Printer Friendly

High Expression of H3K9me3 Is a Strong Predictor of Poor Survival in Patients With Salivary Adenoid Cystic Carcinoma.

Salivary adenoid cystic carcinoma (ACC) is one of the most common malignancies of salivary glands in eastern China populations, comprising 9.75% of all salivary gland neoplasms and 30.42% of all salivary gland malignancies. (1) Patients with ACC have a relatively high 5-year survival rate, but their longer-term prognosis is less optimistic as a result of tumor recurrence after treatment and distant metastasis to sites that include the lung, bone, and liver. (2,3) The precise evaluation of the prognosis of patients with ACC might help to determine the most appropriate treatment strategies, thereby improving their survival and quality of life. Therefore, the discovery of accurate prognostic biomarkers is an important area of ACC research.

In addition to genetic alterations, epigenetic modifications contribute to the development and progression of many malignant tumors. (4-6) Alterations in DNA methylation, histone modifications, and microRNA expression (7,8) can change gene expression without changing the DNA sequence. Epigenetic alterations may influence several aspects of tumor cell biology, including cell proliferation, differentiation, and apoptosis, and might be linked to the clinical outcomes of patients. (9) Many studies have focused on aberrant DNA methylation in cancer, and our previous studies demonstrated that hypermethylation of the promoter regions of CDKN2A, RASSF1A, and DAPK commonly occurs in ACC; furthermore, E-cadherin is frequently inactivated in ACC tumors through promoter hypermethy lation. (10,11)

In comparison to DNA methylation, histone modifications are less well characterized in ACC. However, histone modifications appear to play important roles in organizing the nuclear architecture, with consequent effects on the regulation of gene transcription. (4,12) Histone H3 lysine 9 trimethylation (H3K9me3) is a histone modification that occurs in the N-terminal of lysine residues. It has been correlated with tumor suppressor gene silencing in colorectal cancer, (13) hepatocellular cancer, (14) and breast cancer. (15) Histone acetylation is an important epigenetic mechanism that regulates gene transcription through altering the chromatin structure. (16) Specific marks, such as lysine acetylation (H3K9Ac, H3K18Ac, and H4K12Ac), are associated with activating gene promoters. (17,18)

The prognostic value of histone modifications remains controversial. High levels of H3K9me3 have been correlated with a poor survival rate in gastric adenocarcinoma, (19) but a decrease in H3K9me3 expression is associated with a poor prognosis of patients with non-small cell lung cancer. Low expression levels of H3K9Ac have also been correlated with poor survival outcomes in non-small cell lung cancer. (20) Together, H3K9me3 and H3K9Ac expression could be potential prognostic biomarkers for malignant tumors. However, whether the expression of H3K9me3 and H3K9Ac is correlated with prognosis in patients with ACC is not known. Therefore, we determined H3K9me3 and H3K9Ac expression levels in 66 ACC biopsy specimens in a Chinese population and correlated these histone modifica tions with both clinicopathologic parameters and patient outcomes.



Tumor samples of 66 patients with ACC were included in the present study. Patients underwent radical surgery without either preoperative chemotherapy or radiotherapy at the Department of Oral and Maxillofacial Surgery, 9th People's Hospital, Shanghai Jiao Tong University (Shanghai, China) between 1999 and 2006. At 3 to 4 weeks postoperatively, adjuvant radiotherapy was administered to 51 patients with megavoltage equipment using cobalt 60. For patients with positive surgical margins, a radiation dosage of 66 Gy was given, compared to 60 Gy for patients with negative surgical margins. Thirteen patients received concomitant chemotherapy using a combination of either cisplatin and 5-fluorouracil or cisplatin, 5-fluorouracil, and adriamycin. Clinicopathologic parameters, including patient age, sex, tumor site, histologic pattern, perineural invasion, surgical margin status, TNM stage at the time of surgery, and follow-up information, such as the presence of recurrence and distant metastasis, were obtained from pathologic reports and medical records. According to the histologic typing of salivary gland tumors by the World Health Organization, (21) ACC is classified into cribriform, tubular, and solid patterns.

The histologic pattern was determined in this study according to the following criteria--cribriform-tubular pattern: no solid component or solid component less than 30%; solid pattern: solid component of 30% or greater. The TNM stage of each case was determined according to the American Joint Committee on Cancer staging system. (22) The study was approved by the Shanghai 9th People's Hospital Ethics Committee and informed consent was obtained from each patient.


