Abstract
Hyalinizing clear cell carcinoma (HCCC) is a salivary gland carcinoma characterized by the presence of clear and eosinophilic cells within a hyalinized stroma and the EWSR1 rearrangement. Aiming to identify prognostic factors and establish a grading system, we herein conducted a detailed clinicopathology review of a large retrospective cohort of 87 HCCCs from seven tertiary centers.
Most HCCCs (91%) originated from minor salivary glands, although major salivary glands were affected in 8%. The most common sites were base of tongue, palate, nasopharynx, and maxilla . Eosinophilic cells were more prevalent than clear cells. Histologic features included Intraosseous component (19%), perineural invasion (48%), lymphovascular invasion (LVI, 16%), nuclear pleomorphism (14%), tumor necrosis (26%), and a mitotic index (MI) ≥5/2 mm2, 9%). Factors associated with increased risk of nodal metastasis at presentation included LVI, high MI, and tumor necrosis.
The 10-year disease-specific survival (DSS), disease-free survival (DFS), and distant metastasis-free survival (DMFS) were 80%, 51%, and 87%, respectively. Significant prognostic factors identified on univariate survival analysis included MI≥5/2 mm2, tumor necrosis, atypical mitosis, and nuclear pleomorphism for DSS; LVI, MI≥5/2 mm2, and percentage of clear cells for DFS; and nodal metastasis, LVI, MI≥5/2 mm2, tumor necrosis, and atypical mitosis for DMFS. The only independent prognostic factor for DFS identified on multivariate survival analysis was MI≥5/2 mm2. High grade HCCCs, defined as tumors with MI≥5/2 mm2 and/or tumor necrosis, were associated with increased risk of nodal metastasis at presentation and shortened DSS and DMFS. Among 67 HCCCs examined for EWSR1 rearrangement, 65 (97%) harbored EWSR1 translocation.
In conclusion, we identified multiple prognostic factors in HCCC, including MI, necrosis, atypical mitosis, nuclear pleomorphism, lymphovascular invasion, and percentage of clear cells. We herein proposed a prognostically relevant two-tiered grading system, classifying HCCC with a MI≥5/10 2 mm2 and/or tumor necrosis as high grade.
Introduction
Hyalinizing clear cell carcinoma (HCCC) is a rare salivary gland carcinoma, accounting for less than 3% of all salivary gland malignancies 1. In 2011, Antonescu et al. were the first to identify EWSR1::ATF1 translocation in the majority of HCCCs 2. Since then, HCCC has become a unique type of salivary gland carcinoma that can be distinguished from other salivary gland tumors with clear cell changes by the presence of EWSR1 translocation. The current World Health Organization (WHO) classification (5th edition) defines HCCC as a carcinoma that is composed of clear to eosinophilic cells in a hyalinized stroma, often harboring an EWSR1 rearrangement 3.
Given its rarity, published series describing the clinical and pathologic characteristics have included only limited number of patients, ranging from 3 to 26 patients 1,2,4–15. Among these, Hernandez-Prera et al. was the only study to report 10-year disease-specific survival (DSS), local recurrence free survival, and regional recurrence-free survival, being 97%, 51%, and 85% respectively, based on merely 15 patients with head and neck HCCCs 1. Adverse outcomes, although infrequent, have been reported in patients with HCCC. In a recent literature review of 254 HCCCs from 97 studies, the pooled rate of distant metastasis, regional metastasis, and disease-specific mortality was reported as 4%, 9% and 3%, respectively 16. To date, no prognostic factors have been found in HCCC, possibly due to the small number of cases included in prior studies.
Aiming to identify prognostic pathologic features and to establish a prognostically-relevant grading system for head and neck HCCC, we herein conducted a detailed clinicopathologic review of a retrospective cohort of 87 head and neck HCCCs from seven academic centers, representing the largest HCCC cohort to date.
Material and methods
Study cohort and clinicopathologic review
All candidate cases were reviewed by at least one head and neck pathologist (BX, KV, DL, KM, LR, MBW, ZB, or NK). A proportion of cases (n=47) additionally underwent central review (BX and NK) to confirm the diagnosis of HCCC and to exclude potential diagnostic mimickers. Detailed clinicopathologic characteristics and outcome were collected. The outcomes collected included overall survival (OS), DSS, disease free survival (DFS), and distant metastasis-free survival (DMFS).
The study cohort included 87 cases with a confirmed diagnosis of HCCC from the head and neck region (Memorial Sloan Kettering Cancer Center n=43, Icahn School of Medicine at Mount Sinai n=12, University of Pennsylvania n=10, Emory University n=9, Johns Hopkins University n=7, Yale University n=4, and Centre Hospitalier Universitaire Vaudois n=2).
