- Email: [email protected]
Nonmelanoma cutaneous malignancy with regional metastasis
Nonmelanoma cutaneous malignancy with regional metastasis ANGELA CHU,
MD,
and J. DAVID OSGUTHORPE,
MD,
Charleston, South Carolina
OBJECTIVE AND STUDY DESIGN: We performed a retrospective study of 28 patients to evaluate the management and outcome of regional metastasis from nonmelanoma cutaneous malignancies. RESULTS: There were 25 squamous cell, 1 basal cell, and 2 eccrine carcinomas. The most common primary locations were the temple, ear, and cheek. Median interval from primary resection to regional metastasis was 9 months, and such were commonly of the parotid, level IB, and level IIA nodes. Twenty-four patients underwent parotidectomy and/or neck dissection; 19 patients, postoperative radiation. Ten patients (36%) died from the disease, 2 patients (7%) have persisting disease, 5 (18%) have died from intercurrent disease, and 11 (39%) have no evidence of disease (minimum follow-up, 22 months; median, 34 months). CONCLUSIONS AND SIGNIFICANCE: Nonmelanoma cutaneous malignancies with regional metastasis have a poor prognosis despite aggressive therapy. When high-risk characteristics are detected, examination of the nearest “sentinel node,” prophylactic lymphadenectomy, and/or regional irradiation might be justified; further study is warranted. (Otolaryngol Head Neck Surg 2003;128:663-73.)
B asal cell carcinoma (BCC) and squamous cell carcinoma (SCC) comprise the preponderance of nonmelanoma cutaneous malignancies, the most common cancers in the United States. BCC occurs more frequently than SCC, by a ratio of 4:1. The most frequent location of these malignancies is in the head and neck region of white men in their From the Department of Otolaryngology–Head and Neck Surgery, Medical University of South Carolina. Presented at the Annual Meeting of the American Academy of Otolaryngology–Head and Neck Surgery, San Diego, CA, September 22-25, 2002. Reprint requests: J. David Osguthorpe, MD, Otolaryngology– Head and Neck Surgery, PO Box 250550, 135 Rutledge Ave, Suite 1130, Charleston, SC 29425-0550; e-mail, [email protected]. Copyright © 2003 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. 0194-5998/2003/$30.00 ⫹ 0 doi:10.1016/S0194-5998(03)00226-2
fifth to sixth decade of life, which reflects the common risk factors of chronic sun exposure and fair complexion. Other predisposing factors include burn scars, cutaneous exposure to certain chemicals (eg, arsenic, hydrocarbons), immunosuppression, and genetic factors (eg, xeroderma pigmentosum, basal cell nevoid syndrome, etc). Although uncommon, both SCC and BCC have the potential to metastasize. The regional metastatic rate for BCC is reported from 0.0028% to 0.5%,1-3 whereas the reported rate for SCC ranges from 1% to 12.5%.4-6 Distant metastasis is rare, especially without locoregional disease. Regional metastasis from cutaneous primary lesions in the head and neck region most frequently occur in the parotid nodes, followed by the upper cervical nodes.7,8 This pattern reflects the favored primary sites of BCC and SCC, including the ear, preauricular region, and upper face whose lymphatics drain first to the parotid nodes and then onto the neck nodes.5 Drainage patterns based on melanomas are presumed to be applicable to BCC and SCC as depicted (Fig 1). That is, parotid involvement usually arises from the anterior scalp, temple, face, and ear and much less frequently from the posterior scalp and neck.9 Furthermore, minimal involvement of level VA nodes occurs with lesions of the anterior scalp, temple, face, and ear, whereas level IA is rarely involved with lesions of the posterior scalp and neck.9 In general, most regional metastatic SCC becomes clinically evident within the first 2 years after treatment of the primary site.5,6 For BCC this time period varies from less than 1 year to 45 years, with a median time of 6 to 12 years.2,3,7 Regional metastasis is concurrent with persistence or recurrence at the primary location in approximately 42% of cases.4 Thus, failure of local control is an important risk factor for regional metastasis.3,10 Curative treatment options for regional metastasis include surgery and irradiation. Many studies have demonstrated improved disease control with combined surgery and irradiation versus surgery 663
664 CHU and OSGUTHORPE
Otolaryngology– Head and Neck Surgery May 2003
Fig 1. Lymphatic drainage patterns of the skin of the face, scalp, and upper neck. (From Jackson G, Ballantyne A: Am J Surg 1981;142:464-9, with permission.)
alone, particularly in the management of the parotid nodes.8,11,12 Cassisi et al12 demonstrated that a superficial parotidectomy alone for metastatic cutaneous SCC had a 75% recurrence rate compared with 20% for those treated with combined therapy. This high recurrence can be at least in part explained by the anatomy of lymphatics that traverse the parotid, with the intraglandular nodes located lateral to the posterior facial vein. Depending on the relation of the facial nerve to the posterior facial vein, the former lateral to the latter in about 75% of cases, involved nodes can be medial to the facial nerve and hence be left behind in a superficial parotidectomy.12 Parotidectomy or parotid area irradiation is sometimes performed prophylactically in addition to a neck dissection for positive level I or II cervical nodes associated with primary lesions in the upper lateral face or posterior cheek, as lymphatic spread from these areas to the neck assumes passage through the parotid nodes. Khurana et al7 found an 18% rate of occult parotid metastasis in patients presenting solely with cervical metastasis from cutaneous primary lesions of the face.
