Abstract
Background/Aim: Adenoid cystic carcinoma (ACC) is an aggressive neoplasm even though it has low-grade histological appearance and slow growth. The aim of this study was to identify the immunohistochemical and molecular characteristics of ACC, as well as their correlation with the clinical course of patients. Patients and Methods: This is a retrospective multicenter analysis. We included 50 patients diagnosed with ACC in the head and neck between 2000 and 2021. The expression of MYB proto-oncogene transcription factor (MYB), neurotrophic tyrosine kinase receptor (NTRK), human epidermal receptor-2 (HER-2), and Ki-67 was examined through immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH). We also performed a clinical follow-up of the patients. Results: The median age of the patients was 58.5 years; moreover, 54% of the patients were male. Compared with female patients, male patients were at a higher risk of both recurrence and death. No HER-2-positive cases were revealed. MYB expression was positive in 28 (56%) cases. However, MYB expression did not significantly affect survival. NTRK expression was positive in eight (16%) cases. NTRK-positive patients had worse overall survival (OS) than NTRK-negative patients (p=0.0246). Additionally, the percentage of NTRK-stained cells was negatively correlated with disease-free survival (p=0.0016) and OS (p=0.0027). Conclusion: There was no correlation between MYB positivity and survival. Contrarily, NTRK-positive patients had worse survival, indicating that NTRK is a negative prognostic factor. Tropomyosin receptor kinase inhibitors can be used to treat these patients. Furthermore, MYB-targeted inhibitors are promising therapeutic agents.
Adenoid cystic carcinoma (ACC) accounts for approximately 1% of all head and neck malignancies (1), as well as 10% of all salivary gland tumors (2). It can also originate in regions with secretory glands, including the nasal cavity, paranasal sinuses, ceruminous glands, larynx, and trachea (1). ACC generally appears in the fifth and sixth decades of life and affects women more commonly than men (3).
The three histological subtypes of ACC are cribriform, tubular, and solid (4). A solid growth pattern is associated with an advanced stage at diagnosis and poor prognosis (5). These subtypes can be observed in different proportions within a single tumor (6).
Despite its low-grade histological appearance and slow growth, ACC has a persistent disease course (7). It is characterized by perineural invasion and local recurrence (1). Although regional lymph node metastasis is rare, distant metastases frequently occur (8). The treatment of choice involves radical surgical resection and postoperative radiotherapy, with chemotherapy being associated with a low response rate (9).
The translocation between MYB proto-oncogene transcription factor (MYB) and nuclear factor I/B (NFIB) is considered the molecular hallmark of ACC (10). The MYB-NFIB gene fusion causes MYB protein over-expression, which increases cell proliferation. Furthermore, ACC can produce high levels of human epidermal receptor-2 (HER-2) (11), as well as the Ki-67 protein (12), which is a nuclear antigen in proliferating cells. Furthermore, neurotrophic tyrosine kinase receptor (NTRK) gene fusions are oncogenic drivers (13) that have been identified in ACC specimens (14).
We aimed to investigate the immunohistochemical and molecular characteristics of ACC, as well as their correlation with the clinical course of patients. Among the features examined was NTRK expression and its effect on survival. To our knowledge, few studies have examined NTRK in patients with ACC (14, 15). Elucidating the molecular basis underlying ACC could facilitate the discovery of novel prognostic factors and the development of novel targeted therapies (16).
Patients and Methods
This retrospective multicenter study was conducted in four tertiary referral centers. This study was approved by the local institutional review board. We included patients diagnosed with ACC in the head and neck between 2000 and 2021. We collected data regarding the following characteristics: primary tumor site, age at diagnosis, sex, clinical and pathological tumor characteristics, cancer staging, treatment modalities, time to progression, and survival. A uniform therapy was offered in all participating referral centers. The patients were classified according to two histopathological grading systems based on the degree of the solid pattern. The one grading system was described by Perzin et al. (17) and Szanto et al. (18) and the other by Spiro et al. (19) We will refer to these grading systems as Perzin/Szanto and Spiro, respectively. All cases were reviewed by two pathologists (IP and NK).
