Acta Oto-Laryngologica. 2015; Early Online, 1–7
ORIGINAL ARTICLE
Transoral robotic surgery for the management of head and neck squamous cell carcinoma of unknown primary
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
HANI IBRAHIM CHANNIR1, NICLAS RUBEK1, HANS ULRIK NIELSEN1, KATALIN KISS2, BIRGITTE W. CHARABI1, CHRISTEL B. LAJER1 & CHRISTIAN VON BUCHWALD1 1
Department of Otorhinolaryngology, Head and Neck Surgery and Audiology and 2Department of Pathology, Rigshospitalet, University of Copenhagen, Denmark
Abstract Conclusion: The addition of transoral robotic surgery (TORS) in the diagnostic management of patients classified with head and neck squamous cell carcinoma of unknown primary (SCCUP) is promising and appears to improve detection rates of the primary tumour. The approach presented in this first Scandinavian study could potentially minimize the radiation field to the pharyngeal axis in patients with identified primary tumours. Objectives: The aim of the study was to investigate whether bilateral lingual tonsillectomy performed with TORS is feasible, and whether it could improve the detection rates of primary tumours in patients diagnosed and classified as having SCCUP. Methods: The study was retrospective and included 13 patients with SCCUP who were referred to TORS between October 2013 and January 2015. All 13 patients had previously undergone a full investigation programme following the national guidelines including whole-body PET/CT, examination in general anaesthesia, including random biopsies of the base of the tongue and bilateral palatine tonsillectomy without identification of the primary tumour. Results: The primary tumour was identified by TORS in seven of the 13 patients (54%) at the lingual tonsils. Human papillomavirus DNA and p16 were positive in all identified primary tumour specimens and in the corresponding lymph node metastases.
Keywords: TORS, human papillomavirus, SCCUP, p16
Introduction The management of head and neck squamous cell carcinoma of unknown primary (SCCUP) constitutes a challenge for many clinicians, when the diagnostic work-up fails to identify the site of the primary tumour. Approximately 2–10% of the squamous cell carcinomas of the head and neck are classified as SCCUP [1]. According to the Danish Head and Neck Cancer Group (DAHANCA), 52 patients on average were diagnosed with SCCUP in Denmark each year from 2012–2014, accounting for 4.4% of all new cases of head and neck cancer [2]. The DAHANCA developed a national diagnostic algorithm for SCCUP in 1998, which was recently updated in 2013. The work-up at our institute follows
the national guidelines and includes (successively) physical examination; fine-needle aspiration of the lymph node; flexible fibre-optic examination of the nasal cavity, pharynx, and larynx; whole-body PET/CT scan; pan-endoscopy under general anaesthesia with random biopsies of the base of the tongue; bilateral palatine tonsillectomy, and adenoidectomy. Traditionally, the treatment of the N-site in patients with SCCUP has been primary surgery with selective ipsilateral neck dissection. Based on the pTNM classification, patients with pN1 disease and no extracapsular spread (ECS) are referred to follow-up after neck dissection. These patients will remain untreated at the primary tumour site and they may subsequently develop growth of the primary tumour and run the risk of developing further lymph node metastases and
Correspondence: Hani I. Channir, MD, Department of Otorhinolaryngology, Head & Neck Surgery and Audiology, Rigshospitalet, Copenhagen, Denmark. Tel: +45 20259637. Fax: +45 35452629. E-mail: [email protected]
(Received 16 March 2015; accepted 14 May 2015) ISSN 0001-6489 print/ISSN 1651-2251 online 2015 Informa Healthcare DOI: 10.3109/00016489.2015.1052983
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
2
H. I. Channir et al.
distant metastases. Patients with stage above pN1 or with ECS are offered wide-target bilateral radiation of the neck, including the mucosa of the larynx and pharynx (hypo-, oro-, and nasopharynx), combined with chemotherapy if indicated. The applications of transoral robotic surgery (TORS) have grown rapidly worldwide for salvage and primary surgery in oropharyngeal and laryngeal sites [3]. Our institute is among the first head and neck centres in Scandinavia to start performing TORS and, to our knowledge, no other institutes in Scandinavia have published any series with TORS. Indications for TORS in Denmark may grow as the incidence of oropharyngeal squamous cell carcinoma is rising, which is associated with a high prevalence of HPV, with HPV16 being the predominant HPV type involved [4]. TORS improves the surgeon’s visualization and manoeuvrability in an otherwise challenging operating field with tumours located in the oropharynx, thereby enabling en-bloc resections. Recently, a small number of studies have explored the potential of TORS as a new diagnostic and treatment approach for patients with SCCUP. Overall, rates of detection of primary tumours with TORS have varied from 77–90% [5–9]. Some of the tumours detected have, however, been based on palatine tonsillectomies performed with TORS, which makes it difficult to compare the studies since the diagnostic work-up algorithm for SCCUP in Denmark already includes bilateral palatine tonsillectomy and random biopsies of the base of the tongue. The aim of the present study was to investigate whether bilateral lingual tonsillectomy (BLT) performed with TORS is feasible for improving the detection rates of primary tumours, and thereby potentially reducing the need for wide-field irradiation of the pharynx, which is associated with considerable side-effects and reduced quality-of-life. Materials and methods
from a non-profit organization to purchase a daVinci SI HD robot, which made the study possible. It was conducted according to the tenets of the Helsinki Declaration and approved by the Regional Scientific Ethical Committee (H-1-2014-082) and the Danish Data Protection Agency. TORS procedure The TORS BLT was performed using the daVinci SI HD robot mounted with a 5-mm monopolar cautery (spatula tip) and a 5-mm Maryland dissector (Figure 2). The lingual tonsils were exposed using a Feyh-Kastenbauer Weinstein-O’Malley (FK-WO) mouth retractor. The initial cut was made in the midline from the terminal sulcus of the tongue to the vallecula, without cutting into the epiglottis. The depth was determined as a plane at the junction of lymphoid tissue and the tongue musculature. The second cut was made along the terminal sulcus on one side from the midline to the lateral wall at the junction of the inferior anterior aspect of the palatine tonsillar fossa. Following the lateral wall almost to the lateral glossoepiglottic fold, the entire lingual tonsil was resected, finishing with a cut along the vallecula anteriorly to the epiglottic root. The procedure was then performed on the contralateral side. Hemostasis was insured with the hemoclips on larger vessels if needed, and by mono/bipolar cautery. The resected specimen was pinned and oriented on a corkboard to guide the pathologist in evaluating the resection margins. During the same procedure, the patients then underwent a unilateral level II–IV neck dissection on the SCCUP side. Post-operative adjuvant treatment Following the Danish national guidelines, patients with stage pN1M0 without ECS should not receive adjuvant treatment. With ECS and/or pN2 or pN3 stage, the patient will be offered post-operative irradiation concomitant with cisplatin therapy.
