Elective regional lymphadenectomy for advanced auricular squamous cell carcinoma

Abstract

AIM: To investigate the rate of occult lymph node disease in elective parotidectomy and neck dissection specimens in patients with advanced auricular cutaneous squamous cell carcinoma (cSCC).

METHODS: At a single institution, from 2000 to 2010, 17 patients with advanced auricular cSCC were considered high risk for occult regional parotid and/or neck nodal metastases and, thus, underwent an auriculectomy and elective regional lymphadenectomy (parotidectomy and/or neck dissection). Indications for elective regional lymphadenectomy were large tumor size, locally invasive tumors, post-surgical and post-radiation recurrence, and being an immunosuppressed patient. We determined the presence of microscopic disease in the regional (parotid and neck dissection) pathology specimens.

RESULTS: There were 17 advanced auricular cSCC patients analyzed for this study. Fifteen (88%) patients were men. The average age was 69 (range: 33 to 86). Ten (59%) patients presented with post-surgical recurrence. Five (29%) patients presented with post-radiation recurrence. Four (24%) patients presented with both post-surgical and post-radiation recurrence. Four (24%) patients were immunosuppressed (2 (12%) were liver transplant patients, 2 (12%) were chronic lymphocytic leukemia patients, and 1 (6%) was both). The subsite distribution of cSCC included helix (3, 18%), antihelix (2, 12%), conchal bowl (7, 41%), tragus (2, 12%), and postauricular sulcus (3, 18%). Four (24%) patients presented with multifocal auricular cSCC. No patients had bilateral disease. All patients were confirmed to have cSCC on final pathology. The tumors were well (5, 29%), moderately (10, 59%), and poorly (2, 12%) differentiated SCC. The average size of the cSCC tumor was 2.9 cm (range: 1.7 to 7 cm). Twelve (70%) tumors were greater than 2 cm. Six (35%) patients underwent partial auriculectomy. Eleven (65%) patients underwent total auriculectomy. Eight (47%) patients underwent elective parotidectomy and elective neck dissections; 3 (18%) underwent only elective parotidectomy; 3 (18%) underwent only an elective neck dissection; 2 (12%) underwent an elective parotidectomy and therapeutic neck dissection; and 1 (6%) underwent a therapeutic parotidectomy and an elective neck dissection. None of the elective parotidectomy or neck dissection specimens were found to contain any malignant disease. All therapeutic parotidectomy and neck dissection specimens contained metastatic SCC. Fourteen (82%) underwent parotidectomy. Of these, 10 (71%) underwent superficial parotidectomy whereas 4 (29%) underwent total parotidectomy. Fourteen (82%) underwent neck dissections [levels II/Va (1, 7%), levels II/III/Va (2, 14%), levels I/II/III/Va (2, 14%), and complete levels I-V (9, 64%)]. Three (18%) underwent concurrent temporal bone resections for tumor extension from the auricle. The average follow-up for our patients was 44 mo (range: 4 to 123 mo). At the time of the review, 6 (35%) patients were alive and 11 (65%) had passed away.

CONCLUSION: This study suggests that, in patients with advanced auricular cutaneous SCC, elective regional lymphadenectomy is not necessary. However, furtherprospective studies are necessary to assess the necessity.

Key Words: Advanced auricular squamous cell carcinoma, Elective lymphadenectomy, Elective parotidectomy, Elective neck dissection, Occult regional metastases

INTRODUCTION

Advanced auricular cutaneous squamous cell carcinoma (cSCC) has conventionally been treated with partial or total auriculectomy accompanied by a regional lymphadenectomy, which often includes a parotidectomy and/or a neck lymph node dissection for clinically apparent or occult lymph node disease. However, this treatment paradigm is controversial and without solid data[1]. Auricular cSCC is generally defined as advanced when it has one or more of the following features: large tumor size, increased depth, locally invasive tumors (cartilage and beyond), multifocal disease, clinically apparent lymph node metastases, post-surgical recurrence, post-radiation recurrence, and when the patient is immunosuppressed[2-5]. In patients with auricular cSCC, the practice of performing therapeutic lymphadenectomy follows the logical reasoning of excising known disease whereas the practice of performing elective lymphadenectomy is done out of perceived increased risk of occult nodal disease and a known higher mortality rate associated with advanced auricular cSCC[2,6,7].

