- Email: [email protected]
Lower ataxia telangiectasia mutated (ATM) mRNA expression is correlated with poor outcome of laryngeal and pharyngeal cancer patients
original article
Annals of Oncology 22: 1088–1093, 2011 doi:10.1093/annonc/mdq569 Published online 1 December 2010
Lower ataxia telangiectasia mutated (ATM) mRNA expression is correlated with poor outcome of laryngeal and pharyngeal cancer patients K.-W. Lee1,2,3, Y.-S. Tsai4,5, F.-Y. Chiang1,2, J.-L. Huang6, K.-Y. Ho1,2, Y.-H. Yang7, W.-R. Kuo1,2, M.-K. Chen8,9 & C.-S. Lin4,5* 1
original article
Received 12 September 2009; revised 13 March 2010 & revised 8 June 2010; accepted 16 August 2010
Background: Ataxia telangiectasia mutated (ATM) kinase is a critical regulator in initiating DNA damage response and activating DNA repair. However, the correlation between ATM expression and the outcome of laryngopharyngeal cancer patients is unknown. We hypothesize that ATM expression is correlated with a worse outcome in laryngopharyngeal cancer patients. Patients and methods: The ATM messenger RNA (mRNA) expression of 80 tumors of laryngeal and pharyngeal cancer was examined by real-time quantitative RT-PCR. Overall survival rates were measured using Kaplan–Meier estimates and the log-rank tests. The adjusted hazard rate ratios (HRRs) were computed by multivariate Cox regressions. Results: Reduced ATM mRNA was found in 65 of 80 studied cases. Lower ATM expression [tumor/normal <0.3, HRR = 2.49; 95% confidence interval (CI) 1.27–4.88], younger age (<55 years, HRR = 2.71; 95% CI 1.16–6.32), and larger tumor (T3/T4, HRR = 2.21; 95% CI 1.10–4.44) were independent risk factors for survival. Patients with lower ATM and younger age (HRR = 6.51; 95% CI 2.05–20.66) or with lower ATM and T3/T4 tumor (HRR = 5.23; 95% CI 2.04–13.40) exhibited the poorest outcome. Conclusion: The expression of ATM mRNA, which is frequently downregulated in laryngeal and pharyngeal cancers, could be a valuable prognostic marker. Key words: ataxia telangiectasia mutated, laryngeal cancer, pharyngeal cancer, survival analysis
introduction DNA damage response plays an important role in the maintenance of genome integrity. Defects or dysregulation of genes involved in DNA damage response and DNA repair can result in genomic instability, which is a common feature of cancer cells [1]. Head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer in the world. It has been suggested that genomic instability may be involved in the development of HNSCC [2–4]. Some studies have shown that DNA repair activity is reduced in the peripheral blood cells of HNSCC patients when compared with normal individuals [5, 6]. These indicate that interference of DNA damage repair may contribute
*Correspondence to: Dr C.-S. Lin, Graduate Institute of Medicine, College of Medicine, Kaohsiung Medical University, 100 Shih-Chuan 1st Road, Kaohsiung 807, Taiwan. Tel: +886-7-3121101-2019; Fax: +886-7-3218309; E-mail: [email protected]
to tumorigenesis of HNSCC. Nevertheless, studies on DNA repair gene expression in head and neck cancers are limited so far. Ataxia telangiectasia mutated (ATM) kinase is a critical regulator in DNA damage response and is conserved in all eukaryotes from yeast to human. It is activated immediately upon DNA damage and then phosphorylates p53, BRCA1, CHEK1/2, and many others, which lead to activation of DNA repair and cell cycle arrest [7]. Germline mutation of ATM gene results in ataxia telangiectasia, a syndrome susceptible to various cancers. Loss of ATM gene or decreased ATM expression is found in human cancers [8–14]. Parikh et al. [14] found that the ATM locus was partially lost in 8 of 11 HNSCC cell lines examined and in at least 60% of human malignancies including HNSCC. This partial loss of ATM locus was correlated with a reduced expression of ATM protein level in HNSCC cell lines [14]. In addition, Ai et al. [15] and Bolt et al. [16] have shown hypermethylation of ATM promoter in head and neck cancer, suggesting that ATM expression might be repressed. Indeed, He et al. [10] found an
ª The Author 2010. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected]
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
Department of Otolaryngology, Kaohsiung Medical University Hospital, Kaohsiung; 2Department of Otolaryngology, Faculty of Medicine, College of Medicine; Kaohsiung Medical University, Kaohsiung; 3Department of Otolaryngology, Kaohsiung Municipal Ta-Tung Hospital, Kaohsiung; 4Graduate Institute of Medicine, College of Medicine, Kaohsiung; 5Center of Excellence for Environmental Medicine, Kaohsiung Medical University, Kaohsiung; 6Department of Bioscience Technology, Chang Jung Christian University, Tainan; 7Department of Dental Hygiene, College of Dental Medicine, Kaohsiung Medical University, Kaohsiung; 8Department of Otorhinolaryngology-Head and Neck Surgery, Changhua Christian Hospital, Changhua; 9School of Medicine, Chung Shan Medical University, Taichung, Taiwan
original article
Annals of Oncology
absent or reduced ATM protein expression in 31.25% of oral squamous cell carcinoma, and the patients with reduced/absent ATM expression had poorer differentiated tumor and more lymph node invasion. Bose et al. [8] also showed that ATM messenger RNA (mRNA) and protein were downregulated in Epstein–Barr virus-associated nasopharyngeal carcinoma. However, whether ATM expression is also repressed in HNSCC at larynx and pharynx subsites and whether ATM expression is correlated with HNSCC patient outcome are unclear. In these regards, we examined ATM expression in laryngeal and pharyngeal cancers and correlated ATM expression with clinicopathologic parameters to explore the role of ATM expression in the prediction of patient outcome.
subjects The specimens of 40 laryngeal cancer patients (17 paired tumor and adjacent normal tissues plus 23 tumors only) and 40 pharyngeal cancer patients (15 paired tumor and adjacent normal tissues plus 25 tumors only) were obtained from the Department of Otolaryngology, Kaohsiung Medical University Hospital. These specimens were collected after obtaining patient’s informed consent, and the study was approved by the Institutional Review Board. Specimens were snap frozen in liquid nitrogen and were stored at 280C until use. The age of patients was between 33 and 82 years, with the median of 61 years, the first quartile of 54 years, and the third quartile of 70 years.
RNA isolation, reverse transcription, and real-time quantitative PCR The method for quantitatively analyzing mRNA expression was as described in our previous paper [17]. Briefly, total RNA was isolated by Tri-reagent (Sigma–Aldrich, St Louis, MO). Reverse transcription was conducted with 1 lg of total RNA and the High-Capacity cDNA Archive Kit (Applied BioSystems, Foster City, CA) in 20 ll. The resulting complementary DNA (cDNA) was subsequently diluted to a total volume of 100 ll and 2 ll of diluted cDNA was used as templates for real-time quantitative PCR (Q-PCR). Q-PCR was carried out in a total volume of 20 ll using PowerSYBR Green reagent (Applied BioSystems) and ABI Prism 7500 Sequence Detection System instrument (Applied BioSystems). The PCR condition was 50C for 2 min, 95C for 10 min followed by 50 cycles at 95C for 15 s and 60C for 1 min. Dissociation curve was added at the final step to validate that only the specific target was amplified. The PCR primers were designed using the ProbeFinder software (Roche Applied Science, Mannheim, Germany) and were ATM-F: ATAGATTGTGTAGGTTCCGATGG, ATM-R: CATCTTGTCTCAGGTCATCACG; ACTB-F: ATTGGCAATGAGCGGTTC, ACTB-R: GGATGCCACAGGACTCCAT. The ATM mRNA level in each sample was calculated by 22DDCT method [18] and ACTB was used as an internal control. Each specimen was examined by at least two independent rounds of reverse transcription and Q-PCR reactions for ATM and internal control. In all cases with paired specimens, the ATM mRNA expression of each case was represented by the ratio of ATM mRNA level in each tumor to that of the corresponding normal part. For the cases with tumor parts only, the mean ATM mRNA levels of all normal tissue specimens were used for comparison with those from each individual tumor.