Study design Handling of the specimens and p16/HPV analysis The study was a retrospective case review of 13 patients who had undergone TORS-based unilateral or bilateral lingual tonsillectomy between October 2013 and January 2015 at the Department of Otolaryngology, Head and Neck Surgery at Rigshospitalet, Copenhagen. Patients diagnosed with SCCUP according to the national standardized diagnostic guidelines were included and referred to TORS (Figure 1). None of the patients included had had previous radiation therapy to the oropharynx or neck region. In 2013, our department received a donation
The removed specimens were formalin-fixed for one day, then cut into 2–3-mm thin slides perpendicular to the surgical margins (Figure 3). On average, 21 tissue slides were cut per patient. Surgical margins were not routinely described when the study was started, since the procedure was considered a large biopsy that was not orientated. This was, however, quickly taken into consideration and an orientation method was developed together with reporting the distance from the tumour to the surgical margins. Following the
Transoral robotic surgery and unknown primary
3
Physical and flexible fibereoptic exam of naso-,oro-, hypopharynx and larynx, ultrasonography and fineneedle aspiration
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
Interpretation of fine-needle aspiration
Unsure/non-diagnostic cytology without suspicion of primary tumour
Suspicion of squamous cell carcinoma metastasis with unknown primary
Lymph node extirpation
PET/CT
Suspicion of nonsquamous cell carcinoma
Referral to other diagnostic workup
General anaesthesia: - Pharyngo-laryngoscopy, esophagoscopy - Bilateral tonsillectomy - Adenoidectomy - Random biopsies of base of tongue
Unknown primary? Figure 1. The Danish guidelines for the diagnostic work-up of unknown primary tumour in the head and neck.
Figure 2. Bilateral lingual tonsillectomy with a monopolar cautery (right instrument) and a maryland dissector (left arm). The lingual tonsils are exposed using a Feyh-Kastenbauer Weinstein-O’Malley mouth retractor.
principles of most TORS protocols, margins above 2 mm were considered free, 0–2 mm close, and 0 mm to be involved [10].
Polymerase chain reaction (PCR) for HPV was performed on formalin-fixed paraffin-embedded material using consensus primers GP5+ and GP6+ [4]. Validation of the quality of the DNA and the efficacy of the PCR reaction was carried out using the housekeeping gene GAPDH. Immunohistochemistry for p16 was carried out using a Ventana Benchmark Ultra autostainer with the optiView detection kit and the p16 monoclonal antibody E6H4 (Roche, Basel, Switzerland). The staining was evaluated as being positive if there was a strong and diffuse nuclear and cytoplasmic reaction in more than 75% of the tumour cells [11]. Tumours were morphologically classified as non-keratinizing, as keratinizing, or as hybrid squamous cell carcinoma, the latter when there was a non-keratinizing morphology with more than 10% squamous differentiation as described earlier [12,13]. Results The demographic and clinical characteristics of all 13 patients included are given in Table I. There were nine men and four women with a median age of 60 years (range = 50–79), most of them former
4
H. I. Channir et al. was discharged from hospital after spending 4 days in the intensive care unit.
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
Neck dissection
Figure 3. Formalin fixed lingual tonsil specimen oriented on corkboard, the orange syringe is placed towards the epiglottis. The whole specimen was submitted in 16 sections. The tumour region (measuring 4 mm in diameter) is marked with arrows. The 4 mm tumour is present at the junction of section 1 and 10.
smokers. Eight patients presented with a stage N1 (62%), three a N2a (23%), and two a N2b (15%). There were no macroscopic signs of ulceration or palpable tumour during the pan-endoscopy under general anaesthesia, which included examination of the oropharynx. By performing TORS-based unilateral or bilateral lingual tonsillectomy, we identified the primary tumour in seven patients. Six out of seven primary tumours were classified as non-keratinizing squamous cell carcinoma of oropharyngeal type. One tumour was classified as hybrid squamous cell carcinoma. The median tumour diameter was 1.16 cm (range = 0.4–2.2 cm), with the smallest tumour size being present in at least two tissue blocks (Figure 3). All primary tumours and their corresponding metastases were p16 and HPV positive. None of the patients with p16 and HPV negative metastatic SCC had tumour in the lingual tonsils. ECS was seen in six of the 13 patients overall, but only two patients in the ‘detected primary’ group had ECS. Final pT stage in the ‘detected primary’ group was T1 in six cases and T2 in one patient. The final pN stage was N1 in five patients and N2b in two patients. Following TORS, short-term complications were observed in four patients. One patient developed temporarily impaired sensitivity of the tongue; one patient had difficulty in breathing and was admitted to the intensive care unit for 3 days and was hospitalized further due to complications with small pulmonary embolisms. One patient developed short-lasting severe pain but quickly recovered. Lastly, one patient developed a bleeding at the base of tongue and a haematoma on the neck, which was evacuated and the bleeding managed by electrocautery. The patient
Three patients in the ‘detected primary’ group had ipsilateral neck dissection performed after TORS, and one patient underwent bilateral neck dissection (Tables I and II). Three patients had lymph node extirpation performed prior to TORS. One patient had a lymph node excision due to a suspected infected neck cyst, where the histological analysis revealed a ruptured cystic non-keratinizing squamous cell carcinoma metastasis with heavy inflammatory reaction. In the ‘undetected primary’ group, four patients had ipsilateral neck dissection performed, one patient underwent bilateral neck dissection, and one patient underwent lymph node extirpation only. Neck dissection was not chosen in this patient since histological examination showed ECS in the extirpated lymph node and the patient was planned to receive radiation therapy. Oncological adjuvant treatment In the ‘detected primary’ group, four patients received chemotherapy concomitant with radiotherapy to the primary tumour site and regional lymph nodes (Table II). Chemotherapy was given due to N stage above N1 or ECS. One patient received radiotherapy alone. One patient with pT1N1 underwent BLT and an ipsilateral left neck dissection, but had a close surgical margin (1 mm). He achieved macro- and microradical surgery with re-resection of the T site with TORS and neck dissection on the contralateral side, as he declined the option of radiotherapy. One more patient achieved radical surgery and did not wish to receive adjuvant therapy post-operatively. In the ‘undetected primary’ group, four patients received widespread bilateral radiation to the naso-, oro-, hypopharyngeal, and laryngeal mucosa along with N sites, and two patients also received concomitant cisplatin treatment. Lastly, two patients with pT1N1 stage had surgery with free lymph node margins and did not receive adjuvant treatment. Discussion While assessing the role of TORS in identification of the primary tumour with unilateral or bilateral lingual tonsillectomy, this study showed a detection rate of 54% in a cohort of 13 patients. All these patients had HPV-positive tumours and it is noteworthy that seven out of nine patients (78%) with HPV-positive lymph node metastases had their primary tumour detected.
