Can one composite index capture both inflammatory and metabolic cardiovascular risk in type 2 diabetes?

Abstract

Cardiovascular disease is a leading cause of morbidity and mortality in type 2 diabetes mellitus (T2DM), although optimal biomarker strategies for risk stratification remain incompletely defined. C-reactive protein (CRP) consistently predicts cardiovascular events in the general population, but its incremental value in T2DM - where chronic low-grade inflammation is already prevalent - has been debated. Fasting C-peptide (FCP), a surrogate marker of β-cell function and insulin resistance, shows paradoxical associations with cardiovascular outcomes, with bidirectional risk observed at both high and low levels. Such inconsistencies highlight the limitations of single biomarkers and have led to interest in composite approaches that integrate inflammatory and metabolic pathways. A recent study comprehensively introduced the CRP-FCP product, a multiplicative composite index demonstrating independent associations with cardiovascular, cerebrovascular, and combined vascular events, even when neither component alone achieved consistent significance across all vascular territories. This finding builds on the Danish DD2 cohort, where co-elevation of both biomarkers conferred the highest cardiovascular and mortality risk. In our view, the CRP-FCP product is best understood as a conceptual attempt to integrate inflammatory and metabolic risk signals rather than as a definitive predictive tool. While biologically plausible, its incremental value, methodological robustness, and generalizability across populations remain to be established, particularly in comparison with existing composite indices such as the triglyceride-glucose index.

Key Words: Type 2 diabetes mellitus; C-reactive protein; Fasting C-peptide; Composite biomarker; Cardiovascular risk; Insulin resistance; Inflammation; Triglyceride-glucose index; Cerebrovascular events; East Asian population

Core Tip: The product of C-reactive protein and fasting C-peptide represents a hypothesis-generating composite marker integrating inflammatory and metabolic cardiovascular risk in type 2 diabetes mellitus. By combining two biomarkers with distinct and individually limited predictive profiles, this approach aims to capture the adipose tissue-driven metaflammation axis underlying diabetic vascular disease. The multiplicative formulation preferentially reflects the high-C-reactive protein, high-fasting C-peptide phenotype, but its clinical value remains uncertain given modest discriminative performance, limited comparison with competing composite indices, and the need for prospective validation across diverse populations.

INTRODUCTION

Cardiovascular disease (CVD) is the leading cause of morbidity and mortality among patients with type 2 diabetes mellitus (T2DM), contributing to approximately 30% to 50% of all deaths in this population[1-4]. Despite substantial advances in pharmacological management - including sodium-glucose cotransporter-2 inhibitors, glucagon-like peptide-1 receptor agonists, and non-steroidal mineralocorticoid receptor antagonists - a significant residual cardiovascular risk persists even in well-treated patients[5-7]. This persistent residual risk highlights the need for more refined biomarker-based risk stratification strategies that better capture the multidimensional pathophysiology of diabetic vascular disease[8-10].

Traditional cardiovascular risk prediction models, such as the Framingham Risk Score[11] and the Pooled Cohort Equations[12], were developed predominantly in general populations and incorporate conventional risk factors including age, blood pressure, lipid levels, and smoking status, while treating diabetes merely as a binary present-or-absent variable[10,13]. This dichotomous approach does not account for the heterogeneous metabolic disturbances that characterize T2DM - namely chronic low-grade inflammation, insulin resistance, β-cell dysfunction, and dyslipidemia - all of which contribute to accelerated atherosclerosis through partially distinct yet interacting pathways[14,15]. The 2023 European Society of Cardiology guidelines for CVD management in diabetes explicitly emphasize the necessity of novel approaches to refine risk estimation in this heterogeneous high-risk population[5].

Two biomarkers that have individually attracted considerable attention in this context are high-sensitivity C-reactive protein (hs-CRP), a widely recognized marker of systemic inflammation[16], and fasting C-peptide (FCP), a surrogate for pancreatic β-cell secretory function and insulin resistance[17]. However, neither biomarker alone has proven consistently reliable for cardiovascular risk prediction in T2DM populations: C-reactive protein (CRP)’s predictive value attenuates after adjustment for established risk factors in some diabetic cohorts[18-20], while FCP exhibits paradoxical and bidirectional associations with cardiovascular outcomes[21,22]. These limitations have prompted interest in composite biomarker strategies that may capture the synergistic interaction between inflammatory and metabolic dysregulation.

