Enhanced predictive value of post-treatment lymphocyte-based immunoinflammatory biomarkers for cancer therapy-related cardiac dysfunction in anthracycline-treated breast cancer patients.

Background
Cardiovascular disease is a leading cause of morbidity and mortality among cancer survivors [1, 2]. Anthracyclines are among the most widely used and effective chemotherapeutic agents for breast cancer, leukemia, lymphoma, and other hematologic and solid malignancies [3]. However, anthracycline administration is closely associated with the development of various cardiovascular complications [4], among which cancer therapy–related cardiac dysfunction (CTRCD) is the primary driver of adverse cardiovascular events [5]. CTRCD is predominantly characterized by left ventricular systolic dysfunction, with an estimated 40% of breast cancer patients in the United States experiencing a reduction in left ventricular ejection fraction (LVEF) [1, 6].
Anthracycline-induced cardiac damage is a cumulative and partially irreversible process with insidious onset and no obvious clinical symptoms in the early stages. Without timely detection and intervention, the subclinical cardiac injury will gradually progress to overt heart failure, at which point the myocardial structural damage and systolic function impairment are often irreversible [7]. This irreversible cardiac dysfunction not only severely reduces the quality of life of breast cancer survivors but also significantly increases their long-term cardiovascular mortality, which has become a major cause of non-tumor-related death in cancer survivor [8, 9]. Recognizing the pivotal role of early detection in CTRCD management, the 2022 European Society of Cardiology (ESC) guidelines recommend close surveillance during anthracycline chemotherapy to enable timely diagnosis and inform clinical decision-making. In addition to cardiac imaging, recommended monitoring strategies include serial assessments of cardiac biomarkers such as cardiac troponin (cTnT) and natriuretic peptides (NP) [5, 10]. However, as nonspecific markers of cardiac injury, cTnT and NP, are merely elevated after overt cardiac damage has occurred, which severely limits their utility for the early detection of CTRCD [5]. Thus, there is an urgent need to develop cost-effective and clinically feasible biomarkers for the early screening and monitoring of anthracycline-induced cardiac toxicity.
Inflammation is a key pathogenic mediator of anthracycline - induced CTRCD [11, 12]. Clinical evidence has demonstrated that lymphocyte counts are significantly reduced in breast cancer patients with CTRCD, and lymphocyte-based immunoinflammatory biomarkers are closely correlated with the development of CTRCD in these patients [13]. Compared with single-marker indices, lymphocyte-based immunoinflammatory biomarkers provide a more comprehensive and refined assessment of systemic immunoinflammatory status [14–16]. Specifically, the neutrophil-to-lymphocyte ratio (NLR) and monocyte-to-lymphocyte ratio (MLR) have been identified as independent predictors for both the diagnosis and long-term prognosis of CTRCD [13, 17, 18].
Additionally, an increasing number of novel lymphocyte-based immunoinflammatory biomarkers - including the platelet-lymphocyte ratio (PLR), systemic immune inflammation index (SII), and system inflammation response index (SIRI), and neutrophil-monocyte to lymphocyte ratio (NMLR) - have been shown to be significantly positively correlated with coronary artery stenosis, coronary artery calcification, and acute coronary syndrome [19–21]. However, their predictive value for anthracycline-induced CTRCD remains largely unexplored. Therefore, the present study aims to evaluate the clinical utility of multiple lymphocyte-based immunoinflammatory biomarkers assessed pre- and post-anthracycline therapy as early warning indicators for CTRCD in breast cancer patients.