M
M
F
M
60
60
59
69
52
68
79
3
4
5
6
7
8
9
10 67
11 50
12 64
13 64
M
M
F
F
M
M
F
BLT
BLT
BLT
BLT
BLT
BLT
BLT
BLT
BLT
BLT
BLT
right LT
BL
IL
No
IL
IL
BL
IL
No
No
IL
IL
No
IL
No
Yes
Yes
Yes
No
Yes
No
Yes
No
Yes
No
Yes
No
Neck Primary dissection tumour found?
NKSCC
NKSCC
NKSCC
NKSCC
KSCC
KSCC
NKSCC
NKSCC
NKSCC
NKSCC
NKSCC
NKSCC
NKSCC
Histology of nodal metastases
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
NKSCC
NKSCC
NKSCC
HSCC
NKSCC
NKSCC
NKSCC
+
+
+
+
+
+
+
p16-status in HPV- nodal Histology p16-status metastases status of primary in primary tumour tumor
+
+
+
+
+
+
+
HPV-status in primary tumour
TxN2bMx
TxN1Mx
TxN1Mx
TxN1Mx
TxN2bMx
TxN1Mx
TxN1Mx
TxN2aMx
TxN2aMx
TxN1Mx
TxN1Mx
TxN2aMx
TxN1Mx
cTNM
TxN2bM0
T1N1M0
T1N2bM0
T1N1M0
TxN2bM0
T1N1M0
TxN1M0
T2N2bM0
TxN2aM0
T1N1M0
TxN1M0
T1N1M0
TxN2aM0
pTNM
Free
Close
Free
Free
Close
Close
Close
Margin
0.9 cm
1.8 cm
0.4 cm
0.6 cm
2.2 cm
1.3 cm
0.9 cm
Tumour sizea
Yes
Yes
Yes
No
Yes
No
No
No
Yes
No
No
No
Yes
ECS
Largest tumour diameter. LT, lingual tonsillectomy; BLT, bilateral lingual tonsillectomy; IL, ipsilateral; BL, bilateral; TORS, transoral robotic surgery; HPV, human papillomavirus; NKSCC, non-keratinizing squamous cell carcinoma; KSCC, keratinizing squamous cell carcinoma; HSCC, hybrid squamous cell carcinoma; cTNM, clinical TNM classification; pTNM, pathological TNM classification; ECS, extracapsular spread.
a
M
55
2
right LT
51
1
M
Age Sex TORS type
#
Table I. Patients clinical and final pathology, including status on p16 and HPV.
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
Transoral robotic surgery and unknown primary 5
6
H. I. Channir et al.
Table II. Distribution of adjuvant therapy type and neck dissection. Variables
Detected primary Undetected group group
Neck dissection Ipsilateral
3
4
Bilateral
1
1
Lymph node extirpation 3
1
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
Adjuvant therapy Radiation alone
1
2
Chemoradiation
4
2
None
2
2
Ipsilateral radiation
0
0
Bilateral radiation
5
4
Radiation site
The patients in whom the primary tumour was not identified underwent unnecessary intervention with TORS and two patients unfortunately experienced short-term complications. The overall treatment strategy changed for all seven patients in whom the primary tumour was identified by offering TORS. Four of these patients would have been untreated at the primary tumour site, and thereby sub-optimally treated, if TORS had not been performed. Three of the patients in the ‘detected primary’ group received a reduced radiation field compared to the wide-field radiation they would have been offered due to a SCCUP with stage N2 or ECS. Two patients received only primary surgical treatment. The percentage of patients receiving chemotherapy was higher in the ‘detected primary’ group than in the ‘undetected primary’ group, mostly due to the advanced stages and ECS and not because of involved or close surgical margins. A larger study with long-term oncological outcomes will be required before conclusions can be drawn. Considering the possibility of de-escalation of the oncological treatment as a consequence of finding the primary site, there is a reasonable argument for implementing TORS as a diagnostic addition, as much is gained in reducing long-term morbidity associated with radiotherapy. Early identification of the primary site in the lingual tonsils may offer the possibility of primary radical surgery instead of primary radiation therapy, since the patients often present with very small and superficial tumours (down to 4 mm in this study), as showed previously [6,7]. The tumours are neither palpable nor macroscopically visible during the initial diagnostic work-up. TORS provides the potential for a wide BLT and definitive treatment of the primary tumour. It is difficult to state the overall importance of TORS
based on our small group of selected patients, but TORS seem to improve detection rates of the primary tumour when added, especially if the HPV status of the lymph node metastases is known. The procedure is considered safe, which was confirmed by the shortterm complications associated with TORS – in concordance with previous studies [6,9]. Since the first case report proposing TORS as a diagnostic tool for SCCUP [14], several authors have proposed a transoral robotic approach for the management of SCCUP [5–7,9,15]. Transoral laser microsurgery may also be an option for detecting primary tumours in SCCUP patients with comparable detection rates, and can be recommended in Institutes where robotic surgery is unavailable [9,16]. Studies using a TORS approach have found high detection rates (77–90%), but they used different definitions of a true ‘SCCUP’ syndrome; however, they still managed to identify the majority of primary tumours in the oropharyngeal palatine and lingual tonsils. Some studies have included palatine tonsils as their indication for TORS [5,6] which explains how the definition of a true ‘SCCUP’ syndrome varies in the literature [1]. If we focus on the exact T site and separate that from the published studies, which included 166 patients in total, there were 76 lingual tonsil carcinomas, corresponding to an identification rate of 46% which is comparable to our detection rate. This could encourage clinicians to increase the number of biopsies from the base of the tongue. This may be particularly relevant in institutes with no option for robotic surgery. Many institutes perform similar pre-diagnostic work-ups, but there are still differences – especially in the choice of imaging. We use whole-body PET/CT that often indicates primary focus and excludes further SCCUP work-up if the PET positive site is confirmed by histology. In Denmark, patients are diagnosed with SCCUP after the diagnostic workup including palatine tonsillectomy has failed to identify the primary tumour. Our institute has recently performed the largest unselected, consecutive cohort of tonsillar cancers (TSCCs) to date. This showed the relationship between HPV and oropharyngeal tumours and also showed an increasing incidence of HPV-positive TSCC in Denmark [4]. Based on this study, it can be expected that we will encounter a rising number of patients with SCCUP due to HPV pathogenesis; however, the detection rate of primary tumours has improved in recent years, mainly from adding PET/CT scans. We emphasize the use of cytological cell material obtained from FNA as a substrate for HPV detection. Those patients with HPV-positive cytology in neck metastases should be considered eligible for TORS lingual tonsillectomy, after a palatine tonsillectomy
Transoral robotic surgery and unknown primary with no detection of the primary tumour has been performed, as shown in our diagnostic work-up algorithm. In addition, the handling of the surgical specimens plays a major role in finding the primary tumour, by performing >20 tissue slides of 2–3 mm thickness followed by haematoxylin and eosin staining. TORS may take on a greater role in this diagnostic setting in the future management of patients with SCCUP.