Zhang et al[23] address this challenge by proposing the CRP-FCP product - a composite index derived by multiplying serum CRP and FCP values - and evaluating its associations with cardiovascular events (CVEs), cerebrovascular events (CBVEs), and their combination (CCBVEs) in 8486 patients with T2DM. Unlike organ-specific markers such as N-terminal pro-brain natriuretic peptide (NT-proBNP) or high-sensitivity cardiac troponin (hs-cTn) that reflect existing myocardial stress or injury[24], the CRP-FCP product is positioned as an upstream indicator of the systemic inflammatory-metabolic milieu from which CVEs arise. In this context, we reconsider the conceptual rationale, supporting evidence, and methodological considerations for composite inflammatory-metabolic biomarkers in T2DM cardiovascular risk assessment.

CRP IN T2DM: CURRENT EVIDENCE AND CONTROVERSIES

CRP as a cardiovascular risk marker: General population evidence

CRP is an acute-phase protein synthesized predominantly by hepatocytes in response to interleukin-6 (IL-6) stimulation and one of the most widely requested laboratory tests for inflammatory assessment in clinical practice[25]. Over the past two decades, hs-CRP has emerged as one of the most extensively studied inflammatory biomarkers in cardiovascular medicine[26]. The landmark Emerging Risk Factors Collaboration individual-participant meta-analysis, encompassing over 160000 individuals, demonstrated that CRP was associated with coronary heart disease risk with a hazard ratio of 1.37 [95% confidence interval (CI): 1.27-1.48] per standard-deviation increase after adjustment for established risk factors[27]. More recently, a large-scale 2025 analysis confirmed a linear, graded association between elevated CRP and increased cardiovascular risk across diverse cohorts[28], reinforcing these earlier findings.

The incremental prognostic value of CRP has been quantified in several meta-analytic frameworks. A meta-analysis of 39 independent studies (n = 175778) demonstrated that adding circulating CRP to clinical risk models improved risk discrimination, with a pooled incremental Δc-index of 2.2% for major adverse CVEs[29]. CRP also predicts extreme cardiovascular outcomes: A meta-analysis encompassing over 117000 individuals showed that those in the highest CRP category faced a 1.62-fold increased risk of sudden death compared with those in the lowest category[30]. Furthermore, recent large-scale evidence from over 40000 participants confirms that elevated CRP (≥ 2 mg/L) remains an independent predictor of major adverse CVEs in both primary and secondary prevention settings, even under contemporary therapies including high-intensity statins[31].

However, whether CRP plays a causal role in CVD remains uncertain despite this robust observational evidence. The JUPITER trial demonstrated that statin therapy significantly reduced CVEs in individuals with elevated CRP but without hyperlipidemia, supporting CRP-guided therapeutic stratification[32]. However, bidirectional Mendelian randomization studies have consistently shown that genetically elevated CRP levels do not directly cause CVD[33,34], suggesting that CRP functions as a sensitive marker of the underlying systemic inflammatory state rather than a direct mediator of vascular injury.

CRP in diabetic populations: Attenuated and contested predictive value

A large-scale meta-analysis of over 22000 T2DM patients demonstrated that those with the highest CRP levels had a 76% greater risk of cardiovascular mortality (relative risk = 1.76; 95%CI: 1.46-2.13) and a 2-fold higher risk of all-cause mortality (relative risk = 2.03; 95%CI: 1.49-2.75)[35]. However, the independent predictive value of CRP in T2DM populations is increasingly contested. A pooled analysis of 25979 participants from four United Kingdom prospective cohort studies showed that CRP’s association with cardiovascular mortality was comparable in diabetic and non-diabetic individuals, raising questions about its added value in an already high-risk population[18]. More critically, in the ADVANCE study of 3763 T2DM participants, CRP’s association with major macrovascular events became non-significant after adjustment for IL-6, suggesting that CRP may be a downstream marker of the IL-6 signaling pathway rather than an independent risk contributor[19]. A comprehensive 2024 systematic review on precision cardiovascular prognostics in T2DM corroborated these findings, demonstrating that CRP - despite its broad associations - provides limited incremental predictive power over established clinical risk models[36]. Furthermore, a 2024 observational and Mendelian randomization study from the Copenhagen General Population Study (n = 106177) showed that genetic instruments for CRP revealed no causal cardiovascular effect in either diabetic or non-diabetic individuals[20], reinforcing the interpretation that CRP elevation in T2DM reflects the underlying inflammatory milieu rather than directly driving vascular injury. Taken together, these findings raise concerns about relying on CRP as a standalone cardiovascular predictor in T2DM and suggest that its predictive utility may differ across vascular territories.

