Vol 34 No 1 Original Article

Screening for intraocular metastases in patients with breast carcinoma

Emil Anthony T. Say, MD, Gary John V. Mercado, MD

ACCORDING to the Department of Health (DOH), cancer ranks third among the leading causes of morbidity and mortality in the Philippines, after infectious and cardiovascular diseases.1 The DOH–Rizal Cancer Registry and the Philippine Cancer Society–Manila Cancer Registry listed breast cancer as the most commonly diagnosed type of cancer in 2006. The Cancer Institute of the University of the Philippines–Philippine General Hospital (UP-PGH) also reported breast cancer as the most common carcinoma seen at the outpatient clinic, with more than 10,000 cases each year.

Although intraocular metastasis most frequently originates from the breast,2 it was once a rarity overall. Perls was the first to describe intraocular metastasis in 1872,3 and a subsequent article by Godtfredsen reported its incidence at 0.07%.4 However, post-mortem studies have shown that among patients who died of cancer, 12.2% had histopathologic evidence of intraocular metastases.5 This was supported by another study involving donor eyes with a prevalence of 12.6%.6 Two studies determined the frequency of intraocular metastases from breast carcinoma. Mewis and Young7 examined 250 patients with breast carcinoma and detected 67 (26.8%) positive cases. They also reported that 9% of patients who were asymptomatic also had intraocular metastases. Wiegel and colleagues performed a prospective screening program for 120 asymptomatic patients with disseminated breast carcinoma and found 6 (5%) with intraocular metastases.8 These studies showed that metastases to the eye may not be as rare as once thought.

The goal of cancer care has always been to improve the patient’s quality of life. This should include preservation of vision and the globe. Early diagnosis and prompt treatment is essential to reducing visual morbidity and improving the patient’s way of life.7, 9 Accurate staging is also a must since 9.2% of patients without visual complaints may also have intraocular metastases, with more than 60% of them having the eye as the first sign of systemic spread.7 However, before screening methods are proposed, prevalence rates need to be estimated and risk factors established.

Breast carcinoma is the most commonly diagnosed cancer in the Philippines and the most common origin of intraocular metastases. We conducted this study to determine the prevalence of intraocular metastases among patients with stages II, III, and IV breast cancer. We also compared the prevalence rates of metastases among the different stages of breast cancer and determined factors associated with increased or decreased risk of intraocular metastases.


Patients, 18 years or older, with histopathologically proven breast carcinoma classified as stage II, III, or IV by the tumor-node-metastasis (TNM) staging at the time of examination, and no history of ocular surgery for metastatic intraocular or periocular tumors, were consecutively recruited from the UP–PGH Cancer Institute. Informed consent was obtained from all subjects after thorough explanation of the nature and possible risks and benefits of the study. Excluded were patients with hematologic malignancies involving the breast and those with direct globe involvement secondary to primary or metastatic tumors from periocular structures. Patients with contraindications to a complete ophthalmologic examination, including a dilated-fundus exam, were also excluded.

The study was done in accordance with the Declaration of Helsinki and was approved by the institution’s ethics review board.

All patients underwent a single ophthalmologic examination. Pertinent information regarding breast-cancer characteristics and historical data were gathered and recorded on a standard screening form. Ancillary procedures, such as ultrasonography, ultrasound biomicroscopy, and ocular coherence tomography were performed when necessary. All presumed cases were confirmed by an ocular oncologist. All positive cases were documented with slitlamp and fundus photography.

Data were encoded in Microsoft Excel (Microsoft Corporation, Redmond, WA, USA). Prevalence rate was obtained for the study population and for each stage. Factors, such as receptor status, histopathologic diagnosis, eye symptoms, and previous treatment, were also analyzed using chi-square test for any associations with detection of intraocular metastases and tested for significance with a 95% confidence interval.


A total of 254 patients, all females aged 20 to 81 years, were included in the study. Only 71 patients were symptomatic at the time of examination. The most common symptom was blurring of vision (Table 1).

The predominant histopathologic diagnosis was invasive ductal carcinoma (92.5%), followed by invasive lobular carcinoma (3.1%), and mucinous carcinoma (1.1%). There were 106 patients with stage II (41.7%) cancer, 116 with stage III (45.6%), and 32 with stage IV (12.5%). Receptor-status positivity was present in 43.9% of patients for estrogen receptors (ER) and 43.5% for progesterone receptors (PR) where information on receptor status was available. Majority of patients had negative receptor status. Twenty-three patients had not received treatment at the time of examination, while those who had received treatment, 95.7% had surgery and the remainder received primarily neoadjuvant chemotherapy (Table 1).

Five patients had evidence of intraocular metastases; all had blurring of vision. Three had stage-IV breast carcinoma, underwent modified radical mastectomy, and completed 6 cycles of chemotherapy. One was receiving hormonal therapy. Two who had stage-III breast carcinoma had not received treatment at the time of examination. All had invasive ductal carcinoma. The time from initial diagnosis of breast cancer to detection of intraocular metastases ranged from one month to four years (Table 2). Two patients had bilateral involvement, and three had unilateral involvement, 2 in the left eye and one in the right. Only one patient had positive ER/PR. All positive cases had choroidal metastases, and three patients had iris metastases. There were no cases of ciliary body, opticnerve, or retinal metastases. All cases of choroidal metastases had an associated serous retinal detachment, while 1 eye had secondary angle-closure glaucoma (Table 2).

