Context.—

Nonsalivary primary adenocarcinomas of the base of the tongue (PABOTs) are extremely rare and worth reporting.

Objective.—

To study the detailed clinicopathologic features of PABOT.

Design.—

Cases of PABOT diagnosed on pathology material were retrieved from the archived electronic surgical pathology records.

Results.—

Six cases in 4 men and 2 women (M:F ratio, 2:1), with an age range of 31 to 76 years, satisfied the criteria. The tumor epicenter was the base of the tongue in all (6 of 6; 100%), with extension to the epiglottis in 50% (3 of 6), nodal metastasis in 66.7% (4 of 6), and distant metastasis in 33.3% (2 of 6). On histology, all but one were pure adenocarcinoma. Five of 6 cases (83.3%) had a gastrointestinal (GI) phenotype, of which 2 (40%) had a colonic/lower-GI–type (small groups of cells floating in mucin, CK20+, SATB2+, and CDX2+) and 3 (60%) had an upper-GI–like adenocarcinoma (UGI-LA; malignant glands with intracellular mucin, CK7+) histology. Cystic structure suggestive of teratomatous origin was identified in 2 of 5 cases (40%), both with UGI-LA phenotype. The non–GI-type case had a unique histology with squamous differentiation in addition to adenocarcinoma areas, diffuse nuclear β-catenin on immunohistochemistry, and a corresponding exon 3 CTNNB1 mutation. One patient succumbed to disease, and 4 are alive with disease (follow-up of 1–9 months after completion of therapy).

Conclusions.—

We suggest using the broad term primary adenocarcinomas of the base of tongue (PABOTs), which can be further subdivided into colonic-type adenocarcinoma of the tongue and oral cavity, UGI-LA, and not otherwise specified categories, and reiterate a need for recognition and distinction of PABOT from salivary gland tumors. A subset originates from teratoid/duplication cysts, necessitating extensive sampling. Multicentric studies are essential to clinically and biologically prognosticate each of these categories.

Intraoral and oropharyngeal (including tongue base) adenocarcinomas are rare neoplasms, constituting less than 1% of all head and neck tumors.1  Three main types of adenocarcinomas exist in oral and oropharyngeal regions. The first type and the majority of the reported cases are classified as being from the minor salivary gland tissue,1  that is, salivary-type adenocarcinomas, with palate being the most common site, followed by tongue.2,3  The second type is metastatic adenocarcinomas (MAs) to the oral cavity/oropharynx, which constitute approximately 1% of the total. MAs predominantly present as intraoral masses, with the tongue being a site because of its rich vascularity.4  The sites of origin of the primary tumor are lung, endometrium, adrenal, kidney, thyroid, and colon, among others.4  The third type of adenocarcinoma is classified as primary intestinal/colonic-type adenocarcinoma of the tongue and oral cavity (CATOC, also known as intestinal-type and non–intestinal-type adenocarcinomas, akin to their sinonasal counterparts). They are exceptionally rare, with only 10 cases reported in the English-language literature to date.514  Two of the previously reported cases occurred in the base of tongue (BOT) region.5  Although the name CATOC entails oral cavity, it does not preclude the possible occurrence of these tumors in the oropharynx. Their rarity, coupled with an abstruse origin, their unique location and histology, and their difficult differential from metastatic tumors and salivary gland–type adenocarcinomas, makes these tumors worth reporting. We hereby report a series of 6 cases of primary, nonsalivary, nonmetastatic adenocarcinoma of the BOT (PABOT), with insights related to the histology, possible origin, and nomenclature of these exceedingly rare tumors. This series is an attempt to understand, in detail, the histologic spectrum of these tumors, after carefully ruling out salivary gland tumors and metastases to the oral/oropharyngeal region.

All patients received the standard of care for their condition, which was as per ethical standards. No additional procedures were performed on the participants as a part of this study. Patient identifiers are not included in this manuscript.

Clinical Details and Histologic Material

Six cases of primary, nonsalivary adenocarcinomas of the BOT region, diagnosed on pathology material, were retrieved from the archived electronic surgical pathology records (January 2010–December 2020) using search engine portal Sengine and Synoptic Search Engine. Search words used included adenocarcinoma, colonic type adenocarcinoma, and adenocarcinoma, not otherwise specified (NOS). Routine paraffin-embedded sections were cut at 2 to 4 μm and stained with hematoxylin-eosin followed by relevant immunohistochemical (IHC) markers. IHC stains, which included CK7, CK20, CDX2, p63, androgen receptor (AR), TTF1, PAX8, p16, S100-p, and SATB2, were performed on 3-μm sections cut from paraffin blocks using a fully automated system (Benchmark XT System, Ventana Medical Systems Inc, Tucson, Arizona). The histomorphologic and IHC findings for all cases were reviewed by 2 pathologists. The clinical information, treatment details, and follow-up information were obtained from the electronic medical records and are summarized in Table 1.

Table 1

Clinical and Demographic Profile of the Patients Included in the Study

Clinical and Demographic Profile of the Patients Included in the Study
Clinical and Demographic Profile of the Patients Included in the Study

The possibility of metastases was carefully excluded with the help of available clinical information and examination, imaging, and endoscopies. Further, histologic findings and a broad panel of relevant IHC markers were directed toward distinction of MAs (TTF-1, GATA-3, PAX-8) and salivary-type adenocarcinomas from PABOT. The panel of IHC markers (CerbB2, AR, and S100-P) was sufficiently inclusive of markers of common salivary gland tumors at this site, namely polymorphous adenocarcinoma (PAC), cribriform adenocarcinoma of tongue and minor salivary glands (CATMSG), and salivary duct carcinoma. The possibility of papillary thyroid carcinoma (PTC) arising in a thyroglossal cyst and presenting as a BOT mass was also excluded based on histomorphology and a relevant panel of IHC markers (TTF-1, Pax-8, and thyroglobulin [TG]). p16 (E6H4 clone) IHC was performed on all cases to exclude human papillomavirus (HPV)–associated tumors. Only those cases that remained after excluding all the above-mentioned relevant differential diagnoses were included. In addition, all the cases were a minimum of 6 months past the date of histologic reporting to allow sufficient time for the identification of an occult primary. Thus, the follow-ups in this study were calculated 6 months past the date of histologic diagnosis until the date of death or the last visit to the hospital.

