The Value Of Single Nucleotide Polymorphism By Exon Sequencing In Predicting Phenotype And Prognosis Of ARDS

Part One Variance analysis for single nucleotide polymorphism by exome sequencing and subphenotypes of ARDS based on clinical characteristicsObjective: To understand the variances of single nucleotide polymorphism(SNP)in different ARDS subphenotypes according to the severity of lung injury,pulmonary and extrapulmonary,whether combined with sepsis and shock.Methods: This was an investigator-initiated,singlecenter,prospective trial which enrolled adult ARDS(according to Berlin definition)patients admitted to Intensive care unit,Zhongda Hospital,Southeast University.Baseline characteristics including diagnosis,etiology of ARDS,acute physiology and chronic health evaluation II score,sequential organ failure assessment score,Murray lung injury score were recorded.Peripheral blood samples were drawn.Patients were divided into different subphenotypes according to the severity of lung injury,pulmonary or extrapulmonary,whether combined with sepsis and shock.Whole-exome sequencing was performed by the sequencing platform Illumina,the data were compared with reference genome UCSC hg19.Single nucleotide variants/ Insertion-deletion(SNV/In Del)were tested by plink method to understand the difference between the subphenotypes of ARDS.Results: 1.Baseline characteristics: A total of 105 ARDS patients were enrolled in the study,including 52 severe ARDS patients and 91 pulmonary ARDS patients.On enrollment,89 patients were combined with sepsis,and 66 patients were combined with shock.2.SNV/In Del data by whole-exome sequencing: By whole-exome sequencing,the number of SNV/In Del were 471131.Among them,120830 SNV/In Del were in exonic region.The number of nonsynonymous SNV were 65542,with 436 of frameshift-insertion for In Del and 897 of frameshift-deletion for In Del.GO analysis showed that 52 functions were correlated with ARDS development(p < 0.01),and KEGG enrichment analysis showed that these SNV/In Del were in 10 pathways,such as c GMP-PKG signaling pathway,Platelet activation(p < 0.05).3.SNV/In Del data between ARDS patients with different severity: ARDS patients were divided into severe ARDS group and nonsevere group according to the severity of lung injury.GO analysis showed that 25 functions were correlated with ARDS severity(p < 0.01),and KEGG enrichment analysis showed that these SNV/In Del were in 4 pathways,such as PI3K-Akt signaling pathway,ECM-receptor interaction(p < 0.05).4.SNV/In Del data between pulmonary ARDS and extrapulmonary ARDS patients: ARDS patients were divided into pulmonary ARDS and extrapulmonary ARDS group.GO analysis showed that 19 functions were correlated with pulmonary and extrapulmonary ARDS(p < 0.01),and KEGG enrichment analysis showed that these SNV/In Del were in 8 pathways,such as ECM-receptor interaction(p < 0.05).5.SNV/In Del data between ARDS combined with and without sepsis: ARDS patients were divided into ARDS combined with sepsis and ARDS without sepsis.GO analysis showed that 24 functions were correlated with ARDS combined with sepsis(p < 0.01),and KEGG enrichment analysis showed that these SNV/In Del were in 3 pathways,such as ECM-receptor interaction,Focal adhesion(p < 0.05).6.SNV/In Del data between ARDS combined with and without shock: ARDS patients were divided into ARDS combined with shock and ARDS without shock.GO analysis showed that 46 functions were correlated with ARDS combined with shock(p < 0.01),and KEGG enrichment analysis showed that these SNV/In Del were in 10 pathways,such as c AMP signaling pathway,ECM-receptor interaction(p < 0.05).Conclusion: In different subphenotypes of ARDS,the exome single nucleotide polymorphism was different.Part Two Latent class models of ARDS and the value of single nucleotide polymorphism in predicting phenotypes of ARDSObjective: To identify and describe the characteristics of different phenotypes of ARDS based on latent class analysis,and to understand the variances of single nucleotide polymorphism(SNP)in different ARDS phenotypes.Methods: 1.An investigator-initiated,multiple-centers,prospective trial which enrolled adult ARDS(according to Berlin definition)patients admitted to 11 intensive care units of China was undertaken.Characteristics including baseline information,vital signs and laboratory examinations during the first 24 hours were recorded.The response to therapy and in-hospital mortality were recored as well.Latent class modeling was performed to identify phenotypes using clinical data.The association of phenotypes with therapy and clinical outcomes were tested.2.A single-center,prospective trial which enrolled adult ARDS(according to Berlin definition)patients admitted to Intensive care unit,Zhongda Hospital,Southeast University was undertaken to understand the variances of exome SNP in different ARDS phenotypes.Peripheral blood samples were drawn.Whole-exome sequencing was performed by the sequencing platform Illumina,the data were compared with reference genome UCSC hg19.Patients were divided into different phenotypes according to the latent class analysis.Single nucleotide variants/ Insertion-deletion(SNV/In Del)were tested by plink method to understand the difference between the phenotypes.Results: 1.Baseline characteristics: A total of 353 ARDS patients were enrolled in the study,including 253 survivors and 100 nonsurvivors.Compared with survivors,the following parameters: age,APACHE II score,the percentage of chronic kidney disease and tumor,and pneumonia were significantly different with nonsurvivors.2.Latent class models identified two phenotypes within ARDS: Latent class models identified two phenotypes within ARDS according to the baseline characteristics.Compared with Phenotype 2,Phenotype 1 was defined by hyperinflammatory(higher levels of plasma procalcitonin and C-reactive protein)and hypoperfusion(higher doses of norepinephrine and lactate levels,and lower bicarbonate levels).3.Clinical characteristics and prognosis of different phenotypes: Compared with Phenotype 2,patients in Phenotype 1 received more invasive ventilation and lower tidal volume,higher PEEP and plateau pressure.More patients were responsing to lung recruitment.They also received more prone position,neuromuscular blocking drugs and extracorporeal membrane oxygenation,with higher mortality.4.SNV/In Del data between ARDS patients with different phenotypes: ARDS patients were divided into different phenotypes.GO analysis showed that 544 functions were correlated with ARDS phenotypes(p < 0.01),and KEGG enrichment analysis showed that these SNV/In Del were in 16 pathways,such as Vascular smooth muscle contraction,Endocytosis,ECM-receptor interaction(p < 0.05).Principal component analysis showed that SNV/In Del had a good intraclass correlation,and could be clustered.Conclusion: This analysis showed that two phenotypes within ARDS responded differently to therapy with different prognosis,the exome single nucleotide polymorphism was different.Part Three Lung-injury mutation burden in predicting phenotype and prognosis of patients with ARDSObjective: To identify the value of lung-injury mutation burden(LIMB)in predicting phenotype and prognosis of patients with ARDS.Methods: This was a single-center,prospective trial which enrolled adult ARDS(according to Berlin definition)patients admitted to Intensive care uint,Zhongda Hospital,Southeast University.Baseline characteristics including diagnosis,etiology of ARDS,and P/F ratio were recorded.Peripheral blood samples were drawn.Whole-exome sequencing was performed by the sequencing platform Illumina,the data were compared with reference genome UCSC hg19.Single nucleotide variants/ Insertion-deletion(SNV/In Del)were tested.The LIMB was calculated as nonsynonymous SNV number per megabase(MB)of DNA.Patients were divided into different subphenotypes according to the severity of lung injury,pulmonary or extrapulmonary,whether combined with sepsis and shock(Part 1).Meanwhile,patients were divided into different phenotypes by latent class modeling(Part 2).Prognosis was recorded as the survivors and nonsurvivors in hospital.The area under the receiver operating characteristic curve(ROC)was used for evaluating the predictive values of LIMB in predicting phenotype and prognosis of patients with ARDS.Results: 1.Baseline characteristics: A total of 105 ARDS patients were enrolled in the study,including 52 severe ARDS patients and 91 pulmonary ARDS patients.On enrollment,89 patients were combined with sepsis,and 66 patients were combined with shock.There were 68 patients enrolled in Phenotype 1,and 37 patients enrolled in Phenotype 2.Seventy patients were survived and 35 patients were not survived.2.The value of LIMB in predicting subphenotype and phenotype of patients with ARDS: Patients were divided into different subphenotypes according to the severity of lung injury,pulmonary or extrapulmonary,whether combined with sepsis and shock.Compared with nonsevere group,LIMB was lower in severe ARDS group,with the ROC of predictive value of 0.727(p < 0.0001).LIMB was not significantly altered between the pulmonary and extrapulmonary ARDS.Compared with patients without sepsis,the LIMB was lower in ARDS combined with sepsis,with the ROC of predictive value of 0.6803(p = 0.0084).Compared with patients without shock,the LIMB was lower in ARDS combined with shock,with the ROC of predictive value of 0.6915(p = 0.0008).Patients were divided into two phenotypes by latent class modeling,Phenotype 1 and Phenotype 2.Compared with patients with Phenotype 2,the LIMB was lower in Phenotype 1,with the ROC of predictive value of 0.6459(p = 0.0138).3.The value of predicting scores in predicting prognosis of patients with ARDS: Analysis was carried out on P/F ratio,APACHE II score,SOFA score and Murray lung injury score,with the ROC of prognosis predictive value of 0.568(p = 0.3124),0.6763(p = 0.0053),0.6204(p = 0.1002),0.6614(p = 0.0581),respectively.4.The value of LIMB in predicting prognosis of patients with ARDS: Compared with nonsurvivors,LIMB of survivors was significantly lower,with the ROC of predictive value of 0.6103(p = 0.0807),but could increase to 0.712(p = 0.001)when combined with APACHE II score.Conclusion: This analysis demonstrated that LIMB could be a new predictive biomarker for ARDS subphenotype,phenotype and prognosis.