In 2000, verteporfin for injection (Visudyne®; Novartis), an intravenous photodynamic therapy agent, was approved by the U.S. Food and Drug Administration for the treatment of age related macular degeneration for the treatment of age-related macular degeneration in patients with predominantly classic subfoveal choroidal neovascularization. Subsequently, in 2001, the indication was expanded to include presumed ocular histoplasmosis and pathologic myopia.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

This policy was initially developed as a result of the FDA approval of verteporfin for injection.  A 2000 Blue Cross Blue Shield TEC Assessment offered the following conclusions:

 

 

The TEC Assessment did not specifically address the issue of frequency of treatment or treatment extending beyond 1 year.

 

Since the completion of the TEC Assessment, the 2-year results of the pivotal randomized trial Treatment of Age-Related Macular Degeneration with Photodynamic Therapy (TAP) were published (Bressler, 2001).  Beneficial outcomes regarding visual acuity and contrast sensitivity noted after 12 months were sustained through 24 months. At the end of 2 years, 53% of the treatment group, as compared to 38% of the placebo group, lost fewer than 15 letters. The average number of applications of verteporfin treatment in the second year (2.2) was lower than that required during the first year.   A subgroup analysis was reported, comparing results between those patients with predominantly classic CNV (>50% of lesional area) compared to minimally classic CNV (<50%). For patients with minimally classic disease, no statistically significant differences in visual acuity were noted.

 

The verteporfin in photodynamic therapy (VIP) trial is another randomized study that primarily focused on efficacy of photodynamic therapy in patients with occult but no classic lesions who were presumed to have progressive disease due to visual or anatomic deterioration within the previous 3 months. Of the 339 patients enrolled in the trial, 76% had occult disease; the remainder had early classic CNV with good visual acuity.  Similar to other randomized trials, the primary outcome was the proportion of eyes with fewer than 15 letters of visual acuity loss. While there was no significant difference between the treatment and placebo groups at 12 months, by 24 months, a significantly lower percentage of those with occult CNV had lost vision (55% vs. 68%, p=0.032). These results contrast with those of the TAP trial, although the patient populations are slightly different. The TAP trial required all patients to have some percentage of classic CNV, while the VIP trial recruited patients with occult disease without evidence of classic CNV. In addition, the VIP trial required patients with occult disease to have experienced recent deterioration in vision. Results for the subgroup of patients with classic CNV but good visual acuity were not reported separately.

 

Photodynamic therapy has also been investigated in patients with CNV related to pathologic myopia and presumed ocular histoplasmosis. A second arm of the VIP trial focused on 120 patients with pathologic myopia and CNV, either classic, occult, or mixed (although 90% of patients had classic CNV) who were randomized to receive photodynamic therapy or placebo. At month 12, photodynamic therapy stabilized or improved vision (as defined by a loss of fewer than 8 letters on a standard eye chart) in 72% of patients versus 44% on placebo.  The authors concluded that verteporfin therapy increases the chance of stabilizing or improving vision compared to placebo treatment for at least 1 year. Results were not reported separately for those with predominantly classic CNV versus occult CNV.

 

There are minimal published data regarding the use of photodynamic therapy in patients with CNV related to ocular histoplasmosis. The FDA approval was based on an open-label safety study involving 26 patients with ocular histoplasmosis. Visual acuity improved by an average of more than 1 line on a standard eye chart at 12 months (6.7 letters on a standard eye chart) with 28% of patients experiencing a visual acuity improvement of 3 lines (15 letters) or more. Visual acuity decreased by less than 3 lines of vision in 88% of patients during the same time period.

 

A 2004 review of the peer-reviewed literature based on MEDLINE identified 3 additional reports from the TAP trial.  These reports demonstrated positive outcomes with the use of photodynamic therapy for subfoveal choroidal neovascularization and further supported the findings of the earlier TAP trial reports (Bressler, 2002) (Rubin, 2002) (Blumenkranz, 2002). A Cochrane review was published during this time period that concluded that photodynamic therapy is effective in preventing visual loss in classic and occult CNV due to AMD (Wormald, 2004).  

Ergun and colleagues reported on a case series of photodynamic therapy for CNV secondary to central serous chorioretinopathy on 26 eyes in 24 patients (Ergun, 2004).  Results demonstrated a mean increase in 2.2 lines of visual improvement at 2 years and no adverse outcomes. While promising, larger and comparative studies are warranted for photodynamic therapy for this indication. Therefore, photodynamic therapy for CNV is no longer considered investigational and photodynamic therapy for CNV secondary to central serous chorioretinopathy is added to the policy statement as investigational.