Next-Generation Sequencing

Because of resource constraints, next-generation sequencing (NGS) was performed only on case 6 with a targeted solid tumor panel. The genes included in this panel are detailed in the Supplemental Table (see the supplemental digital content at https://meridian.allenpress.com/aplm in the November 2023 table of contents). The DNA extraction was done from paraffin-embedded tissue, using 3 or 4 scrolls cut at 10 μm. Sequencing libraries were prepared using 20 to 100 ng total DNA followed by paired-end sequencing by synthesis technology followed by bioinformatic analysis.

Clinical Features

The patients’ characteristics are summarized in Table 1. The mean age of the patients was 57.5 years, with the youngest being 31 years of age (cases 3 and 6) and the eldest 79 years (case 1). There were 4 men and 2 women (M:F ratio, 2:1). The most common clinical presentation was with symptoms related to a mass lesion (difficulty in swallowing, “lump in the throat”), except for case 3, where the patient presented with hoarseness of voice. The duration of complaints ranged from 1 month to 2 years. All the cases had BOT (6 cases; 100%) as the epicenter of the tumor, with the tumor extending to the lingual surface of the epiglottis in 3 cases (50%; cases 2, 3, and 4). Four patients (66.7%; cases 1, 2, 3, and 5) had lymph node metastasis, and 2 (33.3%; cases 1 and 3) had distant metastasis at presentation. Case 1, on ultrasonography of the neck as a part of the initial workup of the patient, was found to have 2 large solid-cystic nodules with a well-defined peripheral halo in the left lobe of thyroid, which were confirmed to be MA on fine-needle aspiration cytology. Case 3 had metastatic skin and lung nodules. All but one of the patients were treated with a combination of surgery and at least one nonsurgical modality. Case 4 was treated with radiotherapy alone.

Follow-up

Follow-up information was available in 5 of 6 patients and was calculated after 6 months from the date of diagnosis. Of these 5 patients, 1 patient succumbed to her illness (case 1), 3 were alive with disease with follow-ups of 1, 5, and 9 months, and 1 patient had responded to therapy with no evidence of disease after 1 month of therapy completion.

Pathologic Features

The salient morphologic and IHC features of all the cases in this study are detailed in Table 2. A total of 7 histologic specimens (3 biopsies and 4 resections) from 6 cases were received. The mean size of the tumors was 3.7 cm, with a range of 2 to 5.2 cm. All the tumors occurred in the BOT with or without involvement of the vallecular/lingual surface of epiglottis. All but one (case 6) were pure adenocarcinomas; case 6 showed an admixture of adenocarcinoma and squamous areas. We found a rather heterogeneous group of tumors with varied morphologic and IHC features. Based on the sinonasal adenocarcinoma classification, 5 cases (cases 1, 2, 3, 4 and 5) could be labeled as intestinal/gastrointestinal (GI)–type and 1 case (case 6) as non–intestinal-type adenocarcinoma. Of the GI-type tumors, using the terminology proposed by Bell et al,5  only 2 (cases 3 and 4) could be labeled as colonic-type adenocarcinoma (CATOC) based both on morphology (small groups of cells floating in extracellular pools of mucin, along with signet ring cells) and the lower-GI–type adenocarcinoma (LGI-A) IHC profile (CK20+, SATB2+). As in the present study, both cases reported by Bell et al5  were BOT (oropharyngeal) tumors. Only one of our CATOC cases (case 3) underwent surgical excision of the mass lesion, which, even after extensive sampling, did not reveal any cystic structure/benign-appearing epithelium. Interestingly, this case showed 2 distinct areas on morphology. One was composed of mucinous adenocarcinoma phenotype with abundant extracellular mucin within which were small groups and individual cells, some with signet ring histology. The other area comprised a moderately differentiated colorectal-type adenocarcinoma, which had tumor cells in glands with hyperchromatic elongated columnar cells and was practically devoid of mucin. There was an abrupt transition between the 2 areas; both phenotypes were seen in the nodal metastasis. Only biopsy slides were available for the other case, which showed mucinous adenocarcinoma areas consistent with CATOC. Of the other 3 cases with GI phenotype, case 1 had a morphology different from the other 2 by virtue of its striking resemblance to gastric foveolar-type lining mucosa. The entire tumor appeared to be cystic, with the cyst wall lined by areas showing benign, low cuboidal lining mucosa with varying degrees of proliferation; the proliferative component resembled gastric foveolar epithelium with mild to moderate atypia, thus mimicking an intrapancreatic mucinous neoplasm with gastric-type epithelium. Focal infiltration of the cyst wall with invasion into the stroma was noted. The tumor was also seen involving and ulcerating the surface in a manner that appeared to be the surface opening of a cyst/duct. On IHC, the tumor was positive for CK7 and CDX2 and negative for CK20 and SATB2, confirming the upper-GI–like adenocarcinoma (UGI-LA) profile (Figure 1, A through H). The other 2 cases had a moderately differentiated adenocarcinoma histology with glandular, cribriform, and papillary (case 5) areas and uniform high-grade nuclear atypia, associated with intracellular and focal extracellular mucin. One of these cases showed a dilated cystic structure with stratified ciliated columnar lining epithelium, with the epithelium undergoing a malignant transformation evidenced by complexity of architecture, increasing degree of cellular pleomorphism, and atypia. The wall of the cyst was markedly thickened and fibrotic and showed the infiltrative component of the tumor, with invasion of the surrounding salivary tissue and the overlying epithelium. IHC on both these cases showed UGI-LA phenotype with strong positivity for CK7; CK20 and SATB2 were negative. CDX2 was positive in case 2 and negative in case 5. Other relevant IHC markers such as TTF1, GATA3, PAX8, S100-P, p16, and p63 were negative in all cases (Figure 2, A through L).

