Human papillomavirus (HPV) 16 and the prognosis of head and neck cancer in a geographical region with a low prevalence of HPV infection

Transcription

1 Cancer Causes Control (2014) 25: DOI /s ORIGINAL PAPER Human papillomavirus (HPV) 16 and the prognosis of head and neck cancer in a geographical region with a low prevalence of HPV infection Rossana Verónica Mendoza López José Eduardo Levi José Eluf-Neto Rosalina Jorge Koifman Sergio Koifman Maria Paula Curado Pedro Michaluart-Junior David Livingstone Alves Figueiredo Fabiano Pinto Saggioro Marcos Brasilino de Carvalho Luiz Paulo Kowalski Márcio Abrahão Francisco de Góis-Filho Eloiza Helena Tajara Tim Waterboer Paolo Boffetta Paul Brennan Victor Wünsch-Filho Received: 27 June 2013 / Accepted: 15 January 2014 / Published online: 29 January 2014 Ó Springer International Publishing Switzerland 2014 Abstract Background The role of human papillomavirus (HPV) on head and neck squamous cell carcinoma (HNSCC) survival in regions with low HPV prevalence is not yet clear. We evaluated the HPV16 infection on survival of HNSCC Brazilian patient series. Methods This cohort comprised 1,093 HNSCC cases recruited from 1998 to 2008 in four Brazilian cities and followed up until June HPV16 antibodies were analyzed by multiplex Luminex assay. In a subset of 398 fresh frozen or paraffin blocks of HNSCC specimens, we analyzed for HPV16 DNA by L1 generic primer polymerase chain reaction. HNSCC survival according to HPV16 antibodies was evaluated through Kaplan Meier method and Cox regression. Results Prevalence of HPV16 E6 and E6/E7 antibodies was higher in oropharyngeal cancer than in other head and neck tumor sites. HPV16 DNA positive in tumor tissue was also higher in the oropharynx. Seropositivity for HPV16 E6 antibodies was correlated with improved HNSCC survival and oropharyngeal cancer. The presence of HPV16 E6/E7 R. V. M. López (&) V. Wünsch-Filho Faculdade de Saúde Pública, Universidade de São Paulo, Av. Dr. Arnaldo, 715, São Paulo CEP , Brazil rossana@usp.br R. V. M. López Instituto de Ensino e Pesquisa, Hospital de Câncer de Barretos, Barretos, Brazil J. E. Levi Instituto de Medicina Tropical, Universidade de São Paulo, São Paulo, Brazil J. Eluf-Neto Faculdade de Medicina, Universidade de São Paulo, São Paulo, Brazil R. J. Koifman S. Koifman Escola Nacional de Saúde Pública, Fundação Oswaldo Cruz, Rio de Janeiro, Brazil M. P. Curado Hospital Araujo Jorge, Goiânia, Brazil M. P. Curado P. Boffetta International Prevention Research Institute, Lyon, France P. Michaluart-Junior Hospital das Clínicas, Faculdade de Medicina, Universidade de São Paulo, São Paulo, Brazil D. L. A. Figueiredo F. P. Saggioro Hospital das Clínicas, Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto, Brazil M. B. de Carvalho Hospital Heliópolis, São Paulo, Brazil L. P. Kowalski Hospital do Câncer A.C. Camargo, São Paulo, Brazil M. Abrahão Hospital São Paulo, Universidade Federal de São Paulo, São Paulo, Brazil F. de Góis-Filho Instituto do Câncer Arnaldo Viera de Carvalho, São Paulo, Brazil E. H. Tajara Faculdade de Medicina de São José do Rio Preto, São Paulo, Brazil

2 462 Cancer Causes Control (2014) 25: antibodies was correlated with improved HNSCC survival and oropharyngeal cancer survival. The death risk of oropharyngeal squamous cell carcinoma patients HPV16 E6/ E7 antibodies positive was 78 % lower than to those who test negative. Conclusion Oropharyngeal squamous cell carcinoma is less aggressive in the HPV16 E6/E7 positive serology patients. HPV16 E6/E7 antibody is a clinically sensible surrogate prognostic marker of oropharyngeal squamous cell carcinoma. Keywords Human papillomavirus Serology Prognosis Head and neck cancer Prevalence Introduction Head and neck squamous cell carcinomas (HNSCC) include tumors of the oral cavity, oropharynx, and larynx, which vary with respect to their etiology and prognosis. Worldwide, these tumors account for 650,000 new cases and 350,000 deaths every year [1]. Latin America, particularly Brazil, has a relatively high incidence of HNSCC. Among both sexes, the rates of oral cavity and larynx cancer were 2.8 and 2.2/100,000, respectively, in 2008 [2]. The relative survival rate for HNSCC is approximately %, including all sites, clinical stages, and forms of treatment. In Brazil, the 5-year survival rate for oral and oropharyngeal cancer is \50 % [3]. Tobacco smoking and alcohol consumption are the most important risk factors for HNSCC [4]. The International Agency for Research on Cancer recognizes human papillomavirus (HPV) type 16 as the only type of HPV that is