Prior to the emergence of radiotherapy. Primary transpupillary thermotherapy (TTT) for malignant choroidal melanoma. Patients and Methods

Transcription

1 Primary transpupillary thermotherapy (TTT) for malignant choroidal melanoma B. M. Stoffelns Department of Ophthalmology, Johannes Gutenberg University Mainz, Germany ABSTRACT. Purpose: To evaluate the efficacy of transpupillary thermotherapy (TTT) as the only method of treatment for small choroidal melanoma. Patients and Methods: In a prospective non-randomized analysis, 20 patients with primary choroidal melanoma (posterior to the equator with base 2 and thickness 4.5 mm) were treated with TTT as the only method of treatment (diode laser at 810 nm, beam diameter 3 mm, power setting W, exposure time min). During follow-up of at least 6 months, clinical aspects, ultrasonographic tumour thickness, fluorescein and indocyanine green angiographic patterns, visual acuity and ocular side-effects were recorded. Results: In 17 eyes the tumour regressed significantly within 3 months after one treatment session (to a flat chorioretinal scar in 15 eyes). Despite a clinically flattened chorioretinal scar, fluorescein and indocyanine green angiography revealed that choriocapillary vessels in the heat-treated areas of 15 eyes remained perfused. Three amelanotic melanomas showed almost no response to TTT after repeated treatment at higher power settings. Visual acuity remained unchanged or improved in 12 eyes. Ocular side-effects included posterior synechia of the iris (1), macular oedema (2) and temporary retrobulbar pain (2). No patient showed tumour recurrence or metastases during follow-up. Conclusions: Preliminary results obtained by this study demonstrate good efficacy and visual outcome following TTT as the only method of treatment for small choroidal melanoma. However, indocyanine green angiographic findings suggest that tissue damage in the choroidal layer might be less effective, which perhaps may lead to a higher rate of tumour regrowth. Long-term follow-up is required to obtain data on late ocular side-effects, tumour recurrence and metastasis. Key words: melanoma choroid uvea transpupillary thermotherapy Acta Ophthalmol. Scand. 2002: 80: Copyright c Acta Ophthalmol Scand ISSN Prior to the emergence of radiotherapy as an eye- and vision-sparing alternative, enucleation represented the standard treatment for choroidal melanoma (Finger 1997). Although radiotherapy stabilizes or reduces tumour size in over 90% of cases, more than 40% of treated eyes develop significant radiation-associated complications (Lommatzsch & Kirsch 1988; Finger 1997). In most cases, ophthalmic plaque radiation is applied to the posterior segment of the eye, possibly thereby leading to an increased incidence of radiation retinopathy and optic neuropathy. Since the amount of radiation to normal ocular structures needs to be reduced in order to avoid radiation-associated complications, the amount of radiation required to sterilize uveal melanomas has to be decreased. As hyperthermia is known to enhance the effect of radiotherapy (Marmor et al. 1979; Song 1984; Leeper 1985; Svaasand et al. 1985), various devices have been designed to enable temperature increases of the melanoma tissue to between 42 and 44æC (Marmor et al. 1979; Coleman et al. 1986; Riedel 1988; Finger 1991). These represent an adjuvant treatment to plaque radiotherapy at lower levels of radiation. In 1992 and 1995, Journée-de Korver et al. reported good results for hyperthermia using near infrared radiation at 810nm as the primary treatment in animals (Journée-de Korver et al. 1992, Journée-de Korver et al. 1995). Optimum tumour necrosis was achieved on heating the tumour to a temperature between 45 and 60 æc. Other members of the same working group went on to confirm the efficacy of infrared hyperthermia for human choroidal melanoma (Oosterhuis et al. 1995, Oosterhuis et al. 1998). Near infrared radiation was delivered through the dilated pupil, and the new method was described by the authors as transpupillary thermotherapy. We present preliminary experiences with transpupillary thermotherapy (TTT) used alone in the treatment of 20 patients with posterior uveal melanoma. Patients and Methods Prospective evaluations were performed on all patients with small choroidal melanoma treated with TTT alone at the Department of Ophthalmology at the Johannes Gutenberg University Mainz, Germany, between January and Decemb- 25

