|Ahead of print publication
Eyelid carcinoma: An experience from a tertiary cancer center
Shivakumar Thiagarajan1, Ameya Bahani2, Devendra Chaukar1, Anil K Dcruz1
1 Department of Head and Neck Surgical Oncology, Tata Memorial Centre, Mumbai, Maharashtra, India
2 Department of Head and Neck Surgery, Medicare Hospital, Indore, Madhya Pradesh, India
|Date of Submission||27-Aug-2018|
|Date of Decision||09-Jun-2019|
|Date of Acceptance||22-Aug-2019|
|Date of Web Publication||29-Jan-2020|
Department of Head and Neck Surgical Oncology, Tata Memorial Centre, Mumbai - 400 012, Maharashtra
Source of Support: None, Conflict of Interest: None
Context: Eyelid carcinoma is rare tumors of the head and neck. They are rarely lethal but can be associated with significant morbidity if not treated early and appropriately. There are limited data available from world over and in particular the Indian subcontinent regarding eyelid carcinoma and its prognostic factors influencing treatment outcomes.
Setting and Design:Retrospective study of patients treated in a tertiary cancer center between 2005 and 2016.
Methodology: In this study, 51 patients with eyelid carcinoma treated at single tertiary cancer center were included. The demographic, clinical data, which includes the treatment received, histopathology report and follow–up, were recorded. All the relevant variables influencing disease-free survival (DFS) were analyzed.
Results: Sebaceous carcinoma was the most common eyelid carcinoma followed by squamous cell carcinoma and basal cell carcinoma in descending order in this series. Lower eyelid was involved most often. The incidence of nodal metastasis was low (14%). Multivariate analysis revealed that margin status influenced the DFS (P= 0.001) (hazard ratios = 15.9 [95% confidence interval: 1.8–135.2]). The 5 years' DFS was 70%.
Conclusion: Eyelid tumors are less common cancer with good prognosis if treated appropriately. The morbidity associated with treatment can be reduced if treated early.
Keywords: Eyelid carcinoma, prognostic factors, survival
| > Introduction|| |
Eyelid carcinoma is rare cancers of the head and neck. Description of its incidence in literature varies depending on the region from where the study has been reported. The reported age-adjusted incidences of eyelid malignancies vary between 5.1 cases/100,000 as reported from Singapore to 15.7 cases/100,000 population in the United States of America.,, Its exact incidence in India is not known. The most common type of eyelid carcinoma described in literature is basal cell carcinoma (BCC). However, sebaceous gland carcinoma (SGC) has been reported as the most common eyelid carcinoma from the Asian subcontinent. Eyelid carcinoma is often misdiagnosed and treated as benign lesions, which are much more common. This can add to the morbidity of subsequent treatment after a diagnosis of malignancy is made. The management of eyelid carcinoma requires different considerations keeping in mind its impact on the functions after resection and reconstruction. The treatment for eyelid carcinomas is guided by the type of malignancy and the stage. Surgical excision is the preferred treatment modality. Adjuvant radiotherapy (RT) may be considered in the presence of certain adverse features. There have been few retrospective studies on eyelid carcinoma from the Asian subcontinent, but most of them have small numbers and reports regarding the prognostic factors and survival are often missing.,,
| > Methodology|| |
This retrospective study was done with the aim of studying the pattern of presentation of eyelid carcinoma and also to analyze the various prognostic factors influencing survival in these patients. Patients with eyelid carcinoma of all stages, who received curative treatment at our institute between January 2005 and December 2016 and whose histopathological details were available for analysis were included in the study. We excluded cases with benign lesions. The various demographic, clinical, histopathological details and follow-up details were collected. Statistical analysis was performed with IBM SPSS 20.0 version (IBM, New York, USA). The variables for univariate analysis were selected based on the clinical relevance as well as those previously described in the literature. It was done using log-rank test. All statistically significant (P< 0.05) variables were subsequently tested in the multivariate analysis using the Cox regression analysis model using the forward stepwise selection method.
