Introduction

In recent decades, the prevalence of oropharyngeal squamous cell carcinoma (OPSCC) has been on the rise globally, and this trend is still increasing in Taiwan and the United States.1–3 The presence of neck lymph node metastasis is one of the most significant prognostic factors in treating patients with OPSCC.4,5 The chance of nodal metastasis is high in every tumor stage among patients with tonsil cancer.6 In addition, the occult nodal metastasis rate of the ipsilateral side in OPSCC is approximately 30%, warranting the importance of neck treatment.7,8 Further studies have investigated the risk of developing contralateral nodal metastases (CNM) in patients treated with surgical approaches and have demonstrated a risk of CNM ranging from 16% to 29%.9,10 After the emergence of human papillomavirus (HPV) as the main etiologic agent in the development of OPSCC, Kato et al and Tritter et al have described the incidence of contralateral or bilateral nodal metastasis by clinical or imaging studies to be 14–26%.11,12

More recently, Last et al studied the incidence of CNM in surgically treated HPV-related tongue base cancer, and they reported the rate of CNM to be up to 38.2%.13 Currently, reports are still limited concerning the prevalence rate of pathological nodal positivity in contralateral neck cases among patients with HPV-negative OPSCC, and the incidence of occult contralateral nodal metastasis (OCNM) is not well known in this patient population. Clinically, the expression of p16 is recognized as a surrogate of HPV noted in the tumor cells of OPSCC;14 therefore, this study aimed to identify the prevalence rate and risk factors of OCNM in p16-negative OPSCC. In addition, we aimed to develop a nomogram that could help individually predict the probability of OCNM and thus attempt to tailor treatment options.

Materials and Methods Study Population

From the cancer database of the institute, we retrospectively enrolled patients diagnosed with p16-negative OPSCC who received radical surgery with bilateral neck dissection as primary treatment from January 2005 to December 2018 in Kaohsiung Chang Gung Memorial Hospital, Taiwan. The TNM stage was reclassified according to the eighth edition of the American Joint Committee on Cancer (AJCC) staging system.15 The expression of p16 by immunohistochemistry was also determined according to the suggestion from the same staging system. Only p16-negative OPSCC patients who had upfront radical surgery with bilateral neck dissections were included for the analysis of CNM. The clinical status of neck nodal metastasis was determined by contrast-enhanced image studies preoperatively, including computed tomography scan or magnetic resonance imaging examination. Patients with contralateral cN0 disease who underwent radical surgery with bilateral neck dissections were selected to identify the OCNM, and all neck dissections were performed simultaneously with primary tumor resection. OCNM was defined as pathologically contralateral positive nodal metastases after the surgical intervention. Patients who (a) had any other cancer history, (b) had prior treatment for head and neck cancer, (c) had neoadjuvant therapy, (d) had a non-oropharyngeal primary site, (e) had simultaneous multiple origins of head and neck cancer, and (f) had non-squamous cell carcinoma were excluded from this study. Finally, detailed information on the clinical and pathological features of the 85 patients were retrospectively reviewed.

Predictive Variables

The clinical and pathological characteristics such as gender, age, lifestyle habits, tumor location, laterality of primary tumor, clinical stage, histological grade, tumor thickness and clinical extranodal extension (cENE) in the ipsilateral neck were all statistically analyzed, with laterality of primary tumor being determined by physical examination and image studies. None of the tumors that crossed the midline were classified as lateralized. The cutoff values of tumor thickness were determined by the median values of two studied cohorts: 14 mm in the entire cohort and 12 mm in the contralateral cN0 cohort. The presence of cENE was determined by image studies of lymph nodes with indistinct margins or irregular capsular enhancement or infiltration into adjacent soft tissue. The adjuvant therapy for these patients after surgery was mostly based on the American National Comprehensive Cancer Network (NCCN) guideline.

