key: cord-0949718-foqky30o
authors: Yan, Feifei; Huang, Fengming; Xu, Jun; Yang, Penghui; Qin, Yuhao; Lv, Jingjun; Zhang, Shaogeng; Ye, Lu; Gong, Ming; Liu, Zhibo; Wei, Jie; Xie, Tuxiu; Xu, Kai-Feng; Gao, George F.; Wang, Fu-Sheng; Cai, Lin; Jiang, Chengyu
title: Antihypertensive drugs are associated with reduced fatal outcomes and improved clinical characteristics in elderly COVID-19 patients
date: 2020-10-29
journal: Cell Discov
DOI: 10.1038/s41421-020-00221-6
sha: d84d28e76631862d9add332371c23eecab37a90d
doc_id: 949718
cord_uid: foqky30o

The novel coronavirus (CoV) severe acute respiratory syndrome (SARS)-CoV-2 outbreak began at the end of 2019 in Wuhan, China, and has spread to over 200 countries. In this multicenter retrospective study, we identified 2190 adult patients admitted for laboratory-confirmed COVID-19 in three participating centers. Multivariate logistic regression was conducted in patients with comorbid hypertension to examine the potential association between clinical outcomes, disease severity, and clinical characteristics with the use of ACEI, ARB, calcium-channel blockers (CCB), beta-blockers (BB), and thiazide diuretics. The clinical outcome, dyspnea, and fatigue were significantly improved in patients, especially elderly patients who were older than 65 years, who took ARB drugs prior to hospitalization compared to patients who took no drugs. The reduction of disease severity of elderly COVID-19 patients was associated with CCB and ACEI users. Clinical indices, including CRP, lymphocyte count, procalcitonin D dimer, and hemoglobin, were significantly improved in elderly ARB users. In addition, the clinical outcomes were statistically significantly improved in patients who took antihypertension drugs ARB, BB, and CCB after statistical adjustment by all ages, gender, baseline of blood pressures, and coexisting medical conditions. Our data indicate that hypertension drugs ARB, ACEI, CCB, and BB might be beneficial for COVID-19 patients.

At the end of 2019, a cluster of lethal pneumonia cases was reported in Wuhan, China. A SARS-CoV-like coronavirus in the respiratory tracts of patients was soon isolated and the viral genome sequenced; it was later named SARS-CoV-2 1-6 . In severe cases of COVID-19, patients develop acute respiratory distress syndrome (ARDS) and often die with multiorgan dysfunction syndrome (MODS). By March 11, 2020 , the virus infection had spread to over 100 countries, and SARS-CoV-2 infection was declared a global pandemic by the World Health Organization (https://www.who.int/news-room/detail/30-01-2020statement-on-the-second-meeting-of-the-internationalhealth-regulations-(2005)-emergency-committeeregarding-the-outbreak-of-novel-coronavirus-(2019-ncov)).

There has been growing concern about the use of antihypertensives in patients with COVID-19, mostly due to the fact that angiotensin-converting enzyme 2 (ACE2) 1,3,4,7-10 , a negative regulator of the renin-angiotensin-aldosterone system (RAAS), is a co-receptor for viral entry into human cells by SARS-CoV-2. ACE cleaves angiotensin I to generate angiotensin II, whereas ACE2 converts angiotensin II into angiotensin (1-7) [11] [12] [13] . By counteracting the action of ACE, ACE2 plays a crucial role in maintaining blood-pressure homeostasis, fluid, and salt balance 14, 15 . ACE inhibitor (ACEI) and angiotensin receptor blockers (ARB) are the most commonly prescribed antihypertensive medications [16] [17] [18] [19] . A viewpoint "COVID-19 and angiotensin-converting enzyme inhibitors and angiotensin receptor blockers, what is the evidence?" published on March 24, 2020 in JAMA pointed out that no clinical data are available 20 . In a special report "reninangiotensin-aldosterone system (RAAS) inhibitors in patients with COVID-19" published on March 30, 2020 in the New England Journal of Medicine, the authors discussed the potential for benefit rather than harm of RAAS blockers in COVID-19, but also pointed out that insufficient data are available to determine whether these observations readily translate to humans, and no studies have evaluated the effects of RAAS inhibitors in COVID-19 21 .

