Palladium-103 plaque radiation therapy for macular degeneration: results of a 7 year study

Abstract

Aim: To report 7 year results of ophthalmic plaque radiotherapy for exudative macular degeneration.

Methods: In a phase I clinical trial, 30 patients (31 eyes) were treated with ophthalmic plaque irradiation for subfoveal exudative macular degeneration. Radiation was delivered to a mean 2 mm from the inner sclera (range 1.2–2.4) prescription point calculated along the central axis of the plaque. The mean prescription dose was 17.62 Gy (range 12.5–24) delivered over 34 hours (range 18–65). Early Treatment Diabetic Retinopathy Study (ETDRS) type standardised visual acuity determinations, ophthalmic examinations, and angiography were performed before and after treatment. Clinical evaluations were performed in a non-randomised and unmasked fashion.

Results: At 33.3 months (range 3–4), 17 of 31 (55%) eyes had lost 3 or more lines of vision on the ETDRS chart, five (16%) had improved 3 or more lines, and the remaining nine (29%) were within 2 lines of their pretreatment visual acuity measurement. Overall, 45% of patients were within or improved more than 2 lines of their initial visual acuity. Five eyes developed macular scars, eight developed subsequent neovascularisation or haemorrhage, and three progressed through therapy. Two patients were lost to follow up. The most common finding of patients followed for 6 or more months (n = 18 of 29 (62%)) was regression or stabilisation of the exudative process. No radiation retinopathy, optic neuropathy, or cataracts could be attributed to irradiation.

Conclusion: Ophthalmic plaque radiation can be used to treat exudative macular degeneration. At the dose and dose rates employed, most patients experienced decreased exudation or stabilisation of their maculas. No sight limiting radiation complications were noted during 7 year follow up. Owing to the variable natural course of this disease, a prospective randomised clinical trial should be performed to evaluate the efficacy of plaque radiation therapy for exudative macular degeneration.

Macular degeneration is a leading cause of blindness in the western world. The “dry” form of macular degeneration is most common and characterised by slowly progressive atrophy of macula and mild to moderate loss of central vision. In contrast, 10% of patients develop the “wet” form called neovascular or exudative macular degeneration.^1–6

Exudative macular degeneration is associated with subretinal neovascularisation (SRN), exudation of lipids, serum, blood, and secondary inflammation. This process typically results in scarification of the macula and severe irreversible loss of central vision. Though most commonly age related, exudative macular degeneration can also be associated with ocular histoplasmosis and high myopia.^1–6

Photodynamic therapy (PDT) and laser photocoagulation have been proved (by prospective randomised clinical trials) to be effective treatments for eligible patients with exudative macular degeneration.^7–14 Unfortunately, most eyes with neovascular ARMD have either occult membranes or haemorrhagic leakage, thereby making them ineligible for laser based treatments. Like photocoagulation, the success of each laser activated PDT session is dependent upon visualisation of the subretinal neovascularisation.^7–15 In addition, SRN lesions closed by PDT commonly reopen within 3 months and must be re-treated several times.^12–15

In a meta-analysis of phase I clinical trials, Chakravarthy suggested that low dose external beam radiation therapy (EBRT) exhibited an inhibitory effect on exudative macular degeneration, but higher doses were more effective in the prevention of severe loss of vision (>6 lines on the vision chart).¹⁶ Despite these findings, several multicentre prospective randomised clinical trials (typically utilising 10–24 Gy EBRT delivered in 2–4 Gy daily fractions) have demonstrated conflicting results.^17–22 In a relatively small prospective randomised trial, Bergink et al found that relatively high dose EBRT was effective in limiting vision loss to less than 6 lines on the visual acuity chart.¹⁷

In our opinion, the most striking angiographic evidence of closure of neovascular membranes has been presented by researchers using 8 or 12 Gy of charged particle irradiation (in one 5 minute fraction) and by Jaakola et al using a hand held strontium-90 (⁹⁰Sr) brachytherapy applicator.^23,24 It is important to note that both groups employed high dose rates, the effect of which is equivalent to a larger dose.^23,24

