Enucleation Following Transpupillary Thermotherapy of Choroidal Melanoma: Clinicopathologic Correlations

Transpupillary thermotherapy (TTT) is being used increasingly for the treatment of small and some medium-sized choroidal melanomas.^1-3 Although several studies have outlined the complications following TTT, such as tumor recurrence, ³ vascular occlusions, ⁴ visual field defects, ⁵ and retinal detachment, ² there are only a few studies dealing with histopathologic findings following TTT.^6-8 In this article, we present clinicopathologic correlations in 10 eyes that required enucleation after TTT in an attempt to improve our understanding regarding potential limitations of TTT.

We prospectively collected data from all patients with a diagnosis of uveal melanoma who were treated with planned TTT at the Ocular Oncology Service at Wills Eye Hospital (Philadelphia, Pa) between January 1995 and September 2001.³ Only those patients who eventually required enucleation formed the basis of the present study. Institutional review board approval was obtained.

Tumor progression was defined as any increase in tumor thickness or basal diameter detected by ophthalmoscopy, fundus photography, or ultrasonography during the first 3 treatment sessions, without any documented regression. Tumor recurrence was defined as tumor growth after a period of stable regression. All eyes were fixed in neutral buffered formalin and processed routinely for light microscopy. All available histopathologic sections of each globe were reviewed retrospectively by an experienced ophthalmic pathologist. The findings at the site of primary treatment, margins, and away from the primary site were assessed, tabulated, and correlated with clinical findings.

The general information regarding patient demographics and tumor characteristics is presented in Table 1. Among 357 consecutive patients who received TTT as primary treatment of choroidal melanoma, 10 patients (2.8%) eventually required enucleation (Figure 1). Two tumors were clinically amelanotic. Eight tumors touched the optic disc and were classified as juxtapapillary in location. The mean size of the choroidal melanomas was 7.4 mm in basal diameter and 3.1 mm in thickness, requiring an average of 3 sessions of thermotherapy.

Of the total 10 cases requiring enucleation, 8 cases (80%) were related to the failure of the primary tumor to respond, which was either due to tumor progression (3 cases) or to tumor recurrence (5 cases). One case each was enucleated because of persistent rhegmatogenous retinal detachment and neovascular glaucoma (Figure 2). The mean interval between TTT and enucleation was 19.4 months (range, 6-70 months).

The histopathologic findings based on multiple sections from each globe are summarized in Table 2. At the site of thermotherapy application, well-demarcated, full-thickness retinal atrophy and retinal and choroidal fibrosis were observed in all cases (Figure 3). The extent of fibrosis was variable; in 8 cases, fibrosis was minimal, and in 2 cases, it was massive, extending from the level of the retinal pigment epithelium to the underlying choroid. Viable-appearing choroidal melanoma was observed within the areas of TTT application in 5 cases and was noted at the margins of the treatment area in 6 cases. Tumor cells were evident intrasclerally in 3 cases, and 4 cases had extrascleral extension (Figure 3). The tumor was present within the lumina of the intrascleral emissary canals, or had directly infiltrated the scleral lamellae or both.

The level of hyperthermia during TTT is influenced by many variables such as duration, power, and spot size of the laser beam, and tumor characteristics such as thickness and extent of pigmentation.¹ The potential complications of TTT can be anticipated by effects of hyperthermia on the overlying retina and retinal vessels. The sclera is known to be resistant to hyperthermia.⁹

In all cases in our study, TTT treatment sites were characterized by well-demarcated and abruptly margined areas of retinal atrophy and fibrosis. The underlying sclera seemed normal in cases without extrascleral extension.⁸ In one of our cases, complete atrophy of the overlying retina led to irreparable rhegmatogenous retinal detachment. In 2 cases, there was exuberant retinal fibrosis that obscured the ophthalmoscopic view of the underlying choroidal melanoma that was noted histopathologically. In one case, tumor recurrence was seen only extraocularly without any intraocular component. Although retinal vascular occlusions involving small vessels are fairly common(69%-83% of cases) following TTT, retinal neovascularization is relatively uncommon.⁴ In one of our cases, florid retinal neovascularization led to neovascular glaucoma.

Overall tumor-related complications such as progression and recurrence accounted for the majority of enucleations. Recurrent tumor following TTT can be managed by additional thermotherapy, plaque radiotherapy, or enucleation based on location and extent of the recurrent tumor and the visual potential. The post-TTT enucleation rate of 2.8% in our series compares favorably with the published rates of 2.6% to 7.0%.^4,10,11

In general, the risk of tumor recurrence (about 4% at 1 year and 22% at 3 years) can be minimized by proper case selection and treatment beyond the visible tumor margins.³ The mean time to recurrence is almost 2 years after the initiation of TTT, implying the need for careful long-term follow-up of the patients.³ In our series, 8 cases (80%) were juxtapapillary in location at initial visit—a risk factor identified to predispose to tumor recurrence.³ It is not known whether the risk for recurrence with juxtapapillary tumors is due to difficulty in treating tumors at this location or whether such tumors are inherently more aggressive.

Intrascleral choroidal melanoma was noted histopathologically in 3 cases, and 4 cases developed extrascleral extension. Tumor infiltration of scleral lamellae has been observed in approximately 50% of eyes undergoing primary enucleation.¹² Melanoma cells sheltered in posterior intrascleral emissary canals or scleral lamellae are a potential source of tumor recurrence.⁵ Tumor recurrence may occur intraocularly or extraocularly. Therefore, patients treated with TTT should routinely undergo B-scan ultrasonography to detect extrascleral extension even if they demonstrate satisfactory ophthalmoscopic appearance of regression. Concerns regarding the lack of efficacy of TTT against intrascleral melanoma^2,5,8 have led to a concept of "sandwich therapy."^2,11,13 It is proposed that the combination of TTT and plaque radiotherapy may reduce the risk of tumor recurrence following TTT.¹⁴ Currently, guidelines for use of this sandwich therapy have not been clearly established.

This investigation was supported by the Sarah B. Kant Fund (Dr Singh) from the Eye Tumor Research Foundation, Philadelphia, Pa; the Paul Kayser Award of Merit in Retinal Research (Dr J. A. Shields), Houston, Tex; the Macula Foundation (Dr C. L. Shields), New York, NY; and the Noel T. Simmonds and Sarah L. Simmonds Endowment for Ophthalmic Pathology (Dr Eagle), Wills Eye Hospital, Philadelphia.

Corresponding author: Arun D. Singh, MD, Oncology Service, Wills Eye Hospital, 900 Walnut St, Philadelphia, PA 19107 (e-mail: arunsingh@eyetumors.com).