Association of p53 Codon 72 Genotypes and Clinical Outcome in Human Papillomavirus- Infected Lung Cancer Patients

Transcription

1 GENERAL THORACIC Association of p53 Codon 72 Genotypes and Clinical Outcome in Human Papillomavirus- Infected Lung Cancer Patients Shih-Pin Chen, MD,* Nan-Yung Hsu, MD, PhD,* Jeng-Yuan Wu, MD, PhD, Chih-Yi Chen, MD, Ming-Chih Chou, MD, PhD, Huei Lee, PhD, and Ya-Wen Cheng, PhD Graduate Institute of Cancer Biology and Drug Discovery, Taipei Medical University, Taipei; Institute of Medicine, Chung Shan Medical University, Taichung; Department of Internal Medicine, Chung Shan Medical University Hospital, Taichung; Department of Thoracic Surgery, Taipei Medical University Hospital, Taipei; Division of Thoracic Surgery, Buddhist Tzu Chi General Hospital, Taichung Branch, Taiwan; School of Medicine, Tzu Chi University, Hualien; Department of Thoracic Surgery, China Medical University Hospital, Taichung; and Department of Thoracic Surgery, Chung Shan Medical University Hospital, Taichung, Taiwan, Republic of China Background. We recently reported that high-risk human papillomavirus (HPV) 16/18 E6 protein was associated with p53 protein degradation in lung cancer. The present study addressed the relationship between the different p53 genotypes and HPV oncoprotein expression with respect to p53 protein degradation and clinical outcome in primary lung cancer patients. Methods. We examined whether p53 codon 72 polymorphism and HPV oncoprotein expression could be associated with patients outcome by collecting 319 lung tumors from patients with non-small cell lung cancer to determine p53 codon 72 polymorphisms, HPV 16/18 infection, and HPV 16/18 E6 and p53 protein expression by polymerase chain reaction (PCR)-restriction fragment length polymorphism, nested-pcr, and immunohistochemical analysis. Results. The presence of HPV 16/18 DNA and E6 protein was inversely associated with p53 expression. The frequency of p53 protein degradation was also much higher in HPV 16/18 E6-positive/Arg/Arg lung tumors than in the other 3 groups. After adjusting gender and tumor type, the major contributors to p53 degradation in lung cancer patients were determined to be p53 codon72 polymorphism and HPV 16/18 E6 oncoprotein expression. This association was not found for HPV 16/18 DNA infection. Survival was significantly longer in patients with HPV 16/18 E6-negative/Arg/Arg tumors (median 32.7 months) than in patients with HPV-positive and p53 genetic variant tumors (p 0.008). Conclusions. The HPV 16/18 E6 protein, which is involved in the p53 inactivation that contributes to HPVinfected lung tumorigenesis, is associated with the p53 codon 72 genotype. The combination of HPV 16/18 E6 status and p53 codon72 polymorphism in lung tumors is an important biologic and prognostic parameter. (Ann Thorac Surg 2013;95: ) 2013 by The Society of Thoracic Surgeons Lung cancer has been the leading cause of cancer death for Taiwanese women since Although cigarette smoking is the major cause of lung cancer worldwide, more than 90% of the lung cancer in Taiwanese females is not related to cigarette smoking [1]. Therefore, Taiwanese women may have a unique etiology for lung cancer development. Our previous study indicated that human papillomavirus (HPV) oncogenic subtypes 16/18, which are involved in cervical cancer, may also be involved in the pathogenesis of lung cancer among Taiwanese women [2, 3]. Infection with high-risk HPV has been associated with several types of human carcinoma including cervical, bladder, esophageal carcinoma, as well as head and neck cancer, and lung cancers [2 7]. High-risk HPV DNAs are frequently integrated into cellular DNA in these cancers and transcribe E6 and E7 oncoproteins. The E6 and E7 oncoproteins of high-risk HPVs bind to the proteins encoded by the host p53 and Rb genes, which are tumor-suppressor genes, and disrupt the normal function of these genes in controlling the cell cycle [8]. The p53 gene is a tumor suppressor gene that regulates cell proliferation and it can inhibit the development or survival of cancer cells [9]. Inactivation of the p53 tumor suppressor gene has been found to occur in many types of human cancers [9]. An association between the BstUI Accepted for publication Dec 28, *Drs S-P. Chen and N-Y. Hsu contributed equally to this work. Address correspondence to Dr Cheng, Taipei Medical University, Graduate Institute of Cancer Biology and Drug Discovery, 12F, No.3, Bade Rd, Nangang Dist, Taipei 115, Taiwan; ywc@tmu.edu.tw. An Appendix for this article is available in the Auxiliary Annals section of the STS website: sts.org/annals-thoracic-surgery/auxiliary-annals by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc

2 Ann Thorac Surg CHEN ET AL 2013;95: P53 POLYMORPHISM AND HPV IN LUNG CANCER (Pro/Pro) genotype of the p53 codon 72 polymorphism and lung cancer has been reported previously [10, 11]. A tendency for a worse prognosis is seen in patients with the Pro/Pro and Arg/Arg genotypes than in those with the Arg/Pro genotype [10, 11]. Recently, in vivo and in vitro studies have indicated a greater susceptibility to degradation by HPV E6 for the p53 protein produced by the arginine allele at codon 72 than for the p53 protein produced by the proline allele [12]. Women with a homozygous p53 Arg/Arg genotype have at least a 7 times higher risk of acquiring HPV-induced cervical cancer [12]. This genotype also represents a significant risk factor for the development of other HPV-associated cancers [13, 14]. Our recent report [3] indicated that HPV 16/18 E6 is expressed in HPV DNA-positive lung tumors and is involved in the p53 inactivation that contributes to HPVinfected lung tumorigenesis. In this study, we hypothesized that the p53 protein degradation was associated with p53 codon 72 polymorphism in HPV-infected lung cancers. In addition, the clinical outcome of lung cancer patients may be adversely affected by p53 codon 72 polymorphism and HPV 16/18 E6 status. We therefore examined HPV16/18 DNA, HPV16/18 E6 protein, p53 protein expression and p53 codon72 polymorphism in tumor tissues from 319 primary lung cancer patients. The results obtained are discussed in terms of a role of p53 codon72 polymorphism in HPV-associated p53 protein degradation and lung tumorigenesis. Material and Methods Study Subjects Primary lung cancer patients who were admitted to the Department of Thoracic Surgery, Taichung Veteran s General Hospital, Taichung, Taiwan between 2002 and 2007 were asked to submit a written informed consent approved by the Institutional Review Board. None of these subjects had received radiation therapy or chemotherapy prior to surgery. Some of the collected lung tumors had been previously analyzed for the presence of HPV 16 and 18 DNA [2] and E6 protein [3]. Tumor types and stages were histologically determined according to the WHO classification (World Health Organization, 1981). Pathologic material was processed for conventional histologic procedures. Postoperative follow-up visits were scheduled for 1 and 2 months after surgery, then every 3 months during the first 2 years, and every 6 months thereafter, or more frequently if needed. The median duration of follow-up after a curative resection was 4.8 years. Nested Polymerase Chain Reaction for HPV16/18 DNA Detection Genomic DNA was prepared from tissue sections and isolated by conventional phenol-chloroform extraction and ethanol precipitation. The HPV viral DNA was first amplified with type consensus primers MY09 and MY11, followed by a second round of amplification with type-specific primers flanking the L1 region to identify the subtype [15]. A part of the -actin gene in all of the samples was 1197 amplified to exclude false-negative results and DNA preparations from SiHa cells (containing HPV 16) and HeLa cells (containing HPV 18) were used as positive controls. PCR-RFLP Analysis for p53 Codon72 Polymorphism Genomic DNA prepared from frozen tissue was used in the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) analysis. A 296 bp of fragment covering codon 72 of the p53 gene was amplified by a set of primers (5 ATCTACAGTCCCCCTTGCCG3 & 5 GCAACTGACCGTGCAAGTCA3 ). The PCR product was digested with 10 units of restriction enzyme BstUI (New England Biolabs, Ipswich, MA) at 60 C for 16 