Exploring MEK inhibitor-associated subretinal fluid

By Laird Harrison

Vienna-Subretinal fluid caused by mitogen-activated protein kinase (MEK) inhibitors, a new class of cancer medications, typically resolves on its own without posing long-term risk to visual acuity, researchers say.

”It’s important to let people know that this is not an unusual occurrence,” Marissa Weber, MD, told Ophthalmology Times. “Fortunately, this does not appear to cause permanent vision loss. We would not recommend withholding a life-prolonging treatment because of the presence of subretinal fluid.”

Dr. Weber, a vitreoretinal surgeon in the U.S. Army at Walter Reed National Military Medical Center in Bethesda Maryland, presented the finding at the American Society of Retina Specialists 2015 annual meeting. She conducted the research during her fellowship in vitreoretinal surgery at Ophthalmic Consultants of Boston.

The MEK inhibitor that was studied was binimetinib, which is currently used in many clinical trials in combination with other agents. However, U.S. ophthalmologists are most likely to encounter trametinib, the only MEK inhibitor approved by the FDA. Among the others under study are cobimetinib and selumetinib.

“If ophthalmologists are seeing patients with a history of cancer being treated with chemotherapy, it would be important to delve deeper to find out what chemotherapy they’re on, because it might be one of these MEK inhibitors,” said Dr. Weber.

Previous research has shown MEK inhibitors to improve survival in people with metastatic cancer. Early trials reported the presence of subretinal fluid (SFR), which resembles that seen in central serious retinopathy, but at a much lower incidence than has been reported more recently. 

To determine the incidence and characteriistics of the SRF, Dr. Weber and her colleagues reviewed the records of 52 patients treated twice per day with oral binimetinib during phase 1B and II clinical trials.

These clinical trials included patients with advanced solid tumors such as colorectal, esophageal, non-small-cell lung, pancreatic and breast cancer, and cutaneous malignant melanoma. The researchers excluded those patients with symptomatic or recently treated brain metastasis and history of, or evidence of, central serous retinopathy or retinal vein occlusion.

Clinical examinations revealed discrete, yellow, elevated areas subfoveally and along the arcades in 46 subjects (88%). Optical coherence tomography (OCT) confirmed the presence of SRF. Fluorescein angiography did not show any signs of leakage, distinguishing these serous elevations from those of central serous retinopathy.

In 70% of the subjects, SRF appeared subfoveally and along the arcades. In 20%, SRF was along the arcades only.  And in 8% it was only found subfoveally.  

Nine of the subjects (17%) reported transient visual acuity symptoms, which they typically described as blurry vision, halos or colorful spots, said Dr. Weber. The patients said it took about 45 minutes to 2 hours for these symptoms to develop after taking the drug. Symptoms typically resolved within 4 hours, and stopped occurring altogether after a month. The investigators were able to follow 1 patients each for 9 and or 12 months. Both patients' vision was stable during that time.

In 6 eyes of 5 subjects with SRF,  visual acuity decreased by more than 1 line. However, visual acuity acuity of the fellow eyes in 4 of these subjects improved by 1-2 lines despite the presence of SRF. The changes in visual acuity were felt to be attributed to dry eye, transcription error, and the patient's systemic condition at the time of visual acuity measurement.

The researchers observed one patient using OCT in real time after he took binimetinib. Within 1 hour of dosing, SRF was noted to be present on OCT and increased with a maximum level of SRF at 3 hours. However, by 4 hours after dosing, the subretinal had resolved completely.

It’s not clear why the medication causes SRF to build up, said Dr. Weber. The medication may be toxic to the retinal pigment epithelium (RPE), affects aquaporin 1 channels in the RPE, or the effect could be mediated by auto-antibodies.

“All these theories point to a problem with the RPE, but honestly,  honestly we’re just not sure what it is yet,” she said.