Table 2

Salient Histopathologic Features and Immunoprofile of Primary Adenocarcinomas of the Base of Tongue

Salient Histopathologic Features and Immunoprofile of Primary Adenocarcinomas of the Base of Tongue
Salient Histopathologic Features and Immunoprofile of Primary Adenocarcinomas of the Base of Tongue
Figure 1

Primary intestinal adenocarcinoma of the base of the tongue, lower gastrointestinal (GI)/colonic type (case 3). Total glossectomy specimen (A) showing a submucosal glistening tumor infiltrating the underlying skeletal muscle. Low-power photomicrograph (B) with both the mucin-poor and the mucin-rich areas in this tumor, followed by high-power photomicrograph of both (C). The mucin-rich areas abounded in signet ring cells (D). Immunohistochemical studies showed positivity for CK20 (F), CDX2 (G), and SATB2 (H) and negativity for CK7 (E), confirming the lower-GI phenotype (hematoxylin-eosin, original magnifications ×40 [B] and ×400 [C and D]; DAB, original magnifications ×200 [E] and ×400 [F through H]).

Figure 1

Primary intestinal adenocarcinoma of the base of the tongue, lower gastrointestinal (GI)/colonic type (case 3). Total glossectomy specimen (A) showing a submucosal glistening tumor infiltrating the underlying skeletal muscle. Low-power photomicrograph (B) with both the mucin-poor and the mucin-rich areas in this tumor, followed by high-power photomicrograph of both (C). The mucin-rich areas abounded in signet ring cells (D). Immunohistochemical studies showed positivity for CK20 (F), CDX2 (G), and SATB2 (H) and negativity for CK7 (E), confirming the lower-GI phenotype (hematoxylin-eosin, original magnifications ×40 [B] and ×400 [C and D]; DAB, original magnifications ×200 [E] and ×400 [F through H]).

Close modal
Figure 2

Primary intestinal adenocarcinoma of the base of the tongue, upper-gastrointestinal (GI) type. Case 1: representative images of the gastro-pancreatic–type histology. A submucosal tumor (A) with a large cystic component (B) focally lined by a pseudostratified ciliated lining epithelium (C) with an increasing degree of architectural complexity (D) culminates into an invasive tumor. The other case of upper-GI type (case 5) shows a cyst lined by low columnar to cuboidal epithelium with foci of epithelial proliferation (E and F) leading to an infiltrative tumor, a mucin-secreting adenocarcinoma with glandular (G) and papillary architecture (H). Immunohistochemical evaluation revealed that CK7 was diffusely positive in the cyst (I) as well as the tumor (J), whereas they were negative for CK20 (K) and SATB2 (L) (hematoxylin-eosin, original magnifications ×40 [A], ×100 [B], ×200 [D through H], and ×400 [C]; DAB, original magnifications ×40 [I] and ×200 [J through L]).

Figure 2

Primary intestinal adenocarcinoma of the base of the tongue, upper-gastrointestinal (GI) type. Case 1: representative images of the gastro-pancreatic–type histology. A submucosal tumor (A) with a large cystic component (B) focally lined by a pseudostratified ciliated lining epithelium (C) with an increasing degree of architectural complexity (D) culminates into an invasive tumor. The other case of upper-GI type (case 5) shows a cyst lined by low columnar to cuboidal epithelium with foci of epithelial proliferation (E and F) leading to an infiltrative tumor, a mucin-secreting adenocarcinoma with glandular (G) and papillary architecture (H). Immunohistochemical evaluation revealed that CK7 was diffusely positive in the cyst (I) as well as the tumor (J), whereas they were negative for CK20 (K) and SATB2 (L) (hematoxylin-eosin, original magnifications ×40 [A], ×100 [B], ×200 [D through H], and ×400 [C]; DAB, original magnifications ×40 [I] and ×200 [J through L]).

Close modal

Case 6 had disparate histology, with adenocarcinoma areas representing 70% of the tumor. Though the adenocarcinoma areas in this tumor focally displayed elongated stratified columnar cells reminiscent of colorectal adenocarcinoma, the tumor cells had optically clear nuclei, with foci of psammomatous calcification and squamous morules. This led to PTC, columnar variant, in a lingual thyroid being considered as the probable diagnosis on hematoxylin-eosin evaluation; however, the TTF1 and PAX8 immunostains came back negative. In addition, further sampling of this tumor showed a central cystic area lined by squamous lining with an increasing degree of dysplasia, finally culminating into nonkeratinizing squamous carcinoma areas. Though an admixture of both adenocarcinoma and squamous carcinoma areas was present in this tumor, the morphology was unlike the other adenosquamous carcinomas of the head and neck region reported in the literature, which necessitate that the tumor areas be distinct and not mixed. PAC and cribriform adenocarcinoma of the tongue were ruled out because of their morphologic dissimilarities with this case and also by using the appropriate panel of IHC markers (negative for S100-P and SOX10 and focally positive for CK7). p40 showed a unique pattern of staining with diffuse positivity in squamous areas, complete absence in adenocarcinoma areas, and focal p40 positivity in areas of collision. This case showed diffuse strong nuclear staining for β-catenin; however, the histomorphology and IHC markers were unlike those of basal cell adenoma/adenocarcinoma of the salivary gland (Figure 3, A through H). Thus, this was labeled as a carcinoma with both adeno and squamous differentiation, and molecular analysis was advised because of the unusually rare morphology. NGS was performed using a targeted solid tumor panel, SOPHiA Solid Tumor Plus Solution (Sophia Genetics, Saint-Sulpice, Switzerland), which revealed a missense mutation in exon 3 of the CTNNB1 gene. No other abnormality was detected in any of the genes studied (Supplemental Table).

