Metastatic Tumors to the Eyelid: Title and subTitle BreakReport of 20 Cases and Review of the LiteratureMetastatic Tumors to the Eyelid FREE

Metastases to the eyelids are rare, accounting for less than 1% of all malignant eyelid lesions.^{1 - 4} In a survey of 892 eyelid lesions from an ocular pathology laboratory, Aurora and Blodi¹ reported that 214 (24.0%) were malignant. The findings included basal cell carcinoma (172 [19.3%]), squamous cell carcinoma (15 [1.7%]), malignant melanoma (11 [1.2%]), and sebaceous gland carcinoma (7 [0.8%]). There were only 3 cases (0.3%) of eyelid metastasis. Similarly, in a review of 1502 eyelid lesions evaluated histopathologically, Arnold and coworkers² found only 1 case (0.07%) of eyelid metastasis.

In the ophthalmic literature, information on eyelid metastases has been reported mostly as individual case reports and a few small case series.^{5 - 10} Riley¹¹ reported 15 cases of eyelid metastasis from a pathology laboratory registry and found that the most common primary tumor was breast carcinoma (6 [40%]), followed by skin melanoma (5 [33%]), gastric carcinoma (2 [13%]), uveal melanoma (1 [7%]), and lung carcinoma (1 [7%]). Later, Mansour and Hidayat¹² reviewed 31 cases from the Armed Forces Institute of Pathology and found that the most frequent primary site was the breast (11 [35%]), followed by the skin (4 [13%]), gastrointestinal tract (3 [10%]), genitourinary tract (3 [10%]), and uvea (2 [6%]). These studies focused primarily on the histopathologic features of eyelid metastases, with little information on the clinical aspects and patient outcomes. In this study, we describe the clinical features, systemic and local treatment, and outcome of 20 patients with eyelid metastasis who were diagnosed and managed at a single ocular oncology center across 3 decades.

We reviewed the medical records of all patients with eyelid metastasis that was managed at the Ocular Oncology Service, Wills Eye Institute, from August 1977 through July 2007. The study included patients with eyelid metastasis from noncontiguous solid tumors. Patients who had orbital tumors or tumors that invaded the surrounding conjunctiva were excluded, as were patients with lymphoma because of multiorgan involvement. Internal review board approval at Wills Eye Institute was obtained.

The data collected included information on the primary tumor and the eyelid metastasis. Primary tumor information included patient age at the time of diagnosis, the involved organ, and the primary tumor treatment (ie, surgical treatment, external beam radiotherapy [EBRT], and chemotherapy). We evaluated the sites and management of associated nonocular systemic metastases and the interval between the diagnosis of the primary tumor and the eyelid metastasis (in months).

The patients' ages at the time of ocular diagnosis and symptoms were recorded. Ophthalmic data included laterality and best corrected Snellen visual acuity. Tumor data included the color (melanotic or amelanotic), number, maximum tumor dimensions (in millimeters), and shape (sessile or pedunculated) of the tumor and the presence of other ocular metastases. The anatomic location of the eyelid metastasis was classified as follows: involved quadrant (upper eyelid, lower eyelid, or medial canthal or lateral canthal area [Figure 1A]); sagittal location (cutaneous or subcutaneous [Figure 1B]); and coronal location (eyelid margin, eyebrow or cheek, or in between [Figure 1C]).

The management of eyelid metastasis (surgical excision, EBRT, chemotherapy, and observation) was recorded. Follow-up information was collected regarding visual acuity at the last visit, response to therapy (regression, recurrence, or stable or active disease), complications of therapy, and patient outcome (alive with metastasis, dead of metastasis, or dead of another cause). If there was a recurrence or development of a new eyelid metastasis, the location, date of the recurrence or new metastasis, and treatment were recorded.

From August 1977 through May 2008, 1773 patients had ocular metastasis, only 20 of whom (1.1%) had eyelid metastasis. The mean age at initial diagnosis of eyelid metastasis was 60 years (median, 60 years; range, 33-91 years). There were 11 women (55%) and 9 men (45%). Nineteen patients were white and 1 was African American. Diagnosis of the eyelid metastasis was performed histopathologically in 16 patients (80%) and was based on clinical findings in 4 patients (20%) owing to widespread systemic disease.