CRP and cerebrovascular specificity: An unresolved question

An additional unresolved issue is whether CRP’s associations differ across vascular territories. Although CRP is an established marker for ischemic stroke, meta-analytic evidence indicates that this association carries considerably lower certainty for cerebrovascular outcomes than for major adverse CVEs[27]. A 2025 large-scale analysis further emphasized that while hs-CRP is a robust predictor of coronary heart disease, its independent association with stroke often attenuates after adjusting for systolic blood pressure and other traditional risk factors, suggesting a more complex, multi-factorial etiology for CBVEs[28]. Furthermore, the observation that elevated CRP is equally associated with non-cardiovascular mortality - including respiratory and neoplastic diseases - reinforces its role as a “potent but non-specific” marker of general systemic vulnerability rather than a dedicated vascular mediator[28,37]. Importantly, however, this non-specific pattern appears to be modified by diabetes status. Recent data from the China National Stroke Registry III demonstrate that the hs-CRP-stroke association is most pronounced in patients with impaired glucose metabolism, suggesting that the cerebral vasculature may be particularly vulnerable to inflammatory burden in the diabetic state[38]. Zhang et al’s finding[23] of an independent CRP association specifically with CBVEs - but not CVEs - in their T2DM cohort reinforces this observation. This territory-specific vulnerability in diabetes may reflect distinct mechanisms such as accelerated blood-brain barrier dysfunction, cerebral small-vessel disease, or enhanced neuroinflammation[39]. A recent systematic review further highlighted this discrepancy, reporting high-certainty evidence for CRP’s association with major adverse CVEs but only very low certainty specifically for cerebrovascular outcomes in secondary prevention populations[40]. This inconsistency makes Zhang et al’s observation[23] - that CRP was independently associated only with CBVEs in this T2DM cohort - particularly noteworthy. It raises the possibility that under metabolic dysregulation, the inflammatory-cerebrovascular axis may follow a distinct pathological trajectory.

FCP AS A CARDIOMETABOLIC MARKER: BIDIRECTIONAL RISK AND CLINICAL AMBIGUITY

C-peptide physiology and clinical significance

C-peptide, a 31-amino-acid peptide cleaved from proinsulin during insulin biosynthesis, is released in equimolar amounts with insulin from pancreatic β-cells[17]. Unlike insulin, C-peptide is not subject to substantial first-pass hepatic extraction and has a longer circulating half-life, making FCP a more stable surrogate marker of endogenous insulin secretion than circulating insulin itself[17,41]. In T2DM, however, the clinical meaning of FCP is highly context dependent. In earlier or more insulin-resistant stages of disease, elevated FCP generally reflects compensatory hyperinsulinemia driven by peripheral insulin resistance, whereas in more advanced disease, declining FCP may indicate progressive β-cell failure and reduced secretory reserve[17,41]. Interpretation of FCP therefore requires attention to disease duration, treatment exposure, and renal function, all of which may influence circulating levels independently of the underlying cardiometabolic phenotype[17,41].

Beyond its role as a passive biomarker, emerging evidence suggests that C-peptide may exert biological effects on endothelial signaling, vascular cellular responses, and inflammatory regulation[42]. However, the clinical significance of these effects remains uncertain, and C-peptide’s primary utility in clinical practice continues to center on its function as a marker of β-cell reserve and, by extension, insulin resistance.

The bidirectional cardiovascular risk of C-peptide

One of the key difficulties in interpreting C-peptide’s cardiovascular associations is its bidirectional risk profile. The Skaraborg Diabetes Register demonstrated that elevated FCP predicted all-cause and cardiovascular mortality in 398 patients with newly diagnosed T2DM over a mean follow-up of 12 years [hazard ratio (HR) = 1.53 per unit increase; 95%CI: 1.10-2.11][43]. This finding has been corroborated by a large meta-analysis of 23 studies encompassing diverse populations, which found that elevated C-peptide was associated with CVD mortality (HR = 1.38; 95%CI: 1.08-1.77) and all-cause mortality (HR = 1.28; 95%CI: 1.12-1.46)[44].