We used specific rabbit polyclonal antibodies (all from Abcam Inc, Cambridge, United Kingdom) to detect H3K9me3 and H3K9Ac expression in ACC specimens. The appropriate dilutions and incubation times had been previously determined. After deparaffinization through a graded series of xylene, the sections were rehydrated in a graded series of alcohol. Heat-mediated antigen retrieval using 0.01 M sodium citrate buffer (pH 6.0) was performed and endogenous peroxidase was quenched with 3% hydrogen peroxide for 20 minutes at room temperature. The sections were then incubated with normal goat serum to reduce nonspecific binding. The sections were incubated with rabbit polyclonal H3K9me3 antibody (1:300) and H3K9Ac antibody (1:200) at 4[degrees]C overnight. Detection was accomplished with the Dako EnVision System (Carpinteria, California), and the expression of both H3K9me3 and H3K9Ac was detected with the diamino benzidine chromogen system (Dako). The sections were counterstained with hematoxylin and then dehydrated and mounted. Blank controls were whole-tissue sections stained in the absence of the primary antibody.

Evaluation of Immunohistochemical Staining

All the immunohistochemical staining was evaluated by 2 independent pathologists (W.L. and H.Y.) who were blinded to the clinical data. In the case of discordance, both observers defined the final score of the sample through discussion. The score of each section was based on the percentage and intensity of positively stained cells. The percentage score was as follows: no positive cells (0), fewer than 25% positive tumor cells (1), 25% to 50% positive tumor cells (2), 51% to 75% positive tumor cells (3), and greater than 75% positive tumor cells (4). The intensity score was as follows: no staining (0), light yellow (1), yellow-brown (2), and dark brown (3). The sections were graded into 2 groups by multiplying the percentage and intensity scores, as follows: low expression (score <6) and high expression (score [greater than or equal to]6).

Statistical Analysis

Statistical analysis was performed with SPSS 17.0 software (SPSS Inc, Chicago, Illinois). The [chi square] test was used to determine the relationship between the expression of H3K9me3 or H3K9Ac and clinicopathologic parameters. Survival curves were estimated by using the Kaplan-Meier method, and the significance of differences in the overall survival (OS) and disease-free survival (DFS) rates was tested by the log-rank test. The variables with a significant difference (P [less than or equal to] .05) according to the log-rank test were subjected to a multivariate analysis by using the Cox proportional hazards regression model. The Enter method was then used to determine a final Cox model. All of the tests were 2-sided and P [less than or equal to] .05 was considered to be statistically significant.


Clinicopathologic Characteristics

Detailed clinicopathologic characteristics and follow-up information for each patient are presented in the Supplementary S1 file (see supplemental material file at http:// in the December 2013 table of contents). Of the 66 patients with ACC, 36 (54.55%) were men and 30 (45.45%) were women, ranging in age from 16 to 82 years (mean, 53.02 years). Twenty-seven tumors (40.91%) originated from the major salivary glands and 39 tumors (59.09%) originated from the minor salivary glands. Histologically, 49 cases (74.24%) had a cribriform-tubular pattern and 17 (25.76%) had a solid pattern. Two samples were excluded from the analysis of the correlation between distant metastasis and survival as the disease had metastasized to the lung before surgery in these patients. Perineural invasion, positive surgical margins, local recurrence, and distant metastasis were observed in 50 (75.76%), 21 (31.82%), 21 (31.82%), and 34 (51.52%) patients, respectively. The mean follow-up duration was 77.24 months (range, 6-132 months).

Correlation Between H3K9me3, H3K9Ac Expression, and Clinicopathologic Parameters

The expression of H3K9me3 (Figure 1A, 1B, 1C, and 1D) and H3K9Ac (Figure 2A, 2B, 2C, and 2D) was manifested as nuclear staining. The correlations between the expression levels of H3K9me3 and H3K9Ac expression with clinicopathologic parameters are presented in Table 1. In particular, tumors with a solid pattern had higher levels of H3K9me3 expression than those with a cribriform-tubular pattern (P = .002), while tumors with a cribriform-tubular pattern had higher levels of H3K9Ac than solid pattern tumors (P = .03) (Figure 3A, 3B, 3C, and 3D). As adenoid cystic carcinoma is a biphasic tumor that consists of 2 main cell types, myoepithelial-like cells and ductal cells, we analyzed whether H3K9me3 and H3K9Ac expression patterns correlated to the tumor cell types. We found that myoepithelial cells and ductal cells did not represent different H3K9me3 and H3K9Ac expression patterns. Of the 34 patients who had distant metastasis, 26 (76.5%) had high levels of H3K9me3, whereas only 11 of 32 patients (34.4%) who had no distant metastasis had high levels of H3K9me3 expression; high levels of H3K9me3 expression therefore correlated with distant metastasis (P = .001). There was no statistically significant correlation between the expression level of H3K9me3 or H3K9Ac and other clinicopathologic parameters.