| > Results|| |
Of the 161 patients with eyelid carcinoma registered at our institute in the specified period, 51 patients fulfilled the eligibility criteria. The details of the demographic and clinical features are given in [Table 1]. The distribution of cases among males and females was almost equal. The median age of the patients was 62 years. Lower eyelid (n = 27, 53%) was the most common site. Majority of the patients in this series were per primam (n = 39, 76.5%). Twelve patients (23.5%) had received treatment outside before presenting to our institute. Clinically, conjunctiva was involved in 25 cases (49%), more owing to the pagetoid spread of sebaceous carcinoma, which was the most common histopathological type. Twelve (23.5%) patients had visual disturbance and 3 (6%) patients had restriction of eye movement, all three of them required orbital exenteration as part of their treatment. Visual disturbances were noticed in patients only with conjunctival involvement. Most patients had early-stage cancer at presentation [Figure 1]. Clinically, node-positive disease was witnessed in 6 (12%) patients. All the patients underwent surgery. Wide excision was performed in 48 (94%) patients, the remaining 3 patients (6%) required orbital exenteration due to the extensive nature of the disease.
AJCC 7th edition was used to stage all the patients in this series as majority of them were treated after 2009. Most of the patients presented with early-stage carcinoma [Figure 1]. Sebaceous carcinoma (n = 19, 37%) was the most common type of eyelid carcinoma in our series. This was followed by squamous cell carcinoma (SCC) (n = 14, 27%) and BCC (n = 12, 23%) [Table 2]. Four (8%) patients had undergone surgical excision outside with inadequate margins for which they were operated at our institute. There was no residual carcinoma reported in the final histopathology report in these four patients.
Nodal echelons were addressed in clinically node-positive neck or for suspicious nodes in radiological evaluation or advanced-stage disease [Table 1]. The histopathological type of eyelid carcinoma associated with node-positive disease (n = 5) was SCC in two patients and sebaceous carcinoma in three patients. Both the patients with extracapsular spread of nodal disease were seen in patients with sebaceous carcinoma.
The surgical defects were reconstructed with primary closure in 6 (12%) patients, 16 patients (31%) required a local rotation or advancement flap for reconstruction of the defect. Pedicled flap-like median forehead flap was used in 26 patients (51%). Two out of three patients who underwent exenteration required free flap (anterolateral thigh flap) for reconstruction [Table 1].
We had taken 4-mm margin as an adequate margin for all histopathological variants of eyelid carcinoma. Eighteen patients had close margins, of these seven, were with SCC, eight patients with sebaceous carcinoma, two with BCC, and one of them had mucosal melanoma. Number of the patients had a positive margin. Two (4%) patients had lymphovascular emboli, both of which were seen in patients with sebaceous carcinoma. None of the 51 patients in the present series had perineural invasion [Table 2]. Eleven (21%) patients in our series received adjuvant RT for either advanced T stage or node-positive neck. None of our patients received chemotherapy either in the neoadjuvant or adjuvant setting.
The median follow-up of our patients was 24 months (range: 0–135 months). Majority of the patients were alive and disease-free at last follow-up (n = 45, 88%) [Table 1]. Seven patients had recurrences, all of which were salvaged. Of the many clinical factors that were assessed for its influence on disease-free survival (DFS), conjunctival involvement, close margins (<4 mm), presence of Lymphovascular emboli (LVE), and clinical stage of disease and whether adjuvant treatment was received were found to have a significant influence in univariate analysis [Table 3]. On multivariate analysis, only the margin status, i.e., close margin of <4 mm influenced the DFS the most [Figure 2]. The 5 years' DFS was 70%.
|Figure 2: Graphs depicting the disease-free survival (overall) along with some of the factors influencing the same|
Click here to view
| > Discussion|| |
Eyelid malignancies form 5%–10% of skin cancer. Eyelid carcinoma is mostly attributed to various environmental factors such as exposure to sunlight, apart from certain genetic factors, and ethnicity. Hence, the prevalence of eyelid tumors shows geographical variations. The most common type of eyelid carcinoma described in literature has been BCC. This, however, does not match with the case series that have been published from the Asian subcontinent, where SGC has been reported to be the most common type.,,,,,, This may be due to the genetic predisposition and skin pigmentation of the Asian population, which makes them less prone to sunlight-induced skin cancer, in comparison to the western population. This finding is consistent in the present series, were SGC was the most common eyelid carcinoma (37%) followed by SCC (27%) and BCC (23.5%). Eyelid carcinoma is commonly seen in elderly patients and commonly involves the lower eyelid as seen in our series. As mentioned earlier, eyelid carcinoma is misdiagnosed and treated as inflammatory lesions. Twelve patients in our series (23.5%) presented after having received surgical treatment from outside for the eyelid lesions, which was subsequently reported to be malignant. This could add to the morbidity of treatment but did not have an effect on the treatment outcome on statistical analysis.
