TL;DR
This study identifies two proteins in human tears, S100A4 and PIP, that show consistent changes in levels correlating with the severity of Thyroid Eye Disease (TED). These proteins could serve as valuable, non-invasive biomarkers to predict the progression of TED in patients with autoimmune thyroid disease.
What Problem Does This Paper Address?
Thyroid Eye Disease (TED) is a serious complication of autoimmune thyroid disease (AITD) that can lead to vision loss. Currently, there are no reliable biomarkers to predict which patients will develop severe TED, making early intervention and closer monitoring challenging. The lacrimal gland, which produces tears, is implicated in TED, suggesting that tear composition might offer clues to disease progression. This paper aims to identify specific tear proteins that change with TED severity, potentially serving as predictive biomarkers.
Methods
The researchers employed a two-phase quantitative proteomics approach.
- Discovery Phase: Isobaric tags for relative and absolute quantification (iTRAQ) was used to analyze tear samples from four groups: normal controls, AITD without TED, mild TED, and severe TED. This method allows for the relative quantification of thousands of proteins simultaneously.
- Verification Phase: To confirm the findings from the discovery phase, Sequential Window Acquisition of All Theoretical Fragment Ion Spectra (SWATH-MS) was used on an independent cohort of patients. SWATH-MS is a more targeted and sensitive technique for validating protein changes.
Tear samples were collected using Schirmer's strips, and proteins were eluted. For iTRAQ, samples were labeled and analyzed using nanoLC-MS/MS. For SWATH, a comprehensive spectral library was built from pooled healthy tear samples, and then the independent cohort was analyzed using SWATH acquisition. Statistical analysis, including Mann-Whitney U tests and Kruskal-Wallis tests, was used to compare protein levels between groups.
Datasets
The study utilized tear samples collected from human participants.
- Discovery Cohort: 8 participants per group (normal controls, AITD without TED, mild TED, severe TED), totaling 32 samples.
- Verification Cohort: 10 participants per group (normal controls, AITD without TED, mild TED, severe TED), totaling 40 samples.
- Spectral Library: A pooled sample from 1000 healthy individuals was used to build a comprehensive human tear spectral library for SWATH analysis.
No new benchmark datasets were introduced; the study focused on analyzing existing biological samples.
Key Insights and Ideas
- Tears as a Biomarker Source: The study reinforces the idea that tears, being easily accessible, can reflect the physiological state of the eye and potentially systemic conditions like TED.
- S100A4 and PIP as Key Players: The consistent dysregulation of S100A4 (calcium-binding protein A4) and PIP (Prolactin-Induced Protein) across both discovery and verification phases is a significant finding.
- Opposing Trends: Interestingly, S100A4 levels showed a downward trend with increasing TED severity, while PIP levels showed an upward trend. This suggests they might reflect different aspects of the disease process.
- Potential for Prediction: The consistent trends observed in independent cohorts and using different proteomic techniques lend strong support to the potential of S100A4 and PIP as predictive biomarkers for TED progression.
Key Results
- In the discovery phase, thousands of tear proteins were identified.
- The study identified S100A4 as being down-regulated in severe TED compared to controls (p=0.029).
- PIP was identified as being up-regulated in severe TED compared to controls (p=0.023).
- These trends for S100A4 and PIP were confirmed in the verification phase, with S100A4 showing a significant downward trend (p=0.0022 for mild TED, p=0.0361 for severe TED) and PIP showing a significant upward trend (p=0.0131 for severe TED).
- Table 3 summarizes the fold changes and p-values for S100A4 and PIP, highlighting their statistically significant changes in relation to TED severity.
Limitations and Future Work
- Small Sample Size: The authors acknowledge that the small sample size is a limitation, particularly due to the stringent exclusion criteria for patients (e.g., no prior treatment with steroids or immunosuppressants). This limited the number of eligible participants.
- Confounding Factors: While efforts were made to exclude patients with other medical conditions, the authors note that it's difficult to be completely certain that other medical conditions or medications wouldn't influence S100A4 and PIP levels.
- Future Work: The authors suggest that further prospective evaluation on larger cohorts of patients is necessary to confirm these findings and validate S100A4 and PIP as reliable biomarkers for predicting TED progression. They also propose that S100A4 might indicate limbal normality, while PIP could reflect immune tolerance breakdown and fibrosis in severe TED, suggesting they may complement each other as biomarkers.
Summary updated