sign in sign up
<<
Home , Benazepril

Renoprotection of Optimal Antiproteinuric Doses (ROAD) Study: A Randomized Controlled Study of Benazepril and in Chronic Renal Insufficiency

Fan Fan Hou,* Di Xie,* Xun Zhang,* Ping Yan Chen,† Wei Ru Zhang,* Min Liang,* Zhi Jian Guo,* and Jian Ping Jiang* *Renal Division, Nanfang Hospital, and †Department of Biostatistics, Southern Medical University, Guangzhou, People’s Republic of China

The Renoprotection of Optimal Antiproteinuric Doses (ROAD) study was performed to determine whether titration of benazepril or losartan to optimal antiproteinuric doses would safely improve the renal outcome in chronic renal insufficiency. A total of 360 patients who did not have diabetes and had proteinuria and chronic renal insufficiency were randomly assigned to four groups. Patients received open-label treatment with a conventional dosage of benazepril (10 mg/d), individual uptitration of benazepril (median 20 mg/d; range 10 to 40), a conventional dosage of losartan (50 mg/d), or individual uptitration of losartan (median 100 mg/d; range 50 to 200). Uptitration was performed to optimal antiproteinuric and tolerated dosages, and then these dosages were maintained. Median follow-up was 3.7 yr. The primary end point was time to the composite of a doubling of the serum creatinine, ESRD, or death. Secondary end points included changes in the level of proteinuria and the rate of progression of renal disease. Compared with the conventional dosages, optimal antiproteinuric dosages of benazepril and losartan that were achieved through uptitration were associated with a 51 and 53% reduction in the ,and 0.022, respectively). Optimal antiproteinuric dosages of benazepril and losartan 0.028 ؍ risk for the primary end point (P at comparable BP control, achieved a greater reduction in both proteinuria and the rate of decline in renal function compared with their conventional dosages. There was no significant difference for the overall incidence of major adverse events between groups that were given conventional and optimal dosages in both arms. It is concluded that uptitration of benazepril or losartan against proteinuria conferred further benefit on renal outcome in patients who did not have diabetes and had proteinuria and renal insufficiency. J Am Soc Nephrol 18: 1889–1898, 2007. doi: 10.1681/ASN.2006121372

nterruption of the - axis with an angio- titrating ACEi or ARB to higher dosages is effective at reducing tensin-converting enzyme inhibitor (ACEi) or an angioten- proteinuria (12–15), although there have been conflicting re- I sin II receptor blocker (ARB) slows the progression of ports (16,17). Taken together, these data suggest that the rec- chronic renal insufficiency in the presence or absence of diabe- ommended dosages of ACEi or ARB in current practice, which tes (1–5). However, many patients progress to ESRD despite are based on their BP-lowering effect, might be inadequate to using an ACEi or ARB (1–5). Optimizing renin-angiotensin halt satisfactorily renal progression. However, most of the dos- system (RAS) blockade to provide more effective renal protec- age-response studies to date looked exclusively at reductions in tion has received considerable attention. Increasing evidence proteinuria but failed to evaluate hard end points as the out- suggests that, in addition to effective BP control, proteinuria come parameter. This is partly explained by the short observa- should be considered as an independent risk factor and an tion period (between 8 wk and 9 mo) in these studies. Optimal essential treatment target for renal protection (2,6,7). Reduction antiproteinuric dosages of ACEi or ARB have not been ex- of proteinuria with RAS blockade results in protection against plored thoroughly with respect to long-term renal outcomes protein-induced renal fibrosis, which should translate into and, importantly, tolerability and safety in patients with renal preservation of renal function in the long term (8,9). Animal insufficiency. studies have demonstrated that maximal renal benefit from Several studies have provided evidence that ARB are reno- ACEi or ARB requires higher dosages than those needed to protective in patients with type 2 diabetes (3,4,18). It is to be normalize BP (10,11). Small clinical studies have shown that confirmed whether an ARB is as effective as an ACEi in retard- ing progression in nondiabetic nephropathy, because data from a trial that was designed primarily to compare the effects of Received December 19, 2006. Accepted March 27, 2007. combined therapy with ACEi and ARB to each monotherapy

Published online ahead of print. Publication date available at www.jasn.org. suggest that there is no difference in renal protection between ACEi and ARB (19). Address correspondence to: Dr. Fan Fan Hou, Renal Division, Nanfang Hospital, 1838 North Guangzhou Avenue, Guangzhou 510515, People’s Republic of China. Phone: The primary aim of this study was to evaluate whether the ϩ86-20-61641591; Fax: ϩ86-20-87281713; E-mail: [email protected] optimal antiproteinuric dosages of benazepril (an ACEi) or

