Outcomes of Patients With Single-Node Metastasis of Human Papillomavirus–Related Oropharyngeal Cancer Treated With Transoral Surgery

Key Points

Question  What is the prognostic association of single–lymph node metastasis with regional failure and survival in human papillomavirus–related oropharyngeal squamous cell carcinoma?

Findings  In this multi-institutional cohort study of 207 adults undergoing surgical resection with or without adjuvant therapy, 5% experienced regional recurrence with 70% successfully undergoing salvage treatment. The 5-year overall survival was 95% in the entire cohort.

Meaning  This study suggests that patients undergoing surgical resection for single-node metastasis in human papillomavirus–related oropharyngeal squamous cell carcinoma most commonly experience regional failure, particularly in the absence of adjuvant therapy, and maintain excellent survival.

Importance  Regional lymph node metastasis remains an important prognostic factor in patients with oropharyngeal squamous cell carcinoma (OPSCC). Although survival among patients with regional metastasis in human papillomavirus (HPV)–related OPSCC is more favorable compared with patients who are HPV negative, prognostic variables associated with failure in patients with single-node metastasis are not known.

Objective  To evaluate recurrence and survival in patients with HPV-related OPSCC with single–lymph node metastasis treated with transoral surgery.

Design, Setting, and Participants  A retrospective cohort study was conducted of 207 adults with newly diagnosed p16-positive OPSCC and pathology-confirmed single-node disease who underwent surgical resection with or without adjuvant therapy at 2 tertiary academic medical centers from January 1, 2007, to December 31, 2016. Statistical analysis was performed from September 1, 2018, to September 1, 2020.

Interventions  Surgery alone (n = 59), surgery with adjuvant radiation (n = 75), or surgery with adjuvant chemoradiation (n = 73).

Main Outcomes and Measures  The primary outcome was regional recurrence. Secondary outcomes included overall survival, any recurrence, and identification of factors associated with regional recurrence and overall survival.

Results  Among 207 patients, 178 (86%) were men, with a median age of 57 years (range, 35-82 years) at the time of surgery. Median follow-up was 36.2 months (range, 7-127 months). Regional recurrence occurred in 11 patients (5%). Of these, 1 patient (9%) was lost to follow-up after diagnosis, 1 (9%) was treated with palliative chemotherapy, and 9 (82%) were treated with curative intent. Ultimately, 7 patients received successful salvage treatment, and 3 died with disease. Overall, there were 21 patients (10%) with any recurrence, with 4 patients (19%) experiencing local recurrence, 11 (52%) experiencing regional recurrence, and 6 (29%) experiencing distant metastasis. The 5-year overall survival was 95% (95% CI, 89%-98%) for all patients. Older age (odds ratio [OR], 1.2; 95% CI, 1.1-1.2), advanced T stage (OR, 3.5; 95% CI, 0.9-14.0), and positive margins (OR, 10.9; 95% CI, 1.8-67.5) were associated with increased regional recurrence. Extranodal extension (OR, 0.2; 95% CI, 0.04-0.8), lymph node size greater than 3 cm (OR, 0.2; 95% CI, 0.1-0.7), and adjuvant therapy (OR, 0.08; 95% CI, 0.02-0.4) were associated with decreased regional recurrence. Advanced comorbidities (hazard ratio, 6.20; 95% CI, 1.4-27.7), lymphovascular invasion (hazard ratio, 4.7; 95% CI, 1.0-21.2), and regional recurrence (hazard ratio, 16.0; 95% CI, 3.1-82.0) were associated with worse overall survival.

Conclusions and Relevance  The findings of this cohort study suggest that patients with HPV-related OPSCC and single-node disease undergoing surgical resection with or without adjuvant treatment have excellent survival. Adjuvant therapy appears to improve regional control. Among patients with regional recurrence of OPSCC, there is a high rate of successful salvage treatment.

