SEARCH

SEARCH BY CITATION

Keywords:

  • fellow eye;
  • intravitreal injection;
  • unbound bevacizumab

Abstract.

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Purpose:  To determine the concentration of unbound bevacizumab in untreated fellow eyes after contralateral intravitreal injection of bevacizumab.

Methods:  A total of 18 eyes received intravitreal injections of 1.5 mg bevacizumab. Nine probes were obtained in the injected eye and nine in the fellow eye. Each group contained three individual eyes. Aqueous humour samples were obtained during uneventful phacoemulsification at three intervals 1–7 days (group a), 8–12 days (group b) or 13–28 days (group c).

Results:  In untreated fellow eyes, the concentration of unbound bevacizumab was below the detectable limit of the ELISA (5 ng/ml in all samples). The mean concentration of unbound bevacizumab in the injected eye declined from 28.6 μg/ml (group a), 16.5 μg/ml (group b) to 7.4 μg/ml (group c).

Conclusions:  There are no pharmacological indications for a significant concentration of unbound bevacizumab in the anterior chamber of contralateral eyes in humans.


Introduction

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Bevacizumab (Avastin; Genentech, South San Francisco, CA, USA), a recombinant humanized monoclonal immunoglobulin antibody directed against human vascular endothelial growth factor (VEGF), was approved as a treatment for colorectal cancer. Intravitreal injections of bevacizumab demonstrated beneficial results in patients with macular oedema secondary to neovascular age-related macular degeneration (AMD) and diabetic macular oedema (DME) (Carneiro et al. 2008). Interestingly, unexpected treatment effects have sporadically been reported in untreated fellow eyes, including decreased leakage in fluorescein angiography after intravitreal injection of 1.25 mg in patients with DME (Avery et al. 2006).

Anterior chamber taps to measure the potential involvement of secreted proteins or therapeutic drugs for pharmacokinetic or pharmacodynamic analysis are feasible, according to Campochiaro et al. (2009), whereas vitreous taps are risky and are usually obtained in patients undergoing vitrectomy. To date, the concentration of unbound bevacizumab has not been measured in untreated fellow eyes of humans. However, Sawada et al. (2008) previously treated five patients with DME with an intravitreal injection of 1.25 mg bevacizumab and measured the concentration of VEGF, by enzyme-linked immunosorbent assay (ELISA) in the aqueous humour of fellow eyes. The mean VEGF concentration in the injected eyes was 302 ± 100 pg/ml before the intravitreal bevacizumab injection. One week later, the VEGF concentration declined in all injected eyes to <31 pg/ml (the lower limit of the ELISA) (p < 0.001), while no reduction was detected in the uninjected fellow eyes (382 ± 119 pg/ml). Animal studies by Bakri et al. (2007) indicated a very low concentration of bevacizumab in uninjected fellow eyes with a peak between the fifth and 10th day after surgery.

Recently, we investigated the pharmacokinetics of unbound bevacizumab in the anterior chamber after a single intravitreal injection of 1.5 mg in human eyes (Krohne et al. 2008). The concentration of unbound bevacizumab in the aqueous humour peaked on the first day after the injection with a mean concentration of 33.3 μg/ml and subsequent declined in a monoexponential fashion with a half-time of 9.82 days. With the ELISA employed in our study, bevacizumab in injected eyes was readily detectable even 53 days following injection (our last time-point). The purpose of the present study was to determine the concentration of unbound bevacizumab in untreated fellow eyes after a single intravitreal injection in the contralateral eye and compare this with the concentration in previously injected eyes in humans.

Materials and Methods

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Eighteen patients with DME and bilateral cataracts received an intravitreal injection of 1.5 mg bevacizumab in the first eye. An aqueous humour sample was obtained during elective cataract surgery at three intervals 1–7 days (group a), 8–12 days (group b) or 13–28 days (group c). Each group contained three individual eyes. Nine probes were obtained from uninjected fellow eyes and nine probes from previously injected eyes. The concentration of unbound bevacizumab was measured by ELISA as previously described by (Krohne et al. 2008) with a slight modification for the fellow eye samples. While the probes of injected eyes were diluted 1:100, we measured the samples of the fellow eyes without previous dilution to enhance the chance of detecting even smallest concentrations of bevacizumab in the untreated fellow eyes.

Results

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

The concentration of unbound bevacizumab in the fellow eyes was below the detectable limit of the ELISA test of 5 ng/ml at any time-point in all three groups. The concentration of unbound bevacizumab in the treated eyes was readily detectable in the same range according to our previously published results. A peak concentration was measured initial after the injection (group a) followed by a constantly decline over time (group b and c) (Table 1).

Table 1.   Mean concentration of unbound bevacizumab.
Time-point after injectionMean concentration of unbound bevacizumab
Uninjected fellow eyes, ng/mlInjected eyes, μg/ml
  1. * Below detection limit.

Day 1–7<5*28.6
Day 8–12<5*16.5
Day 13–28<5*7.4

Discussion

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Previous case reports in humans observed clinical effects of bevacizumab in untreated fellow eyes (Shima et al. 2008, Al-Dhibi & Khan 2009, Wu & Sadda 2008). Neri et al. (2008) reported the contralateral vitritis following intravitreal bevacizumab for a choroidal neovascularization secondary to myopia. Yoon et al. (2009) described the rapid progression of subclinical AMD in an untreated fellow eye after intravitreal bevacizumab. We observed after five consecutive bevacizumab injections the tearing of the retinal pigment epithelium in untreated fellow eyes (Mennel et al. 2007).

However, Velez-Montoya et al. (2009) investigated the systemic effects of unilateral intravitreal bevacizumab administration and reported no differences between baseline and follow-up measurements in untreated fellow eyes. Twenty-three consecutive patients with bilateral diffuse DME and a central retinal thickness >275 μm on optical coherence tomography were treated with 2.5 mg bevacizumab intravitreally in the worst eye and observed every 2 weeks for 4 weeks. There were no statistically significant differences between baseline and follow-up measurements in untreated fellow eyes. Visual acuity on early treatment diabetic retinopathy study -letters was 34.46 ± 17.29 at baseline and 38.31 ± 14.64 at 2 weeks, and 37.38 ± 14.59 at 4 weeks. The central retinal thickness (CRT) was 324.77 ± 76.51 μm at baseline, 319 ± 75.7 μm at 2 weeks and 315.54 ± 78.2 μm at 4 weeks. The macular volume was 8.99 ± 1.2 mm3 at baseline, 9.16 ± 1.26 mm3 at 2 weeks and 8.99 ± 1.09 mm3 at 4 weeks. The authors concluded that the systemic effect in untreated fellow eyes after contralateral intravitreal injection of 2.5 mg bevacizumab seemed to be unlikely, because of the lack of significant changes in any parameter. These clinical findings, reporting stable values for functional and anatomical parameters, were favourably supported by our study.

The testing of aqueous samples in ocular drug trials were recommended by Campochiaro et al. to interpret the therapeutic effects of drugs and, most importantly to monitor drug levels after insertion. This approach may give important new insights into pharmacotherapeutics in the eye (Campochiaro et al. 2009). Our study obtained an aqueous humour sample from untreated fellow eyes after contralateral VEGF injections in humans. For sample collection, we selected a time window of 1–28 days after the injection, following previous reports that concentrations of unbound bevacizumab in aqueous of fellow eyes peaked 7 days after contralateral injection in the rabbit model (Bakri et al. 2007) and that serum concentration peaked 5–7 days after injection in humans (Nomoto et al. 2009). In all our samples from uninjected eyes, we determined no measurable unbound bevacizumab concentration. These results are in agreement with the above-mentioned studies by Sawada et al. (2008) or Velez-Montoya et al. (2009) demonstrating no detectable reduction of VEGF-levels in the anterior chamber or CRT as well as unchanged visual acuity values in fellow eyes following contralateral intravitreal bevacizumab injection.

Two studies, examining the pharmacokinetics of unbound bevacizumab in uninjected fellow eyes in animals, measured the intraocular concentration after an intravitreal injection of 1.25 mg bevacizumab. Both detected, in the aqueous humour, a peak at day 7 following the contralateral injection and reported concentrations of 29.4 and 5.4 ng/ml in the uninjected fellow eyes in rabbits, respectively (Bakri et al. 2007; Nomoto et al. 2009). In contrast, maximum concentrations in the untreated vitreous (11.2 ng/ml) were measured significantly later, at 4 weeks after the contralateral injection (Bakri et al. 2007). This finding let the authors to conclude that bevacizumab may predominantly enter the uninjected eye from the circulation into the anterior chamber, diffusing into the vitreous cavity. The studies reported peak serum concentrations in rabbits of 3.3 and 2.1 μg/ml (Bakri et al. 2007; Nomoto et al. 2009), respectively, while maximum concentrations of 59.8–86.5 ng/ml were measured in humans following intravitreal injection of the same dosage of the drug (Ziemssen et al. 2009). This approximate 40-fold difference corresponds remarkably well with the about 40-fold difference in systemic distribution volume (total blood volume) between rabbits (approximately 120 ml at 2 kg body weight as used in the studies) and humans (approximately 5.0 l). Assuming that aqueous concentrations in uninjected eyes correlates with the significantly lower serum concentrations in humans, our finding that aqueous bevacizumab levels in uninjected eyes in humans are below 5 ng/ml is in accordance with the concentrations measured in the rabbit model.

Our concentration in fellow eyes (<5 ng/ml) is <5000-fold lower than the peak concentration in the injected eyes. While our study indicates that unbound bevacizumab is not present in biologically relevant concentrations (Wang et al. 2004) in the aqueous of uninjected fellow eyes in humans, those concentrations may nevertheless be reached in the serum. A theoretical very small dose of unbound bevacizumab in the circulation of the systematic blood may also diffuse in the contralateral eye possibly via leaking retinal/choroidal vessels and bind on the upregulated VEGF in adjacent retinal structures. Following this assumption we may be able to measure no unbound bevacizumab but only bound bevacizumab in the vitreous cavity or anterior chamber. The small amount of unbound drug can bind on VEGF in the vitreous cavity and may become undetectable in the obtained aqueous samples. However, a test for bound bevacizumab has not been developed or used previously in any pharmacokinetic study in the eye, and numerous studies recommended aqueous sampling to be feasible approach for pharmacokinetic or pharmacodynamic analysis (Campochiaro et al. 2009).