Similarly, cervical metastatic disease is treated with surgery, irradiation, or both. Taylor et al11 reported that in clinically positive necks, a neck dissection with postoperative irradiation was favored, with a 89% control rate compared with 63% rate with surgery alone. Furthermore, in clinically negative necks in patients with parotid metastasis, neck dissection or irradiation to the neck might be prudent given the 24% to 35% occult cervical metastatic rate, according to Jackson and Ballantyne8 and O’Brien et al.13 Other rare nonmelanoma skin carcinomas include Merkel cell and eccrine sweat gland carcinomas. Merkel cell carcinoma, first described in 1972 by Toker, is derived from Merkel cells (neuroendocrine cells) located at the dermoepidermal junction. Such are rare, with only approximately 1100 cases reported, but are very aggressive, with high local recurrence (12% to 50%), regional metastatic (17% to 76%), and distant metastatic (12% to 50%) rates.14 If aggressively treated with surgery at an early stage, Allen et al15 reported a 75% 5-year survival rate. However, in the setting of locoregional recurrence or distant metastasis,
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
2-year survival rates are approximately 35% and 17%, respectively. Eccrine (sweat gland) carcinomas are likewise rare malignancies, occurring mostly in the extremities and trunk, with surgical excision the current standard treatment.16 The effectiveness of adjuvant therapies, including radiotherapy, chemotherapy, or antiestrogen therapy, has been variable, mainly in case reports, with no major series results to date.16,17 The objective of this study is to evaluate the management and outcome in our experience with patients at the Medical University of South Carolina with regional metastasis from nonmelanoma cutaneous carcinomas. METHODS A retrospective chart review identified 28 patients referred to the junior author (J.D.O.) between 1984 and 2000 for the treatment of regional metastasis from a nonmelanoma cutaneous malignancy. As summarized in Table 1, information about sites of primary lesions, histology, metastatic sites, and both prior and current treatments (at our institution or elsewhere) were assembled. The interval to metastatic disease was determined as the time from first treatment of a primary lesion to the detection of regional metastasis. Follow-up information was obtained from medical records or telephone interviews. Patients were classified in 4 follow-up categories: dead of disease (DOD), dead of intercurrent disease (DID), alive with disease (AWD), and no evidence of disease (NED). Survivors were followed a minimum of 22 months, and a median of 34 months, and all patients were accounted for. RESULTS The median age of the 28 patients was 59 years. All were white, and 23 were male. Histologically, lesions included 25 SCCs, 1 BCC, and 2 eccrine carcinomas. The most common primary locations were, in order, of the temple, ear, and cheek. The majority of the primary lesions were treated by fresh tissue technique micrographic surgery (22 patients, or 79%, at our institution or elsewhere), with 3 patients having had at least 2 micrographic surgeries for local recurrence. A median of 9 months separated treatment of the primary lesions from clinical evidence of metastatic disease,
CHU and OSGUTHORPE 665
which occurred, in order, in the parotid, level IB, and level IIA nodes. An example of parotid metastasis is shown in the left parotid (Fig 2). Three patients were noted to have metastatic disease involving the external jugular nodes, from temple, posterior cheek, and posterior superior neck primary sites. Two examples of metastasis to the left external jugular nodes are demonstrated (Fig 3). Thirteen patients (46%) presented with parotid nodal disease, whereas 10 (36%) presented with only cervical metastasis. Five patients (18%) had both parotid and cervical metastatic disease. Twenty-four patients received surgical treatments, including parotidectomy and/or selective or modified radical neck dissection. According to pathology results, 7 patients met N1 neck criteria (and had selective neck dissection); 3 patients, N2a, and 14 patients, N2b (had modified radical neck dissection in the latter 2 categories). Seven patients had positive surgical margins, 3 patients had lymphovascular involvement (Fig 4), and 8 had perineural spread at the primary sites (Fig 5); 5 patients had extracapsular spread of tumor from involved parotid or cervical nodes at the time of parotidectomy and/or neck dissection. Four patients did not receive surgical intervention because of unresectability (intracranial perineural spread in 2 patients), poor medical status, or patient refusal (1 patient). Nineteen patients received postoperative irradiation, ranging from 50 to 70 Gy. Of these patients, 1 received irradiation as primary treatment, and 2 patients as palliation. Two patients received chemotherapy for palliation. Follow-up data disclosed 43% (12 patients) either AWD (2 patients) or DOD (10 patients), with distant metastasis the most common situation. Eighteen percent (5 patients) have died of intercurrent disease (pneumonia in 2, noncutaneous malignancy in 2, cardiac disease in 1); and 39% (11 patients, followed a median of 57 months) are NED. Of the 10 patients who were DOD, 4 had positive margins at the primary site (2 had recurrent disease despite micrographic surgery), and 3 of these patients also had perineural involvement. Clinically, perineural spread intracranially via V3 also occurred in 2 patients. Four of these patients had primary sites on the cheek, and 2 on the auricle.
