Objective To analyze the medication rules of traditional Chinese medicine in treating breast cancer based on real-world data mining.Methods Inpatients with breast cancer who received traditional Chinese medicine treatment at the Affiliated Hospital of Chengdu University of Traditional Chinese Medicine from January 2017 to December 2021 were selected.Python 3.10 software was used to mine traditional Chinese medicine prescription data;SPSS 23.0 software was applied for descriptive analysis,and systematic cluster analysis was performed on high-frequency drugs.Results A total of 3026 consultation records of inpatients with breast cancer were collected.The main traditional Chinese medicine syndrome diagnosis of"predominantly liver depression and Qi stagnation"accounted for 60.94％of the total consultations.A total of 240 kinds of traditional Chinese medicine were used,with a cumulative frequency of 35 462 times.Among them,29 kinds of traditional Chinese medicine such as Danggui,Fuling,Baizhu,Chaihu had a cumulative usage frequency exceeding 300 times.Regarding the four natures of drugs,cold-natured(43.55％),warm-natured(30.05％),and neutral-natured(23.34％)drugs were predominant;In terms of five flavors,sweet(46.12％),bitter(30.91％),and pungent(20.02％)were the main ones.The most frequently used drugs were tonifying herbs(32.77％),followed by heat-clearing herbs(15.96％)and phlegm-resolving herbs(14.71％).Systematic cluster analysis yielded 7 groups of drug combinations.Conclusion In real-world clinical practice,traditional Chinese medicine for breast cancer mainly uses tonifying herbs,reflecting the traditional Chinese medicine principle of"strengthening healthy Qi and cultivating the root"in treating tumors.The four natures and five flavors of drugs follow syndrome differentiation and the combination of cold and heat.The clustered drug combinations have extensive therapeutic effects,covering various syndromes of breast cancer at different stages,which can provide a reference for clinical medication.

Objective To analyze the medication rules of traditional Chinese medicine in treating breast cancer based on real-world data mining.Methods Inpatients with breast cancer who received traditional Chinese medicine treatment at the Affiliated Hospital of Chengdu University of Traditional Chinese Medicine from January 2017 to December 2021 were selected.Python 3.10 software was used to mine traditional Chinese medicine prescription data;SPSS 23.0 software was applied for descriptive analysis,and systematic cluster analysis was performed on high-frequency drugs.Results A total of 3026 consultation records of inpatients with breast cancer were collected.The main traditional Chinese medicine syndrome diagnosis of"predominantly liver depression and Qi stagnation"accounted for 60.94％of the total consultations.A total of 240 kinds of traditional Chinese medicine were used,with a cumulative frequency of 35 462 times.Among them,29 kinds of traditional Chinese medicine such as Danggui,Fuling,Baizhu,Chaihu had a cumulative usage frequency exceeding 300 times.Regarding the four natures of drugs,cold-natured(43.55％),warm-natured(30.05％),and neutral-natured(23.34％)drugs were predominant;In terms of five flavors,sweet(46.12％),bitter(30.91％),and pungent(20.02％)were the main ones.The most frequently used drugs were tonifying herbs(32.77％),followed by heat-clearing herbs(15.96％)and phlegm-resolving herbs(14.71％).Systematic cluster analysis yielded 7 groups of drug combinations.Conclusion In real-world clinical practice,traditional Chinese medicine for breast cancer mainly uses tonifying herbs,reflecting the traditional Chinese medicine principle of"strengthening healthy Qi and cultivating the root"in treating tumors.The four natures and five flavors of drugs follow syndrome differentiation and the combination of cold and heat.The clustered drug combinations have extensive therapeutic effects,covering various syndromes of breast cancer at different stages,which can provide a reference for clinical medication.

Objective To investigate the value of serum concentration of VEGF-C in the prognosis of advanced pancreatic cancer. Methods Thirty-five patients with advanced pancreatic cancer were selected from Aug. 2006 to Feb. 2008, ELISA method was used to detect the serum level of VEGF-C, CA19-9 and KPS score was calculated, and survival was analyzed by Kaplan Meier method. The survival difference was calculated by log rank. Cox regression model was used to perform univariate and multivariate analysis. Results The mean serum concentration of VEGF-C was ( 1309 ± 542 ) pg/ml in patients with advanced pancreatic cancer, which were significantly higher than that those in normal control [ (278 ±115) pg/ml, P <0.01 ]. In Cox regression, KPS score, serum CA19-9 and VEGF-C were independent factors (x2 =7.208, 6.908, 3.867, P = 0.007, 0.009, 0.049). In multivariate analysis, serum VEGF-C and KPS score were independent factors (x2 =4.873, P=0.027, x2 =5.274, P =0.022). Using serum concentration of VEGF-C at 1280 pg/ml as the cut-off point, the mean survival of patients with VEGF-C ≤1280 pg/ml was 10.0 months, and the median survival was 11.3 months, 1 year cumulative survival was 50.0% ; while they were 6.0 months, 6.3 months and 5.9% in patients with VEGF-C > 1280 pg/ml, and the difference was statistically significant (x2 = 9.400, P= 0.002). Using KPS score 70 as the cut-off point, the mean survival of patients with KPS <70was 6.0 months, and the median survival was 6.6 months, 1 year cumulative survival was 21.4% ; while they were 9.0 months, 10.1 months, 33.3% in patients with KPS score ≥70,and the difference was statistically significant (x2 =4.040, P =0.044). The difference of the median survival, 1 year cumulative survival in patients with CA19-9 ≤200 U/ml or >200 U/ml was not statistically significant (10.0 months vs. 7.8 months, 37.5% vs. 21.1% ; x2 =1910, P=0. 167). Conclusions Serum concentration of VEGF-C can used as an independent factor for predication of prognosis of patients with advanced pancreatic cancer.