2 er The eligibility criteria for treatment were based on reports in the literature of histological and clinical results gained by transpupillary thermotherapy (Journée-de Korver et al. 1995; Journéede Korver et al. 1997; Oosterhuis et al. 1998). Patients selected for the study were those with primary choroidal melanoma located posterior to the equator of the eye measuring 12mm or less at the base and 4.5mm or less in thickness. All patients had either documented growth or were subject to ophthalmoscopic risk factors for growth (Shields et al. 1995) and/ or metastasis. Patients with retinal, optic nerve, vitreous or extrascleral tumour extension, subretinal haemorrhage or media opacity prohibiting adequate visualization of the tumour were excluded. Patient data included age, race, gender and visual acuity (VA). All patients underwent standard chest radiography, ultrasonic investigation of the abdomen, liver function tests (alkaline phosphatase, lactic dehydrogenase, glutamyl transpeptidase), skull and orbital MR imaging, as well as dermatological and general physical examinations for metastatic evaluation. When these investigations produced no pathological findings, no further metastatic investigation was performed. A chest CT was carried out in patients with an abnormal chest X-ray. Abdominal MR imaging was carried out when ultrasound scans of the liver had produced abnormal results. All patients underwent an ultrasound examination of the liver once per year and liver function tests twice per year during follow-up. The diagnosis of choroidal melanoma was based on results of the ophthalmoscopic examination, ultrasonography and fluorescein angiography. The basal dimensions of the tumour were determined by B-scan ultrasonography. Tumour height was measured by A- and B-scan ultrasonography. Tumour data included tumour location (subfoveal, para- or extrafoveal, juxtapapillary), pigmentation (amelanotic, light and heavy pigmented), presence of subretinal fluid (present, absent), orange pigment (present, absent) and drusen (present, absent). Visual acuity and ultrasonographic tumour thickness were measured during follow-up, initially at 4-week intervals and then at 3-month periods. Fundus photography, fluorescein angiography and indocyanine green angiography were performed before treatment and on regression of the tumour into a flat chorioretinal scar. Additional fundus photographs were taken selectively for documentation of the regression course of selected tumours. Associated findings in the anterior and posterior eye segment were recorded. The presence of cornea, iris or lens damage, retinal artery or vein obstruction, retinal, choroidal or vitreous haemorrhage, macular oedema, surface wrinkling retinopathy, optic atrophy, neovascularization of the disc or retina, or retinal hole formation were recorded. Furthermore, a subgroup analysis based on tumour colour category melanotic versus amelanotic was carried out because the findings of the described evaluation demonstrated a correlation between the effectiveness of TTT and the grade of tumour pigmentation. The pupil was dilated with 5% phenylephrine hydrochloride and 0.25% tropicamid eyedrops prior to treatment. Retrobulbar anaesthesia was achieved with an injection of 2mL xylocaine 2% and 2 ml carbostesine for immobilization and pain prevention. TTT was performed using a specially modified infrared diode laser at 810 nm with an adjustable beam width of 2.0 or 3.0mm (Iris Medical Instruments, Mountain View, California, USA). The infrared diode laser system was adapted to a Haag- Streit slit-lamp biomicroscope (Koniz, Switzerland). The laser light was delivered through two infrared coated panfunduscope contact lenses (Mainster wide field and Mainster ultra wide field OMRA- PRP, Ocular Instruments, Bellevue, Washington, USA). The width of the laser beam used in the majority of patients was 3.0 mm; a diameter of 2.0mm was used under special circumstances to avoid the optic disc or retinal vessels. Treatment was initiated at the centre of the tumour with one spot/1 minute, using a low power setting of 300 mw. Continuous observation of the tumour surface through the slit-lamp biomicroscope was maintained throughout the treatment period. The energy of the laser beam was increased in 50mW increments over 1minute to achieve a treatment end-point where the treated spot attained a grey-white appearance. The