It is considered that the upper eyelid and lateral part of lower eyelid drain into the preauricular nodes the medial canthal area and medial part of lower eyelid drain along the lymph nodes in proximity to the angular vessels and facial vessels which subsequently drains into the submandibular region., Lymphoscintigraphy study looking at the lymphatic drainage pattern of eyelid have found conflicting results. According to Nijhawan et al., most of the eyelid tumor drains in the parotid region irrespective of the site of the tumor. The parotid basin is generally considered as the first echelon for lymphatic drainage from the eyelid, and the lymphatics from here would descend into the neck, especially to Level II. Rates of regional lymph node metastasis in tumors involving the eyelid and conjunctiva are up to 7%–24% for SCC, 7%–30% for Sebaceous carcinoma, and 11% for eyelid melanoma. The incidence of nodal metastasis in our series was 14%. The nodal basins were addressed only when there were clinically or radiologically suspicious/indeterminate nodes or for advanced staged disease. The protocol followed in this series was to perform a superficial parotidectomy with Level II sampling and frozen section analysis, in the presence of metastatic node further nodal clearance was planned. Superficial parotidectomy was done in nine patients however all four nodal metastasis in the parotid gland were reported in the final histopathology report only. Level II sampling was done in four patients, none harbored metastatic nodes. Selective neck dissection (SND) (I–III) and Modified neck dissection (MND) were done in one patient each. Yin et al. in their series had concluded that out of 65 patients, lesion of size with T2b and beyond, especially a pT-size ≥18 mm, had higher chance of nodal metastasis and their 3-year survival was 79%., In the present series, the pathological node-positive disease was seen in lesions more than 20 mm in size and T3A/T3B lesions (>T2B lesions). In the multivariate analysis, however, the nodal positivity did not influence the DFS or overall survival.
Margins to be obtained for eyelid carcinoma are crucial both in terms of obtaining a good locoregional control as well as reducing the morbidity of resection, given the functional importance of the eyelid. Margins suggested for BCC is at least 4 mm, and that for SCC and SGC is 4–6 mm. Similarly, Muquit MM et al. have mentioned that in cases of sebaceous cell carcinoma, it is safe to have at least 4 mm of margin. Hence, we have taken 4 mm margin as our cut off and found that 18 cases had close margin, majority of which were either sebaceous carcinoma or SCC. Of the 18 cases with close margin, 6 cases had a recurrence. On univariate analysis, margins did affect DFS (P = 0.02) [Table 3] and [Figure 2]. None of the patients in our series had positive margins. Eleven patients received adjuvant RT only.
Seven recurrences were seen in our series, with all them occurring within 24–48 months after primary treatment. Local recurrences were seen in three patients, and four patients developed regional recurrences. We had no case with distant metastasis. Various series mention different factors being responsible for recurrence such as the presence of perineural spread, delay of treatment (>6 months), pagetoid spread of SGC, and among others. However, in our series, the margin status (P = 0.001) (hazard ratios = 15.9 [95% confidence interval: 1.8–135.2]) only affected the DFS significantly in the multivariate analysis [Table 3] and [Figure 2]. To the best of our knowledge, our series is one of the few series which exclusively looks at eyelid carcinoma, along with the treatment outcomes and factors influencing the same.
The limitations of our study, however, are its retrospective design with cases inclusive of all histopathological types which have variable biological behavior. There is a need for a prospective multicentric study, given the rarity of the malignancy, to further verify our findings in a larger sample size.
| > Conclusion|| |
Eyelid carcinoma is rare tumors of the head and neck. It was predominantly seen in the elderly population. Sebaceous gland carcinoma was the most common histopathological type with the lower eyelid being the most common subsites involved. All patients underwent surgery and when indicated adjuvant RT. Margin status influenced the DFS the most. Eyelid carcinoma when treated appropriately does well with acceptable morbidity and survival.
