In-stent restenosis (ISR) is the re-narrowing of a previously stented coronary segment, often presenting with recurrent angina or acute coronary syndrome (1, 2). Although new-generation drug-eluting stents (DES) have lowered risk, ISR still occurs in ∼5%–10% of cases worldwide (3, 4), partly because contemporary percutaneous coronary intervention (PCI) increasingly treats diffuse disease and small vessels (5, 6). Management typically involves drug-coated balloons or repeat DES, with intravascular imaging guiding lesion-specific therapy (6, 7). ISR carries substantial consequences, including recurrent symptoms and the need for repeat PCI or coronary artery bypass grafting (CABG) (8).

Traditional clinical and angiographic predictors—such as diabetes, complex lesions, and long or small stents—show limited ability to discriminate against ISR risk (9). Biologically, ISR reflects an immune inflammatory response to vascular injury: endothelial denudation, vascular smooth-muscle proliferation, and extracellular matrix deposition drive neointimal hyperplasia and, later, neoatherosclerosis (2, 10). Systemic inflammation, indexed by elevated high-sensitivity C-reactive protein (hs-CRP) and interleukin-6 (IL-6), has been associated with restenosis and stent failure (10). Consistent with this, higher ISR rates have been reported in antiphospholipid syndrome, systemic lupus erythematosus (SLE), and rheumatoid arthritis, underscoring immune mechanisms in ISR (10). Circulating rheumatic-immune inflammation (RII) biomarkers, including hs-CRP, IL-6, TNF-α, and complement C3/C4, capture systemic immune activation, but their prognostic performance for stent outcomes remains inconsistent across studies (11).

Antiphospholipid antibodies (aPL), including anticardiolipin (aCL), anti-β2-glycoprotein I (anti-β2GPI), and lupus anticoagulant (LA), are central to antiphospholipid syndrome (APS). By engaging endothelial cells, platelets, and monocytes, aPL induce endothelial activation and amplify proinflammatory and procoagulant signaling (e.g., TNF-α, TGF-β, and p38 MAPK), producing a prothrombotic milieu (12, 13). They also activate complement and the mTOR pathway, further increasing thrombotic risk (11, 14). Clinically, aPL positivity is associated with arterial and venous thrombosis and atherosclerotic cardiovascular disease (15), and small studies suggest higher in-stent restenosis (ISR) rates among aPL-positive patients (16).

A major knowledge gap remains regarding the prognostic role of aPL and a comprehensive RII panel in contemporary DES recipients. We therefore asked whether immune inflammatory status at PCI can refine risk stratification for coronary ISR. Specifically, we will prospectively test whether baseline aPL positivity and an expanded RII biomarker panel independently predict angiographic ISR 12 months after implantation of new-generation DES. Secondary aims are to examine associations with clinically driven target-lesion revascularization (TLR), quantify the incremental predictive value of adding immune markers to established clinical models, and explore whether IL-6 statistically mediates any association between aPL and ISR.

Of the 2,780 patients screened between May 2022 and January 2024, 2,503 met eligibility and were enrolled (Figure 1). A 12-month imaging was completed in 2,388 participants (1,757 invasive coronary angiography; 631 contrast-enhanced coronary CT angiography). Follow-up was incomplete in 115 patients due to loss to follow-up (n = 77), non-cardiac death (n = 24), or withdrawal of consent (n = 14).

The cohort's mean age was 62 ± 10 years, and 28.6% were women (Table 1). At 12  ±  1 months, ISR was confirmed in 193 of 2,388 patients (8.1%; 95% CI 6.9–9.1). Patients without 12-month imaging were not classified as “no ISR” and were excluded from ISR analyses. Diabetes mellitus was present in 20% overall and was more frequent among patients who developed ISR (32.1% vs. 19.0%; p < 0.001). Compared with those without ISR, ISR cases had longer treated lesions (28 ± 9 vs. 23 ± 7 mm) and smaller mean stent diameters (2.8 ± 0.3 vs. 3.0 ± 0.4 mm; both p < 0.001).