statistical analyses The ATM mRNA expressions were categorized into two groups (<0.3 and ‡0.3) according to Receiver-Operating Characteristic analysis. Its relationship with clinicopathologic variables was analyzed by Chi-square or Fisher’s exact test. Kaplan–Meier estimates and the log-rank tests were
Volume 22 | No. 5 | May 2011
results ATM mRNA is downregulated in laryngeal and pharyngeal cancers To examine ATM expression in laryngeal and pharyngeal cancers, we harvested total RNA from 80 surgically resected and previously untreated tumors. The characteristics of patients were shown in Table 1. The ATM mRNA expression was determined by real-time quantitative RT-PCR using ACTB as an internal control. The results showed that ATM mRNA expression was decreased (tumor/normal <1) in 65 of 80 (81.3%) laryngeal and pharyngeal cancers (Table 1). The median of ATM mRNA expression (tumor/normal) in the 80 cases was 0.364, and the 25% and 75% quartiles were 0.222 and 0.762, respectively (Figure 1). These results demonstrated that ATM mRNA was frequently downregulated in laryngeal and pharyngeal cancers. ATM mRNA expression is correlated with survival of laryngeal and pharyngeal cancer patients To explore the possible role of decreased ATM expression in laryngeal and pharyngeal cancers, the correlation between ATM mRNA expression and clinicopathologic features was analyzed. We used the ratio of ATM mRNA expression of tumor versus normal at 0.3 as a cut-off point according to the ReceiverOperating Characteristic curve analysis, by which an association (P = 0.048, Chi-square analysis) between patient survival status and ATM mRNA expression was observed (Table 2). Patients
Table 1. Characteristics of patients Variable Total Gender Age (years) Tumor site Tumor size (T)
Lymph node (N)
Distant metastasis (M) Ataxia telangiectasia mutated (ATM) expressionb
Category
Na
%
Male Female <55 ‡55 Larynx Pharynx T1 T2 T3 T4 N0 N1 N2 N3 M0 M1 ‡1
80 73 7 25 55 40 40 19 18 18 20 42 14 14 5 74 1 15
100.0 91.3 8.7 31.3 68.7 50.0 50.0 25.3 24.0 24.0 26.7 56.0 18.7 18.7 7.7 98.7 1.3 18.7
<1
65
81.3
a
Five missing data for tumor size, lymph node, and distant metastasis. Ratio of tumor/normal.
b
doi:10.1093/annonc/mdq569 | 1089
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
patients and methods
employed to compare the survival rates for clinicopathologic variables. The hazard rate ratios (HRRs) and adjusted HRRs were further computed by univariate and multivariate Cox regressions.
original article
Annals of Oncology
with lower ATM mRNA expression (<0.3) exhibited a poorer overall survival (9 alive versus 20 dead) than those with ATM mRNA expression ‡0.3 (25 alive versus 21 dead). There was no significant association between ATM mRNA expression and other clinicopathologic parameters (Table 2).
We next analyzed 3-year and 5-year survival rates using Kaplan– Meier estimates with log-rank tests. The results demonstrated that patients with age <55 years (P = 0.004), with larger tumor size (T3 and T4, P = 0.012), and with lower ATM mRNA expression (<0.3, P = 0.015) had poorer survival rates (Table 3). Figure 2 shows a significant difference in survival rates between individuals with higher and lower ATM mRNA expression. These data indicated that ATM mRNA expression could be a prognostic marker for laryngeal and pharyngeal cancers.
Table 3. Kaplan–Meier survival rates Variable Overall Gender Age (years) Tumor size (T) Figure 1. Ataxia telangiectasia mutated (ATM) messenger RNA (mRNA) was downregulated in 65 of 80 laryngeal and pharyngeal cancers. The expression of ATM mRNA was determined using real-time quantitative RT-PCR and shown as a box plot that comprised the average of, at least, two independent experiments for each sample. The box represents upper (0.762) and lower (0.222) quartiles; the line in the box stands for the median (0.364).
Lymph node (N) Ataxia telangiectasia mutated (ATM) expressionb
Category
3 years (%)
5 years (%)
Male Female <55 ‡55 T1–T2 T3–T4 N0 N1–N3 <0.3
53.1 52.5 60.0 36.4 60.3 75.9 35.4 66.0 41.7 36.1
42.0 41.0 – 19.5 52.3 52.2 35.4 53.2 30.9 28.0
‡0.3
64.0
51.3
P-valuea 0.502 0.004 0.012 0.053 0.015
a
Log-rank test. Ratio of tumor/normal.
b
Table 2. Correlation between clinicopathologic variables and ataxia telangiectasia mutated (ATM) mRNA expressiona Variable
Total Gender Age Tumor size (T) Lymph node (N) Survival
Category
Nb
‡0.3 n % in category
<0.3 n % in category
Male Female <55 ‡55 T1–T2
80 73 7 25 55 37
50 46 4 17 33 25
62.5 63.0 57.1 68.0 60.0 67.6
30 27 3 8 22 12
37.5 37.0 42.9 32.0 40.0 32.4
T3–T4 N0
38 42
21 28
55.3 66.7
17 14
44.7 33.3
N1–N3 No Yes
33 41 34
18 21 25
54.5 51.2 73.5
15 20 9
45.5 48.8 26.5
P-valuec
>0.999d 0.493 0.274
0.285
0.048
a
Ratio of tumor/normal; cut-off point (0.3) was determined by the Receiver-Operating Characteristic curve. b Five missing data for tumor size, lymph node, and survival. c Chi-square test. d Fisher’s exact test.
1090 | Lee et al.
Figure 2. Overall survival curves were analyzed according to ataxia telangiectasia mutated (ATM) messenger RNA expression by using Kaplan–Meier estimate with log-rank test. The analyzed numbers (N) of the two categories are indicated.
Volume 22 | No. 5 | May 2011
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
lower ATM mRNA expression is an independent risk factor for survival of laryngeal and pharyngeal cancer patients To investigate whether lower ATM mRNA expression (<0.3) alone could predict poor outcome of laryngeal and pharyngeal cancers,
original article
Annals of Oncology
patients with lower ATM expression and younger age or larger tumor size have the poorest outcome We examined the effect of combining lower ATM mRNA expression with age or tumor size on predicting patient outcome. As shown in Figure 3, all seven patients in the <55-year-old group
Table 4. Multivariate Cox model analysis of overall survival Variable
Hazard rate ratio
95% Confidence interval Lower Upper
P-value
Ataxia telangiectasia mutated (ATM) expression (<0.3 versus ‡0.3) Gender (male versus female) Age (<55 versus ‡55 years) Tumor size (T3–T4 versus T1–T2) N stage (N1–N3 versus N0)
2.49
1.27
4.88
0.008
1.91 2.71 2.21
0.45 1.16 1.10
8.09 6.32 4.44
0.380 0.021 0.026
0.96
0.42
2.16
0.912
who have lower ATM mRNA expression (<0.3) died within 4 years. In addition, a significant proportion (13/16, 81.3%) of patients with lower ATM mRNA expression (<0.3) and larger tumor size (T3–T4) died within 3 years. On the contrary, patients with T1–T2 tumors or patients whose age was >55 years in combination with higher ATM mRNA expression (‡0.3) had better survival rates (Figure 3). Multivariate Cox regression model (Table 5) showed significant higher risks for individuals with lower ATM mRNA expression and younger age (HRR = 6.51; 95% CI 2.05–20.66; P = 0.001) or with lower ATM mRNA expression and T3/T4 tumor size (HRR = 5.23; 95% CI 2.04–13.40; P = 0.001). These results demonstrated that co-evaluation of ATM mRNA expression and age or tumor size might be a valuable tool for predicting patient outcome.