Copyright © 2007 by the American Society of Nephrology ISSN: 1046-6673/1806-1889 1890 Journal of the American Society of Nephrology J Am Soc Nephrol 18: 1889–1898, 2007 losartan (an ARB), as compared with their conventional dos- entered the titration period. Patients in groups 1 and 3 remained on ages, can safely improve the renal outcome in patients who do fixed dosages of benazepril (10 mg/d) or losartan (50 mg/d), respec- not have diabetes and have proteinuria and chronic renal in- tively. Patients in groups 2 and 4 had their dosages of benazepril or sufficiency. The second aim was to compare the long-term losartan uptitrated as follows: Patients in group 2 received monthly uptitration of benazepril from a starting dosage of 10 mg/d to 20, 30, renoprotection of benazepril and losartan at their optimal an- and 40 mg/d. Patients in group 4 were treated with increasing dosages tiproteinuric dosages. of losartan (from 50 to 100, 150, and 200 mg/d) with each titration period of 4 wk. BP (measured with a mercury-column sphygmoma- Materials and Methods nometer in the sitting position 3 to 4 h after the administration of the Design and Patients study drug) was measured every week by staff who were blind to the Renoprotection of Optimal Antiproteinuric Doses (ROAD) was a study design and recorded as mean of three readings. Urinary protein prospective, randomized, open, blinded end point (PROBE) study that excretion, serum creatinine, and potassium levels were measured every was conducted at Nanfang Hospital Renal Division. An adjudicating 2 wk during the titration period. The dosage of study drug was down- group, whose members were unaware of patients’ treatment assign- titrated to the previous level in the following circumstances: The anti- ments, reviewed the data to determine which patients had reached proteinuric effect of the study drug had reached its plateau (i.e., urinary study end points and to evaluate safety. The study protocol was ap- protein excretion did not fall by Ն10% [22] versus the previous titration proved by the Nanfang Ethics Committee, and all patients provided period [determined by two values that were obtained 4 wk apart at written informed consent. A study period of 3 yr was chosen on the same dosage; the accuracy of urine collection was confirmed by deter- basis of the results of previous trials that involved patients with non- mination of specimen’s creatinine content]), or SBP decreased to Ͻ120 diabetic chronic nephropathy and that suggested that a 3-yr follow-up mmHg despite withdrawal of all additional antihypertensive medica- is adequate to assess efficacy (2,5). tion. Patients who developed (serum potassium Ն6.0 Between January 2002 and May 2003, consecutive patients who were mmol/L) that was refractory to medical treatment or whose serum aged 18 to 70 yr and had chronic disease (CKD) were screened creatinine increased Ͼ30% versus previous values had the study drug for the study at the Nanfang Hospital Renal Division, which has a dosage downtitrated to the previous step or withdrawn. In patients catchment that includes eight cities near Guangzhou with a total pop- who were not responsive to uptitration (defined as Ͻ10% reduction in ulation in the year 2000 of 29.8 million. Eligible patients had not proteinuria from the levels at the start of titration), the study drug was received ACEi or ARB for at least 6 wk before screening and met the uptitrated monthly to the maximum licensed dosage (benazepril 40 following inclusion criteria: A serum creatinine level of 1.5 to 5.0 mg/dl mg/d in group 2; losartan 200 mg/d in group 4). When still no (133 to 442 ␮mol/L) and a creatinine clearance (20) of 20 to 70 ml/min response was observed, the dosage was decreased to the starting dos- per 1.73 m2, with variations of Ͻ30% in the 3 mo before screening; age; patients remained in the study and were included in the analyses. diagnosis of nondiabetic renal disease (as established on the basis of Patients in groups 2 and 4 were maintained on their study drug once patient history and as a result of serum biochemical tests and renal their individual optimal antiproteinuric and tolerated dosage was biopsy); and persistent overt proteinuria (defined as urinary protein achieved. Additional antihypertensive agents (as described previously) excretion of Ͼ1.0 g/d for Ն3 mo without evidence of urinary tract were administered or withdrawn, as required, to maintain BP control. infection or overt [a New York Heart Association class of The patients (including withdrawn cases) were followed up every III or IV]). Exclusion criteria were an immediate need for dialysis; month thereafter for data collection. current treatment with corticosteroids, nonsteroidal anti-inflammatory All patients were instructed to reduce their salt intake to approxi- drugs, or immunosuppressive drugs; hyper- or hypokalemia (serum mately 5 to 7 g/d, to eat 0.5 to 0.7 protein/kg body wt per d, and to potassium concentration Ն5.6 or Յ3.5 mmol/L); renovascular disease; restrict their intake of foods that are rich in potassium. Dietary com- myocardial infarction or cerebrovascular accident in the year preceding pliance was assessed by evaluation of 24-h urinary excretion of urea the trial; connective-tissue disease; and obstructive uropathy. and chloride.

