Adenoid cystic carcinoma (ACC), commonly from salivary glands, is known for its insidious local growth and usually protracted clinical course. ACC developing from non-salivary glands (i.e., non-salivary ACC) is heterogeneous, and its clinicopathological features remain poorly defined. Patients treated for ACC in a single institution between 1995 and 2007 were included in this study. Immunohistochemical evaluation of Ki-67, E-cadherin, p16, and cyclinD1 was performed. The prognostic significance of clinical and immunophenotypic markers was evaluated. 83 cases of salivary ACC and 24 cases of non-salivary ACC were included. The expression levels of Ki-67 (54.8%), E-cadherin (90.4%), p16 (32.9%), and cyclinD1 (19.2%) between ACCs present at various sites were not different. Sinonasal, lacrimal, and tracheobronchial ACCs had significantly worse outcomes than those of ACC of the major salivary glands. Postoperative radiotherapy reduced the recurrence rate of patients with a negative resection margin ( P = 0.028). Older age (age >60 years), advanced stage, positive resection margin, high histological grade, and high expression of Ki-67 were significantly correlated with poor prognosis. In conclusion, the site of origin plays a role in the prognosis of ACC, in which positive resection margin and advanced stage are possible factors underlying the differences in outcomes.
Adenoid cystic carcinoma (ACC) is a rare malignancy, accounting for less than 5% of all head and neck cancer. ACC is known for its insidious local growth and usually protracted clinical course; it usually manifests as a localized disease at diagnosis. Radical surgery plus postoperative radiotherapy (RT) is the mainstay of treatment for ACC. Despite aggressive treatment, the recurrence rate remains high. The 5- and 10-year recurrence-free survival rates are 50–65% and 45–52%, respectively, and the 5- and 10-year overall survival rates are 70–72% and 44–54%, respectively.
ACC arises within secretory glands, most commonly the major and minor salivary glands of the head and neck. Some studies have suggested that ACC originating from minor salivary glands has a poorer prognosis than that associated with ACC originating from the major salivary glands, implying that the site of origin has prognostic significance. ACC can originate from sites other than the salivary glands, such as the lacrimal gland, external ear, tracheobronchial tree, breast, and vulva, and such ACC are called non-salivary ACC. Non-salivary ACC are rare, so the clinicopathological features and treatment outcomes are less defined.
Biological markers have potential value for ACC evaluation. Ki-67, a marker of proliferation, is present during all active phases of the cell cycle, but is absent in resting cells. High Ki-67 expression was associated with poor ACC prognosis. E-cadherin is a transmembrane protein which plays an important role in cell adhesion. Loss of E-cadherin expression was correlated with cancer metastasis and was related to poor ACC prognosis. p16 and cyclinD1 are cell cycle regulatory proteins; p16 inhibits cell cycle progression, whereas cyclinD1 promotes it. Alternation of P16 and cyclinD1 expression was associated with poor prognosis for head and neck cancer. In cases of ACC, cyclinD1 expression correlated with high histological tumor grade and high proliferative index in a small series.
In this study, the authors retrospectively reviewed cases of ACC treated at the National Taiwan University Hospital in Taipei, Taiwan, from 1995 to 2007, and compared clinicopathological features, treatment modalities, patterns of failure, and treatment outcomes among patients with ACCs originating from various sites. Immunohistochemical staining for Ki-67, E-cadherin, p16, and cyclinD1 was performed and correlated with the clinical parameters.
Materials and methods
Patients and clinical characteristics
Patients who were newly diagnosed with ACC of the head and neck and treated at National Taiwan University Hospital from 1995 to 2007 were selected from the Cancer Registry of National Taiwan University Hospital and were included in this study. ACC originating from salivary glands or secretory glands contributing to salivation within or near the oral cavity is defined as salivary ACC. ACC from secretory glands not contributing to salivation beyond the oral cavity of the head and neck is defined as non-salivary ACC.