energy of the laser beam was decreased if the colour change appeared early in the exposure period or in the presence of coagulation of retinal vessels. The maximum power setting, which should not be exceeded, was 900 mw. Spots were targeted in overlapping confluency and included mm of clinically healthy tissue from the tumour margin. Because of the different beam widths used, the energy delivery of the laser to the target area was calculated in W/cm 2. Energy levels ranged from 8 to 19W/cm 2. The number of spots targeted by the laser ranged from three to 16, depending on the diameter of the tumour base. Follow-up examinations were performed at 4-week intervals until the tumour had regressed to a flat chorioretinal scar. When complete tumour regression was assumed, the follow-up intervals were extended to 3 months. In cases of partial regression, additional treatment was performed after 3 months. Postoperatively 0.1% dexamethasone eyedrops were administered three times per day for 5days. Results Patients and tumour data Between January and December 1998, 20 patients with selected choroidal melanoma were treated using transpupillary thermotherapy as the only method of treatment. The minimum follow-up period was 6 months (mean 10 months; range 6 15 months). The mean patient age at treatment was 62.2 years (range years). All patients were white and included eight women and 12 men. Evaluation for metastatic disease in all patients revealed no sign of metastasis at the time of treatment or during follow-up. Twelve patients were asymptomatic, while eight complained of blurred vision in the affected eye at presentation. The mean tumour thickness before treatment was 3.0 mm (range mm), the mean maximum basal diameter measured was 8.2 mm (range mm). The tumour was located below the fovea in six eyes. In two eyes the tumour touched the optic disc. The posterior margin of 12 tumours touched the posterior pole but did not involve the foveal area. Exudative retinal detachment was observed preoperatively in 14 eyes, orange pigment clumps on the tumour surface in 11, and drusen on the tumour surface in three eyes. In 13 cases, tumour growth was documented prior to treatment and seven patients were subject to known risk factors for tumour growth and metastasis (Shields et al. 1995). The results for all patients are shown in Table 1. The treatment course in our series is discussed in two different groups of melanotic versus amelanotic melanoma. Course of tumour regression Group I (Fig.1) represents 17 eyes with melanotic melanoma. These eyes showed 26

3 Table 1. Results of transpupillary thermotherapy (TTT) in choroidal melanoma. Patient No. of Tumour After TTT Visual Acuity no. TTT height before Tumour height Time until Final tumour Control Before Last Location Pigment TTT, mm at 3 months, tumour flat, height, mm period, therapy control mm months months /20 20/20 e ππ /200 20/200 s ππ /200 HM s ππ HM HM s π 5 2(3) /25 20/20 e π /40 CF s ππ /50 20/40 e ππ /20 20/25 p ππ /40 20/30 p π /20 20/20 e ππ /40 20/50 e ππ /20 20/30 p π 13 2(3) /30 20/100 s /100 HM p ππ /50 20/200 p ππ 16 2(3) /30 20/30 e /30 20/20 p ππ /25 20/25 p ππ 19 2(3) /40 20/25 e /30 20/25 s ππ HM: Perception of hand motions CF: Counting fingers. Number in parentheses ( ) is the interval in months between the first and second TTT treatments. The distance in mm of the tumour margin from the fovea is designated as follows: s Ω subfoveal, p Ω parafoveal ( 3 mm), e Ω extrafoveal ( 3 mm). Grade of pigmentation: ππ Ω heavy, π Ω moderate, Ω amelanotic. good response after one treatment session. There was a gradual decrease in the mean tumour thickness from 3.0 mm prior to treatment to 2.1 mm 1 month after treatment, 1.3 mm 2 months after treatment, and 0.7 mm 3 months after treatment. Six months after treatment, the mean tumour thickness had stabilized Fig. 1. Transpupillary thermotherapy for choroidal melanoma in 20 patients: mean tumour thickness response over a 6-month period. The curve with H indicates tumour regression in 17 patients with pigmented melanoma after one treatment session. The curve with ò indicates tumour regression in three patients with amelanotic melanoma after two treatment sessions. at 0.4 mm. Fifteen eyes demonstrated tumour regression to a flattened chorioretinal scar, in some cases with visible underlying sclera. Although a second treatment session was recommended in two eyes with residual tumour prominences