discussion ATM kinase plays a key role in initiating several DNA repair pathways. Downregulation of ATM may lead to broad dysfunction in DNA repair and the accumulation of genome alterations. In this study, we found that ATM mRNA was downregulated in 65/80 (81.3%) of laryngeal and pharyngeal cancers (Figure 1), and the lower ATM expression (<0.3) was an independent risk factor for patient’s survival (Table 4). To our knowledge, this is the first study showing that ATM expression is a valuable prognostic marker for HNSCC. It has been demonstrated that ATM expression was repressed in oral and nasopharyngeal carcinomas [8, 10], but the effect of ATM downregulation on patient’s outcome of these two malignancies was not shown. The prognostic role of ATM expression has been demonstrated in other malignancies, including breast [13], colon [9], lung [12], and gastric cancer [11]. All these studies showed that lower ATM expression was a risk factor for patient’s outcome, except for that in lung cancer [12]. Given the above, it seems that lower ATM expression is usually correlated with poor outcome of cancer patients. Moreover, we noted that the prognostic value of ATM expression could be largely enhanced by co-evaluating patient’s age or tumor size (Table 5). Both were also independent risk factors for patient’s survival (Table 4). Early onset of HNSCC
Figure 3. Overall survival curves according to ataxia telangiectasia mutated (ATM) messenger RNA expression combined with age (left panel) or tumor size (right panel). Analyses were conducted using Kaplan–Meier estimate with log-rank test. The analyzed number (N) of each category is indicated. Others (à and §) stand for ATM ‡ 0.3 and age <55 years or ATM < 0.3 and age ‡55 years (left panel); and ATM ‡ 0.3 and T3–T4 tumor or ATM < 0.3 and T1–T2 tumor (right panel).
Volume 22 | No. 5 | May 2011
doi:10.1093/annonc/mdq569 | 1091
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
we conducted multivariate Cox regression to examine the effect of lower ATM mRNA expression (Table 4). After adjusting for sex, age (<55 versus ‡55 years), tumor size (T3–T4 versus T1–T2), and lymph node metastasis (N1–N3 versus N0), the results showed that lower ATM mRNA expression (<0.3) remained a significant prognostic factor [HRR = 2.49; 95% confidence interval (CI) 1.27– 4.88; P = 0.008] for survival. Moreover, age (HRR = 2.71; 95% CI 1.16–6.32; P = 0.021) and tumor size (HRR = 2.21; 95% CI 1.10– 4.44; P = 0.026) were also independent factors for predicting overall survival (Table 4). These results indicated that lower ATM mRNA expression could be an independent risk factor for survival of laryngeal and pharyngeal cancer patients. When considering the ratio of ATM mRNA expression itself in the multivariate Cox regression analysis, a trend of association between ATM mRNA expression and overall survival could be found (supplemental Table S1, available at Annals of Oncology online), although statistical significance was not reached (P = 0.071).
original article
Annals of Oncology
Table 5. Multivariate Cox model analysis of overall survival (combining two factors) Variables
Hazard rate ratio
95% Confidence interval Lower Upper
P-value
a
0.001 0.017
0.001 0.179
a
Adjusted by sex, tumor size, lymph node invasion. Adjusted by sex, age, lymph node invasion.
b
implies that these patients may have certain genetic susceptibilities for tumor development. Large tumor size may reflect that the tumor cells have accumulated much more genetic alterations. Therefore, it is reasonable to combine ATM expression with these two independent risk factors as predictive parameters to better estimate the effect. One of the mechanisms underlying the downregulation of ATM mRNA expression may be hypermethylation of the ATM promoter. This results in suppression of promoter activity and has been demonstrated in HNSCC [15, 16] and other malignancies [19]. In this regard, our in vitro study showed that ATM promoter activity could be inhibited by ingredients of betel nut (C.-S. Lin, H.-Y. Hsieh, Y.-S. Tsai and J.-L. Huang, unpublished data), which is a popular substance for habitual use in India and Southeast Asia including Taiwan. However, the relationship between lower ATM expression and the habit of betel quid chewing in HNSCC patients remains to be established. Another mechanism accounting for ATM suppression may be the loss of ATM gene locus, by which reduced copy numbers at chromosome 11q22-23 has been demonstrated in HNSCC by comparative genomic hybridization [3, 20, 21]. In our studied cases, only one tumor specimen showed negative RT-PCR amplification (not included in the 80 cases described), implying that deletion of ATM gene might be possible for this case. However, since ATM mRNA could more or less be detected in all 80 tumors, complete gene deletions could not be the reason. Rather, the possibility of loss of one allele at chromosome 11q2223 should not be excluded. Indeed, Parikh et al. [14] found that ATM locus was partially lost in 8 of 11 HNSCC cell lines and all five clinical specimens examined. Lazar et al. [22] also showed loss of heterozygosity (LOH) at 11q23 in 25% (13/52) of primary HNSCC. These results support that loss of ATM locus is possibly one of the mechanisms accounting for lower ATM expression found in HNSCC. In addition to ATM, only a limited number of DNA repair genes have been examined for their expression and prognostic
1092 | Lee et al.
Volume 22 | No. 5 | May 2011
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
Ataxia telangiectasia mutated (ATM) expression and age ATM < 0.3 and age < 55 years 6.51 2.05 20.66 ATM ‡ 0.3 and age < 55 years 2.85 1.21 6.76 or ATM < 0.3 and age ‡ 55 years ATM ‡ 0.3 and age ‡ 55 years 1 ATM expression and tumor sizeb ATM < 0.3 and T3–T4 5.23 2.04 13.40 1.85 0.76 4.50 ATM ‡ 0.3 and T3–T4 or ATM < 0.3 and T1–T2 1 ATM ‡ 0.3 and T1–T2
role in head and neck cancers. Excision repair cross complementation group 1 (ERCC1), which removes cisplatininduced DNA adducts, plays an important role in both damage recognition and incision activities of nucleotide excision repair. It has been shown that lower ERCC1 expression was associated with a lower risk of death from advanced head and neck cancer and also benefited from cisplatin induction chemotherapy [23]. This is possibly correlated with the reduced repair of cisplatin– DNA adducts, which in turn leads to lethal DNA strand breaks and facilitates the eradication of tumor cells. On the other hand, the Nijmegen breakage syndrome gene product (NBN) that works together with MRE11A and RAD50 to recognize DNA double-strand breaks and initiate the signaling for DNA repair was shown to be an independent marker of poor prognosis for advanced HNSCC patients when overexpressed [24]. The authors also found that overexpression of NBN could increase cell transformation. These results suggest that the roles of DNA repair gene expressions in predicting patient outcome may be dependent on the characteristics of each gene and the pathways involved. In the present study, the patients were mostly treated with concurrent chemoradiotherapy after surgical resection. The relationship between ATM expression and patient treatments was not yet explored in detail. However, according to the study of Parikh et al. [14] that partial loss of ATM gene and reduced ATM expression is associated with a reduced sensitivity of HNSCC cells to ionizing irradiation (IR) and the finding of Lazar et al. [22] that LOH at ATM locus in HNSCC is associated with recurrent disease among patients who received radiotherapy, it is proposed that laryngopharyngeal tumors with lower ATM expression may be less responsive to radiotherapy and/or chemotherapy than those with higher ATM expression. As a result, patients with lower ATM mRNA expression may exhibit a poor outcome as shown in this study. This hypothesis suggests that a combination regimen exploiting different cell-killing mechanisms may be superior to monotherapy. Recent results obtained from several clinical trials support this hypothesis, by which another cell-killing mechanism via mitotic catastrophe induced by docetaxel can add the efficacy of chemotherapy using cisplatin and fluorouracil and improve the overall survival rates of patients with no distant metastatic HNSCC [25–27]. Another trial that combines platinum-based chemotherapy and cetuximab (by antagonizing the signaling of epidermal growth factor receptor) also improves the overall survival of head and neck cancer patients with recurrent or metastatic tumors [28]. Therefore, the concomitant findings of lower ATM expression and reduced IR response may, at least partly, explain why these trials that use combination therapy can improve survival of HNSCC patients. Recent studies on modulating DNA repair pathways to improve cancer treatments have become an attractive strategy [29]. Tumor cells survived from innumerable genetic alterations depend largely on multiple DNA repair pathways. However, defects of specific DNA repair pathway may arise during tumor development. Inhibition of another DNA repair pathway can therefore increase selectively killing of tumor cells that contain certain alterations of DNA damage response or DNA repair pathway. For example, poly
Annals of Oncology
(ADP-ribose) polymerase 1 (PARP1) inhibitors that interfere with base excision repair have been successfully used for treating the breast and ovarian cancer patients with inherited defects of BRCA1 or BRCA2 genes [30, 31]. Notably, such kind of treatment is less toxic to normal tissues than conventional chemoradiotherapy. Therefore, investigations of the genes involved in DNA damage responses and DNA repair pathways in HNSCC may help find better prognostic markers and strategies for patient treatment.