Compared to EBRT, brachytherapy techniques can deliver a relatively high dose to the involved macula with less irradiation of most normal ocular structures outside the targeted zone.^23,25–31 In part because of dose gradient effects, ophthalmic plaque radiotherapy also allows for more focused irradiation to the affected choroid. The use of brachytherapy also avoids an anterior segment entry dose, a mobile target volume, and irradiation of the fellow eye, sinuses, and/or brain.²⁶ Published reports on brachytherapy for exudative macular degeneration include our use of palladium-103 (¹⁰³Pd), Jaakola and Freire’s strontium-90 (⁹⁰Sr) applicators, and Berta’s ruthenium-106 (¹⁰⁶Ru).^{23,26,29–31} Immonen et al suggested that ⁹⁰Sr treated eyes lost less vision than controls at the 2001 meeting of the Association for Research in Vision and Ophthalmology.³² In all the aforementioned studies, no complications that might preclude this approach to treatment of exudative macular irradiation have been noted.

In this study, ¹⁰³Pd ophthalmic plaque radiation therapy was used to treat classic, occult, and recurrent exudative macular degeneration. In all, 31 treated eyes have been followed for up to 7 years. Here, we present our methods and clinical findings.

METHODS

Radiation typically induces acute vasculitis and oedema followed by slowly progressive vascular closure (which may take years to develop).^33–37 These effects of radiation are both dose and dose rate dependent.³⁸

Irradiation of a macula containing classic or occult subretinal neovascularisation can directly affect angiogenesis by destroying neovascular endothelial cells and cytokine producing macrophages, or alter the regulatory genes which produce endothelial growth regulating cytokines.³⁹

Widely employed to prevent scar formation, radiation has been used to inhibit the cutaneous keloid and more recently proved to prevent coronary artery stenosis.⁴⁰ Similarly, Hart et al have suggested that radiotherapy inhibited disciform scar formation associated with end stage exudative macular degeneration.⁴¹

Patient selection

To date, 30 patients (31 eyes) with exudative macular degeneration were referred because their disease was considered to be untreatable or they had refused alternative therapy. All lesions were required to have some evidence of subretinal neovascularisation (SRN) demonstrable on fundus photography and fluorescein angiography involving the foveal avascular zone. Such evidence included SRN, blood, exudate, as well as retinal pigment epithelial and neurosensory retinal detachments.

All patients were found to have classic, occult, or recurrent subfoveal exudation secondary to age related macular degeneration (AMD) and had noted recent visual loss or metamorphopsia. All patients had pre treatment visual acuities better than or equal to 20/400, and a minimum post-treatment follow up of 3 months (Table 1).

Table 1.

Patient data and methods

No	Macular degeneration type	Eye	Point Rx	Optic nerve distance	103Pd dose (cGy)	Treatment duration (hours)
1	Recurrent classic	RE	1.9	1.5	1500	45
2	Occult	RE	1.2	2	1500	24
3	Occult	RE	1.4	1.8	1500	39
4	Recurrent classic	RE	1.7	2	1500	24
5	Occult	LE	1.9	2	1500	24
6	Classic	LE	1.9	2.5	1250	24
7	Occult	RE	1.4	3.5	1500	26
8	Occult	RE	1.4	1.8	1250	23
9	Occult	RE	2	1.8	1800	48
10	Recurrent classic	LE	1.4	1	1400	24
11	Occult	LE	1.9	1.2	1800	46
12	Occult, classic	LE	1.4	0	1500/1500*	24/48*
13	Occult	LE	2.4	1	1500	25
14	Occult	RE	1.9	2	2362	18
15	Occult	RE	2	2	1800	33
16	Recurrent occult	LE	2	2.5	1800	26
17	Occult	LE	1.4	1.8	1800	43
18	Classic	LE	2	1.5	1500	65
19	Occult	LE	2	1	1800	48
20	Occult	RE	2	2.5	2400	34
21	Recurrent classic	LE	2	2	1800	42
22	Occult	LE	1.4	1.8	1800	25
23	Occult	LE	1.9	1.2	1500	24
24	Recurrent classic	RE	2	1	2394	25
25	Classic	RE	1.9	1.2	1500	24
26	Classic	LE	2	1.5	2400	24
27	Recurrent classic	RE	1.7	1.8	1800	46
28	Occult	RE	1.7	2	1500	37
29	Classic	LE	2	1.8	1800	24
30	Occult	LE	2	2.1	2166	24
31	Classic	LE	2	1.5	1500	40
	Mean		2	2	1762	34

