Association study of serum growth factors in patient with COPD

• Main Authors: Hong-Yih Tien, 田鴻毅
• Other Authors: Chong-Jen Yu
• Format: Others
• Language: zh-TW
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/a9evkc

碩士 === 國立臺灣大學 === 臨床醫學研究所 === 107 === Background: Chronic Obstructive Pulmonary Disease (COPD) is characterized by complex inflammatory, neuronal and fibrotic changes. Growth factor plays a key regulator of neuronal plasticity, and a crucial role in tissue repair and emphysema pathogenesis. The aims of this study was to determine which blood-based molecules (growth factors) have been evaluated as possible biomarkers to diagnose outcome of chronic obstructive pulmonary disease (COPD), to be correlated with lung function of forced expiratory volume in one second (FEV1% of predicted), and to predict risk of acute exacerbations. Methods: In the present study, patients over 40 and under 100 years of age with COPD (spirometry GOLD stages 1–4) were studied. Lung function, smoking history, medication, eosinophil of peripheral blood and serum concentrations of growth factors were assessed in all participants. There were 11 kinds of growth factors in this study, including NGF(Nerve growth factor-beta), BDNF(Brain-derived neurotrophic factor), EGF(Epidermal growth factor), FGF-2 (fibroblast growth factor-2), HGF( Hepatocyte growth factor), LIF(Leukemia inhibitory factor), PDGF (Platelet-derived growth factor), PLGF Placental growth factor), SCF(Stem cell Factor), VEGF-A and VEGF-D(Vascular endothelial growth factor-D). Spearman rank correlation was used to assess the relationship between growth factors and FEV1 predicted levels. The trend of levels of growth factors across COPD severity (GOLD grading or ABCD group classification) was assessed by the Jonckheere-Terpstra test. We also compared levels of growth factors between subgroups of FEV1% of predicted (< 50% vs. ≥ 50%), eosinophil (≥ 2% vs. < 2%), and frequent acute exacerbation (AE) in the previous year (0-1 vs. 2-3) by the Mann-Whitney U-test. The associations between growth factors and binary outcomes (e.g., GOLD grading 3-4, FEV1% of predicted <50%) were determined using multivariable logistic regression analysis with adjustment with sex, age, smoking status and body mass index. The performance of growth factors to discriminate frequent AE was evaluated by Receiver Operating Characteristic (ROC) curve analysis. The validation using outcome in the next year was also performed. Results: A total of 138 participants were enrolled who male is predominant (98.6%) and with a mean age of 68.3 years (standard deviation [SD], 9 years). The mean FEV1 predicted levels were 55.7% (SD, 18.4%). Number of GOLD stages 1–4 was 16 (11.6%), 67 (48.6%), 45 (32.6%) and 10 (7.2%) respectively. Number of COPD group classification A, B, C and D was 58 (42.0%), 28 (20.3%), 30 (21.7%), and 22 (15.9%) respectively. No significant correlation between the 11 growth factors and FEV1 predicted levels was found. There were no significant trend of levels of the 11 growth factors across GOLD grading or COPD group classification. There were no significant difference in levels of the 11 growth factors between subgroup of FEV1 predicted (< 50% vs. ≥ 50%) and eosinophil (≥ 2% vs. < 2%). Notably, serum levels of both FGF-2 and VEGF-D were significantly decreased in patients with frequency of acute exacerbations of COPD (AECOPD) compared with patients with infrequency of AECOPD (both P < 0.05) in previous year. The ROC curves showed the area under the curve (AUC) of FGF-2 was 0.704 (95% confidence interval [CI], 0.620 to 0.778, P = 0.010), the optimum cut point for Youden index was ≤4.76 with a sensitivity of 81.8% and a specificity of 62.2%. Another AUC of VEGF-D was 0.700 (95% CI, 0.616 to 0.775, P = 0.004), its Youden index is optimally ≤10.02 with a sensitivity of 100% and a specificity of 43.3%. After combining FGF-2 and VEGF-D in the logistic regression, the result showed that the AUC was 0.777 (95% CI, 0.697 to 0.844, P < 0.001). A total of 112 participants were enrolled for validation study in the next year. Higher FGF-2 level was associated with a lower risk of acute exacerbations of COPD one year later (odds ratio, 0.972; 95% CI, 0.949-0.997). The ROC curves showed the area under the curve (AUC) of FGF-2 was 0.614 (95% CI, 0.518 to 0.705, P = 0.032), the optimum cut point for Youden index was ≤9.12. Higher NGF was associated with a lower risk of frequent AE of COPD one year later(odds ratio, 0.93; 95% CI, 0.87-0.99).Similarly, using the ROC curve, the AUC of the NGF was 0.797 (P < 0.001), which reached an acceptable discriminating power for predict of frequency of acute exacerbations of COPD, the optimum cut point for the Younden index is ≤ 25.23. Conclusion: FGF-2 and VEGF-D concentrations are associated with increased frequency of acute exacerbations of COPD in previous year. FGF-2 was used to predict acute exacerbation of COPD after one year later. For predicting acute exacerbation, if combined with other growth factors (eg, FGF-2, VEGF-D, NGF, etc.), the prediction sensitivity should be improved. Therefore, measuring serum VEGF-D, FGF-2 and NGF concentrations in patients with COPD should be a sensitive indicator for predictive acute exacerbation of patients. Therefore, this may be used as a biomarker for COPD.