funding
acknowledgements We thank Ms Jenny Chang and Cathy Yang for critical reading of the manuscript. MK Chen and CS Lin contributed equally to this article.
disclosure The authors declare no conflict of interest.
references 1. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100: 57–70. 2. Friedlander PL. Genomic instability in head and neck cancer patients. Head Neck 2001; 23: 683–691. 3. Bockmuhl U, Schwendel A, Dietel M, Petersen I. Distinct patterns of chromosomal alterations in high- and low-grade head and neck squamous cell carcinomas. Cancer Res 1996; 56: 5325–5329. 4. Veltman JA, Bot FJ, Huynen FC et al. Chromosome instability as an indicator of malignant progression in laryngeal mucosa. J Clin Oncol 2000; 18: 1644–1651. 5. Cheng L, Eicher SA, Guo Z et al. Reduced DNA repair capacity in head and neck cancer patients. Cancer Epidemiol Biomarkers Prev 1998; 7: 465–468. 6. Paz-Elizur T, Ben-Yosef R, Elinger D et al. Reduced repair of the oxidative 8-oxoguanine DNA damage and risk of head and neck cancer. Cancer Res 2006; 66: 11683–11689. 7. Sancar A, Lindsey-Boltz LA, Unsal-Kacmaz K, Linn S. Molecular mechanisms of mammalian DNA repair and the DNA damage checkpoints. Annu Rev Biochem 2004; 73: 39–85. 8. Bose S, Yap LF, Fung M et al. The ATM tumour suppressor gene is downregulated in EBV-associated nasopharyngeal carcinoma. J Pathol 2009; 217: 345–352. 9. Grabsch H, Dattani M, Barker L et al. Expression of DNA double-strand break repair proteins ATM and BRCA1 predicts survival in colorectal cancer. Clin Cancer Res 2006; 12: 1494–1500. 10. He Y, Chen Q, Li B. ATM in oral carcinogenesis: association with clinicopathological features. J Cancer Res Clin Oncol 2008; 134: 1013–1020. 11. Kang B, Guo RF, Tan XH et al. Expression status of ataxia-telangiectasia-mutated gene correlated with prognosis in advanced gastric cancer. Mutat Res 2008; 638: 17–25.
Volume 22 | No. 5 | May 2011
12. Xing J, Wu X, Vaporciyan AA et al. Prognostic significance of ataxiatelangiectasia mutated, DNA-dependent protein kinase catalytic subunit, and Ku heterodimeric regulatory complex 86-kD subunit expression in patients with nonsmall cell lung cancer. Cancer 2008; 112: 2756–2764. 13. Ye C, Cai Q, Dai Q et al. Expression patterns of the ATM gene in mammary tissues and their associations with breast cancer survival. Cancer 2007; 109: 1729–1735. 14. Parikh RA, White JS, Huang X et al. Loss of distal 11q is associated with DNA repair deficiency and reduced sensitivity to ionizing radiation in head and neck squamous cell carcinoma. Genes Chromosomes Cancer 2007; 46: 761–775. 15. Ai L, Vo QN, Zuo C et al. Ataxia-telangiectasia-mutated (ATM) gene in head and neck squamous cell carcinoma: promoter hypermethylation with clinical correlation in 100 cases. Cancer Epidemiol Biomarkers Prev 2004; 13: 150–156. 16. Bolt J, Vo QN, Kim WJ et al. The ATM/p53 pathway is commonly targeted for inactivation in squamous cell carcinoma of the head and neck (SCCHN) by multiple molecular mechanisms. Oral Oncol 2005; 41: 1013–1020. 17. Chiou SS, Huang JL, Tsai YS et al. Elevated mRNA transcripts of nonhomologous end-joining genes in pediatric acute lymphoblastic leukemia. Leukemia 2007; 21: 2061–2064. 18. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using realtime quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 2001; 25: 402–408. 19. Vo QN, Kim WJ, Cvitanovic L et al. The ATM gene is a target for epigenetic silencing in locally advanced breast cancer. Oncogene 2004; 23: 9432–9437. 20. Brieger J, Jacob R, Riazimand HS et al. Chromosomal aberrations in premalignant and malignant squamous epithelium. Cancer Genet Cytogenet 2003; 144: 148–155. 21. van den Broek GB, Wreesmann VB, van den Brekel MW et al. Genetic abnormalities associated with chemoradiation resistance of head and neck squamous cell carcinoma. Clin Cancer Res 2007; 13: 4386–4391. 22. Lazar AD, Winter MR, Nogueira CP et al. Loss of heterozygosity at 11q23 in squamous cell carcinoma of the head and neck is associated with recurrent disease. Clin Cancer Res 1998; 4: 2787–2793. 23. Handra-Luca A, Hernandez J, Mountzios G et al. Excision repair cross complementation group 1 immunohistochemical expression predicts objective response and cancer-specific survival in patients treated by cisplatin-based induction chemotherapy for locally advanced head and neck squamous cell carcinoma. Clin Cancer Res 2007; 13: 3855–3859. 24. Yang MH, Chiang WC, Chou TY et al. Increased NBS1 expression is a marker of aggressive head and neck cancer and overexpression of NBS1 contributes to transformation. Clin Cancer Res 2006; 12: 507–515. 25. Hitt R, Lopez-Pousa A, Martinez-Trufero J et al. Phase III study comparing cisplatin plus fluorouracil to paclitaxel, cisplatin, and fluorouracil induction chemotherapy followed by chemoradiotherapy in locally advanced head and neck cancer. J Clin Oncol 2005; 23: 8636–8645. 26. Vermorken JB, Remenar E, van Herpen C et al. Cisplatin, fluorouracil, and docetaxel in unresectable head and neck cancer. N Engl J Med 2007; 357: 1695–1704. 27. Posner MR, Hershock DM, Blajman CR et al. Cisplatin and fluorouracil alone or with docetaxel in head and neck cancer. N Engl J Med 2007; 357: 1705–1715. 28. Vermorken JB, Mesia R, Rivera F et al. Platinum-based chemotherapy plus cetuximab in head and neck cancer. N Engl J Med 2008; 359: 1116–1127. 29. Helleday T, Petermann E, Lundin C et al. DNA repair pathways as targets for cancer therapy. Nat Rev Cancer 2008; 8: 193–204. 30. Bryant HE, Schultz N, Thomas HD et al. Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) polymerase. Nature 2005; 434: 913–917. 31. Farmer H, McCabe N, Lord CJ et al. Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature 2005; 434: 917–921.
doi:10.1093/annonc/mdq569 | 1093
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
This work was supported by grants from the Ministry of Education to Kaohsiung Medical University (EM97-1.1ab-3); Kaohsiung Medical University Hospital (5D-27); Changhwa Christian Hospital (97-CCH-KMU-003, 95-CCH-KMU-25); and National Science Council (NSC97-2311-B-037-002-MY3, NSC 97-2314-B-371-004-MY3).