hours. The resulting DNA fragments were then separated on a 4% agarose gel and stained with ethidium bromide for analysis. Immunohistochemical Staining for p53 and HPV 16/18 E6 Protein Expression Formalin-fixed and paraffin-embedded specimens were sectioned at 3- m thickness. Sections were deparaffinized in xylene, rehydrated through serial dilutions of alcohol, and then washed in phosphate buffered saline. For p53 and HPV 16/18 E6 detection, sections were heated in a microwave in citrate buffer (ph 6.0) and then incubated with a monoclonal anti-human p53 antibody, DO7 (Dako, Glostrup, Denmark; dilution 1:250), or with polyclonal anti-hpv 16 or 18 E6 antibody (Santa Cruz, CA) [3]. The conventional streptavidin peroxidase method (Dako, LSAB Kit) was performed to develop the signals and the cells were counter-stained with hematoxylin. Negative controls, obtained by leaving out the primary antibody, were performed for every set of experiments to assure the quality of assay. The signals were evaluated independently by 3 observers and scored by the percentage of positive nuclei; score 0, no positive staining; score, positive staining. Positive control slides for p53 protein detection were purchased from Dako and cervical cancer tumor tissues with HPV 16/18 were used as a positive control for HPV 16/18 E6. Statistical Analysis The differences in p53 protein expression and HPV 16/18 infection with respect to clinical parameters were analyzed by a 2 test. Logistic regression analysis was used to assess which variable among gender, smoking status, tumor type, HPV 16/18 infection, and HPV 16/18 E6 expression was important for p53 protein degradation. Survival rates were estimated using the Kaplan-Meier method, and statistical analysis was carried out using the log-rank test for equality of the survival curves. Results Relationship Between p53, HPV16 E6, and HPV18E6 Protein Expression, and Clinical Parameters The associations between HPV 16 or 18 E6 protein expression and clinical factors were statistically analyzed (Table 1). As shown in Table 1, HPV 16/18 E6 protein was more frequently detected in female, adenocarcinoma, GENERAL THORACIC

3 GENERAL THORACIC 1198 CHEN ET AL Ann Thorac Surg P53 POLYMORPHISM AND HPV IN LUNG CANCER 2013;95: Table 1. Association of Clinical Parameters and the Presence of p53, HPV16 E6, and HPV 18E6 Protein in Lung Cancer Patients Parameters p53 HPV16 E6 HPV18 E6 Negative Positive Negative Positive Negative Positive (n 187) (%) (n 132) (%) (n 266) (%) (n 53) (%) (n 260) (%) (n 59) (%) Age (48.1) 66 (50.0) 128 (48.1) 28 (52.8) 120 (46.2) 36 (61.0) (51.9) 66 (50.0) 138 (51.9) 25 (47.2) 140 (53.8) 23 (39.0) p value Gender Female 68 (20.3) 33 (25.0) 71 (26.7) 30 (56.6) 60 (23.1) 32 (54.2) Male 119 (79.7) 99 (75.0) 195 (73.3) 23 (43.4) 191 (76.9) 27 (45.8) p value Smoking status Non-smoking 112 (59.9) 62 (47.0) 137 (51.5) 37 (69.8) 137 (52.7) 37 (62.7) Smoking 75 (40.1) 70 (53.0) 129 (48.5) 16 (30.2) 123 (47.3) 22 (37.3) P value Tumor type AD 114 (61.0) 60 (45.5) 136 (51.1) 38 (71.6) 135 (51.9) 39 (66.1) SQ 73 (39.0) 72 (54.5) 130 (48.9) 15 (28.4) 125 (48.1) 20 (33.9) p value Tumor stage I 69 (36.9) 50 (37.8) 98 (36.8) 21 (36.2) 100 (38.5) 19 (32.2) II 32 (17.1) 27 (20.5) 52 (19.6) 7 (12.1) 48 (18.5) 11 (18.6) III 86 (46.0) 55 (41.7) 116 (43.6) 25 (51.7) 112 (43.0) 29 (49.2) p value T factor 1 5 (2.7) 9 (6.8) 12 (4.5) 2 (3.8) 11 (4.3) 3 (5.1) (72.7) 90 (68.2) 188 (70.7) 38 (71.6) 192 (73.8) 34 (57.6) 3 32 (17.1) 27 (20.5) 48 (18.0) 11 (20.8) 44 (16.9) 15 (25.4) 4 14 (7.5) 6 (4.5) 18 (6.8) 2 (3.8) 13 (5.0) 7 (11.9) p value N factor 0 82 (43.9) 69 (52.3) 125 (47.0) 26 (49.0) 123 (47.3) 28 (47.4) 1 41 (21.9) 18 (13.6) 50 (18.8) 9 (17.0) 47 (18.1) 12 (20.3) 2 61 (32.6) 43 (32.6) 86 (32.4) 18 (34.0) 85 (32.7) 19 (32.3) 3 3 (1.6) 2 (1.5) 5 (1.8) 0 (0) 5 (1.9) 0 (0) p value p53 codon72 polymorphism Arg/Arg 48 (25.7) 63 (47.7) 97 (36.5) 14 (26.4) 91 (35.0) 20 (33.9) Arg/Pro 103 (55.1) 42 (31.8) 116 (43.6) 29 (54.7) 123 (47.3) 22 (37.3) Pro/Pro 36 (19.2) 27 (20.5) 53 (19.9) 10 (18.9) 46 (17.7) 17 (28.8) p value AD adenocarcinoma; Arg arginine; HPV human papillomavirus; Pro proline; SQ squamous