Methods

Study population
In this study, patients with breast cancer who received anthracycline-based chemotherapy at Songjiang Hospital from January 2018 to September 2024 were consecutively enrolled. Anthracycline-based treatment was administered as a standard 4-cycle regimen: each cycle consisted of doxorubicin or epirubicin every 2–3 weeks, with the total treatment duration ranging from 8 to 12 weeks [22]. Given that the entire anthracycline chemotherapy regimen for our breast cancer cohort lasted approximately 3 months, we further clarified two core time points for biomarker detection: pre-treatment refers to the period within 1 week before the first anthracycline dose (for baseline indicator collection); post-treatment refers to the period within 1 week after the completion of the entire chemotherapy regimen. Patients with severe liver and kidney dysfunction [23, 24], LVEF < 55% before treatment initiation, or those with a previous history of heart failure (HF) were excluded. Patient age was defined as their chronological age at the initiation of anthracycline therapy, and body weight was recorded based on the most recent measurement prior to anthracycline treatment initiation. The presence of hypertension, dyslipidemia, and diabetes mellitus was confirmed according to established diagnostic criteria at the start of oncological treatment. The study was approved by the Ethics Committee of Songjiang Hospital Affiliated to Shanghai Jiaotong University School of Medicine (No. 2025-22). The study protocol was conducted in adherence to the Declaration of Helsinki, and all participants provided written informed consent prior to enrollment.

Data collection and CTRCD definition
Blood tests were performed in all patients pre- and post-anthracycline treatment, and lymphocyte-based immunoinflammatory indices (NLR, MLR, NMLR, SIRI, PLR, SII, PIV) were calculated therefrom. Echocardiographic examinations were performed by a trained sonographer in all patients pre- and post-anthracycline therapy. Following completion of anthracycline therapy, patients underwent serial echocardiographic assessments and myocardial biomarker measurements for CTRCD surveillance, with a median follow-up duration of 27 months. Specifically, we assessed patients’ LVEF after anthracycline therapy completion and evaluated post-treatment cardiac function changes based on LVEF variations. CTRCD was defined as an absolute reduction in LVEF > 10% to below 53% [17]. During the first 12 months after anthracycline treatment, all patients were scheduled for standardized echocardiographic assessments at the 12th month time point - this ensured complete LVEF data were available for every enrolled patient at this time point, and aligns with the recommendations of the 2022 ESC Cardio-Oncology Guidelines for cardiac function evaluation post-anthracycline.

Calculation of lymphocyte-based immunoinflammatory biomarkers
Lymphocyte-based immunoinflammatory biomarkers, including neutrophil to lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), monocyte to lymphocyte ratio (MLR), neutrophil monocyte to lymphocyte ratio (NMLR), systemic immune-inflammation index (SIRI), systemic immune inflammation index (SII) and pan-immune-inflammation value (PIV) were calculated based on the absolute counts of these distinct blood cell subpopulations. The formulas were as follows: NLR = neutrophil count / lymphocyte count; PLR = platelet count / lymphocyte count; MLR = monocyte count / lymphocyte count; NMLR = (neutrophil count + monocyte count) / lymphocyte count; SIRI = (neutrophil count × monocyte count) / lymphocyte count; SII = (platelet count × neutrophil count) / lymphocyte count; PIV= (neutrophil count × monocyte count × platelet count) / lymphocyte count [19, 25, 26].

Statistical analysis
This study conducted statistical analyses using the Statistical Package for Social Sciences (SPSS) v.22. and GraphPad software 8.0.1. Continuous variables conforming to a normal distribution were compared using the t-test, while the Mann-Whitney U-test was used for intergroup comparisons of continuous variables that did not conform to a normal distribution. Categorical variables were presented as counts and percentages (%), and intergroup differences were analyzed using the Pearson’s chi-squared (χ²) test or Fisher’s exact test. Correlations between △LVEF and lymphocyte-based immunoinflammatory biomarkers were analyzed using Pearson or Spearman correlation analysis. Receiver operating characteristic curves were utilized to ascertain the optimal cutoff values. Lymphocyte-based immunoinflammatory biomarkers were subjected to statistical analysis as categorical variables, based on the cut-off values that maximized the Youden index. The cumulative risk of the endpoint over time was graphically presented using Kaplan Meier curve, and the log-rank test was used to compare differences between the two groups. Cox proportional hazards regression models were used to estimate hazard ratio (HR) and 95% confidence interval (CI). This study employed three Cox models, namely Model 1 (unadjusted model), Model 2 (model adjusted for clinical risk factors), and Model 3 (model further adjusted for examination indicators at 3rd month). All statistical analysis were performed two- tailed, and P < 0.05 was regarded as statistically significant.

