ABSTRACT
Background
Growth differentiation factor‐15 (GDF‐15) is upregulated in coronary artery disease (CAD). The relationship of intensive statin therapy with circulating GDF‐15 levels has not been explored yet. In the current secondary analysis of an observational cohort, we investigated whether high‐intensity statin treatment is associated with GDF‐15 in patients with advanced CAD.
Methods
In 102 CAD patients we measured GDF‐15 levels before and after median 7 [6–8] months from initiation atorvastatin 80 mg/day (n = 63, 61.8%) or rosuvastatin 40 mg/day (n = 39, 38.2%). At the two time points we also determined C‐reactive protein (CRP), thrombin generation and fibrinolysis inhibitors. There was no concurrent control group.
Results
At baseline GDF‐15 levels (median 942; interquartile range 639–1284 pg/mL) were associated solely with age (R = 0.567, p < 0.001) and CRP (R = 0.464, p < 0.001), but not with lipid profile. On high‐intensity statin therapy GDF‐15 reduction by 17.8% was observed (p = 0.006), which was associated with CRP lowering by 46.4% (R = 0.788, p < 0.001), but not with changes in total cholesterol (by 17.6%) and low‐density lipoprotein cholesterol (LDL‐C) (by 28.1%). However, GDF‐15 levels were lower by 28.5% (p < 0.001) and 18.9% (p = 0.013) in 43 patients (42.2%) who achieved the target LDL‐C < 1.8 mmol/L and 19 (18.6%) with LDL‐C < 1.4 mmol/L, respectively, as compared to the remainder. The type of statin was not associated with follow‐up GDF‐15. Age (p = 0.029), baseline GDF‐15 (p < 0.001) and CRP (p < 0.001), but not LDL‐C, were independently associated with follow‐up GDF‐15.
Conclusions
We demonstrated that high‐dose statin therapy is associated with lower GDF‐15 in CAD patients and this effect is largely related to the anti‐inflammatory properties of statins.
Keywords: coronary artery disease, growth differentiation factor‐15, lipids, statin
We have demonstrated for the first time that in patients with advanced CAD the high‐intensity statin treatment is associated with reduced GDF‐15 levels. This effect of statins has not been associated with the cholesterol‐lowering effect, but with anti‐inflammatory action.
1. Introduction
Growth differentiation factor‐15 (GDF‐15), a cytokine in the transforming growth factor‐β (TGF‐β) family, is involved in multiple physiological (energy balance, appetite regulation, pregnancy) [1] and pathological conditions [2]. GDF‐15 expression is upregulated in response to a wide spectrum of stress stimuli including hypoxia, inflammation, cancer, along with ageing [3].
Growing evidence indicates that GDF‐15 plays a role in cardiovascular diseases, including atrial fibrillation and coronary artery disease (CAD) and it is considered as a useful biomarker for cardiovascular diseases which captures additional aspects of fatal outcomes not reflected by clinical risk factors and other biomarkers [4]. GDF‐15 is overexpressed in cardiomyocytes after ischemic event and reperfusion after myocardial infarction (MI) [3]. It has been recently demonstrated that elevated GDF‐15 predicts mortality and arrhythmic complications in acute MI [5]. Moreover, Schlittenhardt et al. have shown that GDF‐15 is expressed by atherosclerotic plaques [6]. In the meta‐analysis conducted by Zhang et al. serum GDF‐15 concentration on admission was the independent predictor of all‐cause death (hazard ratio [HR] 2.24, 95% confidence interval [CI] 1.95–2.57), cardiovascular death (HR 2.00, 95% CI 1.66–2.42) and MI (HR 1.42, 95% CI 1.21–1.66) in patients with CAD [7].
Statins, potent cholesterol‐lowering agents with well‐established clinical value, display multiple pleiotropic effects, mostly independent of cholesterol reduction, including attenuated inflammatory response [8] through diverse mechanisms, including downregulation of proinflammatory gene expression and leukocyte recruitment [9]. In the randomised placebo‐controlled JUPITER study conducted in 17,802 apparently healthy patients without hyperlipidemia but with elevated C‐reactive protein (CRP), rosuvastatin 20 mg/day led to 37% CRP reduction, which was associated with significant reduction of the major cardiovascular events incidence during median 1.9 years of follow‐up [10].
To our knowledge, the relationship of high‐intensity statin treatment with GDF‐15 levels in patients with CAD has not been reported yet. In the analysis from PROVE IT‐TIMI 22 in patients treated with pravastatin 40 mg/day and atorvastatin 80 mg/day after acute coronary syndrome there was no reduction in GDF‐15 at 4 months of treatment [11]. In a prospective randomised trial Kim et al. did not observe any association of GDF‐15 with 8‐weeks atorvastatin 10 or 40 mg/day treatment in patients with diabetes and dyslipidemia but without CAD [12]. Tentzeris et al. have reported that elevated GDF‐15 levels could be a useful predictor of the effectiveness of statin therapy after percutaneous coronary intervention in acute coronary syndrome [13]. In that analysis 80.3% of the included patients were treated with high‐intensity statins [13]. Recently, in the subanalysis of the ODYSSEY OUTCOMES trial conducted in patients after acute coronary syndrome who were treated with statin therapy at a high‐intensity dose or at the maximum tolerated dose, a significant reduction in GDF‐15 by a PCSK9 inhibitor alirocumab was demonstrated, which was associated with lower all‐cause mortality [14].
