DMLA et accidents thrombo-emboliques (AVC, infarctus)

Les résultats à un an de l’étude SAILOR et une étude parue dans le British Journal of Ophthalmology apportent de nouveaux éléments quant au risque d’accidents thrombo-embolique des patients atteints de DMLA :

  1. Alors que l’analyse intermédiaire (à 6 mois) de SAILOR faisait craindre une augmentation du risque d’AVC sous Lucentis (notamment pour les patients ayant déjà des antécédents d’accidents de ce type), les données à un an, communiquées récemment, sont plutôt rassurants, avec certes toujours une « tendance » à une incidence supérieure d’AVC sous Lucentis 0.5 mg par rapport à 0.3 (1.2% vs 0.7%), mais la différence n’est pas statistiquement significative (p=0.21) :

    One-year data on the phase 3b SAILOR (Safety Assessment of Intravitreal Lucentis for AMD) trial showed the drug was safe and was not associated with higher rate of stroke, according to principal investigator David S. Boyer, MD, at Bascom Palmer Eye Institute’s Angiogenesis, Exudation, and Degeneration 2008 meeting in Key Biscayne, Florida.

    The study included approximately 4,300 patients with all subtypes of new or recurrent active subfoveal wet age-related macular degeneration (AMD) who were treated with 0.3 mg or 0.5 mg doses of intravitreal ranibizumab (Lucentis, Genentech). Results were based off of information gathered from cohort 1, which included 2,378 patients.

    These data come almost 1 year after Genentech issued a letter to physician-users (a so-called « Dear Doctor » letter) in January 2007, noting safety concerns in the interim safety analysis of SAILOR. That preliminary data showed a four-fold difference in the rate of strokes in the group receiving the 0.5-mg (1.2%) than in the group receiving the 0.3-mg dose (0.3%) at an average follow-up of 230 days. Patients with previous history of stroke appeared to be at higher risk for stroke during the trial.

    One-year results, however demonstrated that although the US Food and Drug Administration (FDA) approved dose of Lucentis (0.5 mg) trended toward a higher incidence of stroke (1.2% vs. 0.7% in the 0.3 mg dose group) the results were not statistically significant (P=.21). The stroke rate in the general US population (65 years of age or older) is 1.4%, according to Anne E. Fung, MD, who presented at the same meeting on the rates of arterial thromboembolic events in Medicare patients.

    « I think that we over-analyzed the 6-month data; perhaps it was blown a bit out of proportion in comparison to all of the other studies out there at the time, » Dr. Boyer said in a telephone interview with Retina Today. « Genentech reported the 6-month data, as it should have. But I think that [the results of SAILOR] are one of the reasons you wait for a year to get the complete data. We now see that there was no probable statistical significance — the numbers just caught up by the end of the trial. If we examined the data for a longer period of time, those numbers for the two dosing groups might even come closer in line to each other. »

    As SAILOR was a 1-year study, there are no further plans to continue the trial, Genentech spokeswoman Krysta Pellegrino told Retina Today. Results for the second SAILOR cohort (approximately 1,900 patients), however, will be available sometime this year.

    Sur MDSupport.org, il est précisé que cette tendance (9.6% vs 2.7% quand même !) est également observée dans le sous-groupe de patients aux antécédents d’AVC, mais qu’il est difficile de conclure en raison du faible nombre d’évènements :

    At one-year, patients with a prior history of stroke had a higher rate of stroke in the 0.5 group (9.6%) compared to the 0.3 group (2.7%). However, this trend was inconclusive, as the number of events was small. These data are consistent with epidemiologic data showing that prior history of stroke predisposes patients to subsequent stroke.

  2. Dans une publication en ligne depuis le 29 février, basée sur les examens de patients 10 ans après leur inclusion dans la Blue Mountain study, les sujets de moins de 75 ans avec des signes de maculopathie liée à l’âge à l’inclusion ont un risque de décéder par AVC ou infarctus du myocarde à 10 ans multiplié par 2, par rapport aux patients sans signe de DMLA.
    Pour ceux qui avaient déjà à l’inclusion des signes de DMLA compliquée, le risque est multiplié par 5 (infarctus du myocarde) à 10 (AVC).

    Background/aims: Age-related macular degeneration (AMD) and vascular disease share similar risk factors. Recent data suggest AMD may independently predict stroke or coronary heart disease. We prospectively assessed the relationship between AMD and risk of stroke- or cardiovascular-related death in an Australian population.
    Methods: Of 3654 baseline participants (1992-4) aged 49+years, 2335 were re-examined after 5-years and 1952 after 10-years. Retinal photographs were graded using the Wisconsin System. History and physical examination provided data on possible risk factors. Deaths and cause of death were confirmed by data linkage with the Australian National Death Index. Risk ratios (RR) were estimated in Cox models.
    Results: Among persons aged <75 years at baseline, early AMD predicted a doubling of cardiovascular mortality (RR, 2.32; 95% confidence interval (CI), 1.03-5.19), over the next decade, after controlling for traditional cardiovascular risk factors. Late AMD predicted 5-fold higher cardiovascular mortality (RR, 5.57; CI, 1.35-22.99) and 10-fold higher stroke mortality (RR, 10.21; CI, 2.39-43.60) after adjusting for age and sex only. These associations were not present when persons older than 75 were included. Conclusion: AMD predicted stroke and cardiovascular events over the long-term in persons aged 49-75 years. This may have potential implications for new intravitreal anti-VEGF AMD therapies. Tan JS, Wang JJ, Liew G, Rochtchina E, Mitchell P. Age-related macular degeneration and mortality from cardiovascular disease or stroke. Br J Ophthalmol. 2008 Feb 29 [Epub ahead of print]

Donc, chez les patients traités par anti-VEGF pour DMLA exsudative, il semble que la probabilité de décéder d’un AVC ou d’un infarctus du myocarde soit plus du fait de facteurs de risques partagés entre DMLA et pathologies cardio-vasculaires, que d’un traitement anti-VEGF…
Mais, en raison du (finalement) faible nombre d’évènements, il faudra probablement attendre d’avoir un peu plus de recul pour conclure de manière formelle…

Source : Retina Today eNews