Correlation Between H3K9me3, H3K9Ac Expression, and Survival

Patients whose tumors had high levels of H3K9me3 showed significantly poorer OS outcomes than those with low levels of H3K9me3 (P < .001) (Figure 4, A), while patients with high H3K9Ac expression had a significantly better OS than those with low expression levels (P = .05) (Figure 4, B). In terms of clinicopathologic variables, solid pattern tumors, M1 stage tumors, and distant metastasis were correlated with worse OS (P < .001, P = .004, and P = .001, respectively) (Figure 4, C through E). We also found that high H3K9me3 expression, solid pattern tumors, and positive surgical margins were correlated with poorer DFS rates (P < .001, P = .01, and P = .03, respectively) (Figure 4, F through H).

We then determined whether H3K9me3, H3K9Ac expression, and clinicopathologic parameters were correlated with patient outcomes by using the Cox proportional hazards regression model. In the univariate analysis, high H3K9me3 expression levels, solid pattern tumors, and M1 stage were associated with poor OS, and the H3K9me3 expression was the most striking factor. H3K9me3 expression, histologic pattern, and margin status were associated with poor DFS (Figure 5). To determine which variables were independent survival predictors, a multivariate analysis was performed and showed that H3K9me3 expression (P = .004) and histologic pattern (P = .04) were independent factors for the estimation of OS. H3K9me3 expression (P = .006) was the only independent factor for the estimation of DFS (Table 2).


Histones are subject to posttranslational covalent modifications, such as methylation and acetylation. These modifications can control the activity of the surrounding DNA. (23) The cellular levels of histone modifications, based on tissue specimens, are related to cancer prognosis in different tumor types. (24) Although the information about the histone alterations in ACC tumorigenesis and prognosis was limited, some previous studies (10,25) have reported that CDKN2A might be inactivated by promoter hypermethylation and loss of heterozygosity in ACC. CDKN2A is known to be an important tumor suppressor gene and one recent study (26) showed that CDKN2A could be reactivated by decreasing histone H3K9 methylation and increasing histone H3K9 acetylation level. Therefore, H3K9me3 and H3K9Ac might influence CDKN2A expression in ACC and hence the prognosis of the patients. In the current study, we used immunohistochemistry to evaluate H3K9me3 and H3K9Ac expression levels in ACC tissue specimens. High H3K9me3 and low H3K9Ac expression levels correlated with the solid tumor growth pattern of ACC, in which the tumor cells have a higher proliferative activity and are usually more aggressive. This could be because H3K9me3 and H3K9Ac might regulate the expression of cell proliferation-related genes, such as Cip1/p21, (26) ERK, (27) and Notchl. (28) Another phenomenon in this study was that myoepithelial and ductal tumor cells of ACC did not reveal different H3K9me3 and H3K9Ac expression patterns. We proposed that although myoepithelial and ductal cells represented different phenotypes, they probably still had a similar genotype since both of the cells originated from intercalated duct reserve cells of the salivary gland. We also found that high levels of H3K9me3 expression were significantly correlated with distant metastasis, which may be because H3K9me3 can suppress the expression of cell invasion and distant metastasis-related genes, such as E-cadherin, (29) BRMS1,30 and CXCR4. (31)