Statistical Analysis

Statistical analyses were performed using the SPSS 22 software (SPSS/IBM Inc.). The outcomes of interest were the rate and potential risk factors of OCNM. Simple logistic regression analysis was performed to estimate the odds ratios (ORs) and 95% CIs of contralateral nodal disease for each covariate, while Fisher’s exact test was applied to compare the distribution of categorical characteristics between the groups of patients with ipsilateral and contralateral lymph nodes, and the Log rank test was used to compare survival experiences between groups. A two-sided p value <0.05 was considered significant.

In addition, a predictive nomogram was created incorporating histological grade, tumor lateralization, tumor thickness, and ipsilateral cENE using the R software “rms” package (Version 5.1–0, Vanderbilt University, Nashville, TN, USA) with the endpoints of the pathologic status of lymph nodes in the contralateral cN0 neck. To ascertain the OCNM prediction accuracy of the nomogram, the concordance index (C-index) was derived for the proposed nomogram models with and without ipsilateral cENE; C-index values of 0.5 and 1.0 were considered as signifying random and perfect predictability, respectively. In addition, the predictive model was internally validated using 1000 bootstrap techniques to address model overfit and obtain a relatively unbiased evaluation.

The calibration of this nomogram was assessed using 1000 bootstrap resamples to draw a plot of the predicted probabilities of OCNM against the actual probabilities of OCNM. This study was approved by the Medical Ethics and Human Clinical Trial Committees at Chang Gung Memorial Hospital (Ethical Application Reference number: 202201008B0). This study was conducted in accordance with the Declaration of Helsinki. Patients’ consent to review their medical records was not required by this hospital’s committees because the patient data remained anonymous in this study.

Results

Of the 85 patients, 81 patients (95.3%) were male, and 4 (4.7%) were female. The mean age during diagnosis for all was 53.6 years, ranging from 31 to 71 years; 79 patients (92.9%) had a smoking habit, 66 patients (77.6%) had the habit of alcohol drinking, and 64 patients (75.3%) had the habit of betel nut chewing. The most common tumor subsite was soft palate (n = 39, 45.9%), followed by base of the tongue (n = 36, 42.3%) and tonsil (n = 10, 11.8%). For histologic grade, 18 patients (21.2%) had well differentiated squamous cell carcinoma (WDSCC), 62 patients (72.9%) had moderately differentiated squamous cell carcinoma, and the remaining 5 patients (5.9%) had poorly differentiated squamous cell carcinoma. Sixty primary tumors (70.6%) were found across the midline. For clinical T classification, this cohort included T1 (n = 14, 16.5%), T2 (n = 27, 31.8%), T3 (n = 9, 10.6%), T4a (n = 28, 32.9%), and T4b (n = 7, 8.2%). The clinical status of neck nodal metastasis was present in 56 patients (82.5%), and 27 out of these 56 patients had clinical positive ENE noted by the image findings. The clinicopathological features of all 85 cases are listed in Table 1.

Table 1 Characteristics of Patients Who Had p16 Negative OPSCC Underwent Bilateral Neck Dissection (n = 85)

Overall, 26 patients (26/85 = 30.6%) had presence of contralateral positive neck nodes pathologically. Sixty patients (60/85 = 70.6%) had contralateral cN0 disease, and positive nodes were found in 11 necks after the neck dissections, so the rate of OCNM was 18.3% (11/60).

About 71% of the patients (n = 61) underwent adjuvant therapy after radical surgery with most of these (n = 47) undergoing adjuvant CCRT. During the follow-up period, recurrent tumors were noted in 26 patients (30.6%), including 10 patients (11.8%) who had local recurrence, 5 patients (5.9%) who had regional failure, 2 patients (2.4%) who had locoregional recurrence, and 10 patients (11.8%) who had distant metastasis disease. All the regional recurrences were located at the ipsilateral neck area, and no patient with contralateral nodal recurrence was noted in this cohort. At last contact, 37 patients had died, including 25 patients who had expired from the disease.