However, the relationship is not uniformly linear. A separate meta-analysis reported that low C-peptide levels were associated with increased coronary heart disease and cerebral infarction risk in T2DM patients[45]. In addition, C-peptide’s vascular associations appear to differ according to complication type, with non-identical patterns observed for macrovascular and microvascular disease[46]. Most notably, Yan et al[21] demonstrated a “V”-shaped association between FCP and cardiovascular risk biomarkers in non-diabetic adults, whereas this pattern was not observed in patients with established T2DM[21]. A separate systematic review and meta-analysis further found that, after adjustment for major confounders, the overall association between C-peptide and CVEs became non-significant, underscoring the persistent clinical ambiguity of this biomarker[22].

This apparent inconsistency may instead reflect the fact that FCP can index two biologically distinct high-risk states depending on metabolic stage. At higher levels, elevated FCP is more likely to mark hyperinsulinemic insulin resistance, accompanied by dyslipidemia, adipose tissue dysfunction, and chronic low-grade inflammation. At lower levels, reduced FCP may instead reflect β-cell exhaustion, advanced disease, therapeutic insulin dependence, and greater metabolic fragility[17,21,22]. This stage-dependent interpretation may help explain why FCP alone performs inconsistently as a cardiovascular marker. Its value may lie less in standalone use than in combination with complementary biomarkers - such as CRP - that place it within a broader inflammatory-metabolic context[22,44].

PATHOPHYSIOLOGICAL RATIONALE FOR COMBINING INFLAMMATORY AND METABOLIC BIOMARKERS

The adipose tissue-inflammation-insulin resistance axis

The rationale for combining CRP and FCP into a composite index reflects the well-characterized interrelationship between chronic low-grade inflammation and insulin resistance in T2DM[47,48]. This relationship extends beyond simple correlation and reflects a bidirectional pathogenic loop mediated primarily through adipose tissue-driven immunometabolic signaling, a concept recently expanded by updated metaflammation frameworks[49,50].

In the setting of visceral adiposity - a hallmark of T2DM - expanded and dysfunctional adipose tissue serves as an active endocrine and immune organ, secreting a constellation of pro-inflammatory adipokines and cytokines, such as tumor necrosis factor-α, IL-6, monocyte chemoattractant protein-1, and resistin[51]. tumor necrosis factor-α directly impairs insulin signaling by promoting serine phosphorylation of insulin receptor substrate-1. Conversely, the resulting hyperinsulinemia and localized metabolic stress further amplify the inflammatory response by promoting the recruitment and phenotypic polarization of pro-inflammatory M1 macrophages within the adipose tissue - a dynamic process increasingly recognized as a driver of metabolic dysfunction[50,52,53]. IL-6, largely derived from this inflamed adipose environment, acts as the principal stimulus for hepatic CRP synthesis[54]. This creates a self-sustaining cycle in which systemic inflammation drives peripheral insulin resistance, while insulin resistance, in turn, perpetuates and exacerbates inflammatory signaling[50,55].

This interconnection is reflected in epidemiological data showing consistent positive correlations between CRP and markers of insulin resistance across T2DM populations. The positive correlation between CRP and FCP observed in the Zhang et al’s cohort[23] (r = 0.102, P < 0.001) is modest in magnitude - as expected given the multiple determinants of each biomarker - but consistent with the underlying pathophysiology linking these pathways (Figure 1).

Figure 1 Pathophysiological rationale for the C-reactive protein-fasting C-peptide product. Dysfunctional visceral adipose tissue drives a pro-inflammatory cascade (tumor necrosis factor-α, interleukin-6, monocyte chemoattractant protein-1), resulting in systemic inflammation and insulin resistance with compensatory β-cell activation. These pathways amplify each other bidirectionally. The C-reactive protein × fasting C-peptide product conceptually integrates these two axes and is associated with cardiovascular events, cerebrovascular events, and combined cardiovascular and cerebrovascular events. The dashed box indicates that β-cell failure (low fasting C-peptide) structurally attenuates the product signal regardless of C-reactive protein elevation. TNF-α: Tumor necrosis factor-α; IL-6: Interleukin-6; MCP-1: Monocyte chemoattractant protein-1; CRP: C-reactive protein; FCP: Fasting C-peptide; CVEs: Cardiovascular events; CBVEs: Cerebrovascular events; CCBVEs: Combined cardiovascular and cerebrovascular events.

Mechanistic justification for a multiplicative composite

The multiplicative formulation of the CRP-FCP product is designed to capture the potential synergistic interaction between inflammatory and metabolic pathways. Within this framework, concurrent elevations in both markers are thought to reflect a disproportionately greater cardiovascular risk burden than the sum of either marker alone[40,44,56]. While experimental models sometimes suggest that C-peptide may have independent vasoprotective effects[57], its systemic elevation in the context of T2DM predominantly serves as a surrogate for the hyperinsulinemic, pro-inflammatory state that drives macrovascular complications[22,44].

However, several features of this multiplicative model warrant consideration. This formulation specifically captures the risk associated with the insulin-resistant, pro-inflammatory phenotype - characterized by high-FCP and high-CRP levels. Conversely, it structurally attenuates the signal from individuals with advanced β-cell failure and low FCP values, in whom the product remains low regardless of CRP elevation. Thus, while FCP alone has shown conflicting “U-shaped” or “V-shaped” associations with cardiovascular risk at both pathologically high and low levels[21,22], the CRP-FCP product is not intended to fully resolve this bidirectional ambiguity. Instead, it serves as a targeted operational tool for the metabolic-inflammatory end of the spectrum, where the synergy between these two drivers is most potent.

THE CRP-FCP PRODUCT: EVIDENCE FROM THE DANISH AND CHINESE COHORTS

The Danish DD2 cohort: Categorical co-elevation of CRP and FCP

The rationale for combining inflammatory and metabolic biomarkers is supported by findings from the Danish DD2 cohort study. In that prospective study of 7301 patients with recent-onset T2DM followed for a median of 4.8 years, Gedebjerg et al[58] demonstrated that patients with elevated levels of both CRP and FCP faced the highest risks of CVEs and all-cause mortality. Notably, the Danish study also revealed an outcome-specific pattern, where inflammation and insulin resistance contributed differentially depending on the endpoint of interest. This observation provided the first prospective evidence that these two biomarker axes interact to stratify cardiovascular risk in T2DM beyond what either achieves independently.

A continuous composite index in an East Asian T2DM cohort

The present work by Zhang et al[23] extends this line of evidence in several respects. First, the use of a continuous composite index (the CRP-FCP product) preserves the full range of variation in both biomarkers, avoiding the information loss inherent in the categorical grouping as employed in the Danish study. Second, this study differentiates between CBVEs and CVEs, demonstrating that the CRP-FCP product is associated with both vascular territories - a granularity not explored in the Danish study[58]. Third, the East Asian study population provides valuable insight into the applicability of this concept across populations with distinct body composition, insulin secretory capacity, and metabolic phenotype[59-61].

The study analyzed a substantial registry-based cohort of 8486 patients from Shanghai Sixth People’s Hospital, stratified by CRP-FCP product values. A key finding from the fully adjusted logistic regression analysis was a distinct dissociation in vascular-territory specificity. Specifically, CRP alone was significantly associated with CBVEs but not CVEs, whereas FCP showed the opposite pattern, with significant associations for CVEs but not CBVEs. In contrast to these inconsistent individual patterns, the CRP-FCP product achieved independent and uniform associations across all three outcomes: CVEs, CBVEs, and CCBVEs (odds ratio 1.014-1.017). This divergent pattern, in which the composite retained significance across vascular endpoints that its individual components did not consistently capture, appears to represent a key contribution of the study.

However, the cross-sectional design and modest discriminative accuracy (area under curve 0.550-0.558) of the present study represent methodological limitations compared with the prospective design and stronger effect sizes (HR = 1.61-2.36) observed in the Danish cohort[58].

HOW DOES THE CRP-FCP PRODUCT COMPARE WITH OTHER COMPOSITE APPROACHES?

The triglyceride-glucose index

The triglyceride-glucose (TyG) index, calculated as ln[fasting triglycerides (mg/dL) × fasting glucose (mg/dL)/2], has emerged as the most widely studied composite metabolic biomarker for cardiovascular risk in recent years[62]. Originally proposed as a surrogate for insulin resistance, the TyG index has demonstrated associations with coronary artery disease, myocardial infarction, stroke, and composite cardiovascular outcomes in large meta-analyses[63,64]. The prospective PURE study across 22 countries showed that the TyG index predicted CVD and mortality in low- and middle-income countries, although its predictive utility for CVEs (as opposed to diabetes incidence) was attenuated in high-income settings[65].