The prognostic value of H3K9me3 and H3K9Ac expression has been established in gastric adenocarcinoma, (19) acute myeloid leukemia, (32) lung cancer, (20) and ovarian epithelial tumors, (33) although it remains controversial. For example, high H3K9me3 expression levels have been correlated with poor prognosis in gastric adenocarcinoma, but decreased levels of H3K9me3 and H3K9Ac expression have been associated with poor prognosis in lung cancer. In the current study, we used log-rank test and Cox proportional hazards regression models to analyze the survival data. We found that high H3K9me3 expression indicated poor OS and DFS rates for patients with ACC. These findings were in accordance with the study by Park et al, (19) which showed that stronger H3K9me3 expression corresponded to a worse prognosis in gastric adenocarcinoma. Thus, the prognostic value of H3K9me3 expression might be different in different tumors. We also found that low H3K9Ac expression correlated with poor survival in ACC. Pasini et al (34) reported that histone H3 lysine 27 trimethylation (H3K27me3) and acetylation (H3K27Ac) acted as an antagonistic switch in the transcriptional regulation of target genes. Therefore, to explain our current results, we hypothesized that the methylation and acetylation of H3K9 might have the same antagonistic switch function in suppressing the expression of tumor suppressor genes. This might help to explain why high H3K9me3 and low H3K9Ac expression were correlated with poor survival.

We then analyzed whether the H3K9me3, H3K9Ac, and clinicopathologic parameters were risk factors for ACC. Although the univariate analysis showed that H3K9me3 expression, histologic pattern, and M stage were risk factors for predicting OS, and H3K9me3 expression, histologic pattern, and margin status could predict DFS, the multivariate analysis indicated that only H3K9me3 expression and histologic pattern could act as independent factors for estimating the OS. More importantly, H3K9me3 expression was the only independent factor that could estimate the DFS. This might have been because histone modifications can regulate the expression of certain tumor suppressor genes, such as p16, RlZl, (14) and BRCA1, (35) which leads to the inhibition of tumor cell apoptosis. In addition, histone modifications can also regulate tumor suppressor genes that are associated with cell proliferation and distant metastasis, as described above. Cell growth dysregulation, and the inhibition of apoptosis and distant metastasis, are features that might correlate with poor prognosis in some tumors. Therefore, we hypothesized that high levels of H3K9me3 expression increased tumor cell growth, invasion, and distant metastasis, which result in a poor prognosis in ACC.

With regard to clinicopathologic parameters, with the exception of the histologic pattern, our results revealed that sex, perineural invasion, tumor site, recurrence, and postoperative radiotherapy and chemotherapy had no significant prognostic value in predicting either OS or DFS. Postoperative radiation therapy for ACC was usually performed in most institutions. Recent studies (36,37) have indicated that surgery combined with postoperative treatment could achieve better local control and improve disease-free survival. However, the role of chemotherapy in ACC is controversial and no definitive protocols have become universally accepted. (3) In the current study, our analysis did not show significant relationship between radiation/chemotherapy and OS and DFS, which could be for 2 reasons. First, for radiation therapy, in other research studies, (35,36) which showed that radiation therapy had a positive effect on patients' survival, almost every patient received postoperative radiation therapy, but in our cohort, only 77.3% (51 of 66) received postoperative radiation therapy. Second, for the chemotherapy, in the current study, only 19.7% (13 of 66) of patients underwent postoperative treatment. The lower adjuvant therapy rate in our institution may explain why radiotherapy/chemotherapy was not correlated with OS and DFS.

One limitation of this study was the fact that analyzing histone modifications via immunohistochemistry cannot provide information regarding changes in the activity of specific tumor suppressor genes. Future studies will be needed to address whether H3K9me3 and H3K9Ac expression changes could lead to the silencing of specific tumor suppressor genes in ACC. Nevertheless, our current research detected the expression level of H3K9me3 and H3K9Ac in 66 ACC specimens and demonstrated that an assessment of those histone modifications, especially H3K9me3, might be useful for predicting the prognosis of patients with ACC.


Our findings highlighted that H3K9me3 and H3K9Ac expression was correlated with histologic pattern and the presence of distant metastasis in ACC. H3K9me3 was shown to be a striking independent factor that could predict OS and DFS for patients with ACC, compared to other clinicopathologic parameters. H3K9me3 staining may provide helpful prognostic information in addition to histopathologic evaluation.

This work was supported by grants from the National Natural Science Foundation of China, (81272976, 30872905, and 81072211), Research Fund for the Doctoral Program of Ministry of Education of China (20120073110085), and the Doctoral Innovation Foundation of Shanghai Jiao Tong University, School of Medicine (Shanghai, China) (BXJ201029). We gratefully acknowledge Li Mao, MD, from the Department of Oncology and Diagnostic Sciences, Dental School, University of Maryland, Baltimore, for his critical review of our manuscript, and James Mikkelson, PhD, from Shanghai American School for his editorial assistance.