Clinicopathological features, including age, gender, primary tumor site, stage, pathological features, treatment modality, site of recurrence, disease status, and survival status, were assessed. Each pathological specimen was reviewed under 10 high-power fields by two experienced pathologists (KCC and KTK) and the diagnoses were agreed. Histologically, ACC can be categorized into three growth patterns: cribriform, tubular, and solid. The solid growth pattern has been reported to be associated with worse outcome. Histological grading of ACC in the present study was according to Szanto et al. ; the most widely accepted grading system. Grade I, tumors with tubular and cribriform areas but without solid components; Grade II, cribriform tumors that were either pure or mixed with less than 30% of solid areas; and Grade III, tumors with a predominantly (more than 30%) solid pattern. The positive resection margin was defined as cancer cell involvement at the edge of the specimen, while the negative resection margin was defined as the edge of the specimen free from cancer cell involvement, including a clear margin (the closest distance between cancer cell and edge of the specimen ≧1 mm) and close margin (the closest distance between cancer cell and edge of the specimen <1 mm). Radiography of the chest, abdominal sonography, bone scan, and computed tomography (CT) were performed for evaluation of disease extent. Staging was performed on the basis of the American Joint Committee on Cancer (AJCC) staging system, 7th edition.
Study of molecular markers
For immunohistochemical staining, a 5 mm thick section of the tumor tissue was deparaffinized and rehydrated, and then antigen retrieval was performed. After cooling for 20 min at room temperature, the decanted retrieval solution was washed 2–3 times with phosphate-buffered saline (PBS) at room temperature. The antibodies and detection kits for Ki-67, E-cadherin, p16, and cyclinD1 were used serially as previously described.
For calculating Ki-67 expression levels, the percentage of positive neoplastic cells was determined. The level of E-cadherin expression was determined according to the method described by Maruya et al. The immunohistochemical staining results were classified into the following 4 grades: negative (no positive tumor cells), 1+ (percentage of positive tumor cells <15%), 2+ (percentage of positive tumor cells 15–50%), and 3+ (percentage of positive tumor cells >50%). The criteria for p16 and cyclinD1 scoring were adopted from the study by Higuchi et al. The p16 staining was considered positive if more than 1% tumor cells demonstrated nuclear p16 staining. The cyclinD1 staining results were classified into the following 4 grades: 0+ (percentage of positive tumor cells 0–5%), 1+ (percentage of positive tumor cells 5–20%), 2+ (percentage of positive tumor cells 20–50%), 3+ (percentage of positive tumor cells 50–100%).
Event-free survival (EFS) was calculated from the start of initial treatment to the time of disease progression, relapse, or death. Overall survival (OS) was calculated from the start of initial treatment to the time of all-cause death. Survival was estimated using the Kaplan–Meier method and compared using the log-rank test. The following variables and categories were used for Cox-regression univariate and multivariate analyses: age, gender, primary tumor site, disease stage, resection margin, pathological grade, and biological markers, including Ki-67, E-cadherin, p16, and cyclinD1. Categorical variables were compared using χ 2 or Fisher’s exact test. Differences were considered significant if the P value obtained in the two-sided test was less than 0.05. The SPSS software (Windows version 18.0) was used for statistical analysis.
101 patients, with a median age of 49 years (range 16–86 years), met the inclusion criteria ( Table 1 ). There were 44 male and 57 female patients. 32 patients had ACC of the major salivary gland, 51 of the minor salivary gland, and 18 of non-salivary glands. Minor salivary gland ACC included intraoral and sinonasal ACC, and non-salivary ACC was composed of ear, lacrimal, and tracheobronchial ACC. Histological grade 1 (61.4%), perineural invasion (70.3%), and positive resection margin (62.4%) dominated the pathological findings of specimens. There were more advanced stage ACC (stage III or IV, 55.5%) than early stage ACC (stage I or II, 44.6%). The representative microscopic pictures of positive and negative resection margins are shown in Figs. 1 and 2 , respectively.
|Total patient number||101|
|Male (%)||44 (43.5%)|
|Median age (range)||49 (16–86)|
|Positive/negative a /NA||63/30/8|
|Classification of ACC||Site||No. of patients|
|Major salivary gland (32 patients)||Parotid gland||21|
|Minor salivary gland (51 patients)||Intraoral (26 patients)||Tongue||5|
|Sinonasal (25 patients)||Sinonasal||25|
|Non-salivary glands (18 patients)||Ear||2|