Randomization and Interventions Outcome Variables Eligible patients got their sequence numbers from the coordinator The primary efficacy variable was time to the first event for the and were randomly allocated into four groups according to a comput- composite end point: Doubling of the serum creatinine concentration, er-generated randomization sequence list using a blocking size of 8. ESRD, or death. Doubling of serum creatinine concentration from the The list was prepared by the Department of Biostatistics, Southern baseline value was confirmed by a second serum creatinine value that Medical University. For safety’s sake, all patients entered an 8-wk was obtained at least 4 wk after the initial doubling. ESRD was defined pretitration phase in which patients in groups 1 and 2 received 10 mg by the need for long-term dialysis or renal transplantation. of benazepril (1) and those in groups 3 and 4 received 50 mg of losartan Secondary end points included changes in urinary protein excretion daily (21). After 4 wk of therapy with the study drugs, patients who rate and the progression of renal disease as assessed by estimated GFR i.e. continued to show inadequate BP control ( , a systolic BP [SBP] of (eGFR) that was calculated by Modification of Diet in Renal Disease Ͼ Ͼ 130 mmHg and/or a diastolic BP of 80 mmHg) had an additional (MDRD) equation (23). To avoid the imprecision of the MDRD equation ␤ antihypertensive agent (diuretic, , blockers, in Chinese, we also included evaluated creatinine clearance (24) to centrally acting agent, or combination of these medications, excluding reflect the rate of decline in renal function. ACEi and ARB) added to their treatment regimen. All patients were under close observation, including weekly measurements of BP, serum creatinine, and serum potassium. Statistical Analyses At completion of the pretitration phase, patients with stable renal The sample size was estimated before the study with the nQuery function (Ͻ30% increase from inclusion baseline in serum creatinine, Advisor software 5.0 (Statistical Solutions Ltd., Cork, Ireland). Our confirmed by at least three separate measurements) and serum potas- preliminary study in patients with renal insufficiency showed that the sium levels Ͻ5.6 mmol/L and without other adverse events then rate of the primary end point among patients who were treated with 10 J Am Soc Nephrol 18: 1889–1898, 2007 Benazepril and Losartan in Chronic Renal Insufficiency 1891 mg of benazepril was approximately 33%, whereas in those who re- Results ceived 20 mg of benazepril, 23% reached the primary end point (25). A total of 406 patients were screened during January 2002 to Thus, we estimated that the optimal antiproteinuric dosage of benaz- May 2003, and 360 were identified for study participation and epril would reduce this rate to 10%. The enrollment of 70 patients in randomly assigned to the four groups (Figure 1). The baseline each group of benazepril arm would provide the study with a statistical characteristics of the patients were similar in the four groups power of 80% at a two-sided significance level of 0.05. Because no data on long-term renal effect of ARB in patients with (Table 1). The mean age and the distribution of the primary nondiabetic nephropathy were available at the start of the study, we causes of renal dysfunction across the groups were similar to assumed that treatment with an optimal antiproteinuric dosage of those reported in the registry of the Chinese Society of Nephrol- losartan would also reduce the rate of the primary end point to 10% ogy (26). The median length of follow-up was 3.7 yr (min-max, versus the conventional dosage. It was estimated that to give the study approximately 2.0 to 4.5). an 80% power to detect statistical significance, at least 70 patients in Fifty patients withdrew during the pretitration phase with each group of the losartan arm had to complete the study. main cause of dry cough (30 [60%] of 50). Seventeen of them The primary and secondary end points were analyzed according to were lost to follow-up: Eight because of poor adherence (one in the intention-to-treat (ITT) principle. We included data from all patients group 1, two in group 2, two in group 3, and three in group 4), who entered active treatment. The missing data were treated using lost and nine because of refusal to continue the treatment (five in operation carry forward principle to satisfy ITT analysis. For further group 1 and four in group 2; Figure 1). Four patients were lost confirmation of the result, the primary end points were also analyzed by using the per-protocol principle. The Cox regression model based on to follow-up after the pretitration phase because they moved the assessment of goodness-of-fit by Ϫ2 log likelihood ratio was used to away (one in group 1 and three in group 4; Figure 1). determine the hazard ratio for the primary end point. The model was also adjusted by baseline variables as the covariables, which were Primary Outcomes proteinuria, eGFR, and SBP, at end point to detect their effect. The risk In the benazepril arm, 15 of 84 patients who were treated ϫ Ϫ reduction was calculated as 100% (1 hazard ratio). Event curves with optimal antiproteinuric dosages (group 2) reached the were based on Kaplan-Meier analysis, and significance was assessed by primary end point, compared with 26 of 83 patients who re- log-rank test. ceived the conventional dosage (group 1; 17.9 versus 31.3%; P ϭ Changes in urinary protein excretion, creatinine clearance, and BP 0.025; Figure 2A). Likewise, in the losartan arm, 13 of 84 pa- were analyzed by repeated-measures ANOVA. Multiple comparisons were conducted with LSD t test when ANOVA was significant. The tients who received optimal antiproteinuric dosages (group 4) relationship between proteinuria reduction and the rate of the decline reached the primary end point, as compared with 26 of 88 of renal function was analyzed by Pearson correlation. Two-tailed P Յ patients who received the conventional dosage (group 3; 15.5 0.05 were considered statistically significant. Analyses were performed versus 29.5%; P ϭ 0.022; Figure 2A). Treatment with optimal with SPSS 13.0. antiproteinuric dosages of benazepril or losartan, compared

Figure 1. Study flow diagram. 1892 Journal of the American Society of Nephrology J Am Soc Nephrol 18: 1889–1898, 2007

Table 1. Baseline characteristics of the patientsa

Benazepril Losartan Characteristic Group 1 Group 2 Group 3 Group 4 (n ϭ 90) (n ϭ 90) (n ϭ 90) (n ϭ 90)

Age (yr; mean Ϯ SD) 51.9 Ϯ 12.6 49.1 Ϯ 14.3 51.5 Ϯ 13.3 51.0 Ϯ 13.5 Male gender (n ͓%͔) 59 (66) 56 (62) 56 (62) 55 (61) BMI (kg/m2; mean Ϯ SD) 22.5 Ϯ 3.9 22.8 Ϯ 4.2 22.7 Ϯ 5.8 23.3 Ϯ 3.1 Body surface area (m2; mean Ϯ SD) 1.7 Ϯ 0.2 1.7 Ϯ 0.2 1.7 Ϯ 0.2 1.7 Ϯ 0.2 Renal disease (n ͓%͔)b glomerular 52 (58) 56 (62) 56 (62) 61 (68) 23 (26) 26 (29) 21 (23) 17 (19) polycystic kidney disease 6 (7) 3 (3) 2 (2) 2 (2) interstitial 8 (9) 5 (6) 9 (10) 8 (9) unknown 1 (1) 0 (0) 2 (2) 2 (2) Hypertension (n ͓%͔) 58 (64) 57 (63) 58 (64) 56 (62) No. of antihypertensive drugs median 2222

Q1 to Q3 0to3 0to3 0to3 0to3 BP (mmHg; mean Ϯ SD) systolic 150.7 Ϯ 26.6 148.9 Ϯ 29.7 149.3 Ϯ 24.5 152.0 Ϯ 22.1 diastolic 86.2 Ϯ 17.3 86.3 Ϯ 15.7 86.1 Ϯ 13.5 86.1 Ϯ 13.6 Renal function (mean Ϯ SD)c serum creatinine (mg/dl) 2.7 Ϯ 0.9 2.8 Ϯ 0.9 2.8 Ϯ 1.1 2.9 Ϯ 1.0 eGFR (ml/min per 1.73 m2) 30.6 Ϯ 11.3 30.5 Ϯ 14.0 31.4 Ϯ 14.1 29.9 Ϯ 12.4 creatinine clearance (ml/min per 1.73 m2) 33.9 Ϯ 14.7 35.1 Ϯ 12.2 34.4 Ϯ 15.5 33.8 Ϯ 14.0 urinary protein excretion (g/d)