(1.) Tian Z, Li L, Wang L, Hu Y, Li J. Salivary gland neoplasms in oral and maxillofacial regions: a 23-year retrospective study of 6982 cases in an eastern Chinese population. Int J Oral Maxillofac Surg. 2010; 39(3):235-242.

(2.) van der Wal JE, Becking AG, Snow GB, van der Waal I. Distant metastases of adenoid cystic carcinoma of the salivary glands and the value of diagnostic examinations during follow-up. Head Neck. 2002; 24(8):779-783.

(3.) Papaspyrou G, Hoch S, Rinaldo A, et al. Chemotherapy and targeted therapy in adenoid cystic carcinoma of the head and neck: a review. Head Neck. 2011; 33(6):905-911.

(4.) Esteller M. Cancer epigenomics: DNA methylomes and histone-modification maps. Nat Rev Genet. 2007; 8(4):286-298.

(5.) Lund AH, van Lohuizen M. Epigenetics and cancer. Genes Dev. 2004; 18(19):2315-2335.

(6.) Goel A, Boland CR. Epigenetics of colorectal cancer. Gastroenterology. 2012; 143(6):1442-1460.

(7.) Bhalla KN. Epigenetic and chromatin modifiers as targeted therapy of hematologic malignancies. J Clin Oncol. 2005; 23(17):3971-3993.

(8.) Esteller M. The necessity of a human epigenome project. Carcinogenesis. 2006; 27(6):1121-1125.

(9.) Jones PA, Baylin SB. The fundamental role of epigenetic events in cancer. Nat Rev Genet. 2002; 3(6):415-428.

(10.) Li J, El-Naggar A, Mao L. Promoter methylation of p16INK4a, RASSF1A, and DAPK is frequent in salivary adenoid cystic carcinoma. Cancer. 2005; 104(4): 771-776.

(11.) Zhang CY, Mao L, Li L, et al. Promoter methylation as a common mechanism for inactivating E-cadherin in human salivary gland adenoid cystic carcinoma. Cancer. 2007; 110(1):87-95.

(12.) Lim EJ, Lu TX, Blanchard C, Rothenberg ME. Epigenetic regulation of the IL-13-induced human eotaxin-3 gene by CREB-binding protein-mediated histone 3 acetylation. J Biol Chem. 2011; 286(15):13193-13204.

(13.) Kondo Y, Shen L, Issa JP. Critical role of histone methylation in tumor suppressor gene silencing in colorectal cancer. Mol Cell Biol. 2003; 23(1):206-215.

(14.) Zhang C, Li H, Wang Y, et al. Epigenetic inactivation of the tumor suppressor gene RIZ1 in hepatocellular carcinoma involves both DNA methylation and histone modifications. J Hepatol. 2010; 53(5):889-895.

(15.) Gencheva M, Chen CJ, Nguyen T, Shively JE. Regulation of CEACAM1 transcription in human breast epithelial cells. BMC Mol Biol. 2010; 11:79-93.

(16.) Shen F, Kirmani KZ, Xiao Z, Thirlby BH, Hickey RJ, Malkas LH. Nuclear protein isoforms: implications for cancer diagnosis and therapy. J Cell Biochem. 2011; 112(3):756-760.

(17.) Bernstein BE, Kamal M, Lindblad-Toh K, et al. Genomic maps and comparative analysis of histone modifications in human and mouse. Cell. 2005; 120(2):169-181.

(18.) Pokholok DK, Harbison CT, Levine S, et al. Genome-wide map of nucleosome acetylation and methylation in yeast. Cell. 2005; 122(4):517-527.

(19.) Park YS, Jin MY, Kim YJ, Yook JH, Kim BS, Jang SJ. The global histone modification pattern correlates with cancer recurrence and overall survival in gastric adenocarcinoma. Ann Surg Oncol. 2008; 15(7):1968-1976.

(20.) Song JS, Kim YS, Kim DK, Park SI, Jang SJ. Global histone modification pattern associated with recurrence and disease-free survival in non-small cell lung cancer patients. Pathol Int. 2012; 62(3):182-190.

(21.) Barnes L, Eveson JW, Reichart P, Sidransky D, eds. Pathology and Genetics of Head and Neck Tumours. Lyon, France: IARC Press; 2005. World Health Organization Classification of Tumours; vol 9.

