| Objective:To explore application of the growth model in the evaluation of efficacy of antihypertensive drug in clinical trials, and the associations between efficacy of antihypertensive drug and gene polymorphism of ACE (I/D) in Chinese.Methods:This paper discussed the merits and defects of biostatistical model used usually for evaluating efficacy of new drug through system review .The growth model was introduced to evaluate changes of ambulatory blood pressure in twenty-four hour under the antihytensive drug taken and on the basic of it,a new method was found to calculated the trough versus peak ratio (T/P ratio) .Associations between efficacy of antihypertensive drug and gene polymorphism of ACE (I/D) were explored by a meta-analysis and a example in clinical trial about ACE (I/D) .The seven literatures were selected by two people' score according the same standards .A new method for estimation of the combined effect-size estimates was adopted for meta-analysis.A randomized, double-blind, parallel-group, active-controlled clinical trial was carried out. 243 patients were enrolled and were randomly assigned to receive imidapril or Benazapril for 8 weeks. Polymerase chain reaction-restriction fragment length polymorphism(PCR-RFLP) was used to identify ACE-I/D polymorphism.Results:(1) The growth model (SBP=group+time+time*time+time* group+e ) was identified by the information of AIC and BIC (AIC=17442.8,BIC=17450.2,x~2 =1049.59, P <.0001 ).The result of test for the change of ambulatory blood pressure in average in both group was not significant (F=0.93, P=0.34) .but the results about order one or order two was significant (F =70.46, P=0.0001 and F=8.82, P=0.003 ), which shown blood pressure changed in different time .The interaction of time and group didn't exist.(2) The result from traditional method for T/P ratio: 24hs was divided into 12 time interval of 2 hours. The peak effect was assessed by concentrating on the interval 22h after dose administration, with the largest mean difference the trough blood pressure for each treatment was the difference in the change from baseline between 22h and 24h after dose administration. The better method for (T/P) ratio was calculated from the mean peak and trough values . For example , the t/p ratio for SBP and DBP were 43% and 59% respectively under the population way for group A, and for group B were 51% and 70% respectively with M±QR. The t/p ratio for SBP and DBP were 21%±41% and 20%±51% respectively under the individual way for group A, and for group B were 30%±55% and 4%±63% respectively with M±QR.(3) The result from new method for T/P ratio on the basic of growth model were listed as following: The t/p ratio for SBP and DBP were 80% and 64% respectively under the population way for group A, which for group B were 93% and 63% respectively. The t/p ratio for SBP and DBP were 10%±71% and 13%±90% respectively with M±QR under the individual way for group A, and for group B were 34%±86% and 23%±46% respectively with M±QR. The result under population way was larger than 50%.(4) The seven literature reports were included out and every study included three genotype, that is II, ID and DD .The decreases in diastolic blood pressure and systolic blood pressure reflect the clinical effects for antihypertensive drug. The total of patients from all studies was 2589.There were 936 patients in genotype II, 1089 patients in genotype ID and 564 patients in genotype DD respectively. The result of homogeneity test of variance from individual effect of study show that were significant for four studies, others were not significant.Assuming equal the two group variance, the combined effect-size estimates was-0.02-(-0.07, 0.10) for genotype ID versus II and-0.06- (-0.05, 0.17) for genotype DD versus II which shown the three genotype are similar in the reflection of clinical effects of antihypertensive drug (Q=3.94, P=0.98) . Assuming unequal the two group variance, the combined effect-size estimates was-0.04- (-0.05, 0.13) for genotype ID versus II and 0.32- (0.21, 0.42) for genotype DD versus II which shown the three genotype are different in the reflection of clinical effects of antihypertensive drug (Q=52.50, P=0.00) .Assuming independent effect-size estimates, the combined effect-size estimates was0.07-(-0.02, 0.17) for genotype ID versus II and 0.372- (0.27, 0.48) for genotype DD versus II which shown the three genotype are different in the reflection of clinical effects of antihypertensive drug.(5) 243 patients' genotypes were examined and divided into three kinds of genotype, 65 of them is genotype DD, 63 of them is II and others of them is ID. The frequencies of alleles for D is 50.41% and for I is 49.59%, which meet the equilibrium of Hardy-Weinberg.On the end ,the test for the decrease of systolic blood pressure and diastolic blood pressure in the three kinds of genotype were significant (F=3.26, P=0.04 and F=3.19, P=0.04 ) , and the test for the decrease of systolic pressure and diastolic pressure in the three genotype were significant (F=260.72, P=0.00 and F=25.26, P=0.00) ,which shown the three genotype are different in the reflection of clinical effects of antihypertensive drug.Conclusions: 1,The growth model may be introduced to evaluate the efficacy of antihypertensive drug in clinical trials, especially for the assessment of ambulatory blood pressure. Compared with the traditional biostatistical models, the growth model has some advantages on evaluating the efficacy of antihypertensive drug in clinical trials, which were listed as following:(1) The growth model may cope with missing data or from unbalance design, or unequal of variance or periods for records blood pressure.(2) The growth model may fit the relationships between blood pressure and twenty-four hours.(3) The growth model may reflect the between effect and within effect.(4) The growth model was implemented easily by SAS software.2,The new method on the basis of growth model can be adopted in calculating T/P ratio index and he result is reasonable in the main.3,The combined estimates of multiple correlated effect-sizes was met in meta analysis for multiple treatment, and three genotypes are different in the reflection of clinical effects of antihypertensive drug, which of genotype DD outweighed that of genotype II through an clinical trial about ACE.
|
PDF Full Text Request