The study was approved by the Medical Ethics Committee of Tianjin Chest Hospital. Informed consents were obtained from all subjects. ACS patients who presented to the Tianjin Chest Hospital from May to December 2019 were enrolled, with the inclusion criteria as follows: (1) Informed consent obtained from the participant who voluntarily take the medications, and related documents signed; (2) Patients aged 18 years and older presenting within 72 h after pain onset associated with either one of the following primary diagnoses: ST-elevation myocardial infarction (STEMI), non-ST-elevation myocardial infarction (NSTEMI) or unstable angina pectoris (UAP). The diagnosis of ACS patients follows the emergency rapid diagnosis and treatment guidelines for acute coronary syndrome (2019) [14]; (3) Patients who have a clearly documented LDL-C level higher than 2.6 mmol/l(100 mg/dl) before taking any medication or an LDL-C level higher than 1.8 mmol/l (70 mg/dl) while on lipid-lowering agents. The exclusion criteria include: (1) Severe end-stage diseases, such as renal dysfunction(eGFR < 30 ml/min/1.73 m²), heart failure (left ventricular ejection fraction < 35%), and malignant carcinoma, and liver dysfunction (liver enzymes ≥ 3times the upper limit of normal); (2) Systematic inflammatory disease or severe infection; (3) Thyroid disorder; (4) Pregnant or lactating women; (5) Suspected alcohol or other substance abuse, or other conditions that might impair follow-up or complicate subsequent treatment, deemed by the researcher, e.g. patients with frequent changes of workplace that might become lost on follow-up.

The 99 enrolled ACS patients were divided into two groups according to their medication: 54 in the experimental group receiving the combination therapy of PCSK9 inhibitor (Repatha®, 140 mg, q2w) and moderate statin (rosuvastatin, 10 mg, qn), and 45 in the control group receiving statin monotherapy (rosuvastatin, 10 mg, qn).

At baseline and after 8 weeks treatment, the participants in the two groups were collected peripheral venous blood at the fasting and resting state in the morning for examination.

Routine blood lipid profile measurements: The participants were fasted for 8 h. The blood was collected in a serum tube containing an inert separating gel. After the blood was fully coagulated, it was centrifuged at 3000 rpm for 10 min, and the supernatant was taken for testing. Lipid profile parameters (concentrations of TC, HDL-C, LDL-C, and TG) were measured using Roche c701 automated clinical chemistry analyzer. Apolipoprotein A1 (Apo-A1) and Apolipoprotein B (Apo-B) were measured by immunoturbidimetric methods. Plasma Lp(a) levels were determined by means of latex enhanced immuno-turbidimetry.

Nuclear magnetic resonance (NMR) spectroscopy measurements: 4 ml of the participant's whole blood (fasting for 8 h, BD blood vessels containing EDTA-K2 anticoagulant) was collected, centrifuged at 1700g for 13 min, and transferred the upper plasma into the cryopreserved tube, and then frozen and stored in aliquots at − 80 °C. During the test, the samples were taken out of the refrigerator, and after thawing completely, 400 μl plasma was taken and mixed with NMRS lipid buffer (Bruker Biospin, USA) 1:1, fully mixed, and then placed in a 5 mm NMR tube, and loaded into an automatic sample injector for testing [15].

Aliquots were measured blinded to patients' data by means of numbered codes and then merged with the clinical data set.

According to the standard operating procedure of AVANCE IVDr magnetic resonance spectrometer system (Bruker Biospin) [16, 17]. The spectra were normalized to the same quantitative scale using Bruker’s QuantRef manager within TopSpin which is based on the PULCON method; hence, the spectral intensity is normalized to proton concentration in units of millimoles per liter. For data analysis, the study selected the commercial Bruker IVDr LIpoprotein Subclass Analysis (B.I.-LISA) method [18] as lipoprotein distribution prediction method, which used a PLS-2 regression model as the algorithm for spectral deconvolution [19]. This model provides information on main lipoprotein classes, including very low density lipoprotein (VLDL), intermediate density lipoprotein (IDL), LDL, and high density lipoprotein (HDL), as well as the five VLDL subfractions (VLDL-1 to VLDL-5), six LDL subfractions (LDL-1 to LDL-6), and four HDL subfractions (HDL-1 to HDL-4). Subfractions were sorted according to their increasing density and decreasing size in ascending order, respectively.

Statistical analyses were performed by SPSS 22.0. Basic parameters of descriptive statistics for the analysed continuous variables are showed as a mean and standard deviation (SD) for normal distribution or as a median of the first and third quartiles (Q1–Q3) for a distribution other than normal. Comparisons of groups regarding continuous variables are tested with the t Student test (for a normal distribution of variables) or with the Mann–Whitney U test for nonnormal distributions. The categorical variables were presented as numbers (percentages), and comparisons between groups were conducted with Chi-square test. The concentrations of large-size LDL-P, medium-size LDL-P and small-size LDL-P were compared using ANOVA. Associations were examined by Pearson’s correlational analyses. The interval of two-sided P < 0.05 was considered statistically significant.

The baseline data of the two groups was shown in Table 1.

Demographic data, clinical characteristics, baseline blood lipid profiles, and lipoprotein particle concentrations of the 99 participants were shown in Table 1. 63.0% and 66.7% were males in the experimental group and control group, respectively; while 27.8% and 22.2% were diabetic, respectively. No significant difference was found in age, gender, statin use, comorbidities including hypertension, stroke history, dislipidemia, diabetes mellitus, MI history, prior PCI or CABG, high waist circumference and body mass index (BMI) between the two groups (P > 0.05). The comparisons of the prevalence of STEMI, NSTEMI, and UAP revealed no significant difference (P > 0.05). At baseline, the median and interquartile range of LP(a) levels in the two groups were 73.1 (13.7, 102.3) and 32.7 (8.2, 49.1), respectively, significantly higher in the experimental group with a P value of 0.032. Other baseline lipid profiles and lipoprotein particle concentrations were comparable between the two groups. The LDL-C concentrations in the experimental and control groups were 123.7 ± 32.3 mg/dl and 103.7 ± 28.1 mg/dl, respectively; and LDL-P concentrations were 1573.9 ± 375.3 nmol/l and 1317.8 ± 337.8 nmol/l, respectively. Other baseline characteristics of blood lipid profiles and lipoprotein particles were shown in Table 1.

Pearson correlation analyses were performed to assess the consistency between the NMR spectroscopy and enzymatic method measurements, in terms of the six items that they had in common (triglycerides [TG], total cholesterol [TC], LDL-C, high-density lipoprotein [HDL-C], Apo-A1, and Apo-B). The correlation coefficient was between 0.839 and 0.912, indicating close correlation between the two sets of measurement (Table 2).

In the experimental group, after a combined treatment of evolocumab and moderate statins for 8 weeks, benefits in LDL-C concentrations and other blood lipids parameters were revealed with statistical significance (P < 0.05). TC, LDL-C and its subfractions (LDL-1 to -6), VLDL-C and its subfractions (VLDL-1 to -5), and IDL-C significantly decreased compared to baseline (P < 0.001). The decrease in total LDL-C concentrations was mainly due to a decreased concentration of small-sized LDL particles (LDL 5 + 6). In contrast, a significant increase in HDL-C concentrations was observed (P < 0.05), which could be mainly attributed to an increase in small-sized HDL particles (HDL 3 + 4). The concentrations of TC, LDL-C, VLDL-C and IDL-C were reduced by 48.4%, 65.5%, 26.3%, and 62.0%, respectively, while the HDL-C concentrations was increased by 7.6%, compared to baseline.

After 8 weeks of single moderate statins treatment in the control group, the concentrations of TC, IDL-C, and LDL-C significantly decreased by 16.9%, 20.7%, and 18.1%, respectively (P < 0.05). But the concentrations of VLDL-C and HDL-C did not decrease significantly after treatment (P > 0.05).

Changes in lipoprotein particle concentrations were presented in Table 3. After 8 weeks of combined treatment of evolocumab and moderate statins, statistically significant benefits in the concentrations of LDL-P and other lipoprotein particle concentrations were observed (P < 0.05). VLDL-P, IDL-P, LDL-P and its subfractions (LDL-P1 to 6), ApoB and LP(a) all decreased compared to baseline (P < 0.001). The concentration of total LDL-P before and after treatment was 1573.9 ± 375.3 and 463.6 ± 246.7 respectively, showing a reduction of 71.1% (P < 0.001). The decrease in total LDL-P concentrations was mainly due to a decreased concentration of small-sized LDL particles (LDL 5 + 6). In this study, LDL-P were further classified into large (LDL-P1 + 2), medium (LDL-P3 + 4) and small LDL-P (LDL-P5 + 6) by the size of the particles. The concentrations of small-sized LDL-P decreased by 76.8%, with a significantly larger extent than medium-sized and large-sized LDL-P (P < 0.001). The concentrations of VLDL-P, IDL-P, LP(a), and apoB decreased by 20%, 53.3%, 21%, and 60.9%, respectively, while the size of LDL-P in the experimental group increased by 3.6% (P < 0.001), as compared to an insignificant change in the control group.

In the control group, after 8 weeks of statin monotherapy, IDL-P, LDL-P, and Apo-B concentrations significantly lowered compared to baseline (P < 0.05), but no significant change was found in VLDL-P concentration (P > 0.05). The concentrations of IDL-P, LDL-P and ApoB decreased by 24.5%, 21.7%, and 20.8% compared to baseline. Comparison of LDL-P subfractions (small-sized, medium-sized and large-sized) revealed no significant difference (P > 0.05). LP(a) was seen no significant decrease, instead an increase in the control group.

After 8 weeks treatment, the mean LDL-C concentration in the experimental group decreased from 124 to 44 mg/dl, showing a 65.3% reduction compared to the baseline, in contrast to a 18.1% reduction from 104 to 81 mg/dl in the control group. The difference in the percentage of reduction between the two groups was 47.2%, with an absolute difference of LDL-C concentration of 57 mg/dl.

According to the China Cholesterol Education Program (CCEP) Expert Consensus in 2019: for patients at very high risk, the target of LDL-C concentration should be lower than 55 mg/dl, or at least a 50% reduction from baseline. After 8 weeks treatment, 96.3% patients in the experimental group and 13.3% in the control group had achieved the LDL-C therapeutic target (P < 0.01) (Fig. 1).