(22.) Greene FL, Page DL, Fleming ID, eds. AJCC Cancer Staging Manual. 6th ed. New York, NY: Springer; 2002.

(23.) Jenuwein T, Allis CD. Translating the histone code. Science. 2001; 293(5532):1074-1080.

(24.) Kurdistani SK. Histone modifications in cancer biology and prognosis. Prog Drug Res. 2011; 67:91-106.

(25.) Nishimine M, Nakamura M, Kishi M, et al. Alterations of p14ARF and p16INK4a genes in salivary gland carcinomas. Oncol Rep. 2003; 10(3):555-560.

(26.) Nandakumar V, Vaid M, Katiyar SK. (-)-Epigallocatechin-3-gallate reactivates silenced tumor suppressor genes, Cip1/p21 and p16INK4a, by reducing DNA methylation and increasing histones acetylation in human skin cancer cells. Carcinogenesis. 2011; 32(4):537-544.

(27.) Martin C, Chen S, Heilos D, et al. Changed genome heterochromatinization upon prolonged activation of the Raf/ERK signaling pathway. PLoS ONE. 2010; 5(10):e13322.

(28.) Mysliwiec MR, Carlson CD, Tietjen J, Hung H, Ansari AZ, Lee Y. Jarid2 (Jumonji, AT rich interactive domain 2) regulates NOTCH1 expression via histone modification in the developing heart. J Biol Chem. 2012; 287(2):1235-1241.

(29.) Lakshmikuttyamma A, Scott SA, DeCoteau JF, Geyer CR. Reexpression of epigenetically silenced AML tumor suppressor genes by SUV39H1 inhibition. Oncogene. 2010; 29(4):576-588.

(30.) Metge BJ, Liu S, Riker AI, Fodstad O, Samant RS, Shevde LA. Elevated osteopontin levels in metastatic melanoma correlate with epigenetic silencing of breast cancer metastasis suppressor 1. Oncology. 2010; 78(1):75-86.

(31.) Mori T, Kim J, Yamano T, et al. Epigenetic up-regulation of C-C chemokine receptor 7 and C-X-C chemokine receptor 4 expression in melanoma cells. Cancer Res. 2005; 65(5):1800-1807.

(32.) Muller-Tidow C, Klein HU, Hascher A, et al. Profiling of histone H3 lysine 9 trimethylation levels predicts transcription factor activity and survival in acute myeloid leukemia. Blood. 2010; 116(18):3564-3571.

(33.) Zhen L, Gui-lan L, Ping Y, Jin H, Ya-li W. The expression of H3K9Ac, H3K14Ac, and H4K20TriMe in epithelial ovarian tumors and the clinical significance. Int J Gynecol Cancer. 2010; 20(1):82-86.

(34.) Pasini D, Malatesta M, Jung HR, et al. Characterization of an antagonistic switch between histone H3 lysine 27 methylation and acetylation in the transcriptional regulation of Polycomb group target genes. Nucleic Acids Res. 2010; 38(15):4958-4969.

(35.) Papoutsis AJ, Borg JL, Selmin OI, Romagnolo DF. BRCA-1 promoter hypermethylation and silencing induced by the aromatic hydrocarbon receptor-ligand TCDD are prevented by resveratrol in MCF-7 cells. J Nutr Biochem. 2011; 23(10):1324-1332.

(36.) Al-Mamgani A, van Rooij P, Sewnaik A, Tans L, Hardillo JA. Adenoid cystic carcinoma of parotid gland treated with surgery and radiotherapy: long-term outcomes, QoL assessment and review of the literature. Oral Oncol. 2012; 48(3): 278-283.

(37.) Shen C, Xu T, Huang C, Hu C, He S. Treatment outcomes and prognostic features in adenoid cystic carcinoma originated from the head and neck. Oral Oncol. 2012; 48(5):445-449.

Ronghui Xia, DDS; Rongrui Zhou, DDS, MDS; Zhen Tian, DDS, MDS, PhD; Chunye Zhang, DDS, MDS; Lizhen Wang, DDS, MDS; Yuhua Hu, DDS, MDS; Jing Han, DDS, MDS; Jiang Li, DDS, MDS, PhD

Accepted for publication April 6, 2013.

Supplemental digital content is available for this article at www. in the December 2013 Table of Contents.

