Complement Inhibitor Fails in Late-Stage Trials for Geographic Atrophy

Caroline Helwick

November 13, 2017

NEW ORLEANS  — Complement inhibition as a means of halting the progression of geographic atrophy took a hit this week, with negative findings announced for two pivotal phase 3 clinical trials of lampalizumab.

Researchers reported here at the American Academy of Ophthalmology (AAO) 2017 Annual Meeting that Spectri, the first of the phase 3 trials, failed to meet its primary endpoint: a reduction in progression of geographic atrophy (GA) at 48 weeks compared with the control. Genentech, which is developing lampalizumab, simultaneously issued a press release reporting the same negative conclusion from the second, identically designed phase 3 Chroma trial. The company, however, did not release specific findings from Chroma.  

In his late-breaking presentation, Jeffrey S. Heier, MD, said, "Spectri's primary endpoint was not met. Lampalizumab was not effective in reducing GA lesion growth. Chroma also demonstrated no treatment benefit." Dr Heier, who is an investigator on the trial, is co-president and medical director of the Vitreoretinal Service and director of Retina Research at Ophthalmic Consultants of Boston,  Massachusetts.

"Based on the totality of data from Chroma and Spectri, Genentech will not pursue approval from the U.S. Food and Drug Administration for lampalizumab in geographic atrophy," a company press release stated.

Findings remained encouraging, however, for another investigational complement inhibitor, APL-2 (Apellis) evaluated in a phase 2 trial, according to another late-breaking presentation by David Boyer, MD, senior partner of the Retina Vitreous Medical Group in Los Angeles, California. "In the FILLY trial, APL-2 slowed the growth of GA, independent of genetics," Dr Boyer reported.

Both lampalizumab and APL-2 inhibit factors within the alternative complement pathway, a component of the immune system. Complement is known to play a role in the development of age-related macular degeneration (AMD). Presumably, local inflammation plus complement activation contributes to the pathogenesis and progression of the disease. Dysfunction of the alternative complement pathway has been linked to the pathogenesis of AMD through numerous genetic studies.

Lampalizumab Trials

The monoclonal antibody lampalizumab is directed against complement factor D, a rate-limiting enzyme involved in the activation and amplification of the alternative complement pathway.

Spectri and Chroma were identically designed global phase 3 trials comparing lampalizumab 10 mg administered every 4 or 6 weeks by intravitreal injection vs sham injections. Together, the studies enrolled more than 1800 patients with GA due to AMD.  The primary objective of both trials was to show a smaller change in GA lesion area in patients treated with lampalizumab compared with sham at week 48, as measured by fundus autofluorescence. 

Dr Heier reported only the results of Spectri, which enrolled 975 patients (mean age, 78 years) with bilateral GA and no choroidal neovascularization. At baseline, the mean GA area was 7.5 mm2 in the control group, 8.4 mm2 in the group receiving lampalizumab every 4 weeks, and 8.5 mm2 in patients receiving lampalizumab every 6 weeks. More than 90% of patients completed the first 48 weeks of the study. The two sham groups were pooled for the analysis.

Lampalizumab given every 4 or 6 weeks did not result in a greater reduction in GA area growth than did sham injections, Dr Heier reported. Mean change from baseline, adjusted for potential confounders, was 2.089 mm2 with lampalizumab every 4 weeks, 2.019 mm2 with lampalizumab every 6 weeks, and 1.932 mm2 with sham injections.

The differences between active treatment and controls is shown in the table below. Although the endpoint was statistically significant, the investigators concluded the difference was not clinically meaningful.

A sensitivity analysis of an alternative measure — change from baseline in the square root of the GA area — also showed no significant difference over sham and "no impact of lampalizumab," Dr Heier reported.

Table. Difference in Mean Change From Baseline vs Sham

Endpoint Lampalizumab Every 4 Weeks P Value Lampalizumab Every 6 Weeks P Value
GA area change from baseline (mm2) 0.157 .048 0.087 .274
Change from baseline in square root GA area (mm) 0.014 .276 0.003 .822


Spectri and Chroma also sought to determine whether complement factor I (CFI) might serve as a biomarker of response to lampalizumab. Genetic variants of CFI, a key negative regulator of the alternative complement pathway, have been linked to the risk for advanced AMD, and phase 2 studies have found differential treatment response based on CFI profile, Dr Heier said. 

For CFI-positive patients, outcomes did not differ among the treatment groups. In CFI-negative patients, however, those receiving lampalizumab every 4 weeks had greater GA growth than patients treated in the sham group (0.340 mm2 vs controls; P = .006). "We saw greater progression in the lampalizumab group. However, if you look at the sham arm, there was a lower growth rate than was seen in the other arms," Dr Heier said. "Also, Chroma did not show this difference, so we believe this is due to chance."

The safety profile was in line with that seen in previous lampalizumab trials and with other intravitreal therapies. Rates of conversion to neovascular AMD were low and mirrored rates reported in the literature.  

In conclusion, Dr Heier noted that the lampalizumab development program "is the most comprehensive study of GA to date" and indicated that a deeper evaluation of these studies is underway to help inform the design of future therapies.

"This rich data set of over 1000 patients, including genetics and long-term visual function outcomes, will help advance our understanding of disease pathophysiology and natural history of this disease," he said. 

FILLY Phase 2 Trial of APL-2

APL-2 is an inhibitor of C3, which is involved in the classic and alternative complement pathways. Blocking C3 suppresses inflammation and affects other factors associated with disease.

The FILLY trial enrolled 246 patients with GA, approximately 85% of whom had bilateral disease. Patients were randomly assigned to monthly injections of APL-2, APL-2 injected every other month, or sham injections monthly or every other month. Mean baseline GA lesions size was 8.0 mm2 for APL-2 monthly, 8.9 mm2 for APL-2 every other month, and 8.2 mm2 for the control groups. "The GA areas were smaller for the APL-2 monthly arm but well matched otherwise," Dr Boyer noted.

The primary endpoint was mean change in the square root GA area from baseline to month 12. Two key secondary endpoints were change in best corrected visual acuity (BCVA) and incidence of new-onset exudative AMD.

FILLY Endpoint Met

The phase 2 FILLY trial met its primary endpoint: Patients treated monthly with APL-2 achieved a 28.6% lower GA lesion growth compared with the pooled control groups.  Mean change from baseline in square root GA lesion area was 0.35 mm2 for sham-treated patients and 0.25 mm2 for patients receiving monthly treatments with APL-2 (P = .008), Dr Boyer reported. 

For patients receiving the drug every other month, mean change was 0.28 mm2, a 20.0% reduction over sham (P = .067). The treatment effect was independent of genetic biomarkers, including CFI, he said.

Visual outcomes did not differ between the groups. Mean change from baseline BCVA was –3.3 letters in patients receiving APL-2 monthly, –4.3 letters in those receiving APL-2 every other month, and –5.5 in the sham pooled groups. "This was done to look for toxicity or worsening with drug. The study was not powered to determine differences in vision," he explained.

New-onset wet AMD, however, was observed in 1% of patients in the sham groups compared with 18% of patients receiving APL-2 monthly and 8% of those receiving it every other month. Among patients with a history or presence of choroidal neovascularization in the fellow eye, new-onset wet AMD was seen in 0%, 29%, and 14%, respectively. Without choroidal neovascularization, these percentages were 2%, 10%, and 4%.  

"APL-2 appeared to increase the risk of new-onset wet AMD, though this was a small percentage and did not affect visual outcomes or growth of lesions," Dr Boyer commented.

Adverse events were similar between treatment groups, and most were related to conversion to the exudative form of AMD. Three patients receiving APL-2 developed endophthalmitis.  

Should We Treat Earlier?

Several retinal specialists suggested that treating GA earlier might make a difference with complement inhibitors. 

"Did Spectri treat too late? That's an open question, but we did see disparity in the results of the two studies, where attacking complement factor 3 [in Filly] appeared to have more benefit and to affect the lesion growth rate, vs looking at complement factor D [in Spectri]. There is a signal from Filly that is promising," said Baruch Kuppermann, MD, PhD, the Roger F. Steinert Professor and chair of ophthalmology at the University of California, Irvine.

Another panelist at the late-breaking session, Michael A. Singer, MD, from San Antonio, Texas, echoed Dr Kuppermann in an interview with Medscape Medical News. "The question is whether we need to treat earlier," he said. "To figure out whether we need to treat earlier, we essentially have to try and fail. Do we keep with the same alternative complement pathway and treat earlier, or maybe identify another pathway that may be more successful?" he said.  

"The Filly trial showed promise, but it's a phase 2 trial. We [will be] excited to see results from a phase 3 trial of APL-2," Dr Singer commented. 

"Obviously, it's disappointing that lampalizumab did not work out, but Genentech did a great job in giving us a great understanding of the underlying disease of geographic atrophy. The work they have done will set the stage for new clinical trials, which will hopefully utilize that data to create better drugs." 

Dr Heier reported financial relationships with Genentech/Roche, which sponsored the Spectri trial, as well as 4D Molecular Technologies, Adverum, Aerpio, Allegro, Apellis, Asclepix, Bayer, Beaver-Visitec, Daiichi-Sanyo, Genentech/Roche, Heidelberg, Hemera, Janssen, Kanghong, Kodiak, Neurotech, Notal Vision, Novartis, Ocular Therapeutix, Ocunexus, Optos, Quark, Regeneron, Regenxbio, Santen, SciFluor, Shire, Steal Biotherapeutics, Thrombogenics, and Tyrogenex. Dr Boyer has served as a consultant for Apellis, which sponsored the Filly trial, and reported financial relationships as well with Acucela, Aerpio, Alcon Laboratories, Alimera Sciences, Allegro,  Allergan, Amydis, Bausch+Lomb, Bayer Healthcare Pharmaceuticals, BioMotiv, Boehringer-Ingelheim Pharmaceuticals, Clearside Biomedical, CoDa Therapeutics, DigiSight, Foresight Biotherapeutics, Genentech, GenSight Biologics, Glaukos Corporation, GlaxoSmithKline, GrayBug Vision, Ionis Pharmaceuticals, Isarna Therapeutics, Notal Vision, Novartis, Ocular Therapeutics, Ophthotech, Optos, OptoVue, Ora, pSivida Corporation, Regeneron Pharmaceuticals, Regenxbio, Regulus Therapeutics, River Vision, Roche, Santen, Shire, Sun Pharmaceutical Industries, Taiwan Liposomal Company, and ThromboGenics, and Zeiss. Dr Kuppermann disclosed relationships with Aerpio, Alcon Laboratories, Alimera Sciences, Allegro, Allergan, Ampio, Apellis, Catalyst, Dose, Eleven Biotherapeutics, Genentech, Glaukos Corporation, GlaxoSmithKline, J-Cyte, Lumenis, Inc, Neurotech, Novartis, Alcon Pharmaceuticals, Ohr, Ophthotech, Regeneron Pharmaceuticals, SciFluor, and ThromboGenics Inc. Dr Singer disclosed relationships with Aerpio, Aestelis, Alimera Sciences, Allergan, Ampio, Bayer Healthcare Pharmaceuticals, Clearside, Genentech, Optos, Regeneron Pharmaceuticals, and Sante.

American Academy of Ophthalmology (AAO) 2017 Annual Meeting. Presented on Retina Subspecialty Day. November 11, 2017.

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.