median (Q1 to Q3) 1.4 (1.6 to 2.1) 2.1 (1.8 to 2.4) 1.6 (1.6 to 2.3) 2.0 (1.8 to 2.3) Laboratory variables (mmol/L; mean Ϯ SD)d serum cholesterol 5.4 Ϯ 1.6 5.3 Ϯ 1.5 5.5 Ϯ 1.4 5.4 Ϯ 1.5 serum triglycerides 1.9 Ϯ 1.5 1.9 Ϯ 1.4 2.0 Ϯ 1.2 2.1 Ϯ 1.7 serum potassium 4.2 Ϯ 0.5 4.3 Ϯ 0.6 4.2 Ϯ 0.5 4.3 Ϯ 0.5 Hemoglobin (g/L; mean Ϯ SD) 105.0 Ϯ 24.1 106.5 Ϯ 20.9 105.7 Ϯ 20.9 105.6 Ϯ 19.7 Urinary urea excretion (g/d; mean Ϯ SD) 5.2 Ϯ 1.9 5.2 Ϯ 1.7 5.1 Ϯ 1.7 5.2 Ϯ 1.4 Urinary chloride excretion 106.3 Ϯ 14.0 107.1 Ϯ 12.6 106.4 Ϯ 13.4 107.7 Ϯ 12.3 (mmol/d; mean Ϯ SD)

a There were no significant differences in the groups. eGFR, estimated GFR; Q1, 25th percentile; Q3, 75th percentile. bAs established on the basis of previous medical records (including renal biopsy) and as a result of current clinical evaluation (e.g., ultrasound and computerized tomography examination). cTo convert values for creatinine to mmol/L, multiply by 88.4. Serum and urinary creatinine levels were determined by an enzymatic method (sarcosine oxidase-peroxidase-antiperoxidase). The eGFR was calculated with the use of the four- Ϫ component Modification of Diet in Renal Disease (23) equation: 186 ϫ (serum creatinine level ͓in mg/dl͔) 1.154 ϫ (age ͓in Ϫ years͔) 0.203. For women, this equation was multiplied by a correction factor of 0.742. Urinary protein excretion was determined by biuret method (24) using 24-h urine samples. dTo convert values for cholesterol to mg/dl, divide by 0.02586. To convert values for triglycerides to mg/dl, divide by 0.01129.

with their relevant conventional dosage, resulted in overall for baseline eGFR (P ϭ 0.039 in the benazepril and 0.035 in the reduction in the risk for a primary end point by 51% (95% losartan arm). The ITT analyses of the individual components confidence interval [CI] 4.8 to 73.3) in the benazepril arm and of the primary end point indicated that the risk for doubling of 53% (95% CI 5.5 to 74.1) in the losartan arm (P ϭ 0.028 and serum creatinine level was 49% lower in the benazepril arm and 0.022, respectively). The decrease in the risk remained signifi- 50% lower in the losartan arm among patients who received cant after adjustment for the SBP at end point (hazard ratio optimal antiproteinuric dosages than among those who were 50.5% in benazepril and 47.6% in the losartan arm; P ϭ 0.030 given the conventional dosage (P ϭ 0.041 and P ϭ 0.040, and 0.031, respectively), for baseline urinary protein excretion respectively). Use of the optimal antiproteinuric dosage of (P ϭ 0.037 in the benazepril and 0.039 in the losartan arm), and benazepril or losartan also reduced the risk for ESRD by 47% J Am Soc Nephrol 18: 1889–1898, 2007 Benazepril and Losartan in Chronic Renal Insufficiency 1893

Figure 2. Kaplan-Meier estimates of the percentage of patients who reached the primary composite end point of a doubling of the serum creatinine level, ESRD, or death according to intention-to-treat (A) or per-protocol principal (B).

