Abstract
Background:
The impact of close surgical margins on oncologic outcomes in HPV-related oropharyngeal squamous cell carcinoma (HPV + OPSCC) is unclear.
Methods:
Retrospective case series including patients undergoing single modality transoral robotic surgery (TORS) for HPV + OPSCC at three academic medical centers from 2010 to 2019. Outcomes were compared between patients with close surgical margins (<1 mm or requiring re-resection) and clear margins using the Kaplan–Meier method.
Results:
Ninety-nine patients were included (median follow-up 21 months, range 6–121). Final margins were close in 22 (22.2%) patients, clear in 75 (75.8%), and positive in two (2.0%). Eight patients (8.1%) recurred, including two local recurrences (2.0%). Four patients died during the study period (4.0%). Local control (p = 0.470), disease-free survival (p = 0.513), and overall survival (p = 0.064) did not differ between patients with close and clear margins.
Conclusions:
Patients with close surgical margins after TORS for HPV + OPSCC without concurrent indications for adjuvant therapy may be considered for observation alone.
Keywords: human papillomavirus, oropharynx, squamous cell carcinoma, surgical margins, transoral robotic surgery
1 ∣. INTRODUCTION
Transoral robotic surgery (TORS) accesses otherwise difficult to reach areas in the head and neck. Developed in the early 2000s,1-3 the da Vinci robot (Intuitive Surgical, Inc.) has become well-accepted for various indications in otolaryngology.4-10 TORS has since replaced open pharyngeal surgery in many clinical scenarios outside of advanced or recurrent malignancy, often sparing patients profound morbidity.11 The most common application for TORS remains the extirpation of oropharyngeal tumors residing predominately in the tonsil and base of tongue (BOT). In comparison with nonsurgical strategies for oropharyngeal tumors, TORS achieves equivalent oncologic results.12,13 As human papillomavirus (HPV)-related oropharyngeal squamous cell carcinoma (HPV + OPSCC) has become increasingly prevalent, utilization of TORS has similarly grown.4
Original descriptions of TORS for HPV + OPSCC advocated a subunit-based approach designed to resect tissue based on defined anatomic areas rather than to achieve a particular margin.3,14 These methods fell within a traditional “en-bloc” paradigm in which HPV-unrelated head and neck cancer (HNC) has long been approached, wherein margins less than 5 mm are considered “close” in that they portend a worse prognosis.15-19 In light of the anatomic boundaries present within the oropharynx, 5 mm margins are often not achievable.20 In the traditional paradigm, narrow surgical margins are often addressed with adjuvant radiotherapy or re-resection if possible. In an era of therapeutic de-escalation intended to minimize treatment morbidity, the question of what constitutes an acceptable margin for HPV + OPSCC has not been answered.
Herein, we examine detailed margin data on patients treated with TORS for HPV + OPSCC at three academic institutions. We hypothesized that patients with negative but close margins would have equivalent local control to those with wider margins. In light of the excellent survival outcomes among patients with HPV + OPSCC, the focus has shifted towards treatment morbidity and quality of life (QOL). While no data conclusively demonstrates a relationship between defect volume and QOL, it stands to reason that a tailored approach with narrower margins, if oncologically safe, could reduce morbidity and improve post-operative function. Furthermore, there could be significant functional and financial benefit to patients if radiation could safely be spared in select patients with “close” margins. Given that patients have multiple efficacious treatment options for HPV + OPSCC, better defining the role of close margins in this patient population is necessary to tailor adjuvant therapy.
2 ∣. MATERIALS AND METHODS
Institutional review boards approved the collection of study data at three academic oncology centers; Massachusetts Eye and Ear, Johns Hopkins Hospital, and University of California, San Francisco Hellen Diller Comprehensive Cancer Center. We retrospectively reviewed charts of patients with HPV + OPSCC from 2010 to 2019. We included patients with the following criteria: age ≥ 18 years, histologic diagnosis of squamous cell carcinoma with evidence of HPV infection via p16 immunohistochemistry or high-risk HPV DNA, primary tumor site in the oropharynx, single modality TORS during the study period, and minimum six months follow-up. Patients with recurrent disease, induction chemotherapy, distant metastases at initial presentation, or who underwent adjuvant therapy were excluded. Informed consent was waived due to the retrospective nature of the study.
We assessed race, ethnicity, age, sex, smoking history with total pack-years, tumor site, laterality, surgical procedure performed, tumor size (centimeters [cm]), pathologic T-classification, number of lymph nodes positive, pathologic N-classification, lymph node yield during neck dissection, largest positive lymph node, and the presence of intermediate and high-risk factors including perineural invasion (PNI), lymphovascular invasion (LVI), and extranodal extension (ENE). Both American Joint Committee on Cancer (AJCC) 7th edition and 8th edition nodal classifications were recorded. Risk stratification was applied to each patient based on the Eastern Cooperative Oncology Group (ECOG) 3311 trial (NCT01898494), which classifies patients undergoing TORS for HPV + OPSCC as low, intermediate or high risk based on pathologic features including margin status. We collected detailed margin data including the closest radial margin distance and location when available, use of and status of frozen section margins, which were performed at individual surgeon's discretion, and need for re-resection in an immediate (same day) or secondary (subsequent day) fashion. It is the general practice of surgeons at participating institutions to orient the specimen with the pathologist and grossly evaluate margins together. We are unable to retrospectively assess how frequently this did not occur. However, there is consensus that all surgeons oriented the specimens themselves, although the method varied by institution (e.g. some surgeons personally ink their margins, others place sutures).