Immunohistochemistry studies were performed in 73 cases. The primary antibodies utilized included p40 (clone BC28, dilution 1:400, Biocare) , p63 (clone 4A4, ready to use [RTU], Ventana), CK5/6 (clone D5/1684, RTU, Ventana), 34BE12 (clone 34BE12, RTU, Ventana) , cytokeratin AE1/AE3 (clone AE1/AE3, dilution 1:1600, DAKO), CK7 (clone OV-TL-12/30, dilution 1:800, DAKO), CAM5.2 (clone CAM5.2, dilution 1:75, Becton & Dickinson), S100 (polyclonal, RTU, Leica), SOX10 (clone BC34, dilution 1:50, Biocare), calponin (clone EP798Y, RTU, Cell Marque), and smooth muscle actin (SMA, clone 1A4, dilution 1:200, Cell Marque). Mucicarmine stain was performed in 17 cases.
Detection of EWSR1 translocation
EWSR1 translocation was examined using various platforms in 67 cases (MSK-IMPACT n=5, ARCHER n=13, FoundationOne n=1, and/or fluorescence in situ hybridization [FISH] n=52). MSK-IMPACT is a Food and Drug Administration (FDA)-approved deep-coverage, targeted next-generation sequencing (NGS) assay detecting single nucleotide variants (SNVs), small insertions/deletion (indels), copy number variants (CNVs) and select fusion/structural variants in 505 cancer-related genes 17,18. ARCHER RNA sequencing platform is a clinical molecular diagnostic essay performed in a CLIA-accredited laboratory utilizing multiplex polymerase chain reaction (PCR) to detect oncogenic fusion transcripts involving 123 genes 19. FoundationOne is a commercially-available FDA-approved NGS platform detecting SNV, indels, CNAs and rearrangement in 324 genes 20. FISH for EWSR1 and ATF1 rearrangements was performed using a custom bacterial artificial chromosome clone probes designed to flank the target genes based on the UCSC genome browser (http://genome.ucsc.edu/) as previously described 2.
Outcome and statistical analysis
Statistical analyses were performed using SPSS software, version 29 (IBM, Armonk, NY, USA). The association between clinicopathologic parameters and risk of nodal metastasis at presentation was calculated using Fisher’s exact test for categorical parameters or two-tailed Student’s t test for continuous variables. Follow-up data were available in 68 cases. The outcomes studied included OS, DSS, DFS and DMFS. Univariate survival analysis was performed using log rank test for categorical variables and Cox proportional hazards model for continuous variables. Multivariate survival analysis was subsequently performed using Cox proportional hazards model and clinicopathologic factors significant on univariate survival analysis.
Results
Clinicopathologic features of HCCC
The clinicopathologic features of the study cohort are shown in Table 1. Most HCCCs (n=79, 91%) originated from the minor salivary glands of the upper aerodigestive tract. The most common anatomic sites in a descending order were base of tongue (n=23), palate (n=15), nasopharynx (n=8), and maxilla (n=8). Other sites of origin included floor of mouth, tongue, oropharynx, retromolar trigone, sinonasal tract, lip, mandible, trachea, and buccal mucosa. Major salivary glands were occasionally affected in seven cases (8%), including parotid (n=3), sublingual (n=2), and submandibular glands (n=2).
Table 1. Clinicopathologic characteristics of 87 head and neck hyalinizing clear cell carcinomas stratified by lymph node metastasis.
Values are n (column%) for categorical variables and median (range) for continuous variables. P values were obtained using two-tailed Student’s t test for continuous variables and Fisher’s exact test for categorical variables. NS: not significant (P > 0.05).
