Human papillomavirus infection is not associated with laryngeal squamous cell carcinoma in Taiwan

+

MODEL

Journal of Microbiology, Immunology and Infection (2018) xx, 1e8

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.e-jmii.com

Original Article

Human papillomavirus infection is not associated with laryngeal squamous cell carcinoma in Taiwan Chung-Guei Huang a,b,c, Li-Ang Lee d,e,**, Tuan-Jen Fang d,e, Hsueh-Yu Li d,e, Chun-Ta Liao d,e, Chung-Jan Kang d,e, Tzu-Chen Yen d,f, Kuo-Chien Tsao b,c, Tse-Ching Chen d,g, Shin-Ru Shih a,b,c,* a

Graduate Institute of Biomedical Sciences, Chang Gung University, Taoyuan, Taiwan Department of Medical Biotechnology and Laboratory Science, Chang Gung University, Taoyuan, Taiwan c Department of Laboratory Medicine, Linkou Chang Gung Memorial Hospital, Taoyuan, Taiwan d Department of Otorhinolaryngology e Head and Neck Surgery, Linkou Chang Gung Memorial Hospital, Taoyuan, Taiwan e Faculty of Medicine, Chang Gung University, Taoyuan, Taiwan f Department of Nuclear Medicine and Molecular Imaging Center, Linkou Chang Gung Memorial Hospital, Taoyuan, Taiwan g Department of Pathology, Linkou Chang Gung Memorial Hospital, Taoyuan, Taiwan b

Received 4 September 2017; received in revised form 27 November 2017; accepted 5 February 2018

Available online - - -

KEYWORDS Human papillomavirus; Laryngeal squamous cell carcinoma; Prevalence; Prospective cohort study; Risk factor

Abstract Background/Purpose: To examine whether the prevalence rate of human papillomavirus (HPV) infection in Taiwanese patients with primary laryngeal squamous cell carcinoma (LSCC) is different from that in those with a vocal polyp (VP) or vocal fold leukoplakia (VFL). Methods: This prospective cohort study recruited 41 consecutive patients with primary LSCC and 27 and 20 patients with VP and VFL, respectively. The HPV L1 gene in surgical specimens was detected using polymerase chain reaction. High-risk HPV DNA in tissue microarray specimens was detected using in situ hybridization. Expression of p16INK4a in tissue microarray specimens was determined through immunohistochemistry. Results: The prevalence of HPV L1 DNA in the LSCC group was equivalent to that in the VP and VFL groups (7.3% vs. 7.4% vs. 10.0%; P Z 0.929; effect size Z 0.20). High-risk HPV DNA

* Corresponding author. Research Center for Emerging Viral Infections, Chang Gung University, No. 259, Wenhua 1st Road, Guishan District, Taoyuan City 33302, Taiwan. Fax: þ886 3211 8700. ** Corresponding author. Department of Otorhinolaryngology, Head and Neck Surgery, Linkou Chang Gung Memorial Hospital, No. 5, Fu-Hsing Street, Gueishan District, Taoyuan City 33305, Taiwan. Fax: þ886 3397 9361. E-mail addresses: [email protected] (L.-A. Lee), [email protected] (S.-R. Shih). https://doi.org/10.1016/j.jmii.2018.02.002 1684-1182/Copyright ª 2018, Taiwan Society of Microbiology. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Please cite this article in press as: Huang C-G, et al., Human papillomavirus infection is not associated with laryngeal squamous cell carcinoma in Taiwan, Journal of Microbiology, Immunology and Infection (2018), https://doi.org/10.1016/j.jmii.2018.02.002

+

MODEL

2

C.-G. Huang et al. detected using in situ hybridization was relatively rare in all groups (2.6% vs. 5.3% vs. 0.0%; P Z 0.636; effect size Z 0.81). The prevalence of p16INK4a positivity was significantly lower in the LSCC group than in the VP and VFL groups (5.1% vs. 58.8% vs. 14.3%; P < 0.001). Multivariate analysis results revealed that age 65 years (adjusted odds ratio, 4.09; 95% confidence interval, 1.21e13.91; P Z 0.024) and p16INK4a positivity (adjusted odds ratio, 0.10; 95% confidence interval, 0.02e0.53; P Z 0.006) were LSCC risk factors. Conclusion: HPV infection is uncommon in Taiwanese patients with LSCC and seems not to be associated with an increased LSCC risk. Larger sample size is warranted for further study. Copyright ª 2018, Taiwan Society of Microbiology. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/).