Contralateral Nodal Metastasis

The overall incidence of pathologically positive CNM was 26/85 (30.6%). Regarding these positive contralateral nodes, there were 3/10 (30%), 9/39 (23.1%), and 14/36 (38.9%) over-the-tumor locations in tonsil, soft palate and base of the tongue, respectively. The factors associated with CNM are described in Table 2. Compared with clinical ipsilateral negative nodal disease, clinical ipsilateral positive nodal status had a higher risk of contralateral nodal spread (OR, 4.044; 95% CI, 1.238–13.212). In addition, patients with ipsilateral cENE had an increased risk of contralateral nodal disease (OR, 5.662; 95% CI, 2.079–15.415). For the lateralization of the primary tumor, those with cross-midline tumors had a higher risk of CNM than those without them, but this did not reach statistical significance (OR, 3.039; 95% CI, 0.924–10.003, p = 0.067).

Table 2 Risk Factors for Contralateral Nodal Disease in All Patients (n = 85)

Occult Contralateral Nodal Metastasis

In this cohort, 25 patients had contralateral nodal disease clinically, but only 15 out of these 25 patients had metastatic lymph nodes with pathological proof. The other 60 patients had no contralateral nodal disease clinically, and OCNM was identified in 11 patients (18.3%) after surgical validation. The number of occult lymph nodes was identified—one in nine patients and two in two patients. Pathologic ENE was identified in two contralateral necks (2/11 = 18.2%) among these occult metastatic lymph nodes. The risk factors associated with OCNM are described in Table 3. Only the presence of ipsilateral cENE increased the risk of OCNM (odds ratio, 4.271; 95% CI, 1.045–17.458, p = 0.043).

Table 3 Risk Factors for Contralateral Nodal Disease in Those with Clinically Negative Contralateral Neck Findings (n = 60)

Nomogram for OCNM Prediction

A nomogram incorporating histological grade (WDSCC vs non-WDSCC), tumor lateralization (lateralized vs crossed midline), tumor thickness (<12 mm vs ≥12 mm), and ipsilateral cENE (present vs absent) was developed for the purpose of predicting individualized OCNM in the study population (Figure 1). The longer the line, the greater the impact of risk factors on the rate of OCNM. Each subtype in these variables was assigned a score. The cumulative sum of each “point” is the “total points.” The corresponding “risk of occult contralateral nodal metastasis” of the “total points” is the predicted probability of OCNM suggested by our designed nomogram. Another nomogram incorporating histological grade, tumor lateralization, and tumor thickness was implemented for comparison. The C-index derived for the nomogram incorporating several clinicopathological features without ipsilateral cENE was 0.623, whereas that derived for the nomogram model incorporating ipsilateral cENE and several clinicopathological features was 0.717. These results indicate that the nomogram incorporating clinicopathological factors including ipsilateral cENE had better performance in predicting the OCNM of patients with p16-negative OPSCC upfront radical surgery than the nomogram model with no ipsilateral cENE. The proposed predictive nomogram was internally validated using bootstrap techniques.

Figure 1 (A) Nomogram was constructed based on the data of clinical contralateral N0 group. The points of each feature were added to obtain the total points, and a vertical line was drawn on the total points to obtain the corresponding risk of OCNM. (B) Calibration curves of nomogram prediction of OCNM.

Abbreviations: OCNM, occult contralateral neck metastasis; WDSCC, well differentiated squamous cell carcinoma; cENE, clinical extranodal extension.

The example of nomogram usage in this cohort is shown as follows. A sample was randomly selected from the subjects. A patient who was diagnosed with p16-negative OPSCC with a pretreatment image study was revealed to be clinically negative in contralateral neck metastasis. Its tumor characteristics showed non-WDSCC in the histologic grade (score: 37.20), crossed midline of the primary tumor (score: 56.13), a tumor thickness ≥12 mm (score: 19.49), and the presence of ipsilateral cENE (score: 100). The total points added up to 212.82, and thus, the probability of OCNM was 46.19%.

