Fig 1: ROC curve analysis for the diagnosis of IPF versus controls. (A) Unadjusted ROC curves for ROR2, WNT1, WNT5A, and WNT7A based on single biomarker measurements. (B) Age-adjusted ROC curves from logistic regression models including age. Cut-off values represent predicted probabilities. (C) Summary of AUC, cut-off, 95% CI, sensitivity, specificity, and p-values. For age-adjusted analyses, cut-off values represent predicted probabilities. Abbreviations: AUC: area under the curve; CI: confidence interval; p-values ≥ 0.05 are marked as “(NS)” to indicate non-significance.
Fig 2: Exploratory comparison of 10-year mortality based on combined data-derived thresholds of ROR2 and WNT5A concentrations in BAL fluid. Patients were categorized into four groups according to whether their ROR2 (5.59 ng/mL) and WNT5A (6 ng/mL) concentrations were above or below exploratory, data-derived thresholds. Cumulative mortality curves were used to explore differences in 10-year mortality across the four groups. Numbers at risk shown below the curves represent patients who remained under observation without experiencing the event (death) at each time point. Hazard ratios (HR) with 95% confidence intervals (CI) and corresponding p-values were obtained using unadjusted Cox proportional hazards regression analysis, applying the same exploratory thresholds used for group stratification. These thresholds are preliminary and not intended as validated clinical cut-off values.
Fig 3: ROR2 and Wnt protein concentrations in BAL fluids. (A-D) ROR2, WNT1, WNT5A, and WNT7A protein levels were measured by ELISA in BAL fluids from patients with IPF (n = 124) and controls (n = 17). Values below the manufacturer’s lower limit of detection (LOD) were assigned a value of 0 ng/mL (or pg/mL) for statistical analysis, as they were below the quantifiable range. The LODs were 0.05 ng/mL for ROR2, 0.01 ng/mL for WNT1, 0.07 ng/mL for WNT5A, and 5.2 pg/mL for WNT7A. Detection rates were as follows: ROR2, 109/124 (87.9%) in IPF and 1/17 (5.9%) in controls; WNT1, 100/124 (80.6%) in IPF and 15/17 (88.2%) in controls; WNT5A, 89/124 (71.8%) in IPF and 9/17 (52.9%) in controls; WNT7A, 82/124 (66.1%) in IPF and 8/17 (47.1%) in controls. Data are presented as medians with interquartile ranges (25th and 75th percentiles). Statistical differences were assessed using the Mann–Whitney U test. *p < 0.05, **p < 0.01.
Fig 4: Exploratory comparison of 10-year mortality according to data-derived thresholds of ROR2 and WNT concentrations in BAL fluid from patients with IPF. (A) Mortality comparison based on the exploratory ROR2 threshold (5.59 ng/mL). (B) Mortality comparison based on the exploratory WNT1 threshold (0.01 ng/mL). (C) Mortality comparison based on the exploratory WNT5A threshold (6 ng/mL). (D) Mortality comparison based on the exploratory WNT7A threshold (20.97 pg/mL). Thresholds represent exploratory, data-derived values, not validated clinical cut-offs. Mortality rates were compared between groups above and below each threshold. Number at risk below the cumulative mortality curves represents the number of patients who remained under observation without experiencing the event (death) at each time point. P-values shown on the plots correspond to comparisons between groups, and hazard ratios (HR) with 95% confidence intervals (CI) are from Cox regression analyses presented in Table 2.
Supplier Page from Bioss Inc. for Human Wingless Type MMTV Integration Site Family, Member 5A (WNT5A) ELISA Kit