){kind=link}
Contributed by Nicole Delaney, MD and Raja Seethala, MDFINAL DIAGNOSIS
Metastatic Acinic Cell Carcinoma with High-Grade Transformation
DISCUSSION
Acinic cell carcinoma (AciCCa) is a low-grade malignant salivary gland neoplasm with serous acinar differentiation (Simpson et al., 2017). First described by Nasse in 1892 as a 'blue dot tumor', at least some of the neoplastic cells in AciCCa demonstrate serous acinar differentiation characterized by cytoplasmic zymogen granules; however, salivary ductal-type cells are also often a component of AciCCa (Vander Poorten et al., 2016; Evenson & Nagao, 2009).
AciCCa represents about 6% of all salivary gland tumors and the majority of cases (80-90%) occur in the parotid (Eveson & Thompson, 2012). It typically presents as a slowly growing parotid or facial mass which may be mobile or fixed. Although AciCCa has a wide age distribution from the 2nd to 7th decades, the mean age at presentation is approximately 50 years old with a female-to-male ratio of 1.5:1 (Simpson et al., 2017).
Complete surgical excision is the treatment of choice for AciCCa. Although recurrence rates as high as 35% have been reported (Simpson et al., 2017), the prognosis for AciCCa is usually good with reported 5- and 20-year survival rates of about 97% and 90%, respectively (Patel et al., 2014). Despite this, AciCCa has malignant potential and can metastasize to other sites, including cervical lymph nodes and lung. Patients exhibiting distant metastasis demonstrate worse long-term prognosis with a 20-year disease-specific survival of approximately 22% (Patel et al., 2014).
Microscopically, acinic cell carcinoma may show apparent encapsulation or circumscription, although extension into normal tissue is common. While by definition AciCCa must demonstrate evidence of serous acinar differentiation, historically this tumor can show a range of cell types and architectural patterns. Specifically, while serous acinar cells must be present, this tumor may also contain intercalated ductal, non-specific glandular, vacuolated, and/or clear cells. While the prototypical architecture of AciCCa is a solid/lobular growth pattern, the tumor may also show microcystic, papillary-cystic, or follicular growth patterns. In most cases, AciCCa shows bland cytonuclear features and is considered a low-to-intermediate grade malignancy. AciCCa often has a surrounding lymphoid infiltrate, sometimes referred to as tumor associated lymphoid proliferation (TALP), which may contain germinal centers and could be mistaken for a lymph node metastasis.
Serous acinar-type cells are large and polygonal with abundant lightly basophilic cytoplasm and round, eccentric nuclei. The basophilic cytoplasm contains PAS(+) diastase resistant granules, usually more pronounced near the luminal surface of the cell. By immunohistochemistry, AciCCa is usually diffusely positive for DOG-1 (Chenevert et al., 2012) and SOX-10 (Ohtomo et al., 2013). While CK7 is often positive in AciCCa, the immunophenotype is somewhat variable depending on the proportion of serous acinar-type cells within the tumor as these tend to be negative for CK7 (Schwarz et al., 2012). Although normal serous acinar type cells are usually positive for amylase by immunohistochemistry, AciCCa is negative for amylase immunostain as well as for p63 (Zhu et al., 2015).
Recent studies using molecular techniques have helped to further characterize AciCCa of salivary gland. First, a new low-grade salivary gland carcinoma known as mammary analogue secretory carcinoma (MASC) was defined by Skalova and colleagues in 2010 on the basis of histologic similarity to breast secretory carcinoma as well as a characteristic ETV6-NTRK3 gene fusion (Skalova et al., 2010). Subsequent study has shown that more than half of salivary gland tumors previously diagnosed as zymogen-granule poor AciCCa are likely to represent MASC (Chiosea et al., 2012). Moreover, a genetic fusion involving the MSANTD3 gene has been recently characterized in a subset of AciCCa of the salivary gland (Andreasen et al., 2019). Given these new developments, it is plausible that further refinements to the entity known as salivary AciCCa may take place in the future.
Although salivary gland neoplasms typically are of low-grade histology, a variety of salivary gland carcinomas can rarely undergo a phenomenon known as high-grade transformation, (Nagao, 2013). Sometimes also referred to as 'dedifferentiation' in older literature, tumors with high-grade transformation are composed of a conventional salivary-type carcinoma juxtaposed with areas of high-grade morphology. Recognition of high-grade transformation is important as it is associated with more aggressive behavior than that of its corresponding low-grade salivary-type carcinoma counterpart with poorer prognosis, higher local recurrence rate, and propensity for locoregional lymph node metastasis (Nagao, 2013).
The two largest published case series of AciCCa with high-grade transformation include a report of 9 cases in 2009 published by Skalova and colleagues (Skalova et al., 2009) followed by a report of 25 cases published in 2016 by Thompson and colleagues (Thompson et al., 2016). In both series, high-grade transformation in AciCCa affected older individuals compared to conventional AciCCa (median age of 61 years of age reported by Skalova; median age of 66 years of age reported by Thompson). All tumors in both series occurred in the parotid. More significantly, both series demonstrated an increased risk of lymph node metastases as well as distant metastases. In the Skalova series, 5 of 9 patients (56%) had lymph node metastases and distant metastases to the lung and pleura were seen in 4 of 9 patients (44%). The Thompson series demonstrated that 10 patients had lymph node metastases; however, this parameter was not available for all cases due to consultation-type cases included in the study.
Most importantly, both the Skalova and Thompson series demonstrated that AciCCa with high-grade transformation has a worse prognosis than conventional AciCCa. In the Skalova series, the median overall survival was 4.3 years (range 1-9 years) in a patient population with AciCCa with high-grade transformation who underwent treatment with surgery and post-resection radiation therapy. Similarly, despite aggressive therapy, 86% of patients with AciCCa with high-grade transformation in the Thompson series had recurrent disease with an overall median survival of 2.2 years and an overall mean survival of 3.3 years. In comparison, a 2013 study from MD Andersen Cancer Center demonstrated a median survival of 28.5 years in patients with conventional AciCCa (Neskey et al., 2013). Based on this literature, the presence of high-grade transformation in AciCCa is associated with worse patient outcomes.
Given this important implication for patient outcome, it is important for pathologists to recognize the presence of high-grade transformation in AciCCa. In the Thompson series, the high-grade transformed component in 24 of the 25 cases consisted predominantly of invasive lobules of undifferentiated cells with areas of acinic differentiation while the low-grade AciCCa component was often microcystic in architecture. As four of the cases were biopsy samples only in the Thompson series, certain histologic features could not be reliably assessed. In a majority of cases, the high-grade transformed component of AciCCa demonstrated intermediate to high mitotic activity with atypical mitoses (21/22) and central comedonecrosis (19/22).
In the Skalova series, all 9 cases were composed of two distinct carcinomatous components, a low-grade conventional AciCCa and a high-grade transformed component. The high-grade component appeared as a distinct population of anaplastic cells with abundant cytoplasm and large polymorphic nuclei arranged in irregularly-shaped tumor islands with central comedonecrosis. Skalova reports a spectrum of histologic growth patterns in the high-grade component, including solid undifferentiated carcinoma (5/9 cases), high-grade adenocarcinoma not otherwise specified (2/9 cases), and overlapping solid and glandular features (1/9 cases). Representative images of a case of AciCCa with high-grade transformation from our institution with typical findings described in both the Skalova and Thompson series is provided for reference (see Figure 12).
Both the Skalova and Thompson series reported that low-molecular weight cytokeratin expression was retained in both the low-grade and high-grade transformed components of the AciCCa. In addition, both series demonstrated increased expression of Ki-67/MIB-1 in the high-grade transformed areas of the AciCCa. Mutant p53 immunophenotype was also identified in the subset of the high-grade transformed areas of AciCCa in both series. In the Skalova series, there was overexpression of p53 in 2 of 9 cases whereas the Thompson series demonstrated that about 40% of cases had a mutant p53 immunophenotype (extreme p53 immunopositivity in 33% of cases and loss of p53 immunostaining in 7% of cases).
In our patient, similar to the cases reported by Skalova and Thompson, the tumor consisted of a low-grade AciCCa component as well as a high-grade transformed component which were intimately associated with one another. Unlike the cases previously reported in the literature, the high-grade transformed component of the AciCCa in our case demonstrated an anaplastic and sarcomatoid morphology. Focal positive staining for low-molecular weight cytokeratins and DOG-1 in the high-grade area also supports the diagnosis of AciCCa with high-grade transformation. Furthermore, the low-grade AciCCa area in this temporal bone lesion demonstrates similar morphology to the patient's earlier conventional AciCCa. Our case also demonstrates extreme positive staining for a p53 immunostaining in the high-grade transformed areas.
In addition to metastatic acinic cell carcinoma with high-grade transformation, the differential diagnosis in our case also included osteosarcoma, particularly given this patient's history of previous radiation therapy. For this reason, we also performed immunohistochemistry for Special AT-Rich Sequence Binding Protein-2 (SATB2), a nuclear protein with a critical role in osteoblast lineage commitment. A study by Conner & Hornick in 2013 demonstrated that SATB2 is a marker of osteoblastic differentiation in benign and malignant mesenchymal tumors, including osteosarcoma (Conner & Hornick, 2013). While SATB2 is a sensitive marker for osteosarcoma, it is not necessarily specific for osteosarcoma compared to other high-grade bone sarcomas (Davis & Horvai, 2016).
In our case, the high-grade transformed component was positive for SATB2 expression (Figure 6). Although we considered an osteosarcoma in the differential diagnosis of our original frozen section, no convincing areas of osteoid were identified microscopically at permanent section, the imaging features were not those of an osteosarcoma, and the focal expression of low-molecular weight cytokeratin also argued against this diagnostic consideration.
Finally, the clinical presentation of our patient was also in keeping with AciCCa with high-grade transformation. Furthermore, at the time of resection of the temporal bone lesion, our patient also had radiologic and pathologic evidence of locoregional and metastatic AciCCa. Unfortunately, despite adjuvant radiotherapy after resection of this temporal bone lesion, our patient experienced locoregional recurrence and died of disease less than one year after the diagnosis of metastatic acinic cell carcinoma with high-grade transformation.
REFERENCES
Contributed by Nicole Delaney, MD and Raja Seethala, MD
  |