Introduction

Materials and methods

Laryngeal squamous cell carcinoma (LSCC) is one of the most common cancers worldwide.1 The crude incidence of LSCC was estimated to be 2.2 per 100,000 persons worldwide in 2012 according to the International Agency for Research on Cancer and World Health Organization2 and 2.9 per 100,000 persons in 2013 in Taiwan.3 Tobacco smoking, alcohol consumption, old age, and being of male sex are the major conventional risk factors for LSCC.4,5 A study that evaluated patient records for the period 1980e20106 found that despite a reduction in risky oral habits (i.e., tobacco smoking and alcohol consumption), the incidence and mortality of LSCC in Taiwan still increased during this period (crude incidence: 1.0/100,000 in 1979 vs. 2.9/ 100,000 in 2013; crude mortality: 0.6/100,000 in 1979 vs. 0.9/100,000 in 2013).3 Thus, other risk factors for LSCC must be identified. In 2009, the International Agency for Research on Cancer advocated that “Human papillomavirus (HPV) may be associated with the development of LSCC with limited evidence in humans.”7 A previous meta-analysis of HPV infection and LSCC risk8 demonstrated that the overall prevalence of HPV in LSCC tissue was 28.0% (mainly types 16 and 18) and the summary odds ratio (OR) was 5.4 in patients with HPV infection compared with controls without HPV infection. A subsequent meta-analysis of HPV markers of head and neck squamous cell carcinoma (HNSCC)9 revealed a pooled HPV DNA prevalence estimate of 23.9% for LSCC. A previous study found that in the United States, female patients are more likely to have HPVpositive LSCC than are male patients, with an adjusted OR of 2.8.10 A recent meta-analysis of the prevalence of HPV infection in patients with LSCC concluded that the prevalence of HPV is higher in Chinese patients (overall prevalence: 32.0%) than in non-Chinese patients and that HPV infection significantly increases LSCC risk.11 Moreover, in our previous study, we observed a relatively high incidence of malignant transformation of laryngeal papilloma.12 However, no conclusion has been drawn regarding the significance of HPV infection in LSCC tumor tissue in Taiwan. Moreover, the relationship between HPV infection and LSCC remains to be investigated. The present study used a comprehensive approach to test the hypothesis that HPV infection increases the risk of LSCC.

This study was approved by the Institutional Review Board of Chang Gung Medical Foundation (first approval no.: 1004421B; revised approval no.: 102-3834C) and was conducted following the tenets of the Helsinki Declaration. Written informed consent was obtained from each patient before their enrollment in this study.

Patient cohorts This prospective cohort study initially recruited 89 patients with primary laryngeal lesions who underwent endoscopic surgery at the Outpatient Otolaryngology Clinic of Linkou Chang Gung Memorial Hospital in Taoyun City from August 1, 2012, to October 20, 2015. The inclusion criteria were 1) age >18 years, 2) a pathological diagnosis of laryngeal lesions, and 3) willingness to sign the informed consent form. The exclusion criteria were 1) unwillingness to answer subjective questionnaires and 2) unwillingness to undergo tissue examinations. All participants underwent extensive evaluation. For all participants, data on age at diagnosis, sex, tobacco smoking (ever/never), alcohol consumption (ever/never), and betel quid chewing (ever/never) were recorded at enrollment. Patients were categorized according to their levels of malignancy: 1, low potential (vocal polyp [VP]); 2, high potential (vocal fold leukoplakia [VFL]); and 3, LSCC (newly diagnosed with primary LSCC). Patients with an initial diagnosis of VFL who exhibited malignant transformation to LSCC within 1 year after initial surgery were excluded from statistical analysis. Clinical cancer status was graded according to the staging criteria of the 2009 American Joint Committee on Cancer (AJCC), 7th Edition.13

Detection of HPV L1 DNA Hematoxylin and eosin-stained slides from formalin-fixed paraffin-embedded (FFPE) tumor specimens were histologically evaluated for adequacy of tumor tissue (at least 10% tumor cells) and DNA extraction, as previously described.14 Three FFPE sections, each 5 mm thick, were cut from each biopsy specimen. DNA was extracted using a Lab Turbo 48 automatic nucleic acid extraction system and a Lab Turbo Virus Mini Kit LVN500 (LVN500; TaiGen Biotechnology Co.,

Please cite this article in press as: Huang C-G, et al., Human papillomavirus infection is not associated with laryngeal squamous cell carcinoma in Taiwan, Journal of Microbiology, Immunology and Infection (2018), https://doi.org/10.1016/j.jmii.2018.02.002

+

MODEL

HPV infection in primary laryngeal cancer in Taiwan Taipei city, Taiwan), as previously described.15,16 To produce a 192-bp DNA fragment, 1 mL eluted DNA solution was used as the polymerase chain reaction (PCR) template for the amplification of L1 DNA by using MY11/biotinylated GP6þ primers. HPV infection was diagnosed and subtyped using a commercially available HPV L1 gene PCR assay kit (EasyChip HPV Blot genotyping assay; King Car Biotechnology Co., Ltd., Yilan County, Taiwan).17 A total of 39 HPV types including 13 high-risk types (types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68) were determined.14e18 The detection limit of this assay to detect 39 HPV types was about 10e100 copies per FFPE sample.19 In the pilot study of this study, we confirmed the detection limit of the present protocol to be 100 copies per LSCC sample by serial dilution of plasmid DNA containing the specific segment of the L1 open reading frame of HPV 16 and a HPV 16-positive CaSki cell line. This cell line had been performed STR-PCR profile at the Taiwan Bioresource Collection and Research Center, Food Industry Research and Development Institute, Hsinchu, Taiwan and identified as their original cell lines (BCRC60251 [100% match]; HPV 16 viral load, 520 copies/ genome equivalent) as previously described.17

Tissue microarray construction This study examined 89 routine FFPE tissue blocks. Prior to tissue microarray (TMA) construction, LSCC, VP, and VFL foci were marked by our researcher on the corresponding hematoxylin and eosin-stained slides. Two separate areas per case were selected to ensure the representation of the selected foci for assembling the recipient blocks. We performed core biopsies (each 1.0 mm in diameter) of tissue from the corresponding donor blocks and placed them into a TMA recipient block by using a semiautomatic tissue arrayer (Beecher Instruments, Sun Prairie, WI, USA). For in situ hybridization (ISH) and immunohistochemistry (IHC) experiments, 4-mm-thick tissue sections of the TMAs were used.