In modality-inclusive analyses of the imaging-complete cohort (Table 2), established angiographic predictors were significant (longer stent length, smaller stent diameter). Critically, any aPL positivity and higher IL-6 were each independently associated with ISR (aPL OR 1.92, 95% CI 1.34–2.74; IL-6 per doubling OR 1.25, 95% CI 1.10–1.42). hs-CRP attenuated in the fully adjusted model (OR 1.12; p =  0.09). These estimates reproduce your corrected calculations on the imaging-complete cohort and do not rely on classifying imaging-incomplete patients as “no ISR.”

Adding aPL and IL-6 to the clinical model improved discrimination (AUC 0.79 vs. 0.72; Δ  =  0.07; p =  0.008), with better calibration (slope ≈ 1.01; Hosmer–Lemeshow p =  0.45) and a lower Brier score (0.058 vs. 0.064). Net reclassification (NRI  =  0.18, p =  0.01) and IDI  =  0.04 favored the immune-enhanced model. Bootstrap validation (B  =  1,000) showed minimal optimism (AUC 0.78 vs. 0.71 for the clinical model), supporting internal stability (Table 3 and Figure 2).

To address modality-specific misclassification, an ICA-only analysis (n =  1,757) showed consistent associations: aPL OR 1.95 (95% CI 1.24–3.05) and IL-6 OR 1.27 (95% CI 1.09–1.48) (Table 4). Results were also robust when ISR was redefined as ≥70% stenosis, when patients with systemic autoimmune disease were excluded, with IPTW for ACI  >  0, and after excluding LA-positive cases (Table 4). Isotype/titer analyses indicated that high-titer IgG aPL carried the strongest associations.

Over 12 months, 100/2,503 (4.0%) underwent TLR. Freedom from TLR was lower in aPL-positive patients and declined stepwise across IL-6 quartiles (Figure 3). In adjusted Cox models, aPL positivity (HR 2.08, 95% CI 1.36–3.18) and IL-6 (per doubling; HR 1.29, 95% CI 1.11–1.50) were independently associated with higher TLR risk (Table 5).

Our findings align with contemporary summaries of the immune-coronary axis, in which inflammatory cytokines shape vascular healing and neointimal responses after PCI (19). In this large, prospective Chinese cohort that coupled comprehensive autoantibody testing with cytokine profiling in the era of new-generation DES, baseline aPL positivity and higher IL-6 were each independently associated with the risk of both ISR and clinically driven TLR. Adding these immune markers to a conventional clinical model improved discrimination. In mediation analysis, IL-6 accounted for approximately 27% of the total association between aPL and ISR, supporting an interlinked autoimmune-inflammatory pathway in restenotic healing.

Our findings confirm and extend earlier observations that autoimmunity increases the risk of coronary ISR. Small case reports and retrospective cohorts have consistently shown that patients with APS or SLE experience disproportionately high repeat-revascularization rates after PCI, even in the bare-metal stent era (20). More recent DES era data in RA and SSc similarly document excess ISR and adverse cardiac events (21, 22). In our DES-only cohort, baseline aPL positivity was associated with nearly twofold higher odds of 12-month ISR, in line with meta-analytic estimates in APS/SLE and demonstrating that this risk persists despite contemporary stent technology. The inflammatory milieu appears central. Whereas earlier meta-analyses linked elevated CRP to post-PCI events (23), our multivariable model—and others (24)—identify IL-6 as the stronger ISR predictor, with incremental discrimination comparable to gains achieved by adding HbA1c or on-treatment platelet reactivity in prior predictive modeling studies (25).

