Royal College of Ophthalmologists Guidelines 

Photodynamic therapy for subfoveal choroidal neovascularisation

The vast majority of cases of subfoveal CNV are caused by age-related macular degeneration (AMD) but other causes include pathological myopia, angioid streaks and punctate inner choroidopathy, especially in younger patients. Until very recently confluent argon laser photocoagulation has been the only treatment modality of proven clinical efficacy but after initial enthusiasm its role has become restricted to the treatment of eyes with extrafoveal CNV. In addition to PDT other therapeutic options for subfoveal CNV that have shown promise in pilot clinical studies include radiation with photons or protons, surgical translocation, indocyanine angiography guided laser to feeder vessels and transpupillary thermotherapy. 

PDT for AMD is a two stage process comprising a 10 minute intravenous infusion of 6mg/kg verteporfin followed by activation 5 minutes later by 689nm diode laser for 83 seconds at 50J/cm2. The photosensitive verteporfin is selectively taken up by rapidly proliferating endothelial cells within the target CNV reaching its peak concentration at 15 minutes. Cytotoxic reactive oxygen intermediates damage cellular proteins and cause microvascular thrombosis.² There is no direct thermal effect. 

Initial development of PDT occurred in the fields of oncology and dermatology. Studies of ocular experimental CNV, corneal neovascularisation and melanoma were followed by dose ranging and safety studies in human volunteers. ^3,4 Other photosensitisers are currently undergoing early clinical or pre-clinical trials. 

Randomised controlled clinical trials of PDT with verteporfin 

Two randomised multicentre double-masked clinical trials of PDT with verteporfin (VisudyneTM; Novartis Ophthalmics, Duluth, USA) are currently in progress, the Treatment of AMD with Photodynamic Therapy (TAP) and Verteporfin in Photodynamic Therapy (VIP) trials. In 1999 the one year interim results from the TAP study were published.¹ In addition unpublished details are in the public domain giving headline results from the two year TAP and one year VIP results. 

TAP study 

609 patients with AMD and subfoveal CNV with a classic component and refracted logMAR visual acuity (VA) between 6/12 and 6/60 (Snellen equivalent) underwent a 2:1 randomisation between treatment and control (sham treatment). Retreatment was applied to zones of persistent or new leakage at 3 monthly visits. Follow-up rates were 94% at 12 and 87% at 24 months. The frequency of stable (<15 logMAR letters lost) or improved vision in each group was: 12 months - 61% treated, 46% placebo (p<0.001); 24 months - 53% treated, 37% placebo (p<0.001). At 12 months a visual improvement (15 letters) occurred in 16% treated and 7% placebo patients. 

For predominantly classic lesions the frequency of stable/improved vision was: 12 months - 67% treated, 39% placebo (p<0.001); 24 months - 59% treated, 31% placebo (p<0,001). For lesions with no occult component frequencies at 12 months were: 73% treated, 23% placebo (p<0.01). Treatment benefit was also shown for secondary endpoints. Average number of treatments in the treated group was 5.6 (3.4 in the first 12 months). 

There were a similar number of deaths and serious adverse events in each group. <2% of patients were withdrawn due to adverse events. Commoner in treated patients at 12 months were: transient visual disturbance - 18% treated, 12% placebo; injection site events - 13% treated, 0% placebo; transient photosensitivity reactions - 3% treated, 0% placebo; low back pain - 2% treated, 0% placebo. 

VIP study 

339 patients with occult CNV with evidence of progression but no classic have been enrolled. At the time of writing only limited results are available: interim analysis at 12 months has not detected a significant difference between treated and placebo groups. 

120 patients have been enrolled into the study of CNV in pathological myopia. The frequency of maintenance or improvement of vision at 12 months was 86% in treated patients compared to 67% in placebo patients (p=0.01). 

Implications of currently available TAP and VIP results 

Clinical indications 

Interpretation of the statistical of the positive results from the TAP study is made easier when the NNT analysis of clinical effectiveness is considered. A balanced position has been taken by the TAP study group who have recommended treatment for patients with 50% or more classic (predominantly classic) subfoveal CNV secondary to AMD. In this group the NNT was 3.6 at 12 months and 24 months and this compares well with an NNT calculation for the prevention of severe visual loss after panretinal scatter photocoagulation for proliferative diabetic retinopathy with high risk characteristics of 14.^3,5 The Cochrane Eyes and Vision Group have interpreted the results as indicating that the benefit is restricted to patients with no occult;⁶ the NNT improved to 2.2 in the 143 patients in this subgroup. 
 