MYB immunohistochemistry (IHC). Rabbit monoclonal antibody (Abcam, Cambridge, MA, USA, ab45150, clone EP769Y, 1:250 dilution) was used for MYB IHC (20). Only nuclear staining was evaluated, where we evaluated the intensity and proportion of staining. The staining intensity was defined as none (no labeling), weak (labeled structures were barely colored), weak/medium (labeled structures were clearly but lightly labeled), medium (labeling was obvious but light passed through labeled area easily), medium/strong (labeled areas passed a small amount of light), or strong (labeled areas were essentially opaque to light) (21). The proportion of stained cells was determined.
MYB fluorescence in situ hybridization (FISH). An MYB dual-color break-apart probe (ZTV-Z 2143-200; ZytoVision GmbH, Bremerhaven, Germany) was used to detect translocations involving MYB using FISH (22). Slides were evaluated through fluorescence microscopy.
HER-2 IHC. HER-2 protein over-expression was estimated using IHC (23). The HerceptTest™ kit (polyclonal antibody; 1:800 dilution; Dako, Carpinteria, CA, USA) was used for analysis. Only membrane staining was evaluated. HER-2 was scored as 0 (negative), 1+ (negative), 2+ (weakly positive), and 3+ (positive).
HER-2 FISH. A ZytoLight SPEC HER2/CEN17 dual-color probe kit (ZytoVision GmbH) was used to detect HER-2 gene amplification using FISH (23). Slides were evaluated through fluorescence microscopy.
Ki-67 IHC. The Ki-67 cellular proliferation index was identified using IHC (24). This was determined using a monoclonal mouse anti-human Ki-67 antigen (MIB-1; 1:200 dilution; Dako).
NTRK IHC. A pan-Trk monoclonal antibody clone EPR17341 (Abcam) was used for IHC staining of NTRK (25). This antibody recognizes the C-terminal domains of the Trk A, B, and C proteins. The antibody concentration was 6 μg/ml. Staining intensity was evaluated as negative, weak, moderate, or strong. Additionally, the staining location (membranous, cytoplasmic, nuclear, or mixed) and extent (the percentage of NTRK-stained cells) were determined. A sample was considered positive if ≥1% of the tumor cells exhibited NTRK staining at any intensity (26).
Statistical analysis. Statistical analyses were performed using the SAS software (version 9.4; SAS Institute Inc., Cary, NC, USA) (27). Descriptive values are expressed as medians (Q1-Q3) and mean±standard deviation (SD). Between-group comparisons of qualitative parameters were performed using the chi-squared test. For survival analysis, we used Kaplan–Meier curves. The significance level (p-value) was set at <0.05.
Results
We included 50 patients (median age: 58.5 years); among them, 27 (54%) were male and 45 (90%) patients were primary cases. The parotid gland was the most commonly affected site (34%). Furthermore, 36 (72%) patients were in the early stage, whereas 27 (54%) patients had a T2 primary tumor size based on the AJCC on Cancer. At diagnosis, three (6%) patients had cervical lymph node metastases, whereas none had distant metastases. Perineural infiltration was identified in 24 (48%) patients. All patients underwent primary tumor surgery. Moreover, 86% and 32% of the patients received radiotherapy and chemotherapy, respectively. Five (10%) patients underwent selective neck dissection. Table I presents the baseline characteristics.
Baseline characteristics of the study population.
Male patients with ACC had a higher risk of both recurrence and death than female patients. The mean time to progression in female and male patients was 93.8 (SE: 9.0) and 54.7 (SE: 6.4) months, respectively (p=0.0313). Additionally, the mean time to death in female and male patients was 124.9 (SE: 11.9) and 79.3 (SE: 9.9) months, respectively (p=0.0322).
There were significant differences in disease-free survival (DFS) and overall survival (OS) according to the tumor component (T) (DFS, p=0.0117; OS, p=0.0273) but not the node component (N). Regarding cancer stage, the early-stage group T1+2 had better DFS and OS [mean: 68.4 months (SE: 4.8) and mean: 118.4 months (SE: 10.1), respectively] than the advanced-stage group T3+4 [mean: 60.9 months (SE: 12.5), p=0.0214, and mean: 57.8 months (SE: 6.0), p=0.0390, respectively].
Neck dissection did not significantly influence survival. Moreover, only seven (14%) patients were managed without radiotherapy. Patients who received chemotherapy had worse DFS than patients who did not receive chemotherapy [43.4 months (SE: 5.2) vs. 90.3 months (SE: 8.0), p=0.0371]; however, there was no difference in OS [92.2 months (SE: 19.9) vs. 100.4 months (SE: 9.1), p=0.3481].