Cases were dichotomized based on margin status: those with close margins and those with clear margins. For cases in which radial margins were reported, margins less than 1 mm were considered close, and margins 1 mm or greater were considered clear. This threshold was chosen to determine whether any radial margin is truly necessary or if “no ink on tumor” may be considered an acceptable margin in HPV + OPSCC. Thresholds of 2 and 3 mm were similarly tested to determine if cancer control differences were observable based on these cut-offs. Cases in which no radial margin distance was reported but where margins were reported as widely clear were also assigned to the clear margin group. Cases in which re-resections were required to clear the margin were placed within the close margin group unless a new final radial margin was otherwise reported by the pathologist. No consistent template was used to report specimen orientation and margin assessment. However, most institutions use the College of American Pathologists oropharyngeal synoptic reporting form for final pathology reports. We assessed length of follow-up, vital status, and local, regional, and distant disease status. For patients who experienced recurrent disease, we assessed the salvage treatment, current disease status, and follow-up since salvage treatment.
Statistical analyses were performed using SPSS (Version 25; IBM Corporation). The primary outcome was local control. Secondary outcomes were overall survival (OS) and disease-free survival (DFS). Group comparisons were made using chi-square tests, Fisher's exact tests, two-way Student's t-tests, and simple logistic regressions where appropriate. Local control, DFS, and OS were measured using the Kaplan–Meier method and compared using the log-rank test.
3 ∣. RESULTS
3.1 ∣. Patient and tumor characteristics
Ninety-nine patients were included (28–43 consecutive patients per institution). The mean age was 60.9 years (SD = 10.4); 90.9% of patients were white, and 82.8% were male. Fifty-three (53.5%) patients were current or former smokers, of which 29 (54.7%) had greater than 10 pack-year lifetime smoking history. Fifty-five (55.5%) tumors primarily involved the tonsil or pharyngeal wall, and 42 (42.4%) primarily involved the BOT. All patients underwent single modality TORS, with 59 patients who underwent radical tonsillectomy and 47 patients who underwent BOT resection. Seven patients underwent both given the extent of the tumor. Neck dissection was performed in 99.0% of cases, including 83 (89.8%) cases of ipsilateral neck dissection and 15 cases of bilateral neck dissection, all in patients with BOT tumors. Patients had the following AJCC 8th edition pathologic T classifications: 62.6% T1, 35.4% T2, and 2.0% T3. Patients had the following AJCC 8th edition pathologic N classifications: 27.3% N0/Nx, 69.7% N1, and 3.0% N2 and the following AJCC 7th edition pathologic N classifications: 27.3% N0/Nx, 33.3% N1, 23.2% N2a, 14.1% N2b, 1.0% N2c, and 1.0% N3. Patients had the following AJCC 8th edition overall stages: 94 (94.9%) stage I and 5 (5.1%) stage II. Three (3.0%) tumors had PNI and 17 (17.2%) had LVI. In 72 node-positive specimens, ENE was present in seven (9.7%), including two with microscopic ENE (≤1 mm) and five with macroscopic ENE (>1 mm).21 No patient had distant metastatic disease prior to surgery. Using ECOG 3311 risk stratification, five patients would be classified as high risk (5.1%), 59 as intermediate risk (59.6%), and 31 as low risk (31.3%). Among intermediate-risk patients, 35 (59.3%) would be placed in the intermediate-risk category for margin status alone, whereas 24 (40.7%) had other risk factors contributing to this classification. Four patients (4.0%) would be ineligible for trial inclusion due to a combination of the following: T3 (n = 1), N2c (n = 1), or N3 (n = 1) classification or positive margins (n = 2).
3.2 ∣. Margin status
Final margins were close in 22 (22.2%) patients, 1–1.9 mm in 16 (16.2%), 2–2.9 mm in 16 (16.2%), 3–3.9 mm in eight (8.1%), 4–4.9 mm in five (5.1%), and >5 mm in 15 (15.2%). Margins were widely clear without specified distance in 15 (15.2%) patients. Margins were positive in two (2.0%), both of whom refused re-resection. The closest margin locations were reported in 87 cases with 32.3% being mucosal, 38.4% being deep, and 17.2% being multiple. Tonsillar tumors more often had a closest deep margin (50.9%) relative to mucosal (34.0%) or multiple (15.1%) margins. Mucosal margins were more commonly close in BOT tumors (43.8%) relative to deep (28.1%) or multiple (28.1%) (p = 0.004). Among 73 cases with intraoperative frozen sections, 42 (57.5%) were taken from the surgical defect, 28 (38.4%) from the specimen, and three (4.1%) from both. Among 33 cases in which re-resections were performed, 20 (60.6%) were immediate in response to frozen section margins and/or surgeon assessment of a potential close/positive margin. The remaining 13 (39.4%) were secondary re-resections in response to close or positive final pathology results. Among 97 patients with negative margins, 22 (22.7%) had a close (<1 mm) final margin, and 75 (77.3%) patients had a clear (1 mm or greater) final margin.
Demographic and tumor characteristics are compared between patients with clear and close margins in Table 1. Patients with close margins were significantly older, had significantly higher T and N classifications, were significantly more likely to be in an intermediate risk category by ECOG 3311 classification, and were significantly more likely to have a closest margin on a mucosal surface rather than deep.
TABLE 1.