From the Department of Oral Pathology, 9th People's Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai Key Laboratory of Stomatology, Shanghai, China (Drs Xia, Tian, Zhang, Wang, Hu, Han, and Li); the Department of Oncology and Diagnostic Sciences, Dental School, University of Maryland, Baltimore, Maryland (Dr Xia); and the Department of Oral Medicine, Shanghai Stomatological Disease Center, Shanghai, China (Dr Zhou). Ronghui Xia and Rongrui Zhou contributed equally to this work.

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

Reprints: Jiang Li, DDS, MDS, PhD, Department of Oral Pathology, 9th People's Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai Key Laboratory of Stomatology, Shanghai 200011, China (e-mail:

Caption: Figure 1. Representative photomicrographs of immunohistochemical staining for histone H3 lysine 9 trimethylation (H3K9me3). A, High expression of H3K9me3 in a cribriform-tubular pattern tumor. B, Low expression of H3K9me3 in a cribriform-tubular pattern tumor. C, High expression of H3K9me3 in a solid pattern tumor. D, Low expression of H3K9me3 in a solid pattern tumor (original magnifications X40 [A, B, and C], X20 [D], and X80 [insets B and D]).

Caption: Figure 2. Representative photomicrographs of immunohistochemical staining for histone H3 lysine 9 acetylation (H3K9Ac). A, High expression of H3K9Ac in a cribriform-tubular pattern tumor. B, Low expression of H3K9Ac in a cribriform-tubular pattern tumor. C, High expression of H3K9Ac in a solid pattern tumor. D, No expression of H3K9Ac in a cribriform-tubular pattern tumor (original magnifications X40 [A, B, and D], X20 [C], and X80 [inset C]).

Caption: Figure 3. Histone H3 lysine 9 trimethylation (H3K9me3) and histone H3 lysine 9 acetylation (H3K9Ac) expression was correlated with the histologic pattern of salivary adenoid cystic carcinoma. A, High expression of H3K9Ac in a cribriform-tubular pattern tumor. B, Low expression of H3K9me3 in a cribriform-tubular pattern tumor. C, Low expression of H3K9Ac in a solid pattern tumor. D, High expression of H3K9me3 in a solid pattern tumor (original magnifications X20 [A and B] and X40 [C, D, and inset A]).

Caption: Figure 4. A through E, Kaplan-Meier survival analyses of the relationship between histone H3 lysine 9 trimethylation (H3K9me3) expression (A), histone H3 lysine 9 acetylation (H3K9Ac) expression (B), histologic pattern (C), M stage (D), and distant metastasis (E) with overall survival. F through H, Kaplan-Meier survival analyses of the relationship between H3K9me3 (F), histologic pattern (G), and margin status (H) with disease-free survival. Abbreviation: c/t, cribriform/tubular.

Caption: Please Note Illustration(s) are not available due to copyright restrictions.
Table 1. Correlation Between H3K9me3 and H3K9Ac Expression and
Clinicopathologic Parameters of Adenoid Cystic Carcinoma3

Characteristics                H3K9me3        P     H3K9Ac         P
                               Expression   Value   Expression   Value

                               Low   High           Low   High
Age y
  [less than or equal to] 55   17     19     .56    25     11     .51
  >55                          12     18            23     7
  Male                         19     17     .11    27     9      .65
  Female                       10     20            21     9
Tumor site
  Minor glands                 17     22     .95    28     11     .84
  Major glands                 12     15            20     7
Histologic pattern
  Cribriform-tubular           27     22    .002#   32     17    .03#
  Solid                         2     15            16     1
Perineural invasion
  Yes                          22     28     .99    36     14     .82
  No                            7     9             12     4
T stage
  T1+T2                        19     28     .37    33     14     .47
  T3+T4                        10     9             15     4
N stage
  N+                            4     9      .29    10     3     >.99
  N-                           25     28            38     15
M stage
  M0                           29     35     .50    46     18    >.99
  M1                            0     2              2     0
  Yes                           6     15     .09    15     6      .87
  No                           23     22            33     12
Distant metastasis
  Yes                           8     26    .001#   26     8      .48
  No                           21     11            22     10

Abbreviations: H3K9Ac, histone H3 lysine 9 acetylation;H3K9me3,
histone H3 lysine 9 trimethylation;N+, positive lymph node;N-,
negative lymph node.

(a) Bolded values are statistically significant (P [less than or
equal to] .05).