(95% CI 4.2 to 72.1) in the benazepril arm and 47% (95% CI 3.6 epril and losartan arms at both conventional and optimal anti- to 76.9) in the losartan arm (P ϭ 0.042 and P ϭ 0.046, respec- proteinuric doses (P Ͼ 0.05). tively). The per-protocol analyses of the primary end points showed the similar results (Figure 2B). There was no statisti- cally significant difference between benazepril and losartan in Dosage-Response Relationship Optimal antiproteinuric efficacy of benazepril was obtained the overall relative risk reduction at their respective optimal with a 20-mg dosage in 43 (61%) patients, 30-mg dosage in 11 antiproteinuric dosages or at conventional dosages. (16%) patients, 40-mg dosage in three (4%) patients, and Ͼ40-mg dosage (i.e., proteinuria reduction Ͼ10% of previous Secondary Outcomes level at maximum dosage of 40 mg) in three (4%) patients. The In both the benazepril and losartan arms, there was a signif- mean dosage at the end of the uptitration phase was 20.8 mg icantly greater reduction in the level of proteinuria among (SD 7.4). Dosage titration with losartan revealed a similar pat- patients who were given optimal antiproteinuric dosages than tern. Optimal antiproteinuric efficacy was obtained with a conventional dosages (Figure 3A). The median of final changes 100-mg dosage in 48 (57%) patients, 150-mg dosage in 12 (14%) from baseline in benazepril arm at 4, 12, 24, and 36 mo was 51, patients, 200-mg dosage in nine (11%) patients, and Ͼ200-mg 50, 53, and 50% in group 2 and 37, 35, 36, and 38% in group 1 dosage (i.e., proteinuria reduction Ͼ10% of previous level at (P Ͻ 0.05). Among the patients who received losartan, the maximum dosage of 200 mg) in three (4%) patients. The mean changes of urinary protein excretion at 4, 12, 24, and 36 mo dosage at the end of the uptitration phase was 117.7 mg (SD were 52, 51, 52, and 53% in group 4 and 36, 36, 39, and 41% in 42.6). Four (6%) patients in group 2 and six (7%) patients in group 3 (P Ͻ 0.05). The proteinuria reduction remained signif- group 4 were not responsive to uptitration and were main- icant after adjustment for differences in the SBP (P ϭ 0.033 and tained on the starting dosages of the study drugs. 0.034 for benazepril and losartan, respectively). There was no Uptitration was stopped in five patients before the optimal significant difference in proteinuria reduction between benaz- antiproteinuric dosage had been reached because of SBP Ͻ120 epril and losartan at both conventional and optimal antipro- mmHg (two in group 2 and three in group 4). Two patients teinuric dosages. titrated back because of reversible hyperkalemia (one in group Use of optimal antiproteinuric dosages versus conventional 2 and one in group 4). Five patients (three in group 2 and two dosages reduced the decline in renal function by 60% in the in group 4) withdrew during uptitration because of adverse benazepril arm (P ϭ 0.021) and 55% in the losartan arm (P ϭ events. 0.037), as assessed by creatinine clearance (Figure 3B). In addi- The antihypertensive efficacy was similar within arms and tion, the optimal antiproteinuric dosage was associated with between arms during the study (P Ͼ 0.05; Figure 3D). There slowing in the decline of eGFR in both benazepril (P ϭ 0.02) was no statistically significant difference in SBP between pa- and losartan arms (P ϭ 0.03; Figure 3C). There was a close tients with decline of eGFR Ͼ5orՅ5 ml/min per 1.73 m2 correlation between the extent of the reduction in proteinuria at (Figure 3E). Urinary excretion of urea and chloride was com- 3 mo and the rate of decline in the eGFR (r ϭϪ0.554, P Ͻ 0.001) parable among groups during the study (Figure 3F). Classes of and creatinine clearance at the study end (r ϭϪ0.487, P Ͻ conventional antihypertensive drugs that were used before and 0.001). Changes in renal function were similar between benaz- during the study are listed in Table 2. 1894 Journal of the American Society of Nephrology J Am Soc Nephrol 18: 1889–1898, 2007

Figure 3. Median of changes in urinary proteinuria excretion (A), median of creatinine clearance (B), estimated GFR (eGFR; C), BP in all patients (D), median of systolic BP in patients with decline of eGFR (⌬eGFR) Ͼ5orՅ5 ml/min per 1.73 m2 (E), and median of urinary excretion of urea and chloride (F).

Safety ing the study were comparable in groups. The proportion of No patient died during the study period. The incidence of patients who receiving recombinant human erythropoietin dry cough was significantly higher in the benazepril arm as and the mean dosage of recombinant human erythropoietin compared with the losartan arm, but it did not seem to be were similar among the groups during the follow-up (data dosage related (Table 3). There were no dosage-related differ- not shown). ences between benazepril and losartan with respect to the incidence of nonfatal cardiovascular events or other adverse events. Hyperkalemia occurred in eight (4.4%) patients in the Discussion benazepril arm and eight (4.4%) patients in the losartan arm. Of The results of the ROAD study indicate that treatment with these 16 patients, six were successfully treated with dietary optimal antiproteinuric dosages of the ACEi benazepril or the modifications, concomitant diuretic therapy, and optimized ARB losartan, as compared with their conventional antihyper- acid-base balance. The remaining 10 patients withdrew from tensive dosages, resulted in significant reduction in the risk for the study. The mean hemoglobin levels at baseline and dur- the primary end point (doubling of serum creatinine concen- J Am Soc Nephrol 18: 1889–1898, 2007 Benazepril and Losartan in Chronic Renal Insufficiency 1895

Table 2. Use of conventional antihypertensive medications

Benazepril Losartan Drug (n ͓%͔) Group 1 Group 2 Group 3 Group 4 (n ϭ 90) (n ϭ 90) (n ϭ 90) (n ϭ 90)

Calcium channel blocker at baseline 60 (67) 54 (60) 58 (64) 55 (61) during treatment 72 (80) 69 (77) 70 (78) 63 (70) Dihydropyridine at baseline 48 (53) 44 (49) 46 (51) 44 (49) during treatment 55 (61) 52 (58) 55 (61) 51 (57) Non-dihydropyridine at baseline 12(13) 10 (11) 12 (13) 11 (12) during treatment 17 (19) 17 (19) 15 (17) 12 (13) Diuretic at baseline 46 (51) 46 (51) 47 (52) 43 (48) during treatment 71 (79) 70 (78) 67 (74) 70 (78) ␤ blocker at baseline 43 (48) 44 (49) 42 (47) 40 (44) during treatment 48 (53) 47 (52) 48 (53) 43 (48) Centrally acting agent at baseline 28 (31) 27 (30) 27 (30) 26 (29) during treatment 34 (38) 28 (31) 32 (36) 27 (30)

Table 3. Adverse events after randomization

Benazepril Losartan Adverse Events Group 1 Group 2 Group 3 Group 4 (n ϭ 90) (n ϭ 90) (n ϭ 90) (n ϭ 90)