- To the Department of Medical Records, Tata Memorial Centre
- Dr. Atanu Bhattacharjee, Biostatistician, Tata Memorial Centre, Mumbai.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Kersten RC, Ewing-Chow D, Kulwin DR, Gallon M. Accuracy of clinical diagnosis of cutaneous eyelid lesions. Ophthalmology 1997;104:479-84.
Cook BE Jr., Bartley GB. Epidemiologic characteristics and clinical course of patients with malignant eyelid tumors in an incidence cohort in Olmsted county, Minnesota. Ophthalmology 1999;106:746-50.
Lee SB, Saw SM, Au Eong KG, Chan TK, Lee HP. Incidence of eyelid cancers in Singapore from 1968 to 1995. Br J Ophthalmol 1999;83:595-7.
Lin HY, Cheng CY, Hsu WM, Kao WH, Chou P. Incidence of eyelid cancers in Taiwan: A 21-year review. Ophthalmology 2006;113:2101-7.
Casson P. Basal cell carcinoma. Clin Plast Surg 1980;7:301-11.
Gupta P, Gupta RC, Khan L. Profile of eyelid malignancy in a tertiary health care center in North India. J Cancer Res Ther 2017;13:484-6.
Wang JK, Liao SL, Jou JR, Lai PC, Kao SC, Hou PK, et al.
Malignant eyelid tumours in Taiwan. Eye (Lond) 2003;17:216-20.
Kale SM, Patil SB, Khare N, Math M, Jain A, Jaiswal S. Clinicopathological analysis of eyelid malignancies – A review of 85 cases. Indian J Plast Surg 2012;45:22-8.
] [Full text]
Jahagirdar SS, Thakre TP, Kale SM, Kulkarni H, Mamtani M. A clinicopathological study of eyelid malignancies from central India. Indian J Ophthalmol 2007;55:109-12.
] [Full text]
Takamura H, Yamashita H. Clinicopathological analysis of malignant eyelid tumor cases at Yamagata University hospital: Statistical comparison of tumor incidence in Japan and in other countries. Jpn J Ophthalmol 2005;49:349-54.
Deprez M, Uffer S. Clinicopathological features of eyelid skin tumors. A retrospective study of 5504 cases and review of literature. Am J Dermatopathol 2009;31:256-62.
Paul S, Vo DT, Silkiss RZ. Malignant and benign eyelid lesions in San Francisco: study of a diverse urban population. Am J Clin Med 2011;8:40-6.
Wang CJ, Zhang HN, Wu H, Shi X, Xie JJ, He JJ, et al.
Clinicopathologic features and prognostic factors of malignant eyelid tumors. Int J Ophthalmol 2013;6:442-7.
Muqit MM, Roberts F, Lee WR, Kemp E. Improved survival rates in sebaceous carcinoma of the eyelid. Eye (Lond) 2004;18:49-53.
Jordan DR, Anderson RL. Surgical Anatomy of the Ocular Adnexa: A Clinical Approach. San Francisco, CA: American Academy of Ophthalmology; 1996. p. 95-7.
Whitnall SE. The Anatomy of the Human Orbit and Accessory Organs of Vision. 2nd
ed. New York: Oxford University Press; 1932. p. 174-6.
Nijhawan N, Marriott C, Harvey JT. Lymphatic drainage patterns of the human eyelid: Assessed by lymphoscintigraphy. Ophthalmic Plast Reconstr Surg 2010;26:281-5.
Silverman N, Shinder R. What's new in eyelid tumors. Asia Pac J Ophthalmol (Phila) 2017;6:143-52.
Yin VT, Merritt HA, Sniegowski M, Esmaeli B. Eyelid and ocular surface carcinoma: Diagnosis and management. Clin Dermatol 2015;33:159-69.
Nasser QJ, Roth KG, Warneke CL, Yin VT, El Sawy T, Esmaeli B. Impact of AJCC 'T' designation on risk of regional lymph node metastasis in patients with squamous carcinoma of the eyelid. Br J Ophthalmol 2014;98:498-501.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3]