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
Conclusion Detection of the primary tumour using TORS combined with HPV DNA and p16 testing appears feasible for patients with SCCUP, who would otherwise be untreated (i.e. sub-optimally treated) at the primary site. We found that seven out of nine patients (78%) with HPV DNA/p16-positive lymph node metastases had a primary tumour in the lingual tonsils. The approach presented in this first Scandinavian study could potentially improve functional outcomes in the ‘detected primary’ group of patients by minimizing the radiation field to the pharyngeal axis. Declaration of interest: The authors report no conflicts of interests. The authors alone are responsible for the content of the paper. They certify that they have no affiliation with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials described in the paper. In 2013, the department received a donation from a non-profit organization to purchase a daVinci SI HD robot.
References [1] Straetmans J, Vent J, Lacko M, Speel E-J, Huebbers C, Semrau R, et al. Management of neck metastases of unknown primary origin united in two European centers. Eur Arch Otorhinolaryngol 2015;272:195–205. [2] Danish Head and Neck Cancer Group [Internet]. National guidelines regarding carcinoma metastases on the neck from unknown primary tumor, 2013. Available from: https://www. dahanca.oncology.dk/. [Last accessed 25 February 2015]. [3] Rinaldi V, Pagani D, Torretta S, Pignataro L. Transoral robotic surgery in the management of head and neck tumours. Ecancermedicalscience 2013;7:359.
7
[4] Garnaes E, Kiss K, Andersen L, Therkildsen MH, Franzmann MB, Filtenborg-Barnkob B, et al. A high and increasing HPV prevalence in tonsillar cancers in Eastern Denmark, 2000-2010: the largest registry-based study to date. Int J Cancer 2014;4:2000–10. [5] Durmus K, Rangarajan SV, Old MO, Agrawal A, Teknos TN, Ozer E. Transoral robotic approach to carcinoma of unknown primary. Head Neck 2014;36:848–52. [6] Patel SA, Magnuson JS, Holsinger FC, Karni RJ, Richmon JD, Gross ND, et al. Robotic surgery for primary head and neck squamous cell carcinoma of unknown site. JAMA Otolaryngol. Head Neck Surg 2013;139:1203–11. [7] Byrd JK, Smith KJ, de Almeida JR, Albergotti WG, Davis KS, Kim SW, et al. Transoral robotic surgery and the unknown primary: a cost-effectiveness analysis. Otolaryngol Head Neck Surg 2014;150:976–82. [8] Mehta V, Johnson P, Tassler A, Kim S, Ferris RL, Nance M, et al. A new paradigm for the diagnosis and management of unknown primary tumors of the head and neck: a role for transoral robotic surgery. Laryngoscope 2013; 123:146–51. [9] Graboyes EM, Sinha P, Thorstad WL, Rich JT, Haughey BH. Management of human papillomavirus-related unknown primaries of the head and neck with a transoral surgical approach. Head Neck 2014; doi: 10.1002/hed.23800. [10] Weinstein GS, Quon H, Newman HJ, Chalian JA, Malloy K, Lin A, et al. Transoral robotic surgery alone for oropharyngeal cancer. Arch Otolaryngol Head Neck Head Neck Surg 2012;138:628–34. [11] Lewis JS, Chernock RD, Ma X-J, Flanagan JJ, Luo Y, Gao G, et al. Partial p16 staining in oropharyngeal squamous cell carcinoma: extent and pattern correlate with human papillomavirus RNA status. Mod Pathol 2012;25: 1212–20. [12] El-Mofty SK, Zhang MQ, Davila RM. Histologic identification of human papillomavirus (HPV)-related squamous cell carcinoma in cervical lymph nodes: a reliable predictor of the site of an occult head and neck primary carcinoma. Head Neck Pathol 2008;2:163–8. [13] Lewis JS, Thorstad WL, Chernock RD, Haughey BH, Yip JH, Zhang Q, et al. p16 positive oropharyngeal squamous cell carcinoma:an entity with a favorable prognosis regardless of tumor HPV status. Am J Surg Pathol 2010;34:1088–96. [14] Abuzeid WM, Bradford CR, Divi V. Transoral robotic biopsy of the tongue base: a novel paradigm in the evaluation of unknown primary tumors of the head and neck. Head Neck 2013;35:E126–30. [15] Mehta V, Johnson P, Tassler A, Kim S, Ferris RL, Nance M, et al. A new paradigm for the diagnosis and management of unknown primary tumors of the head and neck: a role for transoral robotic surgery. Laryngoscope 2013; 123:146–51. [16] Nagel TH, Hinni ML, Hayden RE, Lott DG. Transoral laser microsurgery for the unknown primary: role for lingual tonsillectomy. Head Neck 2014;36:942–6.