ISH staining of high-risk HPV DNA For the in situ detection of high-risk HPV (types 16, 18, 31, 33, and 51) infection at the DNA level, biotin-labeled HPV probe solutions (Bond Ready-to-Use ISH HPV Probe, CAT # PB0829; Leica Biosystems Newcastle Ltd., Newcastle, United Kingdom) were applied to FFPE tissue sections by using a fully automated IHC and ISH immunostainer (BondMax; Leica Microsystems GmbH, Wetzlar, Germany) in accordance with the manufacturer’s instructions. The tumor tissues of a known HPV-positive cervical squamous cell carcinoma were included as positive controls. A known negative control (HPV-negative tongue squamous cell carcinoma) was also included in each run. The HPV ISH result was interpreted as positive if a minimum of one tumor cell showed punctate dot-like nuclear positivity.20

IHC staining of p16INK4a protein IHC staining of the cell cycle regulator p16INK4a (CAT #E6H4; Roche Diagnostics GmbH, Heidelberg, Germany) was performed. Using the same immunostainer, tissue sections of

3 TMA were stained with antibodies according to the manufacturer’s protocols. For optimal detection, sections were incubated with a 1:10 dilution of antibodies at room temperature for 20 min. A known HPV 6-positive adulthood recurrent respiratory papillomatosis (Fig. 1A) and known HPV-negative tongue squamous cell carcinoma were included as positive and negative controls in each run, respectively. After performing IHC staining on the sections, the slides were digitalized at 20 magnification by using a scanner (Aperio ScanScope, Leica Biosystems, Richmond, IL, USA). The expression of p16INK4a was then digitally assessed using semiquantitative image analysis software (Tissue Studio v.2.1, Definiens AG, Munich, Germany).21 We determined for staining intensities in the whole cells of four userspecified regions of interest containing laryngeal mucosa lesions and submucosal tissue. The staining intensity in cells was categorized as negative (Fig. 1B) or positive (weak, medium, or high staining intensity; Fig. 1C). Highly diffuse nuclear and cytoplasmic staining were indicative of transcriptionally active high-risk HPV infection (Fig. 1D).22 The cell positive index (%) ([number of positively stained cells/ total number of cells]  100) was calculated. Moreover, p16INK4a positivity was defined as an intensity score indicative of medium or high staining intensity in 30% of tumor cells.23

Statistical analysis Sample sizes were estimated using the method from a previous study on HPV infection22 in patients with LSCC. The prevalence of HPV infection was 45.6% in patients with laryngeal carcinoma and 9.1% in patients with vocal cord leukoplakia or polyps. We estimated the sample size for HPV DNA detection by using a priori calculation (proportions [Fisher’s exact test], two-tailed, a Z 0.05, power Z 0.95; allocation ratio Z 1) and found that a sample size of at least 39 patients was valid. We expected that 10% of the tissue would be insufficient for all examinations; therefore, we determined that 88 patients should be recruited for this study. Implementing the D’Agostino and Pearson normality test,24 the distribution of p16INK4a was found to be nonnormal. The descriptive statistics of this study are presented as median values and ranges. Differences in the variables between groups were analyzed using the ManneWhitney U test or KruskaleWallis test as appropriate. Categorical variables were analyzed using the chisquare test or Fisher’s exact test as appropriate. To estimate the risk of LSCC, coefficients estimated from the univariate and multivariate logistic regression models were used to derive the OR and 95% confidence interval (95% CI) of each covariate. Spearman correlation coefficients for all selected variables were computed. The effects of the variables found to be associated with LSCC (P  0.200) in the VP, VFL, and LSCC groups were analyzed using univariate and multivariate logistic regression models. For each risk factor of LSCC, the results of univariate analysis are expressed as unadjusted ORs, whereas those of multivariate analysis are expressed as adjusted ORs. Two-tailed P values of <0.05 were considered statistically significant. All

Please cite this article in press as: Huang C-G, et al., Human papillomavirus infection is not associated with laryngeal squamous cell carcinoma in Taiwan, Journal of Microbiology, Immunology and Infection (2018), https://doi.org/10.1016/j.jmii.2018.02.002

+

MODEL

4

C.-G. Huang et al.