The skin of the auricle has lymph drainage that flows to the pre-auricular/parotid, infra-auricular/neck, and post-auricular nodal basins[8,9]. Several studies claim a higher rate of metastasis with auricular cSCC compared to other sites on the head and neck[10-14]. Studies show that, for external auditory canal and auricular cSCC, the range of involvement in the parotid or neck lymph nodes is 7.9% to 17.5%[5,7,14,15]. The rate of regional metastases for other sun exposed skin sites (including the head and neck) ranges from 2% to 5%[6]. The 5 years mortality rate directly attributable to regional metastases from cSCC can be as high as 56.6% to 66.7% even after combined surgical and radiation therapy[6,7].

With advanced auricular cSCC, elective lymphadenectomy is performed in these cases in an effort to maximize therapy, increase the chance of cure, and reduce the chance of locoregional recurrence. However, the incidence of occult parotid and neck lymph node malignant disease is currently unknown. Thus, we developed a study of patients with advanced auricular cSCC and clinically negative parotid and neck lymph nodes to determine the rate of occult microscopic lymph node disease in the elective parotidectomy and neck dissection.

MATERIALS AND METHODS

The Committee on Human Research at University of California, San Francisco approved this study. All patients gave informed consent for the operation they underwent.

We performed a retrospective chart review at our single institution from a database of all non-melanoma skin cancer patients treated surgically from 1997 to 2010. We found 17 patients with auricular cSCC and no suspicious regional (parotid or neck) lymphadenopathy on physical exam or imaging [by computed tomography (CT) or magnetic resonance imaging (MRI)] who underwent auriculectomy (partial or total) and elective regional lymphadenectomy (superficial or total parotidectomy and/or neck dissection to any extent) from 2000 to 2010. We included only patients with auricular cSCC confined to or originating from the helix, antihelix, conchal bowl, tragus, antitragus, postauricular sulcus, or lobule and those extending from these areas to the external auditory canal or temporal bone. We excluded patients who had isolated external auditory canal or temporal bone cSCC without auricular involvement or cSCC from other head and neck subsites.

Indications for elective regional lymphadenectomy retrospectively appeared to be large tumor size, locally invasive tumors (into cartilage, the external auditory canal, parotid, or the temporal bone), multifocal auricular disease (multiple lesions involving nonadjacent auricular subunits), recurrence after previous surgery, recurrence after previous radiation, and being immunosuppressed (including transplant or chronic lymphocytic leukemia patients). Additionally, we found elective neck dissections performed for suspicious parotid lymphadenopathy and elective parotidectomy performed for suspicious neck lymphadenopathy.

For each patient, we determined the following factors: gender, age, prior surgical or radiation treatment, immunosuppression, the preoperative radiology performed (CT or MRI) to determine lymph node disease, the extent of a surgical resection they underwent (partial vs total auriculectomy, superficial vs total parotidectomy, extent of neck dissections (levels 1, 2, 3, 4, and/or 5), temporal bone resection, and other procedures performed), the final surgical pathology results for the auricle tumor excision, the parotidectomy, and the neck lymphadenectomy, the size of the primary auricular tumor, the grade of the malignancy, if they subsequently underwent post-operative radiation treatment, and the length of follow-up from the surgical treatment, and their mortality/survival.

RESULTS

Table 1 displays the characteristics and treatment experience in the 17 advanced auricular cSCC patients analyzed for this study.

Table 1 Characteristics and treatment experience of the 17 advanced auricular cutaneous squamous cell carcinoma patients.

