Defining the Prevalence and Prognostic Value of Perineural Invasion and Angiolymphatic Invasion in Human Papillomavirus–Positive Oropharyngeal Carcinoma | Head and Neck Cancer | JAMA Otolaryngology–Head & Neck Surgery | JAMA Network
[Skip to Navigation]
Figure 1.  Kaplan-Meier Curve of Overall Survival Stratified by Absence of Angiolymphatic Invasion (ALI) and Perineural Invasion (PNI) vs Presence of at Least 1 ALI or 1 PNI
Kaplan-Meier Curve of Overall Survival Stratified by Absence of Angiolymphatic Invasion (ALI) and Perineural Invasion (PNI) vs Presence of at Least 1 ALI or 1 PNI

Hazard ratio, 2.78; 95% CI, 1.15-6.76.

Figure 2.  Kaplan-Meier Curve of Disease-Free Survival Stratified by Absence of Angiolymphatic Invasion (ALI) and Perineural Invasion (PNI) vs Presence of at Least 1 ALI or 1 PNI
Kaplan-Meier Curve of Disease-Free Survival Stratified by Absence of Angiolymphatic Invasion (ALI) and Perineural Invasion (PNI) vs Presence of at Least 1 ALI or 1 PNI

Hazard ratio, 3.10; 95% CI, 1.17-8.23.

Figure 3.  Overall Survival in Each Stage Defined by the Eighth Edition of the American Joint Committee on Cancer Manual Stratified by the Presence of Angiolymphatic Invasion (ALI) and/or Perineural Invasion (PNI)
Overall Survival in Each Stage Defined by the Eighth Edition of the American Joint Committee on Cancer Manual Stratified by the Presence of Angiolymphatic Invasion (ALI) and/or Perineural Invasion (PNI)

A, Stage I (hazard ratio [HR], 1.11; 95% CI, 0.28-5.31); B, stage II (HR, 11.7; 95% CI, 1.23-111.67); C, stage III (HR not estimable); D, stage IV (HR not estimable).