Results

Baseline characteristics and dynamic changes of core indicators following anthracycline therapy
A total of 98 breast cancer patients receiving anthracycline-based chemotherapy was initially enrolled. Of these, 17 patients were excluded due to exclusion criteria or incomplete data, and 6 were lost to follow-up (Fig S1). Ultimately, 75 breast cancer patients were included in the final analysis, with a median follow-up duration of 27 months. All patients were female with a mean age of 56 ± 11 years. 1 (1.3%) patient had coronary heart disease, 21 (28%) patients had hypertension, 12 patients had diabetes mellitus (16%), and 3 (4%) patients had dyslipidemia. Additionally, 17.4% of patients received cardioprotective medications, including angiotensin-converting enzyme inhibitors (ACEIs), angiotensin II receptor blockers (ARBs), beta‐blockers, and statins. All patients (100%) completed four cycles of anthracycline-containing chemotherapy (Table 1).

Prior studies have established a close association between lymphocyte counts and cardiac function in cancer patients undergoing anthracycline treatment [13, 18]. In the present study, lymphocyte counts exhibited a downward trend within the first 3 months of anthracycline therapy initiation, then gradually recovered to the pre-treatment baseline level from 3rd month to 12th month (Fig. 1). Concurrently, LVEF showed a progressive decline post-anthracycline therapy, with the most significant reduction observed at the 12th month assessment (Fig. 2), indicating that anthracycline-induced cardiac injury is persistent and progressive over time. We further assessed changes in lymphocyte - based immunoinflammatory biomarkers post-anthracycline therapy, and our findings demonstrated significant elevations in NLR, MLR, NMLR, SIRI, SII, PLR, and PIV levels post-treatment compared with baseline values (Fig. 3).

Incidence of CTRCD and intergroup comparisons of lymphocyte-based Immunoinflammatory biomarkers
The incidence of CTRCD was observed in 9 (12%) patients in this study. A comparison between the CTRCD and non-CTRCD groups revealed a significant reduction in LVEF at 12th month post-anthracycline therapy (Table 2). Subsequently, we compared the levels of lymphocyte-based immunoinflammatory biomarkers between the CTRCD and non-CTRCD patients in both pre- and post- anthracycline treatment. Neutrophil and monocyte counts showed a slight increase, while lymphocyte counts exhibited a marked decrease in the CTRCD group relative to the non-CTRCD group in pre-treatment; however, these differences did not reach statistical significance. Similarly, pre-treatment lymphocyte-based immunoinflammatory biomarkers showed a trend of alteration between groups without significant differences. Notably, at post-anthracycline therapy, monocyte counts were significantly higher, and lymphocyte counts were significantly lower in the CTRCD group compared with the non-CTRCD group. Concomitantly, lymphocyte-based immunoinflammatory biomarkers (NLR, MLR, NMLR, SIRI, SII, and PIV) were significantly elevated in the CTRCD group relative to the non-CTRCD group (Table 3).