Human papillomavirus (HPV)-related oropharyngeal squamous cell carcinoma (OPSCC) is a growing clinical entity, with an estimated 85 000 newly diagnosed oropharyngeal carcinomas occurring per year worldwide.¹ Although HPV-associated OPSCC is associated with a higher rate of lymph node (LN) metastases, it generally has an improved prognosis compared with HPV-negative OPSCC.^2-4 The recently updated American Joint Commission on Cancer (AJCC) 8th edition of the Cancer Staging Manual reflects the improved survival outcomes and prognosis in HPV-associated OPSCC.⁵

Although nodal metastasis remains an important prognostic factor, the impact of single-LN metastasis is unclear. The proportion of patients with a pathologic single-LN metastasis has been reported in only 1 prior study as 42% among patients who underwent surgical resection for HPV-associated OPSCC.⁶ Previous studies have demonstrated that metastatic LN number and size are associated with survival and failure in this patient population.^3,7-9 This finding has led to recent changes in the AJCC staging system that delineate the importance of the number of pathologically positive LNs, with a cutoff of 1 to 4 vs 5 or more involved LNs, without LN size, laterality of LNs, or extranodal extension (ENE) associated with survival.⁵ Current National Comprehensive Cancer Network (NCCN) guidelines state that patients with HPV-related OPSCC and a single pathologic LN smaller than 3 cm can be observed after surgery in the absence of adverse features, with such features being an area of active investigation.¹⁰

Although studies have investigated the prognostic value of a high number of metastatic LNs,^3,8,9 to our knowledge, none have specifically examined the outcomes and prognostic factors in single-node metastasis. Thus, the objectives of this study are to assess oncologic outcomes after transoral surgery with or without adjuvant therapy in patients with single pathologic nodal disease and identify factors associated with failure.

We reviewed the electronic medical records of all consecutive patients with newly diagnosed, biopsy-proven HPV-related OPSCC who underwent transoral surgery from January 1, 2007, to December 31, 2016, at Washington University in St Louis and the Mayo Clinic. Human papillomavirus status was assessed by p16 immunohistochemistry. This study was approved by the institutional review boards of Washington University in St Louis (St Louis, Missouri) and the Mayo Clinic (Rochester, Minnesota). Written informed consent was obtained from participants to be included in the prospectively collected database.

All patients underwent transoral surgical resection of the primary site with either transoral robotic surgery or transoral laser microsurgery, with or without adjuvant therapy. Neck dissection was performed concurrently at the time of surgical resection, routinely levels II, III, and IV unless there was clinical evidence of nodal disease elsewhere. Only patients with pathologically proven metastasis to a single LN and at least 6 months of follow-up were included. Patients were excluded if they had prior treatment for their head and neck cancer, no or multiple pathologic LN metastases, p16-negative OPSCC, or unknown p16 status. The recommendation for adjuvant radiotherapy or chemotherapy was based on NCCN guidelines and was individualized for patients through a discussion with a multidisciplinary treatment team and took into account overall disease stage, high-risk pathologic features such as positive margins and ENE, comorbidities, functional status, and patient preference. Historical practice is to recommend adjuvant therapy for the following adverse pathologic features: stage T3 or T4 disease, positive margins, ENE, more than 1 involved LN, or LN metastasis larger than 3 cm for HPV-positive disease.

Clinical and pathologic data were collected retrospectively and included demographic data, age as a continuous variable, comorbidities, smoking history, site of tumor, clinical and pathologic stage, type of surgery (transoral laser vs robotic), extent of neck dissection if performed, pathologic features including positive margins, ENE, lymphovascular invasion, LN size, complications of neck dissection, adjuvant therapy, radiation dose, site of first recurrence, date of recurrence, treatment and outcome of recurrence, date of last follow-up or death, and status at last follow-up. The severity of comorbidities was defined by the Adult Comorbidity Evaluation 27 at the time of presentation with cancer¹¹ and was among the clinical factors assessed for association with survival and recurrence. The T and N staging was based on the 8th edition of the AJCC Cancer Staging Manual. Extranodal extension was assessed according to pathology reports stating present ENE and/or soft tissue metastasis. We chose to stratify by LN size of 3 cm or less vs more than 3 cm in our analysis based on NCCN guidelines using 3 cm as a cutoff for close observation vs further treatment, in addition to prior studies assessing its prognostic value.⁶

The primary outcome was regional recurrence. Secondary outcomes were overall survival (OS), any recurrence, and factors associated with regional recurrence and OS. Overall survival and time to recurrence were calculated as the time from the date of surgery to the date of death or recurrence, respectively.

Statistical analysis was performed from September 1, 2018, to September 1, 2020. Descriptive statistics were used to characterize the study sample. Univariable analysis was performed using logistic regression and Cox proportional hazards regression analysis was used to identify factors associated with regional recurrence and OS. The proportional assumption was tested and satisfied. The Kaplan-Meier method was used to generate survival curves and statistics. Data were analyzed using Stata statistical software, version 14.2 (StataCorp) and IBM SPSS Statistics, version 25.0 (IBM Corp).