Table 1.  Baseline Characteristics of 201 HPV-Positive Patients With OPSCC Included in the Study
Baseline Characteristics of 201 HPV-Positive Patients With OPSCC Included in the Study
Table 2.  Prevalence of ALI and PNI in a Cohort of Surgically Treated HPV-Positive Patients With OPSCC Stratified by AJCC-8 Pathologic Stage
Prevalence of ALI and PNI in a Cohort of Surgically Treated HPV-Positive Patients With OPSCC Stratified by AJCC-8 Pathologic Stage
1.
Chaturvedi  AK, Engels  EA, Pfeiffer  RM,  et al.  Human papillomavirus and rising oropharyngeal cancer incidence in the United States.  J Clin Oncol. 2011;29(32):4294-4301.PubMedGoogle ScholarCrossref
2.
Gillison  ML.  Human papillomavirus-associated head and neck cancer is a distinct epidemiologic, clinical, and molecular entity.  Semin Oncol. 2004;31(6):744-754.PubMedGoogle ScholarCrossref
3.
Fischer  CA, Zlobec  I, Green  E,  et al.  Is the improved prognosis of p16 positive oropharyngeal squamous cell carcinoma dependent of the treatment modality?  Int J Cancer. 2010;126(5):1256-1262.PubMedGoogle Scholar
4.
Wierzbicka  M, Szyfter  K, Milecki  P, Składowski  K, Ramlau  R.  The rationale for HPV-related oropharyngeal cancer de-escalation treatment strategies.  Contemp Oncol (Pozn). 2015;19(4):313-322.PubMedGoogle Scholar
5.
Huang  SH, Xu  W, Waldron  J,  et al.  Refining American Joint Committee on Cancer/Union for International Cancer Control TNM stage and prognostic groups for human papillomavirus–related oropharyngeal carcinomas.  J Clin Oncol. 2015;33(8):836-845.PubMedGoogle ScholarCrossref
6.
Dahlstrom  KR, Calzada  G, Hanby  JD,  et al.  An evolution in demographics, treatment, and outcomes of oropharyngeal cancer at a major cancer center: a staging system in need of repair.  Cancer. 2013;119(1):81-89.PubMedGoogle ScholarCrossref
7.
American Joint Committee on Cancer.  American Joint Committee on Cancer (AJCC) Staging Manual. 8th ed. New York, NY: Springer; 2016.
8.
Haughey  BH, Sinha  P, Kallogjeri  D,  et al.  Pathology-based staging for HPV-positive squamous carcinoma of the oropharynx.  Oral Oncol. 2016;62:11-19.PubMedGoogle ScholarCrossref
9.
Woolgar  JA.  Histopathological prognosticators in oral and oropharyngeal squamous cell carcinoma.  Oral Oncol. 2006;42(3):229-239.PubMedGoogle ScholarCrossref
10.
Binmadi  NO, Basile  JR.  Perineural invasion in oral squamous cell carcinoma: a discussion of significance and review of the literature.  Oral Oncol. 2011;47(11):1005-1010.PubMedGoogle ScholarCrossref
11.
Massano  J, Regateiro  FS, Januário  G, Ferreira  A.  Oral squamous cell carcinoma: review of prognostic and predictive factors.  Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2006;102(1):67-76.PubMedGoogle ScholarCrossref
12.
Jardim  JF, Francisco  ALN, Gondak  R, Damascena  A, Kowalski  LP.  Prognostic impact of perineural invasion and lymphovascular invasion in advanced stage oral squamous cell carcinoma.  Int J Oral Maxillofac Surg. 2015;44(1):23-28.PubMedGoogle ScholarCrossref
13.
Rahima  B, Shingaki  S, Nagata  M, Saito  C.  Prognostic significance of perineural invasion in oral and oropharyngeal carcinoma.  Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2004;97(4):423-431.PubMedGoogle ScholarCrossref
14.
Fagan  JJ, Collins  B, Barnes  L, D’Amico  F, Myers  EN, Johnson  JT.  Perineural invasion in squamous cell carcinoma of the head and neck.  Arch Otolaryngol Head Neck Surg. 1998;124(6):637-640.PubMedGoogle ScholarCrossref
15.
Yilmaz  T, Hosal  AS, Gedikoğlu  G, Onerci  M, Gürsel  B.  Prognostic significance of vascular and perineural invasion in cancer of the larynx.  Am J Otolaryngol. 1998;19(2):83-88.PubMedGoogle ScholarCrossref
16.
Mesolella  M, Iorio  B, Misso  G,  et al.  Role of perineural invasion as a prognostic factor in laryngeal cancer.  Oncol Lett. 2016;11(4):2595-2598.PubMedGoogle Scholar
17.
Amit  M, Eran  A, Billan  S,  et al.  Perineural spread in noncutaneous head and neck cancer: new insights into an old problem.  J Neurol Surg B Skull Base. 2016;77(2):86-95.PubMedGoogle ScholarCrossref
18.
Jones  HB, Sykes  A, Bayman  N,  et al.  The impact of lymphovascular invasion on survival in oral carcinoma.  Oral Oncol. 2009;45(1):10-15.PubMedGoogle ScholarCrossref
19.
Adel  M, Kao  H-K, Hsu  CL,  et al.  Evaluation of lymphatic and vascular invasion in relation to clinicopathological factors and treatment outcome in oral cavity squamous cell carcinoma.  Medicine (Baltimore). 2015;94(43):e1510.PubMedGoogle ScholarCrossref
20.
Sandu  K, Nisa  L, Monnier  P, Simon  C, Andrejevic-Blant  S, Bron  L.  Clinicobiological progression and prognosis of oral squamous cell carcinoma in relation to the tumor invasive front: impact on prognosis.  Acta Otolaryngol. 2014;134(4):416-424.PubMedGoogle ScholarCrossref
21.
Tai  SK, Li  WY, Chu  PY,  et al.  Risks and clinical implications of perineural invasion in T1-2 oral tongue squamous cell carcinoma.  Head Neck. 2012;34(7):994-1001.PubMedGoogle ScholarCrossref
22.
Chatzistefanou  I, Lubek  J, Markou  K, Ord  RA.  The role of perineural invasion in treatment decisions for oral cancer patients: A review of the literature.  J Craniomaxillofac Surg. 2017;45(6):821-825.PubMedGoogle ScholarCrossref
23.
Liao  CT, Chang  JTC, Wang  HM,  et al.  Does adjuvant radiation therapy improve outcomes in pT1-3N0 oral cavity cancer with tumor-free margins and perineural invasion?  Int J Radiat Oncol Biol Phys. 2008;71(2):371-376.PubMedGoogle ScholarCrossref
24.
Rath  TJ, Narayanan  S, Hughes  MA, Ferris  RL, Chiosea  SI, Branstetter  BF  IV.  Solid lymph nodes as an imaging biomarker for risk stratification in human papillomavirus-related oropharyngeal squamous cell carcinoma.  AJNR Am J Neuroradiol. 2017;38(7):1405-1410.PubMedGoogle ScholarCrossref
25.
Chenevert  J, Chiosea  S.  Incidence of human papillomavirus in oropharyngeal squamous cell carcinomas: now and 50 years ago.  Hum Pathol. 2012;43(1):17-22.PubMedGoogle ScholarCrossref
26.
Maxwell  JH, Ferris  RL, Gooding  W,  et al.  Extracapsular spread in head and neck carcinoma: impact of site and human papillomavirus status.  Cancer. 2013;119(18):3302-3308.PubMedGoogle ScholarCrossref
27.
Zenga  J, Haughey  BH, Jackson  RS,  et al.  Outcomes of surgically treated human papillomavirus-related oropharyngeal squamous cell carcinoma with N3 disease.  Laryngoscope. 2017;127(9):2033-2037.PubMedGoogle ScholarCrossref
28.
Mizumachi  T, Homma  A, Sakashita  T, Kano  S, Hatakeyama  H, Fukuda  S.  Confirmation of the eighth edition of the AJCC/UICC TNM staging system for HPV-mediated oropharyngeal cancer in Japan.  Int J Clin Oncol. 2017;22(4):682-689.PubMedGoogle ScholarCrossref
29.
Chen  T-C, Wang  C-P, Ko  JY,  et al.  The impact of perineural invasion and/or lymphovascular invasion on the survival of early-stage oral squamous cell carcinoma patients.  Ann Surg Oncol. 2013;20(7):2388-2395.PubMedGoogle ScholarCrossref
30.
Brandwein-Gensler  M, Teixeira  MS, Lewis  CM,  et al.  Oral squamous cell carcinoma: histologic risk assessment, but not margin status, is strongly predictive of local disease-free and overall survival.  Am J Surg Pathol. 2005;29(2):167-178.PubMedGoogle ScholarCrossref
31.
Tassone  P, Crawley  M, Bovenzi  C,  et al.  Pathologic markers in surgically treated hpv-associated oropharyngeal cancer: retrospective study, systematic review, and meta-analysis.  Ann Otol Rhinol Laryngol. 2017;126(5):365-374.PubMedGoogle ScholarCrossref
32.
Johnson  JT, Barnes  EL, Myers  EN, Schramm  VL  Jr, Borochovitz  D, Sigler  BA.  The extracapsular spread of tumors in cervical node metastasis.  Arch Otolaryngol. 1981;107(12):725-729.PubMedGoogle ScholarCrossref
33.
Sinha  P, Lewis  JS  Jr, Piccirillo  JF, Kallogjeri  D, Haughey  BH.  Extracapsular spread and adjuvant therapy in human papillomavirus-related, p16-positive oropharyngeal carcinoma.  Cancer. 2012;118(14):3519-3530.PubMedGoogle ScholarCrossref
34.
Sinha  P, Kallogjeri  D, Gay  H,  et al.  High metastatic node number, not extracapsular spread or N-classification is a node-related prognosticator in transorally-resected, neck-dissected p16-positive oropharynx cancer.  Oral Oncol. 2015;51(5):514-520.PubMedGoogle ScholarCrossref
35.
Haughey  BH, Sinha  P.  Prognostic factors and survival unique to surgically treated p16+ oropharyngeal cancer.  Laryngoscope. 2012;122(suppl 2):S13-S33.PubMedGoogle ScholarCrossref
Original Investigation
From the American Head and Neck Society
December 2017