Figure 3

Non–intestinal-type primary adenocarcinomas of the base of tongue (case 6). Partial glossectomy specimen (A) with a gray-white submucosal tumor with a central cavity (arrow) with brown hemorrhagic walls. On microscopy, the tumor showed squamous (left half) and nonsquamous (right half) areas (B). High-power view of the area on the right shows an adenocarcinoma with extensive psammomatous calcification (C) and nuclear clearing (D) suggestive of papillary thyroid carcinoma. Squamous morules and papillary architecture were also easily identifiable (not shown here). On immunohistochemistry, this tumor showed variable positivity for CK7 (E), with the lower third of the picture showing diffusely positive foci, the middle third with negative staining, and the upper half with focal staining. A similar pattern was seen with p40 (F), where the squamous areas (right side) showed diffuse staining, the adenocarcinoma areas (left lower part) showed complete absence of staining, and the collision areas showed focal p40 positivity. A diffuse positivity for β-catenin (G) was seen in all the distinct areas in the tumor. S100-P (H) was negative (hematoxylin-eosin, original magnifications ×40 [B] and ×400 [C and D]; DAB, original magnifications ×40 [E], ×200 [F and H], and ×400 [G]).

Figure 3

Non–intestinal-type primary adenocarcinomas of the base of tongue (case 6). Partial glossectomy specimen (A) with a gray-white submucosal tumor with a central cavity (arrow) with brown hemorrhagic walls. On microscopy, the tumor showed squamous (left half) and nonsquamous (right half) areas (B). High-power view of the area on the right shows an adenocarcinoma with extensive psammomatous calcification (C) and nuclear clearing (D) suggestive of papillary thyroid carcinoma. Squamous morules and papillary architecture were also easily identifiable (not shown here). On immunohistochemistry, this tumor showed variable positivity for CK7 (E), with the lower third of the picture showing diffusely positive foci, the middle third with negative staining, and the upper half with focal staining. A similar pattern was seen with p40 (F), where the squamous areas (right side) showed diffuse staining, the adenocarcinoma areas (left lower part) showed complete absence of staining, and the collision areas showed focal p40 positivity. A diffuse positivity for β-catenin (G) was seen in all the distinct areas in the tumor. S100-P (H) was negative (hematoxylin-eosin, original magnifications ×40 [B] and ×400 [C and D]; DAB, original magnifications ×40 [E], ×200 [F and H], and ×400 [G]).

Close modal

Intraoral and oropharyngeal adenocarcinomas (OOPAs) are rare, with the currently accepted nomenclature being adenocarcinoma, NOS, and the proposed origin being minor salivary glands according to the 4th edition of the World Health Organization (WHO) classification of head and neck tumors.13,15  This was a change from the previous 2005 edition, which subclassified mucinous adenocarcinoma (akin to colloid carcinoma) as a stand-alone entity and carcinoma with papillary and mucinous architecture as a component of cystadenocarcinoma.16,17  Thus, a seemingly heterogeneous group of tumors were all grouped into a very broad histologic category that is defined by neither morphology nor IHC nor molecular underpinnings. Even mucin—intracellular or extracellular, focal or extensive—was considered as a nonspecific finding.

Though most of the series on minor salivary gland tumors of the head and neck region mention them, the details of OOPA are unelaborated. Pires et al2  studied 542 intraoral minor salivary gland tumors, of which 21 (3.9%) were NOS and 3 were adenosquamous carcinoma, indicating the rarity of salivary-type adenocarcinoma, NOS category. However, there is no further qualification on the histologic and IHC findings of “adenocarcinoma, NOS” to understand either their histology or IHC expression. The largest study involving minor salivary gland tumors of the oropharynx lists adenocarcinoma as the second most common subtype (25.9% of cases) of minor salivary gland tumors, which is contrary to the above study. However, the details of the methodology reveal that this category of “adenocarcinoma” includes acinic cell carcinoma; polymorphous low-grade adenocarcinoma (PAC); adenocarcinoma, NOS; and other rare adenocarcinomas.3  Histologic details of the latter are not available, and the International Classification of Diseases for Oncology codes used to search for this category of tumors in the SEER database are again nonspecific and belong under the umbrella of “rare epithelial cell tumors of salivary gland origin.”3  Thus, it is difficult to gauge the exact morphologic and molecular entities and their distribution in this category; the reasons for classifying these tumors as of salivary origin, whether or not these tumors have a counterpart in the major salivary glands; and the reasons for their propensity to occur in the BOT region despite the widespread occurrence of minor salivary gland tissues in the head and neck regions. This seems to be the case for most of the large case series of minor salivary gland tumors of the oral/oropharyngeal region, where limitation of space and word limits preclude a detailed account of this rare category. Moreover, this was practically deemed unnecessary, as all OOPAs are currently classified by the 4th edition of the WHO classification of head and neck tumors as adenocarcinoma, NOS, of salivary origin.15 

The focus of this study is a group of unusual and rare adenocarcinomas that preferentially, rather almost exclusively, occur in the BOT region, and presently exist in the literature in the form of a few previously published case reports. These cases have been alluded to using a myriad of names, such as intestinal adenocarcinoma, sinonasal type; primary intestinal/colonic type adenocarcinoma (CATOC); mucinous adenocarcinoma; and signet-ring cell carcinoma; however, there are only 10 reported cases in the literature to date (Table 3).1,512  One of these was a case of a mixed adeno-neuroendocrine carcinoma of the tongue arising within a congenital enteric cyst (Table 3).10 

Table 3

Cases of Intraoral Nonsalivary Lower Gastrointestinal/Colonic Type Adenocarcinomas of Tongue and Oral Cavity Reported in the Literature

Cases of Intraoral Nonsalivary Lower Gastrointestinal/Colonic Type Adenocarcinomas of Tongue and Oral Cavity Reported in the Literature
Cases of Intraoral Nonsalivary Lower Gastrointestinal/Colonic Type Adenocarcinomas of Tongue and Oral Cavity Reported in the Literature

There are 2 schools of thought for the origin of these tumors: that they originate in the remnants of minor salivary tissue1  or as a malignant transformation in a preexisting teratomatous lesion that could be a choristoma containing GI elements,7  a congenital teratoid cyst,13  or a foregut duplication cyst.6  Evidence in favor of both of these theories exists in the literature, albeit in the form of individual case reports.