Therefore, in the current analysis we sought to assess whether high‐intensity statin treatment is associated with GDF‐15 levels in CAD patients and if any changes in this biomarker are linked with alterations to lipid profile, inflammatory or prothrombotic markers.
2. Methods
In this secondary analysis of an observational cohort, we assessed 130 consecutive stable CAD patients with documented advanced CAD (prior MI or coronary artery bypass graft surgery). The characteristics of the study group were described in detail previously [8, 15]. Since 28 patients (21.5%) had no available serum samples to measure GDF‐15, we evaluated in this report 102 patients from the former cohort. In short, the exclusion criteria were combined lipid‐lowering treatment with ezetimibe or fibrates, MI or percutaneous coronary intervention in the last 6 months, end‐stage renal failure, active cancer, acute infection and known autoimmune disease.
All the included patients were initially treated suboptimally with statins (simvastatin 20 mg/day, atorvastatin 10 or 20 mg/day or rosuvastatin 10 mg/day) [16]. They did not meet the dyslipidaemia therapeutic goals, that is low‐density lipoprotein cholesterol (LDL‐C) < 1.8 mmol/L or > 50% reduction in LDL‐C, if baseline level was 1.8–3.5 mmol/L based on the European guidelines [17, 18]. After that, they were switched to high‐intensity statin treatment at the discretion of the investigators, that is, to atorvastatin 80 mg/day (n = 63, 61.8%) or rosuvastatin 40 mg/day (n = 39, 38.2%) as presented previously [8, 15]. With intensive regimen, they were treated for 6–12 months before blood collection [17, 18, 19]. There was no concurrent control group.
The study protocol was approved by the Jagiellonian University Medical College Ethical Committee (1072.6120.59.2018) and the study was conducted in accordance with the Declaration of Helsinki. All participants provided informed written consent. During the preparation of this work, the authors did not use any artificial intelligence‐assisted technology.
2.1. Laboratory Tests
At baseline and after high‐intensity statin treatment, blood was drawn from the antecubital vein into citrated tubes after overnight fasting and centrifuged at 2500g at 20°C for 20 min or into serum tubes and centrifuged at 1600g at 4°C for 10 min.
Serum total cholesterol (TC), LDL‐C, high‐density lipoprotein cholesterol (HDL‐C) and triglycerides (TGs) were assayed in a standard manner with the biochemical analyser Cobas 6000 (Roche Diagnostics GmbH, Mannheim, Germany) [20]. CRP was measured by immunoturbidimetry (Roche Diagnostics GmbH, Mannheim, Germany) and d‐dimer with the Innovance d‐dimer assay (Siemens Healthcare, Marburg, Germany), while fibrinogen was measured with the von Clauss method [21].
Regarding prothrombotic/fibrinolytic markers, we determined thrombin activatable fibrinolysis inhibitor (TAFI) activity (Hyphen‐Biomed, Neuville‐Sur‐Oise, France) with dedicated assay [21]. Clot lysis time (CLT) was measured as previously described [21]. Plasma thrombogenic potential was assessed using calibrated automated thrombography with the use of tissue factor at a final concentration of 5 pM (CAT; Thrombinoscope BV, Maastricht, the Netherlands), as described previously [21]. Endogenous thrombin potential (ETP) was calculated as the area under the curve [21].
2.2. GDF‐15 Analysis
Serum GDF‐15 levels were assessed with enzyme‐linked immunosorbent assay (R&D Systems, Minneapolis, Minnesota, United States; assay range: 23.4–1500 pg/mL) [22].
2.3. Statistical Analysis
Our study was powered to have an 80% chance to detect a 20% reduction in GDF‐15 with a p‐value < 0.05. Based on the previous data [8, 12, 15] at least 82 patients were required. The variables were presented as numbers and percentages or median and interquartile range (IQR). Normality was assessed by Shapiro–Wilk test. Differences between the groups were compared with the Student's t‐test or the Mann–Whitney U‐test depending on the distribution normality. In case of dependent variables for within‐subject pre–post‐comparisons, paired Student's t‐test or the sign test was applied for normal or non‐normal data distribution, respectively. Categorical variables were compared by the Pearson's χ 2 test or Fisher's exact test. Associations between nonparametric variables were assessed by Spearman's rank correlation coefficient. ANCOVA was used to determine statin influence on GDF‐15 at follow‐up. All independent baseline variables potentially associated with GDF at follow‐up or with the absolute change of GDF‐15 following high‐intensity statin treatment were included in the multivariable linear regression model. The selection of variables was based on the results in the univariable model, literature and to avoid collinearity. A two‐sided p < 0.05 was considered statistically significant. All statistical analyses were performed using STATISTICA software Version 13.3 (StatSoft, Krakow, Poland).