Defining the Prevalence and Prognostic Value of Perineural Invasion and Angiolymphatic Invasion in Human Papillomavirus–Positive Oropharyngeal Carcinoma

Author Affiliations
  • 1Department of Otolaryngology–Head and Neck Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
  • 2Department of Otolaryngology, School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania
  • 3Biostatistics Facility at the University of Pittsburgh Medical Center Hillman Cancer Center, Pittsburgh, Pennsylvania
  • 4Veterans Affairs Pittsburgh Health System, Pittsburgh, Pennsylvania
JAMA Otolaryngol Head Neck Surg. 2017;143(12):1236-1243. doi:10.1001/jamaoto.2017.2019
Key Points

Question  Does the presence of perineural invasion or angiolymphatic invasion provide prognostic information in human papillomavirus–positive oropharyngeal squamous cell carcinoma?

Findings  The presence of angiolymphatic invasion or perineural invasion denotes a worse prognosis in this patient population.

Meaning  Attention should be paid to angiolymphatic invasion and perineural invasion in human papillomavirus–positive oropharyngeal squamous cell carcinoma, particularly in ongoing deescalation trials.

Abstract

Importance  Recently, the American Joint Committee on Cancer (AJCC) updated its staging system for human papillomavirus (HPV)–positive oropharyngeal squamous cell carcinoma (OPSCC). The prognostic significance of perineural invasion (PNI) and angiolymphatic invasion (ALI) within this staging system is unknown.

Objective  To examine the prevalence and prognostic significance of PNI and ALI in HPV-positive OPSCC.

Design, Setting, and Participants  A retrospective review was performed of all patients with HPV-positive OPSCC treated surgically at the University of Pittsburgh Medical Center from January 1, 1980, through December 31, 2015, with at least 1 year of follow-up or death within 1 year.

Interventions  Surgical treatment of HPV-positive OPSCC.

Main Outcomes and Measures  The prevalence of PNI and ALI was determined from review of pathologic data, and Kaplan-Meier curves were generated for overall survival and disease-free survival when stratified by the presence of PNI and ALI. Multivariate analysis was performed using a Cox proportional hazards regression model.

Results  A total of 201 patients met the inclusion criteria (mean [SD] age, 57.4 [9.0] years; 79.6% [3.0%] male, and 20.4% [3.0%] female). Perineural invasion was identified in 32 of 201 primary specimens (15.9%), whereas ALI was identified in 74 of 201 primary specimens (36.8%). Both were significantly associated with increasing T stage. On multivariate analysis, the presence of at least 1 risk factor was significantly associated with overall and disease-free survival (overall hazard ratio, 2.78; 95% CI, 1.15-6.76; disease-free survival hazard ratio, 3.10; 95% CI, 1.17-8.23). Among patients classified as having stage II disease according to the eighth edition of the AJCC manual, the presence of at least 1 risk factor was associated with worse overall survival (hazard ratio, 11.7; 95% CI, 1.2-111.7).

Conclusions and Relevance  Both PNI and ALI were commonly found in HPV-positive OPSCC, with increasing prevalence as T stage increased. The presence of at least 1 risk factor was associated with worse overall and disease-free survival. Specifically, among patients classified as having stage II disease according to the eighth edition of the AJCC manual, the presence of ALI or PNI may suggest a poorer prognosis.

Introduction

The incidence of oropharyngeal squamous cell carcinoma (OPSCC) has increased during the past 3 decades, and an increasing proportion of OPSCC cases are human papillomavirus (HPV) positive.1 Compared with traditional, HPV-negative OPSCC, HPV-positive OPSCC is a unique clinical entity that affects a younger population of patients who often lack the traditional risk factors of heavy tobacco and alcohol exposure.2 In addition, HPV-positive OPSCC has an overall better prognosis than HPV-negative OPSCC, a distinction with important implications for disease management.3 Identifying the prognostic power of pertinent clinical and pathologic variables is at the forefront of clinical research, and treatment deescalation strategies are being applied to some HPV-positive patients to minimize treatment-related morbidity.4

TNM staging is an important tool for prognostication and treatment guidance, but several studies5,6 have found that the seventh edition of the American Joint Committee on Cancer (AJCC) pathologic staging system lacks prognostic power when applied to HPV-positive OPSCC. The eighth edition of the AJCC pathologic staging system (AJCC-8) was published in late 2016 and is scheduled to be implemented on January 1, 2018. It features separate staging criteria for HPV-positive patients and incorporates separate pathologic staging that acknowledges the increasing role that surgery is playing in HPV-positive OPSCC.7 Development of this new staging system was motivated by data from the Washington University School of Medicine, but information on its efficacy when used clinically is limited.8 This study therefore included an analysis of this new staging criteria, applying it to a new, independent cohort of patients.

