PERSEUS 24-month analysis: a prospective non-interventional study to assess the effectiveness of intravitreal aflibercept in routine clinical practice in Germany in patients with neovascular age-related macular degeneration.


Eter N(1), Hasanbasic Z(2), Keramas G(2), Rech C(3), Sachs H(4), Schilling H(5), Wachtlin J(6)(7), Wiedemann P(8), Framme C(9); PERSEUS Study Group.
Author information:
(1)Klinik für Augenheilkunde, Universitätsklinikum-Augenklinik Münster, Domagkstr. 15, 48149, Münster, Germany. [Email]
(2)Bayer Vital GmbH, Leverkusen, Germany.
(3)Data Generation, Bayer Vital GmbH, Leverkusen, Germany.
(4)Augenklinik, Städtisches Klinikum Dresden-Friedrichstadt, Dresden, Germany.
(5)Klinik für Augenheilkunde, St.-Johannes-Hospital, Dortmund, Germany.
(6)Abteilung für Augenheilkunde, Sankt Gertrauden-Krankenhaus, Berlin, Germany.
(7)MHB, Medizinische Hochschule Brandenburg, Neuruppin, Germany.
(8)Klinik und Poliklinik für Augenheilkunde, Universitätsklinikum Leipzig, Leipzig, Germany.
(9)Klinik für Augenheilkunde, Medizinische Hochschule Hannover, Hannover, Germany.


PURPOSE: To evaluate the real-world effectiveness of intravitreal aflibercept injections in Germany in patients with neovascular age-related macular degeneration over 24 months. METHODS: PERSEUS was a prospective, non-interventional cohort study. The primary endpoint was the mean change in visual acuity (VA) from baseline. Secondary endpoints included the proportion of patients with a VA gain or loss of ≥ 15 letters and the frequency of injections and examinations. Patients with regular (bimonthly after 3 monthly injections during year 1 and ≥ 4 injections in year 2) and irregular (any other) treatment were analyzed. The last observation carried forward (LOCF) and the observed cases (OC) approach was applied for primary endpoint analysis to account for missing data. RESULTS: 803 patients were considered for effectivity analysis. At month 24, only 38% of the patients were still under observation. The LOCF population included 727, the OC population 279 patients. Treatment-naïve patients improved by 6.3 (LOCF)/8.1 (OC) letters with regular treatment over 24 months but only by 3.3 (LOCF)/3.1 (OC) letters with irregular treatment. The proportion of treatment-naïve patients achieving a VA improvement of ≥ 15 letters was similar between regularly and irregularly treated cohorts. However, considerably more patients in the irregular cohorts experienced a VA worsening of ≥ 15 letters than in the regular cohorts (LOCF: 18.7% vs. 7.4%). CONCLUSIONS: Regular IVT-AFL treatment resulted in better VA outcomes than irregular treatment at month 24. However, only a minority of patients received regular treatment over a 2-year period.