ORIGINAL ARTICLE
Transoral robotic surgery for the management of head and neck squamous cell carcinoma of unknown primary
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
HANI IBRAHIM CHANNIR1, NICLAS RUBEK1, HANS ULRIK NIELSEN1, KATALIN KISS2, BIRGITTE W. CHARABI1, CHRISTEL B. LAJER1 & CHRISTIAN VON BUCHWALD1 1
Department of Otorhinolaryngology, Head and Neck Surgery and Audiology and 2Department of Pathology, Rigshospitalet, University of Copenhagen, Denmark
Abstract Conclusion: The addition of transoral robotic surgery (TORS) in the diagnostic management of patients classified with head and neck squamous cell carcinoma of unknown primary (SCCUP) is promising and appears to improve detection rates of the primary tumour. The approach presented in this first Scandinavian study could potentially minimize the radiation field to the pharyngeal axis in patients with identified primary tumours. Objectives: The aim of the study was to investigate whether bilateral lingual tonsillectomy performed with TORS is feasible, and whether it could improve the detection rates of primary tumours in patients diagnosed and classified as having SCCUP. Methods: The study was retrospective and included 13 patients with SCCUP who were referred to TORS between October 2013 and January 2015. All 13 patients had previously undergone a full investigation programme following the national guidelines including whole-body PET/CT, examination in general anaesthesia, including random biopsies of the base of the tongue and bilateral palatine tonsillectomy without identification of the primary tumour. Results: The primary tumour was identified by TORS in seven of the 13 patients (54%) at the lingual tonsils. Human papillomavirus DNA and p16 were positive in all identified primary tumour specimens and in the corresponding lymph node metastases.
Keywords: TORS, human papillomavirus, SCCUP, p16
Introduction The management of head and neck squamous cell carcinoma of unknown primary (SCCUP) constitutes a challenge for many clinicians, when the diagnostic work-up fails to identify the site of the primary tumour. Approximately 2–10% of the squamous cell carcinomas of the head and neck are classified as SCCUP [1]. According to the Danish Head and Neck Cancer Group (DAHANCA), 52 patients on average were diagnosed with SCCUP in Denmark each year from 2012–2014, accounting for 4.4% of all new cases of head and neck cancer [2]. The DAHANCA developed a national diagnostic algorithm for SCCUP in 1998, which was recently updated in 2013. The work-up at our institute follows
the national guidelines and includes (successively) physical examination; fine-needle aspiration of the lymph node; flexible fibre-optic examination of the nasal cavity, pharynx, and larynx; whole-body PET/CT scan; pan-endoscopy under general anaesthesia with random biopsies of the base of the tongue; bilateral palatine tonsillectomy, and adenoidectomy. Traditionally, the treatment of the N-site in patients with SCCUP has been primary surgery with selective ipsilateral neck dissection. Based on the pTNM classification, patients with pN1 disease and no extracapsular spread (ECS) are referred to follow-up after neck dissection. These patients will remain untreated at the primary tumour site and they may subsequently develop growth of the primary tumour and run the risk of developing further lymph node metastases and
Correspondence: Hani I. Channir, MD, Department of Otorhinolaryngology, Head & Neck Surgery and Audiology, Rigshospitalet, Copenhagen, Denmark. Tel: +45 20259637. Fax: +45 35452629. E-mail: [email protected]
(Received 16 March 2015; accepted 14 May 2015) ISSN 0001-6489 print/ISSN 1651-2251 online 2015 Informa Healthcare DOI: 10.3109/00016489.2015.1052983
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
2
H. I. Channir et al.
distant metastases. Patients with stage above pN1 or with ECS are offered wide-target bilateral radiation of the neck, including the mucosa of the larynx and pharynx (hypo-, oro-, and nasopharynx), combined with chemotherapy if indicated. The applications of transoral robotic surgery (TORS) have grown rapidly worldwide for salvage and primary surgery in oropharyngeal and laryngeal sites [3]. Our institute is among the first head and neck centres in Scandinavia to start performing TORS and, to our knowledge, no other institutes in Scandinavia have published any series with TORS. Indications for TORS in Denmark may grow as the incidence of oropharyngeal squamous cell carcinoma is rising, which is associated with a high prevalence of HPV, with HPV16 being the predominant HPV type involved [4]. TORS improves the surgeon’s visualization and manoeuvrability in an otherwise challenging operating field with tumours located in the oropharynx, thereby enabling en-bloc resections. Recently, a small number of studies have explored the potential of TORS as a new diagnostic and treatment approach for patients with SCCUP. Overall, rates of detection of primary tumours with TORS have varied from 77–90% [5–9]. Some of the tumours detected have, however, been based on palatine tonsillectomies performed with TORS, which makes it difficult to compare the studies since the diagnostic work-up algorithm for SCCUP in Denmark already includes bilateral palatine tonsillectomy and random biopsies of the base of the tongue. The aim of the present study was to investigate whether bilateral lingual tonsillectomy (BLT) performed with TORS is feasible for improving the detection rates of primary tumours, and thereby potentially reducing the need for wide-field irradiation of the pharynx, which is associated with considerable side-effects and reduced quality-of-life. Materials and methods
from a non-profit organization to purchase a daVinci SI HD robot, which made the study possible. It was conducted according to the tenets of the Helsinki Declaration and approved by the Regional Scientific Ethical Committee (H-1-2014-082) and the Danish Data Protection Agency. TORS procedure The TORS BLT was performed using the daVinci SI HD robot mounted with a 5-mm monopolar cautery (spatula tip) and a 5-mm Maryland dissector (Figure 2). The lingual tonsils were exposed using a Feyh-Kastenbauer Weinstein-O’Malley (FK-WO) mouth retractor. The initial cut was made in the midline from the terminal sulcus of the tongue to the vallecula, without cutting into the epiglottis. The depth was determined as a plane at the junction of lymphoid tissue and the tongue musculature. The second cut was made along the terminal sulcus on one side from the midline to the lateral wall at the junction of the inferior anterior aspect of the palatine tonsillar fossa. Following the lateral wall almost to the lateral glossoepiglottic fold, the entire lingual tonsil was resected, finishing with a cut along the vallecula anteriorly to the epiglottic root. The procedure was then performed on the contralateral side. Hemostasis was insured with the hemoclips on larger vessels if needed, and by mono/bipolar cautery. The resected specimen was pinned and oriented on a corkboard to guide the pathologist in evaluating the resection margins. During the same procedure, the patients then underwent a unilateral level II–IV neck dissection on the SCCUP side. Post-operative adjuvant treatment Following the Danish national guidelines, patients with stage pN1M0 without ECS should not receive adjuvant treatment. With ECS and/or pN2 or pN3 stage, the patient will be offered post-operative irradiation concomitant with cisplatin therapy.