Figure 1. Diversity of p16INK4a staining patterns demonstrated using immunohistochemistry. (A) Patchy nuclear and cytoplasmic staining in an patient with HPV 6 L1 DNA-positive recurrent respiratory papillomatosis. (B) Negative staining in a patient with HPV 68 L1 DNA-positive primary LSCC. (C) Patchy nuclear and cytoplasmic staining in a patient with HPV L1 DNA-negative primary LSCC. Weak, medium, and high nuclear and cytoplasmic staining are indicated. (D) Highly diffuse nuclear and cytoplasmic staining in a patient with HPV L1 DNA-negative primary LSCC.

calculations were performed using G*Power 3.1.9.2 software (Heinrich Heine University, Dusseldorf, Germany), GraphPad Prism 7.0 for Windows (GraphPad Software, Inc., San Diego, CA, USA), and SPSS 23.0 statistical package for Windows (SPSS Inc., Chicago, IL, USA).

proportions of patients exhibiting the habits of cigarette smoking, alcohol drinking, and betel quid chewing among the groups. In the LSCC group, the majority of patients had clinical early stage LSCC (80.5%).

HPV-related markers

Results Patient characteristics One VFL patient with malignant transformation in the 6th month was excluded from statistical analysis. Thus, this study included 88 patients (79 men and 9 women), comprising 27 patients with VP (median age, 52 years; range, 31e80 years), 20 with VFL (median age, 53 years; range, 33e78 years), and 41 with LSCC (median age, 64 years; range, 42e89 years). The median age of the LSCC group was significantly higher than those of the VP and VFL groups (P < 0.001). Table 1 depicts the general characteristics of the enrolled participants. The proportion of patients aged 65 years in the LSCC group was also significantly higher than those in the VP and VFL groups. No statistically significant differences were observed in the

Based on the PCR results, the overall prevalence rates of any HPV and high-risk HPV infection were 8.0% (n Z 7) and 3.4% (n Z 3), respectively (Table 2). HPV 6 infection was the most frequently encountered infection type in patients with laryngeal lesions (4.5%, n Z 4), whereas the prevalence rates of HPV 18, 11, 16, and 68 infections were 2.3% (n Z 2), 1.1% (n Z 1), 1.1% (n Z 1), and 1.1% (n Z 1), respectively. In the VFL group, single HPV 6 infection was detected in two patients (10.0%). In the VP group, single HPV 11 infection and single HPV 18 infection were each detected in one patient (3.7% and 3.7%, respectively). Furthermore, in the LSCC group, single HPV 6 infection, single HPV 68 infection, and multiple HPV infections were each detected in one patient (2.4%; types 6, 16 and 18, respectively). In the LSCC group, the prevalence rates of any HPV and high-risk HPV infection were 7.3% and 4.9%, respectively. No statistically significant differences were

Please cite this article in press as: Huang C-G, et al., Human papillomavirus infection is not associated with laryngeal squamous cell carcinoma in Taiwan, Journal of Microbiology, Immunology and Infection (2018), https://doi.org/10.1016/j.jmii.2018.02.002

+

MODEL

HPV infection in primary laryngeal cancer in Taiwan Table 1 Parameter

Clinical characteristics. All VP group VFL LSCC P (n Z 88) (n Z 27) group group values (n Z 20) (n Z 41)

Sex Female 9 (10.2) Male 79 (89.8) Age (year) <65 26 (29.5) 65 62 (70.5) Tobacco smoking Never 15 (17.0) Ever 73 (83.0) Alcohol consumption Never 31 (35.2) Ever 57 (64.8) Betel quid chewing Never 54 (61.4) Ever 34 (38.6) Cancer staging Tise2 e T3e4 e N0 e N1e2 e Stage 0eII e Stage IIIeIV e

0.224 5 (18.5) 1 (5.0) 3 (7.3) 22 (81.5) 19 (95.0) 38 (92.7) 0.001 1 (3.7) 6 (30.0) 19 (46.3) 26 (96.3) 14 (70.0) 22 (53.7) 0.099 8 (29.6) 3 (15.0) 4 (9.8) 19 (70.4) 17 (85.0) 37 (90.2) 0.092 14 (51.9) 6 (30.0) 11 (26.8) 13 (48.1) 14 (70.0) 30 (73.2) 0.977 17 (63.0) 12 (60.0) 25 (61.0) 10 (37.0) 8 (40.0) 16 (39.0) e e e e e e

e e e e e e

e 33 (80.5) 8 (19.5) 39 (95.1) 2 (4.9) 33 (80.5) 8 (19.5)

5 observed in the prevalence rates of any HPV and high-risk HPV infection among the groups (P Z 0.929 and 0.612, respectively; effect size Z 0.20 and 0.81; power Z 0.26 and 0.99, respectively). For some patients, lesion tissue appeared to be of insufficient quality to be used in the analyses of all HPV markers through ISH and IHC examinations. The ISH staining assay showed that positive nuclear staining of high-risk HPV DNA in the laryngeal lesions was observed in one of 27 patients with VP (5.3%) and one of the 41 patients with LSCC (2.6%). The distribution of the nuclear positivity of high-risk HPV DNA was equivalent in the VP and LSCC groups (P Z 0.636). IHC techniques were used to detect p16INK4a. The results showed that the cell positive indices of the VFL group (median, 25.9%, range, 5.5%e98.0%) and the LSCC group (median, 38.1%; range, 1.8%e99.3%) were significantly lower than that of the VP group (median, 66.2%, range, 20.6%e99.7%) (P Z 0.048). Furthermore, the p16INK4a positivity rate in the VP group (58.8%) was significantly higher than those in the VFL group (14.3%) and LSCC group (5.1%) (P < 0.001).