cell carcinoma. and non-smoking lung cancer patients than in male, squamous cell carcinoma, and smoking patients. These results were consistent with our previous reports [2, 3]. A further understanding of the association of p53 and HPV 16/18 E6 protein was therefore sought by analyzing the p53 protein expression in lung tumor tissues. Most of our study population was p53 inactivated (187 of %, Table 1). Inactivation of p53 protein was higher in female lung cancer patients than in males. A higher proportion of p53 inactivation was seen in non-smokers than in smokers; and in adenocarcinomas than in squamous cell carcinomas. The p53 protein expression was inversely associated with expression of HPV16 E6, HPV18 E6, and HPV16/18 E6 (HPV16 E6 vs p53, p 0.006; HPV18 E6 vs p53, p 0.001; HPV16/18 E6 vs p53, p ; Table 2) but was not associated with the presence of HPV 16/18 DNA (HPV16 DNA vs p53, p 0.122; HPV18 DNA vs p53, p 0.242; HPV16/18 DNA vs p53, p 0.053; Table 2, Fig 1). P53 protein expression in HPV 16 E6, HPV 18, and HPV 16/18 E6 positive lung cancer patients were 24.5%,

4 Ann Thorac Surg CHEN ET AL 2013;95: P53 POLYMORPHISM AND HPV IN LUNG CANCER Table 2. Association of the Presence of HPV 16/18 DNA, HPV 16/18 E6 Protein and p53 Protein Degradation in Tumor Tissues of Lung Cancer Patients Parameters p53 Protein Negative Positive (n 187) (%) (n 132) (%) p Value HPV 16 DNA Negative 116 (55.5) 93 (44.5) Positive 71 (64.5) 39 (35.5) HPV 18 DNA Negative 110 (55.8) 87 (44.2) Positive 77 (63.1) 45 (36.9) HPV 16/18 DNA Negative 77 (52.7) 69 (47.3) Positive 110 (63.6) 63 (36.4) HPV 16 E6 Negative 147 (55.3) 119 (44.7) Positive 40 (75.5) 13 (24.5) HPV 18 E6 Negative 141 (54.2) 119 (45.8) Positive 46 (78.0) 13 (22.0) HPV 16/18 E6 Negative 117 (51.3) 111 (48.7) Positive 70 (76.9) 21 (23.1) HPV human papillomavirus. 22.0%, and 23.1%, significantly lower than in HPV E6 negative groups (p53 protein expression in HPV 16 E6 negative, HPV 18 E6 negative, and HPV 16/18 E6 negative were 44.7%, 45.8%, and 48.7%). These results suggested 1199 that the expression of HPV 16/18 E6 protein may be more important than the presence of HPV 16/18 DNA with respect to p53 protein degradation. The P53 Protein Degradation in HPV 16/18 E6-Positive Lung Cancer Patients Was Associated With P53 Codon72 Polymorphism The association between HPV 16/18 E6 and p53 genetic polymorphism in p53 degradation in lung cancer patients was analyzed by immunohistochemistry and PCR-RFLP. As shown in Appendix Table 1 (an Appendix for this article is available in the Auxiliary Annals section of the STS website: Appendix.pdf), the frequency of negative expression of the p53 protein was higher in HPV16 E6 E6-positive/Arg/Arg (85.7%, 18 of 21) lung cancer tumors than in the other 3 categories (59.3% in HPV16 E6- negative/arg/arg, 53.1% in HPV16 E6-negative/Arg/ Pro Pro/Pro, and 68.8% in HPV16 E6-positive/Arg/ Pro Pro/Pro; p 0.019). Similar results were observed for the HPV18 E6 (p 0.006) or HPV16/18 E6 positive groups (p ). Both p53 Codon72 Polymorphism and HPV 16/18 E6 Oncoprotein Are Contributors to p53 Degradation in Lung Cancer Patients A number of variables, including gender, smoking status, tumor type, p53 codon72 polymorphisms, HPV 16/18 DNA, and HPV 16/18 E6, were examined for their effects on p53 protein degradation in lung tissue by multivariate linear regression analysis. Patients with a p53 codon 72 Arg/Arg had an approximately fold risk of p53 GENERAL THORACIC Fig 1. The representative reciprocal relationships between human papillomavirus (HPV)16 or HPV18 E6 and p53 immunostainings in serial paraffin sections of lung tumors. Immunoreactivity was detected by immunoperoxidase-diaminobenzidine with hematoxylin counter stain. Note immunoreactivity (brown color) showing in the nuclei. (B) and (C) show HPV 16 E6 and p53 positive immunostaining in the nuclei; (A) and (B), (C) and (D), 2 serial sections from the same lung tumors. (A) p53 immunostaining positive; (B) HPV16 E6 immunostaining positive; (C) p53 immunostaining positive; (D) HPV16 E6 immunostaining negative. In (A) and (B) the bars 100 micrometers. In (C) and (D) bars 200 micrometers.