The mean time to recurrence and death were 80 and 104 months, respectively. The 5- and 10-year DFS rates were 50% and 27.78%, respectively, while the 5- and 10-year OS rates were 67.50% and 45.71%, respectively.
The histological results were as follows: the Ki-67 percentage mean value was 14%±14% (min: <1%, max: 74%, median: 10%, Q1-Q3: 4%-20%). No case was HER-2 positive upon evaluation using both IHC and FISH. The mean MYB percentage was 20.65%±27.70% (min: 0%, max: 95%, median: 5%, Q1-Q3: 0%-35%). MYB expression was positive in 28 (56%) cases. Table II shows the results based on staining intensity. The cellular homolog (C)-MYB, which was evaluated using FISH, was amplified in 28 (56%) cases. NTRK expression, which was evaluated by IHC, was positive in eight (16%) cases. The staining intensity was strong/cytoplasmic in one (2%) case and weak/cytoplasmic in 11 (22%) cases. Four patients with weak cytoplasmic intensity were NTRK-negative. There was no staining intensity in 38 (76%) patients. The mean percentage of NTRK-stained cells was 3.29% ± 8.51% (min: 0%, max: 40%, median: 0%, Q1-Q3: 0%-0.5%). The NTRK percentage was <1% in 40 (80%) patients.
MYB intensity.
The patients were staged according to Perzin/Szanto and Spiro histopathological grading systems (Table III). Analysis using the Perzin/Szanto system indicated a positive correlation of Ki-67 expression with its grading (p=0.015, Figure 1). Similarly, the Spiro system revealed a correlation between its grading and Ki-67 expression; specifically, grade I tumors had a lower Ki-67 percentage than tumors of other grades (p=0.024, Figure 1). In the Spiro system, grade III tumors showed a non-significant higher recurrence rate than tumors of other grades (p=0.063).
Differentiation according to Perzin/Szanto and Spiro systems.
Grade I of Spiro system has lower Ki67 percentage (p=0.024). Ki67 expression is positively correlated to Perzin/Szanto system (p=0.015).
Age was marginally higher in MYB-positive patients than in MYB-negative patients (p=0.090). Examination of C-MYB through FISH revealed increased perineural infiltration upon C-MYB amplification (95%CI=1.02-10.72, p=0.0423). Analysis of MYB distribution at various sites revealed that ACC in minor salivary glands showed a higher likelihood of being MYB-positive than ACC in other sites (p=0.0196). Moreover, the MYB staining intensity differed according to the primary tumor site. Specifically, ACCs in major salivary glands were more likely to show weak/medium intensity, while ACCs in the remaining locations were more likely to show medium/strong intensity (95%CI=0.02-0.75, p=0.0238). The ACCs of the parotid and submandibular glands mostly showed weak/medium intensity, whereas the nasal cavity and paranasal sinuses mostly showed medium/strong intensity (p=0.0036). Additionally, patients with weak/medium staining intensity for MYB showed a non-significantly higher 5-year OS rate than those with medium/strong intensity (84.62% vs. 50%, 95%CI=0.78-38.7, p=0.0737).
Male patients showed a higher rate of NTRK positivity than female patients (95%CI=0.87-68.19, p=0.0381). Additionally, NTRK expression could affect DFS (p=0.0807). The median DFS in NTRK-positive and NTRK-negative patients was 24 (Q1-Q3: 13.5-59.5) and 53 (Q1-Q3: 28-108) months, respectively. Furthermore, the 5-year DFS rate was higher in NTRK-negative patients (51.52%) than that in NTRK-positive patients (33.33%) (95%CI=0.08-2.93, p=0.4124). The 10-year DFS rates in NTRK-positive and NTRK-negative patients were 0% and 34.48%, respectively, with no significant difference (p=0.0888). Moreover, the NTRK-negative patients showed higher 5-year (73% vs. 33.33%) and 10-year OS rates (50% vs. 16.67%) than NTRK-positive patients (95%CI=0.03-1.21, p=0.0597 and 95%CI=0.02-1.94, p=0.1356, respectively). Generally, NTRK-positive patients had worse OS than NTRK-negative patients [mean OS: 49.4 months (SE: 9.9) vs. 110.8 months (SE: 9.2); p=0.0246] (Figure 2). The percentage of NTRK-stained cells was negatively correlated with both DFS (rs=−0.44, p=0.0016) and OS (rs=−0.42, p=0.0027). Notably, none of our patients were treated with tropomyosin receptor kinase (TRK) inhibitors.