Pt.	Age (yr)	Sex	Tumor size (cm)	Sub-site	Prior Rx	Auriculx type	Parotidx type	Neck diss.extent	Parotid, neck diss. path	Post-op XRT	F/u (mo)
1	71	M	3.0	AH	S, XRT	Total	Elective	-	Neg,	-	28
							Superficial		-
2	86	M	2.5	AH	-	Partial	-	Elective	-,	-	61
								(II/III/Va)	Neg		(D)
3	69	M	4.1	P	S, XRT	Total	Elective	Elective	Neg,	-	80
							Superficial	(complete)	Neg		(D)
4	69	M	1.6	C	S	Total	Elective	Elective	Neg,	-	43
							Superficial	(I/II/III)	Neg		(L)
5	56	M	1.3	C	S	Total	Elective	Elective	Neg,	Yes	48
							Superficial	(complete)	Neg		(L)
6	68	M	7.0	P	-	Total	Elective	Elective	Neg,	Yes	70
							Total	(complete)	Neg		(D)
7	78	M	6.0	T	S, XRT	Total	Elective	Elective	Neg,	-	10
							Total	(complete)	Neg		(D)
8	33	F	1.9	C	-	Total	Elective	Therapeutic	Neg,	-	61
							Total	(complete)	Pos		(D)
9	70	M	1.7	H	S	Partial	Therapeutic	Elective	Pos,	Yes	96
							Total	(complete)	Neg		(L)
10	71	M	1.7	C	-	Partial	Elective	Therapeutic	Neg,	Yes	123
							Superficial	(complete)	Pos		(L)
11	86	M	2.0	P	S, XRT	Total	Elective	-	Neg,	-	53
							Superficial		-		(D)
12	69	F	2.5	T	-	Partial	Elective	Elective	Neg,	Yes	5
							Superficial	(I/II/III)	Neg		(D)
13	61	M	3.0	H	S	Partial	Elective	Elective	Neg,	Yes	LTF
							Superficial	(complete)	Neg
14	65	F	2.4	C	XRT	Total	-	Elective	-,	-	10
								(complete)	Neg		(D)
15	78	M	4.2	C	S	Total	Elective	Elective	Neg,	Yes	5
							Superficial	(II/III/Va)	Neg		(L)
16	84	M	2.0	H	-	Total	-	Elective	-,	-	11
								(II/Va)	Neg		(D)
17	65	M	3.0	C	-	Total	Elective	-	Neg,	Yes	4
							Superficial		-		(L)

M: Male; F: Female; AH: Antihelix; P: Posterior sulcus; C: Concha bowl; T: Tragus; H: Helix; Rx: Treatment; S: Surgery; XRT: Radiation treatment; D: Deceased; L: Live; LTF: Lost to follow-up.

Fifteen (88%) patients were men. The average age was 69 years (range: 33 to 86 years). Ten (59%) patients presented with post-surgical recurrence. Five (29%) patients presented with post-radiation recurrence. Four (24%) patients presented with both post-surgical and post-radiation recurrence. Four (24%) patients were immunosuppressed 2 (12%) were liver transplant patients, 2 (12%) were chronic lymphocytic leukemia patients, and 1 (6%) was both). Six (35%) patients had preoperative CT; 6 (35%) had preoperative MRI; and 5 (29%) had no preoperative imaging. The subsite distribution of cSCC included helix (3, 18%), antihelix (2, 12%), conchal bowl (7, 41%), tragus (2, 12%), and postauricular sulcus (3, 18%). Four (24%) patients presented with multifocal auricular cSCC. No patients had bilateral disease.

All patients were confirmed to have cSCC on final pathology. The tumors were well (5, 29%), moderately (10, 59%), and poorly (2, 12%) differentiated SCC. The average size of the cSCC tumor was 2.9 cm (range: 1.7 to 7 cm). Twelve (70%) tumors were greater than 2 cm. Six (35%) patients underwent partial auriculectomy. Eleven (65%) patients underwent total auriculectomy.

Eight (47%) patients underwent elective parotidectomy and elective neck dissections; 3 (18%) underwent only elective parotidectomy; 3 (18%) underwent only an elective neck dissection; 2 (12%) underwent an elective parotidectomy and therapeutic neck dissection; and 1 (6%) underwent a therapeutic parotidectomy and an elective neck dissection. Six surgeons including Ryan WR, Wang SJ performed the operations (along with 3 different temporal bone surgeons).

None of the elective parotidectomy or neck dissection specimens were found to contain any malignant disease. All therapeutic parotidectomy and neck dissection specimens contained metastatic SCC.

Fourteen (82%) underwent parotidectomy. Of these, 10 (71%) underwent superficial parotidectomy whereas 4 (29%) underwent total parotidectomy. Fourteen (82%) underwent neck dissections [levels II/Va (1, 7%), levels II/III/Va (2, 14%), levels I/II/III/Va (2, 14%), and complete levels I-V (9, 64%)]. Three (18%) underwent concurrent temporal bone resections for tumor extension from the auricle. One (6%) patients underwent a concurrent condyle resection and infratemporal lymph node dissection (for therapeutic purposes).