 

A 2005 literature review identified 2 articles that support the use of photodynamic therapy for CNV due to age-related macular degeneration. The Visudyne in Minimally Classic Choroidal Neovascularization (VIM) Study Group randomized 117 patients to verteporfin or placebo infusion with standard or reduced light fluence rates (Azab, 2004). The authors concluded after 24 months of follow-up that verteporfin therapy was safe and reduced the risks of disease progression to predominantly classic CNV and visual acuity loss of at least 15 letters. In a meta-analysis of the safety of photodynamic therapy, Azab and colleagues analyzed data from the 24-month TAP A and B and VIP trials, totaling 948 patients with AMD (Azab, 2005).  The authors concluded that the safety profile of verteporfin therapy was not statistically different from placebo. An October 2005 TEC Special Report on the treatment of AMD also supports the conclusions given here and notes that the pathway leading to CNV is complex.  

 

The October 2005 TEC Special Report notes that therapies are available for AMD to potentially modify (angiostatic agents) and palliate (PDT) the disease. Combining these modalities concurrently or sequentially has a biological basis. The 2005 TEC Special Report found that a host of trials are in progress combining an angiostatic agent with PDT. The angiostatic agents being studied in trials include pegaptanib, ranibizumab, anecortave acetate, squalamine, vitalinib, and triamcinolone. For example, in the pegaptanib trial, PDT was administered at physician discretion but an analysis was not provided that examined possible synergistic effects (Gradoudas, 2004).

 

Searches of the MEDLINE database for the periods of October 2005 through February 2007 and March 2007 through April 2008 found no evidence to support a change in the policy statement. A Cochrane review evaluated results from 3 randomized controlled trials (total of 1,022 patients), which included the TAP and VIP trials described here (Wormald, 2007).  Meta-analysis showed a 24-month risk ratio of losing 6 or more lines of visual acuity of 0.62 compared to the control group. The authors concluded that photodynamic therapy is probably effective in treating CNV due to AMD, though there is doubt about the size of the effect. Kaiser reported results of a 3-year open-label extension of the TAP study (Kaiser, 2006).  Of 402 verteporfin-treated patients who completed the 24-month randomized study, 320 (80%) enrolled in the extension protocol. Patients who chose not to participate in the TAP extension were more likely to be older, have a poorer level of visual acuity, and have evidence of fluorescein leakage, or evidence of progression, at the 24-month examination. Of the 320 enrolled, 193 (60%) completed the 60-month examination and 122 (38%) discontinued prematurely, 3 (1%) were noncompliant. Yearly treatment rates declined from 3.5 treatments in the first year to 0.1 in the fifth year; subjects who remained in the study lost an additional 2.3 lines of letters over the 3-year extension.

 

Ranibizumab was compared with photodynamic therapy in a multicenter, double-blind study (423 patients) by the ANCHOR study group (Brown, 2006).  Following 12 monthly treatments, patient groups treated with ranibizumab (0.3 or 0.5 mg) and sham verteporfin had 94% to 96% of subjects lose fewer than 15 letters. The patient group treated with monthly sham injection and active verteporfin therapy (average 2.8 times over the year) had 64% of subjects lose fewer than 15 letters. Visual acuity improved by more than 15 letters in 36% and 40% of the ranibizumab groups (average dose-dependent gain of 8.5 and 11.3 letters), in comparison with 5.6% of subjects in the verteporfin group (average loss of 9.5 letters). Intraocular inflammation occurred in 10.2% and 15% of ranibizumab-treated patients, with presumed endophthalmitis in 1.4% and serious uveitis in 0.7% of patients treated with the highest dose.

 

It appears, based on numerous case reports and case series, that photodynamic therapy is being used in an attempt to decrease CNV of many different etiologies. For example, photodynamic therapy has been reported to slow down, but not prevent or reverse, the progression of disease of CNV associated with angioid streaks and inflammatory chorioretinal disease  (Heimann, 2005) (Lim, 2006). Controlled studies are lacking, and photodynamic therapy is considered investigational for ophthalmologic disorders other than AMD, pathologic myopia, or presumed ocular histoplasmosis.

 

The FOCUS study group reported first- and second-year results of a blinded phase I/II multicenter, randomized controlled trial of ranibizumab (0.5 mg) combined with photodynamic therapy (Heier, 2006) (Antoszyk, 2008).  Patients with subfoveal CNV secondary to AMD were randomized in a 2:1 ratio to ranibizumab (n=106) or sham (n=56) injection (initially 7 days) following verteporfin photodynamic therapy. PDT was repeated only if fluorescein angiography revealed persistent or recurrent leakage from CNV at evaluation visits (3-month intervals). A higher than expected rate of serious intraocular inflammation occurred in the first patients, and the 2 treatments were subsequently scheduled no closer than 21 days apart. Intent-to-treat analysis showed an average improvement in acuity of 5 letters at both 12 and 24 months (85% retention) with ranibizumab, compared with a decrease of 8 letters in the PDT-alone group. Twenty-nine percent of patients in the ranibizumab group received additional PDT treatment (average of 0.4 treatment), compared with 93% of patients in the PDT-alone group (average of 3 treatments). Visual acuity improved by 15 or more letters in 25% of patients treated with ranibizumab (plus PDT as needed) compared with 7% of the patients treated with PDT alone. Endophthalmitis or intraocular inflammation was observed in 16 (15%) patients treated with ranizumab. The majority of adverse events (9%) reported for the PDT-alone group were AMD-related (i.e., CNV, macular degeneration, retinal hemorrhage).