of 1.0 and 1.5 mm, the patients refused further treatment. The ocular status of these patients remained unchanged during follow-up. The treatment parameters for this group included power of 650 mw (range mw) for each treatment spot, total treatment time of 27 min (range 20 35min), spot size of 3mm in the majority of cases (spot size of 2mm was used in a few instances). These parameters allowed the tumour surface to achieve a blanched white appearance in all eyes. Group II (Fig. 1) represents three eyes with amelanotic melanoma that showed minimal or no regression over a period of 3 months after initial treatment. The mean tumour thickness decreased from 3.4mm prior to treatment to 3.3mm after 1month, 3.2mm after 2months, and 3.1 mm 3months after treatment. The second treatment session also proved less effective in this group. Mean tumour thickness of the amelanotic melanomas regressed only minimally to 2.3 mm over the 3-month period after the second treatment session. Retinal pigment epithelium hyperplasia at the tumour margins and considerable intravitreal pigment dispersion were noted, while surface wrinkling retinopathy was absent. Mean treatment parameters were power of 870 mw (range mw) for 32 min (range 27 35min) at the first session and 900mW in all cases for 35min (range min) at the second session. Although power levels were higher than those used in the treatment of pigmented melanoma, the end-point of treatment of all amelanotic melanomas was a faint grey colour at both sessions. Tumour spread None of the 20 patients developed local tumour recurrence over a 6-month follow-up period. No sign of intrascleral spread was noted ultrasonographically in the two eyes with iuxtapapillary tumours (Table1: patients no.4 and 11). All patients survived throughout the follow-up period and showed no evidence of melanoma metastasis. 27

4 Fig. 2. Subfoveal choroidal melanoma in a 49-year-old man (see Table 1: patient 6). (A) Fundus photograph before thermotherapy. Note the orange pigment on the tumour surface. (B). A- and B-scan ultrasound showing a thickness of 1.7 mm. (C) Fluorescein angiogram demonstrates tumour vessels with leakage in the early frames. (D) 24 hours after treatment, the tumour surface appears whitish with slight overlying haemorrhages and surrounding oedema. (E) 10 months after treatment the tumour has regressed, leaving a completely flattened scar with partial visibility of the sclera in the centre and retinal pigment epithelium clumps on the posterior sclera margin. (F) Fluorescein angiogram shows the sharp demarcation of the heat-treated area from the surrounding tissue. Note the diffuse leakage in the background. (G) Indocyanine green angiogram reveals persistence of large and medium size choroidal vessels in the outer zone of the heat-treated area. 28

5 Course of visual acuity By the end of follow-up, VA had either improved or remained unchanged in 12 eyes (60%) (Table 1). The improvement in vision was due to the resolution of subretinal fluid. Visual acuity decreased in eight eyes, primarily due to the tumours locations: six of these eyes required treatment through the foveola for subfoveal tumours. In two eyes the development of macular oedema was noted 3 and 5days after thermotherapy. Although the macular oedema partially resolved during follow-up, the finding accounted for a poorer visual outcome. Complications and side-effects Retinal oedema and fine intraretinal haemorrhages on the tumour surface frequently occurred immediately after treatment (Fig. 2). Exudative retinal detachment, present in 14 eyes prior to treatment, increased over the 7 days after treatment but resolved completely in all cases over the following 4 7 weeks. No signs of direct heat-related alterations were recorded in the cornea and the vitreous. In one case the laser beam touched the iris and we observed a local inflammatory reaction at the pupillary border which caused posterior synechia and local subcapsular cataract but did not affect vision. The amelanotic melanomas treated at higher power settings showed considerable pigment dispersion into the vitreous, but there was no occurrence of surface wrinkling retinopathy. Macular oedema was observed in two eyes containing pigmented melanomas that were partly located at the posterior pole, but did not involve the fovea. No overlying branch retinal artery or vein occlusion, neovascularization of the retina, or optic disc oedema