A previous study from this research group investigated the role of RAAS in the acute lung failure caused by SARS-CoV-2 infection 22 . SARS-CoV-2 infection and the spike protein of SARS-CoV-2 reduce ACE2 expression, increase angiotensin II-level signaling through the angiotensin II type 1a receptor (AT1a), promote disease pathogenesis, induce lung edemas, and impair respiratory function 9, 12 . We demonstrated that the ARB losartan could attenuate acute lung failure in a mouse model whose condition had been worsened after injection of the SARS-CoV-2 spike protein 9 . We also showed imbalanced RAAS in many predisposing conditions for ARDS, including sepsis, acid aspiration, bacteria, SARS-CoV-2, avian influenza (H5N1 and H7N9) infections, as well as nanoparticle aspiration 9, [11] [12] [13] 23 . These studies suggested that blocking the RAAS pathway and reducing angiotensin II levels could ameliorate lung injury 9, 11, 12, 16, 23 .

Observational studies have associated the use of ACEIs and ARBs with improved outcomes in patients with pneumonia 24 . Our previous study reported significantly elevated plasma levels of angiotensin II in COVID-19 patients 8 , again indicating RAAS imbalance in COVID-19. More studies of ACEI/ARBs associated with mortality and morbidity of COVID-19 reported different results recently [25] [26] [27] [28] [29] . Here, we conducted a retrospective study to examine the potential association between the use of antihypertensives and COVID-19 disease severity.

This study cohort included 2190 patients with COVID-19 who were admitted to three hospitals in China. A total of 655 participants with hypertension were included in the subsequent analysis, and 19 cases were excluded due to missing medication information (Fig. 1) . The mean (±SE) age was 64.6 ± 11.8 years, 51.8% of the patients were male (Supplementary Table S1 ), and 318 patients were >65 years of age (Supplementary Table S2 ). The distribution of age and sex among total severe (169, 25.8%) and nonsevere (486, 74.2%) COVID-19 patients was significantly different. Severe COVID-19 patients were older (69.6 ± 11.0 years vs 62.9 ± 11.5 years, P < 0.001) and more often men (60.4%, 102/169 vs 49%, 238/486, P = 0.012). Besides, nonsurvivors were significantly older than survivors (71.0 ± 10.9 years vs 64.2 ± 11.7 years, P < 0.001). No association between death and sex was found (P > 0.05). Among elderly patients (>65 years), patients who were older were more likely to develop severe COVID-19 and die. Men were more likely to develop severe COVID-19, but this had no statistically significant effect on mortality.

ARB and CCB are associated with decreased mortality in elderly COVID-19 patients

Using the multivariable logistic regression method, we evaluated the effects of hypertensive comorbidity patients who took antihypertensive drugs prior to hospitalizations, in subgroups, ARB, ACEI, CCB, Thiazide, and BB, on mortality and disease severity compared to a no-drug-used patient group with hypertension comorbidity (Tables 1, 2  and Supplementary Tables S3-S8 ). The clinical outcomes of ARB users were statistically significantly improved with the adjustment of age, gender, baseline of blood pressure, and coexisting medical condition variables among all patients (adjusted OR = 0.421, 95% CI: 0.19-0.934, P = 0.033), as well as in elderly patients who were more than 65 years of age (adjusted OR = 0.202, 95% CI: 0.055-0.745, P = 0.016) ( Table 1 and Supplementary Tables S5, S6 ). The clinical outcomes of CCB users were statistically significantly improved in elderly patients (adjusted OR = 0.22, 95% CI: 0.062-0.778, P = 0.019) ( Table 1 and Supplementary Table S6 ). Our data indicate that ARB and CCB might be beneficial to COVID-19 patients.

CCB and ACEI are associated with reduced disease severity in elderly COVID-19 patients

We observed statistically significant reductions of disease severity in CCB antihypertensive drug subgroups compared to no drug users after adjustment (adjusted OR = 0.472, 95% CI: 0.257-0.865, P = 0.015), especially in elderly patients (adjusted OR = 0.287, 95% CI: 0.114-0.723, P = 0.008) ( Table 2 and Supplementary  Tables S7, S8 ). The disease severity of ACEI users was statistically significantly reduced in elderly patients (adjusted OR = 0.156, 95% CI: 0.036-0.67, P = 0.013) ( Table 2 and Supplementary Table S8 ). Statistically significant reductions in the odds of developing severe disease among ARB users compared to no drug users were observed without adjustment (Table 2) . Thus, CCB and ACEI antihypertensive drugs might associate with disease severity in COVID-19 patients.