According to our designed nomogram, if the scores of total points ≥122 corresponded to the probability of OCNM of more than 20% in the typical threshold, it would be necessary to perform elective neck dissection by surgeons for better clinical outcomes. A new risk stratification scheme based on nomogram scores was then proposed, which stratified p16-negative OPSCC patients into low- and high-risk groups of OCNM. Ipsilateral neck dissection was recommended for patients in the low-risk group, and bilateral neck dissection was recommended for patients in the high-risk group (Table 4).

Table 4 Metastatic Risk Stratification and Individualized Treatment of Contralateral cN0 Disease in p16 Negative OPSCC Patients Based on Nomogram Scores

Discussion

Nodal metastasis is the most important factor in the prognosis of HPV-negative OPSCC.16 Givens et al17 in a retrospective analysis showed that the five-year survival rate was 70% in N0, but this dramatically reduced to 27% in N1 or more advanced cases. Some authors have also shown that delayed neck metastasis from an untreated neck usually presents a higher incidence of extracapsular spread, multiple nodal metastasis, and the presence of skip metastases, reducing the salvage rates to less than 40% despite intensive treatment.18,19 This finding highlights the importance of treating neck lymph nodes at the initial diagnosis of OPSCC.

With a higher proportion of patients of OPSCC presenting with neck nodal disease at the time of diagnosis, it is essential to perform neck dissection when ipsilateral nodal metastasis is present. Clinically, the primary radiation strategy is the mainstay in treating OPSCC, but the detailed histopathologic information is lacking; additionally, treatment guidelines for clinical contralateral N0 neck, especially among patients with HPV-negative OPSCC, are insufficient except for the primary tumor crossing the midline.

In our cohort, the prevalence of positive CNM in patients with p16-negative OPSCC was 30.6% after validation of curative surgery for primary site and bilateral neck dissections. Our results showed that the rate of pathologically positive CNM is higher than other investigations of OPSCC. The reason for these results might be due to more extensive tumor involvement in this cohort.

Before the era of HPV as the main etiology in the development of OPSCC, some authors had found that the chance of developing CNM was between 16% and 29% in patients having OPSCC treated by surgical approaches.9,10 Kato et al11 in a National Cancer Database study found patients with HPV-negative OPSCC had a 14.5% risk of contralateral nodal disease. In our cohort, the prevalence of positive CNM in patients with p16-negative OPSCC was 30.6% after validation of curative surgery for primary site and bilateral neck dissections. Larger and more extensive tumor involvement in this series could be the reason. In addition, our results revealed that the contralateral metastasis rate would be significantly higher than those without if ipsilateral positive nodal disease, the presence of ENE and a higher number of involved ipsilateral nodes were noted pathologically. This correlation has been previously observed among patients with hypopharyngeal cancer. Basically, elective bilateral neck dissections are recommended for surgical management from these reports.20,21 Nowadays, the TNM stage of HPV-negative OPSCC is classified together with hypopharyngeal cancer as a group according to the eighth edition of the AJCC system.

Some previous studies have described that a crossed-midline tumor and the tumor location over the tongue base in OPSCC possess increasing risk of CNM.11,13 In our study, a crossed-midline tumor also showed a higher risk of CNM but did not reach statistical significance (p = 0.06), possibly due to the limited case number in this cohort. The primary tumor location might also play a role because a tumor over the tongue base has relatively less distance from the midline than primary tonsil cancer, but this did not reach significance and might also be due to limited study numbers in our series.

To this day, few reports have focused on OCNM in HPV-negative OPSCC. Lim et al22 have demonstrated a 16% rate of OCNM in 43 patients with clinical contralateral N0 tonsillar cancer. This study might include both HPV-positive or HPV-negative tumors because the HPV status is unknown. McMullen et al23 studied patients with pT1-2 N0-3 (AJCC, eighth edition) OPSCC disease treated with transoral robotic surgery and bilateral neck dissection, including 24 patients with p16-positive tumors and 8 patients with p16-negative tumors, where the OCNM rate was 12.5% (1/8) in p16-negative OPSCC. To the best of our knowledge, this is a large series of investigations about tissue proof by neck dissection in contralateral occult nodal metastases among patients with p16-negative OPSCC. The proportion of OCNM was up to 18.3% (11/60) in this cohort, also revealing that the presence of ipsilateral cENE produced significantly higher risk of OCNM. Our study also established a multivariate nomogram that integrated clinicopathological variables, including histologic grade, tumor lateralization, tumor thickness and ipsilateral cENE, which could yield a feasible result (C-index: 0.717). The nomogram could provide more accurate and individualized predictions of OCNM of patients with p16-negative OPSCC than the nomogram model without ipsilateral cENE being incorporated. It could aid surgeons in the identification of patients who could achieve favorable results from more neck management.