5 GENERAL THORACIC 1200 CHEN ET AL Ann Thorac Surg P53 POLYMORPHISM AND HPV IN LUNG CANCER 2013;95: Table 3. Multivariant Logistic Regression Analysis of the Risk of the p53 Protein Degradation in Lung Cancer Patients Parameters Favorable/Unfavorable OR 95% CI p Value Gender Female/male Smoking status Non-smoking/smoking Tumor type AD/SQ HPV 16/18 DNA Negative/positive HPV 16/18 E6 protein Negative/positive p53 codon 72 polymorphism Arg/Pro Pro/Pro/Arg/Arg AD adenocarcinoma; Arg arginine; CI confidence interval; HPV human papillomavirus; OR odd ratio; Pro proline; SQ squamous cell carcinoma. protein degradation when compared with the persons with p53 codon 72 Arg/Pro or Pro/Pro (95% confidence interval (CI) to 2.935, p 0.025). The patients who were HPV 16/18 E6 positive had an approximately fold risk of p53 protein degradation when compared with the persons who were HPV 16/18 E6 negative (95% CI to 5.988, p 0.001). The odds ratios of the other variables did not reach statistical significance (Table 3). Thus, both p53 codon72 polymorphism and HPV 16/18 E6 oncoprotein were contributors to p53 degradation in lung cancer patients. The Combination of HPV 16/18 E6 Status and a p53 Codon 72 Genetic Variant May Predict Survival in Lung Cancer We examined whether E6 and p53 codon 72 polymorphism could be associated with clinical outcome in lung cancer. In this study population, no patients received adjuvant chemotherapy before surgical resection. Ka- Fig 2. Kaplan-Meier analysis for the influence of (A) tumor stage; (B) human papillomavirus (HPV) 16/18 E6; (C) p53 codon 72 polymorphism; and (D) both combinations in all study subjects. (Arg arginine; Pro proline.)

6 Ann Thorac Surg CHEN ET AL 2013;95: P53 POLYMORPHISM AND HPV IN LUNG CANCER 1201 GENERAL THORACIC Fig 3. Kaplan-Meier analysis for the influence of 53 codon 72 polymorphism in (A) HPV 16/18 E6 negative, (B) HPV 16 E6 negative, and (C) HPV 18 E6 negative patients; and in (D) HPV 16/18 E6 positive, (E) HPV 16 E6 positive, and (F) HPV 16 E6 positive patients. (Arg arginine; HPV human papillomavirus; Pro proline.) plan-meier analysis showed that HPV16/18 E6 expression was not associated with overall survival (OS) in this study population (p 0.444, Fig 2A). Patients with the p53 Arg/ Arg genotype had longer OS than those did patients with Arg/Pro and Pro/Pro (p 0.026, p 0.001; Fig 2B). The prognostic significance of the combination of HPV16/18 E6 and p53 codon 72 polymorphism was indicated by Kaplan- Meier analysis. Patients with p53 Arg/Arg/HPV 16/18 E6- negative tumors had longer OS (59.2 months) than did patients with Arg/Arg/HPV 16/18 E6-positive tumors (21.8 months), Arg/Pro Pro/Pro/HPV 16/18 E6-negative tumors (21.2 months), and Arg/Pro Pro/Pro/HPV 16/18 E6- positive tumors (34.6 months) (p 0.008, Fig 2C). We further verified whether p53 codon 72 polymorphism might play different roles in HPV 16/18 E6 positive and negative groups. In the studied HPV 16/18 E6-negative population, patients with p53 Arg/Arg tumors had longer OS than did those with p53 Arg/Pro and Pro/Pro tumors (p 0.001, Fig 3A). The results were the same in HPV 16 E6 negative and HPV 18 E6 negative groups (p for HPV 16 E6, Fig 3B; p for HPV 18 E6, Fig 3C). However, in HPV 16 E6, HPV 18 E6, and HPV 16/18 E6 positive groups, the OS was not significantly different between patients with p53 Arg/Arg or Arg/Pro Pro/Pro tumors (Fig 3D F). These results clearly suggest that the presence of HPV 16/18 E6 negative plus p53 codon 72 Arg/Arg tumors may independently predict longer survival in lung cancer patients. Comment Several studies have reported better clinical outcomes for non-small lung cancer (NSCLC) patients with p53 codon72 Arg/Arg genotypes than with Arg/Pro or Pro/Pro genotypes [16 18]. However, our previous report [19] revealed that the p53 codon 72 Pro allele potentially increases the prognostic value of the p53 mutation in stage I NSCLC. These results would therefore seem inconsistent. In the present study, patients with Arg/Arg who had better clinical outcomes were only found in the HPV16/18 E6 negative groups and not in the positive groups (Fig 3A C). This means that HPV 16/18 E6 may play an important role in p53 function. Our previous report [20] indicated that p53 inactivation by E6 in lung cancer cells could activate Ape1 transcription and nuclear export of the protein, thereby subsequently promoting tumor malignancy through activation of the NF- B signaling pathway. The effect of p53 codon72 polymorphism on lung cancer clinical outcome seems to depend on the HPV 16/18 E6 status. Therefore,