Mechanistically, aPL bind β2-glycoprotein I (β2GPI) on endothelial cells, upregulating ICAM-1/VCAM-1, triggering complement activation (C3, C5), and engaging the mTOR pathway processes that promote thrombosis and vascular smooth-muscle cell (VSMC) proliferation (26–30). In vitro, aPL enhances endothelial migration and matrix metalloproteinase-9 activity, whereas complement split products amplify IL-6 release and downstream Th17 responses, phenomena well documented in APS and SLE (31, 32). IL-6 is a potent amplifier of cytokine cascades that drives VSMC migration and extracellular matrix deposition; notably, IL-6 blockade attenuates neointimal hyperplasia in NZB/W F1 lupus-prone mice (14). Our mediation analysis, showing that IL-6 accounted for ∼27% of the aPL effect on ISR, fits this biology, while the remaining ∼73% direct effect suggests additional mechanisms, such as fibrin-rich microthrombi or immune-complex deposition within restenotic tissue (12). These mediation estimates are assumption-dependent and exploratory. Given single-time point biomarkers and potential measurement error, the ∼27% figure should be treated as descriptive on the log-odds scale, not as evidence that IL-6 explains the effect.

Although aPL and IL-6 were measured pre-PCI, and ISR was ascertained 12 months later, reducing the risk that the outcome directly influenced biomarker levels, reverse causation cannot be entirely excluded. Systemic inflammation could predispose to both higher biomarker levels and restenosis propensity. We adjusted for clinical covariates and hs-CRP and performed sensitivity analyses with unchanged inferences. Nevertheless, residual confounding from unmeasured inflammatory or immunologic factors remains possible. Consistent with this, we frame the findings as associations rather than causal effects, and the mediation results as assumption-dependent and hypothesis-generating.

Clinically, these findings support integrating immune inflammatory profiling into routine PCI practice. Validation data suggest that a preprocedural panel combining aPL, IL-6, and hs-CRP improves ISR risk prediction, particularly in APS and RA populations (25, 33), and the SII provides complementary prognostic information in acute coronary syndromes and diabetes (25, 34, 35). High-risk patients could undergo earlier surveillance to detect subclinical restenosis (36). Therapeutically, IL-6 blockade (e.g., tocilizumab), low-dose colchicine, or preferential use of sirolimus-eluting (mTOR-inhibiting) stents may mitigate immune-driven ISR; ongoing anti-cytokine trials in CAD offer a framework for testing these strategies (33, 35). Finally, in aPL-positive patients with demonstrable thrombophilia, extended dual antiplatelet therapy or adjunctive low-dose anticoagulation warrants prospective evaluation to personalize secondary prevention after PCI.

Nevertheless, several limitations merit consideration. As an observational study, residual confounding cannot be excluded. Biomarkers were measured at a single baseline time point, precluding assessment of post-PCI dynamics. Finally, this single-center cohort may limit generalizability. aPL was measured at a single baseline time point; thus, our findings reflect aPL serostatus rather than APS classification, which requires persistent positivity at ≥12 weeks at medium/high titers. Moreover, transient elevations in aPL can occur with intercurrent infection, recent vaccination, or acute illness, which we did not systematically adjudicate. Some misclassification toward transient positivity is therefore possible. ISR was ascertained by both ICA and CCTA. Although CCTA enables standardized follow-up, specificity is lower in stented segments due to blooming artifacts, small-lumen segmentation challenges, and heavy calcification; any such misclassification is expected to be nondifferential with respect to aPL/IL-6, biasing associations toward the null. Future work should externally validate the immune-enhanced prediction model across diverse populations; incorporate serial measurements of IL-6 and aPL to determine whether post-procedural trajectories further refine risk stratification; and pair biomarker profiling with mechanistic imaging to link immune signatures with neointimal phenotypes. Randomized trials of targeted anti-inflammatory or immunomodulatory therapies—such as IL-6 blockade or colchicine—in biomarker-positive patients, alongside genomic and epigenomic studies to identify susceptibility loci for immune-mediated ISR, will be essential to translate these findings into precision cardiovascular care.

In summary, antiphospholipid antibodies and heightened interleukin 6-mediated inflammation are independently associated with drug-eluting stent failure, increasing the risks of restenosis and clinically driven repeat revascularization despite contemporary PCI practice and supporting further development of immune-enhanced risk stratification and secondary prevention for ISR.