Despite being a widely used prognostic indicator, TNM stage does not address certain important pathologic variables. Because of the increase in surgical treatment of HPV-positive OPSCC, this limitation raises questions about their clinical importance. In the head and neck cancer literature, perineural invasion (PNI) and angiolymphatic invasion (ALI) have generally been described as poor prognostic factors.9-12 Perineural invasion influences recurrence, distant metastases, and survival in several head and neck sites, including the oropharynx, oral cavity, hypopharynx, and larynx.13-17 Angiolymphatic invasion has received less attention than PNI, likely because it is difficult to recognize and has subsequently been omitted from some histologic grading systems.9,18 Studies18-21 have found an association between ALI and worse treatment outcomes; however, overall the literature is mixed, and additional studies to elucidate the significance of this association are indicated. Although the literature describes a more conclusive role for PNI in prognosis, supplementary data could contribute to discussions about the use of PNI (and perhaps ALI) to guide postsurgical treatment.17,22,23

Much of the literature on PNI and ALI in head and neck cancer makes no distinction between HPV-positive and HPV-negative disease. In addition, the prevalence of these risk factors in HPV-positive OPSCC is poorly defined. The purpose of this study was to examine the prevalence and prognostic value of PNI and ALI in HPV-positive OPSCC using this same cohort to test the external validity of the newly released AJCC-8.

Methods

A retrospective review was performed of all HPV-positive patients with OPSCC treated surgically at the University of Pittsburgh Medical Center from January 1, 1980, through December 31, 2015, with at least 1 year of follow-up or death within 1 year. Surgery was defined as resection of the primary tumor and ipsilateral or bilateral neck dissection. Tumors were considered to be HPV positive if immunostaining of the tumor was positive for HPV or p16. All patients had HPV testing by p16 immunohistochemical analysis or HPV in situ hybridization. Medical records of patients were reviewed for presenting demographic factors, surgical and adjuvant treatment, presence of PNI or ALI, and overall and disease-free survival. Approval was obtained for this retrospective clinical review from the University of Pittsburgh Institutional Review Board, which deemed that informed consent was not required. Data were de-identified for statistical analysis.

The presence of PNI or ALI was judged by review of the original pathology report from the primary tumor. Head and neck subspecialty pathologists who conform to institutional standards reviewed all specimens. The specimens were periodically re-reviewed for quality control and previous publications but were not specifically re-reviewed for the purposes of this study.24-26 If the presence or absence of PNI was not mentioned in the pathology report, the report result was considered to be negative according to institutional standards. If the presence or absence of ALI was not mentioned in the pathology report and patients had pN0 disease, the report results were considered to be negative according to institutional standards. If the presence or absence of ALI was not mentioned in the pathology report and patients had pN+ disease, the report results were considered to be indeterminate according to institutional standards. For the purposes of this study, the patients with indeterminate results were also considered to not have ALI. The prevalence of PNI and ALI was determined on the basis of these definitions and was then stratified by T stage, number of positive lymph nodes (≥5), and AJCC-8 pathologic stage. The demographic and clinical characteristics of sex, age, race, T stage, N stage, AJCC-8 pathologic stage, positive margin status, extracapsular spread (ECS), smoking history (categorical variable stratified by 10–pack-year history), radiation dose, and chemotherapy were summarized for patients with and without PNI and ALI. Differences in proportions and 95% CIs were reported for binary variables. Odds ratios (ORs) and 95% CIs were estimated for categorical variables using logistic regression. The Cohen d values and 95% CIs were reported for age; Hodges-Lehmann estimator of location shift and 95% CIs were reported for radiation dose.

We first validated the AJCC-8 pathologic staging system in the patient cohort in our study; none of the participants were used for initial creation of this staging system. This pathologic staging system was validated with generation of a Kaplan-Meier estimate of the overall survival functions and compared among AJCC-8 stages I through IV using Cox proportional hazards regression models.

The primary outcome measure was overall survival, which was defined as the time from the date of surgery to the date of death. Patients who were alive at the last follow-up were censored observations. The secondary outcome measure was disease-free survival, which was defined as the time from the date of surgery to the date of first recurrence or death (if any). Patients alive at the time of last follow-up and those who died of other causes without recurrence of disease were censored observations. Overall and disease-free survival curves stratified by the presence of ALI, PNI, or both risk factors were generated using the Kaplan-Meier method.

Cox proportional hazards regression models were used to estimate the hazard ratios (HRs) of overall and disease-free survival among patients with PNI or ALI and the presence or absence of either risk factor. Additional covariates included in the model were age at diagnosis, AJCC-8 surgical stage, ECS, smoking history, radiation dose, and the presence of positive margins. Variables were chosen if they were significant on univariate analysis (P < .05, 2-sided) or are commonly known to be associated with survival in this patient population.

Finally, we assessed within each AJCC-8 stage whether PNI and/or ALI was associated with overall survival using a log-rank test. On the basis of these conclusions, an adjusted staging variable that included information about the presence of PNI and/or ALI for patients with AJCC-8 stage II cancer was developed and compared with the basic staging variable using the Akaike information criterion (AIC) in Cox proportional hazards regression models.

Results

A total of 201 surgically treated HPV-positive patients with OPSCC and at least 1 year of follow-up were identified (mean [SD] age, 57.4 [9.0] years; 79.6% [3.0%] male, and 20.4% [3.0%] female). Baseline demographic information for these patients are given in Table 1. Within this patient cohort, the AJCC-8 pathologic staging system appropriately stratified patients by survival outcome. Overall survival was worse among patients with later-stage compared with earlier-stage disease: stage I (n = 152), HR, 1 (reference); stage II (n = 27), HR, 3.5 (95% CI, 1.7-7.1); stage III (n = 18), HR, 2.9 (95% CI, 1.3-6.5); and stage IV (n = 3), HR, 22.4 (95% CI, 6.4-78.2).

