Cancer-associated retinopathy (CAR) is an autoimmune paraneoplastic syndrome that can affect the visual function of cancer patients. The immune system, triggered by specific proteins produced by a tumor, cross-reacts with retinal cells (specifically photoreceptors) leading to visual impairment. Lung cancer is a common underlying malignancy in patients diagnosed with CAR, and recoverin has been identified as a key antigen [1]. However, it can also be diagnosed in patients with other types of cancer, and various causative antigens have been reported [2]. This globally rare disease, with only 53 cases reported since 2010, is seldom reported in Korea [2,3].

Here, we report a case of CAR that occurred in a patient with a diagnosis of small cell lung cancer and esophageal cancer. This study was approved by the Institutional Review Board at Seoul National University Bundang Hospital (No. B-2508-993-701). Written informed consent for publication of the research details and clinical images was obtained from the patient.

A 72-year-old male patient with a history of esophageal cancer (pT1aNX, diagnosed in 2015), recently diagnosed small cell lung cancer (left lower lobe, 2.5 cm nodule, N3 metastases, diagnosed in 2025) presented to our clinic with a chief complaint of rapidly progressive vision loss and visual field defect in the both eyes. Upon presentation, his best-corrected visual acuity was hand motion in the right eye and 20 / 32 in the left eye. Intraocular pressure was 11 mmHg each. Fundus photography and fundus autofluorescence were unremarkable (Fig. 1A-1D), while fluorescein angiography revealed diffuse, mild late-phase leakage in both eyes (Fig. 1E, 1F). Both eyes showed complete visual field loss on the Humphrey 24-2 test. Nevertheless, the left eye retained a central visual acuity of 20 / 32, implying preservation of a small central island corresponding to residual photoreceptors on optical coherence tomography (OCT). OCT showed diffuse photoreceptor disruption in both eyes (Fig. 1G-1L). Based on the electroretinogram results, rod and cone responses in both eyes were nearly extinguished (Fig. 1M). The patient had no remarkable ophthalmic history other than having received multiple intravitreal bevacizumab injections a t a n outside clinic without a definitive diagnosis. There was no family history of inherited retinal diseases. Accordingly, the patient was diagnosed with CAR on both eyes.

To detect tumor-derived antiretinal autoantibodies in the blood, the patient’s serum samples were reacted with human and mouse retinal lysates as well as six human antigens, followed by protein band detection. No autoantibodies against recoverin were detected. However, autoantibodies against α-enolase, carbonic anhydrase II, cone-rod homeobox (CRX), heat shock protein 60, and aldolase C were identified. In particular, the autoantibody against CRX was relatively strong due to the high protein load (Fig. 1N).

He was treated with bilateral intravitreal dexamethasone implant (Ozurdex, AbbVie Inc). Also, the patient was subsequently started on a tapering course of oral corticosteroids. At the 2-week follow-up, the patient reported subjective visual improvement. His uncorrected visual acuity also improved to 20 / 200 in the right eye and 20 / 25 in the left eye.

The low incidence and variable clinical features of CAR make diagnosis challenging, even though it can precede a cancer diagnosis [4]. However, OCT often shows photoreceptor loss, and electroretinogram reveals a significant reduction in rod and cone cell function. Although various treatment options have been proposed, including steroids, rituximab, and plasmapheresis, no established treatment protocol exists. Recently, however, intraocular dexamethasone implant has been reported as an effective therapeutic strategy [5]. Likewise, in our case, despite the new macular hole in the right eye, the visual acuity improvement suggested the favorable treatment effect on the underlying CAR (Fig. 1K, 1L). The most well-known antiretinal autoantibody associated with CAR is anti-recoverin, which was not detected in our case. Furthermore, esophageal cancer as the underlying malignancy is highly atypical in previously reported cases.

The significance of our case, which involves a patient with two primary tumors including small cell lung cancer, is that a similar clinical presentation to typical anti-recoverin positive CAR can occur even in anti-recoverin negative CAR. Secondly, strong expression for the anti-CRX antibody suggests its potential role as the causative autoantibody in this case. Although dual malignancies may increase the complexity of CAR pathogenesis by expanding the antigen pool, the specific role of each tumor is undetermined. Further investigation is required to differentiate between a direct causal link and a coincidental association. Nevertheless, our findings underscore the importance of considering CAR in any cancer patient with unexplained vision loss, regardless of tumor type or autoantibody type.