| All cases (n=87) |
pN0/Nx (n=69, 79%) |
pN+ (n=18, 21%) |
P values | |
|---|---|---|---|---|
| Female : male ratio | 1.2:1 | 1.1:1 | 1.3:1 | NS |
| Age (years) | 57 (13–92) | 57 (13–92) | 65 (41–83) | NS |
| Procedure | NS | |||
| Biopsy/incision | 16 (18%) | 15 (22%) | 1 (6%) | |
| Resection/excision | 70 (80%) | 53 (77%) | 17 (94%) | |
| Unknown | 1 (1%) | 1 (1%) | 0 (0%) | |
| Site of origin | ||||
| Minor salivary glands | 79 (91%) | 62 (90%) | 17 (94%) | NS |
| Base of tongue | 23 | 14 | 9 | |
| Palate | 15 | 13 | 2 | |
| Nasopharynx | 8 | 6 | 2 | |
| Maxilla | 8 | 8 | 0 | |
| Tongue, not otherwise specified | 6 | 4 | 2 | |
| Floor of mouth | 4 | 3 | 1 | |
| Oropharynx, not otherwise specified | 3 | 3 | 0 | |
| Retromolar trigone | 2 | 2 | 0 | |
| Sinonasal tract | 3 | 3 | 0 | |
| Lip | 3 | 2 | 1 | |
| Mandible | 2 | 2 | 0 | |
| Trachea | 1 | 1 | 0 | |
| Buccal | 1 | 1 | 0 | |
| Major salivary glands | 7 (8%) | 6 (9%) | 1 (6%) | |
| Parotid | 3 | 3 | 0 | |
| Sublingual | 2 | 1 | 1 | |
| Submandibular | 2 | 2 | 0 | |
| Unknown | 1 (1%) | 1 (1%) | 0 (0%) | |
| Tumor size (cm) | 2.2 (0.6–9.5) | 2.1 (0.6–9.5) | 3.1 (1.0–6.5) | |
| AJCC pT stage (n=67) | NS | |||
| pT1 | 18 (27%) | 15 (29%) | 3 (19%) | |
| pT2 | 24 (36%) | 17 (33%) | 7 (44%) | |
| pT3 | 11 (16%) | 7 (14%) | 4 (25%) | |
| pT4 | 14 (21%) | 12 (24%) | 2 (13%) | |
| Positive resection margin (n=67) | 13 (19%) | 11 (22%) | 2 (13%) | NS |
| Perineural invasion | 40 (48%) | 28 (42%) | 12 (71%) | NS |
| Lymphovascular invasion | 13 (16%) | 3 (4%) | 10 (59%) | <0.001 |
| Intraosseous component | 16 (19%) | 15 (22%) | 1 (6%) | NS |
| Mitotic index | 0 (0–12) | 0 (0–9) | 2 (0–12) | <0.001 |
| ≥ 5/2 mm2 | 8 (9%) | 4 (6%) | 4 (22%) | NS |
| Tumor necrosis | 22 (26%) | 11 (16%) | 11 (61%) | <0.001 |
| High grade (≥ 5/2 mm2 and/or necrosis) | 24 (28%) | 13 (19%) | 11 (61%) | <0.001 |
| Atypical mitosis | 7 (8%) | 4 (6%) | 3 (17%) | NS |
| Nuclear pleomorphism | 11 (14%) | 6 (10%) | 5 (28%) | NS |
| Architecture | ||||
| Sheets/solid architecture (%) | 43 (0–100) | 40 (0–100) | 48 (0–100) | NS |
| Trabeculae/cord architecture (%) | 30 (0–100) | 33 (0–100) | 25 (0–75) | NS |
| Small clusters/single files (%) | 10 (0–80) | 10 (0–80) | 10 (0–40) | NS |
| Ducts (%) | 0 (0–20) | 0 (0–20) | 0 (0–20) | NS |
| Cribriform architecture (%) | 0 (0–20) | 0 (0–20) | 0 (0–1) | NS |
| Cystic/microcystic (%) | 0 (0–50) | 0 (0–50) | 0 | NS |
| Cellular composition | ||||
| Eosinophilic cells (%) | 50 (0–95) | 50 (0–95) | 48 (5–90) | NS |
| Clear cell (%) | 40 (0–100) | 40 (0–100) | 30 (10–90) | NS |
| Basaloid cells (%) | 10 (0–90) | 10 (0–90) | 10 (0–65) | NS |
| Intracellular mucin/mucoytes | 24 (28%) | 17 (25%) | 7 (39%) | NS |
| Stroma | ||||
| Hyalinized stroma | 83 (97%) | 66 (97%) | 17 (94%) | NS |
| Fibrocellular stroma | 61 (71%) | 47 (69%) | 14 (78%) | NS |
| Follow up period (months) | NS | |||
| Post-operative radiation (n=67) | 25 (37%) | 17 (33%) | 8 (50%) | NS |
| Post-operative chemotherapy (n=67) | 15 (22%) | 8 (16%) | 7 (44%) | 0.035 |
The median age of diagnosis was 57 years (range: 13–92). There was a slight female predominance with a female to male ratio of 1.2:1. Sixteen tumors (19%) had an intraosseous component, including six from maxilla, two from hard palate, two from retromolar trigone, two from floor of mouth, and one each from base of tongue, mandible, nasopharynx, and sinonasal tract. Perineural invasion, lymphovascular invasion, pT4 stage, and positive resection margins were seen in 48%, 16%, 21%, and 19% respectively. The most common architectural pattern was sheet/solid architecture (median 43%), followed by trabeculae/cords (30%), and small clusters/single files (10%). Other rare minor architectural patterns (median <1%) that were noted included ducts, cribriform, macrocystic, and microcystic patterns (Figure 1).
Figure 1. Histologic features of hyalinizing clear cell carcinoma (HCCC).
