Dear Editor

We have read Manzano et al’s insightful paper[1] with interest and appreciated the comments concerning the inhibitory effects of bevacizumab on corneal neovascularisation (NV). However, the manuscript had some important aspects that need to be clarified. Cumulative data from numerous basic and clinical studies strongly implicates the central role of VEGF in ocular and corneal angiogenesis. Binding all free-circulating VEGF by specific antibodies inhibits VEGF-mediated receptor stimulation which normally leads to an abnormal angiogenesis in pathological conditions.[2,3] Bevacizumab (Avastin) is a humanised version of a monoclonal parent antibody originally produced in mice specifically against human VEGF. Since the parent antibody of avastin is a murine protein likely to provoke an immune response and thus unsuitable for use in humans, it has been humanised by genetic engineering to overcome these disadvantages. The humanisation process turns the parent antibody into commercially available avastin molecule, 93% human and 7% murine. However, the specificity of avastin molecule is exactly similar to that of the parent antibody, which binds to human VEGF with a high affinity, but has “no effect on host VEGF” (i.e., that produced by the mouse).[4] This is why preclinical anti-tumor studies of both avastin and its parent antibody have been conducted on human xenografts model in animals, and complete inhibition of tumor growth was not seen (and would not be expected) due to the interfering effects of host VEGF, which is not inhibited by the anti-human VEGF antibody, avastin.[5] Our previous case report [6] demonstrated that bevacizumab, a humanized anti-VEGF antibody, may actually have a therapeutic option for corneal NV in human beings. However, bevacizumab should not reasonably be considered to serve as an antiangiogenic agent in rat corneal tissue for the inherited characteristics of this molecule explained in detail above. Furthermore, we have previously tested this hypothesis in a similar (unpublished) pilot study of silver nitrate induced corneal NV in rats, but on the contrary found no improvement on the degree of corneal NV after intraperitoneal injection of bevacizumab 5 mg/kg, equivalent to its effective intravenous dose applied for the treatment of human colorectal cancers. Therefore, it is problematic to select such a monoclonal anti-human VEGF antibody, as an anti-rat VEGF agent in any experimental study conducted in rats. So, we unfortunately consider that this is a major methodological error in this experimental study.

The possible explanation for the difference encountered in the extent of corneal NV after topically applied avastin in rat eyes could be an unexpected cross-reaction of bevacizumab molecule with rat VEGF. We think that cross-reactivity of human specific monoclonal recombinant antibodies (if such exists), particularly bevacizumab, with other murine subspecies is another interesting topic of extensive researches. Another, but a minor concern in this paper is about the dose of topically applied avastin, yet not explained by the authors how to determine 2 drops of avastin 4mg/ml (i.e, approximately corresponding to 0.4 mg/day) solution daily would be sufficient for inhibiting corneal NV, since its minimal intravitreal dose is 1.25 mg in clinical practice, and also bioavailability of topically applied avastin drops in the cauterized corneal tissue has not been established. Finally, we obviously support the suggestion of using anti-human VEGF agents for prevention and management of corneal NV in the clinical practice [6], yet consider this experimental research has a major methodological error.

References

1. Manzano RP, Peyman GA, Khan P, et al. Inhibition of experimental corneal neovascularisation by bevacizumab (Avastin). Br J Ophthalmol. 2007;91:804-7.

2. Bhisitkul RB.Vascular endothelial growth factor biology: clinical implications for ocular treatments. Br J Ophthalmol. 2006;90:1542-7.

3. Philipp W, Speicher L, Humpel C. Expression of vascular endothelial growth factor and its receptors in inflamed and vascularized human corneas. Invest Ophthalmol Vis Sci 2000;41: 2514–22.

4. Presta LG, Chen H, O’Connor SJ, et al. Humanization of an anti-VEGF monoclonal antibody for the therapy of solid tumors and other disorders. Cancer Res 1997;57:4593–9.

5. Gerber H, Kowalski J, Sherman D, et al. Complete inhibition of rhabdomyosarcoma xenograft growth and neovascularization requires blockade of both tumor and host vascular endothelial growth factor. Cancer Res 2000;60:6253-8.

6. Erdurmus M, Totan Y. Subconjunctival bevacizumab for corneal neovascularization. Graefes Arch Clin Exp Ophthalmol. 2007 Apr 26; [Epub ahead of print]

Dear Editor,

We read with interest the recently published paper by Manzano et al. titled ‘Inhibition of experimental corneal neovascularisation by bevacizumab (Avastin)’. The authors used topical eyedrop application of either saline or bevacizumab in a rat model of corneal neovascularization, with apparent moderate success. Furthermore, in their discussion, they mention that there was incomplete inhibition of the neovascularization and speculate several other cytokines or growth factors may be responsible. However, bevacizumab has previously been reported to be incapable of binding rodent VEGF [2]. The findings in this study suggest that either there is in fact a low affinity interaction with bevacizumab and rat VEGF, or the effects observed are due to other non-VEGF mediated interactions of bevacizumab. If the latter is the case, then the question arises as to whether this interaction is specific to bevacizumab or can be replicated with other type specific normal IgG. Recently Gerber et al. reported the generation of transgenic mice producing humanized VEGF-A [1], which produce a VEGF protein that, in contrast to wild type mouse protein, does bind bevacizumab. This mouse would be suitable for evaluating human specific anti-VEGF therapies in vivo. We anticipate Lucentis will be evaluated in this new model and await the results eagerly.

References

1. Gerber, H.P., Wu, X., Yu, L., Wiesmann, C., Liang, X.H., Lee, C.V., Fuh, G., Olsson, C., Damico, L., Xie, D., Meng, Y.G., Gutierrez, J., Corpuz, R., Li, B., Hall, L., Rangell, L., Ferrando, R., Lowman, H., Peale, F., and Ferrara, N., Mice expressing a humanized form of VEGF-A may provide insights into the safety and efficacy of anti-VEGF antibodies. Proc Natl Acad Sci U S A, 2007. 104(9): p. 3478-83.

2. Lin, Y.S., Nguyen, C., Mendoza, J.L., Escandon, E., Fei, D., Meng, Y.G., and Modi, N.B., Preclinical pharmacokinetics, interspecies scaling, and tissue distribution of a humanized monoclonal antibody against vascular endothelial growth factor. J Pharmacol Exp Ther, 1999. 288(1): p. 371-8.