Associations of HPV-related markers and clinical variables

Data are expressed as sample numbers (%). LSCC Z laryngeal squamous cell carcinoma; Tis Z carcinoma in situ; VFL Z vocal fold leukoplakia; VP Z vocal polyp.

Table 3 shows the associations of HPV-related markers with clinical variables. Any HPV L1 DNA positivity was significantly associated with high-risk HPV L1 DNA, high-risk HPV DNA, and betel quid chewing. High-risk HPV L1 DNA was significantly associated with any HPV L1 DNA, high-risk HPV DNA, and betel quid chewing. p16INK4a positivity was associated with the level of malignancy.

Risk factors for LSCC Table 2

Summary of HPV-related markers.

Parameter All

VP group

VFL group

LSCC group

HPV markers Any HPV L1 DNA (PCR) (n Z 88) (n Z 27) (n Z 20) (n Z 41) Negative 81 (92.0) 25 (92.6) 18 (90.0) 38 (92.7) Positive 7 (8.0) 2 (7.4) 2 (10.0) 3 (7.3) High-risk HPV L1 DNA (PCR) (n Z 88) (n Z 27) (n Z 20) (n Z 41) Negative 85 (96.6) 26 (96.3) 20 (100.0) 39 (95.1) Positive 3 (3.4) 1 (3.7) 0 (0.0) 2 (4.9) High-risk HPV DNA (nuclear positive staining; ISH) (n Z 73) (n Z 19) (n Z 16) (n Z 38) Negative 71 (97.3) 18 (94.7) 16 (100.0) 37 (97.4) Positive 2 (2.7) 1 (5.3) 0 (0.0) 1 (2.6) p16INK4a positivity (IHC) (n Z 70) (n Z 17) (n Z 14) (n Z 39) Negative 56 (80.0) 7 (41.2) 12 (85.7) 37 (94.9) Positive 14 (20.0) 10 (58.8) 2 (14.3) 2 (5.1)

P values 0.929

0.612

0.636

<0.001

Data are expressed as sample numbers (%). HPV Z human papillomavirus; IHC Z immunohistochemistry; ISH Z in situ hybridization; LSCC Z laryngeal squamous cell carcinoma; PCR Z polymerase chain reaction; VFL Z vocal fold leukoplakia; VP Z vocal polyp.

Age, tobacco smoking, alcohol consumption, and p16INK4a positivity were included for the univariate and multivariate analyses (Table 4). The univariate analysis results revealed that age 65 years (unadjusted OR, 4.94; 95% CI, 1.80e13.56; P Z 0.002) and p16INK4a positivity (unadjusted OR, 0.09; 95% CI, 0.02e0.42; P Z 0.003) were risk factors of LSCC. In the multivariate analysis, age 65 years (adjusted OR, 4.09; 95% CI, 1.21e13.91; P Z 0.024) and p16INK4a positivity (adjusted OR, 0.10; 95% CI, 0.02e0.53; P Z 0.006) were associated with the development of LSCC.

Discussion Based on our results, the null hypothesis was rejected; in other words, HPV infections are not associated with an increased risk of LSCC in Taiwan. The study results also revealed that the DNA level was not related to LSCC. Thus, the results do not support the assertion that high-risk HPV infection plays a crucial role in the etiology of primary LSCC. By contrast, our data indicated that age 65 years and p16INK4a inactivation increase the risk of LSCC. In our previous study, low-risk HPV infection (types 6 and 11) was common in patients with laryngeal papilloma and a high malignant transformation rate was found in Taiwan.11 In the present study, we found that the negative conversion

Please cite this article in press as: Huang C-G, et al., Human papillomavirus infection is not associated with laryngeal squamous cell carcinoma in Taiwan, Journal of Microbiology, Immunology and Infection (2018), https://doi.org/10.1016/j.jmii.2018.02.002

+

MODEL

6

C.-G. Huang et al. Table 3

Spearman’s correlations between HPV-related markers and clinical variables.