Adverse events 23 25 7 9 nonfatal cardiovascular event 4 5 5 4 myocardial infarction 2 2 2 2 heart failure 1 2 2 1 stroke 1 1 1 1 Other adverse events hyperkalemia 3 5 3 5 acute decline in renal function 2 3 3 3 dry cough 17 15 0 0 1 2 1 1

tration, ESRD, or death) in patients who did not have diabetes (1–7). In this study, SBP was rigorously controlled under a level and had overt proteinuria and chronic renal insufficiency. The of 130 mmHg but not 120 mmHg, because it has been evi- greater reduction in proteinuria and the slower decline in cre- denced that the risk for both renal and cardiovascular diseases atinine clearance and eGFR that were seen in the titration starts to increase at a SBP as low as 127 mmHg (31,32). BP and groups provide further evidence of the long-term renal protec- the use of conventional antihypertensive drugs were compara- tion that is afforded by this strategy. Although previous clinical ble at baseline and during treatment in the four groups. RAS studies have shown that titrating an ACEi or an ARB to higher blockade reduced the risk for renal progression even when dosages reduces proteinuria more effectively (27–30), to our adjusted for BP. Benazepril and losartan both are long-acting knowledge, this is the first study to demonstrate that titration RAS blockers. Although the study was not able to examine the of these therapies against urinary protein excretion provides 24-h BP, the previous studies that used the similar dosage of further benefit not only on proteinuria but also on a renal hard benazepril (20 mg/d) and losartan (approximately 50 to 100 end point. mg/d) achieved stable control of 24-h BP (33,34). These data BP is one of the most important predictors for renal outcome suggest that the additive renal protection that is conferred by 1896 Journal of the American Society of Nephrology J Am Soc Nephrol 18: 1889–1898, 2007 uptitration of benazepril or losartan might not be completely and losartan arms. Five percent of patients in the titration dependent on BP. This rationale is further supported by the groups experienced an increase in serum potassium levels of data from animal studies that indicated non–BP-dependent Ͼ6.0 mmol/L, but only six patients withdrew from the study. protective effects against renal fibrosis during high-dosage However, we cannot exclude the possibility that the lower rate therapy with ACEi or ARB (10,35). of serious hyperkalemia in patients with higher dosages of The extent of renal dysfunction at baseline may influence the benazepril or losartan may be related to the lower potassium outcome (1,2). However, the baseline renal function was com- intake (suggested by the lower protein intake, approximately parable among the four groups, and renal function was stable 0.5 g/kg per d) in this cohort and the wide use of diuretics as evidenced by the small variations of creatinine clearance (75%) during follow-up. before the study. Furthermore, the decrease in the risk for renal This study also provided direct evidence that the long-term outcome remained significant after adjustment for baseline renoprotective effects of benazepril and losartan were remark- eGFR. Therefore, we may not attribute the observed renal ben- ably similar at the optimal antiproteinuric dosages. These two efit to the difference in baseline renal dysfunction. classes of drugs, at comparable BP control, achieved a similar The median of proteinuria reduction that was observed at 3 reduction in both proteinuria and the risk for the primary end mo of titration treatment was approximately 1.0 g/d. The point in patients who did not have diabetes and had overt Efficacy in Nephropathy (REIN) study showed that proteinuria and renal insufficiency. Data from a trial that was for each 1.0-g/d reduction in proteinuria seen at 3 mo of ACEi designed primarily to compare the effects of combined ACEi therapy, subsequent GFR decline was slowed by 2.0 ml/min and ARB treatment with each monotherapy provided indirect per yr (36). Consistent with the other studies (22,37), protein- evidence supporting this notion (19). Given the finding that uria in this cohort was further reduced by approximately 12 to both drugs have a dosage-response relationship for renal pro- 17% by increase of benazepril or losartan from its conventional tection, it is fair to assume that, in clinical practice, full-dosage dosage to higher dosages. However, this less impressive differ- titration of a RAS blocker aimed at optimal reduction of pro- ence was associated with significant benefit in renal outcome, teinuria might be more important than the choice of the initial suggesting that proteinuria is a strong time-dependent predic- drug. Combination of ACEi and ARB have been shown to be tor for renal progression (2) and should be reduced as much as superior to either agent alone (19). However, the safety of this possible. Furthermore, RAS blockade has been shown to have approach remains to be further confirmed, especially in pa- benefit on renal progression, which is in addition to the anti- tients with advanced CKD. proteinuric effect (10,38). Because this was an open-label study, there is a possibility of Our study was designed to establish the optimal strategy for bias in the recording of the results. However, the Prospective reducing proteinuria with RAS blockers. Considering the in- Randomized Open Blinded End-Point (PROBE) design is a creasing recognition of individual differences in responsiveness well-accepted approach that is as rigorous as the double-blind and tolerability to antiproteinuric interventions (39,40), we in- design, at least when, as in this case, hard end points are dividually titrated benazepril or losartan to its optimal antipro- considered (41). Although the study was of single-center de- teinuric dosage. We found that the optimal antiproteinuric sign, the clinical characteristics of our cohort were similar to efficacy was obtained at the 20-mg dosage of benazepril or those reported in the registry of China. Therefore, our cohort 100-mg dosage of losartan in more than half of the patients who can reflect the general status of nondiabetic CKD, at least in received titration. Approximately 25% of patients needed even Chinese. higher dosages of benazepril or losartan for proteinuria control. It is important, however, to note that Ͻ8% of patients were Conclusion refractory to the antiproteinuric effect of benazepril or losartan. In patients who did not have diabetes and had overt protein- Uptitration of these drugs to the maximum licensed dosages uria and chronic renal insufficiency, use of optimal antiprotein- (benazepril 40 mg, losartan 200 mg) did not seem to overcome uric dosages of benazepril and losartan was associated with a such therapy resistance. Although it is obviously important to significant improvement in renal outcome as compared with determine the patient factors that may be related to the re- their conventional dosages in current clinical practice. The sponse variability, this study was not appropriate to analyze renoprotective effects might be comparable between the ACEi these factors, because only a few patients showed therapy benazepril and the ARB losartan in this population. resistance. Uptitration of benazepril or losartan was generally well tol- Acknowledgments erated in this study. Only 4% of patients stopped titration This study was supported by a National Nature and Sciences Grant because of intolerability before the optimal antiproteinuric dos- for Major Projects (30330300), a People’s Liberation Army Grant for ages were achieved. Long-term use of the optimal antiprotein- Major Clinical Research (2000), and National 11th Five-Years Plan uric dosages of benazepril or losartan seemed to be well toler- Foundation (to F.F.H.). ated in patients with chronic renal insufficiency. Dry cough was The abstract was presented at the annual meeting of the American the most common adverse event in the benazepril arm, but it Society of Nephrology; November 14 through 19, 2006; San Diego CA. did not seem to be dosage related. The overall incidence of other adverse events between groups that were given conven- Disclosures tional and titrated dosages was similar in both the benazepril None. J Am Soc Nephrol 18: 1889–1898, 2007 Benazepril and Losartan in Chronic Renal Insufficiency 1897