Perineural invasion was identified in 32 of 201 primary specimens (15.9%). The prevalence of PNI increased as T stage increased: T1, 12 of 93 (12.9%; OR, 1 [reference]); T2, 12 of 87 (13.8%; OR, 1.08; 95% CI, 0.46-2.55); T3, 4 of 16 (25.0%; OR, 2.25; 95% CI, 0.62-8.12); and T4, 4 of 5 (80.0%; OR, 27.00; 95% CI, 2.78-262.28). The prevalence of PNI also increased as AJCC-8 pathologic stage increased: stage I, 18 of 152 (12%; OR, 1); stage II, 6 of 27 (22%; OR, 2.13; 95% CI, 0.76-5.97); stage III, 6 of 18 (33%; OR, 3.72; 95% CI, 1.24-11.14); and stage IV, 2 of 3 (67%; OR, 14.89; 95% CI, 1.28-172.59) (Table 2). Angiolymphatic invasion was identified in 74 of 201 primary specimens (36.8%). The prevalence of ALI also generally increased as T stage increased: T1, 34 of 93 (36.6%; OR, 1 [reference]); T2, 28 of 87 (32.2%; OR, 0.82; 95% CI, 0.44-1.53); T3, 7 of 16 (43.8%; OR, 1.35; 95% CI, 0.46-3.95); and T4, 5 of 5 (100%; OR, >999.99; 95% CI, 0->999.99). There was no difference in lymph node status when stratified by ALI or PNI. The median radiation dose among patients with ALI was higher than that among those patients without ALI (location shift, 300 cGy; 95% CI, 0-600 cGy), but there was no difference in median radiation dose among patients when stratified by PNI. Similarly, patients with ALI were more likely to receive chemotherapy than were those without ALI (difference, 25.1%; 95% CI, 10.6%-38.8%), but there was no difference in chemotherapy rates when stratified by PNI status. Patients who tested ALI positive had a higher rate of ECS than did ALI-negative patients (difference, 21.9%; 95% CI, 6.4%-36.5%), which likely explains the higher rates of chemotherapy and radiotherapy in this cohort.

On univariate analysis, the presence of PNI inferred a worse overall survival (HR, 2.26; 95% CI, 1.13-4.51) and disease-free survival (HR, 2.66; 95% CI, 1.26-5.60). However, after controlling for the effects of age, smoking, AJCC-8 stage, radiation dose, ECS, and positive margins, there is no longer a statistically significant association between overall survival and PNI (HR, 1.46; 95% CI, 0.53-4.06) or a statistically significant association between disease-free survival and PNI (HR, 1.20; 95% CI, 0.39-3.74) (eTable in the Supplement).

On univariate analysis, the presence of ALI was associated with worse overall survival (HR, 2.20; 95% CI, 1.15-4.18) and disease-free survival (HR, 3.55; 95% CI, 1.76-7.17). After controlling for the effects of age, smoking, AJCC-8 stage, ECS, radiation dose, and positive margins, the presence of ALI was still associated with decreased overall survival (HR, 2.54; 95% CI, 1.01-6.41) and disease-free survival (HR, 2.74; 95% CI, 1.07-6.98) (eTable in the Supplement). Despite sex being different between the 2 groups, it was not associated with survival on univariate analysis and thus was not included in the multivariate analysis.

A significant correlation was found between the presence or absence of PNI and ALI (ρ = 0.23; 95% CI, 0.09-0.37). We therefore stratified patients into having 0 or at least 1 risk factor to be able to include both in our model. This decision was made after testing ALI and PNI alone, number of risk factors (0, 1, or 2), or the presence of at least 1 risk factor that provided the best fit based on AIC comparisons. A total of 115 of 201 patients (57.2%) had no risk factor, and 86 of 201 (42.8%) had at least 1 risk factor. When combining the presence of ALI and/or PNI into 1 risk factor, a significant association was found between number of risk factors and overall survival (Figure 1). This association persisted on multivariate analysis controlling for age, smoking, AJCC-8 stage, ECS, radiation dose, and positive margins (HR, 2.78; 95% CI, 1.15-6.76). Similarly, in disease-free survival (Figure 2), the significant association on univariate analysis remained on multivariate analysis (HR, 3.10; 95% CI, 1.17-8.23) (eTable in the Supplement).

The addition of PNI, ALI, or the presence of at least 1 risk factor to the AJCC-8 pathologic staging was not associated with improved overall survival. In the AJCC-8 stage II group, however, patients with ALI had significantly worse overall survival (HR, 3.40; 95% CI, 0.94-12.25). Similarly, in the stage II group, patients with PNI had significantly worse overall survival (HR, 4.38; 95% CI, 1.17-16.50). Among the patients with AJCC-8 stage II cancer, the HR for those with ALI and/or PNI was 11.7 (95% CI, 1.2-111.7) after controlling for the effects of ECS, positive margins, age at diagnosis, smoking history, and radiation dose (Figure 3). Therefore, a Cox proportional hazards regression model that includes the presence of at least 1 of PNI or ALI in a patient with AJCC-8 stage II cancer is a better model for overall survival than one without this information based on AIC comparisons.