Cases 1 and 5 in our series showed a cystic structure consistent with a foregut duplication cyst; however, both cases showed morphologic and IHC findings of a UGI-LA. One of these cysts had a gastro-pancreatic–type lining epithelium, which was also described in the case reported by Agaimy et al.13  The other case had an adenocarcinoma, mucin-secreting type, arising in a cyst lined by pseudostratified ciliated respiratory-type epithelium, similar to the case reported by Volchok et al,6  potentiating the hypothesized origin in a foregut duplication cyst or a teratoid cyst. On IHC, CDX2 can be variably positive in upper- and lower-GI adenocarcinomas; upper-GI adenocarcinomas are much less frequently positive than lower-GI types.18  Of the UGI-LA cases in our series, case 5 was negative and cases 1 and 2 were positive for CDX2.

On the other hand, LGI-As or CATOCs frequently form large masses invading the underlying tissues and show necrosis and intracellular and extracellular mucin as well as desmoplastic stromal response.5,11  As in the present study, both the cases reported by Bell et al5  were BOT (oropharyngeal) tumors. The resected case of CATOC in our series (case 3) was the largest of all 6 cases in size; lacked any cystic structure/teratomatous elements, even after extensive sampling; had nodal metastasis at presentation; and developed distant metastasis while on treatment, corroborating the poor prognostic profile already reported.5,15  However, it is noteworthy that only 2 of 6 cases in the present series showed a phenotype consistent with CATOC. The LGI-A typically bears the immunoprofile of the colorectal adenocarcinomas, with positivity for CK20, CDX2, and SATB2. However, the expression of CK7 is quite variable in the literature. Slova et al7  and McDaniel et al8  showed positive staining for CK7, whereas Bell et al5  had focally positive CK7 in one case and completely negative CK7 in the other case. Zhang et al19  suggested SATB2 as a consistent and reliable biomarker for identifying the colorectal origin of the tumor, whereas Li et al20  suggested the use of SATB2/CDX2 (nuclear) and CK20 and villin (cytoplasmic) in combination for identifying colorectal primary. Both cases of LGI-A in our series were positive for CDX2 and/or SATB2, in line with the above study.

Despite the rarity of these tumors, the need for a nuanced classification is more than evident. The first step toward achieving that is the use of a terminology that distinguishes it from the other types of adenocarcinomas (of salivary origin, like PAC, MA, etc) that are more common at this subsite. We suggest using the broad term primary adenocarcinomas of the base of tongue (PABOT), which can be further subdivided into CATOC, UGI-LA, and NOS categories. Though it seems like an overembellishment considering the small number of cases in this series and those reported in the literature, the variation in histology (upper-GI/pancreatic type, lower-GI type, or nonintestinal type), the perceived origin (teratomatous versus salivary), the varied IHC profile (CK7+ versus CK20+), the need for a detailed molecular analysis, and thus the implications for management justify appropriate histologic subclassification and nomenclature.

Theoretically, PABOTs, by virtue of being adenocarcinomas, can be classified as adenocarcinoma, NOS, of the minor salivary glands of the BOT; however, subtle clinical, morphologic, and molecular parameters distinguish between the 2 entities. Adenocarcinoma, NOS, of salivary origin is uncommon in the BOT region according to published literature.2  The recent WHO classification15  mentioned that colonic phenotype (on morphology and IHC) is so unusual for adenocarcinoma, NOS, of the salivary glands that the mere presence of this phenotype should lead one to think of a separate, unrelated tumor. Moreover, the immunophenotype of CK20+/CDX2+ and SATB2+ is not explainable or exemplifiable by salivary gland origin, as this IHC profile is not seen in any part of the normal salivary gland tumor or any tumor arising from it. Moreover, the presence of a teratomatous cyst in 50% of our cases (all UGI-LA) and 2 of 9 previously reported cases in the literature suggests a unique tumor with origin in a congenital duplication cyst. This is further reinforced by the fact that the incidence of these tumors is much lower than that of the minor salivary gland tumors of the tongue, and mirrors the incidence of congenital duplication cysts of the tongue, which is less than 0.3% of all tongue tumors.1  It is well known that lingual duplication cysts can have all kinds of lining epithelium, including gastric, respiratory, squamous, and upper- and lower-GI type, upper-GI type being the most common.2124  Similar findings are seen in our study, where both the cases with identifiable cystic structures have UGI-LA. LGI-A is practically nonexistent in the realm of salivary gland tumors.1,15 

Case 6 of our series mandates its own discussion because of its unique histology, IHC profile, and molecular findings. The adenosquamous carcinoma has separate adenocarcinoma and squamous carcinoma areas; however, it is characterized by surface squamous dysplasia and malignant glands with luminal mucin, and hence can also be excluded.25  The adenocarcinoma areas were suggestive of PTC and/or CATMSG because of the optically clear nuclei and psammomatous calcification. However, the IHC profile was not supportive of either. The squamous areas were predominantly nonkeratinizing, with brisk mitosis and focal necrosis; however, negativity for p16 ruled out a BOT HPV-mediated squamous carcinomas. The marker that was expressed uniformly in this tumor was nuclear β-catenin, which is not a marker of a usual BOT adenocarcinoma or squamous carcinoma or of salivary gland tumors, except for basal cell adenoma. The IHC expression of β-catenin was mirrored by a missense mutation in exon 3 of the CTNNB1 gene on NGS testing. Intriguingly, though this mutation has not been reported in head and neck squamous or salivary gland tumors, it is a well-known mutation in colonic and pancreaticobiliary cancers. Similar to the other cases in this series, the dual histology in this case can also be explained by the rare but reported occurrence of squamous mucosa, mostly in combination with other types of GI lining epithelium in enteric duplication cysts.21  On the contrary, the combination of malignant keratinizing squamous and adenocarcinoma areas are not explicable by any salivary gland tumor. In addition, the β-catenin mutation detected in this case is a known mutation in UGI/pancreatic cancers and has not been reported in salivary gland tumors. Though not the exon 3 CTNNB1 mutation, mutations in other exons of the β-catenin gene are seen in some salivary gland tumors. The most common salivary gland tumor with β-catenin gene mutations is adenoid cystic carcinoma, which is not a morphologic differential in our case. These findings coupled with the presence of a cystic structure lined by squamous mucosa suggest a teratomatous origin of this tumor as well.