Eight (47%) underwent post-operative radiation.

The average follow-up for our patients was 44 mo (4 to 123 mo). At the time of the review, 6 (35%) patients were alive and 11 (65%) had passed away.

DISCUSSION

This study shows a 10-year experience by 6 different surgeons at one institution of 17 patients with advanced auricular cSCC all of whom had negative elective lymphadenectomy specimens on final surgical pathologic analysis. This absolute result calls into the question the need for elective parotidectomy and neck dissection in the cases of advanced auricular cSCC.

This study was partly inspired by and corroborates Osborne et al[16] in their study of advanced auricular cSCC and elective parotidectomy. Osborne et al[16] found that, in 19 patients, none of the elective parotidectomy specimens performed for advanced auricular cSCC had any positive final surgical pathology. We found the same results with elective parotidectomy and, uniquely, the same results for the elective neck dissections.

Elective lymphadenectomy is performed for any cancer of the head and neck when the predicted risk of occult nodal disease reaches a certain threshold of 15% to 20%. An elective lymphadenectomy is performed to avoid unsalvageable neck disease and improve survival accepting the fact that some unnecessary surgery will be performed. However, our data showing no nodal metastases in 12 patients who underwent elective neck dissection suggests that elective regional lymphadenectomy for auricular cSCC may not necessarily be beneficial.

Limiting the use of elective parotidectomy and neck dissection could reduce the cost, time, and potential morbidity associated with these operations. The total time associated with the parotidectomy and neck dissection includes operating time, hospital stay, and recovery at home. Both procedures require drain placement and a hospitalization of one to several days. Parotidectomy surgery carries the risks of temporary or permanent facial nerve injury (with possible corneal keratitis, facial droop, asymmetric smile, and oral incompetence), hematoma, seroma, salivoma, cellulitis, abscess, skin flap loss, gustatory sweating (Frey’s Syndrome), unwanted indentation in the face, unwanted incision, and perincisional, great auricular, and auriculotemporal nerve injury-associated numbness. Neck dissection, depending on the extent performed, carries the addition risks of unwanted neck sensory dysfunction, neck soft tissue defect if the sternocleidomastoid muscle is removed, chylous fistula/leak, and motor nerve injury to the spinal accessory, marginal mandibular, hypoglossal, vagus, superior laryngeal, phrenic, sympathetic, and brachial plexus nerves. In addition, conceivably, a total auriculectomy could be performed under local anesthesia preventing the need for a general anesthetic and intubation all together in select cases.

The data on cSCC in this study correlates with the body of evidence showing the lack of benefit for elective lymphadenectomy for higher stage malignant melanoma. Several studies show with different thicknesses of malignant melanoma in different parts of the body that elective lymphadenectomy (including parotidectomy and neck dissections in some studies) brought no measurable increase in locoregional control, disease-specific survival, or overall survival[17-19]. Thus, possibly a similar management protocol could be relevant to advanced cases of cSCC with regards to the use of sentinel lymph node biopsy for assessing the need for elective lymphadenectomy[20].

This study is by no means a complete denouncement on the use of elective lymphadenectomy for advanced auricular or head and neck cSCC. There is the risk of leaving occult disease in a patient when a regional lymph node bed is left untreated. Subclinical neck malignancy lymphadenopathy rates in head and neck cSCC are reported as being as high as 35% in 2 studies[21,22]. Two other studies show rates being lower: at 16%[23,24]. Freedlander showed that, in auricular cSCC, 85% of the metastases to the parotid or neck occurred within 1 year of initial auricular excision[14]. Nonetheless, Byers et al[3] found no difference in survival between elective and therapeutic neck dissection but did not report the numbers of patients in each category.

A more thorough analysis of the primary tumor final pathology may be the deciding factor for the need for elective lymphadenectomy in auricular cSCC and possibly cSCC for other head and neck sites. In a recent study, Clark et al[5] showed that tumors with a depth of invasion > 8 mm had a 56.2% risk of metastatic spread and those with a depth of invasion between 2 and 8 mm and with evidence of cartilage destruction, lymphovascular invasion or a non-cohesive invasive front have a 24.2% risk of metastatic spread. In a meta-analysis, Rowe et al[6] showed an increased likelihood of regional metastases with tumor size over 2 cm, depth of invasion over 4 mm, poorly-differentiated grade, perineural invasion, and local recurrence. Given the higher percentages (being over 15%-20%) in auricular cSCC with these features, we agree that elective regional lymphadenectomy is tempting. However, the retrospective nature of these studies calls into question the time relationship between the development of the primary cancer and regional metastases.

The withholding of elective lymphadenectomy for cSCC does not remove the need for close observation. In a patient with poor prognostic risk factors, more frequent follow-up with careful evaluation of the regional lymph node basins is certainly important if an observational strategy is to be implemented. Ultrasound in particular is a promising and accurate method for surveillance of the parotid and neck in these patients which may be used in conjunction or possibly in lieu of the cross-sectional imaging of CT and MRI[25,26].

Our study has limitations. This is a retrospective series at only one institution involving multiple surgeons with different treatment philosophies over a 10-year period with a small sample size. With heterogenous operations being performed and no standard surgical treatment regimen (superficial or total parotidectomy and different extents of neck dissections being performed), occult lymph nodes could have been missed. The sample size is small reducing its generalizability but does reflect the rare character of this particular clinical scenario. There is no control group to compare regional recurrence or survival rates. However, the use of a comparison untreated group is beyond the scope of the goal of determining the rate of regional nodal basin occult disease.

A prospective randomized controlled trial of advanced cSCC with concern for occult regional metastases would be ideal but difficult to carry out given the multiple subsites of the head and neck and the rare presentation of the disease in this setting. Nevertheless, further studies are necessary to further clarify the extent of occult lymph node involvement in advanced auricular cSCC and other sites in the head and neck and a process by which to appropriately risk stratify patients into undergoing elective lymphadenectomy.

This small sample suggests that, in patients with advanced auricular cSCC, elective regional lymphadenectomy may not be necessary. Larger multi-institutional prospective studies are necessary to assess the necessity of elective regional lymphadenectomy for advanced auricular squamous cell carcinoma.

ACKNOWLEDGMENTS

We would like to thank the database managers who helped create the head and neck non-melanoma skin cancer database at our institution.

COMMENTS

Background

Advanced auricular cutaneous squamous cell carcinoma (cSCC) has conventionally been treated with partial or total auriculectomy accompanied by a regional lymphadenectomy, which often includes a parotidectomy and/or a neck lymph node dissection for clinically apparent or occult lymph node disease. With advanced auricular cSCC, elective lymphadenectomy is performed in these cases in an effort to maximize therapy, increase the chance of cure, and reduce the chance of locoregional recurrence. However, the incidence of occult parotid and neck lymph node malignant disease is currently unknown. Thus, authors developed a study of patients with advanced auricular cSCC and clinically negative parotid and neck lymph nodes to determine the rate of occult microscopic lymph node disease in the elective parotidectomy and neck dissection.

Research frontiers

Several studies claim a higher rate of metastasis with auricular cSCC compared to other sites on the head and neck. Studies show that, for external auditory canal and auricular cSCC, the range of involvement in the parotid or neck lymph nodes is 7.9% to 17.5%. The 5 years mortality rate directly attributable to regional metastases from cSCC can be as high as 56.6% to 66.7% even after combined surgical and radiation therapy. Current research is attempting to further delineate what surgical management is needed in this disease process.

Innovations and breakthroughs

Very little research has been done in evaluating the incidence of occult nodal disease in advanced auricular cutaneous SCC. Historically, regional lymphadenectomy has been performed electively in these cases. This study concludes that regional lymphadenectomy may not be necessary and hopes to spawn further larger prospective trials.

Applications

This small sample suggests that, in patients with advanced auricular cSCC, elective regional lymphadenectomy may not be necessary. Larger multi-institutional prospective studies are necessary to assess the necessity of elective regional lymphadenectomy for advanced auricular squamous cell carcinoma.

Terminology

Advanced auricular SCC is defined as advanced when it has one or more of the following features: large tumor size, increased depth, locally invasive tumors (cartilage and beyond), multifocal disease, clinically apparent lymph node metastases, post-surgical recurrence, post-radiation recurrence, and when the patient is immunosuppressed.

Peer review

This retrospective study focuses on 17 patients with auricular squamous cell carcinoma. Authors determined the rate of occult lymph node disease in elective parotidectomy and neck dissection specimens. Manuscript is well written.