| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |



*
Department of Nephrology, University Hospital, Nancy, France.
Department of Genetics, University Hospital, Nancy, France.
Department of Biostatistics, University Hospital, Nancy,
France.
§
Department of Nephrology, University Hospital Caen, France.
Correspondence to Dr. Luc Frimat, Service de Néphrologie, CHU de Nancy, Hôpitaux de Brabois, 54500 Vandoeuvre-les-Nancy, France. Phone: 33 3 83 15 31 69; Fax: 33 3 83 15 35 31; E-mail: l.frimat{at}chu-nancy.fr
| Abstract |
|---|
|
|
|---|
150 µmol/L and 24-P < 1 g), stage 2 (Scr
> 150 µmol/L and 24-P < 1 g or Scr
150 µmol/L and
24-P
1 g), stage 3 (Scr > 150 µmol/L and 24-P
1 g).
Deletion/insertion polymorphism (D/I) of the angiotensin I converting
enzyme gene, M235T polymorphism (T/M) of the angiotensinogen gene and
A1166C polymorphism (C/A) of the angiotensin II type 1 receptor gene
were determined in 274 Caucasian men with biopsy-proven IgAN (n = 86,
112, and 76 in stages 1, 2, and 3, respectively). Mean global follow-up was 6
± 5 yr after renal biopsy. For stages 1, 2, and 3, ESRF developed in 7
(8.1%), 39 (34.8%), and 49 (64.4%) cases (P < 0.0001), 11.7
± 4, 5.4 ± 4, and 2 ± 2 yr, respectively, after renal
biopsy (P < 0.001). The distributions of the three genotypes into
the three stages were similar. Different distributions were observed when
patients were grouped by stage and genotype: ID+DD: 72% in
stage 1 versus 84.6% in stages 2 + 3 (P = 0.02;
=
0.14); MT+TT: 66.2% in stages 1 + 2 versus 78.9% in
stage 3 (P = 0.04;
= 0.09); and AA+AC:
89.9% in stages 1 + 2 versus 97.4% in stage 3 (P = 0.04;
= -0.1). However, with the use of the Cox proportional hazard model,
none of the three genotypes was found to have predictive value for renal
survival. Compared with Scr and 24-P, genotypes DD,
TT, and AA are unlikely to serve as clinically useful
predictors of ESRF in IgAN. | Introduction |
|---|
|
|
|---|
In a multicentric prospective cohort study on IgAN with a mean 5.6-yr
follow-up after renal biopsy
(14), we demonstrated that
serum creatinine and 24-h proteinuria (24-P) at the time of renal biopsy
together with gender are three independent variables that are predictive of
end-stage renal failure (ESRF). Serum creatinine and 24-P are two commonly
described markers of prognosis
(15,
16). It has also been recently
demonstrated that renal disease progresses more slowly to ESRF in women than
in men (17). In previous work
(14), we established a
prognosis classification of ESRF for men with IgAN: stage 1 (serum creatinine
150 µmol/L and 24-P < 1 g), stage 2 (serum creatinine > 150
µmol/L and 24-P < 1 g or serum creatinine
150 µmol/L and 24-P
1 g), stage 3 (serum creatinine > 150 µmol/L and 24-P
1 g).
Seven years after renal biopsy, ESRF-free survival was 98.5% for stage 1,
86.6% for stage 2, and 21.3% for stage 3 (P < 0.001). In the
present study, we examined the hypothesis that ACE and/or Agt and/or ATR gene
polymorphisms correlate with phenotypic distribution in these three prognostic
stages.
| Materials and Methods |
|---|
|
|
|---|
To test the Hardy-Weinberg equilibrium, genotypic distributions were compared with those in a control group composed of 960 Caucasian men in the Stanislas cohort followed at the Nancy region preventive medicine center (18). In this group, the genotypic distributions were II/20.5%, ID/49.5%, and DD/30% for ACE and AA/49%, AC/42%, CC/9% for ATR.
Clinical Variables
We recorded age, history of gross hematuria, serum creatinine, 24-P,
systolic and diastolic BP, and antihypertensive therapy at the time of the
histologic diagnosis. For patients who progressed to ESRF, the date of onset
of substitution therapy was noted. For patients who did not develop ESRF,
serum creatinine, 24-P, systolic and diastolic BP, and antihypertensive
therapy at the time the genotype was determined were recorded. The duration of
the follow-up since the renal biopsy was determined. To take into account
changes in indications for renal biopsyfrom 1992, biopsies were
obtained only in patients with severe diseasea variable indicating the
period during which the biopsy was performed was included in the multivariate
analyses.
Definition of Outcome
This study was longitudinal and retrospective. ESRF was the outcome
criterion. The risk of ESRF was estimated from serum creatinine and 24-P at
the time of the renal biopsy. At the time of the genotyping, patients were
retrospectively classed into one of the three stages described above.
Genetic Study
Informed consent was obtained from each person. Nuclear DNA was extracted
with the Nucleon Bacc3 system resin (Amersham, England) which gives good
quality high-molecular-weight DNA from leukocyte extracts of 15ml peripheral
venous blood samples.