were recorded. Two patients complained of severe retrobulbar pain after TTT. The pain occurred after 5 and 7 days respectively and was resolved within 2weeks of treatment with oral diclofenac. Fig. 3. Extrafoveal choroidal melanoma 4 months after transpupillary thermotherapy (see Table 1: patient 11). Tumour thickness regressed from initially 2.2 mm to a flattened chorioretinal scar over 2 months after TTT. (A) Fundus photograph shows pigment epithelium alterations but no damage of the overlying large retinal vessels. (B) Indocyanine green angiogram demonstrates medium size choroidal vessels remaining perfused in the heat-treated area, which is sharply demarcated. Angiographic findings Prior to treatment, the patterns of indocyanine green videoangiography of the melanoma were varied in onset and intensity of fluorescence, as well as in the visibility of the choroidal vascular pattern. The three amelanotic melanomas exhibited greater intensity and earlier onset of fluorescence (mean 36 s, range s). Intrinsic tumour vessels were well-contrasted with normal choroidal vessels. In pigmented melanoma the onset of fluorescence was delayed (mean 58s, range s) and intensity of fluorescence was reduced. The choroidal vascular pattern was blurred and in most cases only parts of the intrinsic tumour vascularity could be visualized. Two heavily pigmented melanomas were characterized by persistent dense central hypofluorescence and no tumour vessels could be identified, even on the late frame at 15min. Angiography was not performed postoperatively on the three eyes with amelanotic melanoma because they showed virtually 29

6 no regression after two sessions of TTT. Fluorescein and indocyanine green angiographies were performed after TTT only in those 17 eyes with pigmented melanoma and significant tumour regression. In those cases, the treated area was sharply demarcated by a zone of pigment epithelium atrophy. Comparison with the surrounding normal tissue distinguished the heat-treated area as nonfluorescent in 11 cases, and as hypofluorescent in six cases. Choriocapillary vessels in 15 eyes remained perfused in several areas of the thermotherapy field (Figs 2 and 3). Only two eyes showed complete atrophy of the choriocapillary layer. With indocyanine green angiography, 10 eyes showed persistent perfusion of large choroidal vessels in the treated area despite the presence of a clinically completely flattened chorioretinal scar. Branch retinal vessels remained untouched by TTT and exhibited no indication of occlusion. We then looked further for fluoresceinangiographic evidence of macular oedema and found two eyes with macular microangiopathy and oedema that had developed after TTT during follow-up. Angiographic examination showed no tissue damage outside the heat-treated area in the other 18 eyes. In contrast, VA improved in six eyes with paramacular tumours due to the resolution of subretinal fluid after TTT. Discussion Transpupillary thermotherapy has been described as a method performed at subphotocoagulation temperatures using a specially designed delivery system (Oosterhuis et al. 1995, 1998). Journée-de Korver et al. emphasised that intermediate hyperthermia at a temperature between 45 and 60æ C proved optimal in achieving tumour necrosis, as it is characterized by deep penetration and exerts an immediate cytotoxic effect on the melanoma resulting in tumour necrosis to a depth of 6.0mm in animal experiments and 4.7 mm in human choroidal melanoma (Journée-de Korver et al. 1997). High-level hyperthermia at 60 æc or above induced coagulation identified by the whitening of the tumour surface. The appearance of a white coagulation lesion early in TTT application exerts an adverse effect in that any increase in the quantity of laserlight reflected reduces its transmission into tumour tissue. In our study TTT was interrupted immediately when the tumour tissue changed to a whitish colour. After one treatment session, 15 of 20 heat-treated tumours regressed to a flat chorioretinal scar over periods of 3 6 months. These results are comparable to others reported in the literature (Oosterhuis et al. 1998; Godfrey et al. 1999; Robertson et al. 1999; Schneider et al. 1999; Stoffelns 1999). Tumour regression after TTT is generally more rapid than that following plaque radiotherapy for tumours of a similar size (Leeper 1985; Marmor et al. 1979; Nuijs-Beems et al. 1990). Necrosis and vascular