Architectural patterns of HCCC include solid sheets and trabeculae (A), small nests and single files (B), macrocysts (C), microcysts (D), glands/tubules (E), and cribriform pattern (F). The stroma between tumor nests is hyalinized (H) and/or fibrocellular (FC, A). The tumor cells may exhibit clear cells (A and D), basaloid (B), or eosinophilic features (C and D). Perineural invasion (N) may be seen (panel B). (G) A HCCC of the hard palate with EWSR1::ATF1 fusion is composed of basaloid nests exhibiting peripheral palisading with reverse polarity and central spindling, mimicking ameloblastoma. (H) Extracellular and intracellular mucin may be seen in HCCCs, which can be highlighted using mucicarmine stain (insert). (I) A HCCC of the maxilla with EWSR1::ATF1 fusion is composed of a soft tissue (ST) and an intraosseous (IO) component, indicating pT4a disease. (J) A high grade HCCC shows high grade areas (left) with basaloid appearance and extensive tumor necrosis (N) juxtaposed to a conventional HCCC (right) with clear cell cytomorphology, low mitotic count, and absence of tumor necrosis. (K) High grade HCCCs are characterized by tumor necrosis (N) and elevated mitotic activity (blue arrowheads). (L) Nuclear pleomorphism is defined in this study as variation of nuclear diameter by at least 3 folds.
Despite the name, the most common cell type was eosinophilic cells (median 50%), followed by clear cells (median 40%) and basaloid cells with high nuclear/cytoplasmic ratio (median 10%). Three tumors had no (n=1) or 5% of clear cells only. Mucocytes with intracellular mucin were identified in 24 HCCCs (28%). Cells with intracellular mucin were present focally within the tumor, being seen in 1% to 10% of tumor cells. Mucicarmine stain was performed in 17 cases. Among them, seven tumors (41%) contained tumor cells with intracellular mucin. One HCCC of the palate (1%) exhibited “ameloblastic-like” features and was composed of basaloid nests with prominent peripheral palisading, and central spindle to stellated cells (Figure 1G). Hyalinized and fibrocellular stroma was identified in 97% and 71% of HCCC respectively.
Twenty-two HCCCs (26%) had tumor necrosis. Eight tumors (9%) had a mitotic index ≥ 5/2 mm2. Using an arbitrary grading system defining high grade HCCC as those with a mitotic index ≥ 5/2 mm2 and/or tumor necrosis, 24 (28%) HCCCs were classified as high grade. Nuclear pleomorphism, defined arbitrarily in this study as at least three-fold variation in nuclear diameter, was identified in 11 tumors (14%).
Predictors of nodal metastasis at presentation
Lymph node metastasis was observed in 18 cases (21%) at the time of the initial diagnosis. The clinicopathologic factors associated with regional metastasis were lymphovascular invasion (4% in pN0/Nx patients, 59% in pN+ patients, P < 0.001), MI (as continuous variable, median in pN0/pNx patients 0/2 mm2, in pN+ patients 2/2 mm2, P < 0.001), tumor necrosis (16% in pN0/Nx patients, 61% in pN+ patients, P < 0.001), and high grade (19% in pN0/Nx, 61% in pN+, P < 0.001). Other parameters did not differ significantly between patients with and without nodal metastasis (Table 1). When using a MI ≥ 5/2 mm2 as the binary cutoff value, there was a non-significant association between high MI and nodal metastasis (P = 0.056). Other pathologic features, e.g. cellular composition, architecture, stroma, nuclear pleomorphism, atypical mitosis, perineural invasion, margin status, and AJCC pT stage, did not impact the risk of nodal metastasis at presentation (P > 0.05).
Nodal metastasis was associated with significantly higher frequency of post-operative chemotherapy (pN+: 44%, pN0/pNx: 22%, P = 0.035), possibly due to selection bias for patients with high stage tumors to receive adjuvant therapy.
Immunohistochemical profile of HCCC
Immunohistochemistry studies were performed in 73 cases. The majority of HCCCs were positive for CAM5.2 (12/14, 86%),CK7 (29/31, 94%), cytokeratin AE1/AE3 (24/25, 95%) and squamous markers, e.g. p63 (54/55, 98%), p40 (36/37, 97%), CK5/6 (32/32, 100%), and 34BE12 (10/10, 100%), Myoepithelial markers were universally negative, e.g. S100 (0/50, 0%), calponin (0/31, 0%), and SMA (0/29, 0%). SOX10 immunopositivity was detected in 1/11 (9%).
EWSR1 translocation in HCCC
Among the 67 cases tested for EWSR1 translocation, 65 (97%) harbored EWSR1 translocation. EWSR1 translocation was not detected in two cases by FISH. The fusion partner was known in 20 cases, including EWSR1::ATF1 in 16 (80%) and EWSR1::CREM in 4 (20%). All HCCCs with EWSR1::CREM fusions originated from the nasopharynx (n=3) or maxilla (n=1).