The presence of eye symptoms (RR = 2.6) and the clinical staging (RR = 10.4) were the only significant risk factors for the presence of intraocular metastases. Histopathologic diagnosis, age, and treatment were not associated with the presence of intraocular metastasis (Table 3). Because 2 of the 5 patients with intraocular metastases had not received any treatment at the time of examination, receptor status was not available and correlation could not be accurately determined. The calculated overall prevalence rate from the study was 1.97%; the calculated rate per stage was 9.3% for stage IV, and 1.7% for stage III. There were no positive cases among all stage II patients.


The overall prevalence rate derived from the study was 1.97%; this was different from the rates reported in postmortem studies,5-6 and closer to the reported prevalence from screening of asymptomatic breast-cancer patients.8 However, when the clinical staging was considered, the prevalence rate for stage IV was similar to that in postmortem studies of patients who died of disseminated carcinoma. A possible explanation may be the racedependent variability on the distribution of intraocular metastasis. A report on metastatic choroidal tumors in Taiwan showed that cancer in the lung was the most common origin of metastatic choroidal tumors (18/36 cases, 50%), followed by breast cancer (8/36 cases, 22.2%).10 Another possible explanation is that baseline patient demographics were not similar. In the report by Mewis and Young,7 61.1% of the patients had ocular complaints compared with only 27.9% in this study, which may reflect the large differences in prevalence rates (26.8% vs. 1.97% respectively). The correlation observed in this study between eye symptoms and the presence of intraocular metastasis further supported this.

It is interesting to note that there were 2 patients who had intraocular metastasis as the first sign of disseminated carcinoma; both patients were not receiving any treatment at the time of evaluation and the detection of intraocular metastasis resulted in change of treatment plan. In this study, more cases of intraocular metastases were found in the left eye, with a ratio of 1.5:1, similar to previous reports.9,11 In contrast, Mewis and Young7 found a small but insignificant difference in laterality. Because of the limited number of positive cases, however, a definite association could not be established. The choroid was also the most common site of intraocular metastasis (7/10 eyes, 70%), followed by the iris (3/10 eyes, 30%), consistent with previous reports.2, 12

To the best of our knowledge, this is the first study to correlate age, histopathologic diagnosis, eye symptoms, and treatment with detection of intraocular metastasis. There was a previous report by Wiegel and colleagues8 that showed increased risk of intraocular metastasis with the presence of lung and brain metastases. They were not able to show a similar correlation between nodal status, tumor size, and receptor status. Since all the patients with intraocular metastasis in this study had invasive ductal carcinoma, a definite correlation cannot be derived. Analysis of other factors showed that patients who were symptomatic at the time of the examination and who had disseminated carcinoma were more likely to have intraocular metastasis. As this study had a limited number of positive cases, casecontrol studies are needed to confirm the findings.

In summary, the overall prevalence rate of intraocular metastasis from breast carcinoma was 1.97% at the Philippine General Hospital. The prevalence rates per stage were 9.3% for stage IV and 1.7% for stage III. There was a statistically significant increase in the risk of intraocular metastasis in stage IV breast carcinoma, especially in those with visual symptoms. Case-control studies are recommended to confirm a definite association between risk factors and detection of intraocular metastasis.

Ngelangel CA, Wang EHM. Cancer and the Philippine Cancer Control Program. Jpn J Clin Oncol 2002; 32: S52-S61.
Shields CL, Shields JA, Gross NE, et al. Survey of 520 eyes with uveal metastases. Ophthalmology 1997; 104: 1265-1276.
Perls M. Carcinose capillararembolie det chorioidea. Virchows Arch Pathol Anat 1872; 56: 445-448.
Godtfredsen E. On the frequency of secondary carcinoma of the choroid. Acta Ophthalmol 1944; 22: 394-400.
Bloch RS, Gartner S. The incidence of ocular metastatic carcinoma. Arch Ophthalmol 1971; 85: 673-675.
Eliassi-Rad B, Albert DM, Green WR. Frequency of ocular metastases in patients dying of cancer in eye bank populations. Br J Ophthalmol 1996; 80: 125-128.
Mewis L, Young SE. Breast carcinoma metastatic to the choroid: analysis of 67 patients. Ophthalmology 1982; 89: 147-151.
Wiegel T, Kreusel KM, Bornfeld N, et al. Frequency of asymptomatic choroidal metastasis in patients with disseminated breast cancer: results of a prospective screening programme. Br J Ophthalmol 1998; 82: 1159-1161.
Brady LW, Shields JA, Augsburger JJ, Day JL. Malignant intraocular tumors. Cancer 1982; 49: 578-585.
Wang TJ, Yang CM, Ho TC, et al. Metastatic choroidal tumors in Taiwan: An 11- year experience. Am J Ophthalmol 2005; 140: 735-737.
Stephens RF, Shields JA. Diagnosis and management of cancer metastatic to the uvea: a study of 70 cases. Ophthalmology 1979; 86: 1336-1349.
Shields CL, Shields JA, Gross NE, et al. Survey of 520 eyes with uveal metastases. Ophthalmology 1997; 104: 1265-1276.
Demirci H, Shields CL, Chao AN, Shields JA. Uveal metastasis from breast cancer in 264 patients. Am J Ophthalmol 2003; 136: 264-271.