Predictive significance of lymphocyte-based Immunoinflammatory biomarkers for CTRCD
To evaluate the predictive efficacy of lymphocyte-based immunoinflammatory biomarkers for CTRCD in breast cancer patients, we performed ROC curve analyses pre- and post- anthracycline treatment. The findings indicated that lymphocyte-based immunoinflammatory biomarkers assessed at post-anthracycline therapy exhibited significantly higher predictive accuracy for CTRCD than pre-treatment counterparts. Specifically, the pre-treatment AUC values for SIRI, NLR, MLR, NMLR, PLR, SII, PIV and lymphocyte were 0.734, 0.679, 0.662, 0.700, 0.615, 0.672, 0.554, and 0.612, respectively, and rose to 0.889, 0.768, 0.812, 0.828, 0.710, 0.803, 0.854, and 0.732, respectively, at post-anthracycline therapy (Fig. 4). We further determined the optimal cut-off values for these immunoinflammatory biomarkers, with the optimal cut-off values for post-treatment SIRI, NLR, MLR, NMLR, PLR, SII, and PIV being 3.18, 4.02, 0.82, 5.77, 169.99, 875.9, and 654.9, respectively. Additionally, the Youden indices of lymphocyte-based immunoinflammatory biomarkers were significantly higher post-treatment than pre-treatment, with SIRI exhibited the highest Youden index (sensitivity: 99.9%, specificity: 74.2%) (Table 4).

Correlation between lymphocyte-based Immunoinflammatory biomarkers and cardiac function post-anthracycline therapy
ROC curve analysis demonstrated that lymphocyte-based immunoinflammatory biomarkers post-anthracycline therapy exhibited excellent predictive value for CTRCD, particularly SIRI, MLR, NMLR, SII, and PIV. To further explore the association between these post-treatment indices and cardiac function, we analyzed the correlation between post-anthracycline therapy lymphocyte-based immunoinflammatory biomarkers and ΔLVEF (ΔLVEF, defined as baseline LVEF minus LVEF at 12th month). Consistent with previous study, lymphocyte levels were negatively correlated with ΔLVEF. Our findings also revealed a significant positive correlation between SIRI, MLR, NMLR, SII, and PIV levels and ΔLVEF, with correlation coefficients (r values) of 0.594 (P < 0.001), 0.517 (P < 0.001), 0.552 (P < 0.001), 0.415 (P < 0.001), and 0.559 (P < 0.001), respectively (Fig. 5).

CTRCD-free survival and risk association of post-treatment lymphocyte-based Immunoinflammatory biomarkers for CTRCD
Kaplan–Meier survival analysis revealed that a lymphocyte count < 0.78 was significantly associated with an increased risk of CTRCD (log-rank test, p = 0.004). Additionally, breast cancer patients with SIRI ≥ 3.178 (log-rank test, p < 0.001), MLR ≥ 0.823 (log-rank test, p < 0.001), NMLR ≥ 5.767 (log-rank test, p < 0.001), SII ≥ 876 (log-rank test, p = 0.001), or PIV ≥ 655 (log-rank test, p = 0.001) had a significantly higher risk of CTRCD (Fig. 6).

Univariable Cox regression analysis was performed for lymphocytes count, MLR, NMLR, SIRI, SII, and PIV (assessed at post-anthracycline therapy), and lymphocytes count, MLR, NMLR, SIRI, and PIV showed a significant association with an increased risk of CTRCD (Table 5, model 1). Multivariable Cox regression analyses were conducted after adjustment for age, BMI, heart rate, SBP, DBP, CHD, hypertension, dyslipidemia, diabetes, baseline statin use, baseline ACEI/ARB use, baseline beta-blocker use. After adjustment, the associations of MLR (HR = 5.876, 95% CI = 1.450–23.822, p = 0.013), SIRI (HR = 1.601, 95% CI = 1.090–2.351, p = 0.016) and PIV (HR = 1.003, 95% CI = 1.001–1.005, p = 0.001) remained significant, while lymphocytes count, SII, and NMLR were not significantly associated with CTRCD (Table 5, model 2). Similar results were observed after further adjustment for laboratory parameters (ALT, AST, Cr, Hb, UA, FBG, cTnT) and LVEF at 3rd month (MLR: HR = 18.747, 95% CI = 1.871-187.799, p = 0.013; SIRI: HR = 2.877, 95% CI = 1.243–6.659, p = 0.014; PIV: HR = 1.007, 95% CI = 1.002–1.012, p = 0.011)) (Table 5, model 3).