Discussion

Human papillomavirus–positive OPSCC has been studied extensively in recent years, and these data have demonstrated that HPV-positive patients with OPSCC have better outcomes than do HPV-negative cohorts. Previous management practices for OPSCC, derived from clinical experiences with HPV-unrelated outcomes, do not pertain to this HPV-positive population, and a new edition of the AJCC pathologic staging system was released in 2016 that includes separate HPV-positive and HPV-negative OPSCC chapters.7 Several studies6,8 have reported that the previous edition of the AJCC staging manual (AJCC-7) could not reliably fulfill the prognostic needs of HPV-positive patients. Additional studies8,27 about HPV-positive patients with OPSCC found that the AJCC-7 pT stage was significantly associated with survival, but the nN stage was not. A recent study27 from the Washington University School of Medicine found that pN3 cases were equivalent to N1 cases and pN1 was associated with better survival than pN0. Compared with laterality or size of lymph nodes, the number of positive nodes was most important for prognosis.8

The AJCC-8 pathologic staging system used these data to create new criteria for pN stage; pN0 is used when there is no lymph node metastasis, pN1 includes metastasis in 4 or fewer lymph nodes, and pN2 includes metastasis in more than 4 lymph nodes.7 Because this revised system was newly introduced at the start of 2017, data about its usefulness are limited. Our study supports the external validity of the AJCC-8 pathologic staging system, demonstrating a statistically significant difference in survival among patients with stage I, stage II, and stage III disease. When comparing our data with similar results from a study conducted in Japan,28 the prognostic power and external validity of the AJCC-8 pathologic staging system seem to be acceptable and significantly improved from AJCC-7 staging.

In light of these changes reported by the AJCC, we sought to examine how pathologic variables that are not incorporated into TNM stage contribute to prognosis. Both PNI and ALI are generally considered to be risk factors for aggressive tumor behavior in head and neck cancer.9-12 However, results vary among studies,14,18,19,21 and thus it is difficult to draw conclusions about their significance and usefulness in guiding postoperative treatment. In a study14 of lesions in the oropharynx, oral cavity, hypopharynx, and larynx, the rate of local recurrence was 14% (range, 2%-26%) greater and mortality was 29% (range, 13%-43%) greater among patients with PNI compared with those without PNI. Another study29 found that neither variable was significantly associated with decreased survival in a cohort of patients with T1 and T2 squamous cell carcinoma. Nonetheless, the association between PNI and poorer outcomes has been demonstrated extensively in the literature, and the presence or absence of PNI is often considered when planning adjuvant treatment postoperatively.9-12,17,22,23 Despite its less conclusive association with outcomes, a previous study18 recognized that ALI should also be considered when planning adjuvant therapy and recommended future studies on its significance. However, another study30 advocated for a risk classification system in which patients are scored as low risk, intermediate risk, or high risk based on PNI, lymphocytic infiltrate at interface, and pattern of invasion. The risk score can then be used to plan postsurgical treatment.

The existing literature on PNI and ALI in head and neck cancer focuses on traditional, HPV-negative disease. Because HPV-positive OPSCC is a unique clinical entity, it is critical to explore specifically the roles of PNI and ALI in this patient population. This is especially true when considering that extracapsular extension, which has been considered as a poor prognostic factor in OPSCC, was not associated with poor outcomes in several HPV-positive cohorts.8,26,31,32

A recently published meta-analysis31 on the pathologic markers in surgically treated HPV-positive patients with OPSCC included only 4 studies that examined PNI and ALI, and these preliminary data reveal no significant negative effect on survival. This same study31 published prevalence data on PNI and ALI. The reported prevalence of PNI in the 4 studies31,33-35 included in the meta-analysis ranged from 11% to 19%, and the reported prevalence of ALI ranged from 15% to 50%. In our study, PNI prevalence was 16% and ALI prevalence was 37%, fitting with these existing data. In our study, PNI and ALI were significantly associated with T stage, and prevalence of PNI and ALI increased as T stage increased. In addition, ALI was significantly associated with AJCC-8 stage, the presence of ECS, and adjuvant therapy. Despite these associations, ALI remained significantly associated with overall and disease-free survival on multivariate analysis. In addition, in AJCC-8 pathologic stage II, the presence of at least 1 PNI or 1 ALI was significantly associated with shorter overall survival, suggesting that in this group (T0-2/N2 or T3-4/N0-1 HPV-positive patients with OPSCC treated surgically), PNI and ALI should be considered when adjuvant therapy is being considered and perhaps deescalation is not appropriate. It is unclear why stage II is the only stage in which ALI and PNI provide additional value beyond the existing staging classification. We believe that it is likely that the prognosis for patients with stage I disease is so good that we did not have the power to detect a difference in this study. Furthermore, few patients with stage III or IV disease are treated surgically; thus, larger studies are needed in these groups. However, in stage II, our study suggests that this additional prognostic information could be useful in clinical practice, and we recommend further evaluation by other groups.

Limitations

This study is limited by the retrospective nature of the data. Furthermore, there was selection bias in determining which patients can be treated surgically; thus, these results will apply to only appropriately selected surgical patients. Identification and characterization of PNI and ALI by different pathologists may vary and limit interpretation of results. Finally, there was no adjustment for comorbidity, which is a concern given the primary outcome of overall survival and recognizing competing causes of death among patients with head and neck cancer.

Conclusions

The AJCC-8 pathologic staging system for HPV-positive OPSCC was valid in this patient cohort. Both PNI and ALI are commonly found in HPV-positive OPSCC, with increasing prevalence as T stage and AJCC-8 pathologic stage increase. Angiolymphatic invasion remained prognostic of overall and disease-free survival on multivariate analysis when including AJCC-8 pathologic stage, but PNI did not provide additional prognostic information. However, specifically among patients with AJCC-8 stage II disease, the presence of ALI or PNI may suggest a poorer prognosis.