ARB associated with the reductions of dyspnea and fatigue in COVID-19 patients Dyspnea and fatigue are the characteristics of COVID-19. Especially respiratory distress dyspnea, which is defined by a saturation of peripheral oxygen (SPO 2 ) <93%, and a respiratory rate >30 times per min. Using the multivariable logistic regression method calculated in cohort C, which recorded some clinically detailed manifestations, we found that ARB users compared to no antihypertensive drug users were associated with reduced dyspnea (adjusted OR = 0.4, 95% CI: 0.236-0.678, P = 0.001) ( Table 3 ) and less fatigue (adjusted OR = 0.643, 95% CI: 0.457-0.906, P = 0.012) ( Table 4 ). Similar results were also observed in elderly COVID-19 patients (Tables  3 and 4 ). BB and CCB users were also associated with significantly reduced dyspnea, in COVID-19 patients (Table 3 ). These data suggest that ARB might improve respiratory syndromes in COVID-19 patients.

Using the Mann-Whitney U test, we discovered that clinical indices related to infection and inflammation, including CRP, lymphocyte count, and procalcitonin, were significantly improved to normal levels in elderly COVID-19 patients who took ARB prior to hospitalization compared to non-antihypertensive drug users ( Fig. 2a and Supplementary Table S9 ). Blood clots have been reported in COVID-19 patients 30, 31 . D dimer, a biomarker of blood clots, was significantly reduced in elderly ARB users compared to non-antihypertensive drug users (Fig. 2a) . The level of hemoglobin, which carries oxygen from the lungs to the rest of the body, was also significantly increased, returning to normal levels in elderly ARB users (Fig. 2a) . COVID-19 elderly patients who took Thiazide, BB, or CCB prior to hospitalization were also associated with these improved clinical indices, except for the blood-clot biomarker ( Fig. 2b-d) . Taken together, these data indicate that COVID-19 elderly patients with hypertensive comorbidity who took ARB prior to hospitalizations might have reduced clinical syndromes compared to those who took no antihypertensive drugs.

In this retrospective study, we found an association between the use of ARBs and reduced mortality rate in COVID-19 patients with comorbid hypertension (vs those taking no antihypertensives) ( Table 1) . We also noticed the Ref.

Ref. Fully adjusted model includes the following covariates: age, gender, baseline of blood pressure (including SBP and DBP), and coexisting medical conditions (including chronic heart, lung, renal, liver, and cerebrovascular disease, diabetes, and cancer). decreased disease severity in elderly COVID-19 patients taking CCB and ACEI antihypertensive drugs ( Table 2 ). There are 18 reports studying the association of ARBs and ACEI drugs with COVID-19 26, 29, [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] . Eleven out of the 18 papers reported the analysis of fatal outcome. The results are somewhat controversial due to the different methodologies used and the nature of cohorts. Eight of the eleven papers did not separate ARB and ACEI users, and combined them as one study group. Although both ARBs and ACEI are blockers of RAAS, they target different genes. A more rigorous analysis would be to calculate the COVID-19 association of ARBs and ACEI separately. The other three articles included separated ARBs users' group 32, 42, 47 . One of the three reports published in Hypertension demonstrating similar conclusion as our report here. However, this study reported in a short communication with only 1 figure without detailed methodology 47 . Two reports published in JAMA and JAMA Cardiology and one retracted NEJM paper show that ARBs were not statistically significantly associated with fatal outcome of COVID-19 32, 42, 48 . One of the major differences is that their control group was quite different from our controls. The JAMA article used CCB users as control group 42 . The other articles (published in JAMA Cardiology and the retracted NEJM article) used all hypertension patients excluding the examined drug users (ARBs nonusers) as their control group 32, 48 . Notably, their control group of ARB users included ACEI users. ACEI, as RAAS system drugs, may have similar effects to ARBs. We used their method (i.e., all hypertensive patients excluding ARBs, and including ACEIs in the control group) on our data: the adjusted results showed that ARBs were not statistically significantly associated with fatal outcome of COVID-19. However, ARB drugs statistically significantly improved fatal outcomes of COVID-19 patients when ACEI drug users were removed from the control group (Supplementary Table S10 ). Thus, this might partially contribute to the different conclusion of this paper that ARBs/ACEI drugs did not statistically significantly reduce the fatal outcomes of COVID-19 patients.