NTRK-positive patients had worst overall survival than NTRK-negative patients (p=0.0246).
Discussion
ACC is most commonly reported in the fifth and sixth decades of life (28). The median age at diagnosis of our patients was 58.5 years. In our study, 54% of the patients were male, which is inconsistent with previous reports that ACC occurs more commonly in women (9, 28). Compared with female patients, male patients had a higher risk of both recurrence and death (p=0.0313 and p=0.0322, respectively), which is consistent with previous reports that male sex is a poor prognostic indicator (29, 30).
ACC can occur in any glandular body tissue (4). In our study, the major salivary glands were the most affected primary sites (52%). Previous studies have reported lower proportions of ACC in the major salivary glands (29%-47%) (31, 32). Another common site was the sinonasal cavity (22%), with previous studies reporting lower proportions (8%-19%) (33, 34). In our study, ACC occurrence in the oral cavity was rare (8%). This is inconsistent with previous reports that the oral cavity, which has the highest concentration of minor salivary glands (35), is among the most common occurrence sites (36). This inconsistency could be attributed to the fact that patients with ACC in the oral cavity are often referred to maxillofacial surgeons and our study did not involve the maxillofacial department.
The specimens were classified according to Perzin/Szanto and Spiro pathological grading systems (2). There are considerable differences in the grading of the two systems, which indicates the need for a novel pathological scoring system without disparities in histopathological classification (37, 38).
As expected, patients with early-stage ACC had significantly better DFS and OS than patients with late-stage ACC (p=0.0214 and p=0.0390, respectively). Although ACC presents with a slow clinical course, our findings demonstrate the importance of early diagnosis.
Three patients had cervical lymph node metastases, and none had distant metastases at diagnosis. There was no difference in the DFS and OS according to cervical lymph node metastasis. This result could be attributed to the small sample size of patients with metastatic disease. A study conducted by Jang et al. (39) suggested that nodal involvement has no impact on survival. Contrarily, a study by Oplatek et al. (40) indicated that mean OS and time to recurrence were lower for patients with cervical lymph node positive disease.
All patients underwent surgery for primary cancer, with 84% of patients being treated with a multimodal approach. Moreover, 86% and 32% of patients underwent radiotherapy and chemotherapy, respectively. Patients who received chemotherapy had worse DFS (due to advanced-stage disease at diagnosis) than those managed without chemotherapy (p=0.0371); however, there was no significant difference in OS (p=0.3481). This is consistent with the fact that surgery and radiotherapy are the primary treatment modalities (1), while the treatment utility of chemotherapy for ACC remains unclear.
Five (10%) patients underwent selective neck dissection; among them, two patients had a node component of N0. Neck dissection did not significantly affect DFS or OS, which could be attributed to the small number of patients who underwent this procedure. Amit et al. (41) reported no significant difference in survival between N0 patients who underwent selective neck dissection and those who did not. Xiao et al. (42) recommended selective neck dissection for advance-stage N0 patients since it improved the OS.
ACC has a slow but persistent clinical course (28, 39). The mean time to recurrence and death was 80 and 104 months, respectively. The 5- and 10-year DFS rates were 50% and 27.78%, respectively, whereas the 5- and 10-year OS rates were 67.50% and 45.71%, respectively. Therefore, long-term, practically life-long, follow-up of patients with ACC is fundamental (43).
Ki-67 is a proliferation marker for tumor cells (44). In our study, Ki-67 expression was associated with the tumor grading as described by Perzin/Szanto and Spiro systems. Specifically, the Ki-67 percentage was positively correlated with the grade in both systems (p=0.015 and p=0.024, respectively). Nordgard et al. reported that Ki-67 expression is an independent prognostic factor for ACC (45).
All our included patients were negative for HER-2 expression. There are inconsistent reports regarding HER-2 expression in ACCs, with the proportion considerably varying from no over-expression to 5%, 16%, >50%, and 100% (46). A meta-analysis conducted by Egebjerg et al. (47) indicated that the prevalence of HER-2 positivity was 0.15%.