For each case, the primary tumor and its treatment are described in Table 1. In 17 patients with known primary tumors, eyelid metastasis developed a mean of 62.3 months after the diagnosis of the primary tumor (median, 33 months; range, 1-432 months). In this group of patients, the primary tumors were skin melanoma (4 [20%]) (Figure 2), conjunctival melanoma (1 to the eyebrow and 2 to the area between the eyelid margin and the cheek of the lower eyelid [15%]), uveal melanoma (Figure 2) (3 [15%]), renal cell carcinoma (Figure 3) and breast carcinoma (2 [10%] each), and medullary thyroid carcinoma, lung adenocarcinoma (Figure 4), and salivary gland adenocarcinoma (Figure 5) (1 [5% each]). The eyelid metastasis was the initial manifestation of a cancer in 3 patients (15%). In these patients, subsequent systemic evaluation led to the diagnosis of a primary tumor, including prostate carcinoma, breast carcinoma, and uveal melanoma (1 [5%] each).

The initial ocular symptom included a solitary eyelid nodule (12 [60%]), a flat pigmented lesion (3 [15%]), diffuse eyelid edema (3 [15%]), and epiphora and multiple eyelid nodules (1 [5%] each) (Table 2). The right eye was involved in 12 patients (60%), the left eye in 6 patients (30%), and both eyes in 2 patients (10%). The eyelid metastasis was located in the superior eyelid in 7 patients (35%), the inferior eyelid in 6 patients (30%), the lateral canthal area in 3 patients (15%), the medial canthal area in 2 patients (10%), and both upper and lower eyelids in 2 patients (10%) (Figure 1A). The eyelid tumor was located along the palpebral margin in 5 patients (25%), the area between the eyelid margin and the eyebrow or the cheek in 10 patients (50%), in the eyebrow in 1 patient (5%), and in more than 1 of these locations in 4 patients (20%) (Figure 1B). The metastasis involved only the skin in 2 patients (10%), only the subcutaneous tissue in 17 patients (85%), and both of these areas in 1 patient (5%) (Figure 1C).

The eyelid mass was pigmented or dark in 10 patients (50%) and amelanotic in 10 patients (50%). The shape of the eyelid metastasis was sessile in 16 patients (80%) and pedunculated in 1 patient (5%) and showed diffuse infiltration in 3 patients (15%). The mean maximum basal diameter was 11 (median, 10; range, 1-25) mm. Seven patients had intrinsic vascularity (35%), and 2 patients had intralesional hemorrhages (10%). The mass was firm in 19 patients (95%) and soft in 1 patient (5%). None of the patients had ulceration of the eyelid skin. Overlying eyelash loss was found in 7 patients (35%).

Associated metastasis to the uvea was found in 6 patients (30%), to the orbit in 3 (15%), and to the conjunctiva in 2 (10%). Three patients (15%) had more than 1 eyelid metastatic tumor. Two of these patients had skin melanoma and 1 had uveal melanoma. In these patients, there were small, multiple melanotic eyelid tumors involving different parts of the eyelid.

The initial management of the eyelid metastasis consisted of excisional biopsy alone (6 [30%]), EBRT (7 [35%]), observation alone (3 [15%]), and systemic chemotherapy/immunotherapy (4 [20%]) (Table 2). There were no recurrences among the 6 patients who were treated initially with excisional biopsy alone (patients 6, 7, 13, 14, 15, and 16). Of the 7 patients who were treated with EBRT, 3 tumors (43%) (patients 2, 9, and 18) showed regression, 3 tumors (43%) (patients 4, 12, and 19) were unresponsive, and 1 tumor (14%) (patient 20) remained stable. The tumor of one patient (14%) (patient 18) showed initial regression, with recurrence after 6 months of follow-up. This patient later underwent excisional biopsy followed by systemic chemotherapy with no response. Of the 4 patients who were treated with chemotherapy/immunotherapy, 1 tumor (25%) (patient 11) regressed, 1 tumor (25%) (patient 10) remained stable, and 2 tumors (50%) (patients 1 and 5) were unresponsive. Three tumors (15%) (patients 3, 8, and 17) were observed because of advanced systemic disease, and they remained stable. After a mean follow-up of 16 months (median, 10 months; range 1-48 months), 9 patients (45%) were alive and 11 (55%) had died of metastatic disease. Survival was 88%, 76%, 67%, and 67% at 3, 6, 9, and 12 months of follow-up, respectively.