Compared with homeostasis model assessment (HOMA) of insulin resistance (HOMA-IR), the TyG index offers practical advantages: It requires only routine fasting triglyceride and glucose measurements (no insulin assay), is more reproducible, and shows comparable or superior discrimination for metabolic syndrome and cardiovascular outcomes in several head-to-head comparisons[66,67]. However, the TyG index fundamentally captures the metabolic-lipotoxic axis of insulin resistance and does not incorporate inflammatory information. In a recent precision-prognostic meta-analysis evaluating 195 biomarkers for cardiovascular prediction in T2DM, both CRP and the TyG index independently showed improvement in prediction performance, suggesting that these inflammatory and metabolic axes provide non-overlapping risk information[36]. In this context, the CRP-FCP product may represent an alternative composite approach by integrating systemic inflammation (CRP) with an endogenous insulin-secretion-related marker (FCP), potentially capturing aspects of cardiometabolic risk not reflected by TyG alone.

HOMA-IR and related insulin resistance indices

HOMA-IR (fasting glucose × fasting insulin/22.5) remains the most commonly used surrogate for insulin resistance in epidemiological research. Its cardiovascular prognostic value in non-diabetic populations is well established, but its utility in T2DM populations is limited by the confounding effects of exogenous insulin therapy, progressive β-cell failure (which lowers insulin levels independently of improving sensitivity), and the analytical variability of insulin assays. FCP-based HOMA (HOMA-C, using C-peptide instead of insulin) has been proposed as an alternative that avoids some of these limitations, but it has not been widely validated for cardiovascular prediction[41].

Multi-marker approaches

The emergence of high-throughput “omics” platforms has stimulated interest in multi-marker panels that simultaneously capture inflammatory, metabolic, prothrombotic, and neurohormonal axes of cardiovascular risk[36]. A precision-prognostic systematic review evaluating 195 biomarkers for cardiovascular prediction in T2DM identified NT-proBNP (high evidence), hs-cTn (moderate evidence), the TyG index (moderate evidence), and genetic risk scores (moderate evidence) as those with the strongest evidence for improving prediction beyond established risk factors[36]. CRP was supported by moderate predictive evidence, whereas C-peptide was not individually evaluated within this framework.

The CRP-FCP product can be viewed within this broader landscape of cardiovascular biomarkers in T2DM, where different markers capture distinct pathophysiological domains. Organ-specific markers such as NT-proBNP and hs-cTn reflect existing myocardial wall stress and subclinical cardiomyocyte injury, respectively, and currently demonstrate the strongest incremental predictive evidence for CVEs in T2DM[24,36]. Renal-vascular markers, particularly the urine albumin-to-creatinine ratio, serve as a window into systemic endothelial dysfunction and have long been incorporated into cardiovascular risk stratification guidelines for diabetic patients[5]. Upstream inflammatory mediators such as IL-6 - the principal inducer of hepatic CRP synthesis - and growth differentiation factor-15, a stress-responsive cytokine reflecting oxidative stress and cellular senescence, have shown emerging prognostic value that may surpass CRP in pathophysiological specificity[19,36]. In contrast, the CRP-FCP product does not assess end-organ damage or pathway-specific inflammation but rather quantifies the combined burden of the systemic inflammatory-metabolic milieu that precedes and promotes vascular injury. This upstream positioning may confer complementary value when interpreted alongside organ-specific and pathway-specific markers, although formal studies evaluating such multi-marker strategies incorporating the CRP-FCP product have not yet been conducted.

Against this background, the CRP-FCP product represents a conceptually distinct approach: Rather than introducing a novel biomarker or requiring specialized assays, it derives a composite index from two readily available laboratory measurements that reflect complementary pathophysiological domains. Its principal strength lies in its simplicity and biological interpretability; its main limitation is the absence of formal model comparison studies demonstrating incremental discrimination or reclassification beyond established risk factors and competing composite indices[68,69].

Comparative summary

Table 1 provides a structured comparison of the CRP-FCP product with other representative composite or surrogate cardiovascular risk markers evaluated in T2DM populations, including the TyG index, HOMA-IR, and multi-marker approaches.

Table 1 Comparison of composite cardiovascular risk biomarkers in type 2 diabetes mellitus.

