The serum immune marker CXCL9 and its interaction panel improved stroke risk reclassification (NRI 0.22-0.32, p≤0.01) beyond traditional predictors in older adults.
Does a machine learning-selected serum immune marker panel improve prediction and risk reclassification of incident stroke compared to traditional risk factors in older adults?
A machine learning-derived panel of serum immune markers, particularly CXCL9, significantly improves the prediction and risk reclassification of incident stroke beyond traditional risk factors in older adults.
Absolute Event Rate: 0% vs 0%
Background: Inflammation contributes to cerebrovascular disease and is linked to stroke risk. Serum immune markers may capture these pathways and improve risk stratification beyond traditional factors. We used machine learning to identify an immune panel, and tested its predictive value in the multiethnic, population-based Northern Manhattan Study (NOMAS). Methods: Stroke-free NOMAS participants aged ≥55 years with serum levels of 60 immune markers measured by a customized Luminex assay were studied. We applied the least absolute shrinkage and selection operator (LASSO) Cox model, a penalized regression form of machine learning, with 5-fold cross-validation to log-transformed, standardized markers and all immune marker–marker interactions, using time-to-incident stroke as the outcome. Stability was assessed with event-stratified bootstrap resampling (B=300), retaining terms in ≥80% of resamples. Nested Cox models evaluated incremental predictive value of: (a) any single LASSO-selected marker, and (b) selected marker(s) plus stable interactions over traditional stroke predictors (Models 1–3: age only; +sociodemographics; +vascular risk factors: Table 2), using likelihood ratio tests (LRT) and, Harrell’s C-index. We assessed improvement in risk categorization via continuous net reclassification improvement (NRI). Results: Among 1,176 participants (mean age 70±9 years; 60% female; 68% Hispanic), 131 strokes occurred over 14 (IQR 6.5) years. LASSO selected one marker, CXCL9, and five statistical interaction terms, i.e., marker pairs whose effects varied with each other’s levels: IL13 with βNGF; CXCL10 with CCL2; IL13 with TRAIL; CXCL10 with IL1α; and IL1α with IL21. Adding CXCL9 alone improved model discrimination (ΔC-index≈+0.011, LRT p<0.05 all models), but NRI was not statistically significant (e.g., Model 3: NRI=0.02, p=0.49). Adding the full panel yielded larger C-index gains (Δ≈+0.015–0.016, p<0.05) and significantly improved reclassification in all models (NRI=0.22, 0.28, 0.32; all p≤0.01). In the fully adjusted Cox model, CXCL9 remained associated with stroke (HR=1.21 per SD, 95%CI 1.02–1.43). Conclusions: In this large cohort of urban older adults, a machine learning–selected serum immune marker panel improved risk reclassification beyond traditional factors, supporting its potential clinical utility. CXCL9, a monokine induced by IFN-γ (MIG), and the only single marker independently identified, may be a biologically plausible and modifiable prevention target.
Shahrestani et al. (Thu,) reported a other. The serum immune marker CXCL9 and its interaction panel improved stroke risk reclassification (NRI 0.22-0.32, p≤0.01) beyond traditional predictors in older adults.