Systemic Safety Of Intravitreal VEGF Inhibitors

What concerns remain?

USHA CHAKRAVARTHY, MD, PhD, FRCS, FRCOphth • CHRISTINE ROGERS, PhD

Ever since the discovery of vascular endothelial growth factor as a key mediator of angiogenesis,¹ researchers have studied it extensively. We now know that it plays a multitude of trophic roles in the health and functioning of many mammalian tissues and cells.²

While its roles in the maintenance of a nonthrombogenic vascular endothelium, in the promotion of the regeneration of endothelial buds in newly formed vessels, and in ensuring the patency of fenestrations in microvascular beds come as no surprise, this molecule is also trophic to the neural cells of the developing brain, axons of the spinal cord, lung epithelium, retinal neurons, and RPE, to name but a few.^2,3

As a result, legitimate concerns have arisen that the therapeutic use of antibodies that block this vital molecule to ameliorate the exudative pathology in neovascular AMD may turn out to be a double-edged sword.

THE ORIGINAL DATA

The pharmacokinetics of ranibizumab (Lucentis, Genentech, South San Francisco, CA), the first monoclonal antibody tested for the treatment of neovascular AMD, initially allayed fears of adverse systemic side effects when it revealed almost no detectable drug in peripheral blood.^4,5

The information on ranibizumab from the European Medicines Agency detailing the product characteristics⁶ states that following monthly intravitreal administration of ranibizumab to patients with neovascular AMD, serum concentrations of ranibizumab were generally low, with maximum levels (Cmax) lower than that necessary to inhibit the biological activity of VEGF by 50% (11-27 ng/mL, as assessed in an in vitro cellular proliferation assay). Cmax was dose proportional over the dose range of 0.05 to 1.0 mg/eye.

Revisiting the data from the original pivotal clinical trials, which reported that the rates of systemic adverse events were similar in the groups that received anti-VEGF treatment when compared to natural history groups⁷ or to PDT-treated groups,⁸ subsequent analyses combining the data from MARINA, ANCHOR, and a further smaller phase 2 trial of ranibizumab (FOCUS) found that cerebrovascular events were more common in patients exposed to ranibizumab (2.2%) vs nonexposed patients (0.7%).⁹

In addition, an open-label extension study, SAILOR, which compared two doses of 0.3 mg with two doses of 0.5 mg, showed a 1.2% occurrence rate of stroke in the 0.5-mg arm vs 0.3% in the lower-dose arm.

However, in later analyses, this difference was no longer statistically significant.¹⁰ Notably, the SAILOR data also suggested that the stroke risk was greatest among patients with a history of stroke.¹⁰

SPECIFIC ISSUES WITH BEVACIZUMAB

The widespread off-label use of bevacizumab (Avastin, Genentech), also for neovascular AMD, because of its relative inexpensiveness compared to ranibizumab, began to raise concerns, particularly when systematic reviews of clinical trials comparing the clinical effectiveness of the two drugs, as well as some clinical databases, appeared to suggest a higher incidence of systemic serious adverse events (SAEs).^11,12

This relationship appeared to be the case when the one-year safety outcomes of two large comparative effectiveness trials (CATT and IVAN) underwent pooled analysis.^13–16 While the risk of death or arteriothrombolic events (ATEs) was similar between the drugs, other systemic SAEs were more common in the bevacizumab-treated patients in the pooled dataset.

Notably, 271 of 913 participants in the ranibizumab arm vs 314 of 882 in the bevacizumab arm suffered SAEs, with a statistically significant odds ratio of 0.76 (95% CI: 0.58, 0.95; P = .01). Additional data from the publication of one-year data from the MANTA¹⁷ and GEAFAL¹⁸ studies have accrued, combined with the two-year data from CATT and IVAN, the risk of death and ATEs was not different between these drugs. However, the risk of systemic SAEs remained (362 of 1,291 in the bevacizumab arm vs 310 of 1,315 in the ranibizumab arm), with an OR of 0.77 in favor of ranibi-zumab (0.64, 0.93; Figure).

Figure. Pooled analysis of all SAEs in CATT, IVAN (two-year follow up), MANTA, and GEAFAL (one-year follow-up).

Longer Follow-ups

Interestingly, with longer follow-up, which have only been available in CATT and IVAN, the pooled analysis for any systemic SAE, while remaining significant, disguised the inconsistency between the two trials.

Importantly, the SAEs accruing in the second year in the IVAN trial shifted the two-year OR almost to unity. Possible explanations for the difference in the frequency of harm noted between year 1 and year 2 in the IVAN trial include the absence of any true increase in risk and increased risk only manifesting early in the treatment cycle.