original article
Annals of Oncology 22: 1088–1093, 2011 doi:10.1093/annonc/mdq569 Published online 1 December 2010
Lower ataxia telangiectasia mutated (ATM) mRNA expression is correlated with poor outcome of laryngeal and pharyngeal cancer patients K.-W. Lee1,2,3, Y.-S. Tsai4,5, F.-Y. Chiang1,2, J.-L. Huang6, K.-Y. Ho1,2, Y.-H. Yang7, W.-R. Kuo1,2, M.-K. Chen8,9 & C.-S. Lin4,5* 1
original article
Received 12 September 2009; revised 13 March 2010 & revised 8 June 2010; accepted 16 August 2010
Background: Ataxia telangiectasia mutated (ATM) kinase is a critical regulator in initiating DNA damage response and activating DNA repair. However, the correlation between ATM expression and the outcome of laryngopharyngeal cancer patients is unknown. We hypothesize that ATM expression is correlated with a worse outcome in laryngopharyngeal cancer patients. Patients and methods: The ATM messenger RNA (mRNA) expression of 80 tumors of laryngeal and pharyngeal cancer was examined by real-time quantitative RT-PCR. Overall survival rates were measured using Kaplan–Meier estimates and the log-rank tests. The adjusted hazard rate ratios (HRRs) were computed by multivariate Cox regressions. Results: Reduced ATM mRNA was found in 65 of 80 studied cases. Lower ATM expression [tumor/normal <0.3, HRR = 2.49; 95% confidence interval (CI) 1.27–4.88], younger age (<55 years, HRR = 2.71; 95% CI 1.16–6.32), and larger tumor (T3/T4, HRR = 2.21; 95% CI 1.10–4.44) were independent risk factors for survival. Patients with lower ATM and younger age (HRR = 6.51; 95% CI 2.05–20.66) or with lower ATM and T3/T4 tumor (HRR = 5.23; 95% CI 2.04–13.40) exhibited the poorest outcome. Conclusion: The expression of ATM mRNA, which is frequently downregulated in laryngeal and pharyngeal cancers, could be a valuable prognostic marker. Key words: ataxia telangiectasia mutated, laryngeal cancer, pharyngeal cancer, survival analysis
introduction DNA damage response plays an important role in the maintenance of genome integrity. Defects or dysregulation of genes involved in DNA damage response and DNA repair can result in genomic instability, which is a common feature of cancer cells [1]. Head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer in the world. It has been suggested that genomic instability may be involved in the development of HNSCC [2–4]. Some studies have shown that DNA repair activity is reduced in the peripheral blood cells of HNSCC patients when compared with normal individuals [5, 6]. These indicate that interference of DNA damage repair may contribute
*Correspondence to: Dr C.-S. Lin, Graduate Institute of Medicine, College of Medicine, Kaohsiung Medical University, 100 Shih-Chuan 1st Road, Kaohsiung 807, Taiwan. Tel: +886-7-3121101-2019; Fax: +886-7-3218309; E-mail: [email protected]
to tumorigenesis of HNSCC. Nevertheless, studies on DNA repair gene expression in head and neck cancers are limited so far. Ataxia telangiectasia mutated (ATM) kinase is a critical regulator in DNA damage response and is conserved in all eukaryotes from yeast to human. It is activated immediately upon DNA damage and then phosphorylates p53, BRCA1, CHEK1/2, and many others, which lead to activation of DNA repair and cell cycle arrest [7]. Germline mutation of ATM gene results in ataxia telangiectasia, a syndrome susceptible to various cancers. Loss of ATM gene or decreased ATM expression is found in human cancers [8–14]. Parikh et al. [14] found that the ATM locus was partially lost in 8 of 11 HNSCC cell lines examined and in at least 60% of human malignancies including HNSCC. This partial loss of ATM locus was correlated with a reduced expression of ATM protein level in HNSCC cell lines [14]. In addition, Ai et al. [15] and Bolt et al. [16] have shown hypermethylation of ATM promoter in head and neck cancer, suggesting that ATM expression might be repressed. Indeed, He et al. [10] found an
ª The Author 2010. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected]
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
Department of Otolaryngology, Kaohsiung Medical University Hospital, Kaohsiung; 2Department of Otolaryngology, Faculty of Medicine, College of Medicine; Kaohsiung Medical University, Kaohsiung; 3Department of Otolaryngology, Kaohsiung Municipal Ta-Tung Hospital, Kaohsiung; 4Graduate Institute of Medicine, College of Medicine, Kaohsiung; 5Center of Excellence for Environmental Medicine, Kaohsiung Medical University, Kaohsiung; 6Department of Bioscience Technology, Chang Jung Christian University, Tainan; 7Department of Dental Hygiene, College of Dental Medicine, Kaohsiung Medical University, Kaohsiung; 8Department of Otorhinolaryngology-Head and Neck Surgery, Changhua Christian Hospital, Changhua; 9School of Medicine, Chung Shan Medical University, Taichung, Taiwan
original article
Annals of Oncology
absent or reduced ATM protein expression in 31.25% of oral squamous cell carcinoma, and the patients with reduced/absent ATM expression had poorer differentiated tumor and more lymph node invasion. Bose et al. [8] also showed that ATM messenger RNA (mRNA) and protein were downregulated in Epstein–Barr virus-associated nasopharyngeal carcinoma. However, whether ATM expression is also repressed in HNSCC at larynx and pharynx subsites and whether ATM expression is correlated with HNSCC patient outcome are unclear. In these regards, we examined ATM expression in laryngeal and pharyngeal cancers and correlated ATM expression with clinicopathologic parameters to explore the role of ATM expression in the prediction of patient outcome.
subjects The specimens of 40 laryngeal cancer patients (17 paired tumor and adjacent normal tissues plus 23 tumors only) and 40 pharyngeal cancer patients (15 paired tumor and adjacent normal tissues plus 25 tumors only) were obtained from the Department of Otolaryngology, Kaohsiung Medical University Hospital. These specimens were collected after obtaining patient’s informed consent, and the study was approved by the Institutional Review Board. Specimens were snap frozen in liquid nitrogen and were stored at 280C until use. The age of patients was between 33 and 82 years, with the median of 61 years, the first quartile of 54 years, and the third quartile of 70 years.
RNA isolation, reverse transcription, and real-time quantitative PCR The method for quantitatively analyzing mRNA expression was as described in our previous paper [17]. Briefly, total RNA was isolated by Tri-reagent (Sigma–Aldrich, St Louis, MO). Reverse transcription was conducted with 1 lg of total RNA and the High-Capacity cDNA Archive Kit (Applied BioSystems, Foster City, CA) in 20 ll. The resulting complementary DNA (cDNA) was subsequently diluted to a total volume of 100 ll and 2 ll of diluted cDNA was used as templates for real-time quantitative PCR (Q-PCR). Q-PCR was carried out in a total volume of 20 ll using PowerSYBR Green reagent (Applied BioSystems) and ABI Prism 7500 Sequence Detection System instrument (Applied BioSystems). The PCR condition was 50C for 2 min, 95C for 10 min followed by 50 cycles at 95C for 15 s and 60C for 1 min. Dissociation curve was added at the final step to validate that only the specific target was amplified. The PCR primers were designed using the ProbeFinder software (Roche Applied Science, Mannheim, Germany) and were ATM-F: ATAGATTGTGTAGGTTCCGATGG, ATM-R: CATCTTGTCTCAGGTCATCACG; ACTB-F: ATTGGCAATGAGCGGTTC, ACTB-R: GGATGCCACAGGACTCCAT. The ATM mRNA level in each sample was calculated by 22DDCT method [18] and ACTB was used as an internal control. Each specimen was examined by at least two independent rounds of reverse transcription and Q-PCR reactions for ATM and internal control. In all cases with paired specimens, the ATM mRNA expression of each case was represented by the ratio of ATM mRNA level in each tumor to that of the corresponding normal part. For the cases with tumor parts only, the mean ATM mRNA levels of all normal tissue specimens were used for comparison with those from each individual tumor.