Outcome and prognostic factors
Follow up data were available in 68 patients with a median follow up of 45.8 months (range: 1.4 – 441.3 months). Adverse events were seen in 20 patients, including four with local persistent disease, nine with local recurrence, seven with regional recurrence, six with distant metastasis and eight with disease-related mortality. Late recurrence, defined as recurrence at least 60 months after the primary resection, was seen in six patients, including local recurrence in three, regional recurrence in three, and/or distant metastasis in one. The 5-year OS, DSS, DFS, and DMFS were 90%, 94%, 73%, and 92% respectively. The 10-year OS, DSS, DFS, and DMFS were 73%, 80%, 51%, 87% respectively.
Univariate survival analysis was performed using log rank test for categorical variables and Cox proportional hazards model for continuous variables. Among different thresholds of mitotic index (MI) tested, a cutoff value of ≥5/2 mm2 was the most significant predictor for DSS (P < 0.001, Table 3 and Figure 2), DFS (P < 0.001), and DMFS (P < 0.001). This cutoff was thereafter incorporated in an arbitrary grading system defining high-grade HCCCs as those with a MI ≥5/2 mm2 and/or tumor necrosis. Using this definition, high-grade HCCCs was associated with shortened DSS (P = 0.010) and DMFS (P = 0.002). The 10-year DSS for low-grade and high-grade HCCCs was 97% and 50% respectively, while the 10-year DMFS for low-grade and high-grade HCCCs was 97% and 65% respectively.
Table 3. Univariate survival analysis using log rank test for categorical variables and Cox proportional hazards model for continuous variables.
Values are P values. Bold P values are significant P values. DSS: disease-specific survival, DFS: disease-free survival, DMFS: distant metastasis-free survival, HPFs: high power fields.
| Clinicopathologic features | DSS | DFS | DMFS |
|---|---|---|---|
| Sex | 0.690 | 0.821 | 0.987 |
| Age (continuous variable) | 0.515 | 0.476 | 0.139 |
| Site (major vs. minor) | 0.277 | 0.711 | 0.650 |
| Size (continuous variable) | 0.765 | 0.409 | 0.710 |
| AJCC pT stage | 0.700 | 0.856 | 0.548 |
| Nodal metastasis | 0.994 | 0.308 | <0.001 |
| Resection margin | 0.355 | 0.183 | 0.309 |
| Perineural invasion | 0.561 | 0.422 | 0.181 |
| Lymphovascular invasion | 0.057 | 0.014 | <0.001 |
| Intraosseous component | 0.209 | 0.099 | 0.905 |
| Mitotic index ≥ 5/2 mm2 | <0.001 | <0.001 | <0.001 |
| Tumor necrosis | 0.006 | 0.330 | 0.012 |
| Grade (mitotic index ≥ 2 mm2 and/or necrosis) | 0.010 | 0.436 | 0.017 |
| Atypical mitosis | 0.020 | 0.145 | 0.002 |
| Nuclear pleomorphism | 0.027 | 0.196 | 0.494 |
| Sheets/solid architecture (%) | 0.439 | 0.753 | 0.712 |
| Trabeculae/cord architecture (%) | 0.809 | 0.672 | 0.983 |
| Small clusters/single filing (%) | 0.740 | 0.529 | 0.149 |
| Eosinophilic cells (%) | 0.649 | 0.210 | 0.461 |
| Clear cell (%) | 0.790 | 0.042 | 0.642 |
| Basaloid cells (%) | 0.322 | 0.132 | 0.095 |
| Intracellular mucin/mucoytes | 0.506 | 0.942 | 0.753 |
| Hyalinized stroma | 0.808 | 0.644 | 0.849 |
| Fibrocellular stroma | 0.978 | 0.213 | 0.108 |
| Post-operative radiation | 0.112 | 0.015 | 0.254 |
| Post-operative chemotherapy | 0.030 | 0.023 | 0.290 |
Figure 2. Kaplan Meier curves for disease-specific survival (DSS), disease-free survival (DFS), and distant metastasis-free survival (DMFS).
MI: mitotic index.
Additional adverse prognostic factors identified included: tumor necrosis, atypical mitosis, and nuclear pleomorphism for DSS; lymphovascular invasion and smaller percentage of clear cells for DFS; and nodal metastasis at presentation, lymphovascular invasion, tumor necrosis, and atypical mitosis for DMFS (P < 0.05, Table 3). Post-operative radiation therapy was associated with shortened DFS, whereas adjuvant chemotherapy was associated with shortened DSS and DFS, possibly due to selection bias for aggressive tumors to received adjuvant therapy.
Multivariate survival analysis showed that a MI ≥5/2 mm2 was an independent adverse prognostic factor for DFS (hazard ratio = 6.849, 95% confidence interval 1.370 – 34.241, P = 0.019). No other independent prognostic factors were identified for DSS, DFS, and DMFS (data not shown, P > 0.05).