7 GENERAL THORACIC 1202 CHEN ET AL Ann Thorac Surg P53 POLYMORPHISM AND HPV IN LUNG CANCER 2013;95: we suggest that p53 function is important in lung cancer progression. When tumors progress and become increasingly complex, it is probably difficult for a single genetic abnormality to define tumor behavior. We previously observed that the polymorphism of p53 codon 72 was a prognostic indicator in stage I patients. 19 In the present study, Kaplan-Meier analysis showed that p53 codon 72 polymorphism was associated with the overall survival in the total and the stage I study populations (p for all patients; p for stage I patients, Appendix Fig 1A and B), but not in stage II and stage III patients (p for stage II patients; p for stage III patients, Appendix Fig 1C and D. These results were consistent with our previous report [19]. In addition, as shown in Appendix Figure 2, no significant association was found between HPV 16/18 E6 and patient survival in the total patient population or in stage I, stage II, and stage III patients. In the total and stage I groups, patients with Arg/Arg/ HPV 16/18 E6- tumors had significantly longer survival when compared with the other 3 groups (Appendix Fig 3; an Appendix for this article is available in the Auxiliary Annals section of the STS website: Appendix.pdf). Based on these findings, we suggest that the combination of HPV 16/18 E6 status and p53 codon72 polymorphism is an important biologic and prognostic parameter of lung cancer, especially in stage I patients. Storey and colleagues [12] reported that patients with a p53 gene with homozygous arginine-72 alleles were about 7 times more susceptible to HPV-associated tumorigenesis than were heterozygotes; therefore, the arginine-encoding allele represented a significant risk factor for the development of HPV-associated cancers. The p53 codon 72 polymorphism has also been recently linked to the risk of cervical cancer. 15 The arginine allele at codon 72 of p53 was found to associate with a greater susceptibility for p53 degradation by HPV E6 protein than with the proline allele in both in vivo and in vitro studies [12 14, 21 25]. An association between HPV infection and p53 codon 72 polymorphism has also been reported for several types of cancers [12 14, 21 25], but the data are inconsistent. This inconsistency may perhaps result from the attempt to correlate p53 protein expression with HPV DNA infection rather than with the presence of HPV E6 protein. Current knowledge indicates that the effects of highrisk HPVs on disruption of normal p53 cell cycle control would appear to occur through direct interactions with the E6 protein and that these effects are independent of DNA levels. 8 In the present study, we found that p53 protein expression in lung tumors was inversely associated with HPV 16/18 E6 expression but was not associated with HPV 16/18 DNA levels (Table 2). Therefore, we considered that a discussion of the association of HPV16/18 E6, p53 codon 72 polymorphism, and p53 protein degradation was more pertinent to understanding the role of HPV 16/18 E6 in p53 protein degradation in lung tumors than were HPV DNA levels. Most previous studies used standard solution nested- PCR to detect the HPV infection in tumor tissues, which requires tissue digestion before DNA extraction [2, 15]. This presents 2 disadvantages for the analysis: (1) a high risk of molecular contamination because of the environment and the secondary amplification; and (2) the contamination by other types of cell in the tissues such as leukocytes, fibroblasts, and vascular endothelial cells. Although nested-pcr is more sensitive than in situ hybridization and immunohistochemistry for detection of HPV 16/18, the results from nested-pcr detection may be misleading due to contamination. In contrast, in situ molecular methods circumvent these problems by the use of formalin-fixed tissue sections on glass