Metastatic tumors rarely involve the eyelids. Ferry and Font¹³ reviewed reports from an ocular pathology laboratory of 227 patients with ocular metastases and found that the most commonly affected ocular tissue was the choroid (112 [49%]), followed by the orbit (28 [12%]) and the iris and ciliary body (26 [11%]). The authors reported no eyelid metastases in that series. Similarly, in a postmortem survey of 230 autopsied eyes from patients with cancer-related deaths, Bloch and Gartner¹⁴ described ocular metastasis in 28 patients (12%), including 23 patients (10%) who had intraocular metastasis and 5 (2%) with orbital metastasis. No cases of eyelid involvement were mentioned in that series. Hutchison and Smith¹⁵ described 100 patients with ocular metastasis from an ocular pathology laboratory. The most common ocular location was the choroid (63 [63%]), followed by the orbit (32 [32%]) and the ciliary body (6 [6%]). Three patients (3%) had eyelid metastasis. In our study, we reviewed 1773 cases of ocular metastasis from the ocular oncology service of a single institution and found 20 (1.1%) to have eyelid involvement. We have previously reported specific cases of eyelid metastasis from choroidal melanoma, prostate carcinoma, renal cell carcinoma, and medullary thyroid carcinoma.^{5 - 6 ,8 - 10}

Although eyelid metastasis usually appears in patients with a known systemic cancer, it can be the first symptom of systemic cancer or the first sign of metastasis from a known malignant neoplasm. In a series of 15 patients with eyelid metastases, Riley¹¹ reported that the eyelid metastasis was the first sign of systemic cancer in only 1 patient (7%). In the other 14 patients (93%), eyelid metastases were detected from 4 to 120 months after the diagnosis of the primary malignant neoplasm. In a histopathological review of 31 patients, Mansour and Hidayat¹² reported that eyelid metastasis was the first sign of systemic cancer in 14 patients (45%). In the other 17 patients (55%), eyelid metastases developed after a mean follow-up of 40 months (range, 2 weeks to 120 months). In our series, biopsy of the suspicious eyelid lesions led to the diagnosis of systemic cancer in 3 patients (15%). Immunohistochemistry is an essential tool for the ophthalmic pathologist dealing with an unknown primary tumor. Eyelid metastasis classically is associated with multiorgan, systemic, or other ocular tissue involvement. Riley¹¹ reported that associated multiorgan systemic metastases were found in 73% of patients, orbital metastasis in 33%, and choroidal metastasis in 7%. We observed associated multiorgan systemic metastases in 19 of our patients (95%) and ocular metastasis in 10 (50%). Patients with eyelid metastasis should have a full systemic evaluation and a detailed eye examination and orbital imaging.

The features of eyelid metastasis at the initial examination are nonspecific. Metastases can manifest as solitary or multiple eyelid nodules or as diffuse eyelid swelling. In his review of 15 patients with eyelid metastases, Riley¹¹ reported that 7 patients (47%) had a solitary nodule, 6 (40%) had painless, diffuse eyelid swelling, and 2 (13%) had ulcerative lesions; multiple nodules were not found in any patient. Mansour and Hidayat¹² found that 22 of 31 patients (71%) had a solitary nodule, 4 (13%) had blepharoptosis, 2 (6%) had induration of the eyelid (“woody” eyelid), and 2 (6%) had painless, diffuse eyelid swelling. Because of the wide-spectrum presentation of eyelid metastases, the level of suspicion should be high, especially in patients with a history of systemic cancer. Any suspicious lesion should undergo biopsy.