In this context, immunological sensitization to a drug resulting in unexpected illnesses is a plausible theory, given that bevacizumab is a larger molecule, and egress into the systemic circulation does appear to occur, as the considerable drop in serum VEGF levels in IVAN participants who received this drug has shown.

DOSING SCHEDULES

Particularly intriguing is the finding of an increased frequency of SAEs when treatment does not occur monthly. The only trials that have tested monthly administration vs less frequent dosing are CATT and IVAN.^13–16

CATT had true PRN arms from treatment initiation, with retreatment given if the macula was not fluid-free at any follow-up visit after randomization, on the basis of OCT.

In IVAN, both drug and regimen comparison arms received treatment every month for three months as a loading phase. Subsequently, those assigned to the less frequent dosing arm only received treatment if specific failure criteria were met.

If these criteria, based on the presence of fluid in the macular tissue compartments on OCT or angiography combined with functional assessments, were met, treatment was reinitiated, but with a mandated cycle of three.

The pooling of SAEs observed in the CATT PRN arms with the IVAN discontinuous treatment arms showed a significant difference in systemic SAEs, mainly due to the increased frequency of death.

Specifically, 18 of 895 participants who received monthly doses and 36 of 900 who received less frequent dosing died. The OR of 0.49 (0.27, 0.86) was statistically significant, and this probably represents the finding of greatest concern.

However, we must note that only one-year data from CATT could be included in this analysis because of the rerandomization in CATT of participants at 12 months, which resulted in half of those originally assigned to the PRN arm being switched to monthly fixed dosing.

NEW DRUGS, FEWER CONCERNS?

Ranibizumab was a refinement of the full-length humanized monoclonal antibody bevacizumab to its Fab fragment. Both antibodies recognize all isoforms of VEGF, but ranibizumab binds to VEGF with greater affinity.

Ranibizumab was the first anti-VEGF agent successfully introduced into clinical practice, while bevacizumab has been a common treatment for neovascular AMD for many years. As a result, a number of databases, such as those maintained by organizations such as Medicare, have undergone scrutiny to identify potential adverse events.

Initial analyses of such registries suggested a higher risk of ATEs with bevacizumab. However, with adjustment for potential confounders, such as socioeconomic class, these differences disappeared.¹⁹

A Canadian study, in which the exposure of interest was the use of an intraocular anti-VEGF drug over the 180 days before an ATE (myocardial infarction, stroke, and any venous thromboembolism), found no increase in risk between exposed and unexposed cases. Nor did a difference in risk emerge by drug type, namely bevacizumab vs ranibizumab.^20,21

However, findings of the analyses from routinely accrued datasets are at variance with one another. The analyses of another population-based registry found a higher incidence of ATEs in patients with neovascular AMD, with the 12-month rate of myocardial infarctions nearly two times greater with an OR of 2.3 (1.2, 4.5) in persons exposed to any anti-VEGF drug, compared to historical controls treated with PDT.

Nevertheless, no differences between ranibizumab and bevacizumab emerged, nor were any differences in the rates of stroke or gastrointestinal bleeding detected.²²

CONCERNS PERSIST

Concerns have also arisen regarding the safety of aflibercept (Eylea, Regeneron, Tarrytown, NY), an anti-VEGF drug with a slightly different mechanism of action.²³ Furthermore, anti-VEGFs are becoming the standard of care for DME and RVO, conditions with increased risk of stroke and myocardial infraction compared to age-matched populations.

As a result, a perception exists that anti-VEGF use in these other exudative maculopathies may result in an increased rate of ATEs. Serum concentrations of VEGF inhibitors, measured in a limited number of DME patients, have indicated that slightly higher systemic exposure cannot be excluded, compared to those observed in neovascular AMD patients.

Serum ranibizumab concentrations in RVO patients were similar or slightly higher, compared to those observed in neo-vascular AMD patients.

CONCLUSION

In summary, concerns for systemic safety remain on two fronts when physicians use biologicals that inhibit VEGF. Anti-VEGF agents given intraocularly egress into the systemic circulation and have the potential to influence vascular health and cause small increases in the absolute risk of ATEs.

However, no significant differences appear to exist between ranibizumab and bevacizumab in this context, although the risk of other SAEs may be higher with bevacizumab.

An increased risk also appears to arise when anti-VEGF drugs are given in fits and starts. Immunological sensitization may play a role in determining this risk. RP

REFERENCES

1. Ferrara N. VEGF and the quest for tumour angiogenesis factors. Nat Rev Cancer. 2002;2:795-803.

2. Takahashi H, Shibuya M. The vascular endothelial growth factor (VEGF)/VEGF receptor system and its role under physiological and pathological conditions. Clin Sci (Lond). 2005;109:227-241.

3. Zisa D, Shabbir A, Suzuki G, Lee T. Vascular endothelial growth factor (VEGF) as a key therapeutic trophic factor in bone marrow mesenchymal stem cell-mediated cardiac repair. Biochem Biophys Res Commun. 2009;18:834-838.

4. Gaudreault J, Fei D, Rusit J, Suboc P, Shiu V. Preclinical pharmacokinetics of ra-nibizumab (rhuFabV2) after a single intravitreal administration. Invest Ophthalmol Vis Sci. 2005;46:726-733.

5. Xu L, Lu T, Jumbe N, et al. Pharmacokinetics of ranibizumab in patients with neovascular age-related macular degeneration: a population based approach. Invest Ophthalmol Vis Sci. 2013;54:1616-1624.

6. European Medicines Agency. Production information. Avaialble at: http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000715/WC500043546.pdf. Accessed December 15, 2013.

7. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med. 2006;355:1419-1431.

8. Brown DM, Michels M, Kaiser PK, et al. Ranibizumab versus verteporfin photo-dynamic therapy for neovascular age-related macular degeneration: Two-year results of the ANCHOR study. Ophthalmology. 2009;116:57-65.

9. Ueta T, Yanagi Y, Tamaki Y, Yamaguchi T. Cerebrovascular accidents in ranibi-zumab. Ophthalmology. 2009;116:362.

10. Singer MA, Awh CC, Sadda S, et al. Horizon an open label extension trial of ranibizumab for choroidal neovascularization secondary to age-related macular degeneration. Ophthalmology, 2012;119:1175-1183.

11. Lanzetta P, Mitchell P, Wolf S, Veritti D. Different antivascular endothelial growth factor treatments and regimens and their outcomes in neovascular age-related macular degeneration; a literature review. Br J Ophthalmol. 2013;97:1497-1507.

12. Schmucker C, Ehlken C, Agostini H, Anges G, Ruecker G, Lelgemann M, Loke Y. A safety review and meta analysis of bevacizumab and ranibizumab: off label versus gold standard. PLoS One. 2012;7:e42701.

13. Chakravarthy U, Harding SP, Rogers CA, et al. Ranibizumab versus bevacizumab to treat neovascular age-related macular degeneration: one-year findings from the IVAN randomized trial. Ophthalmology. 2012;119:1399-1411.

14. Chakravarthy U, Harding SP, Rogers CA, et al; IVAN study investigators. Alternative treatments to inhibit VEGF in age-related choroidal neovascularisation: 2-year findings of the IVAN randomised controlled trial. Lancet. 2013;382:1258-1267.

15. CATT Research Group; Martin DF, Maguire MG, Ying GS, et al. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med. 2011;364:1897-1908.

16. Comparison of Age-related Macular Degeneration Treatments Trials (CATT) Research Group; Martin DF, Maguire MG, Fine SL, et al. Ranibizumab and bevacizumab for treatment of neovascular age-related macular degeneration: two-year results. Ophthalmology. 2012;119:1388-1398.

17. Krebs I, Schmettererr S, Bolts A, et al. A randomised double-masked trial comparing the visual outcome after treatment with ranibizumab or bevacizumab in patients with neovascular age-related macular degeneration. Br J Ophthalmol. 2013;97:266-271.

18. Kodjikian L, Souied EH, Mimoun G, et al. Ranibizumab versus bevacizumab for neovascular age-related macular degeneration: results from the GEFAL non inferiority randomized trial. Ophthalmology. 2013;120:2300-2309.

19. Curtis LH, Hammill BG, Schulman KA, Cousins SW. Risks of mortality, myocardial infraction, bleeding and stroke associated with therapies for age-related macular degeneration. Arch Ophthalmol. 2010;128:1273-1279.

20. Campbell RJ, Gill SS, Bronskill SE, Paterson JM, Whitehead M, Bell CM. Adverse events with intravitreal injection of vascular endothelial growth factor inhibitors: nested case control study. BMJ. 2012;345:e4203.

21. Campbell RJ, Bell CM, Paterson JM, et al. Stroke rates after introduction of vascular endothelial growth factor inhibitors for macular degeneration: A time series analysis. Ophthalmology. 2012;119:1604-1668.

22. Kemp A, Preen DB, Morlet N, et al. Myocardial infarction after intravitreal vascular endothelial growth factor inhibitors: A whole population study. Retina. 2013;33:920-927.

23. Beaumont PE, Petocz P, Kang HK. Is there a risk of stroke with aflibercept. Ophthalmology. 2013 Nov 20. [Epub ahead of print]