Demographic characteristics compared between margin groups
| Margin group |
|||||
|---|---|---|---|---|---|
| All patients |
Clear margin >1 mm (n = 75) |
Close margin <1 mm (n = 22) |
p | ||
| Age at diagnosis (mean [SD]) | 60.9 (10.3) | 60.6 (9.6) | 62.1 (13.0) | 0.032* | |
| Sex (n [%]) | Male | 82 (82.8) | 63 (84) | 17 (77) | 0.676 |
| Female | 17 (17.2) | 12 (16) | 5 (23) | ||
| Race (n [%]) | White | 90 (90.9) | 69 (92) | 19 (86.4) | 0.387 |
| Black | 1 (1.0) | - | 1 (4.5) | ||
| Other/unknown | 8 (8.1) | 6 (8) | 2 (9.1) | ||
| Ethnicity (n [%]) | Hispanic/Latino | 6 (6.1) | 6 (6.7) | 1 (4.5) | 0.663 |
| Not Hispanic/Latino | 91 (91.9) | 69 (93.3) | 21 (95.5) | ||
| Smoking status (n [%]) | Current smoker | 12 (12.1) | 7 (9.3) | 6 (28.6) | 0.241 |
| Former smoker | 41 (41.4) | 34 (45.3) | 6 (28.6) | ||
| Never smoker | 45 (45.4) | 34 (45.3) | 10 (45.5) | ||
| Total pack years (n [%]) | <10 | 14 (14.1) | 11 (26.8) | 2 (18.2) | 0.279 |
| 10+ | 29 (29.3) | 24 (58.5) | 5 (45.5) | ||
| Unknown | 10 (10.1) | 6 (14.6) | 4 (36.4) | ||
| Laterality (n [%]) | Left | 53 (53.5) | 43 (57.3) | 9 (40.9) | 0.38 |
| Right | 46 (46.5) | 32 (42.7) | 13 (59.1) | ||
| Anatomic subsite (n [%]) | Tonsil/pharyngeal wall | 55 (55.5) | 43 (57.3) | 12 (54.5) | 0.71 |
| Base of tongue | 42 (42.4) | 32 (42.7) | 10 (45.5) | ||
| Tumor size (cm) (mean [SD]) | 1.9 (1.0) | 1.8 (1.0) | 1.7 (0.9) | 0.913 | |
| Pathologic T Classification (AJCC 8th ed.) (n [%]) | I | 62 (62.6) | 50 (66.7) | 12 (54.5) | 0.031* |
| II | 35 (35.4) | 24 (32.0) | 10 (45.5) | ||
| III | 2 (2.0) | 1 (1.3) | - | ||
| Pathologic N Classification (AJCC 7th ed.) (n [%]) | 0/x | 27 (27.3) | 24 (32.0) | 3 (13.6) | 0.019* |
| I | 33 (33.3) | 26 (34.7) | 6 (27.3) | ||
| IIa | 23 (23.2) | 16 (21.3) | 7 (31.8) | ||
| IIb | 14 (14.1) | 9 (12.0) | 5 (22.7) | ||
| IIc | 1 (1.0) | - | - | ||
| III | 1 (1.0) | - | 1 (4.5) | ||
| Pathologic N Classification (AJCC 8th ed.) (n [%]) | 0/x | 27 (27.3) | 24 (32.0) | 3 (13.6) | 0.032* |
| 1 | 69 (69.7) | 50 (66.7) | 18 (81.8) | ||
| 2 | 3 (3.0) | 1 (1.3) | 1 (4.5) | ||
| Number of pathologic nodes (mean [SD]) | 1.1 (1.8) | 0.99 (1.9) | 1.45 (1.8) | 0.466 | |
| Lymph node yield (mean [SD]) | 32.0 (13.6) | 30.88 (12.9) | 35.23 (15.3) | 0.171 | |
| Largest positive node (cm) (mean [SD]) | 2.7 (1.5) | 2.42 (1.4) | 3.38 (1.7) | 0.51 | |
| ENE (n [%]) | Yes | 7 (9.9) | 4 (7.8) | 2 (10.5) | 0.076 |
| No | 62 (87.3) | 46 (90.2) | 15 (78.9) | ||
| Unknown | 2 (2.8) | 1 (1.0) | 2 (10.5) | ||
| PNI (n [%]) | Yes/suspicious | 3 (3.0) | 3 (4.0) | - | 0.686 |
| No | 83 (83.8) | 63 (84.0) | 18 (81.8) | ||
| Unknown | 13 (13.1) | 9 (12.0) | 4 (18.2) | ||
| LVI (n [%]) | Yes/suspicious | 17 (17.2) | 10 (13.3) | 6 (27.3) | 0.296 |
| No | 74 (74.7) | 58 (77.3) | 15 (68.2) | ||
| Unknown | 8 (8.1) | 7 (9.3) | 1 (4.5) | ||
| Closest margin location (n [%]) | Mucosal | 32 (36.8) | 20 (26.7) | 11 (50.0) | 0.035* |
| Deep | 38 (43.7) | 32 (42.7) | 6 (27.3) | ||
| Multiple | 17 (19.5) | 13 (17.3) | 4 (18.2) | ||
| Frozen section location (n [%]) | Defect | 41 (42.4) | 33 (44.0) | 8 (36.4) | 0.672 |
| Specimen | 28 (28.3) | 19 (25.3) | 9 (40.9) | ||
| Both | 3 (3.0) | 3 (4.0) | - | ||
| None | 27 (27.3) | 20 (26.7) | 5 (22.7) | ||
| ECOG 3311 risk stratification | High | 5 (5.1) | 3 (4.0) | 2 (9.1) | <0.001* |
| Intermediate | 59 (59.6) | 40 (53.3) | 19 (86.4) | ||
| Low | 31 (31.3) | 31 (41.3) | - | ||
| Ineligible | 4 (4.0) | 1 (1.3) | 1 (4.5) | ||
| Months Follow-up (mean [SD]) | 28.5 (21.6) | 30.2 (22.8) | 20.6 (11.2) | 0.002* | |
| Vital status (n [%]) | Alive | 93 (93.4) | 73 (97.3) | 20 (90.9) | 0.282 |
| Deceased | 4 (4.0) | 2 (2.7) | 2 (9.1) | ||
| Recurrence (n [%]) | Any recurrence | 8 (8.1) | 7 (9.3) | 1 (4.5) | 0.901 |
| No recurrence | 89 (89.9) | 68 (90.7) | 21 (95.5) | ||
| Local recurrence (n [%]) | No | 95 (96.0) | 73 (97.3) | 22 (100) | 1 |
| Yes | 2 (2.0) | 2 (2.7) | - | ||
Note: Statistically significant differences are denoted by asterisks. Patients with positive surgical margins excluded from this table.