Discussion
Since Milchgrub et al. first described a series of 11 patients and coined the terminology “hyalinizing clear cell carcinoma” in 1994 15, no prognostic factors have been identified and no grading system has been proposed for HCCC, possibly due to the small number of HCCC cases (up to 26 patients) included in prior studies 1,2,4–15. Three cases of high-grade transformation of HCCC have been reported in the literature 5,21,22. However, the very definition of high-grade transformation has not been clearly outlined. In prior reports, features associated with high-grade transformation included nuclear anaplasia (bizarre nuclei), easily identifiable mitotic activity, high nuclear/cytoplasmic ratio, and necrosis 5,21,22. In this multicenter retrospective cohort of 87 patients, we identify for the first time multiple prognostic factors in HCCC, including MI ≥5/2 mm2 (for DSS, DFS, and DMFS), tumor necrosis (for DSS and DMFS), nuclear pleomorphism (for DSS), atypical mitosis (for DSS and DMFS), lymphovascular invasion (for DFS and DMFS), and percentage of clear cells (for DFS).
Based on these prognostic factors, we herein proposed a two-tiered grading system using MI ≥5/2 mm2 and/or tumor necrosis to define high-grade HCCCs. In our cohort, high-grade HCCCs defined as such were associated with significantly shortened DSS and DMFS and an increased risk of nodal metastasis at presentation. The risk of nodal metastasis and 10-year DSS and DMFS were 11%, 97%, and 97%, respectively for low-grade HCCCs, compared with 46%, 50%, 65%, respectively for high-grade HCCCs. This grading system may provide treating clinicians guidance on management such as the need for surgical evaluation of the neck alone with additional prognostic information.
In a retrospective cohort of 15 patients with HCCC, Hernandez-Prera et al. showed that HCCCs had a 10-year risk of regional and local recurrence of 15% and 49% respectively. In our cohort, the 10-year risk for recurrence/persistent disease was 49% and the 10-year risk for distant metastasis risk was 13%. Furthermore, we have shown here late recurrence (being local, regional, or distant) might occur in patients with HCCCs. Although HCCC is overall regarded as an indolent salivary gland carcinoma, these findings suggest that the risk of recurrence and metastasis is not trivial. Long term surveillance and follow up may be required for patients with HCCC.
In 2011, Antonescu et al. were the first to report EWSR1::ATF1 translocation in HCCCs 2. Since then, EWSR1 fusion has been reported in 82% to 100% of HCCCs 1,2,4–6,8,9,11,23. In a recent literature review, Desai et al. have shown that the pooled frequency of EWSR1 translocation was 96% among 115 HCCC cases subjected to various platforms of molecular testing 16. Similarly, among 67 cases tested for EWSR1 translocation in our study, the frequency of EWSR1 translocation, either EWSR1::ATF1 or EWSR1::CREM, was 97%. Together, these data suggest that EWSR1 translocation is a highly prevalent diagnostic molecular signature for HCCC.
HCCC is often subjected to misdiagnosis, particularly in biopsy specimens. The prevalence of eosinophilic cells, despite the name, along with the unexpected occurrence of mucocytes and ductal structures, likely contributes to this diagnostic challenge. Several prior studies have shown that up to 72% of HCCCs may be mistakenly classified as other entities 1,6,8,15,24,25, such as mucoepidermoid carcinoma 1,15,24, squamous cell carcinoma 15,25, epithelial myoepithelial carcinoma 1,8,15, myoepithelial carcinoma 8, pleomorphic adenoma 15, acinic cell carcinoma 15, and calcifying epithelial odontogenic tumor 15,23. Indeed, HCCC can be histologically mistaken as mucoepidermoid carcinoma as up to 50% of HCCCs (28% in current study) 2,26,27 may contain mucocytes and show macrocystic and microcystic architecture. Squamous cell carcinoma of the mucosal origin may also exhibit clear cell changes but often contain true keratinization and marked nuclear pleomorphism. Notably, we demonstrated that HCCC may rarely exhibit “ameloblastic-like” features, resembling ameloblastoma. As shown in this and multiple prior studies 5,8,9,12–16,28, the immunoprofile of HCCCs is identical to mucoepidermoid carcinoma and squamous cell carcinoma, showing positivity for squamous markers (e.g. 34BE12, 34BE12, p40, and p63), while lacking immunoexpression for S100 and myoepithelial markers (e.g. calponin and SMA). This overlap limits the utility of immunohistochemistry in differentiating HCCC from these tumors. Fortunately, EWSR1 translocation is highly prevalent in HCCC. In diagnostically challenging cases, particularly in small biopsy materials, molecular testing for EWSR1 translocation is a valuable method for distinguishing HCCC from its histologic mimickers.