Any HPV L1 DNA High-risk HPV L1 DNA High-risk HPV DNA p16INK4a positivity Male Age  65 years Tobacco smoking Alcohol consumption Betel quid chewing Level of malignancya

Any HPV L1 DNA

High-risk HPV L1 DNA

High-risk HPV DNA

p16INK4a positivity

e 0.639 (<0.001) 0.515 (<0.001) 0.102 (0.400) 0.099 (0.358) 0.086 (0.427) 0.133 (0.216) 0.129 (0.231) 0.284 (0.007) 0.008 (0.941)

0.639 e 0.811 0.071 0.063 0.016 0.085 0.139 0.237 0.044

0.515 (<0.001) 0.811 (<0.001) e 0.128 (0.296) 0.050 (0.673) 0.067 (0.574) 0.063 (0.597) 0.121 (0.307) 0.213 (0.071) 0.042 (0.727)

0.102 (0.400) 0.071 (0.562) 0.128 (0.296) e 0.153 (0.206) 0.211 (0.080) 0.192 (0.111) 0.046 (0.708) 0.015 (0.903) 0.504 (<0.001)

(<0.001) (<0.001) (0.562) (0.557) (0.885) (0.430) (0.198) (0.026) (0.685)

a

Patients were categorized according to the level of malignancy: 1, low potential (vocal polyp); 2, high potential (vocal fold leukoplakia); and 3, laryngeal squamous cell carcinoma. Data are expressed as r values (P values). HPV Z human papilloma virus.

Table 4

Univariate and multivariate logistic regression analysis results for LSCC risk. Total

Number of LSCC (%)

Univariate analysis Unadjusted OR (95% CI)

All 88 Age (years) 65 26 <65 62 Tobacco smoking Ever 73 Never 15 Alcohol consumption Ever 57 Never 31 p16INK4a positivity Positive 14 Negative 56 Unknown 18

Multivariate analysis P values

Adjusted OR (95% CI)

41 (46.6)

e

19 (73.1) 22 (35.5)

4.94 (1.80e13.56) Reference

0.002

4.09 (1.21e13.91) Reference

37 (50.7) 4 (26.7)

2.83 (0.82e9.70) Reference

0.099

e e

30 (52.6) 11 (35.4)

2.02 (0.82e4.97) Reference

0.126

e e

2 (14.3) 37 (66.1) 2 (11.1)

0.09 (0.02e0.42) Reference e

0.003

0.10 (0.02e0.53) Reference e

P values

e 0.024

0.006

Data are expressed as sample numbers (%). CI Z confidence interval; HPV Z human papillomavirus; OR Z odds ratio; LSCC Z laryngeal squamous cell carcinoma.

of HPV L1 DNA occurred in malignantly transformed tumors in patients with laryngeal papilloma. Our findings are further supported by those of Omland, who revealed that HPV-negative biopsies were more frequent in patients with adult-onset, recurrent respiratory papillomatosis and were associated with an increased risk of laryngeal neoplasia.25 However, Ndiaye et al.9 revealed that approximately onefourth of patients with LSCC had positive HPV DNA. Nevertheless, Young et al.23 and Laprise et al.26 considered the possibility that the prevalence rate of HPV infection in patients with LSCC was overestimated because HPV DNA PCR can detect active and inactive HPV infection.27 Many studies have found a low incidence of active HPV infection in primary LSCC by using HPV E6/E7 mRNA reverse transcriptase PCR,28 ISH,29 or p16INK4a IHC30 techniques. The use of IHC techniques to detect p16INK4a followed by HPV DNA detection has been validated to clinically detect oncogenically active HPV infection in patients with HNSCC.31 To date, the largest single study of HPV involvement enrolled 1042 patients with LSCC and demonstrated that based on

the positivity of HPV DNA and HPV E6 mRNA or p16INK4a, the proportion of LSCCs attributable to HPV was 3.5% (1.5% when requiring simultaneous positivity for all three markers).32 The present study confirmed that the incidence of oncogenically active HPV infection is low in patients with primary LSCC. The proportion of pooled p16INK4a positivity in LSCC was 86.3%.9 A recent study showed that an increased expression of p16INK4a is a robust surrogate marker of HPV infection.33 Notably, we found that p16INK4a positivity was associated with the level of malignancy and LSCC. Thus, we recommend that p16INK4a positivity be used as a surrogate marker of HPV infection. Furthermore, because of its downregulated expression (inactivation), p16INK4a is considered a tumor suppressor protein for HNSCC.34 Similar to a previous study,23 the present study found that the inactivation rather than overexpression of p16INK4a was a risk factor of primary LSCC. Taken together, we found that tobacco smoking and alcohol consumption were related to the level of

Please cite this article in press as: Huang C-G, et al., Human papillomavirus infection is not associated with laryngeal squamous cell carcinoma in Taiwan, Journal of Microbiology, Immunology and Infection (2018), https://doi.org/10.1016/j.jmii.2018.02.002

+

MODEL

HPV infection in primary laryngeal cancer in Taiwan malignancy, although the significance of their association with the development of LSCC was lower in the multivariate analysis results. Previous meta-analyses of tobacco smoking and alcohol consumption have shown strong risk trends in patients with LSCC.35,36 These data highlight the importance of controlling tobacco smoking and alcohol consumption; the impact of these behaviors became larger based on deaths attributable to these causes.37 There is a limitation of our study merits comment. A small effect size for comparing the proportion of any HPV infections among these groups was found in this study. Therefore, larger sample size may be required for comparing any HPV relatedness among three groups. Although the present study cannot tell the whole story of HPV infections in patients with LSCC, we believe that our results are interesting and warrant further study. In summary, this study found that the prevalence of oncogenically active HPV infection was very low in patients with primary LSCC. By contrast, p16INK4a inactivation and age 65 years were found to be independent risk factors of primary LSCC. These results suggest that HPV infection is not an etiological factor of LSCC in Taiwan. However, the association between HPV and LSCC warrants comprehensive laboratory examinations to determine the oncogenetic and prognostic impacts.