Abbreviations: AJCC, American Joint Committee on Cancer; ECOG, Eastern Cooperative Oncology Group; ENE, extranodal extension; LVI, lymphovascular invasion; PNI, perineural invasion.
3.3 ∣. Clinical outcomes
No patients underwent adjuvant therapy after surgery, either by recommendation or patient preference. Mean follow-up was 28.5 months (SD = 21.6, median 21, range 6–121). Two (2.0%) patients developed local recurrences, one from a tonsillar tumor and one from a BOT tumor. One patient with a tonsillar tumor (initial deep margin immediately re-resected to a radial margin of >4 mm) developed a local recurrence 17 months after surgery, underwent salvage radiotherapy, and had no evidence of disease after eight months additional follow-up. A second patient with a BOT tumor (initial margin reported as widely negative) developed a local recurrence 18 months after surgery, underwent successful salvage surgery, and had no evidence of disease after 33 months additional follow-up. Both of these patients had initially clear surgical margins. Observation was recommended after initial surgery over adjuvant therapy to both patients. Three patients (3.0%) developed isolated regional recurrences. Three (3.0%) patients developed distant metastases. When combined, eight patients recurred at any location (8.1%). Adjuvant therapy was recommended but declined by 25 patients (24.2%). Eight of these patients had close margins and two had positive margins. Four of these patients experienced recurrences and three died during the study period.
Four total patients died during the study period. The average age of these patients was 78 years. Two of these patients suffered non-cancer mortality with no evidence of disease at the time of death, one of progression of distant metastasis, and one of unknown causes with unknown disease status. Overall, 94.9% of patients were alive without evidence of disease at the last follow-up. Details of patients who experienced adverse outcomes are outlined in Table 2.
TABLE 2.
Patients who experienced adverse outcomes
| Age at diagnosis |
Sex | Smoking status |
Subsite | T Classification |
N Classification (AJCC 8th ed.) |
N Classification (AJCC 7th ed.) |
ENE | PNI | Overall stage |
ECOG risk class |
Re- resection |
Final margin (mm) |
Event | Time to event (months) |
Salvage therapy |
Disease status |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 61 | Male | Never | Base of tongue | T1 | N1 | N2a | N0 | No | I | Low | No | Negative | Local recurrence | 18 | Surgery | NED |
| 59 | Female | Current | Tonsil | T1 | N0 | N0 | N/A | No | I | Low | Yes | 4–4.9 | Local recurrence | 17 | Radiation | NED |
| 53 | Male | Never | Base of tongue | T1 | N1 | N2a | N0 | No | I | Intermediate | No | Close | Regional recurrence | 27 | Surgery | NED |
| 60 | Male | Current | Base of tongue | T1 | N2 | N2b | Yes | Yes | II | High | Yes | Negative | Regional recurrence | 7 | Chemoradiation | NED |
| 57 | Male | Never | Tonsil | T1 | N1 | N1 | N0 | No | I | Low | No | >5 | Regional recurrence | 24 | Chemoradiation | NED |
| 54 | Male | Former | Base of tongue | T1 | N1 | N1 | N0 | No | I | Low | No | >5 | Distant metastasis | 27 | Chemotherapy | Active disease |
| 69 | Male | Never | Base of tongue | T1 | N0 | N0 | N/A | No | I | Intermediate | Yes | 1–1.9 | Distant metastasis | 41 | Chemotherapy | Active disease |
| 74 | Male | Former | Tonsil | T2 | N0 | N0 | N/A | No | I | Intermediate | No | 1–1.9 | Distant metastasis, death | 12 | Chemoradiation | Deceased |
| 88 | Male | Unknown | Base of tongue | T2 | N2 | N2b | No | No | II | High | No | Close | Death | 8 | N/A | Deceased |
| 88 | Male | Current | Base of tongue | T1 | N1 | N2a | No | Unknown | I | Intermediate | No | Close | Death | 28 | N/A | Deceased |
| 63 | Male | Former | Tonsil | T1 | N1 | N2b | Yes | No | I | High | No | 2–2.9 | Death | 6 | N/A | Deceased |
Abbreviations: AJCC, American Joint Committee on Cancer; ECOG, Eastern Cooperative Oncology Group; ENE, extranodal extension; PNI, perineural invasion.