occlusion develop rapidly in heat-damaged tissues: within 24 hours in hamster melanoma (Journée-de Korver et al. 1992) and within 2days in human choroidal melanoma (Journée-de Korver et al. 1995). Clearance of necrotic debris requires 3 4 months and is demonstrated on ultrasonography by a reduction in tumour height. After radiotherapy, cell damage and tumour regression develop gradually and may continue for several years. The three eyes with minimal or no regression, in which treatment failed even after two sessions, all contained amelanotic choroidal melanoma. Oosterhuis et al. found that the mean energy required in the treatment of amelanotic melanoma was 15% higher than that required for treatment of pigmented tumours, and the reduction of tumour height 3months after TTT was noted to be 33% greater in pigmented than in amelanotic tumours (Oosterhuis et al. 1998). The absence of pigmentation in amelanotic melanoma not only allows for deeper penetration (Svaasand et al. 1985) of infrared laserlight into the tumour tissue, but also implies decreased uptake of light converted into heat. We found that choroidal damage in the heat-treated area was less extensive after TTT than is commonly observed after plaque radiotherapy (Finger 1997). Angiography identified vessels remaining in the choriocapillary layer of the treatment zone in 15 eyes and noted persistence of large choroidal vessels in 10 eyes. Although information obtained by indocyanine green is limited to areas which are not masked by hyperpigmentation, our patterns of indocyanine green angiography after TTT (Figs 2 and 3) confirm that choroidal vascular damage is strictly confined to the treated area. These findings accord with the results of histological investigations (Song 1984; Reinhold & Endrich 1986; Nuijs-Beems et al. 1990), and may be explained by the fact that the choroid adjacent to the area of thermotherapy acts as a heat reducer due to its high blood perfusion rate. This finding may represent the most important benefit of TTT over plaque radiotherapy. Visual loss due to global microvascular damage associated with radiation effects on the retina and the optic nerve might be avoided by using TTT, especially in eyes containing choroidal melanoma near the fovea and the optic disc. Moreover, the visual outcome may be enhanced in such eyes using the described treatment. In our series, visual outcome primarily depended on the location of the tumour, regardless of whether the fovea was included in the treated area or not. When a tumour was located outside the fovea, we observed improvement in vision in six eyes, due to the resolution of subretinal fluid after TTT. Only two eyes with melanoma outside but near the fovea demonstrated reduced VA. This was attributed to developing macular oedema. The results obtained by our study on tumour regression and visual outcome are similar to those reported in the literature (Godfrey et al. 1999; Oosterhuis et al. 1998; Shields et al. 1998; Robertson et al. 1999; Schneider et al. 1999; Stoffelns 1999), although the rate of ocular side-effects was lower in the present series. Shields et al. evaluated the results of primary TTT for choroidal melanoma in 100 consecutive cases in the most extensive study to date, both with regard to the patient population and the duration of follow-up (Shields et al. 1998). The most frequent ocular side-effects in that series included vascular obstruction in 23% and retinal traction in 20% of patients, which occurred within a mean time interval of 4 5months after TTT. In contrast, we observed neither vascular obstruction nor the development of retinal traction over a follow-up period of at least 6months. One reason for the conflicting results may be the different number of treatment sessions used. Seventeen of 20 patients in our series received only one treatment session, whereas Shields et al. generally scheduled three treatment sessions to ensure more complete tumour control (Shields et al. 1998). Conclusion Preliminary observations suggest that transpupillary thermotherapy alone is an 30