The D/I study was performed on a 287 base-pair (bp) fragment of the ACE gene intron 16 using in vitro elective amplification in the presence of Taq polymerase, genomic DNA, and specific primers described by Hunley et al. (11). Because of the preferential amplification of the D allele compared with the larger I allele, ID heterozygotes can be confused with DD homozygotes. To detect these false-positive DD, all of the samples with a DD phenotype were checked with PCR using a pair of primers specific for the insertion sequence whose amplification product is expected at 335 bp.
A search for Agt gene polymorphism was performed with the PCR method described by Russ et al. (19) using specific primers that produce a 165-bp fragment including a nonsense point mutation in exon 2 causing methionine (M) substitution for threonine (T) at position 235 (M235T).
The point mutation with adenosine (A) substitution for cytosine (C) in nucleotide 1166 in the untranslated 3' part of the ATR gene creates a restriction site for the Dde I enzyme. PCR amplification with specific primers produces a 546-bp fragment. Digestion with Dde I generates two 435- and 111-bp fragments uniquely in the presence of the C allele (20). For each of these three polymorphisms, control DNA samples from subjects with known genotype were systematically included in each experiment as were samples totally devoid of DNA.
Statistical Analyses
Deviation from Hardy-Weinberg equilibrium was assessed by goodness of fit
between the observed and expected number of each genotype with
2 test. Qualitative variables for the three stages were
compared with
2 test and continuous variables with ANOVA.
Kruskal-Wallis analysis was performed for variables with a non-normal
distribution.
The prognostic values of each of the three genotypes (ACE, Agt, ATR) were
established in two ways. First, we compared the stage distributions and the
genotype distributions using
2 test and the concordance of
these distributions with
test. Second, to assess the direct prognostic
impact of the genotypes on renal survival and to search gene-variable or
gene-gene interactions, the Cox model was used for univariate then
multivariate analyses. BMDP statistical software (Los Angeles, CA) was used
for all statistical analysis.
Sample Size
The number of subjects required for statistical significance was calculated
with Solo Power Analysis Software. Under the hypothesis that the genes are
independent, search for distribution concordance was done with
test
with an accepted
risk of 5% for a two-way test and 80% power.
Accepting a 90% concordance and
= 0.8, the necessary sample size was
270 patients. The second approximation estimated the number of subjects
necessary to evidence a relative risk corresponding to different renal
survival rates for the three stages. For the smallest estimated relative risk
with 80% power and
risk of 2% for a two-way test, the number of
subjects required was an estimated 500. Finally, we decided to set our
inclusion goal at 300 subjects (100 per stage).
| Results |
|---|
|
|
|---|
|
There was no departure from the Hardy-Weinberg equilibrium for genotypic distributions for ACE (P = 0.72) and ATR (P = 0.86).
As shown in Table 2, the
distributions of the three genotypes into the three prognostic stages were
similar. To achieve more statistical power,
2 tests were run
again after grouping stages and genotypes. Significant differences in
distributions (P < 0.05) were found for the groups presented in
Table 3. Compared with stages 2
and 3 grouped together, the proportion of genotype II in stage 1 was
higher than that of genotypes ID and DD together. The
proportion of genotypes MM and CC was higher than that of
genotypes MT and TT together and AC and AA
together in stages 1 and 2 taken together compared with stage 3. However, for
all of these comparisons, the observer agreement was slight or poor,
i.e., less than 20%, meaning that if there was any resemblance
between stage and genotype, it was weak.
|
|
Using Cox proportional hazards model, univariate analysis showed that ACE, Agt, and ATR did not have a significant prognostic impact. Stage alone remained as an independent variable after multivariate analysis using the same prognostic values as in our preceding study (age at biopsy, stage, systolic BP, gross hematuria, and the period variable). Compared with stage 1, the relative risk of ESRF was 5.5 (95% confidence interval [CI], 3.8 to 7.8) in stage 2 patients and 30.2 (95% CI, 14.4 to 60.8) in stage 3 patients. To assess directly the prognostic impact of genotypes on renal survival in this multivariate analysis, the stage variable was successively replaced by the ACE, Agt, and ATR variables. None of these three genotypes had any prognostic impact. In passing, an interaction between systolic BP and ATR was found, i.e., AA genotype patients with hypertension had a higher relative risk of ESRF than CC genotype patients with hypertension (P = 0.038). No genegene interaction was found.
The IGARAS study, designed to confirm or refute previous findings, yielded a highly significant negative result. Indeed, the power of the study to detect an underlying difference (ß error analysis) was greater than 80%.
| Discussion |
|---|
|
|
|---|
IgAN is one such multifactorial disease with a largely unexplained pathogenesis. The bias as a result of global assessment of glomerular diseases that result from different pathogenic mechanisms (23) was avoided in our study and in others (6,7,8,9,10,11,12) because only patients with IgAN were included.