Several limitations should be addressed in this study. Firstly, this is a retrospective, single‐institute study. Head and neck surgeons might have different treatment opinions in different historical backgrounds, and selection bias in patients and data collection could occur. Secondly, the incidence of contralateral nodal disease could be confined to those patients undergoing surgery in this cohort. The contralateral neck as treated by elective radiotherapy was not available in this cohort, thereby not precisely reflecting the prevalence rate of CNM. Thirdly, the sample size is limited, and finally, the study results and proposed nomogram were not validated using an independent data set. External validation of the results using an independent patient cohort could strengthen the derived evidence revealing that ipsilateral cENE is a good predictor of OCNM in p16-negative OPSCC.

Conclusion

In this study, the incidence of contralateral nodal disease among patients with p16-negative OPSCC was 30.6%, and the chance of OCNM was 18.3%. The presence of ipsilateral cENE denoted significantly higher risk of CNM and OCNM. The ipsilateral cENE–based nomogram developed in this study could be useful in the application of ipsilateral cENE and provide individualized prediction of OCNM for patients with p16-negative OPSCC. Furthermore, larger cohort studies are necessary to validate this issue.

Disclosure

The authors report no conflicts of interest in this work.

References

1. Hsu WL, Yu KJ, Chiang CJ, Chen TC, Wang CP. Head and neck cancer incidence trends in Taiwan, 1980 – 2014. Int J Head Neck Sci. 2017;1:180–189.

2. Wang CP, Chen TC, Hsu WL, et al. Rising incidence of HPV positive oropharyngeal cancer in Taiwan between 1999 and 2014 where betel nut chewing is common. BMC Cancer. 2022;22(1):296. doi:10.1186/s12885-022-09407-5

3. Van Dyne EA, Henley SJ, Saraiya M, Thomas CC, Markowitz LE, Benard VB. Trends in human papillomavirus-associated cancers - United States, 1999–2015. MMWR Morb Mortal Wkly Rep. 2018;67(33):918–924. doi:10.15585/mmwr.mm6733a2

4. Doescher J, Veit JA, Hoffmann TK. Die 8. Ausgabe der TNM‑Klassifikation: Neuerungen für das Fachgebiet Hals‑Nasen‑Ohren‑Heilkunde, Kopf- und Halschirurgie [The 8th edition of the AJCC Cancer Staging Manual: updates in otorhinolaryngology, head and neck surgery]. HNO. 2017;65(12):956–961. German. doi:10.1007/s00106-017-0391-3

5. Tsai MH, Lin YT, Chuang HC, et al. Prognostic value of pathologically positive nodal number in p16-negative oropharyngeal and hypopharyngeal squamous Cell carcinoma with pN3b status. Diagnostics. 2022;12(6):1443. doi:10.3390/diagnostics12061443

6. Chuang HC, Fang FM, Huang CC, et al. Clinical and pathological determinants in tonsillar cancer. Head Neck. 2011;33(12):1703–1707. doi:10.1002/hed.21662

7. Pitman KT, Johnson JT, Myers EN. Effectiveness of selective neck dissection for management of the clinically negative neck. Arch Otolaryngol Head Neck Surg. 1997;123(9):917–922. doi:10.1001/archotol.1997.01900090023004