Otolaryngology– Head and Neck Surgery May 2003
666 CHU and OSGUTHORPE
Table 1. Patient data
Primary site
Histology
Treatment
Metastasis site
Interval to metastasis (mo)
L cheek
SCCE
Moh’s
L exterior jugular and pulmonary metastasis (surgery and radiation for recurrence)
33
L forehead
SCCE
Moh’s
L parotid
6
L lowerlip
SCCE
Wide excision
L level I
24
L temple
SCCE
Wide excision
R parotid, external jugular level II
23
R cheek
SCCE
Moh’s
R parotid
0
L lower lip
SCCE
Moh’s
L level I, II
9
L superior helix
SCCE
Moh’s
L parotid
12
R cheek
SCCE
Curettage ⫻ 3, then wide excision
R parotid
25
L postneck
SCCE
Moh’s
L external jugular
36
L auricle
SCCE
Moh’s
L parotid
12
R superior helix
SCCE
Radiation
R occipital and level V
R temple
SCCE
Moh’s
R parotid and level II
L temple
SCCE
Wide excision
R parotid w/recurrence status post radiation
L auricle
SCCE
Moh’s
L parotid
6
22 1
10
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
CHU and OSGUTHORPE 667
Table 1. (Continued)
Surgical treatment
Pathology
Irradiation (Gy)
Chemotherapy
cisplatin 5FU
Follow-up since treatment for metastasis (mo)
DOD (24) pulmonary metastasis
Total parotid
N1 LN (1) ⫹
59
NED (41)
L MND
N2b level I LN (2) ⫹
63
NED (57)
Total parotid, R MND
N2b parotid LN (1) ⫹ exterior jugular LN (1) ⫹ level II LN (1) ⫹
66
DOD (17) thoracic spinal metastasis
Superficial parotid, R SND
N1 parotid LN (1) ⫹ perineural ⫹ vascular ⫹
L MND
N2b level I LN (1) ⫹/ECS level II (2) ⫹ margins ⫹ perineural ⫹ lymphovascular ⫹
64
DOD (20) perineural intracranial spread via V3
Total parotid, L SND
N1 parotid LN ⫹ margins ⫹ perineural ⫹
60
DOD (21) pulmonary metastasis
59 (palliation)
DOD (7) perineural intracranial spread via V3
Unresectable
DID (10) renal cancer
L SND
N2b exterior jugular LN (2) ⫹
60
NED (58)
Superficial parotid, R SND
N2a parotid LN (1) ⫹
60
DID (20)
R SND
N2b LN (14) ⫹/ECS muscle ⫹ SMGld ⫹/endoneural
64
DID (2) lymphoma recurrence
Refused all treatment Total parotid, local reexcision Medically unsuitable for surgery
DID (20) pneumonia N2b parotid LN (2) ⫹ margin ⫹
64
AWD (45)
25 (incomplete)
DOD (12)
Otolaryngology– Head and Neck Surgery May 2003
668 CHU and OSGUTHORPE
Table 1. (Continued)
Primary site
Histology
Treatment
Metastasis site
Interval to metastasis (mo)
L lower lip
SCCE
Wedge resection
L parotid w/recurrence
18
L cheek
SCCE
Moh’s
L parotid, level I
0
R scalp
SCCE
Moh’s ⫻ 2
L parotid, R level II, V w/satellite lesions
2
L auricle
SCCE
Moh’s
L level II
36
R cheek
SCCE
Moh’s
R level I, II
7
R lower lip
SCCE
Moh’s
R level I
7
L temple
SCCE
Moh’s
L parotid, level I
8
R auricle
SCCE
Moh’s
R parotid
4
L auricle
SCCE
Moh’s
L parotid
3
R temple
SCCE
Moh’s
R parotid
9
R temple/cheek (multiple recurrences)
SCCE
Moh’s multiple
R parotid
0
R nose
SCCE
Moh’s
R level I w/recurrence
96
R temple
Sweat gland cancer
Excision
R parotid
0
R nasolabial
Eccrine cancer
Excision
R level I
4
AWD, alive with disease; DID, dead of intercurrent disease; DOD, dead of disease; MND, modified neck dissection; NED, no evidence of disease; SND, selective neck dissection.
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
CHU and OSGUTHORPE 669
Table 1. (Continued)
Surgical treatment
Pathology
Irradiation (Gy)
Chemotherapy
Follow-up since treatment for metastasis (mo)
Wide excision, total parotid, B SND
N1 parotid LN (1) ⫹ perineural ⫹ (facial N, inf. alveolar N to w/i 2cm f.ovale)
60
NED (77)
Total parotid, L MND
N2b parotid LN (1) ⫹ level I LN (1) ⫹ margin ⫹ perineural ⫹
60
DOD (12) pelvic/ scapular metastasis
After radiation, R RND
N2b level V (2) ⫹ margin ⫹
68
L MND
N2b level II (15) ⫹ lymphovasc ⫹
64
NED (61)
R SND
N2b level I (1) ⫹/ECS level II (1) ⫹/ECS
50
DOD (48) pulmonary metastasis
R SND
N1 level I LN (1) ⫹
Total parotid, L SND
N2b parotid LN (1) ⫹/ECS level I LN (2) ⫹
Total parotid, R SND
N1 parotid LN (1) ⫹
Total parotid
N2b parotid LN (2) ⫹/ECS margin ⫹
Total parotid, R SND
N2a parotid LN (1) ⫹
Total parotid, R SND
N2a parotid LN (1) ⫹ margin ⫹
60
DID (48) pneumonia
Partial rhinectomy, R SND
N2b level I LN (2) ⫹/ECS
60
NED (47)
Total parotid
N1 parotid LN (1) ⫹ lymphatic ⫹
60
NED (168)
R MND
N2b LN (2) ⫹
70
DOD (12)
5-FU Taxol
DOD (11)
NED (54) 60
NED (21)
NED (80) 60
AWD (51)
NED (22)
670 CHU and OSGUTHORPE
Otolaryngology– Head and Neck Surgery May 2003
Fig 2. Metastasis (arrow) from posterior cheek primary (squamous cell carcinoma) just lateral to left parotid gland.
Fig 3. External jugular metastases (left, arrowheads) in 2 patients with posterior inferior cheek primaries (squamous cell carcinoma).
DISCUSSION Cutaneous malignancy with regional metastasis is a disease primarily of the older, male, white population. Although uncommon, regional meta-
static disease portends a poor prognosis. Overall, 79% (22 patients) of the patients in this series developed regional metastasis despite the fact that they received micrographic surgery for their pri-
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
CHU and OSGUTHORPE 671
Fig 4. Lymphatic embolus (E) of tumor cells filling lymphatic vessel.