Local control was 100% (74/74), 95.6% (43/45), and 93.8 (30/32) at one, two, and three years respectively. Local control did not differ between patients with close and clear margins (p = 0.470) (Figure 1). Local control did not appear to be influenced by disease subsite (p = 0.847), PNI (p = 0.722), LVI (p = 0.273), need for re-resection (p = 0.422) or use of (p = 0.354) or location of (p = 0.859) frozen section margins. DFS was 94.6% (69/73), 83.7% (36/43), and 72.4% (21/29) at one, two, and three years respectively. Margin status did not appear to affect DFS (p = 0.513). OS was 97.3% (73/75), 95.7% (45/47), and 87.8% (29/33) at one, two, and three years respectively. Margin status did not appear to affect OS (p = 0.064). Survival analyses performed based on alternative cut-offs for close margins at 2 and 3 mm showed no difference in local control (p = 0.331 at 2 mm, p = 0.173 at 3 mm) or DFS (p = 0.272 at 2 mm, p = 0.568 at 3 mm) between groups.
FIGURE 1.
Local control, disease-free survival, and overall survival by margin status
4 ∣. DISCUSSION
Descriptions of “close” margins date to 1970s studies on oral cancer in which margins less than 5 mm predicted lower survival and local control.15 Multiple cut-offs have been advocated,16,22,23 with recent reports showing favorable results in patients with margins as narrow as 1 mm in oral cancer.24 Nonetheless, a 2005 survey of 476 American Head and Neck Society member surgeons showed that 69% considered 5 mm to be the definition of a close margin.25 Furthermore, this definition is included in National Comprehensive Cancer Network (NCCN) guidelines for HNC.19 Closer margins may be adequate in glottic cancer,26,27 with 1–2 mm margins currently considered acceptable.19 How data on other sites within the head and neck translate to the oropharynx, and how HPV status alters what margin is acceptable, remains unclear.
In 2013, Hinni et al. evaluated 128 patients undergoing transoral laser microsurgery (TLM) for tonsillar carcinoma.20 They reported average deep margins of 1.98 mm and peripheral margins of 2.98 mm, with local control in previously untreated patients of 99% at 5 years, although 59.6% of patients received adjuvant radiotherapy. Radiotherapy was not offered to patients based on close margins alone. There were no local recurrences in the group that received surgery alone. They additionally reported magnetic resonance imaging-based measurements of superior constrictor muscle dimensions in healthy patients, with mean muscle thickness measuring 2.4 mm (SD = 0.8) at its thinnest point. They commented that no additional fascial layer exists deep to this muscle, especially in the region of the tonsillar fossa inferiorly where the styloglossus and stylopharyngeus muscles diverge. Instead, only the parapharyngeal fat lies deep to the constrictor at this point, which serves as a poor margin and contains the carotid artery. They concluded that careful clearance of the margin was adequate and that wide margins are often unobtainable, thereby supporting a “no ink on tumor” approach to OPSCC over a specific radial margin distance. NCCN guidelines now reflect this approach to some extent, allowing for 1.5–2.0 mm margins in TLM.19 Furthermore, ECOG 3311 risk stratification supports a close margin definition of 3 mm. Our multi-institution data provide further support for the “no ink on tumor” approach to margin status, with local control of 93.8% at three years in a larger set of surgery-alone patients.
Few studies have directly assessed the impact of close surgical margins on local control in OPSCC. A meta-analysis by Gorphe et al.28 pooled results of three studies29-31 including 48 patients with close margins and 249 patients with clear margins, with no difference in local control between groups (p = 0.249). Significant heterogeneity existed between studies, with different definitions of close margins, variable HPV reporting, and a portion of patients in each study receiving adjuvant therapy. By contrast, Chao et al.32 identified margin status as an independent predictor of lower local-recurrence-free survival (HR = 1.59, p = 0.027) in a study of 474 patients undergoing TORS for OPSCC. However, positive and close margins (less than 2 mm in this study) were grouped together in their study, and 70.0% of patients received adjuvant radiotherapy. Our results show no impact of close margins on local control, DFS, or OS in a more homogenous study population with uniform margin assessment.
No clinical trials have evaluated differences in outcomes in patients undergoing TORS with or without adjuvant therapy; however, excellent local control has been reported among patients undergoing single modality TORS. Weinstein et al. described one local failure among 30 patients at a mean follow-up of 2.7 years, although they did not report margin status or HPV status.33 Olsen et al. demonstrated 90.9% 3-year local control among 18 patients undergoing single modality TORS for OPSCC, with 72.2% of the study group being HPV positive.34 These results may be compared to those utilizing TORS and adjuvant therapy. For example, Patel et al. described outcomes following TLM with adjuvant radiation for OPSCC in 80 patients, reporting locoregional control (LRC) at three years of 98.6%.35 Kaczmar et al. similarly described outcomes for 114 HPV + OPSCC patients treated with surgery and reported 96.7% 2-year LRC.31 Adjuvant radiation was utilized in 78% and chemotherapy in 43% of patients.
Other traditional risk factors for poor local control in HNC include PNI36 and LVI.37 However, these factors did not appear to significantly impact local control in this study. Additionally, there was no association between frozen section utilization or location and local control. While the use of frozen sections from the specimen has been contended as relating to improved local control in oral cancer,38,39 this does not appear to matter for HPV + OPSCC. In our study, there was a significantly higher rate of close mucosal margins relative to deep margins in the close margin group, particularly in BOT tumors. While the anatomic limitations of the oropharynx principally affect deep margins in tonsillar tumors, our study results suggest that a narrow margin approach can possibly be safely applied to mucosal margins as well as deep margins.
