Synergistic effect of the commonest residual risk factors, remnant cholesterol, lipoprotein(a), and inflammation, on prognosis of statin-treated patients with chronic coronary syndrome

In spite of significant advances in the understanding, prevention, detection, and treatment of atherosclerotic cardiovascular disease (ASCVD), it remains the leading cause of death worldwide [1]. The central, modifiable causal risk factor of the development of ASCVD is dyslipidemia, particularly increased low-density lipoprotein cholesterol (LDL-C) levels [2]. Thus, current guidelines and consensuses have focused on decreasing LDL-C levels with statin, nonstatin, or combination therapies for primary and secondary prevention of ASCVD events [2]. However, patients continue to suffer from cardiovascular events frequently despite being treated effectively with lipid-lowering drugs including intensive statins and even if the LDL-C target is achieved, which is called residual cardiovascular risk (RCVR) [3, 4]. Therefore, there has been a great effort to identify and manage major determinants of RCVR.

Beyond traditional risk factors addressed in usual clinical care [5], residual atherogenic lipoproteins, especially triglyceride-rich lipoproteins (TRLs) and lipoprotein(a) [Lp(a)], and inflammation are suggested to play a critical role in driving RCVR after LDL-C lowering and have been paid great attention in cardiovascular field [2, 6, 7]. Remnant cholesterol (RC) is the cholesterol content of TRLs [8]. There is mounting evidence that elevated RC concentration is causally related to ASCVD risk independent of LDL-C levels [8‐12]. While Lp(a) is a heterogenous glycoprotein, which is an apoB100 containing lipoprotein covalently bound to apolipoprotein(a) [13]. Numerous studies including ours have demonstrated the significance of Lp(a) in the development and progression of ASCVD [14‐21], even in individuals with LDL-C levels < 1.8 mmol/L [22, 23]. In regard to the residual inflammatory risk, previous studies have indicated that a low systemic inflammation burden, as determined by the level of high-sensitivity C-reactive protein (hsCRP), is associated with a better prognosis in statin-treated patients [24‐26]. Inflammation has been suggested to be an important treatment target for the prevention of ASCVD events in high-risk patients [5, 26, 27].

Interestingly, it is indicated that elevated concentration of RC is causally associated with low-grade inflammation [28, 29], and the link of Lp(a) to ASCVD may partially be driven by its pro-inflammatory effects [16]. Based on the evidence regarding the relation of RC, Lp(a), and inflammation to the residual risk following LDL-C lowering therapy and the close relationships among them, we hypothesized that there may be synergetic effects of these three RCVR factors on clinical outcomes in statin-treated patients with ASCVD. Thus, we performed this study to investigate the separate and combined prognostic value of RC, Lp(a), and hsCRP in a large cohort of statin-treated patients with chronic coronary syndrome (CCS).

The data that support the findings of this study are available from the corresponding author on reasonable request.

The holistic view of multiple risk factors maybe a wiser way for risk stratification and management of ASCVD. In this prospective analysis, we observed that elevated RC and Lp(a) levels were associated with a respective 1.47-fold and 1.40-fold increased risk of MACEs among statin-treated patients with CCS, while high levels of hsCRP were related to a slightly but non-significantly increased MACEs risk. When participants were categorized according to RC, Lp(a), and hsCRP levels together, only patients with high levels of them three had a significant 1.99-fold higher risk of MACEs compared with those in the reference group (Low RC-Low Lp(a)-Low hsCRP). Moreover, interaction analyses showed that there was a synergistic relationship between any two of RC, Lp(a), and hsCRP for predicting the risk of MACEs. These findings suggested that the combination of RC, Lp(a), and hsCRP was of great importance for risk stratification in statin-treated patients with CCS.

Statin is considered as the treatment of choice for reducing plasma cholesterols, and is regarded as a corner-stone of primary and secondary prevention of ASCVD by most guidelines [31]. Unlike the recommendation of high-intensity statins in western populations, moderate-intensity statins are recommended as the initial treatment in China due to the relatively lower levels of plasma cholesterols and greater sensitivity to statins of Chinese population [32, 33]. Thus, in the present study, 93.9% of patients were prescribed moderate-intensity statins on discharge, while only 4.5% took high-intensity statins. Recently, antibodies and synthetic small interfering RNA (siRNA) targeting the serine-protease proprotein convertase subtilisin–kexin type-9 (PCSK9), a physiological regulator of LDL-C, have increasingly been used in patients with high ASCVD risk and brought about a further effective decline of LDL-C concentrations with good security [34, 35]. However, a relatively high proportion of patients remain at high risk of cardiovascular events in spite of achieving guideline-recommended treatment targets [36, 37]. Thus, RCVR in the current era poses new challenges on the prevention of ASCVD events despite managing traditional risk factors, including hypertension, DM, and unhealthy lifestyle [5, 38, 39]. Targeting the RCVR has been the primary goal of a large number of pharmacotherapies and has attained a measure of success [36]. Nonetheless, incomplete understanding of the specific mechanisms and the lack of efficient tools to precisely evaluate the RCVR contributed to the relatively high residual risk under optimal medical therapy, which has aroused substantial concern and evoked great efforts to identify most important contributors and more suitable targets to further reduce the RCVR [36, 37].