A total of 207 eligible patients were identified who underwent transoral surgery with or without adjuvant therapy for p16-positive OPSCC with a single pathologic nodal metastasis. A total of 59 patients (29%) underwent surgery alone, 75 (36%) underwent surgery and received adjuvant radiotherapy, and 73 (35%) underwent surgery and received adjuvant chemoradiotherapy. Patients were predominantly male (178 [86%]), with a median age of 57 years (range, 35-82 years) at the time of surgery. Patient demographics and tumor characteristics are shown in Table 1. Median follow-up was 36.2 months (range, 7-127 months).

Among all patients, 11 (5%) developed regional recurrence, which occurred at a median of 12.2 months (range, 2.9-50.4 months) after surgical resection. Of these, 9 patients (82%) had underwent surgery alone and 2 (18%) received adjuvant radiotherapy. Two of the patients declined adjuvant radiotherapy despite its recommendation by the treatment team. Details of patients with regional recurrence and their treatment are described in Table 2. Among patients with regional recurrence, 6 of 9 patients who did not receive prior adjuvant therapy underwent successful salvage treatment, compared with 1 of 2 patients who received prior adjuvant therapy (Figure 1).

For patients with high-risk pathologic features for regional recurrence (165 [80%]), defined as ENE and/or LN size greater than 3 cm, none of the 130 patients who received initial adjuvant therapy developed regional recurrence, while 5 of the 35 patients (14%) who did not receive adjuvant treatment developed regional recurrence (Figure 2). For patients without these adverse features, 2 of 18 patients (11%) who received initial adjuvant therapy developed regional recurrence and 4 of 24 patients (17%) who did not receive adjuvant treatment developed regional recurrence.

Univariable analysis identified age (odds ratio [OR], 1.2; 95% CI, 1.1-1.2), advanced T stage (OR, 3.5; 95% CI, 0.9-14.0), and positive margins (OR, 10.9; 95% CI, 1.8-67.5) as associated with increased odds of regional recurrence (Figure 3). Extranodal extension (OR, 0.2; 95% CI, 0.04-0.8), LN size greater than 3 cm (OR, 0.2; 95% CI, 0.1-0.7), and adjuvant therapy (OR, 0.08; 95% CI, 0.02-0.4) were associated with decreased odds of regional recurrence. Presence of ENE (OR, 5.0; 95% CI, 2.6-9.6) and LN size greater than 3 cm (OR, 2.9; 95% CI, 1.5-5.4) were both associated with greater likelihood of receiving adjuvant therapy.

A total of 21 patients (10%) developed any recurrence: 4 (19%) developed local recurrence, 7 (33%) developed regional recurrence, 4 (19%) developed local and regional recurrence, and 6 (29%) developed distant metastasis. The median time to recurrence was 13.2 months (range, 2.9-95.9 months).

All 6 patients with distant metastasis received adjuvant therapy with either radiotherapy alone (3 [50%]) or chemoradiotherapy (3 [50%]), although 1 of the patients receiving radiotherapy alone did not complete the recommended allotment of radiotherapy secondary to toxic effects of treatment. Pathologic characteristics were notable for ENE in 5 of the 6 patients with distant metastasis (83%). Of patients with distant metastasis, 5 underwent salvage therapy (2 surgical resection alone, 1 surgical resection with adjuvant radiotherapy, 1 radiotherapy alone, and 1 chemoradiotherapy). Of these, 4 (80%) had successful salvage treatment and were alive at a median of 18.6 months (range, 7.9-59.1 months) since diagnosis of recurrence, and 1 (20%) died of disease 5.6 months after diagnosis.

There were 7 (3%) total deaths, with a 5-year OS of 95% (95% CI, 89%-98%). When stratified by regional recurrence, 5-year OS was 98% (95% CI, 89%-100%) for patients with no recurrence compared with 68% (95% CI, 29%-88%) among those with regional recurrence (eFigure in the Supplement).

On univariable analysis, advanced comorbidities (hazard ratio [HR], 6.20; 95% CI, 1.4-27.7), lymphovascular invasion (HR, 4.7; 95% CI, 1.0-21.2), and regional recurrence (HR, 16.0; 95% CI, 3.1-82.0) were associated with decreased OS. Presence of high-risk pathologic features (HR, 0.37; 95% CI, 0.07-2.04) and adjuvant therapy (HR, 0.48; 95% CI, 0.10-2.25) were not correlated with OS.