Thus, what seems to be a group of morphologically and immunohistochemically heterogeneous tumors can be unified by site and the possible underlying origin in an enteric duplication cyst. A purported salivary gland origin cannot offer a similar unification. Hence, the unique location, histology, and immunoprofile, the association with teratomatous elements, and the aggressive biological behavior necessitate the separation of PABOT cases from adenocarcinoma, NOS, of salivary origin.

The term PABOT, rather than being an attempt to create or redefine a small group of histologically and immunohistochemically different entities unified by site, is an attempt to provide an overarching categorization of a rare subgroup of adenocarcinomas of the oral/oropharyngeal region that need to be further studied and defined. This term and the subcategories thereof are an attempt to create a common language among pathologists for a morphology and IHC-based subtyping of these unusual adenocarcinomas, and more importantly their distinction from the established salivary gland tumors. The currently used terms adenocarcinoma, NOS; sinonasal-type adenocarcinoma; and CATOC do not serve the above-mentioned purpose, although we reiterate the rationale and evidence provided by Agaimy1  for using the term CATOC and similarly insist on the need for a thorough sampling for a preexisting cystic lesion.

An apt analogy to further explain the use of this new term would be the category of sinonasal undifferentiated carcinoma (SNUC), which was being used as a default diagnosis until detailed histologic, IHC, and molecular evaluation helped recognize SMARCB1-deficient carcinoma and IDH2-mutated SNUC from the cases previously labeled as SNUC.2629  Another pertinent example is the recent discovery of microsecretory carcinoma involving the major salivary glands, an entity that, until very recently, lay under the umbrella of “adenocarcinoma, NOS” in the major salivary glands.30 

PABOTs are quite different from conventional salivary gland tumors, and this distinction is of clinical value. PAC is a rare salivary gland neoplasm with an annual incidence of 0.051 cases/100 000 population and the second most common minor salivary gland tumor of the tongue region.31  Though it was initially thought to be a low-grade neoplasm with a low risk of metastases, recent studies have shown a high recurrence rate as well as the possibility of high-grade transformation.32  CATMSG is another salivary gland tumor occurring exclusively in the BOT and has histologic features overlapping with PAC. CATMSG is composed of solid, cribriform, and microcribriform tumor nodules containing hybrid secretory-myoepithelial cells. It was considered to be a variant of PAC until now. Its immunoprofile shows positivity for CK7, CAM5.2, CK8, CK18, S100p, and SMA.33,34  Xu et al,33  in their international interobserver study for consensus between PAC and CATMSG, reported 23% of their cases as tumors with indeterminate features, indicating the histomorphologic overlap between the 2 tumors. None of the cases in the present study showed an IHC profile of PAC or CATMSG or any other defined salivary gland tumor.

Salivary duct carcinoma, though rare in the BOT region, also needs exclusion because of its aggressive nature and potential for targeted therapy. The presence of papillary-cribriform areas, necrosis, high nuclear grade, apocrine look, and diffuse and strong positivity for AR are the reported characteristics of salivary duct carcinomas.35  None of the cases in the present study satisfied these criteria.

A relatively newly described “salivary mucinous adenocarcinoma with recurrent AKT1 E17K mutations” needs a mention as well. Lower-GI–type IHC markers, adenosquamous histology, and CDX2 positivity were not reported in the series by Rooper et al.36  This excludes all but case 5 of our series from a possible overlap. The most significant difference between case 5 of our series and the cases included in the series by Rooper et al36  is the presence of a cystic structure in the former and the evidence of an in situ and invasive carcinoma in the cyst wall, which confirms the origin of this tumor in a foregut duplication cyst, the IHC being corroborative. This finding has not been reported in any of the cases in the said series, though 3 of their cases were located in the BOT region. However, limited resources precluded molecular testing for AKT1 mutation in our case.

Second only to minor salivary gland adenocarcinomas, and overall still quite rare, MA may involve the oral soft tissues (33% of cases) or the jawbones.4,37  This differential diagnosis most significantly affects patient management in these cases, and therefore a detailed radiologic examination along with an elaborate panel of IHC is necessary to rule out metastasis. The primary site of origin for the MA was lung, followed by breast and female genital tract, and the most common site of metastatic tumor in the oral cavity was gingiva, followed by tongue, in a large study of 412 cases reported in the literature.4  Notably, 23.3% of these patients presented with oral cavity metastasis as the first sign of malignancy.4  Therefore, the IHC panel should include markers such as TTF-1, HepPar-1, GATA-3, and PAX-8, in addition to CK7, CK20, CDX2, SATB2, S100p, p63, and AR. As a protocol and because of the unusual histology, all the cases included in this series have undergone imaging and/or endoscopy procedures to rule out a GI primary. Also, cases included in this series are at least 6 months from the date of diagnosis to make allowance for late detection of the primary tumor, even though MAs usually present with colonic manifestations first per reported literature.9  Thus, all the cases in this series were primary tumors of the BOT region.

Though morphologic similarities exist, the lack of occupational exposure to wood dust distinguishes PABOT from sinonasal-type adenocarcinomas where exposure to wood dust is well established to initiate the metaplasia-dysplasia-neoplasia sequence.38  None of our cases had a history related to wood- or leather-dust exposure, which is in accordance with all but one case reported in the literature.7 

PTC arising in a thyroglossal duct and presenting as a BOT mass is a differential diagnosis as well. One case (case 6) had areas indistinguishable from the former on histology; however, the difference in IHC profile helped in clinching the diagnosis. PTC is characteristically positive for thyroid-related markers, that is, TTF1, PAX8, and TG. All the cases reported in this series were tested for at least one of these markers and were found to be negative for the same.

Recently, adenocarcinomas have also been added to the spectrum of HPV-mediated oropharyngeal adenocarcinomas (HPVMOAs).39,40  Thus, this entity was also considered in the differential diagnosis while evaluating current cases. p16 IHC was done in all our cases. Cases 1 and 6 showed less than 5% positivity for p16 IHC, interpreted as negative. Case 5 showed a focal area with approximately 20% of tumor cells showing positivity for p16, which is far lower than in the reported cases of HPVMOA where diffuse positivity for p16 has been reported consistently.39,40  Though the role of p16 as a surrogate marker in HPVMOA has not yet been defined by guidelines, the diffuse positivity of p16 in the published cases, and a lack of same thereof in our cases, reasonably excludes HPVMOA in our cases.