 

An open-label assessor-blinded trial (n=165) from Croatia with short-term (3-month) follow-up reported similar results with bevacizumab and PDT (Lazic, 2007).  The frequency of treatment was described only as “standard procedures.” Twenty-two of 52 (42%) patients improved by >0.2 (logarithm of the minimum angle of resolution) following combined treatment, compared with 1 (2%) patient treated with bevacizumab alone and none treated with PDT alone. Adverse events (21%) in the bevacizumab groups (alone or combined) were frequent, and included 3 pigment epithelial tears, 12 posterior vitreous detachments, and 7 cataract progressions.

 

The combined use of corticosteroids and PDT is not supported by recent trial results (Gilson, 2007) Iriyama, 2008).

 

Preferred Practice Patterns (practice guidelines) on photodynamic therapy from the American Academy of Ophthalmology (AAO) are based on the TAP and VIP reports described here.  The AAO recommends PDT with verteporfin for subfoveal CNV where the classic component is greater than 50% of the lesion and the entire lesion is 5400 microns or less in diameter.  While PDT was recommended for recurrent juxtafoveal CNV, evidence was considered insufficient to guide treatment recommendations for the primary use of PDT for juxtafoveal CNV. Combination therapy with VEGF inhibitors and PDT was not discussed.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Semeraro and colleagues published an RCT of 75 patients with treatment-naive exudative CNV due to AMD (Semeraro, 2015). Patients were randomized into 3 groups: ranibizumab monotherapy, ranibizumab combined with reduced-fluence verteporfin PDT, and ranibizumab combined with ketorolac eye drops. At the 12-month follow-up, BCVA (SD) was superior in the ranibizumab plus ketorolac group (-0.25 [0.60] logMAR), compared with ranibizumab monotherapy (-0.14 [0.52] logMAR) or ranibizumab combined with PDT (-0.10 [0.30] logMAR). Change in mean (SD) central retinal thickness was also superior in the ranibizumab plus ketorolac group (-141 [21] μm) compared with the ranibizumab monotherapy group (-125 [15] μm) and the ranibizumab plus PDT group (-130 [15] μm).

 

A Cochrane systematic review on the treatment of CSC, both acute and chronic, was published in 2015 (Salehi, 2015). For PDT monotherapy, there was 1 low-quality trial of patients with acute CSC comparing PDT to no treatment using a sham control. There was a small improvement in visual acuity for the PDT group (mean difference, -0.10 logMAR; 95% CI, -0.18 to -0.02). There was also a decrease in recurrence of CSC (risk ratio, 0.10; 95% CI, 0.01 to 0.81) and a trend toward a lower risk of persistent CSC (risk ratio, 0.12; 95% CI, 0.01 to 1.02). Two low-quality trials compared anti-VEGF agents with PDT and reported no difference in BCVA at 1 year. There was a trend toward less recurrence and less persistent CSC in the PDT group, but these results were inconsistent across trials.

 

Tang and colleagues published a systematic review in 2015 evaluating treatment for polypoidal choroidal vasculopathy (Tang, 2015).  PDT alone was compared to ranibizumab alone and to combination ranibizumab plus PDT. This review included 3 RCTs and 6 retrospective comparative studies. For PDT alone versus ranibizumab alone, 2 RCTs reported the weighted mean difference in visual acuity was 0.06 logMAR (95% CI, -0.01 to 0.12) in favor of ranibizumab alone, but this difference was not statistically significant. For combination therapy versus PDT alone, a single RCT reported that there was a nonsignificant weighted mean difference of-0.08 (95% CI, -0.20 to 0.04) in favor of combination therapy.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NCT02821520) Initial Versus Delayed PDT Combination with Conbercept in PCV; planned enrollment 80; projected completion date December 2019.  

 

 

In 2018, the National Institute for Health and Care Excellence updated its 2003 guidance on the use of PDT for AMD (NICE, 2003; NICE, 2018). The Institute made the following recommendations: it recommended against use of PDT as monotherapy for late (wet) AMD and against use of PDT as first-line adjunctive therapy to anti-VEGF therapies for late (wet) AMD; it recommended for PDT as second-line adjunctive therapy to anti-VEGF therapies for late (wet) AMD in a trial setting.