We are aware of the limitations that apply to our study: undoubtedly, with the small sample size and the limited sampling time-points in our study, we may have missed detectable unbound bevacizumab concentrations because of patient variability or inappropriate timing of analysis. A more sensitive ELISA with the capability of detecting pg/ml concentrations might well have been able to demonstrate bevacizumab in our samples, although the clinical relevance of such low concentrations seems questionable. Our results cannot exclude the possibility that serum-derived bevacizumab exerts therapeutic effects in choroidal neovacularization-membranes and other well-perfused lesions without reaching the aqueous or vitreous in significant amounts.

Conclusion

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Although some clinical studies in humans indicated a possible limited effect of intravitreal bevacizumab injections in untreated fellow eyes, there are no pharmacological indications for a significant concentration of unbound bevacizumab with relevant biological activity in the anterior chamber of human eyes.

Acknowledgements

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Carsten H. Mayer and Tim U. Krohne contributed equally to the manuscript.

References

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  • Al-Dhibi H & Khan AO (2009): Bilateral response following unilateral intravitreal bevacizumab injection in a child with uveitic cystoid macular edema. J AAPOS 13: 400402.
  • Avery RL, Pearlman J, Pieramici D et al. (2006): Intravitreal bevacizumab (Avastin) in the treatment of proliferative diabetic retinopathy. Ophthalmology 113: 16951705.
  • Bakri SJ, Snyder MR, Reid JM, Pulido JS & Singh RJ (2007): Pharmacokinetics of intravitreal bevacizumab (Avastin). Ophthalmology 114: 855859.
  • Campochiaro PA, Choy DF, Do DV, Hafiz G, Shah SM, Nguyen QD, Rubio R & Arron JR (2009): Monitoring ocular drug therapy by analysis of aqueous samples. Ophthalmology 116: 21582164.
  • Carneiro A, Falcão M, Azevedo I, Falcão Reis F & Soares R (2008): Multiple effects of bevacizumab in angiogenesis: implications for its use in age-related macular degeneration. Acta Ophthalmol 87: 517523.
  • Krohne TU, Eter N, Holz FG & Meyer CH (2008): Intraocular pharmacokinetics of bevacizumab after a single intravitreal injection in humans. Am J Ophthalmol 146: 508512.
  • Mennel S, Callizo J, Schmidt JC & Meyer CH (2007): Acute retinal pigment epithelial tear in the untreated fellow eye following repeated bevacizumab (Avastin) injections. Acta Ophthalmol Scand 85: 689691.
  • Neri P, Mariotti C, Mercanti L, Salvolini S & Giovannini A (2008): Vitritis in the contralateral uninjected eye following intravitreal bevacizumab (Avastin). Int Ophthalmol 28: 425427.
  • Nomoto H, Shiraga F, Kuno N et al. (2009): Pharmacokinetics of bevacizumab after topical, subconjunctival and intravitreal administration in rabbits. Invest Ophthalmol Vis Sci 50: 48074813. (Accepted).
  • Sawada O, Kawamura H, Kakinoki M & Ohji M (2008): Vascular endothelial growth factor in fellow eyes of eyes injected with intravitreal bevacizumab. Graefes Arch Clin Exp Ophthalmol 246: 13791381.
  • Shima C, Sakaguchi H, Gomi F et al. (2008): Complications in patients after intravitreal injection of bevacizumab. Acta Ophthalmol 86: 372376.
  • Velez-Montoya R, Fromow-Guerra J, Burgos O, Landers MB III, Morales-Catón V & Quiroz-Mercado H (2009): The effect of unilateral intravitreal bevacizumab (avastin), in the treatment of diffuse bilateral diabetic macular edema: a pilot study. Retina 29: 2026.
  • Wang Y, Fei D, Vanderlaan M & Song A (2004): Biological activity of bevacizumab, a humanized anti-VEGF antibody in vitro. Angiogenesis 7: 335345.
  • Wu Z & Sadda SR (2008): Effects on the contralateral eye after intravitreal bevacizumab and ranibizumab injections: a case report. Ann Acad Med Singapore 37: 591593.
  • Yoon YH, Kim JG, Chung H & Lee SY (2009): Rapid progression of subclinical age-related macular degeneration in the untreated fellow eye after intravitreal bevacizumab. Acta Ophthalmol 87: 685687.
  • Ziemssen F, Zhu Q, Peters S et al. (2009). Intensified monitoring of circadian blood pressure and heart rate before and after intravitreous injection of bevacizumab: preliminary findings of a pilot study. Int Ophthalmol 29: 213224.