Conflicts of interest The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article.

Acknowledgments This project was supported by grants from Chang Gung Medical Foundation, Taoyuan, Taiwan (grants 3C0661, 3C0662, and 3C0663 to L.-A. L.).

References 1. Cattaruzza MS, Maisonneuve P, Boyle P. Epidemiology of laryngeal cancer. Eur J Cancer B Oral Oncol 1996;32B: 293e305. 2. Ferlay J, Soerjomataram I, Ervik M, et al. GLOBOCAN 2012 v1.0, Cancer incidence and mortality worldwide: IARC CancerBase No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer; 2013. Available from: http:// globocan.iarc.fr. [Accessed 10 July 2017]. 3. 2013 Taiwan Cancer Registry Annual Report. Available from: https://cris.hpa.gov.tw. [Accessed 10 July 2017]. 4. De Stefani E, Correa P, Oreggia F, Leiva J, Rivero S, Fernandez G, et al. Risk factors for laryngeal cancer. Cancer 1987;60:3087e91. 5. Zhang SS, Xia QM, Zheng RS, Chen WQ. Laryngeal cancer incidence and mortality in China, 2010. J Cancer Res Ther 2015; 11(Suppl 2):C143e8. 6. Liao CT, Wallace CG, Lee LY, Hsueh C, Lin CY, Fan KH, et al. Clinical evidence of field cancerization in patients with oral cavity cancer in a betel quid chewing area. Oral Oncol 2014; 50:721e31. 7. Bouvard V, Baan R, Straif K, Grosse Y, Secretan B, El Ghissassi F, et al. A review of human carcinogens-Part B: biological agents. Lancet Oncol 2009;10:321e2.

7 8. Li X, Gao L, Li H, Gao J, Yang Y, Zhou F, et al. Human papillomavirus infection and laryngeal cancer risk: a systematic review and meta-analysis. J Infect Dis 2013;207:479e88. 9. Ndiaye C, Mena M, Alemany L, Arbyn M, Castellsague ´ X, Laporte L, et al. HPV DNA, E6/E7 mRNA, and p16INK4a detection in head and neck cancers: a systematic review and meta-analysis. Lancet Oncol 2014;15:1319e31. 10. Hernandez BY, Goodman MT, Lynch CF, Cozen W, Unger ER, Steinau M, et al. Human papillomavirus prevalence in invasive laryngeal cancer in the United States. PLoS One 2014;9. e115931. 11. Zhang C, Deng Z, Chen Y, Suzuki M, Xie M. Is there a higher prevalence of human papillomavirus infection in Chinese laryngeal cancer patients? A systematic review and metaanalysis. Eur Arch Oto-Rhino-Laryngol 2016;273:295e303. 12. Lee LA, Cheng AJ, Fang TJ, Huang CG, Liao CT, Chang JT, et al. High incidence of malignant transformation of laryngeal papilloma in Taiwan. Laryngoscope 2008;118:50e5. 13. Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti A. AJCC cancer staging manual. 7th ed. New York: Springer; 2010. 14. Huang CG, Lee LA, Liao CT, Yen TC, Yang SL, Liu YC, et al. Molecular and serologic markers of HPV 16 infection are associated with local recurrence in patients with oral cavity squamous cell carcinoma. Oncotarget 2017;8:34820e35. 15. Lee LA, Huang CG, Liao CT, Lee LY, Hsueh C, Chen TC, et al. Human papillomavirus-16 infection in advanced oral cavity cancer patients is related to an increased risk of distant metastases and poor survival. PLoS One 2012;7:e40767. 16. Lee LA, Huang CG, Tsao KC, Liao CT, Kang CJ, Chang KP, et al. Increasing rates of low-risk human papillomavirus infections in patients with oral cavity squamous cell carcinoma: association with clinical outcomes. J Clin Virol 2013;57:331e7. 17. Huang CG, Lee LA, Tsao KC, Liao CT, Yang LY, Kang CJ, et al. Human papillomavirus 16/18 E7 viral loads predict distant metastasis in oral cavity squamous cell carcinoma. J Clin Virol 2014;61:230e6. 18. Huang SL, Chao A, Hsueh S, Chao FY, Huang CC, Yang JE, et al. Comparison between the Hybrid Capture II Test and an SPF1/GP6þ PCR-based assay for detection of human papillomavirus DNA in cervical swab samples. J Clin Microbiol 2006; 44:1733e9. 19. Huang HJ, Huang SL, Lin CY, Lin RW, Chao FY, Chen MY, et al. Human papillomavirus genotyping by a polymerase chain reaction-based genechip method in cervical carcinoma treated with neoadjuvant chemotherapy plus radical surgery. Int J Gynecol Cancer 2004;14(4):639e49. 20. Fatima N, Cohen C, Lawson D, Siddiqui MT. Automated and manual human papilloma virus in situ hybridization and p16 immunohistochemistry: comparison in metastatic oropharyngeal carcinoma. Acta Cytol 2013;57:633e40. 21. Braun M, Kirsten R, Rupp NJ, Moch H, Fend F, Wernert N, et al. Quantification of protein expression in cells and cellular subcompartments on immunohistochemical sections using a computer supported image analysis system. Histol Histopathol 2013;28:605e10. 22. Zhao S, Ye Q, Lu S. Detection of human papillomavirus in laryngeal carcinoma with digoxigenin labelled probe prepared by polymerase chain reaction. Zhonghua Shi Yan He Lin Chuang Bing Du Xue Za Zhi 1999;13:13e6. 23. Young RJ, Urban D, Angel C, Corry J, Lyons B, Vallance N, et al. Frequency and prognostic significance of p16INK4A protein overexpression and transcriptionally active human papillomavirus infection in laryngeal squamous cell carcinoma. Br J Cancer 2015;112:1098e104. 24. D’Agostino RB, Stephens MA. Goodness-of-fit techniques. New York: Marcel Dekker; 1986. 25. Omland T, Lie KA, Akre H, Sandlie LE, Jebsen P, Sandvik L, et al. Recurrent respiratory papillomatosis: HPV genotypes and