Discussions
The core findings of this study are highly consistent with existing evidence in the field, further validating the critical role of immunoinflammatory responses in anthracycline-induced cardiac injury, while addressing the limitations of prior studies related to temporal assessment and biomarker scope. Lymphocyte-based immunoinflammatory biomarkers reflect systemic inflammatory status and immune dysfunction, with NLR and MLR having been well established to correlate with CTRCD. Building on this, the present study not only expanded the panel of inflammatory indices investigated (including NMLR, SIRI, SII, and PIV) but also identified the early warning value of lymphocyte-based immunoinflammatory biomarkers at post-anthracycline therapy. Levels of these biomarkers at post-treatment were significantly higher than pre-treatment, addressing the limitation of previous studies that focused solely on baseline biomarkers and failed to reflect dynamic immunoinflammatory changes during treatment. Furthermore, SIRI, MLR, NMLR, SII, and PIV were significantly positively correlated with 12th month ΔLVEF. Using multivariable Cox regression models adjusted for clinical risk factors, this study confirmed that MLR, SIRI, and PIV are robust independent predictors of CTRCD. The high specificity and sensitivity of post-treatment MLR and SIRI further support their clinical applicability, offering a novel solution to the limitation of traditional biomarkers (cTnT, NP), which only reflect established occurred damage.
The present study focuses on lymphocyte-based immunoinflammatory biomarkers, which offer the advantages of simplicity, low cost, and wide accessibility, endowing them with high clinical practicality. Furthermore, our findings demonstrate that lymphocytes exhibit a dynamic trend post- anthracycline treatment, with these changes preceding alterations in LVEF and peaking shortly after therapy completion. This temporal relationship enables early warning of cardiotoxicity in patients immediately post-treatment, facilitating the implementation of targeted cardioprotective therapy and reduction of cardiovascular risk. Notably, multiple post-treatment lymphocyte-based immunoinflammatory biomarkers showed high predictive value for CTRCD, with AUCs for SIRI, MLR, NMLR, SII, and PIV all exceeding 0.8. We also established optimal cut-off values for these biomarkers, and our results confirm their good sensitivity, supporting their utility as early screening and warning markers for breast cancer patients undergoing anthracycline therapy.
Accumulating evidence highlights the pivotal role of lymphocyte-based immunoinflammatory biomarkers in the pathogenesis and prognosis of a broad spectrum of cardiovascular diseases. In a prospective cohort study, MLR correlated strongly with established heart failure biomarkers and independently predicted heart failure-related hospitalizations in patients with coronary artery disease [27]. Similarly, NLR has emerged as an independent predictor of major adverse cardiovascular events [28]. Expanding these observations, Biolo et al. reported that NLR, MLR, and SII as predictor of CTRCD in breast cancer patients [13]. Consistent with their results, we found post-treatment NLR, MLR and SII levels were higher in CTRCD patients receiving anthracycline-based chemotherapy, confirming their predictive value for CTRCD. The optimal cut-off values reported in their study (2.6 for NLR, 0.26 for MLR, 589.2 for SII) are also comparable to those established in the present study (2.29 for NLR, 0.51 for MLR, 432.12 for SII). Furthermore, our study identified MLR, PIV and SIRI as novel independent predictors of CTRCD, with superior predictive performance both pre-treatment and at post-anthracycline therapy.
MLR, SIRI, and PIV are all composite biomarkers that integrate the levels of two or more immune cells (neutrophils, monocytes, lymphocytes, platelets) closely involved in the anthracycline-induced inflammatory response, rather than a single cell type index [29–32]. Besides, these three biomarkers target the key immune cell subsets and are thought to be mediated by the amplification of oxidative stress, induction of cytokine storms, impairment of endothelial function, and subsequent promotion of tissue fibrosis in cardiac injury [33, 34]. Collectively, our findings suggests that MLR, SIRI, and PIV provide valuable insights into inflammatory-immune balance and represent potential novel biomarkers for the prevention and early detection of CTRCD.