Exposure to tobacco was seen in 66.67% of our patients. However, the association between tobacco and PABOTs is unknown, possibly related to the rarity of the latter. Similarly, treatment strategies are nonuniform.8,9,14 

Thus, to conclude, primary adenocarcinomas of the BOT are unique and rare neoplasms with a heterogeneous morphology encompassing upper-GI/pancreatic–like, colonic/lower-GI–type, and non–GI-type histomorphologic features. A subset of these originate from teratoid or duplication cysts in the BOT region. A more detailed subclassification, through multicentric studies, based on histology, IHC, and probable teratomatous origin to clinically and biologically prognosticate each of these categories is essential for a better understanding of these rare tumors. The difference in management and prognosis necessitates sufficient workup (clinical, imaging, histologic with IHC markers) to rule out metastatic disease.

1.
Agaimy
A.
Colonic-type adenocarcinoma of the tongue and oral cavity (CATOC)
.
Head Neck Pathol
.
2018
;
12
(2)
:
291
293
.
2.
Pires
FR,
Pringle
GA,
de Almeida
OP,
Chen
SY.
Intra-oral minor salivary gland tumors: a clinicopathological study of 546 cases
.
Oral Oncol
.
2007
;
43
(5)
:
463
470
.
3.
Goel
AN,
Badran
KW,
Braun
APG,
Garrett
AM,
Long
JL.
Minor salivary gland carcinoma of the oropharynx: a population-based analysis of 1426 patients
.
Otolaryngol Head Neck Surg
.
2018
;
158
(2)
:
287
294
.
4.
Irani
S.
Metastasis to the oral soft tissues: a review of 412 cases
.
J Int Soc Prev Community Dent
.
2016
;
6
(5)
:
393
401
.
5.
Bell
D,
Kupferman
ME,
Williams
MD,
Rashid
A,
El-Nagar
AK.
Primary colonic-type adenocarcinoma of the base of the tongue: a previously unreported phenotype
.
Hum Pathol
.
2009
;
40
(12)
:
1798
1802
.
6.
Volchok
J,
Jaffer
A,
Cooper
T,
Al-Sabbagh
A,
Cavalli
G.
Adenocarcinoma arising in a lingual foregut duplication cyst
.
Arch Otolaryngol Head Neck Surg
.
2007
;
133
(7)
:
717
719
.
7.
Slova
D,
Mondolfi
AP,
Moisini
I,
et al.
Colonic-type adenocarcinoma of the base of the tongue: a case report of a rare neoplasm
.
Head Neck Pathol
.
2012
;
6
(2)
:
250
254
.
8.
McDaniel
AS,
Burgin
SJ,
Bradford
CR,
McHugh
JB.
Clinical problem solving: pathology: pathology quiz case 2
.
JAMA Otolaryngol Head Neck Surg
.
2013
;
13
(6)
:
653
.
9.
Rahimi
S,
Akaev
I,
Repanos
C,
Brennan
PA,
Dubois
JD.
Primary intestinal-type adenocarcinoma of tongue: a case report with immunohistochemical and molecular profiles and review of the literature
.
Head Neck Pathol
.
2017
;
11
(2)
:
186
191
.
10.
Ligthelm
LJ,
Bunn
BK,
Raubenheimer
EJ,
van Heerden
WFP.
Mixed adenoneuroendocrine carcinoma of the tongue arising within a congenital enteric cyst
.
Head Neck
.
2018
;
40
(5)
:
E53
E57
.
11.
Smith
SM,
Old
M,
Iwenofu
OH.
Primary lingual colonic-type adenocarcinoma: a rare and emerging distinct entity!
Head Neck Pathol
.
2017
;
11
(2)
:
234
239
.
12.
Guo
C,
Jia
MQ,
Wang
L,
Jia
J.
Primary intestinal-type adenocarcinoma of the tongue
.
Int J Oral Maxillofac Surg
.
2018
;
47
(12)
:
1523
1526
.
13.
Agaimy
A,
Raab
B,
Bonkowsky
V,
Wünsch
PH.
Intestinal-type adenocarcinoma arising in a congenital sublingual teratoid cyst
.
Virchows Arch
.
2007
;
450
(4)
:
479
481
.
14.
Berg
J,
Manosalva
RE,
Coughlin
A,
Su
YB,
Huang
TS,
Gentry
J.
Primary intestinal-type adenocarcinoma of the oral tongue: case report and review of histologic origin and oncologic management
.
Head Neck
.
2018
;
40
(7)
:
E68
E72
.
15.
Leivo
I,
Brandwein-Gensler
M,
Fonseca
I,
et al.
Adenocarcinoma, NOS
.
In
:
El-Naggar
A,
JK
Chan,
Grandis
JR,
Takata
T,
Slootweg
PJ,
eds
.
WHO Classification of Head and Neck Tumours
. 4th ed.
Lyon, France
:
International Agency for Research on Cancer
;
2017
:
171
172
.
WHO Classification of Tumours; vol 9
.
16.
Sun
KH,
Gao
Y,
Li
TJ.
Mucinous adenocarcinoma
.
In
:
Barnes
L,
Eveson
JW,
Reichart
P,
Sidransky
D,
eds
.
Pathology and Genetics of Head and Neck Tumours: World Health Organization Classification of Tumours
.
Lyon, France
:
IARC
;
2005
:
234
.
17.
Auclair
PL.
Cystadenocarcinoma
.
In
:
Barnes
L,
Eveson
JW,
Reichart
P,
Sidransky
D,
eds
.
Pathology and Genetics of Head and Neck Tumours: World Health Organization Classification of Tumours
.
Lyon, France
:
IARC
;
2005
:
232
.
18.
Park
DY,
Srivastava
A,
Kim
GH,
et al.
CDX2 expression in the intestinal-type gastric epithelial neoplasia: frequency and significance
.
Mod Pathol
.
2010
;
23
(1)
:
54
61
.
19.