Feature	CRP-FCP product	TyG index	HOMA-IR	Multi-marker panels
Formula	CRP × FCP	ln(TG × FG/2)	FG × insulin/22.5	Various combinations
Primary pathophysiological domain	Inflammation + endogenous insulin secretory status	Lipotoxic-metabolic insulin resistance	Insulin resistance	Multiple biological axes
Inflammatory component	Yes (CRP)	No	No	Variable
Requires insulin assay	No	No	Yes	Variable
Prospective evidence in T2DM	Limited (DD2 categorical only)	Relatively extensive	Moderate, heterogeneous	Emerging
Reported predictive performance (AUC)	Modest (0.550-0.558)	Moderate (0.60-0.70)	Variable (0.55-0.65)	Potentially higher (0.65-0.80)
Ethnic diversity of evidence	European + East Asian	Broad, including multi-ethnic cohorts (PURE)	Broad, heterogeneous	Predominantly Western
Practical cost and feasibility	Low	Low	Moderate	Moderate to high

CRP: C-reactive protein; FCP: Fasting C-peptide; TyG: Triglyceride-glucose; HOMA-IR: Homeostasis model assessment of insulin resistance; TG: Triglycerides; FG: Fasting glucose; AUC: Area under the curve; T2DM: Type 2 diabetes mellitus.

METHODOLOGICAL CONSIDERATIONS IN COMPOSITE BIOMARKER DEVELOPMENT

The choice of multiplicative formulation

Several aspects of the study design warrant recognition[23]. The recruitment of a substantial real-world cohort of 8486 patients provides adequate statistical power to detect independent associations. The application of six progressively adjusted regression models - systematically controlling for demographics, lifestyle factors, pharmacological treatments, anthropometric measurements, and metabolic parameters - represents a rigorous analytical framework. Notably, the persistence of significant associations for the CRP-FCP product even through the most stringent adjustment model suggests that the observed associations are not merely attributable to confounding by conventional risk factors.

However, the choice of a multiplicative product appears to be primarily pragmatic and exploratory, rather than the result of a prespecified mathematical hypothesis about the functional form of risk. The present analysis does not formally demonstrate the superiority of a product term over alternative combinations - such as additive scores, standardized sums, or log-scale models with explicit interaction terms - nor does it systematically compare predictive performance across competing functional forms. Accordingly, the CRP-FCP product may be better regarded as a hypothesis-generating composite biomarker that operationalizes the recognized biological synergy between inflammation and metabolic dysfunction, rather than a validated risk score ready for routine clinical implementation[68].

The need for formal model comparison and validation

From a methodological perspective, future work could move beyond hand-crafted transformations toward approaches that formally evaluate the functional form of the CRP-FCP relationship with vascular risk. For example, generalized additive models with spline functions could explore non-linear joint effects within cross-sectional frameworks, while prospective designs would permit Cox proportional hazards models with explicit interaction terms to test for supra-multiplicative effects. Such approaches would help determine whether the multiplicative product represents the optimal functional form or whether alternative specifications provide superior discrimination and calibration.

Furthermore, rigorous biomarker validation requires demonstration of incremental predictive value through metrics such as the C-statistic improvement, net reclassification index, and integrated discrimination improvement - analyses that have not yet been performed for the CRP-FCP product[68,69]. Until such head-to-head comparisons are conducted, the CRP-FCP product remains a promising conceptual framework that requires empirical validation of its comparative utility.

ETHNIC AND POPULATION-SPECIFIC CONSIDERATIONS

East Asian vs Western metabolic phenotypes

The application of the CRP-FCP product in an East Asian T2DM population raises questions about population-specific generalizability. T2DM in East Asian populations differs from that in Western populations in several clinically relevant metabolic characteristics[59]. East Asian individuals tend to develop T2DM at lower body mass index (BMI) thresholds, reflecting a greater propensity for visceral and ectopic fat accumulation despite lower overall adiposity[60]. In addition, East Asian populations generally exhibit more limited insulin secretory reserve and a greater relative contribution of β-cell dysfunction, compared with insulin resistance, in the pathogenesis of T2DM[59,60].

These differences have direct implications for interpretation of the CRP-FCP product. First, the lower BMI threshold for metabolic disease suggests that adipose-tissue-associated CRP elevations may occur at different absolute levels than in Western populations, potentially necessitating population-specific thresholds. Second, the greater contribution of β-cell dysfunction in East Asian T2DM implies that a larger proportion of patients may fall into the low-FCP phenotype, where the multiplicative index structurally attenuates signal regardless of inflammatory burden. Third, baseline CRP concentrations are consistently lower in East Asian populations than in White and Black populations[70], which may further compress the dynamic range of the CRP-FCP product.

Implications for multi-ethnic validation

The heterogeneity of T2DM across ethnic groups - including differences in obesity prevalence, insulin resistance, and β-cell dysfunction - highlights the need to validate the CRP-FCP product in diverse populations[59,60]. Minimal model analyses comparing insulin sensitivity, β-cell function, and hepatic insulin extraction between Japanese and Caucasian populations have demonstrated significant ethnic differences in these parameters[61], suggesting that a single composite index may not perform equivalently across populations without calibration.

