Possibility of Müller Cell Dysfunction as the Pathogenesis of Paclitaxel Maculopathy

Introduction

The taxane class of anticancer drugs (eg, paclitaxel [Taxol], paclitaxel nanoparticle albumin bound [NAB] [Abraxane; Celgene, Summit, NJ], and docetaxel [Taxotere; Sanofi Aventis, Bridgewater, NJ]) is used to treat a wide range of malignancies including breast, lung, gastric, and prostate cancers.¹ They are mitotic inhibitors that prevent normal reorganization of the microtubule network within cells. CME without evidence of leakage on fluorescein angiography (FA) is a rare side effect of these medications.² However, the pathogenesis of CME without fluorescein leakage has been unclear. We report a case to suggest the possible pathogenesis of paclitaxel-induced CME considered from the data of the ERG.

Case Report

A 66-year-old man presented with a 1-month history of blurry vision in both eyes. The patient's medical history was positive for gastric cancer, which was diagnosed 3 years prior, and he was undergoing chemotherapy with intravenous infusion of polyethylated castor oil-based paclitaxel at a dosage of 80 mg/m² three times a month for 10 months (total: 30 treatments). His ocular history was positive for glaucoma. The patient had received latanoprost 0.005% (Xalatan; Pfizer, New York, NY), carteolol 2%, and brimonidine 0.2% topically. His family history was negative for X-linked retinoschisis, as well as retinitis pigmentosa. Best-corrected visual acuity (BCVA) was 20/25 in both eyes at the first examination. Anterior segment examination revealed no inflammation. Fundus examination showed bilateral CME with no apparent vascular abnormality (Figures 1A and B). Spectral-domain optical coherence tomography (SD-OCT) (Cirrus HD-OCT; Carl Zeiss Meditec, Dublin, CA) showed fluid accumulation into multiple well-defined, cyst-like spaces in the outer and inner plexiform layers. The cystic formations were large and centrally located in the outer plexiform layer, whereas in the inner plexiform layer they were relatively small and broadly distributed. Central macular thickness was 609 µm in the right eye and 583 µm in the left eye (Figures 1C, D, E, and F). FA failed to demonstrate CME or any fluorescence leakage (Figures 1G and H). The full-field ERG (Portable LE2000 ERG system; Tomey, Nagoya, Japan) showed the implicit time of cone B-wave significantly delayed and cone B-wave amplitude reduced. The B/A ratio of the single-flash ERG was decreased to 1.03 and 0.98 in the right and left eyes, respectively (Figure 2A). Niacin maculopathy, Goldmann-Favre syndrome, retinitis pigmentosa, and congenital X-linked retinoschisis, which showed non-leakage CME, were ruled out on the basis of history and clinical examinations. Therefore, the patient was diagnosed with bilateral CME secondary to paclitaxel treatment. Paclitaxel treatment was discontinued and irinotecan (Camptosar; Pfizer, New York, NY) was prescribed by the patient's oncologist. Two months after the cessation of paclitaxel treatment, the patient's BCVA improved to 20/20 in both eyes. Fundus examination findings and SD-OCT scans revealed resolution of CME (Figure 3). The single-flash ERG showed the improvement of B/A ratio (right eye: 1.26; left eye: 1.24). However, the implicit time and the amplitude of the cone B-wave were not changed (Figure 2B).

Figure 1. Fundus photos, spectral-domain optical coherence tomography (SD-OCT), and fluorescein angiography (FA) at the first examination. Fundus photos (A, B) and OCT (C, D) reveal cystoid macular edema (CME). CME presents in the inner and especially the outer plexiform layers of the retina. The central foveal thickness is 609 µm in the right eye (E) and 593 µm in the left eye (F). (G, H) Late-phase FA shows normal vascular filling and no leakage in both eyes.

Figure 2. Full-field electroretinogram (ERG) at the first visit and 2 months after cessation of paclitaxel. (A) The full-field ERG shows the implicit time of cone B-wave significantly delayed and cone B-wave amplitude reduced at the first examination. The B/A ratio of the single-flash ERG was decreased to 1.03 and 0.98 in the right and left eyes, respectively. (B) ERG shows the improvement of B/A ratio (right eye: 1.26; left eye: 1.24) 2 months after cessation of paclitaxel.

Figure 3. Spectral-domain optical coherence tomography (SD-OCT) of paclitaxel-related cystoid macular edema (CME). Initial SD-OCT examination showed CME in the both eyes. One, 2, and 3 months after paclitaxel cessation, central foveal thickness decreased gradually.

Discussion

Paclitaxel, a member of the taxane family of microtubule-stabilizing agents, has been used to treat various cancers. The ophthalmic adverse effects include decreased vision, scintillating scotomas, and abnormal visual evoked potentials.³ Bilateral CME is also a significant adverse event with paclitaxel treatment.^4–6 However, there is no specific treatment for paclitaxel-induced CME. Discontinuation of paclitaxel and possible change of the chemotherapeutic regimen are widely accepted as the first choice of treatment. It has been reported that most cases were in remission prior to discontinuation of the drug.^4–6 Our case also demonstrated complete resolution of CME with cessation of paclitaxel.

Paclitaxel treatment cannot be withdrawn in some cases because of the potential adverse effect on the patient's systemic disease. Therefore, understanding of the pathological mechanism is necessary. However, the pathogenesis of CME without fluorescein leakage remains unclear. As a possible pathogenic mechanism, the following four points have been reported: 1) Increased capillary fluid filtration could lead to capillary protein leakage and edema. The proposed pathogenesis for the fluid retention syndrome may explain the development of CME despite the lack of leakage by FA.⁷ 2) Expansion of the intracellular fluid space may also lead to CME. Accumulation of fluid in the intracellular space may lead to CME without evidence of leakage on fluorescein angiograms.⁸ 3) Joshi et al. proposed the mechanism that the toxicity to Müller cells with subsequent intracellular fluid accumulation and subclinical leakage of extracellular fluid.⁹ 4) Kuznetcova et al. described that CME probably originated from retinal pigment epithelium dysfunction by its effect on microtubules functions.¹⁰ In this case, ERG was recorded for the differential diagnosis of retinitis pigmentosa. The attenuation of the B/A ratio and the extension of the implicit time of B-wave could be observed. It is known that the B-wave reflects a function of Müller cells.¹¹ Furthermore, a previous report indicated that the full-field ERG did not show any significant difference in the B-wave amplitudes or implicit times in cases with diabetic macular edema,¹² suggesting that full-field ERG, unlike multifocal ERG, could not detect the abnormality in CME. Therefore, the decrease of the B/A ratio in the single-flash ERG could reflect a decrease in function of Müller cell rather than CME pathologies. Our results suggested that paclitaxel could cause certain toxicity to retinal Müller cells with subsequent intracellular fluid accumulation and subclinical leakage of extracellular fluid. Our observation might support a proposed mechanism by Joshi et al.⁹ Moreover, our ERG finding also suggests that paclitaxel could affect Müller cells in the whole retina in spite of localized edema in macula.

However, this is only a case report. Furthermore, ERG testing is known to be sometime unstable. Therefore, it is important to consider the additional number of cases future in spite of rare disease.