Clinical evaluations

Patients underwent complete ophthalmic eye examinations at each visit. All visual acuity determinations involved protocol refractions by Collaborative Ocular Melanoma Study (COMS) certified personnel, in approved COMS rooms, utilising standard Early Treatment Diabetic Retinopathy (ETDRS) charts. After refraction, pupillary, ocular motor, and slit lamp examinations were performed. Goldmann tonometry was used to measure intraocular pressure. Direct, indirect, and contact lens ophthalmoscopy techniques were used as required. The presence of subretinal neovascular membranes and components was determined by ophthalmoscopy, fundus photography, angiography, and ultrasonography.

Consent

Institutional review board approvals for a phase I clinical trial for evidence of effects of treatment (for example, toxicity, angiographic appearance, visual acuity) were obtained from the participating institutions. Informed consent involved a detailed discussion of current knowledge of radiotherapy for macular degeneration, our comparative dosimetry studies, and the available alternative therapies (for example, photocoagulation, porphyrin laser, drug trials, and submacular surgery as applicable). Based on this information, patients were requested to choose either external beam or plaque radiotherapy. No guarantees were made with respect to visual outcome and incidence of complications.

Radiation dosimetry

We chose to use ¹⁰³Pd versus ¹²⁵I seeds because of our experience with this radionuclide for the treatment of intraocular tumours and comparative dosimetry.^42–48 Similarly, because of the lower photon energy of ¹⁰³Pd (21 versus 28 keV), the use of ¹⁰³Pd offered a slightly higher dose to the targeted volume with less irradiation of most normal ocular structures.⁴³

The prescription point was calculated along the central axis of the plaque. Distances were calculated from the inner scleral surface (Table 1). As has been done in the COMS study, the sclera beneath the plaque was assumed to be 1 mm in thickness.

Plaques were composed of ¹⁰³Pd seeds (Model 200, Theragenics Co, Buford, GA, USA) affixed into 10 or 12 mm standard gold eye plaques (Trachsel Dental Studio Inc, Rochester, MI, USA) with a layer of acrylic fixative. Dosimetric calculations were performed such that the seeds were calculated as point sources without correction for anisotropy (the specific dose rate constant corresponded to that perpendicular to the seeds long axis). The attenuation of the acrylic material was water equivalent. No attenuation was attributed to the 0.5 mm thick gold sidewalls of the plaque. The radial dose function for ¹⁰³Pd in water was obtained from published data.^49–51

The inferior oblique muscle partially inserts into the sclera beneath the macular retina and presents an obstruction to allowing the plaque to lie flat on the sclera. It would be reasonable to assume that one edge of the plaque was slightly tilted during radiotherapy. This effect of plaque tilt on dose distribution is unknown and typically discounted during plaque irradiation of macular uveal melanomas. It was similarly discounted during this series.

Plaque brachytherapy

Our surgical technique has been described.³⁰ In sum, intraoperative indirect ophthalmoscopy was used to confirm the condition of the treatment site. Then, a transconjunctival approach was used to expose the lateral rectus muscle. With the muscle disinserted, the eye could be rotated into adduction and the inferior oblique muscle visualised. The radioactive plaque was placed beneath the macula such that the posterior edge of the plaque met palpable resistance at the optic nerve.