Future multi-ethnic validation studies should consider stratifying results by ethnicity, BMI category, diabetes duration, and treatment modality. Particular attention should be paid to insulin therapy, which directly complicates the interpretation of FCP by dissociating exogenous insulin supply from endogenous β-cell function[41]. Such analyses would help determine whether the CRP-FCP product maintains consistent associations across these subgroups or requires population-adapted thresholds.

LIMITATIONS, GAPS, AND FUTURE DIRECTIONS

Despite these promising findings, several limitations and unresolved questions warrant consideration beyond the scope of the original report[23]. Receiver operating characteristic analyses yielded area under curve values of 0.550 for CVEs, 0.555 for CBVEs, and 0.558 for CCBVEs, indicating modest discriminative performance. The authors appropriately acknowledge that the CRP-FCP product is not intended as a standalone diagnostic or prognostic marker. Its potential value may instead lie in providing complementary information on the combined inflammatory-metabolic burden - a domain not directly captured by conventional cardiovascular risk algorithms[1].

Furthermore, the ORs associated with the CRP-FCP product, although statistically significant, were small in magnitude, ranging from 1.014 to 1.017. The clinical relevance of such incremental risk increases at the individual patient level remains uncertain, particularly in light of recent United Kingdom Biobank data suggesting that longitudinal biomarker trajectories may provide more informative risk stratification than single-point measurements alone[71].

The duration-related patterns reported in the study also warrant cautious interpretation. Longer diabetes duration was associated with lower CRP and lower CRP-FCP product values, together with an apparent biphasic FCP trajectory characterized by an initial rise followed by a decline. However, existing literature does not consistently support a duration-dependent decline in CRP; rather, CRP levels appear to reflect current adiposity, insulin resistance, and metabolic control more closely than diabetes duration itself[72]. Similarly, because the natural history of β-cell function in T2DM generally involves progressive decline[73], the observed biphasic pattern in FCP likely reflects cross-sectional cohort structure, treatment effects, or residual confounding rather than a true biological trajectory.

A roadmap for future research

Several priorities for future investigation can be identified. Prospective cohort studies with repeated biomarker measurements are needed to establish the temporal stability of the CRP-FCP product and its ability to predict incident CVEs[71]. Head-to-head comparisons between the CRP-FCP product and competing composite indices - particularly the TyG index - should evaluate relative discrimination, calibration, and reclassification metrics[68,69]. Multi-ethnic validation across populations with distinct metabolic phenotypes, including South Asian, Black, and Hispanic cohorts, is essential, given the known ethnic differences in adiposity, insulin resistance, and β-cell dysfunction[59,60]. Intervention studies are also needed to examine whether therapies that reduce inflammatory burden or improve insulin sensitivity modify the CRP-FCP product, and whether such changes track with clinical outcomes. In addition, integrating the CRP-FCP concept into machine-learning risk prediction frameworks - where non-linear interactions between inflammatory and metabolic markers can be formally modeled - may provide a promising avenue for improving cardiovascular risk stratification in T2DM (Figure 2).

Figure 2 Research roadmap for validation and clinical translation of the C-reactive protein-fasting C-peptide product. The schematic outlines key steps for future investigation, including prospective validation, model comparison with established indices, multi-ethnic calibration, and integration into machine-learning-based risk prediction frameworks. GAM: Generalized additive model; TyG: Triglyceride-glucose; NRI: Net reclassification index; IDI: Integrated discrimination improvement.

CONCLUSION

The CRP-FCP product represents a conceptually appealing approach to integrating systemic inflammation and metabolic dysfunction into a single clinical measure. While its consistent association with vascular outcomes is promising and biologically plausible, the currently available cross-sectional evidence does not establish incremental predictive value beyond conventional risk factors. The individual limitations of CRP - including non-specificity and attenuated predictive value in T2DM - and of FCP - including its bidirectional risk profile and dependence on disease stage - are partially addressed but not fully resolved by their multiplicative combination, which preferentially captures the high-CRP, high-FCP phenotype. Future prospective research across ethnically diverse populations is essential to determine whether this composite index provides incremental benefit over established risk scores and competing composite biomarkers. Until then, the CRP-FCP product may be best viewed as a hypothesis-generating framework that highlights the interacting contributions of inflammatory and metabolic dysregulation to vascular risk in T2DM.