Article Text

Failure of transpupillary thermotherapy (TTT) for choroidal melanoma: two cases with histopathological correlation
  1. The New York Eye Cancer Centre, New York City, The Departments of Ophthalmology and Laboratory Medicine, The New York Eye and Ear Infirmary, New York City, and The Department of Ophthalmology, The New York University School of Medicine, New York City, New York, USA
  1. pfinger{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Editor,—We are concerned about the widespread use of transpupillary thermotherapy (TTT). Two patients referred for evaluation after TTT failed to destroy their tumours underwent ophthalmic oncology evaluations, were diagnosed as having recurrent choroidal melanoma, and opted for enucleation as definitive treatment. Clinical history and examination revealed evidence of a TTT laser effect with partial chorioretinal attenuation and temporary reductions in tumour height. Histopathology revealed attenuation of the retina and choroid above viable appearing melanoma. Despite temporary reductions in tumour height, resistant and/or untreated choroidal melanoma at the deep margins grew, resulting in TTT failure.


Case 1

A 47 year old man developed a small (8 × 7 × 1.9 mm) macular choroidal melanoma. Before TTT, the tumour's posterior margin was 2.25 mm from the foveola and the patient's visual acuity was 20/20 left eye (with metamorphopsia).

Three months after TTT, the apical tumour height decreased to 1.1 mm but the posterior margin remained elevated and was “heavily” retreated. Three months later a third TTT was administered, resulting in “no visible tumour”. At 30 months a crescent of pigmentation was noted along the posterior margin of the TTT scar (Fig 1) and treated immediately with a fourth and ultimately with a fifth TTT (3 months later).

Figure 1

(A) Patient 1. A colour fundus photograph reveals a large chorioretinal scar with an arc of pigmented tumour along its posterior margin consistent with recurrent melanoma. Residual pigment can also be found within the treatment area. (B) Fluorescein angiography of the posterior TTT margin reveals mottled tumour fluorescence next to the treated margin before an area of arcuate fluorescein blockage as the tumour fans out posteriorly towards the optic nerve. (C) Histopathological evaluation revealed a layer of viable malignant melanoma (arrow) beneath laser attenuated retina (haematoxylin and eosin ×40).

Three years after his initial TTT he presented to the New York Eye Cancer Centre with 20/200 vision left eye, and a marginal recurrence of his choroidal melanoma (Fig 1A). Fluorescein angiography revealed blood flow within the treated zone and recurrent tumour at the posterior margin (Fig 1B). Ultrasonography revealed a dome-shaped choroidal tumour with low internal reflectivity (2 mm in height). After a detailed discussion of observation, radiation, TTT, and enucleation he chose to undergo enucleation. Histopathology revealed viable appearing choroidal melanoma beneath laser attenuated retina (Fig 1C).

Case 2

A 73 year old woman with a (12 × 10 × 6.2 mm) choroidal melanoma, 20/63 vision, cataract, and glaucoma was given one session of TTT. “Ocular pain” and a “secondary membrane” over the tumour precluded further treatment. She sought intermittent evaluations with several physicians over 3.5 years. A series of ultrasound tumour height measurements were extrapolated from her records (6.2, TTT, 2.2, 3.4, 4.4, and 6.7)

She presented with a moderate afferent pupillary defect, a visual acuity of 20/250 (left eye), and a recurrent choroidal melanoma (6.7 mm in apical height). Histopathology again revealed a localised region of atrophic retina above a relatively fibrotic and pigmented tumour. This deep margin contained viable epithelioid melanoma cells extending to the recurrent temporal mass (Fig 2).

Figure 2

Patient 2. Histopathological evaluation revealed tumour necrosis in the path of TTT as evidenced by attenuated retina (arrow). Again, a basal layer of intact choroidal melanoma can be seen at the tumour base and within the zone of treatment (arrow). The tumour mass is largely recurrence into the temporal quadrants.


Transpupillary thermotherapy involves aiming a large (for example, 2.5 mm) spot size diode laser at posterior choroidal melanomas to completely cover the tumour surface and a circumferential (1 mm) safety margin. Power is increased until the 1 minute applications produce a light, white retinal “burn.” TTT is not consistent with clinical hyperthermia where tumour temperatures are measured and radiation is typically added.1 2

Korver and colleagues introduced the use of TTT for choroidal melanoma.3 Their histopathological evaluations demonstrated variable depths of TTT penetration.4 In order to address the possibility of untreated deep tumour, Oosterhuis advocates a “sandwich technique” adding plaque radiotherapy to treat the tumour's base.5 Shields and Capone have advocated the use of TTT alone.6 7

Clinical and histopathological studies suggest that laser (xenon, argon, or krypton) photocoagulation of choroidal melanoma is relatively ineffective.8 9 Laser treated melanomas typically demonstrate residual tumour which can grow over time. Radiation therapy also results in a tumour residua, but irradiated melanoma is considered sterilised (incapable of growth or metastasis).10 Surgeons also define successful TTT as cessation of growth or shrinkage, but like other forms of laser photocoagulation, there is no reason to believe that residual heat treated tumour has been sterilised.1

Radiation has offered mean 5 year local control rates of 92.8%.10 TTT has thus far demonstrated similar control rates of 93%, but only with a mean 16 months' follow up after treatment of small melanomas, and 94% after a mean 14 months (medium sized tumours).6 7

The ability of TTT to preserve vision has been poor. Because of aiming considerations, the perfect tumours for TTT are small, posterior, and melanotic. Case selection makes them relatively close to the optic nerve and/or fovea. This may be why TTT has resulted in a loss of vision in 43% of patients with small melanomas and 41% in Shields' series.6 7 TTT induced scotomas are typical. Traction maculopathy, haemorrhages, and optic nerve damage are common. Though these findings can also occur after ophthalmic radiotherapy they are rarely immediate.10

We present two cases where TTT resulted in inadequate treatment, leaving the ominous finding of malignant tissue at the tumours base adjacent to scleral emissary vessels. We feel that TTT should be scientifically compared with standard treatments before it is presented as standard therapy.


Supported (in part) by The EyeCare Foundation, Inc, New York City (http://www.eyecarefoundation. org).