Typically four 5-0 Vicryl episcleral sutures were placed (through the suture eyelets) to secure the plaque beneath the macula. Then, indirect ophthalmoscopy with scleral indentation was used to confirm the position of the plaque in relation to the macular target zone. The lateral rectus muscle was reattached, the conjunctiva closed, and a lead shield taped over the eye. Plaque removal was performed under local anaesthesia and a similar technique was used. Typically, the episcleral fixation sutures could be transected without disturbing the rectus muscle.

Brachytherapy was delivered in a single continuous session. Patient 12 was treated twice because of recurrent disease. With the plaque in place, radiation travelled through the sclera, choroid (area of neovascularisation), and retina. In this series, ophthalmic plaque brachytherapy was delivered to a mean 2 mm from the inner sclera (range 1.2–2.4) prescription point calculated along the central axis of the plaque. The mean prescription dose of 17.62 Gy (range 12.5–24) was delivered over 34 hours (range 18–65) (Table 1). We currently prescribe to a prescription point 2 mm from the inner sclera along the central axis of the plaque, and to a minimum dose of 24 Gy.

Follow up

Visual function was evaluated by means of standard ETDRS refractions by an unmasked examiner. Lesion growth and recurrence were evaluated by ophthalmoscopy and photography with angiography. Clinical evaluations were used to note any evidence of radiation side effects: eyelid erythema, lash loss, conjunctival injection, dry eye, corneal epitheliopathy, rubeosis iridis, cataract formation, vitreous haemorrhage, radiation retinopathy, and radiation optic neuropathy. After the acute postoperative period, ophthalmic examinations with angiography were performed at 3–6 month intervals (Table 2). Follow up examinations included standardised refraction, pupillary function, ocular motor function, slit lamp examination, Goldmann tonometry, ophthalmoscopy, fundus photography, and ophthalmic angiography.

Table 2.

Results after ¹⁰³Pd plaque for subfoveal neovascularisation

Eye number	Vision (20/x)	Follow up (months)																		Initial vision	Last ETDRS vision	Lines of vision	Last reported vision	Ophthalmic status
		3	6	9	12	15	18	21	24	30	36	42	48	54	60	66	72	78	84
1	400	400	400	400	400	400	320	320	250	250	250	400	400	NA	NA	NA	CF		400	CF	>−6	CF		Regressed to dry AMD
2	63	40	32	32	32	32	NA	32	32	32	NA	32	40	40	32	32				63	32	3	32	Stabilised PED
3	40	63	63	40	63	50	40	40	40	63	NA	50	50	NA	200	NA	100	100	100	40	200	>−6	100	Eccentric recurrence at 82 months
4	250	200	200	400	320	320	320	100	100	400	200	80	80	63	63					250	63	6	63	Regressed to dry AMD
5	63	63	63	50	50	63	NA	80	63	40	40	50	NA	NA	40					63	40	2	40	Regressed to dry AMD+
6	400	200	125	100	100	100	160	NA	100	125	200	250	250	640	NA	NA	CF	CF	CF	400	CF	>−6	CF	Improved but not dry
7	32	20	16	20	NA	25	25	32	32	32	32	32	63	400	NA	CF			32	400	>−6	CF		Regressed to dry AMD
8	40	25	160	250	250	NA	200	NA	200	400	400	400	CF	NA	CF	CF	CF	CF	40	CF	>−6	CF		Regressed to disciform scar
9	40	40	NA	63	80	500	NA	NA	NA	CF	NA	400	400							40	400	>−6	400	Regressed to disciform scar
10	80	100	NA	NA	NA	NA	NA	400	160	CF	160	320	400	200	200	200	320			80	320	−6	320	Regressed to dry AMD
11	100	100	NA	NA	100	NA	NA	NA	NA	NA	NA	HM	HM						100	HM	>−6	HM	Vitreous	haemorrhage at 24 months
12	160	80	63	80*	250*	200*	200*	250*	250*	250*	500*	400*	400*	400*	NA	LP*			160	500	>−6	LP		SRN, re-irradiated at 9 months*
13	200	100	100	100	NA	63	NA	100	500	200	HM	CF	NA	CF	CF	CF			125	HM	>−6	CF	Vitreous haemorrhage at 36 months
14	32	25	20	25	25	20	20	20	20	20	25									32	25	1	25	Regressed to dry AMD
15	50	100	63	100	125	160	200	200	200	250	320	400	CF							50	320	>−6	CF	TTT given at 36 months
16	32	25	20	20	20	40	20	25	25	40	50	200								32	40	−1	200	PDT given at 30 months
17	50	32	32	20	25	50	25	NA	25	32										50	32	2	32	No progression
18	63	80	125	NA	NA	NA	250	250	250	500										63	500	>−6	500	Regressed to disciform scar
19	40	40	40	32	40	NA	50	50	160	320	CF									40	160	−6	CF	PDT given at 24 months
20	125	100	125	125	200	320	400	200	200	CF										125	200	−2	CF	PDT given at 24 months
21	100	100	100	100	125	125	NA	100	125											100	125	−1	125	No progression, deceased
22	50	63	50	63	100	100	125	250	320	320	NA	CF								50	250	>−6	CF	Regressed to disciform scar, deceased
23	32	63	63	100	100	160	200	125												32	125	−6	125	Progressed
24	100	100	100	125	125	125	NA	125												100	125	−1	125	Regressed to dry AMD
25	160	250	200	200	NA	NA	200	200												160	200	−1	200	Improved but not dry, deceased
26	32	63	80	160																32	160	>−6	160	Progressed
27	50	100	160	400	NA	NA	NA	NA	HM	NA	HM	HM	HM	CF						50	400	>−6	CF	Regressed to disciform scar
28	63	50	32																	63	32	3	32	Regressed to dry AMD
29	200	200	320																	200	320	−2	320	Progressed, deceased
30	63	100																		63	100	−2	100	Lost to follow up
31	125	25																		125	25	>+6	25	Lost to follow up
Mean follow up (months)																33.3