8. Lim YC, Koo BS, Lee JS, Lim JY, Choi EC. Distributions of cervical lymph node metastases in oropharyngeal carcinoma: therapeutic implications for the N0 neck. Laryngoscope. 2006;116(7):1148–1152. doi:10.1097/01.mlg.0000217543.40027.1d

9. Chung EJ, Oh JI, Choi KY, et al. Pattern of cervical lymph node metastasis in tonsil cancer: predictive factor analysis of contralateral and retropharyngeal lymph node metastasis. Oral Oncol. 2011;47(8):758–762. doi:10.1016/j.oraloncology.2011.05.013

10. Olzowy B, Tsalemchuk Y, Schotten KJ, Reichel O, Harreus U. Frequency of bilateral cervical metastases in oropharyngeal squamous cell carcinoma: a retrospective analysis of 352 cases after bilateral neck dissection. Head Neck. 2011;33(2):239–243. doi:10.1002/hed.21436

11. Kato MG, Ellis MA, Nguyen SA, Day TA. Predictors of contralateral-bilateral nodal disease in oropharyngeal cancer: a National Cancer Data Base Study. Head Neck. 2018;40(2):338–348. doi:10.1002/hed.24964

12. Tritter AG, Mehta V, Samuelson M, et al. Incidence of contralateral-bilateral nodes in the human papillomavirus era. Laryngoscope. 2017;127(6):1328–1333. doi:10.1002/lary.26439

13. Last AS, Pipkorn P, Chen S, et al. Risk and rate of occult contralateral nodal disease in surgically treated patients with human papillomavirus-related squamous cell carcinoma of the base of the tongue. JAMA Otolaryngol Head Neck Surg. 2019:3:13.

14. Jordan RC, Lingen MW, 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(7):945–954. doi:10.1097/PAS.0b013e318253a2d1

15. Amin MB, Greene FL, Edge SB, et al. The Eighth Edition AJCC Cancer Staging Manual: continuing to build a bridge from a population-based to a more “personalized” approach to cancer staging. CA Cancer J Clin. 2017;67(2):93–99. doi:10.3322/caac.21388

16. Mantravadi RV, Liebner EJ, Ginde JV. An analysis of factors in the successful management of cancer of tonsillar region. Cancer. 1978;41(3):1054–1058. doi:10.1002/1097-0142(197803)41:3<1054::AID-CNCR2820410338>3.0.CO;2-4

17. Givens CD Jr., Johns ME, Cantrell RW. Carcinoma of the tonsil. Analysis of 162 cases. Arch Otolaryngol. 1981;107(12):730–734. doi:10.1001/archotol.1981.00790480006002

18. Ogrady M, Doyle PJ, Flores AD. Cancer of the tonsil. J Otolaryngol. 1985;14(4):221–225.

19. Woolgar JA. Salvage neck dissections in oral and oropharyngeal squamous cell carcinoma: histological features in relation to disease category. Int J Oral Maxillofac Surg. 2006;35(10):907–912. doi:10.1016/j.ijom.2006.08.003

20. Kowalski LP, Santos CR, Magrin J, Scopel A. Factors influencing contralateral metastasis and prognosis from pyriform sinus carcinoma. Am J Surg. 1995;170(5):440–445. doi:10.1016/S0002-9610(99)80325-0

21. Koo BS, Lim YC, Lee JS, Kim YH, Kim SH, Choi EC. Management of contralateral N0 neck in pyriform sinus carcinoma. Laryngoscope. 2006;116(7):1268–1272. doi:10.1097/01.mlg.0000225936.88411.71

22. Lim YC, Lee SY, Lim JY, et al. Management of contralateral N0 neck in tonsillar squamous cell carcinoma. Laryngoscope. 2005;115(9):1672–1675. doi:10.1097/01.mlg.0000184791.68804.0b

23. McMullen CP, Garneau J, Weimar E, et al. Occult contralateral nodal disease in oropharyngeal squamous cell carcinoma patients undergoing primary TORS with bilateral neck dissection. Oral Oncol. 2019;93:96–100. doi:10.1016/j.oraloncology.2019.04.017