The addition of adjuvant therapy after TORS clearly worsens QOL. A series of studies evaluating QOL in patients with OPSCC found that adjuvant therapy independently predicted worse speech, eating, attitude, and overall function.40,41 In addition, xerostomia, odynophagia, thrush, weight loss, need for gastrostomy, and gastrostomy dependence at three months were significantly worse among patients undergoing adjuvant therapy. Achim et al. prospectively evaluated 74 patients treated with TORS for OPSCC, noting improvements in dysphagia, weight loss, and multiple eating QOL scales in single modality surgery patients relative to those undergoing adjuvant therapy.42 Xu et al. found in a study of 76 HPV + OPSCC patients that those who underwent surgery alone, as compared to definitive chemoradiation, adjuvant radiation or adjuvant chemoradiation, reported significantly better salivary, taste, and oral function and less pain, financial, dental, and sexual problems.43 Choby et al. reported functional outcomes in 34 single modality TORS patients and reported improvement in pain, swallowing, activity, chewing, and overall QOL over multiple time intervals.44 It is clear that if oncologically safe, omission of adjuvant therapy after TORS has a meaningful impact on patient QOL. The functional impact of more recent de-escalation trials is pending. Sparing radiotherapy also preserves this therapeutic modality for possible future recurrence or second primary cancers. Although all the presently studied tumors were HPV-related, 53.5% of patients had a smoking history, and over half of these had greater than 10 lifetime pack-years of tobacco use, leaving them at risk for second primary tumors.
This study has several limitations. First, it is a singlearm study, which prohibits direct oncologic comparisons based on the use of adjuvant therapy. Despite this, treatment recommendations were rendered by multidisciplinary tumor boards consistent with NCCN guidelines. Second, it is a retrospective study with heterogeneity of both surgeons and pathologists. Methods of margin assessment, use of frozen sections, and approach to re-resections may have differed between centers and surgeons, or over time. For cases in which re-resections were necessary, assessing final margin status is difficult and ultimately somewhat subjective. Although TORS offers excellent visualization to optimize the accuracy of re-resection, prior studies have highlighted the potential for error in locating these at-risk areas with resultant unidentified positive margins.45 However, these factors are true at all centers; this heterogeneity may be considered a possible strength of the study that improves its external validity. However, it is important to note that each of the study centers is high-volume TORS center with pathologists specialized in HNC. For this study's data to be applied at other centers, surgeons should have close communication with pathologists regarding surgical margins and specimen orientation, as well as a multidisciplinary tumor board for careful review of each patient's case and possible indications for adjuvant therapy.
A third limitation is the heterogeneity in the close and clear margin groups, including patients with and without defined radial margin distances in each group. The described stratification was felt by the authors to most accurately reflect the clinical interpretation of margins, which can be difficult to categorize in the setting of re-resections. A fourth limitation is the small number of local recurrences in the study. While this high local control rate supports the study hypothesis, it also precludes multivariable analysis and limits study power and interpretation of nonsignificant results.
A final and most important limitation is the relatively short median follow-up of 21 months (range 6–121). While this compares favorably to many published studies on oncologic outcomes in HPV + OPSCC, previous work by Guo et al. demonstrated a median time to recurrence of 19.6 months among 80 patients with HPV + OPSCC.46 Thus, patients in our study may remain at risk for recurrence. Therefore, we believe this work offers preliminary and insufficient evidence per se and further long-term data are necessary prior to widespread clinical practice change. However, the patients in this study were followed closely with high-definition, distal chip endoscopy, and surveillance imaging which would be expected to identify an early local recurrence should one occur. Despite these limitations, this is, to our knowledge, the largest report on single modality TORS patients to evaluate the effect of margin status on local control. Thus, we believe this study adds important information to the existing literature on the management of HPV + OPSCC. The data reported herein likely reflect the overall favorable biology of HPV-associated disease and should not be extrapolated to HPV-negative SCC.
5 ∣. CONCLUSION
This study demonstrates excellent oncologic outcomes for patients undergoing single-modality TORS at three academic medical centers. Local control and DFS did not differ based on margin status <1 mm or need for re-resections. These data suggest that “no ink on tumor” may be sufficient for surgically resected HPV + OPSCC and that patients with close surgical margins may be considered for observation without adjuvant therapy if no concurrent indications for adjuvant therapy exist. This should however be validated in a prospective trial where margin analysis and approaches to re-resection of a positive margin are rigorously standardized and longer clinical follow-up of patients is available.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