carcinogenic to organs other than the cervix uteri, including the anus, penis, vagina, vulva, oral cavity, oropharynx, and tonsils [5]. HPV16 has been associated with increased risk of head and neck cancer, particularly oropharynx and tonsil tumors in high as well as low HPV prevalence regions [6 10]. In regions with high HPV prevalence, studies have shown that HPV16-positive patients have a better HNSCC prognosis than those who are HPV16-negative [11 14]. However, the impact of HPV16 on HNSCC survival in regions with low HPV prevalence is not known. T. Waterboer German Cancer Research Center (DKFZ), Heidelberg, Germany P. Boffetta The Tisch Cancer Institute, Mount Sinai School of Medicine, New York, NY, USA P. Brennan International Agency for Research on Cancer, Lyon, France We analyzed the relationship between HPV16 infection and the overall survival of patients with HNSCC and cancer in specific sites (oral cavity, oropharynx, hypopharynx, and larynx) in Brazil, a country with a low prevalence of HPV infection. Materials and methods Study subjects This study includes incident HNSCC cases confirmed by histology and diagnosed between 1998 and The cases originated from two multicentre studies in Brazil. The Latin American study was conducted from 1998 to 2002 in seven cities; for the present study, we only included cases from the cities of Goiânia, Rio de Janeiro, and São Paulo. The Gencapo study was conducted from 2003 to 2008 in the cities of Ribeirão Preto and São Paulo. Both studies were approved by the clinical centres ethics committees and by the National Commission on Ethics in Research. Written consent was given by each patient participating in the studies. HNSCC was classified into one of four categories according to anatomical subsites, each of which has a distinct prognosis: oral cavity, oropharynx, hypopharynx, and larynx. The International Classification of Diseases, version 10 [15], was used to classify the tumor subsite according to the method used by Hashibe and colleagues [16]: oral cavity (C00.3 C00.9, C02.0 C02.3, C02.8, C02.9, C03.0, C03.1, C03.9, C04.0, C04.1, C04.8, C04.9, C05.0, C05.8, C05.9, C06.0 C06.2, C06.8, C06.9, C14.0, C14.2, and C14.8), oropharynx (C01.9, C02.4, C05.1, C05.2, C09.0, C09.1, C09.8, C09.9, C10.0 C10.4, C10.8, and C10.9), hypopharynx (C12.9, C13.0 C13.2, C13.8, and C13.9), and larynx (C32.0 C32.3, C32.8, and C32.9). All patients underwent face-to-face interviews immediately after diagnosis with trained interviewers who used a structured questionnaire to obtain information about variables that could affect HNSCC survival, including tobacco smoking, alcohol consumption, and education. Patient hospital records were reviewed to obtain additional information on tumor clinical stage, treatment, and time since diagnosis until dead or last information. The tumor s clinical stage was classified as CS I IV according to the TNM classification system, 6th edition [17]. Additional information on cause of death was validated through death certificates obtained from the São Paulo State Death Registry (for patients from Ribeirão Preto and São Paulo) and from the Goiânia Population Cancer Registry (for patients from Goiânia). In Rio de Janeiro, death certificates were included in the hospital medical records.

3 Cancer Causes Control (2014) 25: Table 1 Baseline characteristics of HNSCC study patients, Brazil, All patients Oral cavity Oropharynx Hypopharynx Larynx n = 1,093 n = 321 n = 252 n = 115 n = 405 n (%) n (%) n (%) n (%) n (%) Study Latin America 727 (66.5) 181 (56.4) 200 (79.4) 84 (73.0) 262 (64.7) GENCAPO 366 (33.5) 140 (43.6) 52 (20.6) 31 (27.0) 143 (35.3) Sex Female 148 (13.5) 56 (17.4) 39 (15.5) 7 (6.1) 46 (11.4) Male 945 (86.5) 265 (82.6) 213 (84.5) 108 (93.9) 359 (88.6) Age (years) \ (39.8) 145 (45.2) 115 (45.6) 43 (37.4) 132 (32.6) (34.1) 101 (31.5) 91 (36.1) 39 (33.9) 142 (35.1) (20.6) 55 (17.1) 37 (14.7) 24 (20.9) 109 (26.9) C75 60 (5.5) 20 (6.2) 9 (3.6) 9 (7.8) 22 (5.4) Education (years) (14.6) 41 (12.8) 37 (14.7) 18 (15.7) 64 (15.8) (53.9) 163 (50.8) 144 (57.1) 63 (54.8) 219 (54.1) (20.7) 73 (22.7) 39 (15.5) 25 (21.7) 89 (22.0) C (10.8) 44 (13.7) 32 (12.7) 9 (7.8) 33 (8.1) Tobacco smoking Nonsmoker 53 (4.8) 21 (6.5) 15 (6.0) 3 (2.6) 14 (3.5) Former smoker 273 (25.0) 61 (19.0) 50 (19.8) 30 (26.1) 132 (32.6) Current smoker 767 (70.2) 239 (74.5) 187 (74.2) 82 (71.3) 259 (64.0) Alcohol consumption Nondrinker 97 (8.9) 35 (10.9) 13 (5.2) 5 (4.3) 44 (10.9) Former drinker 390 (35.7) 82 (25.5) 101 (40.1) 50 (43.5) 157 (38.8) Current drinker 606 (55.4) 204 (63.6) 138 (54.8) 60 (52.2) 204 (50.4) Tumor stage T1/T2 366 (33.5) 133 (41.4) 83 (32.9) 24 (20.9) 126 (31.1) T3 325 (29.7) 60 (18.7) 84 (33.3) 44 (38.3) 137 (33.8) T4 402 (36.8) 128 (39.9) 85 (33.7) 47 (40.9) 142 (35.1) Treatment Surgery 347 (31.7) 129 (40.2) 39 (15.5) 20 (17.4) 159 (39.3) Radiotherapy 214 (19.6) 36 (11.2) 73 (29.0) 35 (30.4) 70 (17.3) Surgery? radiotherapy 335 (30.6) 114 (35.5) 66 (26.2) 31 (27.0) 124 (30.6) Other 197 (18.0) 42 (13.1) 74 (29.4) 29 (25.2) 52 (12.8) Patients were recruited into the study from November 1998 to December 2008, and they were followed until 30 June Of the 1,275 eligible cases at the start of the study, 182 were excluded because no information was available on tobacco smoking, alcohol consumption, education, tumor clinical stage, or treatment. Ultimately, 1,093 cases with complete follow-up records were included in the analysis. Biological samples In both studies, blood was obtained from each patient at the time of the interview. Tumor tissue samples from biopsies or surgical procedures were obtained for a subset of 398 cases (36 %), including 198 fresh-frozen samples and 200 paraffin-embedded samples. DNA extraction and detection of HPV DNA For patient samples from the Gencapo study, DNA was extracted from paraffin slices using the Nucleon HT kit (GE Life Sciences, São Paulo, Brazil) and examined using the Inno-LiPA HPV Genotyping kit (Innogenetics, Gent, Belgium) [18]. For patient samples from the Latin American study, DNA was extracted from fresh tumor tissue using the QIAamp DNA Mini kit (Qiagen, Valencia, CA,

4 464 Cancer Causes Control (2014) 25: Table 2 Characteristics of HNSCC patients by HVP DNA and HPV16 DNA status, Brazil, Total n = 398 HPV DNA p HPV16 DNA p Negative Positive Negative Positive n = 363 n = 35 n = 384 n = 14 n (%) n (%) n (%) n (%) Sex Female (84.0) 8 (16.0) 46 (92.0) 4 (8.0) Male (92.2) 27 (7.8) 338 (97.1) 10 (2.9) Age (years) \ (87.0) 21 (13.0) 155 (95.7) 7 (4.3) (93.6) 9 (6.4) 136 (96.5) 5 (3.5) (93.4) 5 (6.6) 74 (97.4) 2 (2.6) C (100.0) 0 19 (100.0) 0 Tobacco smoking Nonsmoker (86.7) 2 (13.3) 13 (86.7) 2 (13.3) Former smoker (88.8) 12 (11.2) 103 (96.3) 4 (3.7) Current smoker (92.4) 21 (7.6) 268 (97.1) 8 (2.9) Alcohol consumption Nondrinker (84.2) 6 (15.8) 34 (89.5) 4 (10.5) Former drinker (87.1) 18 (12.9) 136 (97.1) 4 (2.9) Drinker (95.0) 11 (5.0) 214 (97.3) 6 (2.7) Tumor Site Oral cavity (93.4) 8 (6.6) 117 (96.7) 4 (3.3) Oropharynx (93.4) 6 (6.6) 87 (95.6) 4 (4.4) Hypopharynx (93.2) 3 (6.8) 43 (97.7) 1 (2.3) Larynx (87.3) 18 (12.7) 137 (96.5) 5 (3.5) Treatment Surgery (89.5) 15 (10.5) 136 (95.1) 7 (4.9) Radiotherapy (100) 0 76 (100) 0 Surgery? radiotherapy (89.3) 13 (10.7) 118 (96.7) 4 (3.3) Other (87.7) 7 (12.3) 54 (94.7) 3 (5.3) 1 Fisher s exact test. 2 chi-square test USA). HPV DNA detection was performed by polymerase chain reaction (PCR) with the generic primers PGMY09/ 11, which amplify a fragment spanning *450 bp of the L1 region of most mucosal HPV types [19], in the presence of human b-globin primers, which amplify a fragment of 268 bp as a positive control. The reaction conditions were as follows: 200 lm dntps, 4 mm magnesium chloride, 80 nm PGMY09/11, and 20 nm PCO4/GH20 (b-globin) oligonucleotides, 250 ng of template DNA, and 1 U of Taq polymerase (Invitrogen, São Paulo, Brazil). The thermocycling profile consisted of an initial incubation of 5 min at 94 C, followed by 40 cycles of 94 C (1 min), 55 C (1 min), and 72 C (1 min) and a final elongation step of 5 min at 72 C in a PE 2400 (Applied Biosystems, Foster City, CA, USA) or a MasterCycler gradient thermal cycler (Eppendorf AG, Hamburg, Germany). The PCR products were analyzed by electrophoresis on a 2 % agarose gel stained with ethidium bromide and observed under ultraviolet light. Further genotyping of PGMY09/11-positive samples was conducted by restriction fragment length polymorphism using the enzymes and patterns described by Bernard et al. [20]. These procedures and analyses were conducted at the Institute of Tropical Medicine at the University of São Paulo in Brazil. Detection of HPV antibodies Antibodies against HPV were analyzed by multiplex Luminex serology assay. This antibody detection method is based on a glutathione S-transferase capture ELISA [21] in combination with fluorescent bead technology [22]. To determine a positive serological response, mean fluorescence intensity values were dichotomised as antibody positive or negative for the serological response to oncoproteins E1, E2, E4, E6, E7, and L1 of HPV16; oncoproteins E6, E7, and L1 of HPV 18, 31, 33, and 35;

5 Cancer Causes Control (2014) 25: Table 3 Characteristics of HNSCC patients by HPV16 E6, E7, L1, E6/E7 serology antibodies, Brazil, Total HPV16 E6 HPV16 E7 HPV16 L1 HPV16 E6/E7 Negative Positive Negative Positive Negative Positive Negative Positive n = 1,013 n = 80 n = 833 n = 260 n = 1,018 n = 75 n = 1,066 n = 27 n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) Sex Female (91.9) 12 (8.1) 116 (78.4) 32 (21.6) 132 (89.2) 16 (10.8) 141 (95.3) 7 (4.7) Male (92.8) 68 (7.2) 717 (75.9) 228 (24.1) 886 (93.8) 59 (6.2) 925 (97.9) 20 (2.1) p Age (years) \ (90.8) 40 (9.2) 337 (77.5) 98 (22.5) 403 (92.6) 32 (7.4) 420 (96.6) 15 (3.4) (92.5) 28 (7.5) 285 (76.4) 88 (23.6) 349 (93.6) 24 (6.4) 365 (97.9) 8 (2.1) (96.0) 9 (4.0) 164 (72.9) 61 (27.1) 208 (92.4) 17 (7.6) 221 (98.2) 4 (1.8) C (95.0) 3 (5.0) 47 (78.3) 13 (21.7) 58 (96.7) 2 (3.3) 60 (100.0) 0 p Tobacco smoking Nonsmoker (81.1) 10 (18.9) 36 (67.9) 17 (32.1) 47 (88.7) 6 (11.3) 47 (88.7) 6 (11.3) Former smoker (92.7) 20 (7.3) 208 (76.2) 65 (23.8) 255 (93.4) 18 (6.6) 265 (97.1) 8 (2.9) Current smoker (93.5) 50 (6.5) 589 (76.8) 178 (23.2) 716 (93.4) 51 (6.6) 754 (98.3) 13 (1.7) p \0.001 Alcohol Nondrinker (92.8) 7 (7.2) 77 (79.4) 20 (20.6) 88 (90.7) 9 (9.3) 95 (97.9) 2 (2.1) Former drinker (94.9) 20 (5.1) 302 (77.4) 88 (22.6) 365 (93.6) 25(6.4) 384 (98.5) 6 (1.5) Drinker (91.3) 53 (8.7) 454 (74.9) 152 (25.1) 565 (93.2) 41 (6.8) 587 (96.9) 19 (3.1) p Tumor Site Oral cavity (93.8) 20 (6.2) 246 (76.6) 75 (23.4) 302 (94.1) 19 (5.9) 316 (98.4) 5 (1.6) Oropharynx (89.3) 27 (10.7) 191 (75.8) 61 (24.2) 237 (94.0) 15 (6.0) 238 (94.4) 14 (5.6) Hypopharynx (98.3) 2 (1.7) 86 (74.8) 29 (25.2) 105 (91.3) 10 (8.7) 115 (100.0) 0 Larynx (92.3) 31 (7.7) 310 (76.5) 95 (23.5.) 374 (92.3) 31 (7.7) 397 (98.0) 8 (2.0) p Treatment Surgery (91.9) 28 (8.1) 254 (73.2) 93 (26.8) 319 (91.9) 28 (8.1) 339 (97.7) 8 (2.3) Radiotherapy (93.5) 14 (6.5) 166 (77.6) 48 (22.4) 199 (93.0) 15 (7.0) 210 (98.1) 4 (1.9) Surgery? radiotherapy (92.5) 25 (7.5) 259 (77.3) 76 (22.7) 313 (93.4) 22 (6.6) 324 (96.7) 11 (3.3) Other (93.4) 13 (6.6) 154 (78.2) 43 (21.8) 187 (94.9) 10 (5.1) 193 (98.0) 4 (2.0) p chi-square test or Fisher s exact test oncoproteins E6 and L1 of 45, 52, and 58; and oncoprotein L1 of HPV 6, 11, and 77. Means plus three standard deviations were calculated to define cutoffs, excluding outliers [23]. These procedures and analyses were conducted at the German Cancer Research Center in Heidelberg, Germany. Statistical analysis Frequencies and percentages were calculated for categorical variables, and the associations between variables were tested using Fisher s exact test or chi-square tests. The concordance between HPV DNA and serological response was calculated using the Kappa statistic. Cox model regression was performed to calculate the hazard ratios (HR) and 95 % confidence intervals (95 % CI) for the association of HPV on HNSCC overall survival, adjusting for gender, age (years\55, 55 64, 65 74, and C75), study group (Latin America or Gencapo), education (years of schooling: 0, 1 7, 8 10, and C11), tobacco smoking (nonsmoker, former smoker, current smoker), alcohol consumption (nondrinker, former drinker, current drinker),

6 466 Cancer Causes Control (2014) 25: Table 4 Concordance between HPV16 DNA and HPV16 serological antibodies by HNSCC subsite, Brazil, Oral cavity Kappa Oropharynx Kappa Hypopharynx Kappa Larynx Kappa HPV16 DNA DNA p 1 HPV16 DNA p 1 HPV16 DNA p 1 HPV16 p 1 Negative Positive Negative Positive Negative Positive Negative Positive n = 117 n = 4 n = 87 n = 4 n = 43 n = 1 n = 137 n = 5 n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) HPV16 E NA Negative 110 (94.0) 4 (100.0) (96.6) 1 (25.0) \ (100.0) (89.8) 3 (60.0) (100.0) Positive 7 (6.0) 0 3 (3.4) 3 (75.0) (10.2) 2 (40.0) HPV16 E Negative 82 (70.1) 2 (50.0) (72.4) 2 (50.0) (69.8) 1 (100.0) (73.7) 3 (60.0) Positive 35 (29.9) 2 (50.0) 24 (27.6) 2 (50.0) 13 (30.2) 0 36 (26.3) 2 (40.0) HPV16 L Negative 108 (92.3) 4 (100.0) (96.6) 2 (50.0) \ (88.4) 1 (100.0) (91.2) 4 (80.0) Positive 9 (7.7) 0 3 (3.4) 2 (50.0) 5 (11.6) 0 12 (8.8) 1 (20.0) HPV NA E6/E7 Negative 114 (97.4) 4 (100.0) (97.7) 2 (50.0) \ (100.0) 133 (97.1) 4 (80.0) (100.0) Positive 3 (2.6) 0 2 (2.3) 2 (50.0) (2.9) 1 (20.0) 1 p for Kappa statistic NA not available tumor clinical stage (T1/T2, T3, and T4), and treatment (surgery, radiotherapy, surgery plus radiotherapy, another treatment). The dependence between oral cavity, hypopharynx, and larynx cancer (combined) survival and oropharynx cancer (separately) survival on the presence of HPV16 antibodies was evaluated using the Kaplan Meier method, and the differences between the curves were assessed using log-rank tests. For the descriptive analysis and multiple survival analyses, we used IBM Ò SPSS Ò Statistics version 18 (IBM Corp, Somers, NY, USA) and Stata/SE 11.0 for Windows (StataCorp LP, College Station, TX, USA). All reported p values are two-sided with a significance level of Results At the end of the follow-up period, 356 of the study patients were alive, 626 had died, and 111 were lost to follow-up. The median survival of the cohort members was 32 months (95 % CI months). HNSCC 5-year overall survival after diagnosis was 38 %, and diseasespecific survival was 43 %. The description of study group, demographic characteristics, tobacco and alcohol consumption, clinical stage and treatment for all patients with serological response to HPV is presented in Table 1 and corresponding to all patients with serological samples. Of the HNSCC cases, 321 (29.4 %) affected the oral cavity, 252 (23.1 %) affected the oropharynx, 115 (10.5 %) affected the hypopharynx, and 405 (37.1 %) affected the larynx. The majority of the patients were male, and 37 % were between 51 and 60 years old. The education level was low; 68.5 % had \8 years of schooling. Only 4.8 % of the HNSCC patients reported that they did not smoke, and 8.9 % did not report alcohol consumption at the time of the interview. An advanced clinical tumor stage (T3 or T4) at diagnosis was observed in many of the HNSCC cases, and the most common treatment approach was surgery or a combination of surgery plus radiotherapy. The overall prevalence of HPV DNA was 8.8 % and for HPV16 DNA was 3.5 % in tumor tissue. Table 2 shows the prevalence of HPV DNA and HPV16 DNA according to gender, age, tobacco smoking, alcohol consumption, anatomical site, and treatment. Association statistically significant was observed in alcohol consumption and treatment with HPV DNA status. The HPV16 DNA prevalence in tumor tissue was higher in women, patients 50 years old or younger, nonsmokers, and nondrinkers. In addition, the prevalence of HPV16 DNA was higher in patients with oropharyngeal cancer than in patients with cancers of other head and neck sites. No association was observed between treatment and HPV16 DNA status; however, we observed that all patients who received radiotherapy were negative to HPV16 DNA. A total of 8.8 % of HNSCC patients tested positive for HPV DNA in tumor tissue (35/398), of which 40 % had HPV16 (14/35). Table 3 displays the prevalence of HPV16 antibodies by gender, age, tobacco smoking, alcohol consumption, anatomical site, and treatment. The overall prevalence of

7 Cancer Causes Control (2014) 25: Table 5 Overall survival of HNSCC patients by HPV16 serological antibodies and HPV16 DNA, Brazil, n HNSCC p n Oral cavity p n Oropharynx p n Hypopharynx p n Larynx p HR 1 (95 % CI) HR 1 (95 % CI) HR 1 (95 % CI) HR 1 (95 % CI) HR 1 (95 % CI) HPV16 E6 Negative 1, Positive ( ) ( ) ( ) ( ) ( ) HPV16 E7 Negative Positive ( ) ( ) ( ) ( ) ( ) HPV16 L1 Negative 1, Positive ( ) ( ) ( ) ( ) ( ) HPV16 E6/E7 Negative 1, NA Positive ( ) ( ) ( ) NA ( ) HPV16 DNA Negative NA Positive ( ) ( ) ( ) NA ( ) Hazard ratio adjusted for study, gender, age, education, tobacco smoking, alcohol consumption, clinical tumor stage, and treatment. NA: not available 1 HPV16 antibodies among HNSCC patients was as follows: E6 (7.3 %), E7 (23.8 %), L1 (6.9 %), and E6/E7 (2.5 %). HPV16 E6 and E6/E7 antibodies had the lowest prevalence overall among the HNSCC patients, but the prevalence of these antibodies was higher among younger patients, nonsmokers, and those with oropharyngeal cancer. According to treatment, the data showed no association with serological response to HPV. Relative frequencies were distributed similarly in each treatment and antibodies. In general, the concordance of HPV16 DNA and HPV16 serological antibodies was low. However, in cases of oropharyngeal cancer, the concordance between HPV16 DNA and HPV16 E6, L1, and E6/E7 was moderate (0.578, 0.416, and 0.477, respectively) (Table 4). Seropositivity for antibodies against HPV16 E6 and E6/ E7 was significantly associated with a better HNSCC prognosis. Oropharyngeal cancer patients with HPV16 E6/ E7 seropositivity had a 78 % lower risk of death compared with those with negative serology (HR = 0.22; 95 % CI ). No association between HPV16 DNA in tumor tissue and HNSCC survival was observed (HR = 0.74; 95 % CI ). Greater survival rates were observed among patients with oropharyngeal tumors who were HPV16 DNA-positive, but this result was not precise (HR = 0.04; 95 % CI ) (Table 5). We observed a poorer prognosis among men, among patients with advanced clinical tumor stage (T3 and T4), and among those treated with surgery and radiotherapy. Considering only oropharyngeal cancer, older patients and smokers (former and current) had lower overall survival (data no showed). Based on the results in Table 5, Kaplan Meier survival curves were calculated for oral cavity, hypopharyngeal, and laryngeal cancers combined (Fig. 1a, b) and for oropharyngeal cancer cases separately (Fig. 1c, d), considering seropositivity for HPV16 E6 (Fig. 1a, c) and E6/E7 (Fig. 1b, d), respectively. The prognosis was clearly better for oropharyngeal cancer patients with a positive serological response to HPV16 E6/E7 (Log-rank test; p = 0.016) (Fig. 1d). Seropositivity for antibodies against HPV16 E6 and E6/ E7 was significantly associated with a better HNSCC prognosis. Oropharyngeal cancer patients with HPV16 E6/ E7 seropositivity had a 78 % lower risk of death compared with those with negative serology. Table 6 shows the characteristics of HNSCC patients by availability of biological samples for HPV DNA status. No differences were observed in patients according to gender, age, tobacco and alcohol consumption, anatomical site, and tumor stage. Tumor tissue samples were only available for surgery cases and association was observed for HPV16 DNA (p = 0.041).

8 468 Cancer Causes Control (2014) 25: Fig. 1 Kaplan Meier curves of overall survival according to serological response to HPV16 E6 and HPV16 E6/E7 in patients with oral cavity, hypopharynx, and larynx cancers (combined) and in patients with oropharynx cancer. A HPV16 E6 in patients with oral cavity, hypopharynx, and larynx cancers. B HPV16 E6/E7 in patients with oral cavity, hypopharynx, and larynx cancers. C HPV16 E6 in patients with oropharyngeal cancer. D HPV16 E6/E7 in patients with oropharyngeal cancer. Overall survival of 841 patients with oral cavity, hypopharynx, and larynx cancers (A and B). Overall survival in 252 patients with oropharynx cancer (C and D) Discussion This is the largest study conducted in Brazil to assess the prevalence of HPV in a sample of HNSCC patients and to determine the corresponding survival rates. Although the prevalence of HPV16 DNA in this cohort of HNSCC patients was only 3.5 % (14/398), the groups with the highest prevalence of positivity were females (8 %), those 50 years old or younger (4.9 %), nonsmokers (13.3 %), nondrinkers (10.5 %), and patients with oropharynx tumors (4.4 %). These prevalence levels of HPV16 DNA are lower than those observed in other regions of the world [24]. Regarding HPV antibodies in serum, we observed a high prevalence of HPV16 E7 (positive in about 24 % of all HNSCC cases), regardless of tumor site. Oropharyngeal tumors had a higher prevalence of HPV16 E6 (10.7 %) and HPV16 E6/E7 (5.6 %) than tumors in other anatomical sites of the head and neck. Notably, these levels of HPV16 E6 and E6/E7 antibodies are much lower than those previously reported in the United States [7, 14]. The presence of HPV16 DNA alone does not seem to be a good indicator of prognosis for HNSCC patients. We observed that HPV16 DNA had different prognostic value depending on the tumor site: poorer for oral cavity cancer and better for oropharynx and larynx cancers, although the estimates lacked precision. For oropharyngeal cancer, we observed a moderate concordance between HPV16 DNA and HPV16 E6 and E6/ E7 serology. Previous studies have detected significantly higher oropharyngeal cancer death rates for patients who are seronegative for HPV16 E6 or E6/E7 [14, 25 29]. In our study, HPV16 E6 and E6/E7 seropositivity were associated with a reduction in death from HNSCC. However, when we examined the Kaplan Meier survival curves for the combined tumors of the oral cavity, hypopharynx

9 Cancer Causes Control (2014) 25: Table 6 Characteristics of HNSCC patients by availability of HPV16 DNA, Brazil, Total n = 1,093 n (%) No samples n = 695 n (%) Samples available n = 398 p 1 HPV16 DNA Negative Positive n = 384 n = 14 n (%) n (%) Sex Female 148 (13.5) 98 (14.1) 46 (12.0) 4 (28.6) Male 945 (86.5) 597 (85.9) 338 (88.0) 10 (71.4) Age (years) \ (39.8) 273 (39.3) 155(40.4) 7 (50.0) (34.1) 232 (33.4) 136 (35.4) 5 (35.7) (20.6) 149 (21.4) 74 (19.3) 2 (14.3) C75 60 (5.5) 41 (5.9) 19 (4.9) 0 Tobacco smoking Nonsmoker 53 (4.8) 38 (5.5) 13 (3.4) 2 (14.3) Former smoker 273 (25.0) 166 (23.9) 103 (26.8) 4 (28.6) Current smoker 767 (70.2) 491 (70.6) 268 (69.8) 8 (57.1) Alcohol consumption Nondrinker 97 (8.9) 59 (8.5) 34 (8.9) 4 (28.6) Former drinker 390 (35.7) 250 (36.0) 136 (35.4) 4 (28.6) Drinker 606 (55.4) 386 (55.5) 214 (55.7) 6 (42.9) Site Oral cavity 321 (29.4) 200 (28.8) 117 (30.5) 4 (28.6) Oropharynx 252 (23.1) 161 (23.2) 87 (22.7) 4 (28.6) Hypopharynx 115 (10.5) 71 (10.2) 43 (11.2) 1 (7.1) Larynx 405 (37.1) 263 (37.8) 137 (35.7) 5 (35.7) Tumor stage T1/T2 366 (33.5) 228 (32.8) 133 (34.6) 5 (35.7) T3 325 (29.7) 219 (31.5) 102 (26.6) 4 (28.6) T4 402 (36.8) 248 (35.7) 149 (38.8) 5 (35.7) Treatment Surgery 347 (31.7) 204 (29.4) 136 (35.4) 7 (50.0) Radiotherapy 214 (19.6) 138 (19.9) 76 (19.8) 0 Surgery? radiotherapy 335 (30.7) 213 (30.6) 118 (30.7) 4 (28.6) Other 197 (18.0) 140 (20.1) 54 (14.1) 3 (21.4) 1 chi-square test for association between availability of tumor tissues (with samples or not) and characteristic and larynx and, separately, oropharynx tumors, the strong association of HPV16 E6/E7 with oropharyngeal cancer prognosis became clear. There are some limitations to our study. We were not able to obtain information on crucial covariates for 14.3 % (182/1,275) of the initial cases, and we were also unable to follow up with 10.2 % (111/1,093) of the patients included in the study. The patients who were lost might have affected the study s precision. Another limitation is that tissue samples, from which HPV16 DNA was extracted, were only available for a portion of the cases (36.4 %), which could compromise the evaluation of the association of HPV16 DNA and HNSCC survival. Tumor tissue availability depends of the tumor size and also if the patient was submitted to surgery or not. However, the subset for which no tissue samples were available was comparable to the entire sample with respect to demographics, lifestyle, and clinical characteristics (Table 6). We observed that patients who are positive for HPV16 E6/E7 antibodies represent a group with a higher percentage of oropharyngeal cancer, females and nonsmokers. This profile explains the better prognosis of this group. However, the survival advantage associated with HPV16 infection in patients with oropharyngeal squamous cell

10 470 Cancer Causes Control (2014) 25: carcinoma likely reflects the interplay between the virus and lymphoid tissue, accounting for the differences in the biology of the disease and the treatment response [30]. In conclusion, patients with oropharyngeal cancer who test positive for HPV16 E6/E7 antibodies survived longer than those who test negative, then HPV16 E6/E7 antibodies are a clinically relevant indicator of prognosis for oropharyngeal squamous cell carcinoma. This result, which was obtained in a geographical region with a low prevalence of HPV infection, is similar to results reported in regions with a high prevalence of HPV infection. Test for HPV 16 E6/E7 antibodies may highlight to clinicians those patients who could be treated for a shorter duration and lowering the risk of unnecessary toxic effects. Acknowledgments This work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) [10/ ; 09/ and 04/ ] and by the European Commission [IC18-CT ]. Conflict of interest References None declared. 1. Argiris A, Karamouzis MV, Raben D, Ferris RL (2008) Head and neck cancer. Lancet 371: Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM (2010) GLOBOCAN 2008 v2.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No 10 [Internet]. Lyon, France: International Agency for Research on Cancer; [cited 2012 Aug 30] Carvalho AL, Ikeda MK, Magrin J, Kowalski LP (2004) Trends of oral and oropharyngeal cancer survival over five decades in 3267 patients treated in a single institution. Oral Oncol 40: Szymanska K, Hung RJ, Wünsch Filho V, Eluf Neto J, Curado MP, Koifman S et al (2011) Alcohol and tobacco, and the risk of cancers of the upper aerodigestive tract in Latin America: a case control study. Cancer Causes Control 22: Human papillomaviruses (2007) Lyon, France: International Agency for Research on Cancer- IARC Monograph on the Evaluation of Carcinogenic Risks to Humans 6. Herrero R, Castellsagué X, Pawlita M, Lissowska J, Kee F, Balaram P et al (2003) Human papillomavirus and oral cancer: The International Agency for Research on Cancer Multicenter Study. J Natl Cancer Inst 95: Smith EM, Ritchie JM, Pawlita M, Rubenstein LM, Haugen TH, Turek LP et al (2006) Human papillomavirus seropositivity and risks of head and neck cancer. Int J Cancer 120: D Souza G, Kreimer AR, Viscidi R, Pawlita M, Fakhry C, Koch WM et al (2007) Case-control study of human papillomavirus and oropharyngeal cancer. N Engl J Med 356: Näsman A, Attner P, Hammarstedt L, Du J, Eriksson M, Giraud G et al (2009) Incidence of human papillomavirus (HPV) positive tonsillar carcinoma in Stockholm, Sweden: an epidemic of viralinduced carcinoma? Int J Cancer 125: Ribeiro KB, Levi JE, Pawlita M, Koifman S, Matos E, Eluf-Neto J et al (2011) Low human papillomavirus prevalence in head and neck cancer: results from two large case control studies in highincidence regions. Int J Epidemiol 40: Fakhry C, Westra WH, Li S, Cmelak A, Ridge JA, Pinto H et al (2008) Improved survival of patients with human papillomavirus positive head and neck squamous cell carcinoma in a prospective clinical trial. J Natl Cancer Inst 100: Smith EM, Rubenstein LM, Ritchie JM, Lee JH, Haugen TH, Hamsikova E et al (2008) Does treatment seropositivity to human papillomavirus have prognostic significance for head and neck cancers? Cancer Epidemiol Biomarkers Prev 17: Ang KK, Harris J, Wheeler R, Weber R, Rosenthal DI, Nguyen- Tân PF et al (2010) Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med 363: Liang C, Marsit CJ, McClean MD, Nelson HH, Christensen BC, Haddad RI et al (2012) Biomarkers of HPV in head and neck squamous cell carcinoma. Cancer Res 72: World Health Organization (1992) International statistical classification of diseases and related health problems. 10th rev. World Health Organization, Washington, DC 16. Hashibe M, Brennan P, Benhamou S, Castellsague X, Chen C, Curado MP et al (2007) Alcohol drinking in never users of tobacco, cigarette smoking in never drinkers, and the risk of head and neck cancer: pooled analysis in the International Head and Neck Cancer Epidemiology Consortium. J Natl Cancer Inst 99: Sobin LH, Wittekind C (2002) TNM classification of malignant tumours. International Union Against Cancer, 6th edn. Wiley- Liss, New York 18. Kleter B, Van Doorn LJ, Schrauwen L, Molijn A, Sastrowijoto S, Ter Schegget J et al (1999) Development and clinical evaluation of a highly sensitive PCR-reverse hybridization line probe assay for detection and identification of anogenital human papillomavirus. J Clin Microbiol 37: Coutlée F, Gravitt P, Kornegay J, Hankins C, Richardson H, Lapointe N et al (2002) Use of PGMY primers in L1 consensus PCR improves detection of human papillomavirus DNA in genital samples. J Clin Microbiol 40: Bernard HU, Chan SY, Manos MM, Ong CK, Villa LL, Delius H et al (1994) Identification and assessment of known and novel human papillomaviruses by polymerase chain reaction amplification, restriction fragment length polymorphisms, nucleotide sequence, and phylogenetic algorithms. J Infec Dis 170: Sehr P, Zumbach K, Pawlita M (2001) A generic capture ELISA for recombinant proteins fused to glutathione-s-transferase: validation for HPV serology. J Immunol Methods 253: Waterboer T, Sehr P, Michael KM, Franceschi S, Nieland JD, Joos TO et al (2005) Multiplex human papillomavirus serology based on in situ-purified glutathione s-transferase fusion proteins. Clin Chem 51: Clifford GM, Shin HR, Oh JK, Waterboer T, Ju YH, Vaccarella S et al (2007) Serologic response to oncogenic human papillomavirus types in male and female university students in Busan, South Korea. Cancer Epidemiol Biomarkers Prev 16: Kreimer AR, Clifford GM, Boyle P, Franceschi S (2005) Human papillomavirus types in head and neck squamous cell carcinomas worldwide: a systematic review. Cancer Epidemiol Biomarkers Prev 14: Dayyani F, Etzel CJ, Liu M, Ho CH, Lippman SM, Tsao AS (2010) Meta-analysis of the impact of human papillomavirus (HPV) on risk and overall survival in head and neck squamous cell carcinomas (HNSCC). Head Neck Oncol 2:15. doi: / Smith EM, Pawlita M, Rubenstein LM, Haugen TH, Hamsikova E, Turek LP (2010) Risk factors and survival by HPV-16 E6 and E7 antibody status in human papillomavirus positive head and neck cancer. Int J Cancer 127: Rotnáglová E, Tachezy R, Saláková M, Procházka B, Košl abová E, Veselá E et al (2011) HPV involvement in tonsillar cancer:

11 Cancer Causes Control (2014) 25: prognostic significance and clinically relevant markers. Int J Cancer 129: Rubenstein LM, Smith EM, Pawlita M, Haugen TH, Hamsikova E, Turek LP (2011) Human papilomavírus serologic follow-up response and relationship to survival in head and neck cancer: a case-comparison study. Infec Agent Cancer 6:9. doi: / Smith EM, Rubenstein LM, Haugen TH, Pawlita M, Turek LP (2012) Complex etiology underlies risk and survival in head and neck cancer human papillomavirus, tobacco, and alcohol: a case for multifactor disease. J Oncol. doi: /2012/ Méndez E (2012) Biomarkers of HPV infection in oropharyngeal carcinomas: can we find simplicity in the puzzle of complexity? Cancer Res 72: