Ophthalmological assessment of crizotinib in advanced non-small-cell lung cancer

Objectives: During crizotinib clinical evaluation, visual disturbances, generally of grade 1 severity, were frequently reported adverse events (AE). Consequently, ophthalmologic assessments were included in a patient subgroup enrolled in PROFILE 1001 (NCT00585195), a phase 1, open-label, single-arm trial of crizotinib in patients with advanced non-small-cell lung cancer and are reported here.

Materials and methods: At least 30 patients were required to undergo ophthalmologic assessments, including: best-corrected visual acuity (BCVA), refractive error, pupil size, slit-lamp anterior segment biomicroscopy, intraocular inflammation, intraocular pressure, retinal fundoscopic exams, fundus photography, ocular characteristics, and optical coherence tomography (OCT). Scheduled assessments included those at baseline, Cycle 1 Day 15, Cycle 3 Day 1 (C3D1), annually during treatment, and end of treatment (28 days after last crizotinib dose).

Results: Thirty-three patients completed all required ophthalmologic assessments through C3D1, and 22 (66.7 %) had abnormal findings on ≥1 ophthalmologic test. Clinically important changes were ≥2-line loss in BCVA in 10 patients (30.3 %), >±1.25-diopter change in refractive error in 3 patients (9.1 %), >±2-mm change pupillary diameter change in 3 patients (9.1 %), and >50 μm increase in OCT center point thickness in 7 patients (21.2 %). Three patients (15 %) reported clinically significant abnormalities in anterior segment biomicroscopy (grade 1 cataract [n = 2], grade 1 Visual Impairment [n = 1]). No permanent treatment discontinuations were associated with ophthalmologic findings changes. Twenty-four patients (72.7 %) reported ≥1 ocular all-causality treatment-emergent AE (TEAE); none required dose reduction or permanent discontinuation, but 2 required temporary dosing interruption. Although TEAEs and ophthalmologic findings may not have occurred concurrently, of 24 patients with ≥1 all-causality ocular TEAE, 18/24 (75.0 %) had ≥1 abnormal ophthalmologic finding and 6/24 (25 %) had none; and of 9 patients without an all-causality ocular TEAE, 4/9 (44.4 %) had ≥1 abnormal ophthalmologic finding and 5/9 (55.6 %) had none. Of the 18 patients with ≥1 abnormal ophthalmologic finding, 9 (50 %) had preexisting ocular conditions.

Conclusion: During crizotinib treatment, ophthalmologic changes from baseline did not appear to be associated with patient-reported ocular TEAEs. Abnormal ophthalmologic findings occurred in the context of preexisting conditions for a number of patients. No ophthalmologic changes from baseline or ocular all-causality TEAEs required permanent treatment discontinuation.