- 1.O'Malley BW Jr, Weinstein GS, Snyder W, Hockstein NG. Transoral robotic surgery (TORS) for base of tongue neoplasms. Laryngoscope. 2006;116(8):1465–1472. [DOI] [PubMed] [Google Scholar]
- 2.Weinstein GS, O'Malley BW Jr, Hockstein NG. Transoral robotic surgery: supraglottic laryngectomy in a canine model. Laryngoscope. 2005;115(7):1315–1319. [DOI] [PubMed] [Google Scholar]
- 3.Weinstein GS, O'Malley BW Jr, Snyder W, Sherman E, Quon H. Transoral robotic surgery: radical tonsillectomy. Arch Otolaryngol Head Neck Surg. 2007;133(12):1220–1226. [DOI] [PubMed] [Google Scholar]
- 4.Gal TJ, Slezak JA, Kejner AE, Chen Q, Huang B. Treatment trends in oropharyngeal carcinoma: surgical technology meets the epidemic. Oral Oncol. 2019;97:62–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Patel SA, Magnuson JS, Holsinger FC, et al. Robotic surgery for primary head and neck squamous cell carcinoma of unknown site. JAMA Otolaryngol Head Neck Surg. 2013;139(11):1203–1211. [DOI] [PubMed] [Google Scholar]
- 6.Chan JY, Richmon JD. Transoral robotic surgery (TORS) for benign pharyngeal lesions. Otolaryngol Clin North Am. 2014;47(3):407–413. [DOI] [PubMed] [Google Scholar]
- 7.Miller SC, Nguyen SA, Ong AA, Gillespie MB. Transoral robotic base of tongue reduction for obstructive sleep apnea: a systematic review and meta-analysis. Laryngoscope. 2017;127(1):258–265. [DOI] [PubMed] [Google Scholar]
- 8.Montevecchi F, Caranti A, Cammaroto G, Meccariello G, Vicini C. Transoral robotic surgery (TORS) for bilateral Eagle syndrome. ORL J Otorhinolaryngol Relat Spec. 2019;81(1):36–40. [DOI] [PubMed] [Google Scholar]
- 9.Montevecchi F, Cammaroto G, Meccariello G, et al. Trans-oral robotic surgery (TORS) for the treatment of lingual tonsillitis. When conventional therapies fail. Int J Med Robot. 2017;13(3):e1763. [DOI] [PubMed] [Google Scholar]
- 10.Dziegielewski PT, Kang SY, Ozer E. Transoral robotic surgery (TORS) for laryngeal and hypopharyngeal cancers. J Surg Oncol. 2015;112(7):702–706. [DOI] [PubMed] [Google Scholar]
- 11.Ford SE, Brandwein-Gensler M, Carroll WR, Rosenthal EL, Magnuson JS. Transoral robotic versus open surgical approaches to oropharyngeal squamous cell carcinoma by human papillomavirus status. Otolaryngol Head Neck Surg. 2014;151(4):606–611. [DOI] [PubMed] [Google Scholar]
- 12.Baliga S, Kabarriti R, Jiang J, et al. Utilization of transoral robotic surgery (TORS) in patients with oropharyngeal squamous cell carcinoma and its impact on survival and use of chemotherapy. Oral Oncol. 2018;86:75–80. [DOI] [PubMed] [Google Scholar]
- 13.Mahmoud O, Sung K, Civantos FJ, Thomas GR, Samuels MA. Transoral robotic surgery for oropharyngeal squamous cell carcinoma in the era of human papillomavirus. Head Neck. 2018;40(4):710–721. [DOI] [PubMed] [Google Scholar]
- 14.Holsinger FC, McWhorter AJ, Ménard M, Garcia D, Laccourreye O. Transoral lateral oropharyngectomy for squamous cell carcinoma of the tonsillar region: I. Technique, complications, and functional results. Arch OtolaryngolHead Neck Surg. 2005;131(7):583–591. [DOI] [PubMed] [Google Scholar]
- 15.Looser KG, Shah JP, Strong EW. The significance of “positive” margins in surgically resected epidermoid carcinomas. Head Neck Surg. 1978;1(2):107–111. [DOI] [PubMed] [Google Scholar]
- 16.Loree TR, Strong EW. Significance of positive margins in oral cavity squamous carcinoma. Am J Surg. 1990;160(4):410–414. [DOI] [PubMed] [Google Scholar]
- 17.Anderson CR, Sisson K, Moncrieff M. A meta-analysis of margin size and local recurrence in oral squamous cell carcinoma. Oral Oncol. 2015;51(5):464–469. [DOI] [PubMed] [Google Scholar]
- 18.Yamada S, Kurita H, Shimane T, et al. Estimation of the width of free margin with a significant impact on local recurrence in surgical resection of oral squamous cell carcinoma. Int J Oral Maxillofac Surg. 2016;45(2):147–152. [DOI] [PubMed] [Google Scholar]
- 19.https://www.nccn.org/professionals/physician_gls/pdf/head-and-neck.pdf NCCNHancvRf. [Google Scholar]
- 20.Hinni ML, Zarka MA, Hoxworth JM. Margin mapping in transoral surgery for head and neck cancer. Laryngoscope. 2013;123(5):1190–1198. [DOI] [PubMed] [Google Scholar]
- 21.Lewis JS, Carpenter DH, Thorstad WL, Zhang Q, Haughey BH. Extracapsular extension is a poor predictor of disease recurrence in surgically treated oropharyngeal squamous cell carcinoma. Mod Pathol. 2011;24(11):1413–1420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Nason RW, Binahmed A, Pathak KA, Abdoh AA, Sándor GK. What is the adequate margin of surgical resection in oral cancer? Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2009;107(5):625–629. [DOI] [PubMed] [Google Scholar]
- 23.Brennan JA, Mao L, Hruban RH, et al. Molecular assessment of histopathological staging in squamous-cell carcinoma of the head and neck. N Engl J Med. 1995;332(7):429–435. [DOI] [PubMed] [Google Scholar]