mary lesion, a technique that currently affords the most precise margin control and the ability to detect local perineural or lymphovascular involvement.10 The median time from primary lesion to regional metastasis of 9 months in this study was comparable to other studies.2,3,7 Most (86%) of our patients received surgery, including parotidectomy and/or neck dissection, depending on the primary sites and nodal echelon(s) involved. Sixty-eight percent of our patients received postoperative irradiation for indications of extracapsular spread, perineural or lymphovascular invasion, or parotid nodal involvement. A comment is warranted about the metastasis to the external jugular nodes (Fig 3), which occurred in 3 patients. Such location for metastasis is rarely reported, and there is no description of an appropriate regional lymphadenectomy.9 The external jugular vein and associated lymphatics drain the mid to inferior posterior cheek and adjacent neck skin, and such passes between the sternocleidomastoid muscle and platysma to enter the posterior triangle a bit below Erb’s point, continuing inferiorly to drain into the subclavian vein in most,
although connections to the lower internal jugular vein occur in some. A dissection of potentially involved nodes ought to include, at least, the lateral fascial sheath of the sternocleidomastoid and the adjacent platysma, along with level VA and VB lymphatics of the posterior triangle, and possibly level IV (depending on whether the external jugular vein drains into the subclavian or internal jugular vein).9 Several studies have examined prognostic factors for the development of regional metastasis. These include the location of the primary site (ear, temple, lip), tumor in scar tissue, positive surgical margins, local recurrence, size of greater than 2 cm, depth of greater than 4 mm, perineural or lymphovascular invasion, desmoplastic characteristics, and immunosuppression 1,3-7,10,18 Our series revealed 43% (12 patients) who were either AWD or DOD. Five of these had surgical or clinical evidence of perineural spread, 6 had positive surgical margins, and 1 had evidence of extracapsular spread. Of these patients, the most common locations of primary lesions were the cheek and auricle, compared with the temple and
Otolaryngology– Head and Neck Surgery May 2003
672 CHU and OSGUTHORPE
Fig 5. Perineural spread with tumor cells (arrows) encasing, just under the perineurium, 2 small peripheral nerves.
auricle for all the study patients. Six patients have had local recurrences. CONCLUSION Of 28 patients with regional metastasis from a nonmelanoma cutaneous primary in the head and neck, most underwent the currently conventional treatments of regional lymphadenectomy, frequently with postoperative irradiation. With a 34-month mean follow-up, only 39% are without evidence of disease (although an additional 18% in this elderly group died of unrelated causes, without evidence of disease). Improvements in management are obviously required to achieve better outcomes, preferably before clinical metastasis has occurred. At the least, and has been suggested by others,1,3-7,9,10,18 examination of the nearest “sentinel node” (commonly performed for melanomas),19,20 prophylactic lymphadenectomy and/or regional irradiation must be considered in histologically aggressive nonmelanoma cutaneous malignancy, or with evidence of lymphovascular or perineural spread in the primary specimen. Ultimately, multi-institutional studies on larger numbers of patients are warranted.
REFERENCES
1. Malone JP, et al. Basal cell carcinoma metastatic to the parotid: report of a new case and review of the literature. ENT 2000;79:511-9. 2. Lo JS, et al. Metastatic basal cell carcinoma: report of twelve cases with a review of the literature. J Am Acad Dermatol 1991;24:715-9. 3. Tavin E, Persky MS, Jacobs J. Metastatic basal cell carcinoma of the head and neck. Laryngoscope 1995; 105:814-7. 4. Kraus DH, Carew JF, Harrison LB. Regional lymph node metastasis from cutaneous squamous cell carcinoma. Arch Otolaryngol Head Neck Surg 1998;124: 582-7. 5. Dinehart SM, Pollack SV. Metastases from squamous cell carcinoma of the skin and lip. J Am Acad Dermatol 1989;21:241-7. 6. Cherpelis BS, Marcusen C, Lang PG. Prognostic factors for metastasis in squamous cell carcinoma of the skin. Dermatol Surg 2002;28:268-73. 7. Khurana VG, et al. Parotid and neck metastases from cutaneous squamous cell carcinoma of the head and neck. Am J Surg 1995;170:446-50. 8. Jackson GL, Ballantyne AJ. Role of parotidectomy for skin cancer of the head and neck. Am J Surg 1981;142:464-9. 9. Brown RO, Osguthorpe JD. Management of the neck in nonmelanocytic cutaneous carcinomas. Otolaryngol Clin North Am 1998;31:841-56. 10. Rowe DE, Carroll RJ, Day CL. Prognostic factors for local recurrence, metastasis, and survival rates in squa-
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
11.
12.
13.
14.
15. 16.
mous cell carcinoma of the skin, ear, and lip. J Am Acad Dermatol 1992;26:976-90. Taylor BW, et al. Carcinoma of the skin metastatic to parotid area lymph nodes. Head Neck 1991;Sept/Oct: 427–33. Cassisi NJ, Dickerson DR, Million RR. Squamous cell carcinoma of the skin metastatic to parotid nodes. Arch Otolaryngol 1978;104:336-9. O’Brien CJ, et al. Incidence of cervical node involvement in metastatic cutaneous malignancy involving the parotid gland. Head Neck 2001;Sept:744 – 8. Akhtar S, Oza KK, Wright J. Merkel cell carcinoma: report of 10 cases and review of the literature. J Am Acad Dermatol 2000;43:775-67. Allen PJ, Zhang ZF, Coit DG. Surgical management of Merkel cell carcinoma. Ann Surg 1999;229:97-105. Otero-Garcia JE, Carlo VJ, Trinidad-Pinedo J. Malignant
CHU and OSGUTHORPE 673
17.
18.
19.
20.
eccrine spiradenoma of the neck: a case report. Otolaryngol Head Neck Surg 2001;125:428. Daniel SJ, et al. Facial sweat gland carcinoma metastasizing to neck nodes. Arch Otolaryngol Head Neck Surg 2001;127:1495-8. Osguthorpe JD, et al. Neurotropic cutaneous tumors of the head and neck. Arch Otolaryngol Head Neck Surg 1997;123:871-6. Dessureault S, et al. American Joint Committee on Cancer (AJCC) Melanoma Staging Committee. Improved staging of node-negative patients with intermediate to thick melanoma with the use of lymphatic mapping and sentinel lymph node biopsy. Ann Surg Oncol 2001;8: 766-70. Patel SG, et al. Sentinel lymph node biopsy for cutaneous head and neck melanoma. Arch Otolaryngol Head Neck Surg 2002;128:285–91.