Note: Statistically significant (P [less than or
equal to] .05) values are indicated with #.

Table 2. Multivariate Analysis Using Cox
Proportional Hazards Regression Models
to Estimate Overall Survival (OS)
and Disease-Free Survival (DFS) (a)

Variables                   P Value   Hazard
                                      Ratio       95% CI

OS analysis
  Histologic pattern (c/t
    versus s)               .04#      0.409    0.178-0.938
  M stage (M0 versus
    M1)                     .09       3.810    0.831-17.473
  H3K9me3 expression
    (low versus high)       .004#     0.111    0.025-0.501
DFS analysis
  Histologic pattern (c/t
    versus s)               .60       0.816    0.387-1.724
  Margin status (negative
    versus positive)        .21       0.208    0.352-1.255
  H3K9me3 expression
    (low versus high)       .006#     0.351    0.165-0.745

Abbreviations: CI, confidence interval; c/t, cribriform/tubular
pattern; H3K9me3, histone H3 lysine 9 trimethylation;s, solid

(a) Bolded values are statistically significant (P [less than
or equal to] .05).

Note: statistically significant (P [less than or equal to] .05)
values are indicated with #.

Figure 5. Summary of the univariate analysis for overall survival and
disease/free survival for patients with salivary adenoid cystic
carcinoma. The forest plot shows that patients with solid pattern
tumors, M1 stage tumors, and higher levels of histone H3 lysine 9
trimethylation (H3K9me3) expression had a poorer overall survival.
Patients with a solid pattern tumor, positive margins, and higher
levels of histone H3K9me3 expression had a poorer disease/free
survival. * Statistically significant values (P [less than or equal to]
.05). Abbreviations: CI, confidence interval; CT, chemotherapy; c/t,
cribriform/tubular pattern; HRs, hazard ratios; n, negative; H3K9Ac,
histone H3 lysine 9 acetylation; H3K9me3, histone H3 lysine 9
trimethylation; p, positive; RT, radiation therapy; s, solid pattern.

                                    Overall Survival

Age (55[less than                  0.491 (0.218-1.107)
  or equal to] vs. >55)
Gender (male vs. female)           0.773 (0.347-1.721)
Site (major vs. minor)             0.811 (0.355-1.855)
Histological pattern (c/t vs. s)   0.209 (0.093-0.468)
Perineural invasion (p vs. n)      0.633 (0.262-1.529)
Margin (n vs. p)                   0.573 (0.253-1.294)
T stage (T1/T2 vs. T3/T4)          0.603 (0.257-1.415)
N stage (N- vs. N+)                0.786 (0.293-2.107)
M stage (MO vs Ml)                 0.151 (0.034-0.681)
RT and/or CT (yes vs. no)          0.634 (0.251-1.602)
H3K9mc3 (low vs. high)             0.076 (0.018-0.326)
H3K9Ac (high vs. low)              0.312 (0.093-1.049)
                                       HRs (95%C1)

                                   Disease-free Survival

Age (55[less than                   1.096 (0.591-2.032)
  or equal to] vs. >55)
Gender (male vs. female)            0.565 (0.307-1.038)
Site (major vs. minor)              0.713 (0.379-1.342)
Histological pattern (c/t vs. s)    0.435 (0.222-0.855)
Perineural invasion (p vs. n)       0.600 (0.301-1.196)
Margin (n vs. p)                    0.513 (0.275-0.957)
T stage (T1/T2 vs. T3/T4)           0.602 (0.311-1.166)
N stage (N- vs. N+)                 0.873 (0.387-1.969)
M stage (MO vs Ml)
RT and/or CT (yes vs. no)           0.687 (0.337-1.399)
H3K9mc3 (low vs. high)              0.295 (0.151-0.576)
H3K9Ac (high vs. low)               0.736 (0.370-1.467)
                                        HRs (95%CI)
COPYRIGHT 2013 College of American Pathologists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Xia, Ronghui; Zhou, Rongrui; Tian, Zhen; Zhang, Chunye; Wang, Lizhen; Hu, Yuhua; Han, Jing; Li, Jian
Publication:Archives of Pathology & Laboratory Medicine
Date:Dec 1, 2013
Previous Article:How to Evaluate a Potential Merger or Acquisition.
Next Article:Unintended Reporting of Misleading Hb [A.sub.1c] Values When Using Assays Incapable of Detecting Hemoglobin Variants.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters