Advertisement

Impact of HPV-associated p16-expression on radiotherapy outcome in advanced oropharynx and non-oropharynx cancer

Open AccessPublished:November 26, 2014DOI:https://doi.org/10.1016/j.radonc.2014.11.032

      Abstract

      Background and purpose

      HPV is found in head and neck cancer from all sites with a higher prevalence in oropharynx cancer (OPC) compared to non-OPC. HPV/p16-status has a significant impact on radiotherapy (RT) outcome in advanced OPC, but less is known about the influence in non-OPC. We analyzed HPV-associated p16-expression in a cohort of patients with stage III–IV pharynx and larynx cancer treated with primary, curatively intended (chemo-)RT, aiming to test the hypothesis that the impact of HPV/p16 also extends to tumors of non-oropharyngeal origin.

      Material and methods

      1294 patients enrolled in previously conducted DAHANCA-trials between 1992 and 2012 were identified. Tumors were evaluated by p16-immunohistochemistry and classified as positive in case of staining in >70% of tumors cells.

      Results

      Thirty-eight percent (490/1294) of the tumors were p16-positive with a significantly higher frequency in OPC (425/815) than in non-OPC (65/479), p < .0001. In OPC p16-positivity significantly improved loco-regional control (LRC) (adjusted HR [95% CI]: 0.43 [0.32–0.57]), event-free survival (EFS) (HR 0.44 [0.35–0.56]), and overall survival (OS) (HR: 0.38 [0.29–0.49]), respectively, compared with p16-negativity. In non-OPC no prognostic impact of p16-status was found for either endpoint: LRC (HR: 1.13 [0.75–1.70]), EFS (HR: 1.06 [0.76–1.47]), and OS (HR: 0.82 [0.59–1.16]).

      Conclusions

      The independent influence of HPV-associated p16-expression in advanced OPC treated with primary RT was confirmed. However, RT-outcome in the group of non-OPC did not differ by tumor p16-status, indicating that the prognostic impact may be restricted to OPC only.

      Keywords

      Squamous cell carcinomas of the head and neck (HNSCCs) have traditionally been grouped as a cancer entity, mainly based on the anatomy, common etiology and sensitivity to treatment. HNSCCs arise from the epithelium of the upper aerodigestive tract and as such represent a rather heterogeneous group of neoplasias including carcinomas of the oral cavity, pharynx (oropharynx, nasopharynx and hypopharynx) and larynx. With the emergence of human papillomavirus (HPV) as contributing etiological factor in carcinogenesis of a subgroup of HNSCC [
      • Gillison M.L.
      • Koch W.M.
      • Capone R.B.
      • et al.
      Evidence for a causal association between human papillomavirus and a subset of head and neck cancers.
      ], the apparent epidemic incidence increase of HPV-associated HNSCC [
      • Lassen P.
      The role of human papillomavirus in head and neck cancer and the impact on radiotherapy outcome.
      ,
      • Hammarstedt L.
      • Dahlstrand H.
      • Lindquist D.
      • et al.
      The incidence of tonsillar cancer in Sweden is increasing.
      ,
      • Chaturvedi A.K.
      • Engels E.A.
      • Pfeiffer R.M.
      • et al.
      Human papillomavirus and rising oropharyngeal cancer incidence in the United States.
      ] and the potent impact on treatment sensitivity and prognosis for patients with HPV-associated disease, the traditional understanding and therapeutic management of HNSCC has been more or less revolutionized. The strongest association with HPV is found in oropharynx cancer (OPC) where tumours of the tonsils are particularly associated with HPV infection [
      • Fakhry C.
      • Gillison M.L.
      Clinical implications of human papillomavirus in head and neck cancers.
      ,
      • Klussmann J.P.
      • Weissenborn S.J.
      • Wieland U.
      • et al.
      Human papillomavirus-positive tonsillar carcinomas: a different tumor entity?.
      ] but high-risk HPV, predominantly HPV-16, has been found in HNSCC from all sites although with a significantly higher prevalence in OPC compared to tumours arising outside the oropharyngeal region (non-OPC) [
      • Kreimer A.R.
      • Clifford G.M.
      • Boyle P.
      • Franceschi S.
      Human papillomavirus types in head and neck squamous cell carcinomas worldwide: a systematic review.
      ,
      • Isayeva T.
      • Li Y.
      • Maswahu D.
      • Brandwein-Gensler M.
      Human papillomavirus in non-oropharyngeal head and neck cancers: a systematic literature review.
      ].
      The Danish Head and Neck Cancer Group (DAHANCA) has a longstanding tradition for conducting nationwide randomized trials [
      • Overgaard J.
      • Hansen H.S.
      • Overgaard M.
      • et al.
      A randomized double-blind phase III study of nimorazole as a hypoxic radiosensitizer of primary radiotherapy in supraglottic larynx and pharynx carcinoma. Results of the Danish Head and Neck Cancer Study (DAHANCA) Protocol 5–85.
      ,
      • Overgaard J.
      • Hansen H.S.
      • Specht L.
      • et al.
      Five compared with six fractions per week of conventional radiotherapy of squamous-cell carcinoma of head and neck: DAHANCA 6 and 7 randomised controlled trial.
      ] encompassing all eligible Danish HNSCC patients aiming to improve RT-outcome for these patients in general. The use of HPV-associated p16-expression as stratification parameter within these trials has demonstrated that tumor p16-status is presently the strongest known independent factor in primary RT of head and neck cancer [
      • Lassen P.
      • Eriksen J.G.
      • Hamilton-Dutoit S.
      • Tramm T.
      • Alsner J.
      • Overgaard J.
      Effect of HPV-associated p16INK4A expression on response to radiotherapy and survival in squamous cell carcinoma of the head and neck.
      ,
      • Lassen P.
      • Eriksen J.G.
      • Hamilton-Dutoit S.
      • Tramm T.
      • Alsner J.
      • Overgaard J.
      HPV-associated p16-expression and response to hypoxic modification of radiotherapy in head and neck cancer.
      ,
      • Lassen P.
      • Eriksen J.G.
      • Krogdahl A.
      • et al.
      The influence of HPV-associated p16-expression on accelerated fractionated radiotherapy in head and neck cancer: evaluation of the randomised DAHANCA 6&7 trial.
      ,
      • Lassen P.
      • Overgaard J.
      • Eriksen J.G.
      Expression of EGFR and HPV-associated p16 in oropharyngeal carcinoma: correlation and influence on prognosis after radiotherapy in the randomized DAHANCA 5 and 7 trials.
      ,
      • Overgaard J.
      • Hoff C.M.
      • Hansen H.S.
      • Specht L.
      • Overgaard M.
      • Grau C.
      • et al.
      Randomized study of darbepoetin alfa as modifier of radiotherapy in patients with primary squamous cell carcinoma of the head and neck (HNSCC): Final outcome of the DAHANCA 10 trial.
      ,

      Bentzen J., Toustrup K., Eriksen J.G., Primdahl H., Andersen L.J., Overgaard J., Locally advanced Head and Neck cancer treated with accelerated radiotherapy, the hypoxic modifier nimorazole and weekly cisplatin. Results from the DAHANCA 18 phase II study. Acta Oncol [in press].

      ,
      • Eriksen J.G.
      • Maare C.
      • Johansen J.
      • Primdahl H.
      • Evensen J.F.
      • Kristensen C.A.
      • et al.
      DAHANCA19: a randomized phase III study of primary (chemo-) radiotherapy and zalutumumab in head and neck carcinomas.
      ]. Numerous other studies have focused on sub-group analysis of patients with OPC within previously conducted randomized trials, concordantly demonstrating a highly significant impact of tumor HPV/p16-status on treatment outcome [
      • Ang K.K.
      • Harris J.
      • Wheeler R.
      • et al.
      Human papillomavirus and survival of patients with oropharyngeal cancer.
      ,
      • Rischin D.
      • Young R.J.
      • Fisher R.
      • et al.
      Prognostic significance of p16INK4A and human papillomavirus in patients with oropharyngeal cancer treated on TROG 02.02 phase III trial.
      ]. These observations are believed to be caused in part by a higher sensitivity of HPV/p16-positive tumors to RT [
      • Sorensen B.S.
      • Busk M.
      • Olthof N.
      • et al.
      Radiosensitivity and effect of hypoxia in HPV positive head and neck cancer cells.
      ,
      • Rieckmann T.
      • Tribius S.
      • Grob T.J.
      • et al.
      HNSCC cell lines positive for HPV and p16 possess higher cellular radiosensitivity due to an impaired DSB repair capacity.
      ] combined with a different and more favorable risk factor profile [
      • Gillison M.L.
      • D’Souza G.
      • Westra W.
      • et al.
      Distinct risk factor profiles for human papillomavirus type 16-positive and human papillomavirus type 16-negative head and neck cancers.
      ] and better general health status in the group of patients with HPV/p16-positive disease. Much less is known about the impact of HPV/p16-status in non-OPC although a recent paper indicated, that tumor p16-status also has prognostic value in tumors of non-oropharyngeal origin [
      • Chung C.H.
      • Zhang Q.
      • Kong C.S.
      • et al.
      P16 protein expression and human papillomavirus status as prognostic biomarkers of non oropharyngeal head and neck squamous cell carcinoma.
      ].
      Clinical trials are presently investigating whether de-intensified treatment strategies could result in avoidance of excessive toxicity without compromising outcome for selected patients with HPV/p16-positive OPC. On the other hand, the European Organisation for Research and Treatment of Cancer (EORTC) in collaboration with the DAHANCA has launched a trial (EORTC1219 ROG-HNCG/DAHANCA29) investigating whether additional intensification of treatment could be beneficial for patients with HPV/p16-negative HNSCC based on the observed poor outcome for this patient group []. Thus presently there is substantial variation in the treatment strategies considered for patients with head and neck cancer dependent of the HPV/p16-status of the tumors. Consequently, provision of further knowledge regarding outcome for patients with p16-positive non-OPC is needed in order to secure optimal and safe treatment for these patients also. Accordingly, in the present study we analyzed the influence of HPV-associated p16-expression in a large cohort of patients with advanced pharynx and larynx carcinoma treated with primary, curatively intended RT, aiming to test the hypothesis that the prognostic impact of HPV/p16 also extends to tumors of non-oropharyngeal origin.

      Materials and methods

      Study population

      The study cohort was identified in the nationwide DAHANCA database which contains information on all Danish head and neck cancer patients since 1970. Since a very high proportion of patients with HPV-associated OPC present with advanced disease stage due to regional lymphnode involvement at time of diagnosis, we selected patients with advanced (stage III–IV) pharynx (oropharynx and hypopharynx) and larynx carcinoma to avoid skewness caused by stage differences between the two groups (OPC vs. non-OPC). All patients were treated with curatively intended primary (chemo)radiotherapy within DAHANCA trials from 1992 to 2012 and the impact of tumor p16-status within these individual trials have been reported previously [
      • Lassen P.
      • Eriksen J.G.
      • Hamilton-Dutoit S.
      • Tramm T.
      • Alsner J.
      • Overgaard J.
      Effect of HPV-associated p16INK4A expression on response to radiotherapy and survival in squamous cell carcinoma of the head and neck.
      ,
      • Lassen P.
      • Eriksen J.G.
      • Hamilton-Dutoit S.
      • Tramm T.
      • Alsner J.
      • Overgaard J.
      HPV-associated p16-expression and response to hypoxic modification of radiotherapy in head and neck cancer.
      ,
      • Lassen P.
      • Eriksen J.G.
      • Krogdahl A.
      • et al.
      The influence of HPV-associated p16-expression on accelerated fractionated radiotherapy in head and neck cancer: evaluation of the randomised DAHANCA 6&7 trial.
      ,
      • Lassen P.
      • Overgaard J.
      • Eriksen J.G.
      Expression of EGFR and HPV-associated p16 in oropharyngeal carcinoma: correlation and influence on prognosis after radiotherapy in the randomized DAHANCA 5 and 7 trials.
      ,
      • Overgaard J.
      • Hoff C.M.
      • Hansen H.S.
      • Specht L.
      • Overgaard M.
      • Grau C.
      • et al.
      Randomized study of darbepoetin alfa as modifier of radiotherapy in patients with primary squamous cell carcinoma of the head and neck (HNSCC): Final outcome of the DAHANCA 10 trial.
      ,

      Bentzen J., Toustrup K., Eriksen J.G., Primdahl H., Andersen L.J., Overgaard J., Locally advanced Head and Neck cancer treated with accelerated radiotherapy, the hypoxic modifier nimorazole and weekly cisplatin. Results from the DAHANCA 18 phase II study. Acta Oncol [in press].

      ,
      • Eriksen J.G.
      • Maare C.
      • Johansen J.
      • Primdahl H.
      • Evensen J.F.
      • Kristensen C.A.
      • et al.
      DAHANCA19: a randomized phase III study of primary (chemo-) radiotherapy and zalutumumab in head and neck carcinomas.
      ]. Due to the specific epidemiology (Epstein Barr virus), histology, growth-pattern and prognosis, nasopharyngeal carcinomas were not included. Moreover, patients with carcinoma of the oral cavity were deliberately excluded since standard treatment in Denmark of these tumors often consists of primary surgery followed by postoperative radiotherapy if indicated. The DAHANCA trials were performed in accordance with the Helsinki Declaration II, and the patients gave written informed consent. Both the main studies and the tumor tissue analyses performed in the present study were approved by the relevant regional and national ethics committees according to Danish legislation and regulations.

      Evaluation of p16 immunohistochemistry

      Routine paraffin-embedded, formalin-fixed pretreatment tumor tissues were available from 1294 patients and immunohistochemistry (IHC) for p16-expression was performed on whole tumor sections as previously described [
      • Lassen P.
      • Eriksen J.G.
      • Hamilton-Dutoit S.
      • Tramm T.
      • Alsner J.
      • Overgaard J.
      Effect of HPV-associated p16INK4A expression on response to radiotherapy and survival in squamous cell carcinoma of the head and neck.
      ]. Briefly, paraffin sections were cut at 5 μm on Superfrost® plus charged glass slides (Menzel–Glaser), heated at 60 °C for 1 h and deparaffinized in the instrument with EZ prep solution (Ventana Medical Systems). Heat induced antigen retrieval was carried out using Cell Conditioning 1 solution (CC1, Ventana Medical Systems). p16 was detected by incubating sections with antibody clone JC8 (sc-56330; Santa Cruz Biotechnology Inc., Santa Cruz, CA) diluted 1:25 for 32 min. JC8 is a mouse monoclonal immunoglobulin G2a antibody raised against full-length human p16, particularly suitable for use on formalin-fixed, paraffin-embedded sections. The specificity was confirmed by Western blotting [
      • Lassen P.
      • Eriksen J.G.
      • Hamilton-Dutoit S.
      • Tramm T.
      • Alsner J.
      • Overgaard J.
      Effect of HPV-associated p16INK4A expression on response to radiotherapy and survival in squamous cell carcinoma of the head and neck.
      ]. Specific reactions were detected using ultraView Universal DAB Detection Kit (Ventana Medical Systems), and the slides were counterstained with hematoxylin. Sections of p16-positive cervical carcinoma were used as positive controls. The scoring and classification of the tumors were based in part on a 70% cutpoint but also in consideration of the typical microscopic appearance of an HPV-related tumor in order to optimise the correlation with HPV as described and recommended by El-Naggar et al. for oropharyngeal carcinomas [
      • El-Naggar A.K.
      • Westra W.H.
      P16 expression as a surrogate marker for HPV-related oropharyngeal carcinoma: a guide for interpretative relevance and consistency.
      ]. Up until 2006 the scoring was done retrospectively and a tumor was classified as p16-positive in case of strong, diffuse nuclear and cytoplasmatic staining in more than 70% of carcinoma cells, but tumors displaying a conventional keratinizing morphology together with a variable or patchy p16-expression staining pattern were classified as p16-negative [
      • Lassen P.
      • Overgaard J.
      Scoring and classification of oropharyngeal carcinoma based on HPV-related p16-expression.
      ]. Since then p16-IHC has been implemented in the routine work-up of patients with HNSCC in Denmark and as such the staining and interpretation of p16-IHC has been performed prospectively by pathologists at the respective oncological centers.

      Statistical analysis

      The primary endpoint was loco-regional control (LRC) 5 years after completion of RT and secondary endpoints were event-free survival (EFS) [
      • Michiels S.
      • Le M.A.
      • Buyse M.
      • et al.
      Surrogate endpoints for overall survival in locally advanced head and neck cancer: meta-analyses of individual patient data.
      ] and overall survival (OS). LRC was defined as complete and persistent absence of the disease in T-site and regional lymph nodes (N-site) after radiotherapy. Failure was recorded in the event of recurrent tumor, or if the primary tumor never completely disappeared. In the latter situation the tumor was then assumed to have failed at the time of randomization. Consequently the endpoint does not include the effect of a successful procedure with salvage surgery. EFS was defined as loco-regional recurrence or progression, distant metastasis or death resulting from any cause and OS was defined as death resulting from any cause. Regarding LRC patients were followed for at least 5 years and the survival status of the patients was tracked until the end of 2012.
      The χ2-test for categorical variables was used to compare characteristics of p16-positive and negative tumors and OPC vs. non-OPC, respectively. The actuarial values of the endpoints were estimated by the Kaplan–Meier method and compared with a log-rank test. Multivariable hazard ratios (HRs) were estimated by use of the Cox proportional-hazard method to evaluate prognostic parameters and treatment with respect to the risk of loco-regional failure, event-free death and overall death. Additional multivariable analyses were done including an assessment of the interaction between p16-status and tumor site (OPC vs. non-OPC). All results were considered significant at levels less than 5% (two-sided tests) and HRs presented with 95% confidence intervals.

      Results

      Table 1 shows patient and tumor characteristics according to tumor site and p16-status. Sixty-three percent (815) of the tumors were OPC, and 37% (479) were of non-OPC origin (larynx: 25% (321) and hypopharynx: 12% (158), respectively). Overall, 38% (490/1294) of the tumors were p16-positive and the frequency of p16-positivity was significantly higher in OPC (425/815, 52%) than in non-OPC (65/479, 14%), p < .0001. The site-specific distribution of p16-positivity in non-OPC was as follows: larynx 44/321 (14%) and hypopharynx 21/158 (13%).
      Table 1Patient and tumor characteristics by p16-status and tumor-site.
      All patients (n = 1294)Oropharynx (n = 815)Non-oropharynx (n = 479)
      p16-positive

      (n = 425)
      p16-negative

      (n = 390)
      p
      Chi-square test for comparison between p16-positive and p16-negative tumors.
      p16-positive

      (n = 65)
      p16-negative

      (n = 414)
      p
      Chi-square test for comparison between p16-positive and p16-negative tumors.
      No. (%)No. (%)No. (%)No. (%)No. (%)
      Tumor site
       Oropharynx815 (63)425 (100)390 (100)<.0001
      Oropharynx vs. non-oropharynx.
       Larynx321 (25)44 (68)277 (67)<.0001
      Oropharynx vs. non-oropharynx.
       Hypopharynx158 (12)21 (32)137 (33)
      Age (years)
       Median (range)59 (28–84)57 (34–83)58 (28–84)n.s59 (34–82)60 (35–83)n.s
      Gender
       Female270 (21)88 (21)91 (23)n.s19 (29)72 (17).02
       Male1024 (79)337 (79)299 (77)46 (71)342 (83)
      Tumor stage
      International Union of Cancer Research (UICC) 1982 classification.
       T1–2619 (48)279 (66)166 (43)<.000127 (42)147 (36)n.s
       T3–4675 (52)146 (34)224 (57)38 (58)267 (64)
      Nodal status
      International Union of Cancer Research (UICC) 1982 classification.
       N0235 (18)19 (4)60 (15)<.000115 (23)141 (34)n.s
       N+1059 (82)406 (96)330 (85)50 (77)273 (66)
      Disease stage
      International Union of Cancer Research (UICC) 1982 classification.
       III434 (34)101 (24)132 (34)<.000129 (45)172 (42)n.s
       IV860 (66)324 (76)258 (66)36 (55)242 (58)
      Treatment
       5fx/week380 (29)71 (17)118 (30)<.000116 (24)175 (42).007
       6fx/week914 (71)354 (83)272 (70)49 (76)239 (58)
       No Nimorazole140 (11)20 (5)45 (12)<.00014 (6)71 (17).00
       Nimorazole1154 (89)405 (95345 (88)
      61 (94)343 (83)
       No chemotherapy931 (72)190 (45)334 (86)<.000156 (86)351 (85)n.s
       Chemotherapy363 (28)235 (55)56 (14)9 (14)63 (15)
      Abbreviation: n.s, not significant.
      low asterisk International Union of Cancer Research (UICC) 1982 classification.
      a Chi-square test for comparison between p16-positive and p16-negative tumors.
      b Oropharynx vs. non-oropharynx.
      Median age did not differ between groups but in non-OPC there were significantly more women in the p16-positive group compared to p16-negative non-OPC whereas gender was equally distributed in OPC. p16-positive OPC were characterized by small T-size (T1–2) and lymphnode involvement (N+) as opposed to patients with p16-negative OPC whereas such differences were not present in non-OPC. The proportion of OPC increased from 52% in the beginning of the recruitment period whereas late in the period these tumors accounted for approximately 80% of the patients. The corresponding frequency of p16-positivity increased from 22% to 70% in the same period. These findings reflect the demographic shift observed in HNSCC in some parts of the western world, including Denmark [
      • Lassen P.
      The role of human papillomavirus in head and neck cancer and the impact on radiotherapy outcome.
      ] and which in all probability can be explained by an epidemic increase in the incidence of HPV-associated p16-positive OPC during the recruitment period. Based on a combination of this demographic shift and the successive implementation of new treatment strategies for HNSCC throughout the recruitment time, significant differences in terms of treatment components were observed in the study cohort, Table 1. Three-hundred and eighty patients were treated with conventionally fractionated radiotherapy (5fx/week) with (N = 240) or without (N = 140) Nimorazole, reflecting the use of this treatment strategy in the beginning of the period. The majority (N = 914) of the patients were treated with moderately accelerated radiotherapy (68 GY/34fx, 6fx/week) concomitant with the hypoxic cell radiosensitizer Nimorazole and a minor proportion of the patients (N = 363) received chemotherapy (weekly Cisplatinum 40 mg/m2) in addition to moderate acceleration and Nimorazole. This resulted in significant treatment differences between p16-positive and p16-negative tumors of both oropharyngeal and non-oropharyngeal origin, as can be seen in Table 1.
      At the time of evaluation, 460 patients had failed to achieve persistent loco-regional control within the irradiated volume. Distant metastasis occurred in 128 patients, 37 in the p16-positive group of which 24 were isolated distant failures and 91 in the p16-negative group of which 39 were distant metastasis alone. A total of 426 patients had died with or from the cancer in question, and overall, 666 patients had died.
      In the total cohort of 1294 patients, the 5-year actuarial probability of LRC, EFS and OS was significantly better for patients with p16-positive tumors compared with p16-negativity (LRC: 76% vs. 50%, adjusted HR [95%CI]: 0.51 [0.41–0.65], EFS: 61% vs. 39%, adjusted HR: 0.50 [0.42–0.60] and OS: 74% vs. 36%, adjusted HR: 0.42 [0.34–0.51], respectively). Similarly, in the group of OPC, p16-positivity was correlated with a significant improvement in LRC (80% vs. 52%, HR: 0.32 [0.25–0.43], Fig. 1A) EFS (67% vs. 31%, HR 0.34 [0.27–0.42], Fig. 1C), and OS (80% vs. 37%, HR: 0.26 [0.20–0.33], Fig. 1E), respectively, compared with p16-negativity. However, no prognostic impact of p16-status was found in tumors of non-oropharyngeal origin for either endpoint: LRC (53% vs. 48%, HR: 0.94 [0.63–1.40], Fig. 1B), EFS (28% vs. 28%, HR: 0.94 [0.68–1.30], Fig. 1D), and OS (38% vs. 35%, HR: 0.81 [0.56–1.13], Fig. 1F), respectively. To further investigate whether the impact of p16 would differ by non-OPC sub-site, analysis was also done for larynx and hypopharynx individually. This revealed that p16-status did not impact on outcome in neither tumors of the larynx (LRC adjusted HR: 1.09 [0.66–1.79], EFS: 0.91 [0.60–1.37] and OS: 0.77 [0.51–1.18]) nor hypopharynx (LRC adjusted HR: 0.86 [0.42–1.78], EFS: 1.09 [0.63–1.89] and OS: 0.74 [0.41–1.31]), respectively.
      Figure thumbnail gr1
      Fig. 1(A) Actuarial estimated loco-regional tumor control (LRC), (C) event-free survival (EFS) and (E) overall survival (OS) by p16-status in stage III–IV oropharyngeal carcinomas. (B), (D) and (F) illustrate the actuarial estimated LRC, EFS and OS in the group of advanced non-oropharyngeal carcinoma, respectively.
      In the final Cox proportional hazards analysis adjusting for age (<60 years vs. >60 years), gender (female vs. male), T-size (T1–2 vs. T3–4), lymphnode involvement (N0 vs. N+), hypoxic modification (Nimorazole vs. no Nimorazole), fractionation schedule (6fx/week vs. 5fx/week) and chemotherapy (chemotherapy vs. no chemotherapy), p16-status remained a statistical significant independent prognostic factor in the group of patients with OPC (locoregional failure: HR: 0.43 [0.32–0.57], event-free death: HR: 0.44 [0.35–0.56] and overall death: HR: 0.38 [0.29–0.49]), whereas no prognostic impact of p16-status could be demonstrated in the group of patients with non-OPC: (locoregional failure: HR: 1.13 [0.75–1.70], event-free death: HR: 1.06 [0.76–1.47] and overall death: HR: 0.82 [0.59–1.16]), Table 2. We did an additional analysis involving all 1294 patients including an assessment of the interaction between p16-status and tumor site (OPC vs. non-OPC) for all three endpoints in question. Table 2 shows that p16-positivity interacted significantly (p < 0.0001) with oropharyngeal tumor origin only. Thus, outcome was significantly better for all endpoints in patients with p16-positive OPC compared with any other patient group. Moreover, outcomes did not differ significantly between the remaining three groups: p16-neg OPC, p16-pos non-OPC and p16-neg non-OPC.
      Table 2Multivariable analysis for all patients and stratified by tumor-site.
      Endpoint/variableOropharynxNon-oropharynxAll patients
      Analysis included an assessment of the interaction between p16-status and tumor-site.
      HR [95% CI]
      Estimates <1 is in favor of the first-mentioned stratum.
      HR [95% CI]
      Estimates <1 is in favor of the first-mentioned stratum.
      HR [95% CI]
      Estimates <1 is in favor of the first-mentioned stratum.
      Loco-regional failure
       Age <60 years vs. >60 years0.89 [0.68–1.15]1.31 [1.00–1.71]1.06 [0.88–1.28]
       Female vs. male0.72 [0.52–1.01]0.51 [0.35–0.76]0.64 [0.50–0.83]
       T1–2 vs. T3–40.52 [0.40–0.69]0.64 [0.47–0.88]0.57 [0.46–0.70]
       N0 vs. N+0.41 [0.27–0.65]0.46 [0.33–0.64]0.44 [0.34–0.57]
       p16pos vs. p16neg0.43 [0.32–0.57]1.13 [0.75–1.70]
       Nim vs. no nim0.78 [0.52–1.15]0.72 [0.50–1.03]0.74 [0.57–0.97]
       6fx/week vs. 5fx/week0.42 [0.31–0.58]0.59 [0.43–0.82]0.50 [0.40–0.63]
       Chemo vs. no chemo0.58 [0.39–0.87]0.32 [0.17–0.58]0.47 [0.34–0.65]
      p16pos OPC0.41 [0.31–0.55]
      p16pos non-OPC1.12 [0.75–1.68]
      p16neg OPC0.95 [0.77–1.18]
      p16neg non-OPC1 (Ref)
      Event-free death
       Age <60 years vs. >60 years0.78 [0.63–0.96]0.99 [0.80–1.22]0.86 [0.74–0.99]
       Female vs. male0.84 [0.65–1.08]0.6 0 [0.45–0.81]0.76 [0.63–0.91]
       T1–2 vs. T3–40.47 [0.38–0.59]0.60 [0.46–0.77]0.52 [0.44–0.62]
       N0 vs. N+0.52 [0.38–0.72]0.47 [0.36–0.61]0.48 [0.39–0.59]
       p16pos vs. p16neg0.44 [0.35–0.56]1.06 [0.76–1.47]
       Nim vs. no nim0.69 [0.50–0.97]0.95 [0.69–1.29]0.82 [0.66–1.03]
       6Fx/week vs. 5Fx/week0.57 [0.44–0.74]0.65 [0.50–0.84]0.60 [0.51–0.73]
       Chemo vs. no chemo0.57 [0.42–0.77]0.49 [0.32–0.72]0.53 [0.42–0.68]
      p16pos OPC0.41 [0.33–0.51]
      p16pos non-OPC1.06 [0.77–1.47]
      p16neg OPC0.97 [0.82–1.15]
      p16neg non-OPC1 (Ref)
      Overall death
       Age <60 years vs. >60 years0.72 [0.58–0.90]0.85 [0.68–1.06]0.77 [0.66–0.90]
       Female vs. male0.84 [0.64–1.09]0.65 [0.49–0.88]0.79 [0.65–0.95]
       T1–2 vs. T3–40.43 [0.34–0.54]0.62 [0.48–0.81]0.50 [0.42–0.60]
       N0 vs. N+0.48 [0.35–0.67]0.44 [0.33–0.58]0.43 [0.35–0.53]
       p16pos vs. p16neg0.38 [0.29–0.49]0.82 [0.59–1.16]
       Nim vs. no nim0.78 [0.56–1.09]1.25 [0.90–1.73]1.03 [0.81–1.29]
       6Fx/week vs. 5Fx/week0.60 [0.46–0.78]0.69 [0.53–0.90]0.64 [0.53–0.77]
       Chemo vs. no chemo0.36 [0.24–0.53]0.52 [0.34–0.79]0.42 [0.31–0.55]
      p16pos OPC0.35 [0.27–0.44]
      p16pos non-OPC0.84 [0.60–1.18]
      p16neg OPC1.00 [0.85–1.21]
      p16neg non-OPC1 (Ref)
      Abbreviation: HR, hazard ratio.
      a Analysis included an assessment of the interaction between p16-status and tumor-site.
      b Estimates <1 is in favor of the first-mentioned stratum.

      Discussion

      Knowing that HPV-associated p16-positve HNSCC is predominantly characterized by oropharyngeal origin and advanced disease stage, we addressed the question of the potential prognostic impact of p16-status in non-OPC in a selected cohort of patients with stage III–IV disease. Moreover, we only included patients who were treated with primary curatively intended (chemo)radiotherapy, deliberately excluding patients with oral cavity cancer. Within this cohort we confirmed the well-known highly significant independent influence of tumor p16-expression on loco-regional control and survival for patients with advanced OPC. In contrast we could not demonstrate a similar prognostic impact in advanced tumors of the larynx and hypopharynx where radiotherapy-outcome did not differ by tumor p16-status. As opposed to our findings, Chung et al. recently found significant outcome differences within a cohort of non-OPC where the p16-positive group was found to have significant better outcome than the p16-negative non-OPC. [
      • Chung C.H.
      • Zhang Q.
      • Kong C.S.
      • et al.
      P16 protein expression and human papillomavirus status as prognostic biomarkers of non oropharyngeal head and neck squamous cell carcinoma.
      ]. A possible explanation for the diverging results between the two studies might be found in the rather distinct differences between the compositions of the study-cohorts examined. In the RTOG trial one third of the p16-positive tumors were carcinomas of the oral cavity treated with primary surgery followed by postoperative chemoradiation and Cetuximab in the phase II RTOG 0234-trial [
      • Harari P.M.
      • Harris J.
      • Kies M.S.
      • et al.
      Postoperative chemoradiotherapy and cetuximab for high-risk squamous cell carcinoma of the head and neck: Radiation Therapy Oncology Group RTOG-0234.
      ], whereas our analyses only comprised larynx and pharynx cancer treated with primary (chemo)radiotherapy.
      Tobacco-smoking negatively affects outcome for head and neck cancer patients treated with radiotherapy [
      • Hoff C.M.
      • Grau C.
      • Overgaard J.
      Effect of smoking on oxygen delivery and outcome in patients treated with radiotherapy for head and neck squamous cell carcinoma – a prospective study.
      ]. Moreover, it has been shown that also for p16-positive OPC an accumulated smoking history of more or less than 10 packyears independently influences outcome [
      • Ang K.K.
      • Harris J.
      • Wheeler R.
      • et al.
      Human papillomavirus and survival of patients with oropharyngeal cancer.
      ] indicating the existence of a separate group of tumors with mixed aetiology (tobacco and HPV) of importance for clinical outcome. We did not investigate the impact of smoking in the present study. However based on a previous study involving DAHANCA patients [
      • Hoff C.M.
      • Grau C.
      • Overgaard J.
      Effect of smoking on oxygen delivery and outcome in patients treated with radiotherapy for head and neck squamous cell carcinoma – a prospective study.
      ] we know, that the vast majority of patients with non-OPC were smokers and as such the possibility that differences in smoking habits within the group of non-OPC affect the results seems modest. In the group of OPC, on the other hand, it cannot be ruled out, that there are some “never-smokers” (<10 packyears) among p16-positive patients that further lift their outcome and render the difference in prognosis between p16-positive and p16-negative OPC patients even more pronounced.
      The reasons for the apparent site-specific difference in the prognostic impact of p16-expression observed in the present study remain unsolved. Observations from preclinical studies indicate that the enhanced in vitro radiosensitivity of HPV/p16-positive cell lines derived from OPC is linked to the individual tumor cell [
      • Sorensen B.S.
      • Busk M.
      • Olthof N.
      • et al.
      Radiosensitivity and effect of hypoxia in HPV positive head and neck cancer cells.
      ,
      • Rieckmann T.
      • Tribius S.
      • Grob T.J.
      • et al.
      HNSCC cell lines positive for HPV and p16 possess higher cellular radiosensitivity due to an impaired DSB repair capacity.
      ] and impaired DSB repair capacity of the tumor cells has been suggested to be the predominant mechanism behind the pronounced radiosensitivity [
      • Rieckmann T.
      • Tribius S.
      • Grob T.J.
      • et al.
      HNSCC cell lines positive for HPV and p16 possess higher cellular radiosensitivity due to an impaired DSB repair capacity.
      ]. Whether HPV/p16-positive tumor cells from non-oropharyngeal sites also possess such enhanced in vitro radiosensitivity remains to be elucidated. Moreover, one might speculate that the improved radiotherapy-response observed for HPV/p16-positive OPC may also be associated with the proximity of these tumors to lymphoid tissue and that interplay between the immunesystem and the viral antigens may contribute to the enhanced radiosensitivity. Naturally, these hypotheses are far from consolidated and warrant further investigation.
      Investigation of the frequency and site-specific distribution of p16-expression (diffuse staining in >70% tumor cell) in this large cohort of advanced carcinoma of the pharynx and larynx revealed that p16-positivity is significantly more frequent in the oropharynx than in the larynx and hypopharynx, and that p16-positivity at the two last-mentioned sites is a relatively rare event. The use of p16-IHC as a marker of infection with HPV in OPC is widely used in clinical trials based on the high concordance between this method and various other HPV detection methods, including type specific HPV-DNA detection by situ hybridization (ISH) [
      • Jordan R.C.
      • Lingen M.W.
      • Perez-Ordonez B.
      • et al.
      Validation of methods for oropharyngeal cancer HPV status determination in US cooperative group trials.
      ]. Outside oropharynx the correlation is less robust and viral detection rates show considerable variation which in part can be ascribed to tumor-site differences but also the use of diverse detection methods is of importance [
      • Kreimer A.R.
      • Clifford G.M.
      • Boyle P.
      • Franceschi S.
      Human papillomavirus types in head and neck squamous cell carcinomas worldwide: a systematic review.
      ,
      • Isayeva T.
      • Li Y.
      • Maswahu D.
      • Brandwein-Gensler M.
      Human papillomavirus in non-oropharyngeal head and neck cancers: a systematic literature review.
      ,
      • Braakhuis B.J.
      • Snijders P.J.
      • Keune W.J.
      • et al.
      Genetic patterns in head and neck cancers that contain or lack transcriptionally active human papillomavirus.
      ]. Literature reviews estimate that the prevalence of HPV in larynx cancer is approximately 24%, based on PCR-detection methods [
      • Kreimer A.R.
      • Clifford G.M.
      • Boyle P.
      • Franceschi S.
      Human papillomavirus types in head and neck squamous cell carcinomas worldwide: a systematic review.
      ,
      • Isayeva T.
      • Li Y.
      • Maswahu D.
      • Brandwein-Gensler M.
      Human papillomavirus in non-oropharyngeal head and neck cancers: a systematic literature review.
      ], whereas data on the frequency in hypopharynx is sparse. Using combined HPV-testing with detection of HPV DNA by PCR and HPV E6/E7 RNA by RT-PCR in 222 HNSCCs, Salazar et al. [
      • Salazar C.R.
      • Smith R.V.
      • Garg M.K.
      • et al.
      Human papillomavirus-associated head and neck Squamous cell carcinoma survival: a comparison by tumor site and initial treatment.
      ] found a frequency of HPV-positivity of 35% in OPC, 29% in hypopharynx cancer and 5% in laryngeal tumors. They found significant prognostic impact of HPV-status in OPC, whereas HPV-status appeared to be of no significance for outcome in non-OPC regardless of the applied treatment modality. Expression of p16 is not restricted to HPV-positive tumors and using this marker alone as a surrogate of HPV-induced carcinogenesis inevitably entails the risk of including some (virally) false-positive results. We found that the frequency of p16-positivity was 14% for both larynx and hypopharynx carcinomas which is in agreement with the findings of Chung et al. [
      • Chung C.H.
      • Zhang Q.
      • Kong C.S.
      • et al.
      P16 protein expression and human papillomavirus status as prognostic biomarkers of non oropharyngeal head and neck squamous cell carcinoma.
      ] who also investigated the correlation between p16-IHC and HPV-16 ISH and found a moderate correlation for tumors of these sites.
      The strong and independent prognostic impact of tumor p16-status on radiotherapy outcome in advanced OPC is indisputable. In this study we observed, that the number of p16-positive tumors in the larynx and hypopharynx is relatively rare. Moreover, we did not find any indication that the prognosis for patients with p16-positive larynx and hypopharynx carcinoma differs from that of patients with p16-negative OPC and p16-negative non-OPC. Rather, these patient-groups have a significantly worse outcome to primary radiotherapy than patients with p16-positive OPC, and as such our findings failed to prove the hypothesis that the impact of tumor p16-status also extends to non-OPC. Consequently, our data suggests that patients with p16-positive tumors of the larynx and hypopharynx should be considered candidates for enhanced, multimodality treatment schedules in line with p16-negative HNSCC.

      Conflict of interest statement

      The authors have nothing to disclose.

      Acknowledgements

      This work was supported by grants from The Danish Cancer Society and Cirro-The Lundbeck Foundation Centre for Interventional Research in Radiation Oncology .

      Reference

        • Gillison M.L.
        • Koch W.M.
        • Capone R.B.
        • et al.
        Evidence for a causal association between human papillomavirus and a subset of head and neck cancers.
        J Natl Cancer Inst. 2000; 92: 709-720
        • Lassen P.
        The role of human papillomavirus in head and neck cancer and the impact on radiotherapy outcome.
        Radiother Oncol. 2010; 95: 371-380
        • Hammarstedt L.
        • Dahlstrand H.
        • Lindquist D.
        • et al.
        The incidence of tonsillar cancer in Sweden is increasing.
        Acta Otolaryngol. 2007; 127: 988-992
        • Chaturvedi A.K.
        • Engels E.A.
        • Pfeiffer R.M.
        • et al.
        Human papillomavirus and rising oropharyngeal cancer incidence in the United States.
        J Clin Oncol. 2011; 29: 4294-4301
        • Fakhry C.
        • Gillison M.L.
        Clinical implications of human papillomavirus in head and neck cancers.
        J Clin Oncol. 2006; 24: 2606-2611
        • Klussmann J.P.
        • Weissenborn S.J.
        • Wieland U.
        • et al.
        Human papillomavirus-positive tonsillar carcinomas: a different tumor entity?.
        Med Microbiol Immunol (Berl). 2003; 192: 129-132
        • Kreimer A.R.
        • Clifford G.M.
        • Boyle P.
        • Franceschi S.
        Human papillomavirus types in head and neck squamous cell carcinomas worldwide: a systematic review.
        Cancer Epidemiol Biomarkers Prev. 2005; 14: 467-475
        • Isayeva T.
        • Li Y.
        • Maswahu D.
        • Brandwein-Gensler M.
        Human papillomavirus in non-oropharyngeal head and neck cancers: a systematic literature review.
        Head Neck Pathol. 2012; 6: S104-S120
        • Overgaard J.
        • Hansen H.S.
        • Overgaard M.
        • et al.
        A randomized double-blind phase III study of nimorazole as a hypoxic radiosensitizer of primary radiotherapy in supraglottic larynx and pharynx carcinoma. Results of the Danish Head and Neck Cancer Study (DAHANCA) Protocol 5–85.
        Radiother Oncol. 1998; 46: 135-146
        • Overgaard J.
        • Hansen H.S.
        • Specht L.
        • et al.
        Five compared with six fractions per week of conventional radiotherapy of squamous-cell carcinoma of head and neck: DAHANCA 6 and 7 randomised controlled trial.
        Lancet. 2003; 362: 933-940
        • Lassen P.
        • Eriksen J.G.
        • Hamilton-Dutoit S.
        • Tramm T.
        • Alsner J.
        • Overgaard J.
        Effect of HPV-associated p16INK4A expression on response to radiotherapy and survival in squamous cell carcinoma of the head and neck.
        J Clin Oncol. 2009; 27: 1992-1998
        • Lassen P.
        • Eriksen J.G.
        • Hamilton-Dutoit S.
        • Tramm T.
        • Alsner J.
        • Overgaard J.
        HPV-associated p16-expression and response to hypoxic modification of radiotherapy in head and neck cancer.
        Radiother Oncol. 2010; 94: 30-35
        • Lassen P.
        • Eriksen J.G.
        • Krogdahl A.
        • et al.
        The influence of HPV-associated p16-expression on accelerated fractionated radiotherapy in head and neck cancer: evaluation of the randomised DAHANCA 6&7 trial.
        Radiother Oncol. 2011; 100: 49-55
        • Lassen P.
        • Overgaard J.
        • Eriksen J.G.
        Expression of EGFR and HPV-associated p16 in oropharyngeal carcinoma: correlation and influence on prognosis after radiotherapy in the randomized DAHANCA 5 and 7 trials.
        Radiother Oncol. 2013; 108: 489-494
        • Overgaard J.
        • Hoff C.M.
        • Hansen H.S.
        • Specht L.
        • Overgaard M.
        • Grau C.
        • et al.
        Randomized study of darbepoetin alfa as modifier of radiotherapy in patients with primary squamous cell carcinoma of the head and neck (HNSCC): Final outcome of the DAHANCA 10 trial.
        J Clin Oncol. 2009; 27 ([suppl.; abstr. 6007])
      1. Bentzen J., Toustrup K., Eriksen J.G., Primdahl H., Andersen L.J., Overgaard J., Locally advanced Head and Neck cancer treated with accelerated radiotherapy, the hypoxic modifier nimorazole and weekly cisplatin. Results from the DAHANCA 18 phase II study. Acta Oncol [in press].

        • Eriksen J.G.
        • Maare C.
        • Johansen J.
        • Primdahl H.
        • Evensen J.F.
        • Kristensen C.A.
        • et al.
        DAHANCA19: a randomized phase III study of primary (chemo-) radiotherapy and zalutumumab in head and neck carcinomas.
        Radiother Oncol. 2014; 111: 154-155
        • Ang K.K.
        • Harris J.
        • Wheeler R.
        • et al.
        Human papillomavirus and survival of patients with oropharyngeal cancer.
        N Engl J Med. 2010; 363: 24-35
        • Rischin D.
        • Young R.J.
        • Fisher R.
        • et al.
        Prognostic significance of p16INK4A and human papillomavirus in patients with oropharyngeal cancer treated on TROG 02.02 phase III trial.
        J Clin Oncol. 2010; 28: 4142-4148
        • Sorensen B.S.
        • Busk M.
        • Olthof N.
        • et al.
        Radiosensitivity and effect of hypoxia in HPV positive head and neck cancer cells.
        Radiother Oncol. 2013; 108: 500-505
        • Rieckmann T.
        • Tribius S.
        • Grob T.J.
        • et al.
        HNSCC cell lines positive for HPV and p16 possess higher cellular radiosensitivity due to an impaired DSB repair capacity.
        Radiother Oncol. 2013; 107: 242-246
        • Gillison M.L.
        • D’Souza G.
        • Westra W.
        • et al.
        Distinct risk factor profiles for human papillomavirus type 16-positive and human papillomavirus type 16-negative head and neck cancers.
        J Natl Cancer Inst. 2008; 100: 407-420
        • Chung C.H.
        • Zhang Q.
        • Kong C.S.
        • et al.
        P16 protein expression and human papillomavirus status as prognostic biomarkers of non oropharyngeal head and neck squamous cell carcinoma.
        J Clin Oncol. 2014; https://doi.org/10.1200/JCO.2013.54.5228
      2. Available at: http://clinicaltrials.gov/show/NCT01880359.

        • El-Naggar A.K.
        • Westra W.H.
        P16 expression as a surrogate marker for HPV-related oropharyngeal carcinoma: a guide for interpretative relevance and consistency.
        Head Neck. 2012; 34: 459-461
        • Lassen P.
        • Overgaard J.
        Scoring and classification of oropharyngeal carcinoma based on HPV-related p16-expression.
        Radiother Oncol. 2012; 105: 269-270
        • Michiels S.
        • Le M.A.
        • Buyse M.
        • et al.
        Surrogate endpoints for overall survival in locally advanced head and neck cancer: meta-analyses of individual patient data.
        Lancet Oncol. 2009; 10: 341-350
        • Harari P.M.
        • Harris J.
        • Kies M.S.
        • et al.
        Postoperative chemoradiotherapy and cetuximab for high-risk squamous cell carcinoma of the head and neck: Radiation Therapy Oncology Group RTOG-0234.
        J Clin Oncol. 2014; 32: 2486-2495
        • Hoff C.M.
        • Grau C.
        • Overgaard J.
        Effect of smoking on oxygen delivery and outcome in patients treated with radiotherapy for head and neck squamous cell carcinoma – a prospective study.
        Radiother Oncol. 2012; 103: 38-44
        • Jordan R.C.
        • Lingen M.W.
        • Perez-Ordonez B.
        • et al.
        Validation of methods for oropharyngeal cancer HPV status determination in US cooperative group trials.
        Am J Surg Pathol. 2012; 36: 945-954
        • Braakhuis B.J.
        • Snijders P.J.
        • Keune W.J.
        • et al.
        Genetic patterns in head and neck cancers that contain or lack transcriptionally active human papillomavirus.
        J Natl Cancer Inst. 2004; 96: 998-1006
        • Salazar C.R.
        • Smith R.V.
        • Garg M.K.
        • et al.
        Human papillomavirus-associated head and neck Squamous cell carcinoma survival: a comparison by tumor site and initial treatment.
        Head Neck Pathol. 2014; 8: 77-87