Tear protein biomarkers for pain after refractive surgery: candidate discovery

Abstract

Purpose : To identify tear proteins that are differentially expressed in subjects reporting ocular pain 3 months after refractive surgery.

Methods : Subjects (n=103) were recruited from two clinic sites: Portland, OR and Miami, FL. Three months after refractive surgery, tears were collected with anesthetized Schirmer strips. Subjects completed questionnaires on dry eye and rated their ocular pain on a numeric rating scale (NRS) where 0 = no pain and 10 = worst pain imaginable. Other information collected included age, sex, contact lens (CL) use, and Schirmer strip wet length. Proteomic analyses were conducted on subsets of subjects with ocular pain versus subjects with no ocular pain at 3 months after surgery. Proteins were extracted from strips and digested with trypsin using an S-Trap protocol. 18-plex Tandem Mass Tag (TMT) mass spectrometry analyses were done with pooled controls within each run for normalization. Open-source tools for peptide identification and summed reporter ion intensities were used to measure relative protein abundances. Statistical analyses were performed to identify differential abundance candidates between groups (“Pain” vs. “No Pain”). Log linear regression models with 1 additional covariate (age, sex, site, procedure, CL use, or Schirmer length) were used to investigate potential confounds from other variables.

Results : Sixteen “Pain” subjects reported an NRS score ≥ 3 at 3 months and no ocular pain (NRS ≤1) prior to surgery; these were compared to 32 “No Pain” subjects reporting no pain at baseline or 3 months (NRS scores ≤1). Age, sex, Schirmer strip wet length, site, and CL use were not different between the groups. TMT analyses identified 31 tear proteins increased in “Pain” compared to “No Pain” and 6 proteins decreased in “Pain” (≥ 1.5 fold change; p < 0.05). Heat maps of protein fold change showed similar patterns for the main group effect and no differences with each additional covariate, showing a robust main effect of “Pain” vs. “No Pain” on tear protein changes.

Conclusions : Proteomic analyses revealed proteins associated with subjects reporting pain at 3 months after refractive surgery. Proteins identified did not differ when analyzed with other individual features as covariates. These data show potential for detection of tear protein biomarkers for ocular pain, and patterns of concurrent protein changes (some increased, others decreased) that may strengthen the predictive value.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.