slides and observation of the end result with bright field microscopy. Immunohistochemistry uses specific antibodies for direct binding of the target protein. Both of these in situ molecular methods allow identification of the cell type that is positive for the target protein. In this study, we analyzed the p53 protein and HPV 16/18 E6 protein expression in lung tumor serial sections (Fig 1) and we observed an inverse association for expression of these proteins that was not evident by analysis of HPV DNA levels. Therefore, we considered that the use of immunohistochemistry or in situ hybridization methods may be more judicious for determining the role of HPV oncoprotein in p53 protein degradation in lung tumorigenesis. In summary, our findings suggest that the presence of HPV 16/18 E6 protein, but not HPV 16/18 DNA, was inversely associated with p53 expression in lung tumors. The frequency of p53 protein degradation in HPV 16/18 E6-positive/Arg homozygous lung cancer tumors was also much higher than that seen in the other 3 tumor categories investigated. A similar association was not found for HPV DNA infection. Patients with an HPV 16/18 E6 negative and a p53 Arg/Arg genotype had longer overall survival, but not in HPV-positive patients. In conclusion, HPV 16/18 E6 protein involvement in the p53 inactivation that contributes to HPV-infected lung tumorigenesis is associated with p53 codon 72 genotypes. The HPV status is an important biologic and prognostic parameter in NSCLC patients. This work was supported by grants from National Science Council ( B MY3; B MY3) and Buddhist Tzu Chi Taichung General Hospital (TTCRD-9801 and TTCRD-9909) of Taiwan. References 1. Chen CJ, Wu HY, Chuang YC, et al. Epidemiologic characteristics and multiple risk factors of lung cancer in Taiwan. Anticancer Res 1990;10: Cheng YW, Chiou HL, Sheu GT, et al. The association of human papillomavirus 16/18 infection with lung cancer among nonsmoking Taiwanese women. Cancer Res 2001;61: Cheng YW, Wu MF, Wang J, et al. Human papillomavirus 16/18 E6 oncoprotein is expressed in lung cancer and related with p53 inactivation. Cancer Res 2007;67: Leemans CR, Braakhuis BJ, Brakenhoff RH. The molecular biology of head and neck cancer. Nat Rev Cancer 2011;11: Veldhuijzen NJ, Snijders PJ, Reiss P, Meijer CJ, van de Wijgert JH. Factors affecting transmission of mucosal human papillomavirus. Lancet Infect Dis 2010;10:

8 Ann Thorac Surg CHEN ET AL 2013;95: P53 POLYMORPHISM AND HPV IN LUNG CANCER 6. Baltzell K, Buehring GC, Krishnamurthy S, Kuerer H, Shen HM, Sison JD. Limited evidence of human papillomavirus on breast tissue using molecular in situ methods. Cancer 2012;118: Hoots BE, Palefsky JM, Pimenta JM, Smith JS. Human papillomavirus type distribution in anal cancer and anal intraepithelial lesions. Int J Cancer 2009;124: zur Hausen H. Papillomaviruses causing cancer: evasion from host-cell control in early events in carcinogenesis. J Natl Cancer Inst 2000;92: Graziano V, De Laurenzi V. Role of p53 in cancer development. Biochim Biophys Acta 2011;1816: Yan L, Zhang D, Chen C, et al. TP53 Arg72Pro polymorphism and lung cancer risk: a meta-analysis. Int J Cancer 2009;125: Matakidou A, Eisen T, Houlston RS. TP53 polymorphisms and lung cancer risk: a systematic review and meta-analysis. Mutagenesis 2003;18: Storey A, Thomas M, Kalita A, et al. Role of a p53 polymorphism in the development of human papillomavirusassociated cancer. Nature 1998;393: Ji X, Neumann AS, Sturgis EM, et al. p53 codon 72 polymorphism associated with risk of human papillomavirusassociated squamous cell carcinoma of the oropharynx in never-smokers. Carcinogenesis 2008;29: Klug SJ, Ressing M, Koenig J, et al. TP53 codon 72 polymorphism and cervical cancer: a pooled analysis of individual data from 49 studies. Lancet Oncol 2009;10: Soini Y, Nuorva K, Kamel D, et al. Presence of human papillomavirus DNA and abnormal p53 protein accumulation in lung carcinoma. Thorax 1996;51: Han JY, Lee GK, Jang DH, Lee SY, Lee JS. Association of p53 codon 72 polymorphism and MDM2 SNP309 with clinical outcome of advanced nonsmall cell lung cancer. Cancer 2008;113: Nelson HH, Wilkojmen M, Marsit CJ, Kelsey KT. TP53 mutation, allelism and survival in non-small cell lung cancer. Carcinogenesis 2005;26: Wang YC, Chen CY, Chen SK, Chang YY, Lin P. p53 codon 72 polymorphism in Taiwanese lung cancer patients: association with lung cancer susceptibility and prognosis. Clin Cancer Res 1999;5: Chien WP, Wong RH, Wu TC, Cheng YW, Chen CY, Lee H. Potential increase in the prognostic value of p53 mutation by Pro72 allele in stage I non-small-cell lung cancer. Ann Surg Oncol 2009;16: Wu HH, Chu YC, Wang L, et al. Cytoplasmic Ape1 expression elevated by p53 aberration may predict survival and relapse in resected non-small cell lung cancer. Ann Surg Oncol 2012; Jun 12 [Epub ahead of print]. 