- 24.Tasche KK, Buchakjian MR, Pagedar NA, Sperry SM. Definition of “close margin” in oral cancer surgery and association of margin distance with local recurrence rate. JAMA Otolaryngol Head Neck Surg. 2017;143(12):1166–1172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Meier JD, Oliver DA, Varvares MA. Surgical margin determination in head and neck oncology: current clinical practice. The results of an International American Head and Neck Society Member Survey. Head Neck. 2005;27(11):952–958. [DOI] [PubMed] [Google Scholar]
- 26.Ansarin M, Santoro L, Cattaneo A, et al. Laser surgery for early glottic cancer: impact of margin status on local control and organ preservation. Arch Otolaryngol Head Neck Surg. 2009;135(4):385–390. [DOI] [PubMed] [Google Scholar]
- 27.Hartl DM, De Mones E, Hans S, Janot F, Brasnu D. Treatment of early-stage glottic cancer by transoral laser resection. Ann Otol Rhinol Laryngol. 2007;116(11):832–836. [DOI] [PubMed] [Google Scholar]
- 28.Gorphe P, Simon C. A systematic review and meta-analysis of margins in transoral surgery for oropharyngeal carcinoma. Oral Oncol. 2019;98:69–77. [DOI] [PubMed] [Google Scholar]
- 29.Laccourreye O, Hans S, Ménard M, Garcia D, Brasnu D, Holsinger FC. Transoral lateral oropharyngectomy for squamous cell carcinoma of the tonsillar region: II. An analysis of the incidence, related variables, and consequences of local recurrence. Arch OtolaryngolHead Neck Surg. 2005;131(7): 592–599. [DOI] [PubMed] [Google Scholar]
- 30.Lörincz BB, Möckelmann N, Busch CJ, Knecht R. Functional outcomes, feasibility, and safety of resection of transoral robotic surgery: single-institution series of 35 consecutive cases of transoral robotic surgery for oropharyngeal squamous cell carcinoma. Head Neck. 2015;37(11):1618–1624. [DOI] [PubMed] [Google Scholar]
- 31.Kaczmar JM, Tan KS, Heitjan DF, et al. HPV-related oropharyngeal cancer: risk factors for treatment failure in patients managed with primary transoral robotic surgery. Head Neck. 2016;38(1):59–65. [DOI] [PubMed] [Google Scholar]
- 32.Chao H, Schonewolf C, Lukens J, Swisher-McClure S, Ghiam AF, Lin A. Outcomes for oropharyngeal squamous cell carcinoma (OPSCC) following transoral robotic surgery (TORS). Int J Radiat Oncol Biol Phys. 2018;102(3):e234–e235. [Google Scholar]
- 33.Weinstein GS, Quon H, Newman HJ, et al. Transoral robotic surgery alone for oropharyngeal cancer: an analysis of local control. Arch Otolaryngol Head Neck Surg. 2012;138(7):628–634. [DOI] [PubMed] [Google Scholar]
- 34.Olsen SM, Moore EJ, Laborde RR, et al. Transoral surgery alone for human-papillomavirus–associated oropharyngeal squamous cell carcinoma. Ear Nose Throat J. 2013;92(2):76–83. [PubMed] [Google Scholar]
- 35.Patel SH, Hinni ML, Hayden RE, et al. Transoral laser microsurgery followed by radiation therapy for oropharyngeal tumors: the Mayo Clinic Arizona experience. Head Neck. 2014;36(2):220–225. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Nair D, Mair M, Singhvi H, et al. Perineural invasion: independent prognostic factor in oral cancer that warrants adjuvant treatment. Head Neck. 2018;40(8):1780–1787. [DOI] [PubMed] [Google Scholar]
- 37.Adel M, Kao H-K, Hsu C-L, et al. Evaluation of lymphatic and vascular invasion in relation to clinicopathological factors and treatment outcome in oral cavity squamous cell carcinoma. Medicine. 2015;94(43):e1510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Maxwell JH, Thompson LD, Brandwein-Gensler MS, et al. Early oral tongue squamous cell carcinoma: sampling of margins from tumor bed and worse local control. JAMA Otolaryngol Head Neck Surg. 2015;141(12):1104–1110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Varvares MA, Walker RJ, Chiosea S. Does a specimen-based margin analysis of early tongue cancer better predict local control? Laryngoscope. 2016;126(11):2426–2427. [DOI] [PubMed] [Google Scholar]
- 40.Dziegielewski PT, Teknos TN, Durmus K, et al. Transoral robotic surgery for oropharyngeal cancer: long-term quality of life and functional outcomes. JAMA Otolaryngol Head Neck Surg. 2013;139(11):1099–1108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Sethia R, Yumusakhuylu AC, Ozbay I, et al. Quality of life outcomes of transoral robotic surgery with or without adjuvant therapy for oropharyngeal cancer. Laryngoscope. 2018;128(2):403–411. [DOI] [PubMed] [Google Scholar]
- 42.Achim V, Bolognone RK, Palmer AD, et al. Long-term functional and quality-of-life outcomes after transoral robotic surgery in patients with oropharyngeal cancer. JAMA Otolaryngol Head Neck Surg. 2018;144(1):18–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Xu MJ, Plonowska KA, Gurman ZR, et al. Treatment modality impact on quality of life for human papillomavirus–associated oropharynx cancer. Laryngoscope. 2020;130(2):E48–E56. [DOI] [PubMed] [Google Scholar]
- 44.Choby GW, Kim J, Ling DC, et al. Transoral robotic surgery alone for oropharyngeal cancer: quality-of-life outcomes. JAMA Otolaryngol Head Neck Surg. 2015;141(6):499–504. [DOI] [PubMed] [Google Scholar]
- 45.Kerawala CJ, Ong TK. Relocating the site of frozen sections—is there room for improvement? Head Neck. 2001; 23(3):230–232. [DOI] [PubMed] [Google Scholar]
- 46.Guo T, Qualliotine JR, Ha PK, et al. Surgical salvage improves overall survival for patients with HPV-positive and HPV-negative recurrent locoregional and distant metastatic oropharyngeal cancer. Cancer. 2015;121(12):1977–1984. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.