Another key finding of this study is that post-treatment lymphocyte-based immunoinflammatory biomarkers outperform their pre-treatment counterparts in predicting CTRCD, which can be explained by the inflammatory cascade triggered by anthracycline and the dynamic remodeling of the inflammatory - immune microenvironment during chemotherapy. Anthracycline-induced cardiotoxicity is intrinsically linked to a dysregulated inflammatory response, initiated by excessive reactive oxygen species (ROS) production in cardiomyocytes [35]. ROS can directly modulate and activate the NLRP3 inflammasome, and mitochondrial damage further promotes NLRP3 activation by releasing mitochondrial DNA into the cytoplasm [17, 36]. Additionally, ROS activates IκB kinase, leading to phosphorylation and degradation of NF-κB inhibitors. This degradation enables NF-κB nuclear translocation, where it induces the expression of pro-inflammatory genes such as TNF-α and IL-6, as well as chemokines and adhesion molecules [35]. Anthracyclines also promote cardiac inflammation through various other pathways, including endoplasmic reticulum stress and autophagy dysregulation, which drive cardiac function deterioration and exacerbate myocardial fibrosis, ultimately leading to the progression of cardiac dysfunction [37, 38]. Importantly, this inflammatory cascade is not pre-existent but is dynamically amplified by doxorubicin exposure, thus, pre-treatment lymphocyte-based immunoinflammatory biomarkers merely reflect baseline systemic inflammation, rather than the drug-specific pro-inflammatory milieu that directly contributes to cardiotoxicity. In contrast, post-treatment of NLR, MLR, NMLR, SIRI, PLR, SII, and PIV levels are quantitative reflections of the anthracycline-induced pro-inflammatory shift in patients. This explains why the AUCs of post-treatment lymphocyte-based immunoinflammatory biomarkers for assessing CTRCD risk are significantly higher than that those of their pre-treatment counterparts, as post-treatment levels accurately reflect anthracycline-induced inflammatory activity, enabling more precise assessment of immunoinflammatory status during treatment and accurate prediction of CTRCD risk. Notably, post-treatment lymphocyte-based immunoinflammatory biomarkers are closely associated with cardiac function. As demonstrated by the correlation analysis, higher levels of post-treatment MLR, SIRI, SII, PIV are associated with a greater 12th month LVEF reduction (ΔLVEF), reflecting the direct promotional effect of anthracycline-induced inflammatory imbalance on cardiac dysfunction.
Despite the valuable insights of this study regarding the predictive utility of lymphocyte-derived immunoinflammatory biomarkers for anthracycline-induced CTRCD in breast cancer patients, several limitations should be noted. First, the retrospective study design and relatively small sample size limit the generalizability of our findings. As a single-center study, our results may also be subject to selection bias. Second, this study focused exclusively on lymphocyte-based immunoinflammatory biomarkers, without including immune cell subsets, cytokines, and other inflammatory indicators, precluding a direct comparison of the predictive value of these biomarkers with other inflammatory markers for CTRCD. Third, the small sample size also prevents us from clarifying the applicability of the established optimal cut-off values across different patient subpopulations. In summary, while our findings have promising clinical implications, future prospective, large-sample, multi-center studies are needed to fully validate the clinical value of these immunoinflammatory biomarkers for CTRCD.

Conclusion
Lymphocyte-based immunoinflammatory biomarkers have predictive value for CTRCD in breast cancer patients undergoing anthracycline-based chemotherapy, with their predictive efficacy significantly enhanced at post-anthracycline therapy. Among these biomarkers, MLR, SIRI, and PIV exhibit robust predictive performance and serve as independent predictors of anthracycline-induced CTRCD.

Supplementary Information