Clear cell odontogenic carcinoma is currently classified under odontogenic carcinoma in the WHO classification 3. It shares histologic features with HCCC, including sheets and cords of clear cells within fibrocellular/hyalinized stroma, and a high frequency (>80%) of EWSR1 translocation 1,3,26,27. By definition, clear cell odontogenic carcinoma is an intraosseous tumor, most frequently occurring in the mandible. However, soft tissue extension is also reported 3. In our series, all tumors were classified as HCCC and contained a soft tissue component. However, 19% had an intraosseous component (Figure 1I). Although it remains debatable whether clear cell odontogenic carcinoma represents a distinctive type of intraosseous odontogenic carcinoma, the overlap in histologic and molecular features with HCCC suggests that it may be a centrally located subtype of HCCC.
In conclusion, in this large multicenter retrospective cohort of 87 patients with head and neck HCCC, we identify for the first time multiple prognostically relevant pathologic features, including MI, tumor necrosis, atypical mitosis, nuclear pleomorphism, lymphovascular invasion, and percentage of clear cells. Additionally, we proposed a two-tiered grading system for HCCC, with high grade tumors defined by a MI ≥5/2 mm2 and/or tumor necrosis. This grading system can serve as a valuable tool for stratifying patients by prognosis and guiding their clinical management.
Table 2. Immunohistochemical (IHC) profile of hyalinizing clear cell carcinoma.
SMA: smooth muscle actin.
| IHC | Positive/tested (%) |
|---|---|
| CK5/6 | 32/32 (100%) |
| 34BE12 | 10/10 (100%) |
| p63 | 54/55 (98%) |
| p40 | 36/37 (97%) |
| AE1/3 | 24/25 (96%) |
| CK7 | 29/31 (94%) |
| CAM5.2 | 12/14 (86%) |
| SOX10 | 1/11 (9%) |
| S100 | 0/50 (0%) |
| Calponin | 0/31 (0%) |
| SMA | 0/29 (0%) |
Funding statement:
Research reported in this publication was supported in part by the Cancer Center Support Grant of the National Institutes of Health/National Cancer Institute under award number P30CA008748.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts of interest: No competing financial interests exist for all contributory authors.
Ethics/Patient Consent: The research meets the ethics guidelines, including adherence to the legal requirements of the country where the study was performed.
Data Availability statement:
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
References
- 1.Hernandez-Prera JC et al. Reappraising hyalinizing clear cell carcinoma: A population-based study with molecular confirmation. Head Neck 39, 503–511 (2017). [DOI] [PubMed] [Google Scholar]
- 2.Antonescu CR et al. EWSR1-ATF1 fusion is a novel and consistent finding in hyalinizing clear-cell carcinoma of salivary gland. Genes Chromosomes Cancer 50, 559–570 (2011). [DOI] [PubMed] [Google Scholar]
- 3.WHO Classification of Tumours Editorial Board. World Health Organization Classification of Tumours: pathology and genetics of head and neck tumours (5th edition). (International Agency for Research on Cancer (IARC), 2024). [Google Scholar]
- 4.Zhai C, Yuan C, Sun J, Song W, Wang S & Lin L Clinical and Histopathologic Analyses of Nasopharyngeal Hyalinizing Clear Cell Carcinoma: A Series of 26 Cases With Molecular Confirmation. Am J Surg Pathol 47, 1168–1175 (2023). [DOI] [PubMed] [Google Scholar]
- 5.Xuan L, Wang S, Wei J, Yuan J & Liu H Clinicopathological and molecular study of 10 salivary gland clear cell carcinomas, with emphasis on rare cases with high grade transformation and occurring in uncommon sites. Diagnostic pathology 17, 18 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gubbiotti MA, Montone K, Zhang P, Livolsi V & Baloch Z A contemporary update on hyalinizing clear cell carcinoma: compilation of all in-house cases at our institution and a literature review spanning 2015–2020. Hum Pathol 111, 45–51 (2021). [DOI] [PubMed] [Google Scholar]
- 7.Oliver J et al. Patterns of Care and Outcome of Clear Cell Carcinoma of the Head and Neck. Otolaryngol Head Neck Surg 161, 98–104 (2019). [DOI] [PubMed] [Google Scholar]
- 8.Zhao YN, Wang X, Liang FH, Zhang WJ & Song XT Hyalinizing clear cell carcinoma of salivary glands: A retrospective study focused on uncommon morphology, immunohistochemistry, and detection of gene fusion using fluorescence in situ hybridization. Pathol Res Pract 214, 380–384 (2018). [DOI] [PubMed] [Google Scholar]