References hypertensive nondiabetic nephropathies. Am J Kidney Dis 1. Maschio G, Alberti D, Janin G, Locatelli F, Mann JF, Mo- 46: 1074–1079, 2005 tolese M, Ponticelli C, Ritz E, Zucchelli P; The Angiotensin- 14. Weinberg MS, Weinberg AJ, Zappe DH: Effectively target- Converting-Enzyme Inhibition in Progressive Renal Insuf- ting the renin-angiotensin-aldosterone system in cardio- ficiency Study Group: Effect of the angiotensin-converting- vascular and renal disease: Rationale for using angiotensin enzyme inhibitor benazepril on the progression of chronic II receptor blockers in combination with angiotensin-con- renal insufficiency. N Engl J Med 334: 939–945, 1996 verting enzyme inhibitors. J Renin Angiotensin Aldosterone 2. The GISEN Group (Gruppo Italiano di Studi Epidemiolo- Syst 1: 217–233, 2000 gici in Nefrologia): Randomised placebo-controlled trial of 15. Andersen S, Rossing P, Juhl TR, Deinum J, Parving HH: effect of ramipril on decline in glomerular filtration rate Optimal dose of losartan for renoprotection in diabetic and risk of terminal renal failure in proteinuric, non-dia- nephropathy. Nephrol Dial Transplant 17: 1413–1418, 2002 16. Haas M, Leko-Mohr Z, Erler C, Mayer G: Antiproteinuric betic nephropathy. Lancet 349: 1857–1863, 1997 versus antihypertensive effects of high-dose ACE inhibitor 3. Brenner BM, Cooper ME, de Zeeuw D, Keane WF, Mitch therapy. Am J Kidney Dis 40: 458–463, 2002 WE, Parving HH, Remuzzi G, Snapinn SM, Zhang Z, Sha- 17. Ruggenenti P, Mise N, Pisoni R, Arnoldi F, Pezzotta A, hinfar S; RENAAL Study Investigators: Effects of losartan Perna A, Cattaneo D, Remuzzi G: Diverse effects of in- on renal and cardiovascular outcomes in patients with type creasing doses on lipid abnormalities in chronic 2 diabetes and nephropathy. N Engl J Med 345: 861–869, nephropathies. Circulation 107: 586–592, 2003 2001 18. Viberti G, Wheeldon NM; MicroAlbuminuria Reduction 4. Mann JF, McClellan WM, Kunz R, Ritz E: Progression of With (MARVAL) Study Investigators: Mi- renal disease: Can we forget about inhibition of the renin- croalbuminuria reduction with valsartan in patients with angiotensin system? Nephrol Dial Transplant 21: 2348–2351, type 2 diabetes mellitus: A blood pressure-independent 2006 effect. Circulation 106: 672–678, 2002 5. Hou FF, Zhang X, Zhang GH, Xie D, Chen PY, Zhang WR, 19. Nakao N, Yoshimura A, Morita H, Takada M, Kayano T, Jiang JP, Liang M, Wang GB, Liu ZR, Geng RW: Efficacy Ideura T: Combination treatment of angiotensin-II receptor and safety of benazepril for advanced chronic renal insuf- blocker and angiotensin-converting-enzyme inhibitor in ficiency. N Engl J Med 354: 131–140, 2006 non-diabetic renal disease (COOPERATE): A randomised 6. Jafar TH, Schmid CH, Landa M, Giatras I, Toto R, Remuzzi controlled trial. Lancet 361: 117–124, 2003 [published erra- G, Maschio G, Brenner BM, Kamper A, Zucchelli P, Becker tum appears in Lancet 361: 1230, 2003] G, Himmelmann A, Bannister K, Landais P, Shahinfar S, de 20. Levey AS: Clinical evaluation of renal function. In: Primer Jong PE, de Zeeuw D, Lau J, Levey AS: Angiotensin- on Kidney Diseases, 2nd Ed., edited by Greenberg A, San converting enzyme inhibitors and progression of nondia- Diego, Academic Press, 1998, pp 20–26 betic renal disease. A meta-analysis of patient-level data. 21. Gradman AH, Arcuri KE, Goldberg AI, Ikeda LS, Nelson Ann Intern Med 135: 73–87, 2001 [published erratum ap- EB, Snavely DB, Sweet CS: A randomized, placebo-con- pears in Ann Intern Med 137: 299, 2002] trolled, double-blind, parallel study of various doses of 7. Jafar TH, Stark PC, Schmid CH, Landa M, Maschio G, de losartan potassium compared with maleate in Jong PE, de Zeeuw D, Shahinfar S, Toto R, Levey AS; patients with essential hypertension. Hypertension 25: AIPRD Study Group: Progression of chronic kidney dis- 1345–1350, 1995 ease: The role of blood pressure control, proteinuria, and 22. Bos H, Andersen S, Rossing P, De Zeeuw D, Parving HH, angiotensin-converting enzyme inhibition—A patient- De Jong PE, Navis G: Role of patient factors in therapy level meta-analysis. Ann Intern Med 139: 244–252, 2003 resistance to antiproteinuric intervention in nondiabetic 8. Wolf G, Ritz E: Combination therapy with ACE inhibitors and . Kidney Int 57[Suppl 75]: S32– and angiotensin II receptor blockers to halt progression of S37, 2000 chronic renal disease: Pathophysiology and indications. 23. Levey AS, Greene T, Kusek JW, Beck GJ: A simplified Kidney Int 67: 799–812, 2005 equation to predict glomerular filtration rate from serum 9. Ruggenenti P: Angiotensin-converting enzyme inhibition creatinine [Abstract]. J Am Soc Nephrol 11: 155A, 2000 and angiotensin II antagonism in nondiabetic chronic ne- 24. Adler SG. Quantitative and qualitative measurement of phropathies. Semin Nephrol 24: 158–167, 2004 urinary protein. In: Textbook of Nephrology, 4th Ed., edited 10. Peters H, Border WA, Noble NA: Targeting TGF-beta over- by Massry SG, Glassock RJ, Philadelphia, Lippincott Wil- expression in renal disease: Maximizing the antifibrotic liams & Wilkins, 2001, pp 1784–1786 action of angiotensin II blockade. Kidney Int 54: 1570–1580, 25. Zhang GH, Hou FF, Zhang X, Liu QF: Can angiotensin- 1998 converting enzyme inhibitor be used in chronic kidney 11. Fogo AB: Progression and potential regression of glomer- disease patients with serum creatinine level greater than ulosclerosis. Kidney Int 59: 804–819, 2001 266 micromol/L [in Chinese]? Zhonghua Nei Ke Za Zhi 44: 12. Laverman GD, Henning RH, de Jong PE, Navis G, de 592–596, 2005 Zeeuw D: Optimal antiproteinuric dose of losartan in non- 26. Dialysis and Transplantation Registration Group, Chinese diabetic patients with nephrotic range proteinuria. Am J Society of Nephrology: The report about the registration of Kidney Dis 38: 1381–1384, 2001 dialysis and transplantation in China 1999. Chin J Nephrol 13. Aranda P, Segura J, Ruilope LM, Aranda FJ, Frutos MA, 17: 77–78, 2001 Lopez V, Lopez de Novales E: Long-term renoprotective 27. Pisoni R, Ruggenenti P, Sangalli F, Lepre MS, Remuzzi A, effects of standard versus high doses of in Remuzzi G: Effect of high dose ramipril with or without 1898 Journal of the American Society of Nephrology J Am Soc Nephrol 18: 1889–1898, 2007