Study design Handling of the specimens and p16/HPV analysis The study was a retrospective case review of 13 patients who had undergone TORS-based unilateral or bilateral lingual tonsillectomy between October 2013 and January 2015 at the Department of Otolaryngology, Head and Neck Surgery at Rigshospitalet, Copenhagen. Patients diagnosed with SCCUP according to the national standardized diagnostic guidelines were included and referred to TORS (Figure 1). None of the patients included had had previous radiation therapy to the oropharynx or neck region. In 2013, our department received a donation
The removed specimens were formalin-fixed for one day, then cut into 2–3-mm thin slides perpendicular to the surgical margins (Figure 3). On average, 21 tissue slides were cut per patient. Surgical margins were not routinely described when the study was started, since the procedure was considered a large biopsy that was not orientated. This was, however, quickly taken into consideration and an orientation method was developed together with reporting the distance from the tumour to the surgical margins. Following the
Transoral robotic surgery and unknown primary
3
Physical and flexible fibereoptic exam of naso-,oro-, hypopharynx and larynx, ultrasonography and fineneedle aspiration
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
Interpretation of fine-needle aspiration
Unsure/non-diagnostic cytology without suspicion of primary tumour
Suspicion of squamous cell carcinoma metastasis with unknown primary
Lymph node extirpation
PET/CT
Suspicion of nonsquamous cell carcinoma
Referral to other diagnostic workup
General anaesthesia: - Pharyngo-laryngoscopy, esophagoscopy - Bilateral tonsillectomy - Adenoidectomy - Random biopsies of base of tongue
Unknown primary? Figure 1. The Danish guidelines for the diagnostic work-up of unknown primary tumour in the head and neck.
Figure 2. Bilateral lingual tonsillectomy with a monopolar cautery (right instrument) and a maryland dissector (left arm). The lingual tonsils are exposed using a Feyh-Kastenbauer Weinstein-O’Malley mouth retractor.
principles of most TORS protocols, margins above 2 mm were considered free, 0–2 mm close, and 0 mm to be involved [10].
Polymerase chain reaction (PCR) for HPV was performed on formalin-fixed paraffin-embedded material using consensus primers GP5+ and GP6+ [4]. Validation of the quality of the DNA and the efficacy of the PCR reaction was carried out using the housekeeping gene GAPDH. Immunohistochemistry for p16 was carried out using a Ventana Benchmark Ultra autostainer with the optiView detection kit and the p16 monoclonal antibody E6H4 (Roche, Basel, Switzerland). The staining was evaluated as being positive if there was a strong and diffuse nuclear and cytoplasmic reaction in more than 75% of the tumour cells [11]. Tumours were morphologically classified as non-keratinizing, as keratinizing, or as hybrid squamous cell carcinoma, the latter when there was a non-keratinizing morphology with more than 10% squamous differentiation as described earlier [12,13]. Results The demographic and clinical characteristics of all 13 patients included are given in Table I. There were nine men and four women with a median age of 60 years (range = 50–79), most of them former
4
H. I. Channir et al. was discharged from hospital after spending 4 days in the intensive care unit.
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
Neck dissection
Figure 3. Formalin fixed lingual tonsil specimen oriented on corkboard, the orange syringe is placed towards the epiglottis. The whole specimen was submitted in 16 sections. The tumour region (measuring 4 mm in diameter) is marked with arrows. The 4 mm tumour is present at the junction of section 1 and 10.
smokers. Eight patients presented with a stage N1 (62%), three a N2a (23%), and two a N2b (15%). There were no macroscopic signs of ulceration or palpable tumour during the pan-endoscopy under general anaesthesia, which included examination of the oropharynx. By performing TORS-based unilateral or bilateral lingual tonsillectomy, we identified the primary tumour in seven patients. Six out of seven primary tumours were classified as non-keratinizing squamous cell carcinoma of oropharyngeal type. One tumour was classified as hybrid squamous cell carcinoma. The median tumour diameter was 1.16 cm (range = 0.4–2.2 cm), with the smallest tumour size being present in at least two tissue blocks (Figure 3). All primary tumours and their corresponding metastases were p16 and HPV positive. None of the patients with p16 and HPV negative metastatic SCC had tumour in the lingual tonsils. ECS was seen in six of the 13 patients overall, but only two patients in the ‘detected primary’ group had ECS. Final pT stage in the ‘detected primary’ group was T1 in six cases and T2 in one patient. The final pN stage was N1 in five patients and N2b in two patients. Following TORS, short-term complications were observed in four patients. One patient developed temporarily impaired sensitivity of the tongue; one patient had difficulty in breathing and was admitted to the intensive care unit for 3 days and was hospitalized further due to complications with small pulmonary embolisms. One patient developed short-lasting severe pain but quickly recovered. Lastly, one patient developed a bleeding at the base of tongue and a haematoma on the neck, which was evacuated and the bleeding managed by electrocautery. The patient
Three patients in the ‘detected primary’ group had ipsilateral neck dissection performed after TORS, and one patient underwent bilateral neck dissection (Tables I and II). Three patients had lymph node extirpation performed prior to TORS. One patient had a lymph node excision due to a suspected infected neck cyst, where the histological analysis revealed a ruptured cystic non-keratinizing squamous cell carcinoma metastasis with heavy inflammatory reaction. In the ‘undetected primary’ group, four patients had ipsilateral neck dissection performed, one patient underwent bilateral neck dissection, and one patient underwent lymph node extirpation only. Neck dissection was not chosen in this patient since histological examination showed ECS in the extirpated lymph node and the patient was planned to receive radiation therapy. Oncological adjuvant treatment In the ‘detected primary’ group, four patients received chemotherapy concomitant with radiotherapy to the primary tumour site and regional lymph nodes (Table II). Chemotherapy was given due to N stage above N1 or ECS. One patient received radiotherapy alone. One patient with pT1N1 underwent BLT and an ipsilateral left neck dissection, but had a close surgical margin (1 mm). He achieved macro- and microradical surgery with re-resection of the T site with TORS and neck dissection on the contralateral side, as he declined the option of radiotherapy. One more patient achieved radical surgery and did not wish to receive adjuvant therapy post-operatively. In the ‘undetected primary’ group, four patients received widespread bilateral radiation to the naso-, oro-, hypopharyngeal, and laryngeal mucosa along with N sites, and two patients also received concomitant cisplatin treatment. Lastly, two patients with pT1N1 stage had surgery with free lymph node margins and did not receive adjuvant treatment. Discussion While assessing the role of TORS in identification of the primary tumour with unilateral or bilateral lingual tonsillectomy, this study showed a detection rate of 54% in a cohort of 13 patients. All these patients had HPV-positive tumours and it is noteworthy that seven out of nine patients (78%) with HPV-positive lymph node metastases had their primary tumour detected.