ACC is characterized by a translocation involving the oncogene MYB and the transcription factor NFIB (48). In our study, 56% of the patients were MYB-positive. MYB expression was not significantly correlated with DFS or OS, which is consistent with a previous report by Liu et al. where no difference in survival between MYB-positive and-negative patients was reported (49). Contrastingly, Broz et al. reported that MYB-negative patients were likely to have a better prognosis (50). In our study, C-MYB amplification was observed in 56% of the patients. IHC can stain ACCs with and without MYB-NFIB translocation, while FISH can only detect ACCs with MYB-NFIB translocation (51, 52). In our study, IHC showed identical sensitivity to FISH.
We evaluated MYB expression at various tumor locations. ACCs in the minor salivary glands showed a higher frequency of MYB positivity than those in other sites (p=0.0196), which is consistent with a previous report by Rettig et al. (53).
Perineural infiltration was positively correlated with C-MYB amplification (95%CI=1.02-10.72, p=0.0423); however, there have been contrasting reports. West et al. reported a propensity for perineural infiltration in patients with MYB-NFIB translocation (51), whereas another study on 158 patients with ACC indicated that perineural invasion was not associated with MYB-NFIB status (53). Broz et al. (50) reached a similar conclusion.
In our study, only eight (16%) patients were NTRK-positive. Ivanov et al. reported high NTRK3 expression in 17 of 18 ACC primary tumor specimens (15). Other studies have reported high frequencies (>90%) of NTRK fusions in rare cancer types, including mammary analogue secretory carcinoma and infantile fibrosarcoma, and lower frequencies (<1%) in other tumor types (54). In our study, male patients showed a higher frequency of NTRK positivity than female patients (95%CI=0.87-68.19, p=0.0381).
NTRK expression could be associated with survival. NTRK-positive patients had worse OS than NTRK-negative patients [mean OS: 49.4 months (SE: 9.9) vs. 110.8 months (SE: 9.2); p=0.0246]. The percentage of NTRK-positive cells was negatively correlated with both DFS (rs=−0.44, p=0.0016) and OS (rs=−0.42, p=0.0027). This finding has clinical application since NTRK can be used as a negative prognostic marker for ACC. A study on 28 patients with NTRK-positive solid tumors reported that NTRK fusions could be negative prognostic factors (54). Finally, TRK inhibitors can be used to treat NTRK-positive patients. Numerous studies have confirmed the significant antitumor activity of TRK inhibitors, with acceptable toxicity (55-57), with patients showing a remarkable and durable therapeutic response (55).
An important limitation of our study is the small number of ACC patients. This entity is rare and, in certain subgroups, only single cases (e.g., in the trachea and larynx) were included in the study. This makes the statistical power low. Furthermore, the number of NTRK-positive patients was limited. Studies with larger numbers of NTRK-positive patients are required in order to confirm the role of NTRK as a negative prognostic factor for ACC. Another limitation is that so far none of our NTRK-positive patients is treated with TRK inhibitors. This therapy can be used in our NTRK-positive patients with inoperable recurrences. Research is needed in order to explore the therapeutic effects of TRK inhibitors for the treatment of ACC associated with NTRK activation.
In conclusion, 56% and 16% of patients were MYB- and NTRK-positive, respectively. MYB positivity did not influence survival, indicating that MYB is not a good prognostic marker for ACC. Contrarily, NTRK-positive patients had worse OS than NTRK-negative patients; therefore, NTRK can be considered a negative prognostic factor. Further research on targeted molecular therapies, such as TRK and MYB inhibitors, could lead to the development of alternative effective treatment for patients with advanced ACC.
Acknowledgements
The Authors are grateful to the patients and their families for their willingness to participate in this study. The Authors also thank all members of the Department of Pathology of “Attikon” University Hospital for their contribution.
Footnotes
Authors’ Contributions
AD, IP, AP, IG and PM designed the study strategy and supervised the study and analysis. IP, CG, MT, GK, CK, PK, PA, HS, PE, and TF provided the pathological specimens and study material. TP, DL, NK, AP, and IP performed the analysis and interpretation of the results. TP, DL, NK, and AP wrote the manuscript. AD and IP corrected and approved the final manuscript. All Authors discussed the results and implications and reviewed the manuscript.
Conflicts of Interest
The Authors have no conflicts of interest to declare in relation to this study.
- Received January 30, 2023.
- Revision received February 13, 2023.
- Accepted February 14, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.