The novel proteomic signature for the detection of cardiac allograft vasculopathy

Background

Cardiac allograft vasculopathy (CAV) is the major long-term complications after heart transplantation, leading to mortality and re-transplantation. As available noninvasive biomarkers are scarce for CAV screening, we aimed to identify a proteomic signature for CAV detection.

Methods

Urinary proteome was measured by capillary electrophoresis coupled to mass spectrometry in 217 heart transplantation recipients. Participants were further randomly and evenly divided into the derivation cohort and validation cohort. The proteomic signature for CAV was identified by decision tree-based machine learning in the derivation cohort and further tested in the validation cohort. The pathway analysis was investigated with Reactome Pathway Database.

Results

We identified a proteomic signature with 27 urinary peptides, which yielded areas under the curve (AUC) of 0.83 and 0.71 in the derivation and validation cohort, respectively. In the validation cohort, it had a sensitivity of 68.4%, specificity of 73.2%, accuracy of 71.6%, negative predictive value of 81.3%. Including the proteomic signature into the basic model further improved the diagnostic accuracy with an relative integrated discrimination improvement of 25.9% and the continuous net reclassification improvement of 83.3% (p≤0.023). The pathways analysis on revealed that collagen turnover, platelet aggregation and coagulation, cell adhesion and motility might involve in the pathogenesis of CAV.

Conclusions

The proteomic signature might be valuable for the surveillance of CAV thereby reduce the frequency of invasive procedures after HTx. Moreover, the highlighted pathways might provide insights in the potential novel treatment targets for CAV.

Funding Acknowledgement

Type of funding sources: Public grant(s) – EU funding. Main funding source(s): European Research Council Advanced Researcher Grant and Proof-of-Concept Grant ROC curves of the urinary proteomicThe 25 highlighted enrichment pathways

[Analysis of prognostic factors for hyperamylasemia following pancreaticoduodenectomy]

Objective: To investigate the prognostic factors of hyperamylasemia following pancreaticoduodenectomy (PD) . Methods: Clinical data of 359 patients were collected prospectively who underwent PD by the same group at Changhai Hospital of Navy Medical University from January 2017 to June 2018.There were 212 males and 147 females.The median age was 63 years old (range: 23 to 82 years old) .According to whether the patient's serum amylase was greater than 120 U/L at 0 or 1 day after surgery,the patients were divided into hyperamylasemia group and non-hyperamylasemia group. Univariate analysis and multivariate analysis were used to find out the prognostic factors of hyperamylasemia after PD. Results: Of the 359 patients, 238 cases (66.3%) developed hyperamylasemia.The incidence rate of clinically related pancreatic fistula (15.1% vs.2.5%, P<0.01) , grade B/C post pancreatectomy hemorrhage (8.8% vs. 2.5%, P<0.01) , and surgical site infection (9.2% vs. 3.3%, P=0.04) was significantly higher in the hyperamylasemia group.The severity of complications (CD grade≥Ⅲ: 11.3% vs.4.1%, P=0.023) and postoperative hospital stay (11 days vs. 9 days, P=0.001) were higher in the hyperamylasemia group.In the multivariate analysis, the main pancreatic duct diameter (MPD) ≤3 mm (OR=4.469, 95% CI: 2.563-7.793, P<0.01) , pathological type of disease (pancreatic cancer or pancreatitis) (OR=0.230, 95% CI: 0.122-0.436, P<0.01) and soft texture of pancreas (OR=3.297, 95%CI: 1.930-5.635, P<0.01) were independent prognostic factors for hyperamylasemia. Conclusions: Post-PD hyperamylasemia increased the incidence and severity of postoperative complications after PD.MPD≤3 mm, soft texture of pancreas and pathological type of disease were independent prognostic factors of hyperamylasemia.

["Effort index"--estimating abilities of respiratory muscles]

This paper analyzes the actual values of FEV1 and MBC in 240 normal subjects, 147 patients with COPD and 85 cases of cancer tumors, and shows that the actual values of MBC in both COPD patients and cancerous cases are much lower than the calculated values from FEV1. The authors believe that the ratio between the actual values of MBC and its calculated values may be used to reflect the strengths or abilities of the respiratory muscles efforted, called "Effort Index". The formula of MBC calculated from FEV1 is based on the linear regression equation formulated from the actual values of FEV1 and MBC in 240 normal subjects (aged 16-80 yrs). The "Effort Index" is equal to the actual value of MBC/the calculated value of MBC. The "Effort Index" in 240 normal subjects is 1.00 +/- 0.14 (M +/- SD), in mild, moderate and severe cases of COPD being 0.907, 0.807 and 0.626 respectively, and in markedly wasted cancer cases, 0.861 +/- 0.130. The results show that the formula of MBC calculated from FEV1 is useful only for the subjects with normal respiratory muscular abilities; while in COPD patients and markedly emaciated cancerous cases, owing to the chronically tired-out or exhausted respiratory muscular strength, the actual values of MBC are actually lower than those calculated from FEV1, hence the decreased "Effort Index".