3. Results
3.1. At Baseline
A total of 102 patients with CAD aged 63 (59–69) were evaluated, including 43.1% men. There were 75 individuals with prior non–ST‐segment elevation MI (NSTEMI) or STEMI (73.5%) or 27 with a history of coronary artery bypass grafting surgery (CABG) (26.5%). The study group was characterised by typical distribution of cardiovascular risk factors (Table 1). None of the analysed patients were treated with dual antiplatelet therapy or anti‐inflammatory agents, such as colchicine or corticosteroids. Median LDL‐C was 3.2 (2.8–3.8) mmol/L and 63 patients (61.8%) had levels above 3 mmol/L. Patients in whom GDF‐15 was not determined (n = 28) were older, more often male, diagnosed with peripheral arterial disease, but were less often treated with beta‐blockers as compared to the analysed group (Table S1).
TABLE 1.
Baseline clinical characteristics.
| Age, years | 63 (59–69) |
| Male, n (%) | 44 (43.1) |
| BMI, kg/m2 | 28.5 (24.6–31.9) |
| Comorbidities, n (%) | |
| Hypertension | 60 (58.8) |
| Diabetes mellitus | 29 (28.4) |
| Smoking | 34 (33.3) |
| Previous MI | 75 (73.5) |
| Previous CABG | 27 (26.5) |
| Heart failure | 22 (21.6) |
| Peripheral arterial disease | 21 (20.6) |
| Medications, n (%) | |
| Aspirin | 91 (89.2) |
| Beta‐blocker | 50 (49.0) |
| ACE inhibitor | 37 (36.3) |
Abbreviations: ACE, angiotensin converting enzyme; BMI, body mass indexCABG, coronary artery bypass graft; MI, myocardial infarction.
Median GDF‐15 level was 942 (639–1284) pg/mL (Table 1). Patients with GDF‐15 level above a median of 942 pg/mL were older compared with the remainder (p = 0.001), but the two groups did not differ in sex, BMI or the prevalence of cardiovascular risk factors (Table S2). GDF‐15 was positively correlated with age (R = 0.567, p < 0.001; Figure 1A), but not with sex, BMI, or comorbidities. At baseline, GDF‐15 levels were not associated with lipid profile. However, GDF‐15 was positively correlated with CRP (R = 0.464, p < 0.001; Figure 1B) and patients with GDF‐15 level above median had 27.3% higher CRP concentrations (p < 0.001, Table S2). CRP was associated with fibrinogen (R = 0.313, p < 0.001). No significant associations were noted for other tested laboratory variables (Table S2).
FIGURE 1.
Linear associations of GDF‐15 with age and CRP at baseline (A, B), the change of GDF‐15 during high‐intensity statin treatment and change of TAFI activity and CRP (C, D) and GDF‐15, LDL‐C and CRP levels on high‐intensity statin therapy (E, F). CRP, C‐reactive protein; GDF‐15, growth differentiation factor‐15; LDL‐C, low‐density lipoprotein cholesterol; TAFI, thrombin activatable fibrinolysis inhibitor.
3.2. On High‐Intensity Statin Treatment
After 7 [6, 7, 8] months of intensive statin treatment, significant reductions in TC (by 17.6%, p < 0.001), LDL‐C (by 28.1%, p < 0.001) and HDL‐C (by 13.3%, p < 0.001) were observed without difference in TG (p = 0.62) (Table 2). Moreover, CRP (by 46.4%, p < 0.001), fibrinogen (by 19.4%, p < 0.001), d‐dimer (by 25.5%, p < 0.001), CLT (by 15.6%, p < 0.001) and TAFI activity (by 9%, p < 0.001) were also substantially decreased (Table 2). After statin regimen intensification, ETP was slightly decreased (by 2.8%, p = 0.024) (Table 2).
TABLE 2.
Laboratory parameters before and after high‐intensity statin treatment.