 

Indications for Photodynamic Therapy with Verteporfin

Indicated subfoveal/juxtafoveal CNV secondary to AMD, predominantly classic, VA6/60 or better, lesions <5400um  CNV secondary to pathological myopia

Probably indicated lesions greater than 5400 um subject to limitations of laser spot size

Possibly indicated juxtrapapillary lesions with subfoveal extension; CNV from other causes

Not indicated <50% classic; pure occult; RPE tears

To date the TAP and VIP studies have not demonstrated a clinically significant benefit for lesions with minimally (<50%) classic or no classic and so the treatment of these lesions is not indicated. Conversely the 12 month data on pathological myopes has shown a statistically significant benefit and probably justifies inclusion of this cause of CNV at least at this stage. Whether the results in myopes can be generalised into other causes of CNV such as punctate inner choroidopathy and angioid streaks remains unclear. Further work on case selection is required. Indications based on the currently available date are summarised in the table. 

No major safety issues have been identified in verteporfin PDT. A transient visual loss developing by 48 hours and persisting up to 4 weeks is seen in some patients. A potential longer term effect on the RPE needs to be borne in mind: some atrophy was seen after multiple treatments administered at 2-4 week intervals over a 3 month period but not during the TAP study. 

Service development in the UK

Estimating the numbers of patients that might meet criteria for verteporfin PDT is difficult as no direct data exist. Information from BD registrations ⁷ and observational data on classification of lesion components⁸ is available but limited. If 5% of eyes with new exudative AMD remain extrafoveal and treatable with confluent laser then it seems reasonable that a further 30% might be eligible for PDT giving a rough estimate of 5,000 patients per annum in England and Wales. 

For verteporfin PDT to be established in the UK under the NHS the capacity of the hospital eye service will need to expand. Training of ophthalmologists and photographers in stereoscopic angiography will be required as will an increase in numbers of medical, nursing and other ancillary staff. Several centres in the UK are far advanced in business planning/service development to meet the anticipated need. 

Health authorities and other purchasers have adopted varying positions on PDT while they consider proposals for service contracts. The treatment is relatively expensive if the TAP protocol is adhered to but the cost is likely to fall. Guidance on the implementation of PDT has been issued by two UK bodies. He Safety and Efficacy Register of New Interventional Procedures (SERNIP) group have categorised the treatment of classic CNV as 'B' - efficacy established, further evaluation required to establish safety. A Royal College of Ophthalmologists working party have recommended that treatment should commence in a limited number of centres with retinal expertise, access to stereoscopic angiography and experience in collecting data in clinical trials. The National Institute for Clinical Effectiveness (NICE) will not consider the treatment in the near future. A national database of outcomes and adverse events should help provide information to purchasers and regulators and act as a surveillance system for SERNIP. 

Summary

Current evidence supports the use of PDT with verteporfin for a proportion of patients with subfoveal CNV. To minimise the danger of indiscriminate use at this early stage in the introduction of PDT, treatment should be restricted to those patients in whom a treatment benefit has been definitely demonstrated. Further research is needed to establish the optimum treatment regime, its cost-effectiveness and its impact on health services. 

Simon Harding St Paul's Eye Unit, Royal Liverpool University Hospital Email: simonharding1@compuserve.com

Mr Harding has received departmental research funding from QLT/CIBA Vision in his capacity as principal investigator of a clinical centre within the TAP and VIP studies. He has no proprietary, commercial or financial interest in any photodynamic therapy agent or device. 

References:

• TAP study group. Photodynamic therapy of subfoveal choroidal neovascularisation in age-related macular degeneration with verteporfin. One-year results of 2 randomised clinical trials - TAP report 1. Arch Ophthalmol 1999; 117:1329-1345 
• Zhou C. Mechanisms of tumour necrosis induced by photodynamic therapy. J Photochem Photobiol B 1989; 3: 299-318 
• Miller JW, Schmidt-Erfurth U, Sickenberg M, et al. Photodynamic therapy with verteporfin for choroidal neovascularisation caused by age-related macular degeneration: results of a single treatment in a phase 1 and 2 study. Arch Ophthalmol 1999;117:1161-1173 
• Schmidt-Erfurth U, Miller JW, Sickenberg M, et al. Photodynamic therapy with verteporfin for choroidal neovascularization caused by age-related macular degeneration: results of a single treatment in a phase 1 and 2 study. Arch Ophthalmol 1999; 1999; 117:1177-1187 
• Hart PA, Harding SP. Is it time for a national screening programme for sight-threatening diabetic retinopathy? Editorial Eye 1999; 13:129-130 
• Wormald R, Evans J, Smeeth L. Photodynamic therapy for neovascular age-related macular degeneration (Cochrane Review). Cochrane Eyes and Vision Group. In: The Cochrane Library, Issue 3, 2000. Oxford: Update Software. http://www.cochrane.co.uk 
• Evans J, Rooney C, Ashwood F, Dattani J, Wormald R. Blindness and partial sight in England and Wales: April 1990 - March 1991. Health Trends 1996; 28:5-12 
• Moisseiev J, Alhalel A, Masuri R, Treister G. The impact of the Macular Photocoagulation Study results on the treatment of exudative age-related macular degeneration. Arch Ophthalmol 1995;113:185-189.