The results of our study demonstrate that OS is excellent at 95% in patients with a single pathologic node and HPV-related OPSCC undergoing transoral surgery with or without adjuvant therapy. Regional recurrence occurred in 11 patients (5%) in our population and was particularly higher among patients who did not receive adjuvant therapy. Although patients undergoing surgery alone had a higher rate of regional recurrence, adjuvant therapy had no association with OS. This finding is consistent with prior studies^12-14 and may be owing to the overall high cure rate among patients undergoing salvage therapy. However, there was a 14% recurrence rate among the subset of patients with adverse features who did not receive adjuvant therapy. These findings are particularly relevant as more studies aim to identify subsets of patients who may benefit from de-escalated treatment to reduce morbidity while maintaining excellent outcomes.

Among patients with treatment failure, the most common site of recurrence was regional. Patients with regional recurrence had lower OS (68% at 5 years). A total of 7 of 10 patients had successful salvage treatment. Factors associated with increased regional recurrence included age, advanced pathologic T stage, positive margins, and lack of adjuvant therapy. Larger size of LN metastasis and ENE were associated with decreased odds of regional recurrence. It is important to interpret these results in the context of the small sample size, with only 11 patients having regional recurrence. One possible explanation for these findings is that patients with larger LNs and ENE were more likely to receive adjuvant therapy, which was supported by our results, given the 3 to 5 times greater odds of receiving adjuvant therapy in this subset of patients, respectively. In addition, while we were able to identify factors associated with regional recurrence and OS on univariable analysis, the limited sample size and low number of events did not allow for multivariable analysis to assess potential confounders that may have included ENE or LN size. Our study was also underpowered to assess the clinical significance of the higher rate of successful salvage therapy among patients who did not receive initial adjuvant therapy. Among the 130 patients with ENE and/or larger LN size who received adjuvant treatment, none developed regional recurrence, suggesting that adjuvant therapy improves regional control for these high-risk patients.

The association of largest LN size with prognosis has been debated. For example, multiple studies have shown that LN size was not associated with recurrence or survival.^3,9,15,16 Recent changes to the 8th edition of the AJCC Cancer Staging Manual reflect the outcomes of several surgical studies demonstrating that metastatic node number, not pathologic nodal size or laterality, is associated with prognosis.^3,8,16,17 In contrast, a recent study by El Asmar et al⁷ showed that stratifying LN by size (≤4 cm and >4 cm) was associated with decreased survival among patients treated with surgery alone. Another study by An et al⁶ showed that size of LN metastasis (≤3 cm and >3 cm) was associated with ENE, but not with OS. Previous studies did not separately analyze LN size in patients with a single nodal metastasis. This finding illustrates that the understanding of nodal metastasis and its association with prognosis remains incompletely understood.

In addition, there is no consensus on the association of ENE with recurrence and survival in patients with HPV-related OPSCC. Although multiple studies have shown that ENE is not associated with worse prognosis in this population,^16,18,19 Bauer et al²⁰ showed a worse prognosis when evaluating National Cancer Database data. In addition, the presence of ENE remains an indication for adjuvant therapy as recommended by the NCCN.¹⁰ Cramer et al¹⁸ assessed outcomes in patients with p16-positive OPSCC with 4 or fewer LN metastases, and among these patients with low- to intermediate-risk disease, only higher T stage and increased nodal metastatic number were associated with survival; ENE was not prognostic for survival. Similarly, other studies showed that the presence of ENE did not correlate with OS or locoregional control.^15,18 In contrast, some data have shown that ENE is associated with decreased OS^6,7,21 or worse control.²² A study of patients who underwent surgical treatment for HPV-related OPSCC demonstrated that ENE continued to be associated with survival for patients with pathologic N1 but not N2 disease after stratification by nodal status.⁷ An et al⁶ demonstrated a decreased OS in ENE-positive patients among those with HPV-associated OPSCC undergoing surgery with or without adjuvant therapy, which also persisted for patients with only 1 metastatic LN. Additional studies regarding prognostic factors of LN size and ENE are necessary.