Epidemiology and experimental studies have shown that the renal disease progresses more slowly to ESRF in women than in men (4,14,17). To avoid the potential confounding effect of gender on prognosis, the IGARAS study, unlike previous studies (6,7,8,9,10,11,12), included only men. Men in the three prognosis stages have very different relative risks for ESRF. Unrecognized stratification of the populations that possibly occurred in certain studies was avoided. Among the seven studies on the prognostic impact of gene polymorphism in IgAN, only one used ESRF as the end point (8). The others used slope of serum creatinine (6), loss of clearance of creatinine (7,9,12), doubling serum creatinine or ESRF (10), or a cutoff point for serum creatinine (11). In IgAN, the increase of serum creatinine and the decrease of clearance of creatinine are known to fluctuate with time. Estimations of prognostic impact are subsequently less precise when these variables are used as the end point (1,14,16,24).
In the IGARAS study, all patients who were taking a drug that interfered with 24-P (ACEi or AT1 blockers) were excluded. In addition, the study was based on a classification established at a period (1987 through 1996) when slower progression of chronic renal failure was not yet an indication for ACEi. Finally, it is reasonable to assume homogeneous salt intake in our Caucasian population from eastern and northern France. The environmental factors involved in IgAN were thus accounted for (21).
Before the IGARAS study, we estimated the number of subjects required to demonstrate a significant difference, taking into account the classification (14) and the distribution of the genotypes in our control population (18). This approach led us to include a large group of 274 patients to avoid the biases generated by small sample sizes in certain studies. In addition, the IGARAS study is in agreement with the Hardy-Weinberg equilibrium theoretical model. In 1997, Pei et al. (12) observed that only three (6,11,12) of the six studies devoted to D/I polymorphism in patients with progressing chronic renal failure were in agreement with the Hardy-Weinberg law. In the other studies, this bias could have caused less accurate assessment of prognostic markers (13).
The results of IGARAS study show that the distributions of the three genotypes for the three markers into the three prognosis stages were similar (Table 2). After grouping genotypes and stages, the proportions of genotypes II, MM, and CC were higher in stages with better prognosis (Table 3). However, the very small observer agreement is not in favor of a cause-and-effect relationship. The multivariate analysis demonstrated clearly that compared with proteinuria, high BP, and serum creatinine, genotypes have no independent prognostic value. Finally, our study suggests that in glomerular disease, the respective contributions of each of the three RAS gene polymorphisms to predicting progression to irreversible renal parenchymal damage are weak. These results confirm those reported by Schmidt et al. (8), who used the same end point and the largest study population. The earlier studies on ACE polymorphism reported seemingly contradictory results (6,7,8,9,10,11,12). Indeed, a meta-analysis of five studies (6,7,8,9,11) did not evidence any prognostic impact for genotypes ID and DD compared with genotype II (odds ratio and corresponding 95% CI, 0.7 [0.5 to 0.9], 1.2 [0.8 to 1.8]) (2). For Agt polymorphism, our study did not confirm the negative prognostic impact of genotype TT reported by Pei et al. (12). These differences could be explained by variations in their designs. Like two previous studies (11,12), our study did not evidence any impact of genotype AA on prognosis. Unlike ACE and Agt polymorphism, ATR polymorphism is considered to be nonfunctional (4). The weak synergetic effect of genotype AA and high BP on renal prognosis evidenced here was previously suspected (5).
In conclusion, the IGARAS study, designed to quantify the prognostic impact of RAS gene polymorphism in IgAN, has yielded negative results with powerful statistical significance. The relationships between genotype and phenotype explain, in part, the variance of serum ACE and Agt levels. This observation highlights only a restricted part of the overall participation of the RAS in the pathophysiology of chronic nephropathies. Furthermore, on the basis of a recent study (25), it seems that variations in ACE gene are much more complex than envisioned. Thus, although our results tend to exclude determined genetic differences of the RAS as factors that have an important impact on the progression of IgAN to ESRF, they do not exclude a causal role for the RAS in the pathogenesis of proteinuric nephropathies (1).
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Buraczynska, P. Ksiazek, A. Drop, W. Zaluska, D. Spasiewicz, and A. Ksiazek Genetic polymorphisms of the renin-angiotensin system in end-stage renal disease Nephrol. Dial. Transplant., April 1, 2006; 21(4): 979 - 983. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. C.-c. Hsu, M. S. Bray, W.H.L. Kao, J. S. Pankow, E. Boerwinkle, and J. Coresh Genetic Variation of the Renin-Angiotensin System and Chronic Kidney Disease Progression in Black Individuals in the Atherosclerosis Risk in Communities Study J. Am. Soc. Nephrol., February 1, 2006; 17(2): 504 - 512. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. K. Jacobsen Preventing end stage renal disease in diabetic patients -- genetic aspect (part I) Journal of Renin-Angiotensin-Aldosterone System, March 1, 2005; 6(1): 1 - 14. [Abstract] [PDF] |
||||
![]() |
L. Frimat and M. Kessler Controversies concerning the importance of genetic polymorphism in IgA nephropathy Nephrol. Dial. Transplant., April 1, 2002; 17(4): 542 - 545. [Full Text] [PDF] |
||||
![]() |
N. Lapointe and J.-L. Rouleau Activation of vascular tissue angiotensin-converting enzyme (ACE) in heart failure: Effects of ACE inhibitors J. Am. Coll. Cardiol., March 6, 2002; 39(5): 776 - 779. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |