nuclei are enlarged, hyperchromatic and angulated. Nucleoli are not evident (DQ stain, ×100). (c) Loosely-cohesive cluster of tumor cells with large, somewhat irregular nuclei with occasional nucleoli (top of the group). Cytoplasm is scant-moderate and ill-defined with a finely granular texture (DQ stain, ×100). (d) Adenocarcinoma cells on a ThinPrep (TP) display coarse irregularly clumped chromatin, occasional nucleoli and dense to granular cytoplasm (TP, Pap stain, ×100)

Export to PPT

The patient underwent a facial nerve-sparing right total parotidectomy and a right selective neck dissection.

Gross and histological findings

Grossly, a 2.2 cm well-demarcated tumor was seen. The cut-surface was firm, yellow-white, with minute areas of hemorrhage [Figure 2a]. Initial histologic evaluation, at low magnification, demonstrated a PA with extensive hyalinization and an intact fibrous capsule [Figure 2b]. Careful evaluation showed that the PA was interspersed with ductal structures and single cells with cellular anaplasia, exhibiting pleomorphic nuclei and eosinophilic cytoplasm. Mitotic figures were identified, and apoptic cells were present [Figure 2c]. No extracapsular invasion was present. Immunohistochemical stain for Ki-67 showed ~ 10% proliferation rate in the malignant ducts. The benign PA ducts were negative for Ki-67. A p63 stain showed focal staining in some of the ductal structures indicating an intact myoepithelial layer and the presence of an in-situ component. The histologic diagnosis was NI-CXPA with the carcinomatous component a high-grade adenocarcinoma, NOS. Thirteen level II lymph nodes and one external jugular lymph node were negative for carcinoma.

Close

Figure 2

(a) The resection specimen showed a well-demarcated tumor with a yellow-white cut-surface and minute areas of hemorrhage. (b) Low magnification of carcinoma ex pleomorphic adenoma (PA) admixed with PA. Note the uninvolved intact capsule of PA (H and E, ×4). (c) High magnification reveals highly atypical ductal structures and single cells with cellular anaplasia, exhibiting pleomorphic nuclei and eosinophilic cytoplasm (H and E, ×40)

Export to PPT

DISCUSSION

Malignancies in a PA are uncommon. CXPA is a rare epithelial malignancy that arises from a primary or recurrent PA. Overall, CXPA comprises 3.6% of all salivary gland neoplasms and approximately 12% of all salivary gland malignancies. It usually occurs in the sixth decade, is slightly more common in women and is mostly asymptomatic. The classification and histological details of all types of CXPA have been previously published.[125]

Briefly, there are three main categories of CXPA based on the degree of invasion of the tumor: (1) NI-CXPA in which the carcinoma is confined within the PA either as an intraductal/in-situ component (surrounded by an intact layer of myoepithelial cells) or intracapsular malignant cells extending beyond the ducts but still confined within the PA (2) Minimally invasive CXPA (<1.5 mm invasion beyond the capsule) (3) Invasive CXPA (>1.5 mm invasion beyond the capsule). Prognosis of NI-CXPA is good and similar to that of PA, while invasive CXPA is usually a high-grade carcinoma with a poor prognosis.[25]

Molecular studies have shown that malignant transformation of PA follows a multi-step model of carcinogenesis and specific genes including mutation or loss of p53, deregulation of p16 and overexpression of PLAG1 and Her2 are associated with the progression of CXPA.[10] The role of radiation in the pathogenesis of CXPA, as in our patient, remains speculative.