There are several strengths to the study. To our knowledge, this is the first study that focuses solely on outcomes in single-LN metastasis in HPV-related OPSCC. Second, our data suggest that for a single nodal metastasis, larger size of LN metastasis and ENE were not prognostic variables for regional recurrence or survival, but this finding must be interpreted in the setting of a limited sample size precluding the ability to perform a multivariable analysis. These features were associated with the greater likelihood of receiving adjuvant therapy, which may have confounded our findings and will require further study. Our results also demonstrate that adjuvant radiotherapy reduces the risk of regional recurrence but may not have an association with survival in this specific patient population. These findings are particularly relevant with the development of deintensification treatment strategies for specific low-risk patients. In addition, salvage therapy with curative intent for patients with regional recurrence is often successful and may improve OS in certain subsets of patients. Larger prospective studies are necessary to further explore the role of salvage therapy on OS.

Our study is limited by the inherent biases of a retrospective design, smaller sample size, and dual institutional design, with resulting heterogeneity in patient population and treatment practices between institutions. Treatment recommendations and patterns may also have evolved during the study period. In addition, 37 patients did not receive adjuvant therapy despite having high-risk features. Owing to the retrospective nature of the study, specific reasons were not evaluable. Reasons were likely patient refusal, comorbidities, and institutional practices. Given the small number of recurrences and deaths, we were unable to perform multivariable analyses to adjust for confounders. Certain treatment details including radiotherapy dose, chemotherapy regimen, and reason for treatment decisions were also missing, which may have led to potential confounders not being addressed in the analysis.

This study suggests that patients with HPV-related OPSCC with a solitary LN metastasis have excellent outcomes. The addition of adjuvant therapy in HPV-related OPSCC with a single nodal metastasis may decrease the risk of regional recurrence, but may not be associated with OS. Additional prospective studies with a larger study population should further explore outcomes in single-LN metastasis.

Accepted for Publication: August 28, 2020.

Published Online: November 5, 2020. doi:10.1001/jamaoto.2020.3870

Corresponding Author: Ryan S. Jackson, MD, Department of Otolaryngology–Head and Neck Surgery, Washington University School of Medicine in St Louis, 660 S Euclid Ave, Campus Box 8115, St Louis, MO 63110 (jackson.ryan@wustl.edu).

Author Contributions: Drs Chen and Jackson had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Sinha, Ettyreddy, Pipkorn, Rich, Zevallos, Moore, Daly, Jackson.

Acquisition, analysis, or interpretation of data: Chen, Sinha, Last, Ettyreddy, Kallogjeri, Pipkorn, Rich, Zevallos, Paniello, Puram, Van Abel, Oppelt, Palka, Adkins, Gay, Thorstad, Jackson.

Drafting of the manuscript: Chen, Sinha, Jackson.

Critical revision of the manuscript for important intellectual content: Sinha, Last, Ettyreddy, Kallogjeri, Pipkorn, Rich, Zevallos, Paniello, Puram, Van Abel, Moore, Oppelt, Palka, Adkins, Daly, Gay, Thorstad, Jackson.

Statistical analysis: Chen, Kallogjeri.

Administrative, technical, or material support: Ettyreddy, Zevallos, Moore, Oppelt, Adkins, Daly, Thorstad.

Supervision: Pipkorn, Rich, Zevallos, Puram, Moore, Adkins, Jackson.

Conflict of Interest Disclosures: Dr Kallogjeri reported owning stock and serving as consultant for PetentiaMetrics outside the submitted work. Dr Zevallos reported that he holds equity stocks in and serves as chief medical officer at Summit Biolabs outside the submitted work. Dr Oppelt reported receiving personal fees from Merck, Bristol Myers Squibb, and Eisai outside the submitted work. Dr Palka reported receiving personal fees from Merck, Genentech, and Boehringer Ingelheim outside the submitted work. Dr Adkins reported receiving grants and personal fees from Merck, Eli Lilly, Celgene/Bristol Myers Squibb, Pfizer, and Cue Biopharma; personal fees from Loxo Oncology; and grants from Novartis, Roche, Aduro, Atara, Matrix, Blueprint Medicine, Celldex, Enzychem, Exelixis, Shanghai De Novo, AstraZeneca, Medimmune, Kura, and Innate outside the submitted work. Dr Thorstad reported that his wife works for Elekta, a company that makes radiotherapy hardware and software, but his radiotherapy department does not use any Elekta hardware of software. No other disclosures were reported.

Funding/Support: This research was supported by the Foundation for Barnes Jewish Hospital.

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 decision to submit the manuscript for publication.

Disclaimer: Dr Kallogjeri is statistics editor of JAMA Otolaryngology–Head & Neck Surgery, but she was not involved in any of the decisions regarding review of the manuscript or its acceptance.