Morgan and associates⁷ reviewed the literature to evaluate whether eyelid metastases from certain organs display specific clinical features. They reported that metastases from breast cancer tended to be painless diffuse eyelid swelling in 7 of 9 patients (78%), whereas those from lung cancer were solitary nodular lesions in all 4 patients (100%). In our study, breast and lung carcinoma showed no specific clinical manifestation. The only suggestive feature consisted of flat, multiple pigmented eyelid lesions in 3 of the 4 patients with cutaneous melanoma.

Eyelid metastases are more common in middle-aged or older patients, although they can be seen in any age group. They have been reported in a 3-year-old child with embryonal carcinoma arising from the hip region.¹² Eyelid metastases are more often reported in women.^{11 - 12} The most common primary systemic cancer is breast carcinoma, which accounts for 35% to 40% of patients with eyelid metastases.^{11 - 12} Similarly, we have observed that the most common primary tumor for choroidal, orbital, and conjunctival metastasis is breast carcinoma^{16 - 19} (Table 3). In our study, cutaneous and uveal melanoma were the most common primary cancers. This could be related to the increasing incidence of cutaneous melanoma, to a referral bias toward our interest in melanoma, or to the biologic features of these tumors. The management of eyelid metastases includes excisional biopsy, EBRT, systemic chemotherapy/immunotherapy, and observation. The clinical features of the tumor, such as the shape, number, and location of the metastases, and the systemic features play a role in the choice of treatment. A small, solitary nodular lesion can be managed with excisional biopsy. In general, EBRT is used to treat patients with multiple eyelid metastases or recurrent lesions. If the patient is undergoing systemic chemotherapy/immunotherapy for widespread systemic disease, the tumor could be observed for response. Patients who are at the terminal stage and not receiving systemic treatment are generally observed. Newer therapies such as topical imiquimod have been used successfully as local treatment for patients with eyelid metastases from cutaneous melanoma.

Patients with multiple metastases carry a worse prognosis, regardless of treatment. This may also be associated with the better prognosis in patients with single nodules that were excised (as the only treatment) and did not recur, compared with patients with multiple or large eyelid lesions who also had multiple metastases.

Eyelid metastasis can display a variety of clinical features and should be considered in patients with known systemic cancer. These patients usually have multiple metastatic sites, ocular and nonocular. We found that the most common primary malignant neoplasms to promote eyelid metastasis include skin and uveal melanoma (20% of the patients each), followed by breast carcinoma and conjunctival melanoma (15% of the patients each). Most of the patients (90%) were older than 50 years. Eyelid metastasis was the first sign of systemic cancer in 15% of patients, stressing the importance of histopathological examinations for all excised eyelid lesions. Ten patients (50%) had 2 concomitant ocular site metastases at the initial examination. Excision was successful in 100% of the patients with a solitary nodular lesion, EBRT was successful in 43% of the patients with multiple nodular lesions, and systemic chemotherapy was successful in 50% of the patients with multiorgan involvement. Patients with eyelid metastasis have a poor systemic prognosis, and survival was 88%, 76%, 67%, and 67% at 3, 6, 9, and 12 months of follow-up, respectively.

Correspondence: Carol L. Shields, MD, Ocular Oncology Service, Ste 1440, Wills Eye Institute, Thomas Jefferson University, 840 Walnut St, Philadelphia, PA 19107 (Carol.Shields@Shieldsoncology.Com).

Submitted for Publication: September 24, 2008; final revision received November 28, 2008; accepted December 2, 2008.

Financial Disclosure: None reported.

Funding/Support: This study was supported by a donation from Michael, Bruce, and Ellen Ratner (Drs J. A. Shields and C. L. Shields); Mellon Charitable Giving from the Martha W. Rogers Charitable Trust (Dr C. L. Shields); the Eye Tumor Research Foundation (Drs C. L. Shields and J. A. Shields); and the Noel T. and Sara L. Simmonds Endowment for Ophthalmic Pathology, Wills Eye Institute (Dr Eagle).