In our study, ARBs and ACEI exhibited different effects associated with the mortality and morbidity of COVID-19 patients, albeit both are RAAS blockers. Previous literature reported that ARBs could effectively block AT1R, whose stimulation is involved in multiorgan injuries 28, 49, 50 , suggesting that ARBs might be more effective in treating MODS. Besides antihypertensive effects, ARBs could directly reduce lung edema, epithelial and endothelial cell injury, and pro-inflammatory cytokines and chemokines, decrease apoptosis and fibrosis, protect mitochondrial functions, maintain insulin and lipid metabolism, and normalize the coagulation cascade 16, 51, 52 . These reports may provide mechanisms that ARBs were associated with significantly improved clinical characteristics of COVID-19 patients in this study. Our unpublished data demonstrated that ARBs, especially losartan, enhanced the survival rate more than ACEI in an avian influenza A H5N1 mouse model 16 . Further studies are necessary to elucidate the mechanisms involved in both ACEI and ARBs.

Although a wide variety of conditions, including SARS-CoV, sepsis, acid aspiration, bacteremia, and avian influenza (H5N1 and H7N9) infections, predispose patients to ARDS, a common mechanism seems to be an imbalance of RAAS 9, [11] [12] [13] 23 . Our previous studies in mice suggested that ACE, angiotensin II, and the type 1a angiotensin II receptor participate in disease pathogenesis, whereas ACE2 and the type 2a angiotensin II receptor could protect the mice from severe acute lung injury 9, 12 . Blood levels of angiotensin II have been reported to be elevated in patients with COVID-19 (https://www.who.int/news-room/detail/30-01-2020-state ment-on-the-second-meeting-of-the-international-health-reg ulations-(2005)-emergency-committee-regarding-the-out break-of-novel-coronavirus-(2019-ncov)) and influenza 9, 11 . Interventions that restore the balance of RAAS, for example, ARB antihypertensives, thus could be beneficial. A trial of losartan, a member of ARBs, as treatment for COVID-19 is currently ongoing at the University of Minnesota (NCT0 43111777 and NCT04312009) (https://clinicaltrials.gov/ct2/ show/NCT04311177?cond=COVID-19&lead=University+ of+Minnesota&cntry=US&draw=2&rank=2; https://clini caltrials.gov/ct2/show/NCT04312009?cond=COVID-19& lead=University+of+Minnesota&cntry=US&draw=2& rank=1).

We noted the limitations of this study. The patients' numbers who took ACEI and thiazide were too small and may induce statistical bias. The outcomes of hypertensive and nonhypertensive COVID-19 patients were not (see figure on previous page) Fig. 2 Clinical indices between the use of antihypertensive agents and no use among total cohorts and a subgroup of patients older than 65 years with hypertension comorbidity. a Lymphocyte counts, hemoglobin, C-reactive protein, procalcitonin, and D-dimer expression levels of COVID-19 patients between ARB users and no-use antihypertensive agents. b Hemoglobin expression levels of COVID-19 patients between used thiazide diuretics and no-use antihypertensive agents. c Lymphocyte counts, percentages, and hematocrit expression levels of COVID-19 patients between used BB and no-use antihypertensive agents. d Lymphocyte and basophilic granulocyte count expression levels of COVID-19 patients between CCB users and no-use antihypertensive agents, among the total cohort and older than 65-year cohort with hypertension comorbidity. Boxplots with all points are shown in the graph. *P value < 0.05, **P value < 0.01, Mann-Whitney U test. Detailed available numbers of each laboratory result group are shown in Supplementary Table S5. compared. However, the uniqueness of our study is to statistically adjust all combined factors of age, gender, baseline of blood pressure (including SBP and DBP), and coexisting medical conditions (including chronic heart, lung, renal, liver, and cerebrovascular disease, diabetes, and cancer), although BB has no statistical significance before the adjustment.

In summary, the results from this statistical human study link RAAS blockers with COVID-19 patients, and indicate that the use of ARBs in hypertensive patients is associated with decreased mortality of COVID-19 patients. Therefore, hypertensive patients should continue to take ARBs and ACEIs in the current COVID-19 pandemic. Our study also indicates that antihypertensive drugs ARB, ACEI, CCB, and BB might also be potentially economic and effective remedies for COVID-19 patients, especially elderly patients, calling for clinical trials to test this directly.

In this retrospective study, we identified all the adult patients (≥18 years of age) with laboratory-confirmed COVID- 19 

Demographical, epidemiological, clinical, disease severity, and outcome data were extracted from medical records by two trained physicians according to the unified data collection criteria. The classes of antihypertensive medication include ARB, ACE inhibitors, thiazide diuretics, beta-blockers, and calcium-channel blockers, before admission was summarized. Patients diagnosed with hypertension but not taking any antihypertensive drugs were also collected and compared as a control group. Routine blood tests of critically ill patients contained complete blood cell count, serum biochemical examinations (including renal and liver function), coagulation profile, D dimer, interleukin-6 (IL-6), and Creactive protein (CRP). Disease severity was classified as severe or mild according to the guidelines of 2019-nCoV infection from the National Health Commission of the People's Republic of China (http://www.nhc.gov.cn/yzygj/ s7653p/202003/46c9294a7dfe4cef80dc7f5912eb1989/files/ ce3e6945832a438eaae415350a8ce964.pdf). All data were independently checked by more than one physician.

The date of onset was defined as the day when the symptom was noticed. The status at admission and worst status from onset to the final clinical endpoint (discharge/ death) of COVID-19 are based on the Chinese official management guidelines for COVID-19 (http://www.nhc. gov.cn/yzygj/s7653p/202003/46c9294a7dfe4cef80dc7f59 12eb1989/files/ce3e6945832a438eaae415350a8ce964.pdf): severe COVID-19 comprised any one of the following: at resting condition, the respiratory rate was more than 30 times per minute, blood oxygen was less than or equal to 93%, and an obvious progression of the pulmonary infiltrative lesion as shown by chest radiography was more than 50% within 24-48 h. Hypertension is defined as office SBP values >140 mmHg and/or diastolic BP (DBP) values >90 mmHg according to the 2018 ESC/ESH Guidelines for management of arterial hypertension 55 . Fever was defined as the axillary temperature of at least 37.3°C.

The Mann-Whitney U test was used to compare differences in continuous variables between two groups of COVID-19 cases. The chi-squared test was used to compare differences in categorical variables between two groups of COVID-19 cases. Odds ratio (OR) and 95% confidence interval (CI) were calculated for risk evaluation. Adjusted OR was calculated using "Enter" stepwise multivariable logistic regression. Statistical analyses were performed with SPSS 16.0 for Windows (SPSS, Inc.).

A familial cluster of pneumonia associated with the 2019 novel coronavirus indicating person-to-person transmission: a study of a family cluster

Clinical features of patients infected with 2019 novel coronavirus in Wuhan

Recent advances in the detection of respiratory virus infection in humans

The continuing 2019-nCoV epidemic threat of novel coronaviruses to global health-the latest 2019 novel coronavirus outbreak in Wuhan, China

A novel coronavirus outbreak of global health concern

A novel coronavirus from patients with pneumonia in China

Elevated levels of plasma cytokines in the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection reflect viral load and lung injury

Clinical and biochemical indexes from 2019-nCoV infected patients linked to viral loads and lung injury

A crucial role of angiotensin converting enzyme 2 (ACE2) in SARS coronavirus-induced lung injury

Pathological findings of COVID-19 associated with acute respiratory distress syndrome

Angiotensin II plasma levels are linked to disease severity and predict fatal outcomes in H7N9-infected patients

Angiotensin-converting enzyme 2 protects from severe acute lung failure

Angiotensin-converting enzyme 2 protects from lethal avian influenza A H5N1 infections

Angiotensin II signal transduction: an update on mechanisms of physiology and pathophysiology

Effects of calcium channel blockers comparing to angiotensinconverting enzyme inhibitors and angiotensin receptor blockers in patients with hypertension and chronic kidney disease stage 3 to 5 and dialysis: a systematic review and meta-analysis

Angiotensin II receptor blocker as a novel therapy in acute lung injury induced by avian influenza A H5N1 virus infection in mouse

ACE inhibitors and pneumonia

ACE inhibitors and pneumonia in elderly people

Examination of selected clinical factors and medication use as risk factors for pneumonia during stroke rehabilitation: a case-control study

COVID-19 and angiotensin-converting enzyme inhibitors and angiotensin receptor blockers: what is the evidence?

Renin-angiotensin-aldosterone system inhibitors in patients with Covid-19

SARS coronavirus entry into host cells through a novel clathrin-and caveolae-independent endocytic pathway

Cationic nanoparticles directly bind angiotensin-converting enzyme 2 and induce acute lung injury in mice

Population-based study of statins, angiotensin II receptor blockers, and angiotensin-converting enzyme inhibitors on pneumonia-related outcomes

Association of inpatient use of angiotensin converting enzyme inhibitors and angiotensin II receptor blockers with mortality among patients with hypertension hospitalized with COVID-19

Renin-angiotensin-aldosterone system inhibitors and risk of Covid-19

Renin-angiotensin system inhibitors improve the clinical outcomes of COVID-19 patients with hypertension

Circulating plasma concentrations of angiotensin-converting enzyme 2 in men and women with heart failure and effects of reninangiotensin-aldosterone inhibitors

Renin-angiotensinaldosterone system blockers and the risk of Covid-19

Diagnostic utility of clinical laboratory data determinations for patients with the severe COVID-19

Prominent changes in blood coagulation of patients with SARS-CoV-2 infection

Association of renin-angiotensin system inhibitors with severity or risk of death in patients with hypertension hospitalized for coronavirus disease 2019 (COVID-19) infection in Wuhan

Association of use of angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers with testing positive for coronavirus disease 2019 (COVID-19)

COVID-19 patients with hypertension have more severity condition, and ACEI/ARB treatment have no infulence on the clinical severity and outcome

Association of renin-angiotensinaldosterone system inhibitors with COVID-19-related outcomes in Korea: a nationwide population-based cohort study

Pharmacological characteristics of patients infected with SARS-Cov-2 admitted to intensive care unit in South of France

Associations between angiotensin converting enzyme inhibitors and angiotensin II receptor blocker use, gastrointestinal symptoms, and mortality among patients with COVID-19

Clinical characteristics and outcomes of patients with diabetes and COVID-19 in association with glucose-lowering medication

Clinical characteristics of coronavirus disease 2019 (COVID-19) patients with hypertension on renin-angiotensin system inhibitors

Association of inpatient use of angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers with mortality among patients with hypertension hospitalized with COVID-19

COVID-19 with different severities: a multicenter study of clinical features

Association of angiotensin-converting enzyme inhibitor or angiotensin receptor blocker use with COVID-19 diagnosis and mortality

Continued in-hospital ACE inhibitor and ARB use in hypertensive COVID-19 patients is associated with positive clinical outcomes

Renin-angiotensin system inhibitors improve the clinical outcomes of COVID-19 patients with hypertension

Long-term ACE inhibitor/ARB use is associated with severe renal dysfunction and acute kidney injury in patients with severe COVID-19: results from a referral center cohort in the North East of France

Clinical characteristics and outcomes of 112 cardiovascular disease patients infected by 2019-nCoV

Comparative impacts of ACE (angiotensin-converting enzyme) inhibitors versus angiotensin II receptor blockers on the risk of COVID-19 mortality

Cardiovascular disease, drug therapy, and mortality in Covid-19

Comparative efficacy and safety of blood pressure-lowering agents in adults with diabetes and kidney disease: a network meta-analysis

Update on role of direct renin inhibitor in diabetic kidney disease

Angiotensin receptor blockers and COVID-19

Angiotensin receptor blockers as tentative SARS-CoV-2 therapeutics

Laboratory Testing for 2019 Novel Coronavirus (2019-nCoV) in Suspected Human Cases: Interim Guidance (World Health Organization

ESC/ESH guidelines for the management of arterial hypertension

The authors would like to thank Prof. Linzhao Cheng of Johns Hopkins University and USTC for the fruitful discussions and editing of this paper. The authors would also like to thank Dr. Yingxia Liu, Lei Liu, and Hong Gao for providing clinical data of patients from ShenZhen Third people's hospital when this project started. These data were used in another project. This work was supported by the National Natural Science Foundation of China (81788101), the Chinese Academy of Medical Sciences Innovation Fund for