indomethacin on glomerular selectivity. Kidney Int 62: tion: Can it improve renoprotection? Am J Kidney Dis 40: 1010–1019, 2002 664–666, 2002 28. Schmieder RE, Klingbeil AU, Fleischmann EH, Veelken R, 36. Ruggenenti P, Perna A, Remuzzi G; GISEN Group Inves- Delles C: Additional antiproteinuric effect of ultrahigh tigators: Retarding progression of chronic renal disease: dose : A double-blind, randomized, prospec- The neglected issue of residual proteinuria. Kidney Int 63: tive study. J Am Soc Nephrol 16: 3038–3045, 2005 2254–2261, 2003 29. Palla R, Panichi V, Finato V, Parrini M, Andreini B, Bianchi 37. Parving HH, Lehnert H, Brochner-Mortensen J, Gomis R, AM, Giovannini L, Migliori M, Bertelli AA: Effect of in- Andersen S, Arner P; in Patients with Type 2 creasing doses of lisinopril on proteinuria of normotensive Diabetes and Microalbuminuria Study Group: The effect of patients with IgA nephropathy and normal renal function. irbesartan on the development of diabetic nephropathy in Int J Clin Pharmacol Res 14: 35–43, 1994 patients with type 2 diabetes. N Engl J Med 345: 870–878, 30. Laverman GD, Navis G, Henning RH, de Jong PE, de 2001 Zeeuw D: Dual renin-angiotensin system blockade at op- 38. Miyata T, van Ypersele de Strihou C: Renoprotection of timal doses for proteinuria. Kidney Int 62: 1020–1025, 2002 angiotensin receptor blockers: Beyond blood pressure low- 31. Klag MJ, Whelton PK, Randall BL, Neaton JD, Brancati FL, ering. Nephrol Dial Transplant 21:846–849, 2006 Ford CE, Shulman NB, Stamler J: Blood pressure and end- 39. Dickerson JE, Hingorani AD, Ashby MJ, Palmer CR, Brown stage renal disease in men. N Engl J Med 334: 13–18, 1996 MJ: Optimisation of antihypertensive treatment by cross- 32. Sheinfeld GR, Bakris GL: Benefits of combination angio- tensin-converting enzyme inhibitor and calcium antagonist over rotation of four major classes. Lancet 353: 2008–2013, therapy for diabetic patients. Am J Hypertens 12: 80S–85S, 1999 1999 40. Lufft V, Kliem V, Hamkens A, Bleck JS, Eisenberger U, 33. Andersen S, Tarnow L, Rossing P, Hansen BV, Parving Petersen R, Ehlerding G, Maschek H, Pichlmayr R, Brunk- HH: Renoprotective effects of angiotensin II receptor horst R: Antiproteinuric efficacy of after renal blockade in type 1 diabetic patients with diabetic nephrop- transplantation is determined by the extent of vascular and athy. Kidney Int 57: 601–606, 2000 tubulointerstitial damage. Clin Transplant 12: 409–415, 34. Jacobsen P, Andersen S, Jensen BR, Parving HH: Additive 1998 effect of ACE inhibition and angiotensin II receptor block- 41. Hansson L, Hedner T, Dahlof B: Prospective randomized ade in type I diabetic patients with diabetic nephropathy. open blinded end-point (PROBE) study. A novel design for J Am Soc Nephrol 14: 992–999, 2003 intervention trials. Prospective Randomized Open Blinded 35. Navis G, Kramer AB, de Jong PE: High-dose ACE inhibi- End-Point. Blood Press 1: 113–119, 1992