In recent years, emerging evidence suggests the crucial significance of residual atherogenic lipoprotein burden in causing ASCVD events [2, 6, 37]. Among non-LDL-C lipid parameters, RC and Lp(a) are two well-demonstrated factors inducing RCVR [2]. Elevated RC levels were shown to be associated with a high cardiovascular risk in both primary and secondary prevention cohorts [8‐10, 12, 40]. Mendelian randomization study suggested that per 1.0 mmol/L increase of RC was associated with a 2.8-fold causal risk of ischemic heart disease, independent of LDL-C or HDL-C levels [11]. Moreover, a post hoc analysis from TNT trial showed that the reduction of RC levels with statins resulted in a significant lower risk of MACE beyond the reduction in LDL-C concentrations [10]. As to Lp(a), it may mediate atherogenicity via its LDL moiety and induce prothrombotic effect through the plasminogen-like apolipoprotein(a) component [16]. In addition, it also potentially exerts proinflammatory properties via accumulation of oxidized phospholipids [16]. There has been considerable evidence that elevated baseline and on-statin Lp(a) concentrations are independently related to the risk of ASCVD events [14‐21, 41]. It has been reported that Lp(a) becomes a more potent predictor of residual risk when LDL-mediated risk is diminished with statins, since statin cannot reduce Lp(a)-mediated risk [41]. Furthermore, the FOURIER and ODYSSEY OUTCOMES trials have suggested that elevated Lp(a) level remains a significant risk factor for ASCVD events in patients with on-treatment LDL-C levels < 1.3 mmol/L [42, 43]. All the above evidence reveals that RC and Lp(a) are two crucial and promising non-traditional RCVR factors. Undoubtedly, our data further confirmed that elevated RC and Lp(a) levels were significantly related to the risk of MACEs in statin-treated patients with CCS after adjusting for traditional risk factors including LDL-C. Stratification analysis according to LDL-C levels showed that the trends of relationship of RC and Lp(a) to the MACEs risk were consistent in both strata. However, in patients with LDL-C levels ≥ 2.6 mmol/L, elevated RC and Lp(a) were associated with a slightly but non-significantly increased risk for MACEs, which may be due to the small sample size.

In addition, residual inflammatory risk after statin and other lipid-lowering drugs has also been greatly appreciated in recent years [7]. From the earlier JUPITER [24] to the recently published FOURIER [25] and a post hoc analysis of the SPIRE trials [44], inflammation plays a significant role in the occurrence of cardiovascular events independent of circulating LDL-C levels. HsCRP, a most common used biomarker reflecting systemic inflammation, was reported to be an important and independent predictor of the incidence and prognosis of ASCVD [25, 44‐46]. The benefits of anti-inflammatory effects of statins have already been proved [47‐49]. Furthermore, recent trials of anti-inflammatory therapies targeted at specific inflammatory pathways have further demonstrated the efficacy for the prevention of cardiovascular events [26, 27]. Therefore, interfering residual inflammatory risk is necessary in addition to managing LDL-C and residual cholesterol risk [50]. In the present study, elevated hsCRP level was also associated with a higher risk of MACEs, but this was not statistically significant, which may be related to the cut-off value (median) used in this study.

Currently, the importance of composite risk factor evaluation and control for stratifying and reducing ASCVD risk has gained increased attention. It has been reported that when multiple risk factors are present, the increase of risk is often synergistic rather than additive [51]. For example, in the MRFIT (Multiple Risk Factor Intervention Trial) with an average follow-up of 12 years, the coronary mortality for nonsmoking men, with SBP < 120 mmHg and serum TC concentration < 4.7 mmol/L, was 3.1 per 10,000 person-years. For those with SBP over 142 mmHg, it was 13.7; and for those with serum TC level over 6.3 mmol/L, it was 12.2. Moreover, when three risk factors were present together, as in smokers with SBP > 142 mmHg and serum TC concentration > 6.3 mmol/L, the coronary mortality was 62.6 per 10,000 person-years [52]. As stated above, Lp(a) may accelerate the cardiovascular risk through its pro-inflammatory effects. Meanwhile, recent researches indicated that RC was causally associated with low-grade inflammation marked by elevated hsCRP levels [28, 29]. Given the potential close connections among RC, Lp(a), and inflammation and their associations with the RCVR, there may be synergistic effects of them on clinical prognosis in statin-treated patients. To clarify this issue, we performed an analysis categorizing the study population according to RC, Lp(a) and hsCRP levels together and found that the combination of elevated levels of the three indicators indeed greatly worsened the outcomes of statin-treated patients with CCS.

Moreover, compared with patients with low levels of RC, Lp(a), and hsCRP, only those with elevated levels of all them three had a significantly 1.99-fold higher risk of MACEs and this phenomenon mainly existed in patients with LDL-C levels less than 2.6 mmol/L. Positive correlations and synergistic relationships for predicting MACEs risk were observed between any two of these three RCVR factors. The insignificant association of the combination of RC, Lp(a), and hsCRP with MACEs risk in subjects with LDL-C concentrations over 2.6 mmol/L may be limited to the relatively small sample size of this subgroup. In aggregate, our data suggested that a joint assessment of RC, Lp(a), and hsCRP might be a better strategy to further improve RCVR stratification in patients with established CAD under statin therapy.

Nonetheless, this study has several limitations. First, RC concentration was acquired by calculation, but not direct measurement. However, these two kinds of methods show a good correlation and the calculated method used in this study has been adopted in multiple previous studies [9‐11]. Second, the concentrations of Lp(a) are obviously different among various ethnicities, which might impact the generalizability of our findings. Third, because of the nature of observational studies, we didn’t have follow-up data of RC, Lp(a), and hsCRP levels, which might provide incremental value for the stratification of MACEs risk. Fourth, the prescribed medications of the patients may be improved based on the update guidelines over the follow-up period, which might influence the outcomes.

In summary, the present study for the first time indicated that there were synergistic relationships of elevated levels of RC, Lp(a), and hsCRP for predicting MACEs risk and the combination of them could greatly worsen the outcomes of statin-treated patients with CCS, especially in those with LDL-C levels < 2.6 mmol/L. These novel findings suggested that combined evaluation of the major RCVR factors, RC, Lp(a), and inflammation, may be a better strategy for risk stratification in the statin era.