MD,
and J. DAVID OSGUTHORPE,
MD,
Charleston, South Carolina
OBJECTIVE AND STUDY DESIGN: We performed a retrospective study of 28 patients to evaluate the management and outcome of regional metastasis from nonmelanoma cutaneous malignancies. RESULTS: There were 25 squamous cell, 1 basal cell, and 2 eccrine carcinomas. The most common primary locations were the temple, ear, and cheek. Median interval from primary resection to regional metastasis was 9 months, and such were commonly of the parotid, level IB, and level IIA nodes. Twenty-four patients underwent parotidectomy and/or neck dissection; 19 patients, postoperative radiation. Ten patients (36%) died from the disease, 2 patients (7%) have persisting disease, 5 (18%) have died from intercurrent disease, and 11 (39%) have no evidence of disease (minimum follow-up, 22 months; median, 34 months). CONCLUSIONS AND SIGNIFICANCE: Nonmelanoma cutaneous malignancies with regional metastasis have a poor prognosis despite aggressive therapy. When high-risk characteristics are detected, examination of the nearest “sentinel node,” prophylactic lymphadenectomy, and/or regional irradiation might be justified; further study is warranted. (Otolaryngol Head Neck Surg 2003;128:663-73.)
B asal cell carcinoma (BCC) and squamous cell carcinoma (SCC) comprise the preponderance of nonmelanoma cutaneous malignancies, the most common cancers in the United States. BCC occurs more frequently than SCC, by a ratio of 4:1. The most frequent location of these malignancies is in the head and neck region of white men in their From the Department of Otolaryngology–Head and Neck Surgery, Medical University of South Carolina. Presented at the Annual Meeting of the American Academy of Otolaryngology–Head and Neck Surgery, San Diego, CA, September 22-25, 2002. Reprint requests: J. David Osguthorpe, MD, Otolaryngology– Head and Neck Surgery, PO Box 250550, 135 Rutledge Ave, Suite 1130, Charleston, SC 29425-0550; e-mail, [email protected]. Copyright © 2003 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. 0194-5998/2003/$30.00 ⫹ 0 doi:10.1016/S0194-5998(03)00226-2
fifth to sixth decade of life, which reflects the common risk factors of chronic sun exposure and fair complexion. Other predisposing factors include burn scars, cutaneous exposure to certain chemicals (eg, arsenic, hydrocarbons), immunosuppression, and genetic factors (eg, xeroderma pigmentosum, basal cell nevoid syndrome, etc). Although uncommon, both SCC and BCC have the potential to metastasize. The regional metastatic rate for BCC is reported from 0.0028% to 0.5%,1-3 whereas the reported rate for SCC ranges from 1% to 12.5%.4-6 Distant metastasis is rare, especially without locoregional disease. Regional metastasis from cutaneous primary lesions in the head and neck region most frequently occur in the parotid nodes, followed by the upper cervical nodes.7,8 This pattern reflects the favored primary sites of BCC and SCC, including the ear, preauricular region, and upper face whose lymphatics drain first to the parotid nodes and then onto the neck nodes.5 Drainage patterns based on melanomas are presumed to be applicable to BCC and SCC as depicted (Fig 1). That is, parotid involvement usually arises from the anterior scalp, temple, face, and ear and much less frequently from the posterior scalp and neck.9 Furthermore, minimal involvement of level VA nodes occurs with lesions of the anterior scalp, temple, face, and ear, whereas level IA is rarely involved with lesions of the posterior scalp and neck.9 In general, most regional metastatic SCC becomes clinically evident within the first 2 years after treatment of the primary site.5,6 For BCC this time period varies from less than 1 year to 45 years, with a median time of 6 to 12 years.2,3,7 Regional metastasis is concurrent with persistence or recurrence at the primary location in approximately 42% of cases.4 Thus, failure of local control is an important risk factor for regional metastasis.3,10 Curative treatment options for regional metastasis include surgery and irradiation. Many studies have demonstrated improved disease control with combined surgery and irradiation versus surgery 663
664 CHU and OSGUTHORPE
Otolaryngology– Head and Neck Surgery May 2003
Fig 1. Lymphatic drainage patterns of the skin of the face, scalp, and upper neck. (From Jackson G, Ballantyne A: Am J Surg 1981;142:464-9, with permission.)
alone, particularly in the management of the parotid nodes.8,11,12 Cassisi et al12 demonstrated that a superficial parotidectomy alone for metastatic cutaneous SCC had a 75% recurrence rate compared with 20% for those treated with combined therapy. This high recurrence can be at least in part explained by the anatomy of lymphatics that traverse the parotid, with the intraglandular nodes located lateral to the posterior facial vein. Depending on the relation of the facial nerve to the posterior facial vein, the former lateral to the latter in about 75% of cases, involved nodes can be medial to the facial nerve and hence be left behind in a superficial parotidectomy.12 Parotidectomy or parotid area irradiation is sometimes performed prophylactically in addition to a neck dissection for positive level I or II cervical nodes associated with primary lesions in the upper lateral face or posterior cheek, as lymphatic spread from these areas to the neck assumes passage through the parotid nodes. Khurana et al7 found an 18% rate of occult parotid metastasis in patients presenting solely with cervical metastasis from cutaneous primary lesions of the face.
Similarly, cervical metastatic disease is treated with surgery, irradiation, or both. Taylor et al11 reported that in clinically positive necks, a neck dissection with postoperative irradiation was favored, with a 89% control rate compared with 63% rate with surgery alone. Furthermore, in clinically negative necks in patients with parotid metastasis, neck dissection or irradiation to the neck might be prudent given the 24% to 35% occult cervical metastatic rate, according to Jackson and Ballantyne8 and O’Brien et al.13 Other rare nonmelanoma skin carcinomas include Merkel cell and eccrine sweat gland carcinomas. Merkel cell carcinoma, first described in 1972 by Toker, is derived from Merkel cells (neuroendocrine cells) located at the dermoepidermal junction. Such are rare, with only approximately 1100 cases reported, but are very aggressive, with high local recurrence (12% to 50%), regional metastatic (17% to 76%), and distant metastatic (12% to 50%) rates.14 If aggressively treated with surgery at an early stage, Allen et al15 reported a 75% 5-year survival rate. However, in the setting of locoregional recurrence or distant metastasis,
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
2-year survival rates are approximately 35% and 17%, respectively. Eccrine (sweat gland) carcinomas are likewise rare malignancies, occurring mostly in the extremities and trunk, with surgical excision the current standard treatment.16 The effectiveness of adjuvant therapies, including radiotherapy, chemotherapy, or antiestrogen therapy, has been variable, mainly in case reports, with no major series results to date.16,17 The objective of this study is to evaluate the management and outcome in our experience with patients at the Medical University of South Carolina with regional metastasis from nonmelanoma cutaneous carcinomas. METHODS A retrospective chart review identified 28 patients referred to the junior author (J.D.O.) between 1984 and 2000 for the treatment of regional metastasis from a nonmelanoma cutaneous malignancy. As summarized in Table 1, information about sites of primary lesions, histology, metastatic sites, and both prior and current treatments (at our institution or elsewhere) were assembled. The interval to metastatic disease was determined as the time from first treatment of a primary lesion to the detection of regional metastasis. Follow-up information was obtained from medical records or telephone interviews. Patients were classified in 4 follow-up categories: dead of disease (DOD), dead of intercurrent disease (DID), alive with disease (AWD), and no evidence of disease (NED). Survivors were followed a minimum of 22 months, and a median of 34 months, and all patients were accounted for. RESULTS The median age of the 28 patients was 59 years. All were white, and 23 were male. Histologically, lesions included 25 SCCs, 1 BCC, and 2 eccrine carcinomas. The most common primary locations were, in order, of the temple, ear, and cheek. The majority of the primary lesions were treated by fresh tissue technique micrographic surgery (22 patients, or 79%, at our institution or elsewhere), with 3 patients having had at least 2 micrographic surgeries for local recurrence. A median of 9 months separated treatment of the primary lesions from clinical evidence of metastatic disease,
CHU and OSGUTHORPE 665
which occurred, in order, in the parotid, level IB, and level IIA nodes. An example of parotid metastasis is shown in the left parotid (Fig 2). Three patients were noted to have metastatic disease involving the external jugular nodes, from temple, posterior cheek, and posterior superior neck primary sites. Two examples of metastasis to the left external jugular nodes are demonstrated (Fig 3). Thirteen patients (46%) presented with parotid nodal disease, whereas 10 (36%) presented with only cervical metastasis. Five patients (18%) had both parotid and cervical metastatic disease. Twenty-four patients received surgical treatments, including parotidectomy and/or selective or modified radical neck dissection. According to pathology results, 7 patients met N1 neck criteria (and had selective neck dissection); 3 patients, N2a, and 14 patients, N2b (had modified radical neck dissection in the latter 2 categories). Seven patients had positive surgical margins, 3 patients had lymphovascular involvement (Fig 4), and 8 had perineural spread at the primary sites (Fig 5); 5 patients had extracapsular spread of tumor from involved parotid or cervical nodes at the time of parotidectomy and/or neck dissection. Four patients did not receive surgical intervention because of unresectability (intracranial perineural spread in 2 patients), poor medical status, or patient refusal (1 patient). Nineteen patients received postoperative irradiation, ranging from 50 to 70 Gy. Of these patients, 1 received irradiation as primary treatment, and 2 patients as palliation. Two patients received chemotherapy for palliation. Follow-up data disclosed 43% (12 patients) either AWD (2 patients) or DOD (10 patients), with distant metastasis the most common situation. Eighteen percent (5 patients) have died of intercurrent disease (pneumonia in 2, noncutaneous malignancy in 2, cardiac disease in 1); and 39% (11 patients, followed a median of 57 months) are NED. Of the 10 patients who were DOD, 4 had positive margins at the primary site (2 had recurrent disease despite micrographic surgery), and 3 of these patients also had perineural involvement. Clinically, perineural spread intracranially via V3 also occurred in 2 patients. Four of these patients had primary sites on the cheek, and 2 on the auricle.