21. Rogler A, Rogenhofer M, Borchardt A, et al. P53 codon 72 (Arg72Pro) polymorphism and prostate cancer risk: association between disease onset and proline genotype. Pathobiology 2011;78: Chen X, Sturgis EM, Etzel CJ, Wei Q, Li G. p73 G4C14-to- A4T14 polymorphism and risk of human papillomavirusassociated squamous cell carcinoma of the oropharynx in never smokers and never drinkers. Cancer 2008;113: Tornesello ML, Duraturo ML, Guida V, et al. Analysis of TP53 codon 72 polymorphism in HPV-positive and HPVnegative penile carcinoma. Cancer Lett 2008;269: Oliveira S, Sousa H, Santos AM, et al. The p53 R72P polymorphism does not influence cervical cancer development in a Portuguese population: a study in exfoliated cervical cells. J Med Virol 2008;80: Yang W, Zhang Y, Tian X, Ning T, Ke Y. p53 Codon 72 polymorphism and the risk of esophageal squamous cell carcinoma. Mol Carcinogene 2008;47: GENERAL THORACIC INVITED COMMENTARY A viral etiology of cancer is not a novel concept in the oncology field, inasmuch as some of the earliest scientific breakthroughs to an understanding of human cancers came from seminal laboratory work with tumors caused by viruses such as the Rous sarcoma virus. In a recent Lancet Oncology article, it was estimated that approximately 16%, or 2 million, of cancer cases worldwide can be attributable to an infectious cause [1], and just over 600,000 cases worldwide are associated specifically with human papillomavirus (HPV) infection [2]. The worldwide HPV prevalence averages 11%, but it can be as high as 20% to 30% in regions such as eastern Europe, eastern Africa, and the Caribbean. In the remainder of Europe, northern Africa, and Asia, the prevalence ranges between 7% and 10%. In North America, the prevalence of HPV is estimated at 5%. Although HPV prevalence is higher in less developed countries and lower in more developed ones, even the low prevalence rates seen in developed countries translates into real patients with real cancers. Much is now known about the oncogenic properties and mechanistic pathways of HPV infection in carcinogenesis. Although HPV 16 and HPV 18 are the subtypes most commonly associated with cervical carcinoma, the current article by Chen and colleagues [3] provides data to suggest that in lung cancer, more important than mere HPV infection are the combined effects of HPV oncogenic proteins E6 and E7 (although only E6 demonstrated strong associations in this study) and p53 polymorphisms, which result in increased p53 degradation and further loss of a key tumor suppressor in cells at risk. Another important point is that not all lung cancers are derived from inhaled carcinogens of smoking tobacco, and the mechanistic pathways discussed here constitute another possible pathway for lung cancer to develop in the nonsmoking patient. From an epidemiologic standpoint, many are cautiously optimistic that the recently developed HPV vaccine may be able to prevent many such cancers in the future. King F. Kwong, MD, FACS Thoracic Oncology, Surgery Branch National Cancer Institute, NIH 10 Center Dr Bethesda, MD kwongk2@mail.nih.gov References 1. de Martel C, Ferlay J, Franceschi S, et al. Global burden of cancers attributable to infections in 2008: a review and synthetic analysis. Lancet Oncol 2012;13: Forman D, de Martel C, Lacey CJ, et al. Global burden of human papillomavirus and related diseases. Vaccine 2012;30 (Suppl 5):F Chen SP, Hsu NY, Wu JY, et al. Association of p53 codon 72 genotypes and clinical outcome in human papillomavirusinfected lung cancer patients. Ann Thorac Surg 2013;95: by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc