Longer survival in patients with human papillomavirus related head and neck cancer after positive postradiation planned neck dissection

Transcription

1 ORIGINAL ARTICLE Longer survival in patients with human papillomavirus related head and neck cancer after positive postradiation planned neck dissection Shao Hui Huang, MSc, MD, MRT(T), 1 Samip Patel, MD, 2 Brian O Sullivan, MD, 1,2 Xiaowei Shen, MMath, 3 Wei Xu, PhD, 3 Ilan Weinreb, MD, 4 Bayardo Perez Ordonez, MD, 4 Jonathan Irish, MD, 2 John Waldron, MD, 1,2 Patrick Gullane, MD, 2 Ralph Gilbert, MD, 2 Dale Brown, MD, 2 John Kim, MD, 1 Jeremy Freeman, MD, 2 John R. de Almeida, MD, MSc, 2 David Goldstein, MD 2* 1 Department of Radiation Oncology, Princess Margaret Cancer Centre/University of Toronto, Toronto, Canada, 2 Department of Otolaryngology Head and Neck Surgery/Surgical Oncology, Princess Margaret Cancer Centre/University of Toronto, Toronto, Canada, 3 Department of Biostatistics, Princess Margaret Cancer Centre, Toronto, Canada, 4 Department of Pathology, Princess Margaret Centre, Toronto, Canada. Accepted 11 March 2014 Published online 25 September 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI /hed ABSTRACT: Background. The purpose of this study was to compare outcomes between human papillomavirus (HPV)-related versus -unrelated head and neck cancer after pathological positive planned neck dissection. Methods. Positive planned neck dissection for head and neck cancers from 1998 to 2010 were included in this study. Outcomes after planned neck dissection were compared between HPV-related versus -unrelated cohorts. Multivariate analysis identified survival predictors. Results. HPV-related head and neck cancer (n 5 32) had better 5-year overall survival (48% vs 27%; p 5.021), marginally lower second malignancy (7% vs 16%; p 5.13), but similar local, regional, and distant control (87% vs 89%; 94% vs 89%; 62% vs 58%, respectively) versus HPV unrelated (n 5 38). HPV status conferred reduced risk of death (hazard ratio [HR], 0.5; p 5.038) after adjusting for age, smoking, and initial T and N classifications. Conclusion. This study reveals that positive planned neck dissection for HPV-related head and neck cancer represents a biologic unfavorable subset of the HPV population with unsatisfactory survival attributable to distant metastasis. The longer survival compared to the HPV-unrelated counterpart is likely related to a marginally lower second malignancy rather than better disease control. VC 2014 Wiley Periodicals, Inc. Head Neck 37: , 2015 KEY WORDS: human papillomavirus, planned neck dissection, head and neck cancer, radiotherapy, survival INTRODUCTION It has been well-recognized that two predominantly different disease entities originate in the mucosa of the head and neck: smoking-related (human papillomavirus [HPV]-unrelated) and HPV-related head and neck cancers. HPV-related head and neck cancer typically arise from oropharyngeal regions or may present as an unknown primary. 1 They are uncommon in the larynx, hypopharynx, and oral cavity. 2 4 Although cervical lymph node involvement is common in HPV-related head and neck cancer, they are highly responsive to radiotherapy and have superior local and regional control and survival compared to their HPV-unrelated counterparts. 5,6 In the organ-preservation era, primary radiotherapy with or without concurrent chemotherapy is one of the main treatment strategies for patients with locally advanced laryngeal and pharyngeal cancer, as well as for *Corresponding author: D. Goldstein, Department of Otolaryngology Head and Neck Surgery/Surgical Oncology, Princess Margaret Cancer Centre/University of Toronto, 610 University Ave, Room 3-952, Toronto, Ontario M5G 2M9, Canada. David.Goldstein@uhn.on.ca Contract grant sponsor: This work was supported by the Bartley Smith/Wharton, the Gordon Tozer, the Wharton Head and Neck Translational, Dr. Mariano Elia, and the Joe s Team Funds at the Princess Margaret Hospital Foundation. unknown primary. Neck dissection after radiation is generally offered to those with radiological residual disease in the absence of disease at primary or distant sites. 7 Previously, neck dissections after radiotherapy were frequently planned based on the extent of pretreatment nodal burden. Over time, particularly with improvements in imaging, there has been a shift to performing neck dissection based on clinical and radiographic response to treatment. In the latter scenario, imaging is planned 10 to 12 weeks after completion of treatment with a neck dissection performed in those with radiographic evidence of residual nodal disease. Some may consider this as salvage 8 whereas others still consider it as planned because a neck dissection is considered part of the initial treatment plan, and viable tumor foci are not always present in the postradiation neck dissection specimen. 7,9 For the purpose of differentiating neck dissection for potential persistent neck disease immediately after radiation (<20 weeks) from a salvage dissection for recurrent neck disease manifesting later (in which viable tumor is almost always present), we have used the Radiation Therapy Oncology Group (RTOG) 1016 protocol definition 9 and refer to neck dissection after radiotherapy for potential persistent neck disease based on planned posttreatment imaging as a planned neck dissection. 10 This should be contrasted with neck dissections 946 HEAD & NECK DOI /HED JULY 2015

2 SURVIVAL AFTER POSITIVE PLANNED NECK DISSECTION performed for progression of regional disease or as a part of surgery for local persistent disease after radiotherapy. Although patients with HPV-related head and neck cancer are less likely to have a pathological positive planned neck dissection, 11 residual neck disease may still be present in a small subset. 10,11 Several studies have demonstrated that the presence of viable tumor in the planned neck dissection specimen is associated with diminished survival outcome, mainly attributed to higher rates of distant metastasis. 7,8,10,12,13 Whether a survival difference by HPV status exists in this setting has not been investigated, although secondary analyses of clinical trials have demonstrated a longer survival in patients with HPV-related head and neck cancer with recurrent or metastatic disease after palliative systemic therapy We recently reported outcomes of patients with head and neck cancer after pathological positive planned neck dissection and identified that extracapsular extension, carcinoma within soft tissue, and/or lymphovascular invasion, were all adverse survival predictors for patients with positive planned neck dissection. 10 The prognostic value of HPV status was not analyzed in that report. With emerging evidence that patients with HPV-related head and neck cancer have a longer survival even in the recurrent/metastatic setting, we sought to determine whether HPV status is associated with improved outcomes in patients with residual disease in the planned neck dissection specimens compared to patients with HPV-unrelated head and neck cancer, and also to explore its underlying causes. PATIENTS AND METHODS Patient selection After institutional research ethics board approval, a retrospective chart review was conducted for all newly diagnosed patients with head and neck cancer with pathologically positive planned neck dissection after primary radiotherapy or chemoradiotherapy in our institution from 1998 through Planned neck dissection was defined based on the Radiation Therapy Oncology Group (RTOG) 1016 protocol (ie, a neck dissection performed immediately after radiotherapy [<20 weeks]) in patients without clinical evidence of persistent primary tumor, progression of regional disease, or detection of distant metastases. 9 Inclusion criteria were: (1) squamous cell carcinoma of the oropharynx, hypopharynx, larynx, or oral cavity, and unknown primary; (2) primary radiotherapy with curative intent and received at least 60 Gy of radiation dose; (3) viable tumor foci identified in the neck dissection specimen, reviewed by the 2 specialized head and neck pathologists in our institution (I.W. and B.P.-O.). For oropharyngeal cancer and unknown primary, tumor HPV status was ascertained by p16 staining, whereas those with insufficient tissue for p16 staining in these groups were excluded from final analysis. Oropharyngeal cancer or unknown primary with strongly and/or diffusely p16 staining was classified as HPV-related tumor, whereas those with negative p16 staining were classified as HPV-unrelated tumor. 17 Because of the known low prevalence of HPV driven head and neck cancer among non-oropharyngeal cancer cases, 3 5,18,19 and because HPV-unrelated oropharyngeal cancer/unknown primary and non-oropharyngeal cancer behaved similarly in a preliminary analysis, cancers of the larynx, hypopharynx, and oral cavity were analyzed as HPV-unrelated head and neck cancer. This follows the method used in the RTOG 0129 analysis by Ang et al 5 and our previous reports. 20 Radiation treatment and neck management after radiotherapy All patients included in the present analysis underwent curative treatment with definitive radiotherapy with or without concurrent chemotherapy. Patients were routinely managed, evaluated, and imaged after treatment, as previously described. 10 Before 2003, a planned neck dissection was generally performed 8 to 12 weeks after radiotherapy in patients who had nodal metastases at initial presentation as part of a radiotherapy alone (without chemotherapy) clinical trial. 21 From 2003 onward, planned neck dissection was generally performed for patients who had radiologic residual disease (residual lymph nodes >1.0 cm in axial diameter on CT, MRI, and/or more recently a positive [PET] scan 8 to 12 weeks after completion of radiotherapy). The detailed neck management protocol and type of planned neck dissection are available in a previous report from our group. 10 Statistical analysis Clinical characteristics and outcomes after positive planned neck dissection were compared between the HPV-related and HPV-unrelated cohorts. Chi-square test (Fisher s test for frequency <5) was used for comparison of categorical variables and 2-tailed t test was used for continuous variables. Actuarial rates of overall survival (OS) and late toxicity ( grade 3 by RTOG/European Organization for Research and Treatment of Cancer criteria 22 ) were estimated by Kaplan Meier analysis, whereas cause-specific survival (CSS), local control, regional control, locoregional control, distant control, and second malignancy rates were estimated using competing risk methods. The log-rank test was used for outcome comparison. All time-to-event analyses were calculated from the date of planned neck dissection. All tests were 2- tailed with a probability of <.05 considered statistically significance. Finally, multivariate analysis was applied to identify survival predictors using a Cox regression model that included the following parameters: HPV status, age (by 5-year increments), smoking pack-years (by 10 packyear increments), N classification (N1 N2b vs N2c N3), and T classification (T3 4 vs T0 T2). RESULTS Planned neck dissection was performed in 291 patients during the study period, of whom 81 (28%) had viable residual tumor foci in the planned neck dissection specimen. The distribution of planned neck dissection positivity and disease site are shown in Table 1. Eleven patients (5 patients with oropharyngeal cancer and 6 with unknown primary carcinoma) with insufficient tissue blocks for p16 staining were excluded from this study. HEAD & NECK DOI /HED JULY

3 HUANG ET AL. TABLE 1. site. Results of postradiation planned neck dissection by disease Disease site P16 Negative PND Positive PND Oropharynx Positive Negative 10 9 Not tested 68 5 Unknown primary Positive 1 3 Negative 0 1 Not tested 27 6 Larynx Not tested 8 8 Hypopharynx Not tested Oral cavity Not tested 0 4 The remaining 70 patients were eligible for the study. P16-negative oropharyngeal cancer/unknown primary and non-oropharyngeal cancer (larynx, hypopharynx, and oral cavity) had similar 5-year OS (24% [0 52] vs 24% [7 4]; p 5.71), locoregional control (74% [43 100] vs 72%, [48 95]; p 5.77), and distant control (58% [27 90] vs 41%, [18 64]; p 5.83), and were considered as a single HPV unrelated head and neck cancer cohort in the further analysis. The clinical characteristics of the 32 HPVrelated (39 oropharyngeal cancer and 3 unknown primary) and 38 patients with HPV-unrelated (9 oropharyngeal cancer, 1 unknown primary, 8 larynx, 16 hypopharynx, 4 oral cavity) head and neck cancer are listed in Table 2. Compared to the HPV-unrelated, the HPV-related cohort had a lower proportion of heavy smokers (>10 pack-years) (66% vs 89%; p 5.02) and a similar distribution of initial T classification (T3 4, 47% vs 47%; p >.999) and N classification (N2c N3, 47% vs 56%; p 5.48). Severe late toxicity rates were similar between the HPV-related versus HPV-unrelated cohorts (25% vs 32%; p 5.49). Median follow-up after planned neck dissection for surviving patients was 5.2 years. Recurrences (local, regional, or distant) were identified in 14 of 32 patients (44%) with HPV-related and 19 of 38 patients (50%) with HPV-unrelated cancer (Figure 1). Salvage surgery was performed in 6 of 14 patients (43%) with HPVrelated and 4 of 19 patients (21%) with HPV-unrelated cancer (p 5.13). Distant metastasis, the primary form of subsequent failure for both cohorts, without locoregional failure was evident in 9 of 14 (64%) HPV-related and 12 of 19 (63%) HPV-unrelated subsequent recurrent cases (Figure 1). There was no difference in local control (87% vs 89%; p 5.76), regional control (94% vs 89%; p 5.38), or distant control (62% vs 58%; p 5.68) after positive planned neck dissection between the HPV-related versus HPV-unrelated patients. Although not statistically significant, there was a trend toward lower second malignancy rates in the HPV-related cohort (7% vs 16%; p 5.13; Figure 2). At the time of analysis, 18 HPV-related and 31 HPVunrelated patients were deceased, of whom 13 (72%) HPV-related and 19 (65%) HPV-unrelated patients died of their index cancer (almost all from distant metastasis) in both cohorts (Table 2). Compared to the HPV-unrelated cohort, HPV-related patients had a significantly better OS at 5 years (48% vs 27%; p 5.02), but no apparent difference in CSS (58% [36 73] vs 49% [29 63]; p 5.33) (Figure 2). On multivariate analysis, HPV status was independently associated with survival. After adjusting for age, smoking pack-years, initial T classification, and N classification, HPV-related tumors were associated with a reduced risk of death (hazard ratio [HR], 0.51; ; p 5.038) after positive planned neck dissection (Table 3). DISCUSSION Several reports have shown that viable tumor in the neck after primary radiotherapy or chemoradiotherapy is Abbreviation: PND, planned neck dissection. TABLE 2. Clinical characteristics of human papillomavirus positive versus human papillomavirus hx2010;negative patients with positive planned neck dissection. Variable, no. (%) HPV(1) (n 5 32) HPV(-) (n 5 38) p value Age, y, median (range) 59 (43 78) 62 (41 81).238 Sex Female 7 (22) 4 (11) Male 25 (78) 34 (89).320 Disease site Hypopharynx 0 (0) 16 (42) Larynx 0 (0) 8 (21) Oral cavity 0 (0) 4 (10) Oropharynx 29 (91) 9 (24) Unknown primary 3 (9) 1 (3) Tobacco exposure, 21 (66) 34 (89).015 >10 pack-years T classification T (53) 17 (53) T (47) 15 (47) >.999 N classification N1 N2b 20 (53) 14 (44) N2c N3 18 (47) 18 (56).480 Chemotherapy 10 (31) 10 (26).648 Interval to PND, wk, median (range) ( ) ( ) Recurrence, any Total 14 (44) 19 (50) Local Regional Distant Outcomes at 5 y (95% CI) OS 48% (77 70) 27% (16 47).021 CSS 58% (36 73) 49% (29 63),.33 Local control 87% (68 95) 89% (72 96).760 Regional control 94% (75 98) 89% (73 96).381 Distant control 62% (40 76) 58% (38 71).680 Second 7% (2 27) 16% (8 35).130 malignancy rate Severe late 25% (7 40) 32% (11 47).490 toxicity rate Death Total 18 (56) 31 (82).035 Index cancer 13 (41) 21 (55).900 Other cancer 2 (6) 6 (16) Other cause 3 (9) 5 (13) Abbreviations: HPV(1), human papillomavirus-related; HPV(2), human papillomavirusunrelated; PND, planned neck dissection; CI, confidence interval; OS, overall survival; CSS, cause-specific survival. 948 HEAD & NECK DOI /HED JULY 2015

4 SURVIVAL AFTER POSITIVE PLANNED NECK DISSECTION FIGURE 1. Site of subsequent failure after positive planned neck dissection in human papillomavirus-related (HPV(1)) (14 of 32) versus HPVunrelated (HPV(2)) (19 of 38) patients. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] associated with poor survival, 7,8,10,12 but the prognostic value of HPV status in this setting has not been explored. This modest sized retrospective review demonstrates that HPV status is an independent prognostic factor in patients with residual neck disease after radiotherapy or chemoradiotherapy. Compared with HPV-unrelated patients, patients with HPV-related head and neck cancer had higher survival rates despite no difference in local, regional, and distant control rates. Subsequent distant failure predominates in both patient subgroups and is the main cause of death for both cohorts, whereas subsequent local or regional failure is infrequent. The present study confirms that there is a subgroup of patients with HPV-related head and neck cancer with pathologically residual neck disease that fare less well with <50% surviving 5 years. However, they still live longer compared to their HPV-unrelated counterparts. Our evaluation suggests that this is not entirely related to anatomic disease control. A contributing factor for longer survival in HPV-related patients may relate to their lighter smoking history. Smoking is likely to have contributed to comorbidity mortality and to second primary causation in this group of patients. The 3 leading causes of death from smoking are cardiovascular disease, chronic obstructive pulmonary disease, and lung cancer. 23 Smoking can also result in increased radiation-related late toxicities, 24 including severe fibrosis that may ultimately be fatal (eg, silent aspiration pneumonia). Interestingly, smoking packyears was not an adverse predictor of survival in the multivariate analysis. This is likely because of the small sample size of light/never smokers in this heavy smoker dominating cohort (66% HPV-related and 89% HPVunrelated cohort were >10 pack-year smokers) who tended to present with a more adverse disease profile. In turn, this placed them at higher risk of requiring a planned neck dissection that yielded residual disease, compared to light and never smokers. Longer survival in patients with HPV-related head and neck cancer may also relate to tumor biology. The prognostic value of HPV is well-proven in the initial treatment setting. 5,6 Emerging evidence also suggests a better response to systemic agent for HPV-related versus HPVunrelated head and neck cancer in the recurrent/metastatic setting, which may prolong their survival after recurrence. 16 Although anatomic disease control is similar, it is also possible that some HPV-related patients may have a relatively indolent course after recurrence, a higher chance of being salvaged, and/or higher response rate to palliative chemotherapy. Our data could not definitively confirm this as there was no statistical significant difference in CSS (p 5.33; Figure 2E), although a 9% absolute difference (58% vs 49%) in CSS at 5 years is observed between the 2 cohorts. This is probably explained by the small sample size. In addition, our data showed that a slightly higher proportion of patients with HPV-related head and neck cancer underwent successful salvage surgery (6 of 14 vs 4 of 19; p 5.126), although the number is too small to make a definitive conclusion. It has been observed that a small subset of patients with HPV-positive oropharyngeal cancer with oligometastasis may survive longer than expected after salvage procedures. 25,26 Last, the difference in survival between the 2 groups may be related to second primaries. Although not statistically significant, there was a trend toward a lower second primary rate in the patients with HPV-related head and neck cancer. The presence of second primaries would potentially increase the risk of death, but the numbers in this study are too small to address this; conceivably, factors associated with a difference in survival between the 2 groups could further be delineated with a larger sample size. Over the past decade, HPV-related head and neck cancer has rapidly emerged as a disease entity that is HEAD & NECK DOI /HED JULY

5 HUANG ET AL. FIGURE 2. Comparison of outcomes between human papillomavirus-related (HPV(1)) and HPV-unrelated (HPV(2)) cohorts. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] transforming the paradigm of mucosal head and neck cancer management. The strikingly different outcomes, irrelevant of treatment modality, between HPV-related and HPV-unrelated head and neck cancer have necessitated reinterpretation of previous clinical trials in which mucosal head and neck cancer was considered as one homogeneous disease without knowledge of HPV status. Although specific mature clinical trial data for HPVrelated head and neck cancer is still lacking, insight may be gained by observing the natural course of HPV-related head and neck cancer through other sources, such as available retrospective data from clinical trials or observation cohorts. The current study is limited by its retrospective nature, although clinical information for the majority of patients included in this study was recorded prospectively in a prospective database. 27 HPV status was evaluated using p16 staining as a surrogate marker rather than directly detecting HPV using polymerase chain reaction methods, which may be susceptible to misclassification. 28 However, p16 is a well-accepted surrogate for HPV- TABLE 3. Multivariate analysis of survival. Variables HR (95% CI) p value HPV(1) vs HPV(2) 0.51 ( ).038 Age, by 5-y increments 1.11 ( ).25 Smoking pack-years, by 1.01 ( ) pack-year increments T classification: T3 T4 vs T0 T ( ).90 N classification: N2c N3 vs T1 N2b 1.35 ( ).32 Abbreviations: HR, hazard ratio; CI, confidence interval; HPV(1), human papillomavirusrelated; HPV(2), human papillomavirus-unrelated. 950 HEAD & NECK DOI /HED JULY 2015

6 SURVIVAL AFTER POSITIVE PLANNED NECK DISSECTION driven tumor in the oropharynx 29 and our pathologists have extensive experience in interpretation of the required staining characteristics. Whether p16 is a reliable surrogate for HPV-driven tumors in the unknown primary setting is less defined. However, overexpression of p16 correlates significantly with an occult primary cancer in the oropharynx. 30 HPV status was not evaluated in nonoropharyngeal cancer (laryngeal, hypopharyngeal, and oral cavity cancer) because of unavailability of tissue. Although there is little possibility of misclassifying HPVrelated non-oropharyngeal cancer as HPV-unrelated tumors and potentially overestimating survival in HPV unrelated cohort, the prevalence of HPV-related tumors in non-oropharyngeal cancer head and neck cancer is relatively low and their classification as HPV-unrelated tumors was considered appropriate in the recent RTOG 0129 report. 5 In addition, our preliminary analysis showed the outcomes were essentially the same between p16- confirmed HPV-unrelated oropharyngeal cancer/unknown primary and non-oropharyngeal cancer head and neck cancer. Thus, combining both cohorts as HPV-unrelated head and neck cancer seems reasonable. Finally, the main purpose of this study was to understand the natural course of HPV-related head and neck cancer with a positive planned neck dissection, in which the HPV status determination is relatively robust and the survival estimation is considered accurate. The conclusion that longer survival in HPV-related versus HPV-unrelated head and neck cancer with the positive planned neck dissection population is unlikely to be altered by misclassifying a few HPVrelated non-oropharyngeal cancers as HPV-unrelated head and neck cancer, even though the survival might be overestimated in the HPV-unrelated head and neck cancer cohort. Although HPV-positive status usually implies increased tumor burden within the neck, p16 positivity (a surrogate marker for HPV) heralds excellent response rates to primary radiotherapy or chemoradiotherapy and patients with HPV-related head and neck cancer are less likely to have residual neck disease after primary radiotherapy. 11 However, a proportion of patients with HPVrelated head and neck cancer exist with residual neck disease after treatment. These patients seem to represent a less favorable biologic subset of the HPV-related head and neck cancer population with a relatively high distant metastasis rate (40% at 5 years in the current study). Continued efforts to better understand their biology and optimizing better treatments is warranted for these patients. CONCLUSIONS This modest sized cohort study attests that patients with HPV-related head and neck cancer with residual neck disease represent a less favorable biologic subset of the HPV-related head and neck cancer population and have poor survival. However, they generally still live longer than their HPV-unrelated counterparts. Subsequent regional failure was infrequent. Distant metastasis was the main form of failure after positive planned neck dissection in both cohorts. The longer survival seems not directly related to subsequent disease control, but rather attributed to the marginally lower second primary malignancy rate, lower smoking-related comorbidity, slightly higher salvage opportunities, better response to palliative chemotherapy, and potentially a slower natural course after recurrence. REFERENCES 1. Gillison ML, 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: Halec G, Holzinger D, Schmitt M, et al. Biological evidence for a causal role of HPV16 in a small fraction of laryngeal squamous cell carcinoma. Br J Cancer 2013;109: Wendt M, Romanitan M, Nasman A, et al. Presence of human papillomaviruses and p16 expression in hypopharyngeal cancer. Head Neck 2014;36: Lingen MW, Xiao W, Schmitt A, et al. Low etiologic fraction for high-risk human papillomavirus in oral cavity squamous cell carcinomas. Oral Oncol 2013;49: Ang KK, Harris J, Wheeler R, et al. Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med 2010;363: O Sullivan B, Huang SH, Siu LL, et al. Deintensification candidate subgroups in human papillomavirus-related oropharyngeal cancer according to minimal risk of distant metastasis. J Clin Oncol 2013;31: Lango MN, Andrews GA, Ahmad S, et al. Postradiotherapy neck dissection for head and neck squamous cell carcinoma: pattern of pathologic residual carcinoma and prognosis. Head Neck 2009;31: Ganly I, Bocker J, Carlson DL, et al. Viable tumor in postchemoradiation neck dissection specimens as an indicator of poor outcome. Head Neck 2011;33: Trotti A, Gillison M. Phase III trial of radiotherapy plus cetuximab versus chemoradiotherapy in HPV-associated oropharynx cancer. [NCT ], (ed June 9, 2011), RTOG, Huang SH, O Sullivan B, Weinreb I, et al. Patterns of failure and histopathologic outcome predictors following definitive radiotherapy and planned neck dissection with residual disease. Head Neck 2012;34: Shonka DC Jr, Shoushtari AN, Thomas CY, et al. Predicting residual neck disease in patients with oropharyngeal squamous cell carcinoma treated with radiation therapy: utility of p16 status. Arch Otolaryngol Head Neck Surg 2009;135: Simon C, Goepfert H, Rosenthal DI, et al. Presence of malignant tumor cells in persistent neck disease after radiotherapy for advanced squamous cell carcinoma of the oropharynx is associated with poor survival. Eur Arch Otorhinolaryngol 2006;263: Sandulache VC, Ow TJ, Daram SP, et al. Residual nodal disease in patients with advanced-stage oropharyngeal squamous cell carcinoma treated with definitive radiation therapy and posttreatment neck dissection: association with locoregional recurrence, distant metastasis, and decreased survival. Head Neck 2013;35: Stoehlmacher Williams J, Villanueva C, Foa P, et al. Safety and efficacy of panitumumab (pmab) in HPV-positive (1) and HPV-negative (2) recurrent/metastatic squamous cell carcinoma of the head and neck (R/M SCCHN): analysis of the global phase III SPECTRUM trial. J Clin Oncol 2012;30(Suppl): abstract Mehra R, Egloff AM, Li S, et al. Analysis of HPV and ERCC1 in recurrent or metastatic squamous cell carcinoma of the head and neck (R/M SCCHN). J Clin Oncol 2013;31(Suppl): abstract Vermorken JB, St ohlmacher J, Oliner KS, et al. Safety and efficacy of panitumumab (pmab) in HPV positive (1) and HPV negative (2) recurrent/ metastatic (R/M) squamous cell carcinoma of the head and neck (SCCHN): analysis of the phase III SPECTRUM trial. Eur J Cancer 2011;47: Shi W, Kato H, Perez Ordonez B, et al. Comparative prognostic value of HPV16 E6 mrna compared with in situ hybridization for human oropharyngeal squamous carcinoma. J Clin Oncol 2009;27: Mehanna H, Beech T, Nicholson T, et al. Prevalence of human papillomavirus in oropharyngeal and nonoropharyngeal head and neck cancer systematic review and meta-analysis of trends by time and region. Head Neck 2013;35: Hobbs CG, Sterne JA, Bailey M, et al. Human papillomavirus and head and neck cancer: a systematic review and meta-analysis. Clin Otolaryngol 2006;31: Huang SH, O Sullivan B, Xu W, et al. Temporal nodal regression and regional control after primary radiation therapy for N2-N3 head-and-neck cancer stratified by HPV status. Int J Radiat Oncol Biol Phys 2013;87: Waldron J, Warde P, Irish J, et al. A dose escalation study of hyperfractionated accelerated radiation delivered with integrated neck surgery (HARD- WINS) for the management of advanced head and neck cancer. Radiother Oncol 2008;87: Cox JD, Stetz J, Pajak TF. Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 1995;31: HEAD & NECK DOI /HED JULY

7 HUANG ET AL. 23. Ezzati M, Lopez AD. Estimates of global mortality attributable to smoking in Lancet 2003;362: Chen AM, Chen LM, Vaughan A, et al. Tobacco smoking during radiation therapy for head-and-neck cancer is associated with unfavorable outcome. Int J Radiat Oncol Biol Phys 2011;79: Huang SH, Perez Ordonez B, Weinreb I, et al. Natural course of distant metastases following radiotherapy or chemoradiotherapy in HPV-related oropharyngeal cancer. Oral Oncol 2013;49: Sinha P, Thorstad WT, Nussenbaum B, et al. Distant metastasis in p16- positive oropharyngeal squamous cell carcinoma: a critical analysis of patterns and outcomes. Oral Oncol 2014;50: Wong K, Huang SH, O Sullivan B, et al. Point-of-care outcome assessment in the cancer clinic: audit of data quality. Radiother Oncol 2010;95: Robinson M, Sloan P, Shaw R. Refining the diagnosis of oropharyngeal squamous cell carcinoma using human papillomavirus testing. Oral Oncol 2010;46: El-Naggar AK, Westra WH. p16 expression as a surrogate marker for HPV-related oropharyngeal carcinoma: a guide for interpretative relevance and consistency. Head Neck 2012;34: Vent J, Haidle B, Wedemeyer I, et al. p16 Expression in carcinoma of unknown primary: diagnostic indicator and prognostic marker. Head Neck 2013;35: HEAD & NECK DOI /HED JULY 2015