statistical analyses The ATM mRNA expressions were categorized into two groups (<0.3 and ‡0.3) according to Receiver-Operating Characteristic analysis. Its relationship with clinicopathologic variables was analyzed by Chi-square or Fisher’s exact test. Kaplan–Meier estimates and the log-rank tests were
Volume 22 | No. 5 | May 2011
results ATM mRNA is downregulated in laryngeal and pharyngeal cancers To examine ATM expression in laryngeal and pharyngeal cancers, we harvested total RNA from 80 surgically resected and previously untreated tumors. The characteristics of patients were shown in Table 1. The ATM mRNA expression was determined by real-time quantitative RT-PCR using ACTB as an internal control. The results showed that ATM mRNA expression was decreased (tumor/normal <1) in 65 of 80 (81.3%) laryngeal and pharyngeal cancers (Table 1). The median of ATM mRNA expression (tumor/normal) in the 80 cases was 0.364, and the 25% and 75% quartiles were 0.222 and 0.762, respectively (Figure 1). These results demonstrated that ATM mRNA was frequently downregulated in laryngeal and pharyngeal cancers. ATM mRNA expression is correlated with survival of laryngeal and pharyngeal cancer patients To explore the possible role of decreased ATM expression in laryngeal and pharyngeal cancers, the correlation between ATM mRNA expression and clinicopathologic features was analyzed. We used the ratio of ATM mRNA expression of tumor versus normal at 0.3 as a cut-off point according to the ReceiverOperating Characteristic curve analysis, by which an association (P = 0.048, Chi-square analysis) between patient survival status and ATM mRNA expression was observed (Table 2). Patients
Table 1. Characteristics of patients Variable Total Gender Age (years) Tumor site Tumor size (T)
Lymph node (N)
Distant metastasis (M) Ataxia telangiectasia mutated (ATM) expressionb
Category
Na
%
Male Female <55 ‡55 Larynx Pharynx T1 T2 T3 T4 N0 N1 N2 N3 M0 M1 ‡1
80 73 7 25 55 40 40 19 18 18 20 42 14 14 5 74 1 15
100.0 91.3 8.7 31.3 68.7 50.0 50.0 25.3 24.0 24.0 26.7 56.0 18.7 18.7 7.7 98.7 1.3 18.7
<1
65
81.3
a
Five missing data for tumor size, lymph node, and distant metastasis. Ratio of tumor/normal.
b
doi:10.1093/annonc/mdq569 | 1089
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
patients and methods
employed to compare the survival rates for clinicopathologic variables. The hazard rate ratios (HRRs) and adjusted HRRs were further computed by univariate and multivariate Cox regressions.
original article
Annals of Oncology
with lower ATM mRNA expression (<0.3) exhibited a poorer overall survival (9 alive versus 20 dead) than those with ATM mRNA expression ‡0.3 (25 alive versus 21 dead). There was no significant association between ATM mRNA expression and other clinicopathologic parameters (Table 2).
We next analyzed 3-year and 5-year survival rates using Kaplan– Meier estimates with log-rank tests. The results demonstrated that patients with age <55 years (P = 0.004), with larger tumor size (T3 and T4, P = 0.012), and with lower ATM mRNA expression (<0.3, P = 0.015) had poorer survival rates (Table 3). Figure 2 shows a significant difference in survival rates between individuals with higher and lower ATM mRNA expression. These data indicated that ATM mRNA expression could be a prognostic marker for laryngeal and pharyngeal cancers.
Table 3. Kaplan–Meier survival rates Variable Overall Gender Age (years) Tumor size (T) Figure 1. Ataxia telangiectasia mutated (ATM) messenger RNA (mRNA) was downregulated in 65 of 80 laryngeal and pharyngeal cancers. The expression of ATM mRNA was determined using real-time quantitative RT-PCR and shown as a box plot that comprised the average of, at least, two independent experiments for each sample. The box represents upper (0.762) and lower (0.222) quartiles; the line in the box stands for the median (0.364).
Lymph node (N) Ataxia telangiectasia mutated (ATM) expressionb
Category
3 years (%)
5 years (%)
Male Female <55 ‡55 T1–T2 T3–T4 N0 N1–N3 <0.3
53.1 52.5 60.0 36.4 60.3 75.9 35.4 66.0 41.7 36.1
42.0 41.0 – 19.5 52.3 52.2 35.4 53.2 30.9 28.0
‡0.3
64.0
51.3
P-valuea 0.502 0.004 0.012 0.053 0.015
a
Log-rank test. Ratio of tumor/normal.
b
Table 2. Correlation between clinicopathologic variables and ataxia telangiectasia mutated (ATM) mRNA expressiona Variable
Total Gender Age Tumor size (T) Lymph node (N) Survival
Category
Nb
‡0.3 n % in category
<0.3 n % in category
Male Female <55 ‡55 T1–T2
80 73 7 25 55 37
50 46 4 17 33 25
62.5 63.0 57.1 68.0 60.0 67.6
30 27 3 8 22 12
37.5 37.0 42.9 32.0 40.0 32.4
T3–T4 N0
38 42
21 28
55.3 66.7
17 14
44.7 33.3
N1–N3 No Yes
33 41 34
18 21 25
54.5 51.2 73.5
15 20 9
45.5 48.8 26.5
P-valuec
>0.999d 0.493 0.274
0.285
0.048
a
Ratio of tumor/normal; cut-off point (0.3) was determined by the Receiver-Operating Characteristic curve. b Five missing data for tumor size, lymph node, and survival. c Chi-square test. d Fisher’s exact test.
1090 | Lee et al.
Figure 2. Overall survival curves were analyzed according to ataxia telangiectasia mutated (ATM) messenger RNA expression by using Kaplan–Meier estimate with log-rank test. The analyzed numbers (N) of the two categories are indicated.
Volume 22 | No. 5 | May 2011
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
lower ATM mRNA expression is an independent risk factor for survival of laryngeal and pharyngeal cancer patients To investigate whether lower ATM mRNA expression (<0.3) alone could predict poor outcome of laryngeal and pharyngeal cancers,
original article
Annals of Oncology
patients with lower ATM expression and younger age or larger tumor size have the poorest outcome We examined the effect of combining lower ATM mRNA expression with age or tumor size on predicting patient outcome. As shown in Figure 3, all seven patients in the <55-year-old group
Table 4. Multivariate Cox model analysis of overall survival Variable
Hazard rate ratio
95% Confidence interval Lower Upper
P-value
Ataxia telangiectasia mutated (ATM) expression (<0.3 versus ‡0.3) Gender (male versus female) Age (<55 versus ‡55 years) Tumor size (T3–T4 versus T1–T2) N stage (N1–N3 versus N0)
2.49
1.27
4.88
0.008
1.91 2.71 2.21
0.45 1.16 1.10
8.09 6.32 4.44
0.380 0.021 0.026
0.96
0.42
2.16
0.912
who have lower ATM mRNA expression (<0.3) died within 4 years. In addition, a significant proportion (13/16, 81.3%) of patients with lower ATM mRNA expression (<0.3) and larger tumor size (T3–T4) died within 3 years. On the contrary, patients with T1–T2 tumors or patients whose age was >55 years in combination with higher ATM mRNA expression (‡0.3) had better survival rates (Figure 3). Multivariate Cox regression model (Table 5) showed significant higher risks for individuals with lower ATM mRNA expression and younger age (HRR = 6.51; 95% CI 2.05–20.66; P = 0.001) or with lower ATM mRNA expression and T3/T4 tumor size (HRR = 5.23; 95% CI 2.04–13.40; P = 0.001). These results demonstrated that co-evaluation of ATM mRNA expression and age or tumor size might be a valuable tool for predicting patient outcome.
discussion ATM kinase plays a key role in initiating several DNA repair pathways. Downregulation of ATM may lead to broad dysfunction in DNA repair and the accumulation of genome alterations. In this study, we found that ATM mRNA was downregulated in 65/80 (81.3%) of laryngeal and pharyngeal cancers (Figure 1), and the lower ATM expression (<0.3) was an independent risk factor for patient’s survival (Table 4). To our knowledge, this is the first study showing that ATM expression is a valuable prognostic marker for HNSCC. It has been demonstrated that ATM expression was repressed in oral and nasopharyngeal carcinomas [8, 10], but the effect of ATM downregulation on patient’s outcome of these two malignancies was not shown. The prognostic role of ATM expression has been demonstrated in other malignancies, including breast [13], colon [9], lung [12], and gastric cancer [11]. All these studies showed that lower ATM expression was a risk factor for patient’s outcome, except for that in lung cancer [12]. Given the above, it seems that lower ATM expression is usually correlated with poor outcome of cancer patients. Moreover, we noted that the prognostic value of ATM expression could be largely enhanced by co-evaluating patient’s age or tumor size (Table 5). Both were also independent risk factors for patient’s survival (Table 4). Early onset of HNSCC
Figure 3. Overall survival curves according to ataxia telangiectasia mutated (ATM) messenger RNA expression combined with age (left panel) or tumor size (right panel). Analyses were conducted using Kaplan–Meier estimate with log-rank test. The analyzed number (N) of each category is indicated. Others (à and §) stand for ATM ‡ 0.3 and age <55 years or ATM < 0.3 and age ‡55 years (left panel); and ATM ‡ 0.3 and T3–T4 tumor or ATM < 0.3 and T1–T2 tumor (right panel).
Volume 22 | No. 5 | May 2011
doi:10.1093/annonc/mdq569 | 1091
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
we conducted multivariate Cox regression to examine the effect of lower ATM mRNA expression (Table 4). After adjusting for sex, age (<55 versus ‡55 years), tumor size (T3–T4 versus T1–T2), and lymph node metastasis (N1–N3 versus N0), the results showed that lower ATM mRNA expression (<0.3) remained a significant prognostic factor [HRR = 2.49; 95% confidence interval (CI) 1.27– 4.88; P = 0.008] for survival. Moreover, age (HRR = 2.71; 95% CI 1.16–6.32; P = 0.021) and tumor size (HRR = 2.21; 95% CI 1.10– 4.44; P = 0.026) were also independent factors for predicting overall survival (Table 4). These results indicated that lower ATM mRNA expression could be an independent risk factor for survival of laryngeal and pharyngeal cancer patients. When considering the ratio of ATM mRNA expression itself in the multivariate Cox regression analysis, a trend of association between ATM mRNA expression and overall survival could be found (supplemental Table S1, available at Annals of Oncology online), although statistical significance was not reached (P = 0.071).
original article
Annals of Oncology
Table 5. Multivariate Cox model analysis of overall survival (combining two factors) Variables
Hazard rate ratio
95% Confidence interval Lower Upper
P-value
a
0.001 0.017
0.001 0.179
a
Adjusted by sex, tumor size, lymph node invasion. Adjusted by sex, age, lymph node invasion.
b
implies that these patients may have certain genetic susceptibilities for tumor development. Large tumor size may reflect that the tumor cells have accumulated much more genetic alterations. Therefore, it is reasonable to combine ATM expression with these two independent risk factors as predictive parameters to better estimate the effect. One of the mechanisms underlying the downregulation of ATM mRNA expression may be hypermethylation of the ATM promoter. This results in suppression of promoter activity and has been demonstrated in HNSCC [15, 16] and other malignancies [19]. In this regard, our in vitro study showed that ATM promoter activity could be inhibited by ingredients of betel nut (C.-S. Lin, H.-Y. Hsieh, Y.-S. Tsai and J.-L. Huang, unpublished data), which is a popular substance for habitual use in India and Southeast Asia including Taiwan. However, the relationship between lower ATM expression and the habit of betel quid chewing in HNSCC patients remains to be established. Another mechanism accounting for ATM suppression may be the loss of ATM gene locus, by which reduced copy numbers at chromosome 11q22-23 has been demonstrated in HNSCC by comparative genomic hybridization [3, 20, 21]. In our studied cases, only one tumor specimen showed negative RT-PCR amplification (not included in the 80 cases described), implying that deletion of ATM gene might be possible for this case. However, since ATM mRNA could more or less be detected in all 80 tumors, complete gene deletions could not be the reason. Rather, the possibility of loss of one allele at chromosome 11q2223 should not be excluded. Indeed, Parikh et al. [14] found that ATM locus was partially lost in 8 of 11 HNSCC cell lines and all five clinical specimens examined. Lazar et al. [22] also showed loss of heterozygosity (LOH) at 11q23 in 25% (13/52) of primary HNSCC. These results support that loss of ATM locus is possibly one of the mechanisms accounting for lower ATM expression found in HNSCC. In addition to ATM, only a limited number of DNA repair genes have been examined for their expression and prognostic
1092 | Lee et al.
Volume 22 | No. 5 | May 2011
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
Ataxia telangiectasia mutated (ATM) expression and age ATM < 0.3 and age < 55 years 6.51 2.05 20.66 ATM ‡ 0.3 and age < 55 years 2.85 1.21 6.76 or ATM < 0.3 and age ‡ 55 years ATM ‡ 0.3 and age ‡ 55 years 1 ATM expression and tumor sizeb ATM < 0.3 and T3–T4 5.23 2.04 13.40 1.85 0.76 4.50 ATM ‡ 0.3 and T3–T4 or ATM < 0.3 and T1–T2 1 ATM ‡ 0.3 and T1–T2
role in head and neck cancers. Excision repair cross complementation group 1 (ERCC1), which removes cisplatininduced DNA adducts, plays an important role in both damage recognition and incision activities of nucleotide excision repair. It has been shown that lower ERCC1 expression was associated with a lower risk of death from advanced head and neck cancer and also benefited from cisplatin induction chemotherapy [23]. This is possibly correlated with the reduced repair of cisplatin– DNA adducts, which in turn leads to lethal DNA strand breaks and facilitates the eradication of tumor cells. On the other hand, the Nijmegen breakage syndrome gene product (NBN) that works together with MRE11A and RAD50 to recognize DNA double-strand breaks and initiate the signaling for DNA repair was shown to be an independent marker of poor prognosis for advanced HNSCC patients when overexpressed [24]. The authors also found that overexpression of NBN could increase cell transformation. These results suggest that the roles of DNA repair gene expressions in predicting patient outcome may be dependent on the characteristics of each gene and the pathways involved. In the present study, the patients were mostly treated with concurrent chemoradiotherapy after surgical resection. The relationship between ATM expression and patient treatments was not yet explored in detail. However, according to the study of Parikh et al. [14] that partial loss of ATM gene and reduced ATM expression is associated with a reduced sensitivity of HNSCC cells to ionizing irradiation (IR) and the finding of Lazar et al. [22] that LOH at ATM locus in HNSCC is associated with recurrent disease among patients who received radiotherapy, it is proposed that laryngopharyngeal tumors with lower ATM expression may be less responsive to radiotherapy and/or chemotherapy than those with higher ATM expression. As a result, patients with lower ATM mRNA expression may exhibit a poor outcome as shown in this study. This hypothesis suggests that a combination regimen exploiting different cell-killing mechanisms may be superior to monotherapy. Recent results obtained from several clinical trials support this hypothesis, by which another cell-killing mechanism via mitotic catastrophe induced by docetaxel can add the efficacy of chemotherapy using cisplatin and fluorouracil and improve the overall survival rates of patients with no distant metastatic HNSCC [25–27]. Another trial that combines platinum-based chemotherapy and cetuximab (by antagonizing the signaling of epidermal growth factor receptor) also improves the overall survival of head and neck cancer patients with recurrent or metastatic tumors [28]. Therefore, the concomitant findings of lower ATM expression and reduced IR response may, at least partly, explain why these trials that use combination therapy can improve survival of HNSCC patients. Recent studies on modulating DNA repair pathways to improve cancer treatments have become an attractive strategy [29]. Tumor cells survived from innumerable genetic alterations depend largely on multiple DNA repair pathways. However, defects of specific DNA repair pathway may arise during tumor development. Inhibition of another DNA repair pathway can therefore increase selectively killing of tumor cells that contain certain alterations of DNA damage response or DNA repair pathway. For example, poly
Annals of Oncology
(ADP-ribose) polymerase 1 (PARP1) inhibitors that interfere with base excision repair have been successfully used for treating the breast and ovarian cancer patients with inherited defects of BRCA1 or BRCA2 genes [30, 31]. Notably, such kind of treatment is less toxic to normal tissues than conventional chemoradiotherapy. Therefore, investigations of the genes involved in DNA damage responses and DNA repair pathways in HNSCC may help find better prognostic markers and strategies for patient treatment.
funding
acknowledgements We thank Ms Jenny Chang and Cathy Yang for critical reading of the manuscript. MK Chen and CS Lin contributed equally to this article.
disclosure The authors declare no conflict of interest.
references 1. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100: 57–70. 2. Friedlander PL. Genomic instability in head and neck cancer patients. Head Neck 2001; 23: 683–691. 3. Bockmuhl U, Schwendel A, Dietel M, Petersen I. Distinct patterns of chromosomal alterations in high- and low-grade head and neck squamous cell carcinomas. Cancer Res 1996; 56: 5325–5329. 4. Veltman JA, Bot FJ, Huynen FC et al. Chromosome instability as an indicator of malignant progression in laryngeal mucosa. J Clin Oncol 2000; 18: 1644–1651. 5. Cheng L, Eicher SA, Guo Z et al. Reduced DNA repair capacity in head and neck cancer patients. Cancer Epidemiol Biomarkers Prev 1998; 7: 465–468. 6. Paz-Elizur T, Ben-Yosef R, Elinger D et al. Reduced repair of the oxidative 8-oxoguanine DNA damage and risk of head and neck cancer. Cancer Res 2006; 66: 11683–11689. 7. Sancar A, Lindsey-Boltz LA, Unsal-Kacmaz K, Linn S. Molecular mechanisms of mammalian DNA repair and the DNA damage checkpoints. Annu Rev Biochem 2004; 73: 39–85. 8. Bose S, Yap LF, Fung M et al. The ATM tumour suppressor gene is downregulated in EBV-associated nasopharyngeal carcinoma. J Pathol 2009; 217: 345–352. 9. Grabsch H, Dattani M, Barker L et al. Expression of DNA double-strand break repair proteins ATM and BRCA1 predicts survival in colorectal cancer. Clin Cancer Res 2006; 12: 1494–1500. 10. He Y, Chen Q, Li B. ATM in oral carcinogenesis: association with clinicopathological features. J Cancer Res Clin Oncol 2008; 134: 1013–1020. 11. Kang B, Guo RF, Tan XH et al. Expression status of ataxia-telangiectasia-mutated gene correlated with prognosis in advanced gastric cancer. Mutat Res 2008; 638: 17–25.
Volume 22 | No. 5 | May 2011
12. Xing J, Wu X, Vaporciyan AA et al. Prognostic significance of ataxiatelangiectasia mutated, DNA-dependent protein kinase catalytic subunit, and Ku heterodimeric regulatory complex 86-kD subunit expression in patients with nonsmall cell lung cancer. Cancer 2008; 112: 2756–2764. 13. Ye C, Cai Q, Dai Q et al. Expression patterns of the ATM gene in mammary tissues and their associations with breast cancer survival. Cancer 2007; 109: 1729–1735. 14. Parikh RA, White JS, Huang X et al. Loss of distal 11q is associated with DNA repair deficiency and reduced sensitivity to ionizing radiation in head and neck squamous cell carcinoma. Genes Chromosomes Cancer 2007; 46: 761–775. 15. Ai L, Vo QN, Zuo C et al. Ataxia-telangiectasia-mutated (ATM) gene in head and neck squamous cell carcinoma: promoter hypermethylation with clinical correlation in 100 cases. Cancer Epidemiol Biomarkers Prev 2004; 13: 150–156. 16. Bolt J, Vo QN, Kim WJ et al. The ATM/p53 pathway is commonly targeted for inactivation in squamous cell carcinoma of the head and neck (SCCHN) by multiple molecular mechanisms. Oral Oncol 2005; 41: 1013–1020. 17. Chiou SS, Huang JL, Tsai YS et al. Elevated mRNA transcripts of nonhomologous end-joining genes in pediatric acute lymphoblastic leukemia. Leukemia 2007; 21: 2061–2064. 18. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using realtime quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 2001; 25: 402–408. 19. Vo QN, Kim WJ, Cvitanovic L et al. The ATM gene is a target for epigenetic silencing in locally advanced breast cancer. Oncogene 2004; 23: 9432–9437. 20. Brieger J, Jacob R, Riazimand HS et al. Chromosomal aberrations in premalignant and malignant squamous epithelium. Cancer Genet Cytogenet 2003; 144: 148–155. 21. van den Broek GB, Wreesmann VB, van den Brekel MW et al. Genetic abnormalities associated with chemoradiation resistance of head and neck squamous cell carcinoma. Clin Cancer Res 2007; 13: 4386–4391. 22. Lazar AD, Winter MR, Nogueira CP et al. Loss of heterozygosity at 11q23 in squamous cell carcinoma of the head and neck is associated with recurrent disease. Clin Cancer Res 1998; 4: 2787–2793. 23. Handra-Luca A, Hernandez J, Mountzios G et al. Excision repair cross complementation group 1 immunohistochemical expression predicts objective response and cancer-specific survival in patients treated by cisplatin-based induction chemotherapy for locally advanced head and neck squamous cell carcinoma. Clin Cancer Res 2007; 13: 3855–3859. 24. Yang MH, Chiang WC, Chou TY et al. Increased NBS1 expression is a marker of aggressive head and neck cancer and overexpression of NBS1 contributes to transformation. Clin Cancer Res 2006; 12: 507–515. 25. Hitt R, Lopez-Pousa A, Martinez-Trufero J et al. Phase III study comparing cisplatin plus fluorouracil to paclitaxel, cisplatin, and fluorouracil induction chemotherapy followed by chemoradiotherapy in locally advanced head and neck cancer. J Clin Oncol 2005; 23: 8636–8645. 26. Vermorken JB, Remenar E, van Herpen C et al. Cisplatin, fluorouracil, and docetaxel in unresectable head and neck cancer. N Engl J Med 2007; 357: 1695–1704. 27. Posner MR, Hershock DM, Blajman CR et al. Cisplatin and fluorouracil alone or with docetaxel in head and neck cancer. N Engl J Med 2007; 357: 1705–1715. 28. Vermorken JB, Mesia R, Rivera F et al. Platinum-based chemotherapy plus cetuximab in head and neck cancer. N Engl J Med 2008; 359: 1116–1127. 29. Helleday T, Petermann E, Lundin C et al. DNA repair pathways as targets for cancer therapy. Nat Rev Cancer 2008; 8: 193–204. 30. Bryant HE, Schultz N, Thomas HD et al. Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) polymerase. Nature 2005; 434: 913–917. 31. Farmer H, McCabe N, Lord CJ et al. Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature 2005; 434: 917–921.
doi:10.1093/annonc/mdq569 | 1093
Downloaded from http://annonc.oxfordjournals.org/ at Karolinska Institutet University Library on July 20, 2015
This work was supported by grants from the Ministry of Education to Kaohsiung Medical University (EM97-1.1ab-3); Kaohsiung Medical University Hospital (5D-27); Changhwa Christian Hospital (97-CCH-KMU-003, 95-CCH-KMU-25); and National Science Council (NSC97-2311-B-037-002-MY3, NSC 97-2314-B-371-004-MY3).
original article