M
M
F
M
60
60
59
69
52
68
79
3
4
5
6
7
8
9
10 67
11 50
12 64
13 64
M
M
F
F
M
M
F
BLT
BLT
BLT
BLT
BLT
BLT
BLT
BLT
BLT
BLT
BLT
right LT
BL
IL
No
IL
IL
BL
IL
No
No
IL
IL
No
IL
No
Yes
Yes
Yes
No
Yes
No
Yes
No
Yes
No
Yes
No
Neck Primary dissection tumour found?
NKSCC
NKSCC
NKSCC
NKSCC
KSCC
KSCC
NKSCC
NKSCC
NKSCC
NKSCC
NKSCC
NKSCC
NKSCC
Histology of nodal metastases
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
NKSCC
NKSCC
NKSCC
HSCC
NKSCC
NKSCC
NKSCC
+
+
+
+
+
+
+
p16-status in HPV- nodal Histology p16-status metastases status of primary in primary tumour tumor
+
+
+
+
+
+
+
HPV-status in primary tumour
TxN2bMx
TxN1Mx
TxN1Mx
TxN1Mx
TxN2bMx
TxN1Mx
TxN1Mx
TxN2aMx
TxN2aMx
TxN1Mx
TxN1Mx
TxN2aMx
TxN1Mx
cTNM
TxN2bM0
T1N1M0
T1N2bM0
T1N1M0
TxN2bM0
T1N1M0
TxN1M0
T2N2bM0
TxN2aM0
T1N1M0
TxN1M0
T1N1M0
TxN2aM0
pTNM
Free
Close
Free
Free
Close
Close
Close
Margin
0.9 cm
1.8 cm
0.4 cm
0.6 cm
2.2 cm
1.3 cm
0.9 cm
Tumour sizea
Yes
Yes
Yes
No
Yes
No
No
No
Yes
No
No
No
Yes
ECS
Largest tumour diameter. LT, lingual tonsillectomy; BLT, bilateral lingual tonsillectomy; IL, ipsilateral; BL, bilateral; TORS, transoral robotic surgery; HPV, human papillomavirus; NKSCC, non-keratinizing squamous cell carcinoma; KSCC, keratinizing squamous cell carcinoma; HSCC, hybrid squamous cell carcinoma; cTNM, clinical TNM classification; pTNM, pathological TNM classification; ECS, extracapsular spread.
a
M
55
2
right LT
51
1
M
Age Sex TORS type
#
Table I. Patients clinical and final pathology, including status on p16 and HPV.
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
Transoral robotic surgery and unknown primary 5
6
H. I. Channir et al.
Table II. Distribution of adjuvant therapy type and neck dissection. Variables
Detected primary Undetected group group
Neck dissection Ipsilateral
3
4
Bilateral
1
1
Lymph node extirpation 3
1
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
Adjuvant therapy Radiation alone
1
2
Chemoradiation
4
2
None
2
2
Ipsilateral radiation
0
0
Bilateral radiation
5
4
Radiation site
The patients in whom the primary tumour was not identified underwent unnecessary intervention with TORS and two patients unfortunately experienced short-term complications. The overall treatment strategy changed for all seven patients in whom the primary tumour was identified by offering TORS. Four of these patients would have been untreated at the primary tumour site, and thereby sub-optimally treated, if TORS had not been performed. Three of the patients in the ‘detected primary’ group received a reduced radiation field compared to the wide-field radiation they would have been offered due to a SCCUP with stage N2 or ECS. Two patients received only primary surgical treatment. The percentage of patients receiving chemotherapy was higher in the ‘detected primary’ group than in the ‘undetected primary’ group, mostly due to the advanced stages and ECS and not because of involved or close surgical margins. A larger study with long-term oncological outcomes will be required before conclusions can be drawn. Considering the possibility of de-escalation of the oncological treatment as a consequence of finding the primary site, there is a reasonable argument for implementing TORS as a diagnostic addition, as much is gained in reducing long-term morbidity associated with radiotherapy. Early identification of the primary site in the lingual tonsils may offer the possibility of primary radical surgery instead of primary radiation therapy, since the patients often present with very small and superficial tumours (down to 4 mm in this study), as showed previously [6,7]. The tumours are neither palpable nor macroscopically visible during the initial diagnostic work-up. TORS provides the potential for a wide BLT and definitive treatment of the primary tumour. It is difficult to state the overall importance of TORS
based on our small group of selected patients, but TORS seem to improve detection rates of the primary tumour when added, especially if the HPV status of the lymph node metastases is known. The procedure is considered safe, which was confirmed by the shortterm complications associated with TORS – in concordance with previous studies [6,9]. Since the first case report proposing TORS as a diagnostic tool for SCCUP [14], several authors have proposed a transoral robotic approach for the management of SCCUP [5–7,9,15]. Transoral laser microsurgery may also be an option for detecting primary tumours in SCCUP patients with comparable detection rates, and can be recommended in Institutes where robotic surgery is unavailable [9,16]. Studies using a TORS approach have found high detection rates (77–90%), but they used different definitions of a true ‘SCCUP’ syndrome; however, they still managed to identify the majority of primary tumours in the oropharyngeal palatine and lingual tonsils. Some studies have included palatine tonsils as their indication for TORS [5,6] which explains how the definition of a true ‘SCCUP’ syndrome varies in the literature [1]. If we focus on the exact T site and separate that from the published studies, which included 166 patients in total, there were 76 lingual tonsil carcinomas, corresponding to an identification rate of 46% which is comparable to our detection rate. This could encourage clinicians to increase the number of biopsies from the base of the tongue. This may be particularly relevant in institutes with no option for robotic surgery. Many institutes perform similar pre-diagnostic work-ups, but there are still differences – especially in the choice of imaging. We use whole-body PET/CT that often indicates primary focus and excludes further SCCUP work-up if the PET positive site is confirmed by histology. In Denmark, patients are diagnosed with SCCUP after the diagnostic workup including palatine tonsillectomy has failed to identify the primary tumour. Our institute has recently performed the largest unselected, consecutive cohort of tonsillar cancers (TSCCs) to date. This showed the relationship between HPV and oropharyngeal tumours and also showed an increasing incidence of HPV-positive TSCC in Denmark [4]. Based on this study, it can be expected that we will encounter a rising number of patients with SCCUP due to HPV pathogenesis; however, the detection rate of primary tumours has improved in recent years, mainly from adding PET/CT scans. We emphasize the use of cytological cell material obtained from FNA as a substrate for HPV detection. Those patients with HPV-positive cytology in neck metastases should be considered eligible for TORS lingual tonsillectomy, after a palatine tonsillectomy
Transoral robotic surgery and unknown primary with no detection of the primary tumour has been performed, as shown in our diagnostic work-up algorithm. In addition, the handling of the surgical specimens plays a major role in finding the primary tumour, by performing >20 tissue slides of 2–3 mm thickness followed by haematoxylin and eosin staining. TORS may take on a greater role in this diagnostic setting in the future management of patients with SCCUP.
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 07/18/15 For personal use only.
Conclusion Detection of the primary tumour using TORS combined with HPV DNA and p16 testing appears feasible for patients with SCCUP, who would otherwise be untreated (i.e. sub-optimally treated) at the primary site. We found that seven out of nine patients (78%) with HPV DNA/p16-positive lymph node metastases had a primary tumour in the lingual tonsils. The approach presented in this first Scandinavian study could potentially improve functional outcomes in the ‘detected primary’ group of patients by minimizing the radiation field to the pharyngeal axis. Declaration of interest: The authors report no conflicts of interests. The authors alone are responsible for the content of the paper. They certify that they have no affiliation with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials described in the paper. In 2013, the department received a donation from a non-profit organization to purchase a daVinci SI HD robot.
References [1] Straetmans J, Vent J, Lacko M, Speel E-J, Huebbers C, Semrau R, et al. Management of neck metastases of unknown primary origin united in two European centers. Eur Arch Otorhinolaryngol 2015;272:195–205. [2] Danish Head and Neck Cancer Group [Internet]. National guidelines regarding carcinoma metastases on the neck from unknown primary tumor, 2013. Available from: https://www. dahanca.oncology.dk/. [Last accessed 25 February 2015]. [3] Rinaldi V, Pagani D, Torretta S, Pignataro L. Transoral robotic surgery in the management of head and neck tumours. Ecancermedicalscience 2013;7:359.
7
[4] Garnaes E, Kiss K, Andersen L, Therkildsen MH, Franzmann MB, Filtenborg-Barnkob B, et al. A high and increasing HPV prevalence in tonsillar cancers in Eastern Denmark, 2000-2010: the largest registry-based study to date. Int J Cancer 2014;4:2000–10. [5] Durmus K, Rangarajan SV, Old MO, Agrawal A, Teknos TN, Ozer E. Transoral robotic approach to carcinoma of unknown primary. Head Neck 2014;36:848–52. [6] Patel SA, Magnuson JS, Holsinger FC, Karni RJ, Richmon JD, Gross ND, et al. Robotic surgery for primary head and neck squamous cell carcinoma of unknown site. JAMA Otolaryngol. Head Neck Surg 2013;139:1203–11. [7] Byrd JK, Smith KJ, de Almeida JR, Albergotti WG, Davis KS, Kim SW, et al. Transoral robotic surgery and the unknown primary: a cost-effectiveness analysis. Otolaryngol Head Neck Surg 2014;150:976–82. [8] Mehta V, Johnson P, Tassler A, Kim S, Ferris RL, Nance M, et al. A new paradigm for the diagnosis and management of unknown primary tumors of the head and neck: a role for transoral robotic surgery. Laryngoscope 2013; 123:146–51. [9] Graboyes EM, Sinha P, Thorstad WL, Rich JT, Haughey BH. Management of human papillomavirus-related unknown primaries of the head and neck with a transoral surgical approach. Head Neck 2014; doi: 10.1002/hed.23800. [10] Weinstein GS, Quon H, Newman HJ, Chalian JA, Malloy K, Lin A, et al. Transoral robotic surgery alone for oropharyngeal cancer. Arch Otolaryngol Head Neck Head Neck Surg 2012;138:628–34. [11] Lewis JS, Chernock RD, Ma X-J, Flanagan JJ, Luo Y, Gao G, et al. Partial p16 staining in oropharyngeal squamous cell carcinoma: extent and pattern correlate with human papillomavirus RNA status. Mod Pathol 2012;25: 1212–20. [12] El-Mofty SK, Zhang MQ, Davila RM. Histologic identification of human papillomavirus (HPV)-related squamous cell carcinoma in cervical lymph nodes: a reliable predictor of the site of an occult head and neck primary carcinoma. Head Neck Pathol 2008;2:163–8. [13] Lewis JS, Thorstad WL, Chernock RD, Haughey BH, Yip JH, Zhang Q, et al. p16 positive oropharyngeal squamous cell carcinoma:an entity with a favorable prognosis regardless of tumor HPV status. Am J Surg Pathol 2010;34:1088–96. [14] Abuzeid WM, Bradford CR, Divi V. Transoral robotic biopsy of the tongue base: a novel paradigm in the evaluation of unknown primary tumors of the head and neck. Head Neck 2013;35:E126–30. [15] Mehta V, Johnson P, Tassler A, Kim S, Ferris RL, Nance M, et al. A new paradigm for the diagnosis and management of unknown primary tumors of the head and neck: a role for transoral robotic surgery. Laryngoscope 2013; 123:146–51. [16] Nagel TH, Hinni ML, Hayden RE, Lott DG. Transoral laser microsurgery for the unknown primary: role for lingual tonsillectomy. Head Neck 2014;36:942–6.