| Patients before high‐dose statin treatment (n = 102) | Patients on high‐dose statin treatment (n = 102) | p | |
|---|---|---|---|
| Basic laboratory investigations | |||
| INR | 1 (0.97–1.06) | 1 (0.97–1.06) | 0.88 |
| APTT, s | 27 (25–30) | 27 (25–31) | 0.65 |
| Haemoglobin, g/dL | 14.1 (13.2–15.2) | 13.8 (13.1–14.7) | 0.17 |
| White blood cells, 103/μL | 6.4 (5–7.4) | 6.1 (5.2–7.4) | 0.66 |
| Platelets, 103/μL | 244 (206–296) | 234 (200–286) | 0.35 |
| ALT, IU/L | 20 (14–33) | 23 (16–30) | 0.29 |
| Creatinine, μmol/L | 70 (64–80) | 70 (62–79) | 0.57 |
| GFR, mL/min/1.73 m2 | 92 (76–101) | 93 (77–102) | 0.60 |
| Glucose, mmol/L | 5.3 (5–5.9) | 5.2 (4.9–5.6) | 0.025 |
| CRP, mg/L | 2.8 (2.1–3.7) | 1.5 (1.2–1.9) | < 0.001 |
| Lipid profile, mmol/L | |||
| Total cholesterol | 5.1 (4.5–5.7) | 4.2 (3.2–5.1) | < 0.001 |
| LDL‐C | 3.2 (2.8–3.8) | 2.3 (1.5–3.3) | < 0.001 |
| HDL‐C | 1.5 (1.3–1.9) | 1.3 (1.1–1.7) | < 0.001 |
| Triglycerides | 1.1 (0.8–1.4) | 1 (0.8–1.5) | 0.62 |
| Coagulation variables | |||
| Fibrinogen, g/L | 3.1 (2.4–3.7) | 2.5 (2.4–3) | < 0.001 |
| d‐dimer, ng/mL | 306 (233–411) | 228 (200–271) | < 0.001 |
| ETP, nM × min | 1580 (1507–1675) | 1536 (1467–1670) | 0.024 |
| CLT, min | 90 (79–102) | 76 (65–89) | < 0.001 |
| TAFI activity, % | 100 (90–112) | 91 (85–99) | < 0.001 |
| Other parameters | |||
| GDF‐15, pg/mL | 942 (639–1284) | 774 (582–1014) | 0.006 |
Note: p values < 0.05 were bolded.
Abbreviations: ALT, alanine aminotransferase; APTT, activated partial thromboplastin time; CLT, clot lysis time; CRP, C‐reactive protein; ETP, endogenous thrombin potential; GDF‐15, growth differentiation factor‐15; GFR, glomerular filtration rate; HDL‐C, high‐density lipoprotein cholesterol; INR, international normalised ratio; LDL‐C, low‐density lipoprotein cholesterol; TAFI, thrombin activatable fibrinolysis inhibitor.
The high‐intensity statin treatment was associated with marked reduction in GDF‐15 levels (by 17.8%, p = 0.006) (Table 2, Figure 2A). Of note, ΔGDF‐15 was not linked with any changes in lipid variables or their concentrations both at baseline and on high‐dose statins. ΔGDF‐15 was associated closely with ΔCRP (R = 0.788, p < 0.001) and slightly with ΔTAFI activity (R = 0.196, p = 0.048) (Figure 1C,D). No such association was found for CLT and ETP.
FIGURE 2.
Reduction in GDF‐15 levels observed during intensive statin treatment (A) and after atorvastatin or rosuvastatin treatment (B). GDF‐15, growth differentiation factor‐15.
GDF‐15 levels were lower by 28.5% (p < 0.001) in patients who achieved the target LDL‐C value < 1.8 mmol/L (n = 43, 42.2%). They were also lower in patients with LDL‐C < 1.4 mmol/L (n = 19, 18.6%) compared to the remainder (by 18.9%, p = 0.013; Figure 3). Moreover, a non‐significant trend towards a greater reduction of GDF‐15 in patients treated with atorvastatin 80 mg/day in comparison to rosuvastatin was noted though LDL‐C and CRP on therapy and their fall were similar in both subgroups (p = 0.06) (Figure 2B). Patients treated with atorvastatin and those on rosuvastatin did not differ in terms of baseline clinical characteristics (Table S3).
FIGURE 3.
GDF‐15 levels in patients achieving target LDL‐C levels. GDF‐15, growth differentiation factor‐15.
On high‐intensity statin treatment in the CAD patients, residual GDF‐15 levels were positively correlated with TC (R = 0.364, p < 0.001), LDL‐C (R = 0.381, p < 0.001; Figure 1E), TG (R = 0.213, p = 0.031), along with CRP (R = 0.466, p < 0.001; Figure 1F) and fibrinogen (R = 0.436, p < 0.001).
The type of statins (atorvastatin versus rosuvastatin) did not influence GDF‐15 at follow‐up (Table S4). By the multivariable linear regression (R 2 = 0.751, p < 0.001), the level of GDF‐15 at follow‐up after high‐intensity statin treatment was independently associated with age (β = 0.146, p = 0.029), baseline GDF‐15 (β = 0.865, p < 0.001) and CRP (β = −0.383, p < 0.001), but not with LDL‐C (Table 3).
TABLE 3.
Multivariable linear regression with independent predictors of GDF‐15 at follow‐up and of a change of GDF‐15 during high‐dose statin therapy.
| GDF‐15 at follow‐up, R 2 = 0.751 | Change of GDF‐15, R 2 = 0.565 | |||||
|---|---|---|---|---|---|---|
| β | 95% CI for β | p | β | 95% CI for β | p | |
| Atorvastatin vs. Rosuvastatin | 0.031 | −0.063 to 0.126 | 0.560 | −0.040 | −0.161 to 0.098 | 0.559 |
| Age | 0.146 | 0.016–0.304 | 0.029 | −0.193 | −0.364 to −0.022 | 0.029 |
| Baseline GDF‐15 | 0.865 | 0.741–0.985 | < 0.001 | 0.418 | 0.285–0.551 | < 0.001 |
| Baseline CRP | −0.383 | −0.604 to −0.165 | < 0.001 | 0.507 | 0.386–0.630 | < 0.001 |
| Baseline LDL‐C | −0.021 | −0.130 to 0.095 | 0.698 | 0.027 | −0.084 to 0.150 | 0.698 |
| Baseline fibrinogen | 0.026 | −0.083 to 0.134 | 0.612 | −0.035 | −0.156 to 0.096 | 0.612 |
Note: p values < 0.05 were bolded.
