There are concerns about nephropathy after intravitreal anti-vascular endothelial growth factor (VEGF) treatment, which is commonly used in ophthalmic clinical practice for diseases such as age-related macular degeneration (nAMD) and diabetic macular oedema (DMO) [1, 2]. A recent systematic review and meta-analysis in this journal found that trials of these drugs do not often report renal side effects, yet there is increasing evidence that systemic absorption of anti-VEGF drugs can cause renal dysfunction due to systemic VEGF downregulation or direct drug–kidney interactions [1, 3].
Cases reporting nephropathy tend to fall into one of two categories: those with underlying renal disease, who have accelerated disease progression following treatment; and those with new-onset kidney disease attributed to reduced systemic VEGF levels or allergic reactions (e.g. thrombotic microangiopathy and acute interstitial nephritis). This suggests that systemic drug absorption might impact different compartments within the kidney [4].
We conducted a prospective pharmacodynamic study of 53 participants comparing the effects of ranibizumab and aflibercept on systemic serum markers of renal function. Subgroup analyses were pre-planned for drug given and the baseline estimated glomerular filtration rate (eGFR, >60 or <60 mL/min/1.73 m2). We excluded participants who had received an anti-VEGF injection in the preceding 8 weeks, those undergoing renal dialysis, and those taking any form of immunosuppressive or anti-inflammatory medication. Ethical approval for the study was obtained from the National Research Ethics Service committee. All participants provided written informed consent and procedures adhered to the Helsinki Declaration.
Blood samples were taken immediately before the intravitreal anti-VEGF injection, and at 1 week and 1 month afterwards. Aliquots of serum were stored at –80°C prior to renal marker (serum electrolytes, urea and creatinine) testing using the Advia chemistry system (Siemens Healthcare Diagnostics Inc., Munich, Germany). The eGFR was calculated from serum creatinine using the Chronic Kidney Disease Epidemiology Collaboration equation [5].
Baseline characteristics were well balanced between the groups, however there was a significantly higher number of previous intravitreal injections in the aflibercept group than the ranibizumab group (17.7 ± 2.9 vs 10.8 ± 1.9; P = .05), perhaps because aflibercept was introduced with a mandated regular dosing regimen, versus an initial pro re nata regimen for ranibizumab. None of the participants had a history of kidney transplantation.
The mean serum concentrations of sodium, potassium, urea and creatinine remained unchanged at 1 month following anti-VEGF compared with baseline. The mean eGFR significantly reduced from 73.6 mL/min/1.73 m2 at baseline to 71.6 mL/min/1.73 m2 at 1 month (P = .04, 95% CI –3.8, –0.1).
A subgroup analysis of renal function in different anti-VEGF groups (ranibizumab versus aflibercept) showed that mean serum electrolyte and urea concentrations at 1 month compared with baseline remained stable. There was no clinically or statistically significant change in serum creatinine at 1 month compared with baseline for aflibercept (72.4–75.0 µmol/L; P = .11) or ranibizumab (83.1–82.5 µmol/L; P = .85) participants.
There was no significant change over time in mean serum electrolytes or urea in either renal function group (baseline eGFR >60 or <60 mL/min/1.73 m2). There was a statistically significant but trivial increase in mean serum creatinine concentration in participants with baseline eGFR ≥60 mL/min/1.73 m2, from 67.9 µmol/L at baseline to 71.5 µmol/L at 1 month (P < .01, 95% CI 1.6, 5.5), with a corresponding decrease in eGFR from 81.0 mL/min/1.73 m2 to 77.6 mL/min/1.73 m2 (P < .01, 95% CI 1.6, 5.0). Participants with CKD 3–5 had no significant change in mean serum creatinine concentration, measuring 117.8 µmol/L at baseline and 108.1 µmol/L at 1 month (P = .22, 95% CI –11.2, 30.5).
Our results offer some reassurance that intravitreal anti-VEGF for nAMD does not significantly affect the renal function over the course of 1 month. However, we recognize the increasing number of published case reports and series, often with confirmatory renal biopsies, showing worsening hypertension and proteinuria, and new or worsening glomerular disease following intravitreal anti-VEGF treatment [3, 4, 6, 7]. It is possible, therefore, that cumulative anti-VEGF treatment over many months could result in renal dysfunction, as has been suggested by other groups [4].
Based on our study, it is uncertain whether intravitreal administration of anti-VEGF directly influences renal function. However, there are structural differences between the anti-VEGF drugs that may make renal dysfunction more likely with aflibercept. Drugs with an Fc domain (e.g. aflibercept) undergo slower systemic elimination than those without (e.g. ranibizumab), due to their binding affinity for endothelial cells, and therefore they may be more likely to cause VEGF suppression and secondary nephropathy. Studies in monkeys have shown that aflibercept binds to renal glomeruli, reducing the number of endothelial fenestrations [8]. No similar binding was observed with ranibizumab, which might suggest ranibizumab confers a lower potential risk of nephrotoxicity and is safer for at-risk groups. Studies of renal function on larger numbers of patients receiving anti-VEGF agents could further investigate this.
Our study has some limitations. We did not measure blood pressure or urine protein content, which are also mechanistically likely to change with sustained anti-VEGF treatment. Also, the study could have been carried out over a longer duration of treatment to assess the effect of VEGF suppression over time. In addition, a larger sample size would make it more possible to detect subtle changes in renal function. A strength of this study is its prospective design, compared with the retrospective nature of other studies, in which the timing of renal function measurement after anti-VEGF was highly variable [9]. Furthermore, our work was carried out on patients with nAMD, which is the most common disease indication for intravitreal anti-VEGF treatment. These patients are less likely to have confounding effects from co-existing renal disease than those with DMO.
In conclusion, we did not find any clinically significant change in serum markers of renal function during the 1 month period after intravitreal ranibizumab or aflibercept. Further studies on larger cohorts are warranted to investigate renal function after intravitreal anti-VEGF, ideally over a long time period, to assess the risk of chronic VEGF suppression. Such studies could help identify at-risk groups, and offer more bespoke anti-VEGF drug and dosing selection.
ACKNOWLEDGEMENTS
The authors would like to thank the patients who participated in the study. In addition, we thank the Clinical Research Faculty at King's College Hospital for their processing of the blood samples and both Tracey Mare and James Luxton at Synnovis (King's College Hospital) for performing the assays.
FUNDING
None.
CONFLICT OF INTEREST STATEMENT
T.L.J. and J.E.N.’s employer recruits to commercial clinical trials sponsored by Bayer and Novartis, for which the employer receives site payments. T.L.J. leads, and J.E.N. supports an AMD trial for which Zeiss/Oraya provide free use of equipment. J.E.N. is an advisor to Solvemed and has received lecture fees from Novartis. T.L.J. is an advisor to Ophthea, LumiThera, 2CTech, Regeneron and iLumen. T.L.J. and V.K.’s employer has received a free loan of equipment from LKC Technologies Inc. for research purposes. C.C.S. has received consulting fees from Novartis Pharmaceuticals, Travere Pharmaceuticals and Vifor. None of the above disclosures directly relate to this study.
The results presented in this paper have not been published previously in whole or part, except in abstract format. The results have been included in an academic thesis that has been submitted at King's College London.
AUTHORS’ CONTRIBUTIONS
All co-authors contributed sufficiently in the work to take public responsibility for the content. All co-authors participated in the conception or design, and/or analysis and interpretation of data. All co-authors either drafted the article or revised it. All co-authors provided intellectual content of critical importance to the work described and give final approval of the version to be published. Data available on request.
Comments