Please cite this article in press as: Huang C-G, et al., Human papillomavirus infection is not associated with laryngeal squamous cell carcinoma in Taiwan, Journal of Microbiology, Immunology and Infection (2018), https://doi.org/10.1016/j.jmii.2018.02.002

+

MODEL

8

26.

27.

28.

29.

30.

31.

C.-G. Huang et al. risk of high-grade laryngeal neoplasia. PLoS One 2014;9: e99114. Laprise C, Madathil SA, Schlecht NF, Castonguay G, Soulie `res D, Nguyen-Tan PF, et al. Human papillomavirus genotypes and risk of head and neck cancers: results from the HeNCe Life casecontrol study. Oral Oncol 2017;69:56e61. Wang H, Sun R, Lin H, Hu WH. P16INK4A as a surrogate biomarker for human papillomavirus-associated oropharyngeal carcinoma: consideration of some aspects. Cancer Sci 2013; 104:1553e9. Chernock RD, Wang X, Gao G, Lewis Jr JS, Zhang Q, Thorstad WL, et al. Detection and significance of human papillomavirus, CDKN2A(p16) and CDKN1A(p21) expression in squamous cell carcinoma of the larynx. Mod Pathol 2013;26: 223e31. Bhosale PG, Pandey M, Desai RS, Patil A, Kane S, Prabhash K, et al. Low prevalence of transcriptionally active human papilloma virus in Indian patients with HNSCC and leukoplakia. Oral Surg Oral Med Oral Pathol Oral Radiol 2016;122: 609e18. Thomas J, Primeaux T. Is p16 immunohistochemistry a more cost-effective method for identification of human papilloma virus-associated head and neck squamous cell carcinoma? Ann Diagn Pathol 2012;16:91e9. Rietbergen MM, Leemans CR, Bloemena E, Heideman DA, Braakhuis BJ, Hesselink AT, et al. Increasing prevalence rates of HPV attributable oropharyngeal squamous cell carcinomas in

32.

33.

34.

35.

36.

37.

The Netherlands as assessed by a validated test algorithm. Int J Cancer 2013;132:1565e71. Castellsague ´ X, Alemany L, Quer M, et al. HPV involvement in head and neck cancers: comprehensive assessment of biomarkers in 3680 patients. J Natl Cancer Inst 2016;108:djv403. Young RJ, Rischin D, Fisher R, McArthur GA, Fox SB, Peters LJ, et al. Relationship between epidermal growth factor receptor status, p16INK4A, and outcome in head and neck squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev 2011;20: 1230e7. Romagosa C, Simonetti S, Lo ´pez-Vicente L, Mazo A, Lleonart ME, Castellvi J, et al. p16Ink4a overexpression in cancer: a tumor suppressor gene associated with senescence and high-grade tumors. Oncogene 2011;30:2087e97. Zuo JJ, Tao ZZ, Chen C, Hu ZW, Xu YX, Zheng AY, et al. Characteristics of cigarette smoking without alcohol consumption and laryngeal cancer: overall and time-risk relation. A meta-analysis of observational studies. Eur Arch Oto-RhinoLaryngol 2017;274:1617e31. Corrao G, Bagnardi V, Zambon A, La Vecchia C. A meta-analysis of alcohol consumption and the risk of 15 diseases. Prev Med 2004;38:613e9. Lo WC, Ku CC, Chiou ST, Chan CC, Chen CL, Lai MS, et al. Adult mortality of diseases and injuries attributable to selected metabolic, lifestyle, environmental, and infectious risk factors in Taiwan: a comparative risk assessment. Popul Health Metr 2017;15:17.

Please cite this article in press as: Huang C-G, et al., Human papillomavirus infection is not associated with laryngeal squamous cell carcinoma in Taiwan, Journal of Microbiology, Immunology and Infection (2018), https://doi.org/10.1016/j.jmii.2018.02.002