Zhang
YJ,
Chen
JW,
He
XS,
et al.
SATB2 is a promising biomarker for identifying a colorectal origin for liver metastatic adenocarcinomas
.
EBioMedicine
.
2018
;
28
:
62
69
.
20.
Li
Z,
Rock
JB,
Roth
R,
et al.
Dual stain with SATB2 and CK20/villin is useful to distinguish colorectal carcinomas from other tumors
.
Am J Clin Pathol
.
2018
;
149
(3)
:
241
246
.
21.
Savran
B,
Kucur
C,
Kocak
C,
Ozbay
I,
Metineren
MH,
Karakus
YT.
Enteric duplication cyst located at the posterior tongue: a rare case report and review of the literature
.
Case Rep Otolaryngol
.
2015
;
2015
:
951878
.
22.
Tucker
R,
Maddalozzo
J,
Chou
P.
Sublingual enteric duplication cyst
.
Arch Pathol Lab Med
.
2000
;
124
(4)
:
614
615
.
23.
Satish Kumar
KV,
Joshi
M,
Vishwanath
N,
Akhtar
T,
Oak
SN.
Neonatal lingual gastric duplication cyst: a rare case report
.
J Indian Assoc Pediatr Surg
.
2006
;
11
:
97
98
.
24.
Azañero
WD,
Mazzonetto
R,
León
JE,
Vargas
PA,
Lopes
MA,
de Almeida
OP.
Lingual cyst with respiratory epithelium: a histopathological and immunohistochemical analysis of two cases
.
Int J Oral Maxillofac Surg
.
2009
;
38
(4)
:
388
392
.
25.
Dubal
PM,
Unsal
AA,
Echanique
KA,
et al.
Laryngeal adenosquamous carcinoma: a population-based perspective
.
Laryngoscope
.
2016
;
126
(4)
:
858
863
.
26.
Haerle
SK,
Gullane
PJ,
Witterick
IJ,
Zweifel
C,
Gentili
F.
Sinonasal carcinomas: epidemiology, pathology, and management
.
Neurosurg Clin N Am
.
2013
;
24
(1)
:
39
49
.
27.
Guilmette
J,
Sadow
PM.
High-grade sinonasal carcinoma classification through molecular profiling
.
Arch Pathol Lab Med
.
2019
;
143
(11)
:
1416
1419
.
28.
Stelow
EB,
Bishop
JA.
Update from the 4th edition of the World Health Organization Classification of Head and Neck Tumors: Tumors of the Nasal Cavity, Paranasal Sinuses and Skull Base
.
Head Neck Pathol
.
2017
;
11
(1)
:
3
15
.
29.
Bishop
JA.
Recently described neoplasms of the sinonasal tract
.
Semin Diagn Pathol
.
2016
;
33
(2)
:
62
70
.
30.
Bishop
JA,
Weinreb
I,
Swanson
D,
et al.
Microsecretory adenocarcinoma: a novel salivary gland tumor characterized by a recurrent MEF2C-SS18 fusion
.
Am J Surg Pathol
.
2019
;
43
(8)
:
1023
1032
.
31.
Mimica
X,
Katabi
N,
McGill
MR,
et al.
Polymorphous adenocarcinoma of salivary glands
.
Oral Oncol
.
2019
;
95
:
52
58
.
32.
Kimple
AJ,
Austin
GK,
Shah
RN,
et al.
Polymorphous low-grade adenocarcinoma: a case series and determination of recurrence
.
Laryngoscope
.
2014
;
124
(12)
:
2714
2719
.
33.
Xu
B,
Barbieri
AL,
Bishop
JA,
et al.
Histologic classification and molecular signature of polymorphous adenocarcinoma (PAC) and cribriform adenocarcinoma of salivary gland (CASG): an international interobserver study
.
Am J Surg Pathol
.
2020
;
44
(1)
:
545
552
.
34.
Michal
M,
Kacerovska
D,
Kazakov
DV.
Cribriform adenocarcinoma of the tongue and minor salivary glands: a review
.
Head Neck Pathol
.
2013
;
7
(suppl 1)
:
3
11
.
35.
Udager
AM,
Chiosea
SI.
Salivary duct carcinoma: an update on morphologic mimics and diagnostic use of androgen receptor immunohistochemistry
.
Head Neck Pathol
.
2017
;
11
(3)
:
288
294
.
36.
Rooper
LM,
Argyris
PP,
Thompson
LDR,
et al.
Salivary mucinous adenocarcinoma is a histologically diverse single entity with recurrent AKT1 E17K mutations: clinicopathologic and molecular characterization with proposal for a unified classification
.
Am J Surg Pathol
.
2021
;
45
:
1337
1347
.
37.
Jham
BC,
Salama
AR,
McClure
SA,
Ord
RA.
Metastatic tumors to the oral cavity: a clinical study of 18 cases
.
Head Neck Pathol
.
2011
;
5
(4)
:
355
358
.
38.
Bignami
M,
Lepera
D,
Volpi
L,
et al.
Sinonasal non–intestinal-type adenocarcinoma: a retrospective review of 22 patients
.
World Neurosurg
.
2018
;
120
:
e962
e969
.
39.
Hanna
J,
Reimann
JDR,
Haddad
RI,
Krane
JF.
Human papillomavirus-associated adenocarcinoma of the base of the tongue
.
Hum Pathol
.
2013
;
44
(8)
:
1516
1523
.
40.
Perez-Ordoñez
B,
Irish
JC,
Yu
ES,
Gillison
ML.
Human papillomavirus-16 associated adenocarcinoma NOS of base of tongue
.
Head Neck Pathol
.
2013
;
7
(3)
:
268
273
.

Author notes

Supplemental digital content is available for this article at https://meridian.allenpress.com/aplm in the November 2023 table of contents.

The authors have no relevant financial interest in the products or companies described in this article.

. Yadav and Mittal contributed equally toward this research work and writing the manuscript and share first authorship.

Supplementary data