Otolaryngology– Head and Neck Surgery May 2003
666 CHU and OSGUTHORPE
Table 1. Patient data
Primary site
Histology
Treatment
Metastasis site
Interval to metastasis (mo)
L cheek
SCCE
Moh’s
L exterior jugular and pulmonary metastasis (surgery and radiation for recurrence)
33
L forehead
SCCE
Moh’s
L parotid
6
L lowerlip
SCCE
Wide excision
L level I
24
L temple
SCCE
Wide excision
R parotid, external jugular level II
23
R cheek
SCCE
Moh’s
R parotid
0
L lower lip
SCCE
Moh’s
L level I, II
9
L superior helix
SCCE
Moh’s
L parotid
12
R cheek
SCCE
Curettage ⫻ 3, then wide excision
R parotid
25
L postneck
SCCE
Moh’s
L external jugular
36
L auricle
SCCE
Moh’s
L parotid
12
R superior helix
SCCE
Radiation
R occipital and level V
R temple
SCCE
Moh’s
R parotid and level II
L temple
SCCE
Wide excision
R parotid w/recurrence status post radiation
L auricle
SCCE
Moh’s
L parotid
6
22 1
10
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
CHU and OSGUTHORPE 667
Table 1. (Continued)
Surgical treatment
Pathology
Irradiation (Gy)
Chemotherapy
cisplatin 5FU
Follow-up since treatment for metastasis (mo)
DOD (24) pulmonary metastasis
Total parotid
N1 LN (1) ⫹
59
NED (41)
L MND
N2b level I LN (2) ⫹
63
NED (57)
Total parotid, R MND
N2b parotid LN (1) ⫹ exterior jugular LN (1) ⫹ level II LN (1) ⫹
66
DOD (17) thoracic spinal metastasis
Superficial parotid, R SND
N1 parotid LN (1) ⫹ perineural ⫹ vascular ⫹
L MND
N2b level I LN (1) ⫹/ECS level II (2) ⫹ margins ⫹ perineural ⫹ lymphovascular ⫹
64
DOD (20) perineural intracranial spread via V3
Total parotid, L SND
N1 parotid LN ⫹ margins ⫹ perineural ⫹
60
DOD (21) pulmonary metastasis
59 (palliation)
DOD (7) perineural intracranial spread via V3
Unresectable
DID (10) renal cancer
L SND
N2b exterior jugular LN (2) ⫹
60
NED (58)
Superficial parotid, R SND
N2a parotid LN (1) ⫹
60
DID (20)
R SND
N2b LN (14) ⫹/ECS muscle ⫹ SMGld ⫹/endoneural
64
DID (2) lymphoma recurrence
Refused all treatment Total parotid, local reexcision Medically unsuitable for surgery
DID (20) pneumonia N2b parotid LN (2) ⫹ margin ⫹
64
AWD (45)
25 (incomplete)
DOD (12)
Otolaryngology– Head and Neck Surgery May 2003
668 CHU and OSGUTHORPE
Table 1. (Continued)
Primary site
Histology
Treatment
Metastasis site
Interval to metastasis (mo)
L lower lip
SCCE
Wedge resection
L parotid w/recurrence
18
L cheek
SCCE
Moh’s
L parotid, level I
0
R scalp
SCCE
Moh’s ⫻ 2
L parotid, R level II, V w/satellite lesions
2
L auricle
SCCE
Moh’s
L level II
36
R cheek
SCCE
Moh’s
R level I, II
7
R lower lip
SCCE
Moh’s
R level I
7
L temple
SCCE
Moh’s
L parotid, level I
8
R auricle
SCCE
Moh’s
R parotid
4
L auricle
SCCE
Moh’s
L parotid
3
R temple
SCCE
Moh’s
R parotid
9
R temple/cheek (multiple recurrences)
SCCE
Moh’s multiple
R parotid
0
R nose
SCCE
Moh’s
R level I w/recurrence
96
R temple
Sweat gland cancer
Excision
R parotid
0
R nasolabial
Eccrine cancer
Excision
R level I
4
AWD, alive with disease; DID, dead of intercurrent disease; DOD, dead of disease; MND, modified neck dissection; NED, no evidence of disease; SND, selective neck dissection.
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
CHU and OSGUTHORPE 669
Table 1. (Continued)
Surgical treatment
Pathology
Irradiation (Gy)
Chemotherapy
Follow-up since treatment for metastasis (mo)
Wide excision, total parotid, B SND
N1 parotid LN (1) ⫹ perineural ⫹ (facial N, inf. alveolar N to w/i 2cm f.ovale)
60
NED (77)
Total parotid, L MND
N2b parotid LN (1) ⫹ level I LN (1) ⫹ margin ⫹ perineural ⫹
60
DOD (12) pelvic/ scapular metastasis
After radiation, R RND
N2b level V (2) ⫹ margin ⫹
68
L MND
N2b level II (15) ⫹ lymphovasc ⫹
64
NED (61)
R SND
N2b level I (1) ⫹/ECS level II (1) ⫹/ECS
50
DOD (48) pulmonary metastasis
R SND
N1 level I LN (1) ⫹
Total parotid, L SND
N2b parotid LN (1) ⫹/ECS level I LN (2) ⫹
Total parotid, R SND
N1 parotid LN (1) ⫹
Total parotid
N2b parotid LN (2) ⫹/ECS margin ⫹
Total parotid, R SND
N2a parotid LN (1) ⫹
Total parotid, R SND
N2a parotid LN (1) ⫹ margin ⫹
60
DID (48) pneumonia
Partial rhinectomy, R SND
N2b level I LN (2) ⫹/ECS
60
NED (47)
Total parotid
N1 parotid LN (1) ⫹ lymphatic ⫹
60
NED (168)
R MND
N2b LN (2) ⫹
70
DOD (12)
5-FU Taxol
DOD (11)
NED (54) 60
NED (21)
NED (80) 60
AWD (51)
NED (22)
670 CHU and OSGUTHORPE
Otolaryngology– Head and Neck Surgery May 2003
Fig 2. Metastasis (arrow) from posterior cheek primary (squamous cell carcinoma) just lateral to left parotid gland.
Fig 3. External jugular metastases (left, arrowheads) in 2 patients with posterior inferior cheek primaries (squamous cell carcinoma).
DISCUSSION Cutaneous malignancy with regional metastasis is a disease primarily of the older, male, white population. Although uncommon, regional meta-
static disease portends a poor prognosis. Overall, 79% (22 patients) of the patients in this series developed regional metastasis despite the fact that they received micrographic surgery for their pri-
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
CHU and OSGUTHORPE 671
Fig 4. Lymphatic embolus (E) of tumor cells filling lymphatic vessel.
mary lesion, a technique that currently affords the most precise margin control and the ability to detect local perineural or lymphovascular involvement.10 The median time from primary lesion to regional metastasis of 9 months in this study was comparable to other studies.2,3,7 Most (86%) of our patients received surgery, including parotidectomy and/or neck dissection, depending on the primary sites and nodal echelon(s) involved. Sixty-eight percent of our patients received postoperative irradiation for indications of extracapsular spread, perineural or lymphovascular invasion, or parotid nodal involvement. A comment is warranted about the metastasis to the external jugular nodes (Fig 3), which occurred in 3 patients. Such location for metastasis is rarely reported, and there is no description of an appropriate regional lymphadenectomy.9 The external jugular vein and associated lymphatics drain the mid to inferior posterior cheek and adjacent neck skin, and such passes between the sternocleidomastoid muscle and platysma to enter the posterior triangle a bit below Erb’s point, continuing inferiorly to drain into the subclavian vein in most,
although connections to the lower internal jugular vein occur in some. A dissection of potentially involved nodes ought to include, at least, the lateral fascial sheath of the sternocleidomastoid and the adjacent platysma, along with level VA and VB lymphatics of the posterior triangle, and possibly level IV (depending on whether the external jugular vein drains into the subclavian or internal jugular vein).9 Several studies have examined prognostic factors for the development of regional metastasis. These include the location of the primary site (ear, temple, lip), tumor in scar tissue, positive surgical margins, local recurrence, size of greater than 2 cm, depth of greater than 4 mm, perineural or lymphovascular invasion, desmoplastic characteristics, and immunosuppression 1,3-7,10,18 Our series revealed 43% (12 patients) who were either AWD or DOD. Five of these had surgical or clinical evidence of perineural spread, 6 had positive surgical margins, and 1 had evidence of extracapsular spread. Of these patients, the most common locations of primary lesions were the cheek and auricle, compared with the temple and
Otolaryngology– Head and Neck Surgery May 2003
672 CHU and OSGUTHORPE
Fig 5. Perineural spread with tumor cells (arrows) encasing, just under the perineurium, 2 small peripheral nerves.
auricle for all the study patients. Six patients have had local recurrences. CONCLUSION Of 28 patients with regional metastasis from a nonmelanoma cutaneous primary in the head and neck, most underwent the currently conventional treatments of regional lymphadenectomy, frequently with postoperative irradiation. With a 34-month mean follow-up, only 39% are without evidence of disease (although an additional 18% in this elderly group died of unrelated causes, without evidence of disease). Improvements in management are obviously required to achieve better outcomes, preferably before clinical metastasis has occurred. At the least, and has been suggested by others,1,3-7,9,10,18 examination of the nearest “sentinel node” (commonly performed for melanomas),19,20 prophylactic lymphadenectomy and/or regional irradiation must be considered in histologically aggressive nonmelanoma cutaneous malignancy, or with evidence of lymphovascular or perineural spread in the primary specimen. Ultimately, multi-institutional studies on larger numbers of patients are warranted.
REFERENCES
1. Malone JP, et al. Basal cell carcinoma metastatic to the parotid: report of a new case and review of the literature. ENT 2000;79:511-9. 2. Lo JS, et al. Metastatic basal cell carcinoma: report of twelve cases with a review of the literature. J Am Acad Dermatol 1991;24:715-9. 3. Tavin E, Persky MS, Jacobs J. Metastatic basal cell carcinoma of the head and neck. Laryngoscope 1995; 105:814-7. 4. Kraus DH, Carew JF, Harrison LB. Regional lymph node metastasis from cutaneous squamous cell carcinoma. Arch Otolaryngol Head Neck Surg 1998;124: 582-7. 5. Dinehart SM, Pollack SV. Metastases from squamous cell carcinoma of the skin and lip. J Am Acad Dermatol 1989;21:241-7. 6. Cherpelis BS, Marcusen C, Lang PG. Prognostic factors for metastasis in squamous cell carcinoma of the skin. Dermatol Surg 2002;28:268-73. 7. Khurana VG, et al. Parotid and neck metastases from cutaneous squamous cell carcinoma of the head and neck. Am J Surg 1995;170:446-50. 8. Jackson GL, Ballantyne AJ. Role of parotidectomy for skin cancer of the head and neck. Am J Surg 1981;142:464-9. 9. Brown RO, Osguthorpe JD. Management of the neck in nonmelanocytic cutaneous carcinomas. Otolaryngol Clin North Am 1998;31:841-56. 10. Rowe DE, Carroll RJ, Day CL. Prognostic factors for local recurrence, metastasis, and survival rates in squa-
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
11.
12.
13.
14.
15. 16.
mous cell carcinoma of the skin, ear, and lip. J Am Acad Dermatol 1992;26:976-90. Taylor BW, et al. Carcinoma of the skin metastatic to parotid area lymph nodes. Head Neck 1991;Sept/Oct: 427–33. Cassisi NJ, Dickerson DR, Million RR. Squamous cell carcinoma of the skin metastatic to parotid nodes. Arch Otolaryngol 1978;104:336-9. O’Brien CJ, et al. Incidence of cervical node involvement in metastatic cutaneous malignancy involving the parotid gland. Head Neck 2001;Sept:744 – 8. Akhtar S, Oza KK, Wright J. Merkel cell carcinoma: report of 10 cases and review of the literature. J Am Acad Dermatol 2000;43:775-67. Allen PJ, Zhang ZF, Coit DG. Surgical management of Merkel cell carcinoma. Ann Surg 1999;229:97-105. Otero-Garcia JE, Carlo VJ, Trinidad-Pinedo J. Malignant
CHU and OSGUTHORPE 673
17.
18.
19.
20.
eccrine spiradenoma of the neck: a case report. Otolaryngol Head Neck Surg 2001;125:428. Daniel SJ, et al. Facial sweat gland carcinoma metastasizing to neck nodes. Arch Otolaryngol Head Neck Surg 2001;127:1495-8. Osguthorpe JD, et al. Neurotropic cutaneous tumors of the head and neck. Arch Otolaryngol Head Neck Surg 1997;123:871-6. Dessureault S, et al. American Joint Committee on Cancer (AJCC) Melanoma Staging Committee. Improved staging of node-negative patients with intermediate to thick melanoma with the use of lymphatic mapping and sentinel lymph node biopsy. Ann Surg Oncol 2001;8: 766-70. Patel SG, et al. Sentinel lymph node biopsy for cutaneous head and neck melanoma. Arch Otolaryngol Head Neck Surg 2002;128:285–91.