Abbreviations: CI, confidence interval; CRP, C‐reactive protein; GDF‐15, growth differentiation factor‐15; LDL‐C, low‐density lipoprotein cholesterol.
4. Discussion
In the current study we have demonstrated that in patients with CAD the high‐intensity statin treatment is associated with reduced GDF‐15 levels. However, that observation should be interpreted with caution due to the non‐randomised, pre–post‐study design. Moreover, this effect of statins has not been associated with the cholesterol‐lowering effect, but with anti‐inflammatory action. Nevertheless CAD patients who achieved the target LDL‐C of 1.8 or 1.4 mmol/L had significantly lower GDF‐15 as compared to the remainder. Together with previous studies, our findings provide additional arguments for the use of statins at appropriate daily doses and achieving cholesterol treatment goals in clinical practice.
The key novel observation is that high‐intensity statin therapy is associated with reduced GDF‐15 levels by 17.8% in advanced CAD patients already on lower daily doses of statins who did not reach the target LDL‐C. As mentioned above in the subanalysis from PROVE IT‐TIMI 22 study lower GDF‐15 levels were not noted in patients after acute coronary syndrome treated with statins [11]. In comparison with that study the current study group were characterised with more advanced CAD and were treated with more potent statins for a longer time period which probably makes it possible to observe these differences [11]. Kim et al. assessed 10 or 40 mg/day of atorvastatin in patients with only cardiovascular risk factors (mean LDL‐C at baseline 82.7 ± 18.8 mg/dL) and CAD was not reported [12]. Interestingly GDF‐15 levels at baseline and after statin treatment were significantly higher than in our study, which may be related to the different enzyme‐linked immunosorbent assay used for GDF‐15 determination [12]. Moreover, statins were used for only 8 weeks [12], in contrast to more than 6 months in the current report. Kim et al. did not assess whether or not the LDL‐C treatment goals were achieved [12], while we addressed this key issue which enhances practical implications of our study.
There is evidence supporting the involvement of GDF‐15 in atherogenesis and its clinical manifestations [3]. GDF‐15 is inducible locally in atherosclerotic plaques by activated macrophages [6], which was confirmed by Ackermann et al., who showed that GDF‐15 regulates lipid homeostasis and autophagy in human macrophages [23]. This may lead to increased oxidative stress and to atherosclerotic plaque instability. In our current study GDF‐15 was not associated with cholesterol or TGs at baseline. ΔGDF‐15 was also not associated with baseline lipid profile, also in multivariable analysis. However, the present observations should be interpreted with caution. In the literature there are conflicting data regarding the association between GDF‐15 and cholesterol levels. In a recent article by Girona et al. GDF‐15 levels were positively associated with very low‐density lipoprotein cholesterol (VLDL‐C) and LDL‐C and inversely associated with HDL‐C in 156 patients with metabolic‐associated liver injury and atherosclerotic cardiovascular disease [24]. A similar positive relationship of VLDL‐C and TG with GDF‐15 levels was observed in patients who undergone caloric deprivation [25]. In turn, in patients infected with human immunodeficiency virus (HIV), GDF‐15 levels were positively correlated with age and negatively associated with LDL‐C levels [26]. The cohort analysed in the present study was treated previously sub‐optimally with statins. Therefore, we did not observe the LDL‐C reduction typical of statin‐naïve patients [17]. Our findings suggest that in advanced CAD during statin use, the interplay between GDF‐15 expression and cholesterol metabolism is complex as reflected by a significant difference in GDF‐15 related to achieving the target LDL‐C. It might be speculated that specific clinical conditions (e.g., CAD, HIV infection, caloric deprivation etc.) and their molecular mechanisms modulate the complex interplay between lipid metabolism and GDF‐15 expression, which deserves further research. Further studies in which evaluation of GDF‐15 levels will be combined with other metabolic factors should be performed to show association of diverse pathways involved in inflammation, oxidative stress and lipid metabolism. Noteworthy, none of the patients developed HF diagnosis during the follow‐up, which is important due to a close association of HF with GDF‐15 levels reported previously [19].
The association of GDF‐15 with CRP reported here is of paramount importance. GDF‐15 has been shown to play both pro‐inflammatory and anti‐inflammatory roles depending on the context [2]. In most instances, GDF‐15 induces a tissue protective programme during inflammation [3]. However, under severe or chronic disease conditions, such as CAD, GDF‐15 triggered responses that might contribute to their pathogenesis [3]. In our study, baseline CRP is independently associated with ΔGDF‐15. Moreover, GDF‐15 and CRP were positively associated also at baseline before high‐intensity statin treatment. A decrease of CRP during high‐intensity statin treatment is well‐documented [10] given evidence for a key role of interleukin‐6 as a primary mediator of CRP synthesis, promoting the transcription of the CRP gene. Moreover, other inflammatory cytokines such as IL‐1β and TNF‐α can also contribute to CRP production [10]. And it has been suggested that statins can attenuate their pathways [10]. Abulizi et al. observed that deficiency of GDF‐15 induces more severe cardiac and renal injury during sepsis due to increased expression of TNF‐α and IL‐6 [27]. GDF‐15 can suppress TNF‐α production by macrophages through the inhibition of NF‐κB signalling [28]. Moreover, GDF‐15 also exerts a suppressive effect on IL‐1β responses by decreasing expression of key inflammasome components [29]. In that aspect, our conclusions regarding the non‐significant trend towards a lower GDF‐15 in patients treated with atorvastatin in comparison to rosuvastatin are interesting. However, we did not observe the difference in CRP reduction between both statins. Due to the small number of patients treated with rosuvastatin and atorvastatin, our conclusions should be treated with extreme caution, mainly as an underpowered exploratory finding and further studies are needed to compare the magnitude of the effects on GDF‐15 produced by diverse statins at varying daily doses.
It is known that high‐intensity statin treatment exerts beneficial antithrombotic effects as supported also by the previous findings in this cohort [8, 15]. In the context of GDF‐15, our CAD patients with higher GDF‐15 did not show any differences in ETP and fibrinolytic markers. Most likely, advanced CAD with low‐grade prothrombotic state and moderate GDF‐15 levels (as compared to acute cardiovascular conditions) differs from, for instance, post‐pulmonary embolism syndrome, where we observed weak correlations of GDF‐15 with ETP at baseline (R = 0.19, p = 0.032) and with CLT at baseline (R = 0.21, p = 0.018) and after 3 months follow‐up (R = 0.23, p = 0.022) [22]. Potential links of elevated GDF‐15 with haemostasis in CAD should be explored in future studies.
The results from the current analysis may be a preliminary and very speculative basis for further work investigating the prognostic role of pharmacological GDF‐15 lowering. The alternative explanations of the observed associations including regression to the mean, temporal effects, lifestyle modifications, changes in adherence or other unmeasured confounding factors cannot be excluded in the absence of a concurrent comparator group. There have not been effective pharmacological methods for lowering GDF‐15 yet. In this context, the data on ponsegromab, the first humanised monoclonal antibody inhibiting GDF‐15, are very interesting [1, 30]. As has been demonstrated in the randomised trial published in recent years, 12‐week treatment with ponsegromab resulted in increased weight gain and overall activity level in patients with cancer cachexia [30]. There are also similar promising initial data on heart failure and cardiac cachexia [1]. It is tempting to speculate that combined use of statins and ponsegromab might be evaluated in future randomised studies in patients with CAD.
Our current study has several limitations. First, the number of study participants was limited though representative for advanced CAD. However, it was sufficiently powered. Second, a lower dose statin control group was not included since it was considered ethically inappropriate in high‐risk CAD patients. In this study, there was no well‐matched healthy cohort. Third, since we studied patients with advanced CAD after acute MI or CABG, the current results do not likely apply to the entire CAD population [18]. Since no acute ischemia was observed between the two time points, we cannot address the issue of potential clinical relevance of changes in GDF‐15 in CAD. Statin adherence remains a challenge in everyday practice and observational studies. The compliance assessment in the current study was based on patient declarations, which may undoubtedly involve partial error. Since GDF‐15 measurements can be performed using diverse methodology, some effects of the assay used cannot be excluded [12]. GDF‐15 levels before initiation of statin use were not available, but most patients started statin therapy following acute coronary syndrome or PCI, the conditions known to affect GDF‐15 in blood [5]. Finally, in vitro and animal studies should be performed to understand the mechanisms underlying statin‐associated GDF‐15 reduction.
5. Conclusions
We demonstrated that high‐intensity statin therapy is associated with reduced circulating GDF‐15 levels in CAD patients, which is largely related to the anti‐inflammatory effect. It should be established in larger studies whether the observed association between high‐intensity statin therapy and lower GDF‐15 translates into additional clinical benefits in advanced CAD.
Author Contributions
K.S., J.J. and A.U. conceived the concept of the study. K.S., M.Z., J.N., J.J. and A.U. contributed to the design of the research. K.S., M.Z., J.N. and J.Z. performed the review of literature and were involved in data acquisition. All authors analysed and interpreted the data. J.Z., J.J. and A.U. supervised data processing. All authors edited and approved the final version of the manuscript.
Funding
The authors have nothing to report.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: Clinical characteristics of patients analysed in the current study and not included due to unavailable serum samples.
Table S2: Characteristics of patients with GDF‐15 levels above and below median value at baseline.
Table S3: Clinical characteristics of patients treated with atorvastatin 80 mg/day or rosuvastatin 40 mg/day.
Table S4: The association of type of statin with GDF‐15 at follow‐up.
Acknowledgements
The authors have nothing to report.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Takaoka M., Tadross J. A., Al‐Hadithi A. B. A. K., et al., “GDF15 Antagonism Limits Severe Heart Failure and Prevents Cardiac Cachexia,” Cardiovascular Research 120 (2024): 2249–2260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Reyes J. and Yap G. S., “Emerging Roles of Growth Differentiation Factor 15 in Immunoregulation and Pathogenesis,” Journal of Immunology 210 (2023): 5–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Wang D., Day E. A., Townsend L. K., Djordjevic D., Jørgensen S. B., and Steinberg G. R., “GDF15: Emerging Biology and Therapeutic Applications for Obesity and Cardiometabolic Disease,” Nature Reviews. Endocrinology 17 (2021): 592–607. [DOI] [PubMed] [Google Scholar]
- 4. Kato E. T., Morrow D. A., Guo J., et al., “Growth Differentiation Factor 15 and Cardiovascular Risk: Individual Patient Meta‐Analysis,” European Heart Journal 44 (2023): 293–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Mo Q., Zhuo L., Liao Z., Li R., Chen Y., and Geng J., “Serum GDF‐15 Predicts In‐Hospital Mortality and Arrhythmic Risks in Patients With Acute Myocardial Infarction,” Clinical and Applied Thrombosis/Hemostasis 28 (2022): 10760296211063875. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Schlittenhardt D., Schober A., Strelau J., et al., “Involvement of Growth Differentiation Factor‐15/Macrophage Inhibitory Cytokine‐1 (GDF‐15/MIC‐1) in oxLDL‐Induced Apoptosis of Human Macrophages In Vitro and in Arteriosclerotic Lesions,” Cell and Tissue Research 318 (2004): 325–333. [DOI] [PubMed] [Google Scholar]
- 7. Zhang S., Hao P., Li J., et al., “Prognostic Value of Growth Differentiation Factor‐15 in Patients With Coronary Artery Disease: A Meta‐Analysis and Systematic Review,” Frontiers in Cardiovascular Medicine 10 (2023): 1054187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Stępień K., Natorska J., Ząbczyk M., Zalewski J., Jawień J., and Undas A., “High‐Dose Atorvastatin and Rosuvastatin Reduce the Levels of Neutrophil Extracellular Trap‐Related Proteins in Coronary Artery Disease: Association With Prothrombotic State,” Polskie Archiwum Medycyny Wewnętrznej 134 (2024): 16852. [DOI] [PubMed] [Google Scholar]
- 9. Diamantis E., Kyriakos G., Quiles‐Sanchez L. V., Farmaki P., and Troupis T., “The Anti‐Inflammatory Effects of Statins on Coronary Artery Disease: An Updated Review of the Literature,” Current Cardiology Reviews 13 (2017): 209–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Ridker P. M., Danielson E., Fonseca F. A., et al., “Rosuvastatin to Prevent Vascular Events in Men and Women With Elevated C‐Reactive Protein,” New England Journal of Medicine 359 (2008): 2195–2207. [DOI] [PubMed] [Google Scholar]
- 11. Bonaca M. P., Morrow D. A., Braunwald E., et al., “Growth Differentiation Factor‐15 and Risk of Recurrent Events in Patients Stabilized After Acute Coronary Syndrome: Observations From PROVE IT‐TIMI 22,” Arteriosclerosis, Thrombosis, and Vascular Biology 31 (2011): 203–210. [DOI] [PubMed] [Google Scholar]
- 12. Kim J. M., Back M. K., Yi H. S., Joung K. H., Kim H. J., and Ku B. J., “Effect of Atorvastatin on Growth Differentiation Factor‐15 in Patients With Type 2 Diabetes Mellitus and Dyslipidemia,” Diabetes and Metabolism Journal 40 (2016): 70–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Tentzeris I., Farhan S., Freynhofer M. K., et al., “Usefulness of Elevated Levels of Growth Differentiation Factor‐15 to Classify Patients With Acute Coronary Syndrome Having Percutaneous Coronary Intervention Who Would Benefit From High‐Dose Statin Therapy,” American Journal of Cardiology 120 (2017): 747–752. [DOI] [PubMed] [Google Scholar]
- 14. Jukema J., Van Neer N. J., Szarek M., et al., “Plasma Levels of Human Growth Differentiation Factor 15 (GDF‐15) and Their Effect of Alirocumab on Major Adverse Cardiovascular Events and All‐Cause Death After Acute Coronary Syndrome,” European Heart Journal 45 (2024): ehae666.1577. [Google Scholar]
- 15. Stępień K., Siudut J., Konieczyńska M., Nowak K., Zalewski J., and Undas A., “Effect of High‐Dose Statin Therapy on Coagulation Factors: Lowering of Factor XI as a Modifier of Fibrin Clot Properties in Coronary Artery Disease,” Vascular Pharmacology 149 (2023): 107153. [DOI] [PubMed] [Google Scholar]
- 16. Atar D., Munkhaugen J., and Sverre E., “Statin Discontinuation: How Can We Improve on the Multiple Pathways That Contribute to Suboptimal Statin Adherence?,” Kardiologia Polska 82 (2024): 828–830. [DOI] [PubMed] [Google Scholar]
- 17. Mach F., Koskinas K. C., Roeters van Lennep J. E., et al., “2025 Focused Update of the 2019 ESC/EAS Guidelines for the Management of Dyslipidaemias,” European Heart Journal 46 (2025): 4359–4378. [DOI] [PubMed] [Google Scholar]
- 18. Banach M., Toth P. P., Bielecka‐Dąbrowa A., and Lewek J., “Primary and Secondary Cardiovascular Prevention: Recent Advances,” Kardiologia Polska 82 (2024): 1200–1210. [DOI] [PubMed] [Google Scholar]
- 19. Anand I. S., Kempf T., Rector T. S., et al., “Serial Measurement of Growth‐Differentiation Factor‐15 in Heart Failure: Relation to Disease Severity and Prognosis in the Valsartan Heart Failure Trial,” Circulation 122 (2010): 1387–1395. [DOI] [PubMed] [Google Scholar]
- 20. Chlebus K., Zdrojewski T., Jagiełło K., et al., “Prevalence of Hypercholesterolemia in the Elderly and Very Elderly Members of the Polish Population: Results of a National Cross‐Sectional Representative Survey,” Polskie Archiwum Medycyny Wewnętrznej 134 (2024): 16882. [DOI] [PubMed] [Google Scholar]
- 21. Stępień K., Ząbczyk M., Kopytek M., Natorska J., Zalewski J., and Undas A., “Reduced Fibrin Clot Permeability on Admission and Elevated E‐Selectin at 3 Months as Novel Risk Factors of Residual Pulmonary Vascular Obstruction in Patients With Acute Pulmonary Embolism,” Journal of Thrombosis and Thrombolysis 57 (2024): 248–259. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Ząbczyk M., Natorska J., Janion‐Sadowska A., et al., “Isoprostane‐8 and GDF‐15 as Novel Markers of Post‐PE Syndrome: Relation With Prothrombotic Factors,” European Journal of Clinical Investigation 52 (2022): e13660. [DOI] [PubMed] [Google Scholar]
- 23. Ackermann K., Bonaterra G. A., Kinscherf R., and Schwarz A., “Growth Differentiation Factor‐15 Regulates oxLDL‐Induced Lipid Homeostasis and Autophagy in Human Macrophages,” Atherosclerosis 281 (2019): 128–136. [DOI] [PubMed] [Google Scholar]
- 24. Girona J., Guardiola M., Barroso E., et al., “GDF15 Circulating Levels Are Associated With Metabolic‐Associated Liver Injury and Atherosclerotic Cardiovascular Disease,” International Journal of Molecular Sciences 26 (2025): 2039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Chrysafi P., Valenzuela‐Vallejo L., Stefanakis K., Kelesidis T., Connelly M. A., and Mantzoros C. S., “Total and H‐Specific GDF‐15 Levels Increase in Caloric Deprivation Independently of Leptin in Humans,” Nature Communications 15 (2024): 5190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Wang L., Zhao J., Schank M., et al., “Circulating GDF‐15: A Biomarker for Metabolic Dysregulation and Aging in People Living With HIV,” Frontiers in Aging 5 (2024): 1414866. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Abulizi P., Loganathan N., Zhao D., et al., “Growth Differentiation Factor‐15 Deficiency Augments Inflammatory Response and Exacerbates Septic Heart and Renal Injury Induced by Lipopolysaccharide,” Scientific Reports 7 (2017): 1037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Ratnam N. M., Peterson J. M., Talbert E. E., et al., “NF‐κB Regulates GDF‐15 to Suppress Macrophage Surveillance During Early Tumor Development,” Journal of Clinical Investigation 127 (2017): 3796–3809. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Wang X., Chrysovergis K., Kosak J., and Eling T. E., “Lower NLRP3 Inflammasome Activity in NAG‐1 Transgenic Mice Is Linked to a Resistance to Obesity and Increased Insulin Sensitivity,” Obesity (Silver Spring) 22 (2014): 1256–1263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Groarke J. D., Crawford J., Collins S. M., et al., “Ponsegromab for the Treatment of Cancer Cachexia,” New England Journal of Medicine 391 (2024): 2291–2303. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Clinical characteristics of patients analysed in the current study and not included due to unavailable serum samples.
Table S2: Characteristics of patients with GDF‐15 levels above and below median value at baseline.
Table S3: Clinical characteristics of patients treated with atorvastatin 80 mg/day or rosuvastatin 40 mg/day.
Table S4: The association of type of statin with GDF‐15 at follow‐up.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.