Back to top
Article Information

Corresponding Author: William G. Albergotti, MD, Department of Otolaryngology, University of Pittsburgh Medical Center, 200 Lothrop St, Ste 500, Pittsburgh, PA 15213 (albergot@musc.edu).

Accepted for Publication: August 12, 2017.

Published Online: October 26, 2017. doi:10.1001/jamaoto.2017.2019

Author Contributions: Dr Albergotti had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Albergotti, Schwarzbach, Ferris, Johnson, Kim.

Acquisition, analysis, or interpretation of data: Albergotti, Schwarzbach, Abberbock, Duvvuri.

Drafting of the manuscript: Albergotti, Schwarzbach, Ferris.

Critical revision of the manuscript for important intellectual content: Abberbock, Ferris, Johnson, Duvvuri, Kim.

Statistical analysis: Albergotti, Schwarzbach, Abberbock, Duvvuri.

Obtained funding: Abberbock.

Administrative, technical, or material support: Ferris.

Study supervision: Ferris, Johnson, Duvvuri, Kim.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.

Funding/Support: This project used the University of Pittsburgh Medical Center Hillman Cancer Center Biostatistics Facility, which is supported in part by award P30CA047904 from the National Institutes of Health (Dr Abberbock).

Role of the Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and the decision to submit the manuscript for publication.

Additional Contributions: Fei Ding, MS, University of Pittsburgh Cancer Institute Biostatistics Facility, Pittsburgh, Pennsylvania, assisted with statistical analysis and manuscript editing; he was not compensated for his work.

References
1.
Chaturvedi  AK, Engels  EA, Pfeiffer  RM,  et al.  Human papillomavirus and rising oropharyngeal cancer incidence in the United States.  J Clin Oncol. 2011;29(32):4294-4301.PubMedGoogle ScholarCrossref
2.
Gillison  ML.  Human papillomavirus-associated head and neck cancer is a distinct epidemiologic, clinical, and molecular entity.  Semin Oncol. 2004;31(6):744-754.PubMedGoogle ScholarCrossref
3.
Fischer  CA, Zlobec  I, Green  E,  et al.  Is the improved prognosis of p16 positive oropharyngeal squamous cell carcinoma dependent of the treatment modality?  Int J Cancer. 2010;126(5):1256-1262.PubMedGoogle Scholar
4.
Wierzbicka  M, Szyfter  K, Milecki  P, Składowski  K, Ramlau  R.  The rationale for HPV-related oropharyngeal cancer de-escalation treatment strategies.  Contemp Oncol (Pozn). 2015;19(4):313-322.PubMedGoogle Scholar
5.
Huang  SH, Xu  W, Waldron  J,  et al.  Refining American Joint Committee on Cancer/Union for International Cancer Control TNM stage and prognostic groups for human papillomavirus–related oropharyngeal carcinomas.  J Clin Oncol. 2015;33(8):836-845.PubMedGoogle ScholarCrossref
6.
Dahlstrom  KR, Calzada  G, Hanby  JD,  et al.  An evolution in demographics, treatment, and outcomes of oropharyngeal cancer at a major cancer center: a staging system in need of repair.  Cancer. 2013;119(1):81-89.PubMedGoogle ScholarCrossref
7.
American Joint Committee on Cancer.  American Joint Committee on Cancer (AJCC) Staging Manual. 8th ed. New York, NY: Springer; 2016.
8.
Haughey  BH, Sinha  P, Kallogjeri  D,  et al.  Pathology-based staging for HPV-positive squamous carcinoma of the oropharynx.  Oral Oncol. 2016;62:11-19.PubMedGoogle ScholarCrossref
9.
Woolgar  JA.  Histopathological prognosticators in oral and oropharyngeal squamous cell carcinoma.  Oral Oncol. 2006;42(3):229-239.PubMedGoogle ScholarCrossref
10.
Binmadi  NO, Basile  JR.  Perineural invasion in oral squamous cell carcinoma: a discussion of significance and review of the literature.  Oral Oncol. 2011;47(11):1005-1010.PubMedGoogle ScholarCrossref
11.
Massano  J, Regateiro  FS, Januário  G, Ferreira  A.  Oral squamous cell carcinoma: review of prognostic and predictive factors.  Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2006;102(1):67-76.PubMedGoogle ScholarCrossref
12.
Jardim  JF, Francisco  ALN, Gondak  R, Damascena  A, Kowalski  LP.  Prognostic impact of perineural invasion and lymphovascular invasion in advanced stage oral squamous cell carcinoma.  Int J Oral Maxillofac Surg. 2015;44(1):23-28.PubMedGoogle ScholarCrossref
13.
Rahima  B, Shingaki  S, Nagata  M, Saito  C.  Prognostic significance of perineural invasion in oral and oropharyngeal carcinoma.  Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2004;97(4):423-431.PubMedGoogle ScholarCrossref
14.
Fagan  JJ, Collins  B, Barnes  L, D’Amico  F, Myers  EN, Johnson  JT.  Perineural invasion in squamous cell carcinoma of the head and neck.  Arch Otolaryngol Head Neck Surg. 1998;124(6):637-640.PubMedGoogle ScholarCrossref
15.
Yilmaz  T, Hosal  AS, Gedikoğlu  G, Onerci  M, Gürsel  B.  Prognostic significance of vascular and perineural invasion in cancer of the larynx.  Am J Otolaryngol. 1998;19(2):83-88.PubMedGoogle ScholarCrossref
16.
Mesolella  M, Iorio  B, Misso  G,  et al.  Role of perineural invasion as a prognostic factor in laryngeal cancer.  Oncol Lett. 2016;11(4):2595-2598.PubMedGoogle Scholar
17.
Amit  M, Eran  A, Billan  S,  et al.  Perineural spread in noncutaneous head and neck cancer: new insights into an old problem.  J Neurol Surg B Skull Base. 2016;77(2):86-95.PubMedGoogle ScholarCrossref
18.
Jones  HB, Sykes  A, Bayman  N,  et al.  The impact of lymphovascular invasion on survival in oral carcinoma.  Oral Oncol. 2009;45(1):10-15.PubMedGoogle ScholarCrossref
19.
Adel  M, Kao  H-K, Hsu  CL,  et al.  Evaluation of lymphatic and vascular invasion in relation to clinicopathological factors and treatment outcome in oral cavity squamous cell carcinoma.  Medicine (Baltimore). 2015;94(43):e1510.PubMedGoogle ScholarCrossref
20.
Sandu  K, Nisa  L, Monnier  P, Simon  C, Andrejevic-Blant  S, Bron  L.  Clinicobiological progression and prognosis of oral squamous cell carcinoma in relation to the tumor invasive front: impact on prognosis.  Acta Otolaryngol. 2014;134(4):416-424.PubMedGoogle ScholarCrossref
21.
Tai  SK, Li  WY, Chu  PY,  et al.  Risks and clinical implications of perineural invasion in T1-2 oral tongue squamous cell carcinoma.  Head Neck. 2012;34(7):994-1001.PubMedGoogle ScholarCrossref
22.
Chatzistefanou  I, Lubek  J, Markou  K, Ord  RA.  The role of perineural invasion in treatment decisions for oral cancer patients: A review of the literature.  J Craniomaxillofac Surg. 2017;45(6):821-825.PubMedGoogle ScholarCrossref
23.
Liao  CT, Chang  JTC, Wang  HM,  et al.  Does adjuvant radiation therapy improve outcomes in pT1-3N0 oral cavity cancer with tumor-free margins and perineural invasion?  Int J Radiat Oncol Biol Phys. 2008;71(2):371-376.PubMedGoogle ScholarCrossref
24.
Rath  TJ, Narayanan  S, Hughes  MA, Ferris  RL, Chiosea  SI, Branstetter  BF  IV.  Solid lymph nodes as an imaging biomarker for risk stratification in human papillomavirus-related oropharyngeal squamous cell carcinoma.  AJNR Am J Neuroradiol. 2017;38(7):1405-1410.PubMedGoogle ScholarCrossref
25.
Chenevert  J, Chiosea  S.  Incidence of human papillomavirus in oropharyngeal squamous cell carcinomas: now and 50 years ago.  Hum Pathol. 2012;43(1):17-22.PubMedGoogle ScholarCrossref
26.
Maxwell  JH, Ferris  RL, Gooding  W,  et al.  Extracapsular spread in head and neck carcinoma: impact of site and human papillomavirus status.  Cancer. 2013;119(18):3302-3308.PubMedGoogle ScholarCrossref
27.
Zenga  J, Haughey  BH, Jackson  RS,  et al.  Outcomes of surgically treated human papillomavirus-related oropharyngeal squamous cell carcinoma with N3 disease.  Laryngoscope. 2017;127(9):2033-2037.PubMedGoogle ScholarCrossref
28.
Mizumachi  T, Homma  A, Sakashita  T, Kano  S, Hatakeyama  H, Fukuda  S.  Confirmation of the eighth edition of the AJCC/UICC TNM staging system for HPV-mediated oropharyngeal cancer in Japan.  Int J Clin Oncol. 2017;22(4):682-689.PubMedGoogle ScholarCrossref
29.
Chen  T-C, Wang  C-P, Ko  JY,  et al.  The impact of perineural invasion and/or lymphovascular invasion on the survival of early-stage oral squamous cell carcinoma patients.  Ann Surg Oncol. 2013;20(7):2388-2395.PubMedGoogle ScholarCrossref
30.
Brandwein-Gensler  M, Teixeira  MS, Lewis  CM,  et al.  Oral squamous cell carcinoma: histologic risk assessment, but not margin status, is strongly predictive of local disease-free and overall survival.  Am J Surg Pathol. 2005;29(2):167-178.PubMedGoogle ScholarCrossref
31.
Tassone  P, Crawley  M, Bovenzi  C,  et al.  Pathologic markers in surgically treated hpv-associated oropharyngeal cancer: retrospective study, systematic review, and meta-analysis.  Ann Otol Rhinol Laryngol. 2017;126(5):365-374.PubMedGoogle ScholarCrossref
32.
Johnson  JT, Barnes  EL, Myers  EN, Schramm  VL  Jr, Borochovitz  D, Sigler  BA.  The extracapsular spread of tumors in cervical node metastasis.  Arch Otolaryngol. 1981;107(12):725-729.PubMedGoogle ScholarCrossref
33.
Sinha  P, Lewis  JS  Jr, Piccirillo  JF, Kallogjeri  D, Haughey  BH.  Extracapsular spread and adjuvant therapy in human papillomavirus-related, p16-positive oropharyngeal carcinoma.  Cancer. 2012;118(14):3519-3530.PubMedGoogle ScholarCrossref
34.
Sinha  P, Kallogjeri  D, Gay  H,  et al.  High metastatic node number, not extracapsular spread or N-classification is a node-related prognosticator in transorally-resected, neck-dissected p16-positive oropharynx cancer.  Oral Oncol. 2015;51(5):514-520.PubMedGoogle ScholarCrossref
35.
Haughey  BH, Sinha  P.  Prognostic factors and survival unique to surgically treated p16+ oropharyngeal cancer.  Laryngoscope. 2012;122(suppl 2):S13-S33.PubMedGoogle ScholarCrossref
×