7 effective and vision-conserving treatment, especially for small choroidal melanomas near the optic disc and the posterior pole of the eye. Our angiographic findings confirm that, despite a completely flattened tumour scar, choroidal damage is significantly less marked after TTT than after plaque radiotherapy. Indocyanine green angiography may therefore be useful in monitoring late ocular side-effects or tumour recurrences after TTT, particularly in hypopigmented tumours. Additionally, we recommend the use of ultrasonographic imaging of the eye and the retrobulbar region in all eyes successfully treated with TTT for the detection of transscleral tumour growth. This should be a mandatory practice, but it applies particularly to iuxtapapillary tumours, which are more likely to develop intrascleral spread than are tumours away from the optic disc. Despite the presence of an ophthalmoscopically flattened scar, surviving tumour cells in the choriocapillary layer or sclera may be capable of initiating transscleral regrowth. All patients with choroidal melanoma treated with TTT only should undergo close follow-up over years to allow for the collection of data on possible late ocular side-effects in the heattreated tissue, ultimate visual outcomes, the rate of local tumour recurrence and metastatic spread. References Coleman DJ, Lizzi FL, Burgess SEP et al. (1986): Ultrasonic hyperthermia and radiation in the management of intraocular malignant melanoma. Am J Ophthalmol 101: Finger PT (1991): Microwave plaque thermoradiotherapy for choroidal melanoma. Br J Ophthalmol 76: Finger PT (1997): Radiation therapy for choroidal melanoma. Surv Ophthalmol 42: Godfrey DG, Waldron RG & Capone A Jr. (1999): Transpupillary thermotherapy for small choroidal melanoma. Am J Ophthalmol 128: Journée-de Korver JG, Oosterhuis JA, Kakebeeke-Kemme HM & de Wolff-Rouendaal D (1992): Transpupillary thermotherapy (TTT) by infrared irradiation of choroidal melanoma. Doc Ophthalmol 82: Journée-de Korver JG, Osterhuis JA & Vrensen GFJM (1995): Light and electron microscopic findings on experimental melanomas after hyperthermia at 50æC. Melanoma Res 5: Journée-de Korver JG, Oosterhuis JA, de Wolff-Rouendaal D & Kemme H (1997): Histopathological findings in human choroidal melanomas after transpupillary thermotherapy. Br J Ophthalmol 81: Leeper D (1985): Molecular and cellular mechanisms of hyperthermia alone or combined with other modalities. In: Overgaard, J, ed. Hyperthermic Oncology. London. Taylor & Francis Ltd. 2: Lommatzsch PK & Kirsch JH (1988): Ru 106/ Rh 106 plaque-radiotherapy for malignant melanomas of the choroid. Doc Ophthalmol 68: Marmor JB, Hilerio FJ & Hahn GM (1979): Tumour eradication and cell survival after localized hyperthermia induced by ultrasound. Cancer Res 39: Nuijs-Beems EM, Oosterhuis JA, Verburg-van der Marel EH, de-wolff-rouendaal D, van Delft JL & van Best JA (1990): Tumour destruction by intermediate level hyperthermia. Curr Eye Res 9: Oosterhuis JA, Journée-de Korver HG, Kakebeeke-Kemme HM & Bleeker JC (1995): Transpupillary thermotherapy in choroidal melanomas. Arch Ophthamol 113: Oosterhuis JA, Journée-de Korver HG & Keunen JEE (1998): Transpupillary thermotherapy. Results in 50 patients with choroidal melanoma. Arch Ophthalmol 116: Reinhold HS & Endrich B (1986): Invited review: tumour microcirculation as a target for hyperthermia. Int J Hyperthermia 2: Riedel KG (1988): Hypertherme Therapieverfahren in Ergänzung zur Strahlenbehandlung maligner intraocularer Tumoren. Klin Monatsbl Augenheilkd 193: Robertson DM, Buettner H & Bennett SR (1999): Transpupillary thermotherapy as primary treatment for small choroidal melanomas. Arch Ophthalmol 117: Schneider H, Fischer K, Fietkau R & Guthoff RF (1999): Transpupillare Thermotherapie des malignen Aderhautmelanoms. Klin Monatsbl Augenheilkd 214 (2): Shields CL, Shields JA & Kiratli H et al. (1995): Risk factors for growth and metastasis of small choroidal melanocytic lesions. Ophthalmology 102: Shields CL, Shields JA, Cater J, Lois N, Edelstein C, Gündüz K & Mercado G (1998): Transpupillary thermotherapy for choroidal melanoma. Ophthalmology 105 (4): Song CW (1984): Effect of local hyperthermia on blood flow and microenvironment: a review. Cancer Res 44: Stoffelns BM (1999): Der infrarote Diodenlaser (810 nm) zur transpupillaren Thermotherapie des malignen Aderhautmelanoms. Lasermedizin 14: Svaasand O, Boerslid T & Oeveraasen M (1985): Thermal and optical properties of living tissue: application to laser-induced hyperthermia. Laser Surg Med 5: Received on June 11th, Accepted on November 25th, Correspondence: Bernhard M. Stoffelns, MD Department of Ophthalmology Johannes Gutenberg University Mainz Langenbeckstrasse 1 Dª55131 Mainz Germany Tel: π Fax: π stoffelns/augen.klinik.uni-mainz.de 31