- 9.Yang XH, Liu L, Shi YY, Hu YJ, Hu QG & Zhang P Hyalinizing clear cell carcinoma of salivary gland origin in the head and neck: clinical and histopathological analysis. Int J Oral Maxillofac Surg 47, 692–698 (2018). [DOI] [PubMed] [Google Scholar]
- 10.Bishop JA, Rooper LM, Chiosea SI & Westra WH Clear Cell Carcinoma of Salivary Glands Is Frequently p16 Positive: A Pitfall in the Interpretation of Oropharyngeal Biopsies. Am J Surg Pathol 42, 367–371 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Albergotti WG, Bilodeau EA, Byrd JK, Mims MM, Lee S & Kim S Hyalinizing clear cell carcinoma of the head and neck: Case series and update. Head Neck 38, 426–433 (2016). [DOI] [PubMed] [Google Scholar]
- 12.Nakano T et al. Hyalinizing clear cell carcinoma with EWSR1-ATF1 fusion gene: report of three cases with molecular analyses. Virchows Arch 466, 37–43 (2015). [DOI] [PubMed] [Google Scholar]
- 13.Solar AA, Schmidt BL & Jordan RC Hyalinizing clear cell carcinoma: case series and comprehensive review of the literature. Cancer 115, 75–83 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.O’Sullivan-Mejia ED, Massey HD, Faquin WC & Powers CN Hyalinizing clear cell carcinoma: report of eight cases and a review of literature. Head Neck Pathol 3, 179–185 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Milchgrub S, Gnepp DR, Vuitch F, Delgado R & Albores-Saavedra J Hyalinizing clear cell carcinoma of salivary gland. Am J Surg Pathol 18, 74–82 (1994). [DOI] [PubMed] [Google Scholar]
- 16.Desai A, Faquin WC, Iafrate AJ, Rivera MN, Jaquinet A & Troulis MJ Clear cell carcinoma: a comprehensive literature review of 254 cases. Int J Oral Maxillofac Surg 51, 705–712 (2022). [DOI] [PubMed] [Google Scholar]
- 17.Cheng DT et al. Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT): A Hybridization Capture-Based Next-Generation Sequencing Clinical Assay for Solid Tumor Molecular Oncology. The Journal of molecular diagnostics : JMD 17, 251–264 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Morris LG et al. The Molecular Landscape of Recurrent and Metastatic Head and Neck Cancers: Insights From a Precision Oncology Sequencing Platform. JAMA oncology (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhu G et al. Diagnosis of known sarcoma fusions and novel fusion partners by targeted RNA sequencing with identification of a recurrent ACTB-FOSB fusion in pseudomyogenic hemangioendothelioma. Mod Pathol 32, 609–620 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.FDA Summary of Safety and Effectiveness Data of FoundationOne CDx™, <https://www.accessdata.fda.gov/cdrh_docs/pdf17/P170019B.pdf> (2017).
- 21.Jin R, Craddock KJ, Irish JC, Perez-Ordonez B & Weinreb I Recurrent hyalinizing clear cell carcinoma of the base of tongue with high-grade transformation and EWSR1 gene rearrangement by FISH. Head Neck Pathol 6, 389–394 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.O’Regan E, Shandilya M, Gnepp DR, Timon C & Toner M Hyalinizing clear cell carcinoma of salivary gland: an aggressive variant. Oral Oncol 40, 348–352 (2004). [DOI] [PubMed] [Google Scholar]
- 23.Bilodeau EA et al. Clear cell odontogenic carcinomas show EWSR1 rearrangements: a novel finding and a biological link to salivary clear cell carcinomas. Am J Surg Pathol 37, 1001–1005 (2013). [DOI] [PubMed] [Google Scholar]
- 24.Hsieh MS, Wang H, Lee YH, Ko JY & Chang YL Reevaluation of MAML2 fusion-negative mucoepidermoid carcinoma: a subgroup being actually hyalinizing clear cell carcinoma of the salivary gland with EWSR1 translocation. Hum Pathol 61, 9–18 (2017). [DOI] [PubMed] [Google Scholar]
- 25.Xu B et al. Diagnostic discrepancy in second opinion reviews of primary epithelial neoplasms involving salivary gland: An 11-year experience from a tertiary referral center focusing on useful pathologic approaches and potential clinical impacts. Head Neck 43, 2497–2509 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Tanguay J & Weinreb I What the EWSR1-ATF1 fusion has taught us about hyalinizing clear cell carcinoma. Head Neck Pathol 7, 28–34 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Weinreb I Hyalinizing clear cell carcinoma of salivary gland: a review and update. Head Neck Pathol 7 Suppl 1, S20–29 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Chapman E et al. Molecular Profiling of Hyalinizing Clear Cell Carcinomas Revealed a Subset of Tumors Harboring a Novel EWSR1-CREM Fusion: Report of 3 Cases. Am J Surg Pathol 42, 1182–1189 (2018). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.