CF  =  counting fingers, LP  =  light perception, HM  =  hand movements, RR  =  radiation retinopathy; AMD  =  age related macular degeneration, PDT  =  photodynamic therapy, TTT  =  transpupillary thermotherapy.

RESULTS

According to Macular Photocoagulation Study criteria, the macular lesions were initially diagnosed as classic (n = 6), occult (n = 17), or recurrent (n = 8) subfoveal choroidal neovascularisation (Table 1). Classic subretinal neovascular (SRN) lesions were characterised by well demarcated areas of hyperfluorescence discerned on the early phases of a fluorescein angiogram that progressed into the late phases. Occult lesions were defined as those demonstrating diffuse leakage of unknown origin with retinal pigment epithelial detachment, haemorrhage, and/or exudates. Recurrent lesions were defined as those that developed evidence of classic or occult leakage with any history of previous macular laser treatment.

According to Treatment of Age related Macular Degeneration with Photodynamic Therapy (TAP) Study Group criteria the macular lesions were initially diagnosed as having completely classic exudative macular degeneration (n = 6), mostly (⩾50%) classic (n = 8), and less than 50% classic (n = 3). Fourteen maculas were purely occult. This includes both the recurrent and untreated lesions.¹²

From April 1994 to September 1999, we treated 31 eyes in 30 patients. These 31 irradiated lesions were followed in our centre for a mean 33.3 months (range 3–84). Additional follow up was obtained by contacting the patients and their physicians (Table 2). Ophthalmic status was determined by chart review except for the two patients who were lost to follow up. The most common finding (n = 18 of 26 (69%)) was regression or stabilisation of the exudative process. These maculas were either described as “regressed to dry macular degeneration,” “stabilised PED,” “improved by not dry,” “regressed to disciform scar,” “regressed to subretinal fibrosis,” or “no progression.” Five of these eyes developed a disciform macular scar.