Preoperative diagnosis of a parotid gland neoplasm is based on history, clinical findings, imaging and FNA findings. With the high specificity and diagnostic accuracy of salivary gland FNA (86–98%) and a sensitivity ranging from 64% to 94%, a FNA diagnosis of PA is reliable and has a high positive predictive value.[678] However, CXPA can be mistaken for a PA or other benign/malignant salivary gland tumors. To avoid diagnostic pitfalls, all three elements of PA that is three-dimensional cohesive clusters of ductal cells, background of myoepithelial cells and dense fibrillary metachromatic matrix with partially attenuated entrapped myoepithelial cells should be noted.[11] Although occasional cytological atypia may be encountered in some PA, the presence of numerous atypical cells, an abnormal chromatin pattern, and necrosis are features important in distinguishing PA from malignant tumors including CXPA.[3712] Utilizing a combination of clinical and radiologic features can help in making a preoperative diagnosis. A long standing mass with a recent growth spurt as seen in our patient should raise the suspicion for CXPA. In addition to being mistaken for a benign neoplasm, CXPA may be misdiagnosed as other malignancies including salivary duct carcinoma and metastasis. Metastasis to the parotid gland can be distinguished from CXPA by review of the primary tumor and use of an appropriate immunohistochemical panel.

The preoperative diagnosis of CXPA by FNA is difficult with low sensitivity and accuracy. A few studies have examined the cytologic features of CXPA. In the series reported by Zbären et al.,[4] FNA was performed on 16 CXPA and compared to FNA in 147 PA. For CXPA cases, the FNA diagnosis was a true-positive in seven (44%) and false-negative in nine cases (PA in seven, benign salivary gland tissue in one, nondiagnostic in one case). The sensitivity of FNA for diagnosing CXPA was only 47%, much lower than the 91% sensitivity for PA. Nouraei et al.[9] identified malignancy in 4 of 14 (29%) cases, and Klijanienko et al.[3] in 13 of 26 (50%) cases of CXPA on FNA. The authors in these studies contribute the false-negative diagnoses most likely to sampling errors as extent of the carcinomatous component in CXPA can vary considerably. Zbaren reported only 8 CXPAs with a greater than 66% carcinomatous component. Klijanienko noted that the diagnosis of malignancy was more likely to be rendered if the carcinomatous component was high grade and composed a significant proportion of the PA. In our case, the carcinomatous component made up a significant proportion of the tumor (>50%) and a diagnosis of high-grade adenocarcinoma NOS was made. However, the stromal component of the underlying PA on FNA was scant and not recognized as PA.

The low sensitivity of FNA may hamper appropriate planning of the extent of surgery. Although, the diagnosis of a NI-CXPA would require less aggressive surgery, cytologically, it is difficult to almost impossible to distinguish between NI-CXPA and invasive CXPA. As this case demonstrates, while the diagnosis of malignancy can be made, the infiltrative nature of the malignancy may not be possible to discern.

CONCLUSION/SUMMARY

Preoperative diagnosis of NI-CXPA on FNA is difficult and poses a diagnostic challenge to clinicians and pathologists. High-grade salivary adenocarcinoma, NOS should raise the possibility of CXPA especially in the clinical context of recent growth in a long standing mass. Accurate diagnosis is important for proper surgical management.

COMPETING INTERESTS STATEMENT BY ALL AUTHORS

The authors declare that they have no competing interests.

AUTHORSHIP STATEMENT BY ALL AUTHORS

All authors of this article declare that we qualify for authorship as defined by ICMJE http://www.icmje.org/#author.

Each author has participated sufficiently in the work and takes public responsibility for appropriate portions of the content of this article.

TS and RH helped to draft the manuscript and provided the images. WK, RJ, ST, and MR helped to draft the manuscript. All authors read and approved the final manuscript.

ETHICS STATEMENT BY ALL AUTHORS

As this is a case report without identifiers, our institution does not require approval from Institutional Review Board (IRB) (or its equivalent).

LIST OF ABBREVIATIONS

CXPA = Carcinoma Expleomorphic Adenoma

DQ = Diff-Quik

FNA = Fine Needle Aspiration

NI = Noninvasive

NOS = Not Otherwise Specified

PA = Pleomorphic Adenoma

TP = ThinPrep

EDITORIAL/PEER-REVIEW STATEMENT

To ensure the integrity and highest quality of CytoJournal publications, the review process of this manuscript was conducted under a double-blind model (authors are blinded for reviewers and vice versa) through automatic online system.