Refractive Shifts and Changes in Corneal Curvature Associated With Antibody-Drug Conjugates

Abstract

Purpose: Antibody-drug conjugates (ADCs) are a class of cancer drug wherein some are associated with corneal abnormalities, but there is a dearth of published information on refractive shifts in patients receiving ADCs. Here, we evaluated the dynamics of refractive error and keratometry readings in patients with ADC-related keratopathy and microcyst-like epithelial changes (MECs).

Methods: This study is a retrospective case series including 58 eyes of 29 patients with ADC-related keratopathy from a single tertiary care cancer referral center (MSKCC). One eye (29 total) was randomly assigned for statistical analysis. In addition, a subset analysis of MEC location-refractive error correlation was performed on 20 eyes. Clinical records including slitlamp examination, indirect ophthalmoscopy, calculated spherical equivalence (SE), keratometry, and visual acuity were recorded at baseline, during, and off treatment.

Results: A subset analysis of MEC location-refractive error correlation of 20 eyes revealed the following: Peripheral MECs were significantly associated with hyperopic shifts (P value < 0.001) and paracentral/central associated with myopic shifts (P value < 0.001). In the full cohort and on drug, the greatest change in SE from baseline was myopic (68%, as high as -4.75 D) and hyperopic (32%, as much as +3.75 D). Eighty-nine percent had a change in vision from baseline while on drug, but at the 3-month follow-up off drug, SE and vision returned to baseline in 33% and 82% of eyes.

Conclusions: Peripheral MECs were significantly associated with hyperopic shifts, and paracentral/central MECs were associated with myopic shifts. While on drug, most eyes had a myopic refractive shift, which corresponded with corneal steepening.

Figure 1:

Graphical representation of MEC location and it’s correlation with spherical equivalent. Each circle represents a follow-up visit (at various time points) for each eye and is plotted against the y-axis, which represents spherical equivalent at each visit. The black circles represent eyes with peripheral MECs and the white circles represent eyes with paracentral/central MEC location. Peripheral MECs were significantly associated with hyperopic shifts and paracentral/central MECs associated with myopic shifts.

Figure 2:

Spherical equivalent for each randomized right eye over time. The zero time point represents the start of drug and negative time points represent baseline visits prior to starting drug.

Figure 3:

Spherical equivalent for each randomized left eye over time. The zero time point represents the start of drug and negative time points represent baseline visits prior to starting drug.

Figure 4:

The greatest shift in spherical equivalent (Rx change) compared to baseline for the right eye (black column) and left eye (grey column), mapped alongside the greatest shift in corneal curvature (K change = keratometry) compared to baseline for the right eye (dark blue column) and left eye (light blue column) while on drug. The greatest decline in visual acuity at any time point on drug compared to baseline is represented by a black cross for the right eye and a black circle for the left eye and correspond to the y axis on the right-hand side. Each of these 3 data points is mapped per randomized eye in each patient, found along the x-axis.