Other events occurred after treatment

Six maculas (21%) initially were stable to improved but required subsequent treatment 9–36 months after radiotherapy (Table 2). Three lesions (10%) clearly progressed through treatment, and two eyes (7%) experienced vitreous haemorrhages 24 and 36 months after radiation (Table 2).

Analysis of our visual acuity results was complicated by events that occurred after treatment. For example, patient 6 initially improved after treatment, then suffered a recurrence of his disease, was re-irradiated, then improved but not to stabilisation of his vision. By reviewing the vision data in Table 2, the reader will note that any individual patient’s vision could vary through the duration of follow up. Any one patient could improve, worsen, stabilise, or experience a combination of these results (Table 2).

Our vision results represent the last best corrected visual acuity performed at our centre compared to the patient’s pretreatment visual acuity. In this analysis and at their last protocol visit, 17 (55%) eyes had lost 3 or more lines of vision on the ETDRS chart, five (16%) had improved 3 or more lines, and the remaining nine (29%) were within 2 lines of their pretreatment visual acuity measurement. Overall, 14 patients (45%) were within or improved more than 2 lines of their initial visual acuity.

In order to evaluate vision at certain time intervals, we provide the following additional analysis: 88% of patients examined at their 6 month visit (n = 26) were within or improved 3 lines from their baseline vision, 85% at 12 months (n = 20), 65% at 18 months (n = 17), 74% at 24 months (n = 19), 73% at 36 months (n = 11), and 75% at 48 months (n = 8). Severe loss of vision is often defined as a decrease of 6 or more lines of vision. In this study 4% of patients had lost 6 lines or more from their baseline vision at their 6 month visit (n = 26), 5% at 12 months (n = 20), 24% at 18 months (n = 17), 21% at 24 months (n = 19), 18% at 36 months (n = 11), and 25% at 48 months (n = 8). This analysis is limited to those patients (n) who came to the study centre at these reported time intervals.

Other findings have included one (previously strabismic) patient who noted postoperative diplopia that resolved after 3 months follow up. Another patient (No 5) developed a macular hole 6 months after treatment, which was successfully repaired. Evidence of resolution of haemorrhage, exudates, and subretinal fluid suggestive of a therapeutic effect has been noted after plaque radiotherapy. To date, no radiation retinopathy, optic neuropathy, or cataracts have been noted.

DISCUSSION

In review of our results it is important to note that all of our patients were symptomatic of vision loss or metamorphopsia before treatment and that clinical evaluations suggested that only three of 31 (10%) progressed despite treatment. Two additional patients were lost to follow up after 3 months. It is striking that 18 of 26 (69%) treated maculas were described as stabilising or evolving from exudative to dry macular degeneration or scarification (Table 2). This finding not only suggests that radiation affected the exudative process, but warns us that resolution of exudation (in itself) will not erase the damage created by this process or that due to subsequent scarring. In addition, seven eyes (23%) suffered recurrent subretinal neovascularisation or haemorrhages over the years after treatment. Post-treatment vision loss occurred for a variety of factors.

We found that 45% of eyes were within or improved more than 2 lines (compared to their pretreatment visual acuities). Though remarkable, this phase I study does not prove efficacy. The natural course of exudative macular degeneration is known to be variable. For example, the classic and recurrent forms typically result in a rapid loss of vision, while the occult types usually progress slowly. This is why prospective randomised clinical trials are typically employed for proof of efficacy. In this study, our small numbers, lack of a control group, and the possibility of susceptibility bias prevent such a proof.

This study has shown that up to 24 Gy plaque brachytherapy resulted in no complications that might preclude its use for age related macular degeneration associated with subretinal neovascularisation. Chakravarthy and Bergink’s studies have suggested that a higher dose may be required to control exudative macular degeneration. Clearly, brachytherapy offers a method to increase the dose to the affected macula with relative sparing of normal ocular, sinus, and intracranial structures.^{26,30,52–58} Implant radiation therapy is an investigational treatment that should be subjected to a prospective randomised efficacy trial.