DEPARTMENT OF HEALTH AND HUMAN SERVICES

                PUBLIC HEALTH SERVICE

                FOOD AND DRUG ADMINISTRATION

                CENTER FOR DRUG EVALUATION AND RESEARCH

 

 

                   STATISTICAL REVIEW AND EVALUATION

 

NDA #:                20-386

SERIAL #:          SE1-S028

DRUG NAME:   COZAAR (Losartan Potassium) tablets

INDICATION:   Renal Protection

SPONSOR:         Merck & Co., Inc.

DOCUMENT REVIEWED:

1.         Vols. 1 and 2 (CDER REC’D Date: November 13, 2001)

2.         SAS data base in EDR

STATISTICAL REVIEWER:    H.M. James Hung, Ph.D. (HFD-710)

MEDICAL REVIEWER:           Juan Carlos Pelayo, M.D. (HFD-110)

PROJECT MANAGER:             Sandra Birdsong (HFD-110)

 

STATISTICAL KEY WORDS: Interim analysis, DSMB, sample size change, log rank test, O’Brien-Fleming boundary, Cox regression, geographical region

 

Distribution:   NDA 20-386, SE1-S028

                HFD-110/Dr. Throckmorton

                HFD-110/Dr. Stockbridge

                HFD-110/Dr. Pelayo

                HFD-110/Ms. Birdsong

                HFD-700/Dr. Anello

                HFD-710/Dr. Chi

                HFD-710/Dr. Mahjoob

                HFD-710/Dr. Hung

                HFD-710/chron

 

JHung/301-594-5436/DB1/cozaar.doc/3-1-2002


TABLE OF CONTENTS

0.  SUMMARY.................................................................................................................................................................................. 3

1.  INTRODUCTION....................................................................................................................................................................... 3

2. OVERVIEW OF RENAAL STUDY RESULTS...................................................................................................................... 3

Sample Size Planning........................................................................................................................................................ 4

Interim Analysis Plan..................................................................................................................................................... 4

Randomization.................................................................................................................................................................... 5

Analysis methods.............................................................................................................................................................. 5

Protocol Amendments.................................................................................................................................................... 6

Changes in the Conduct of the Study or Planned Analyses.................................................................. 6

Efficacy results.................................................................................................................................................................. 7

Baseline.................................................................................................................................................................................... 7

Primary endpoint (doubling serum creatinine, ESRD or death).................................................................................. 7

Table 1.  Incidence of adjudicated primary events (Reviewer’s Analysis)............................................................................ 8

Time between components of primary endpoint............................................................................................................... 8

Table 2. Time from Doubling of Serum Creatinine to ESRD and From ESRD to death (Sponsor’s analysis, not confirmed by Reviewer)           8

Losartan effect on primary endpoint over time................................................................................................................. 8

Subgroup results.................................................................................................................................................................... 9

Table 3. Subgroup results on primary endpoint (Reviewer’s analysis)................................................................................ 9

Results of primary endpoint by geographical region................................................................................................... 10

Table 4. Primary endpoint by geographical region (Reviewer’s analysis)........................................................................... 10

Table 5. Primary endpoint by country (Reviewer’s analysis)............................................................................................ 10

Secondary endpoints........................................................................................................................................................... 12

Table 6.  Incidence of secondary endpoints (Reviewer’s analysis)..................................................................................... 12

Progression of Renal Disease............................................................................................................................................ 12

Table 7. Summary of slopes of reciprocal of serum creatinine at quartiles (Sponsor’s results confirmed by Reviewer’s analysis)        12

Table 8. Comparison of mean slopes for renal progression................................................................................................. 13

Change in Proteinuria over Time..................................................................................................................................... 13

Predictibility of Proteinuria for Treatment Effect on Primary Endpoint.................................................................. 13

Table 9. Adjustment of the Primary endpoint for proteinuria at Month 6 and as a time-varying covariate (Sponsor’s results confirmed by reviewer’s analysis)............................................................................................................................................................. 13

4. CONCLUSIONS....................................................................................................................................................................... 14

5. APPENDIX................................................................................................................................................................................. 15

Table A.1  Distribution of time (days) to non-event censoring (Reviewer’s analysis)....................................................... 15

Table A.2. Primary endpoint by subgroups in Asia region excluding Israel (Reviewer’s analysis).................................... 16

Table A.3. Primary endpoint by subgroups in non-Asia region including Israel (Reviewer’s analysis)............................. 17

Table A.4  Baseline characteristics for Asia versus Non-Asia regions (Reviewer’s analysis)............................................ 18

Table A.5  Distribution of proteinuria level (mg/g) [Reviewer’s analysis].......................................................................... 19

Figure 1. log(-log(survival)) versus log(days) for the primary endpont, doubling serum creatlinie, ESRD, or death.......... 20


0.  SUMMARY

 

    This submission contains only one study.  The conclusions in Section 4 constitute the summary.

 

1.  INTRODUCTION

 

            This statistical review pertains to the results of the RENAAL trial which is the major component of this submission.  Most of the sponsor’s main results have been confirmed by this reviewer’s analyses.  Only the results of the reviewer’s analyses will be presented in this review unless stated otherwise.

 

 

2. OVERVIEW OF RENAAL STUDY RESULTS

 

          The RENAAL study is a multi-center, randomized, parallel group, double-blind trial of losartan versus placebo in patients with type 2 diabetes and nephropathy on a background of conventional antihypertensive therapy (ACE inhibitor or AIIA therapy excluded). For inclusion and exclusion criteria, the readers are referred to the medical reviewer’s review.

 

Following a 6-week of screening period, eligible patients were stratified by baseline level of proteinuria (urine albumin to urine creatinine artio from a first morning void above or below 2000 mg/gCr), and randomized 1:1 to either losartan 50 mg or placebo.  The losartan should be increased to 100 mg daily if at the first month of the study or at any point following trough blood pressure did not reach the goal of 140/90 mmHg.  After titration occurred, the study design provided the investigators the flexibility to individualize treatment by adding, increasing, or changing patient’s background medication in order to achieve trough goal blood pressure of 140/90 mmHg.  Additionally, at the investigator’s discretion, patients could begin the study on the lower dose of 25 mg losartan or placebo or be reduced to that dose if necessary during the study.

 

The primary efficacy endpoint was a composite endpoint of doubling serum creatinine, end-stage renal disease (ESRD), or death for any reason.  To be considered an endpoint, doubling of serum creatinine must have been a rise in serum creatinine that was at least twice the baseline serum creatinine value and confirmed by a second serum creatinine measurement, analyzed by the central laboratory, and drawn no earlier than 4 weeks following the initial doubling.  ESRD was defined as the need for dialysis or transplantation.   The prespecified secondary efficacy endpoint was a composite of cardiovascular morbidity and mortality endpoints, which included cardiovascular death, hospitalization for heart failure, hospitalization for unstable angina, myocardial infarction, stroke, and revascularization.  Other secondary measures included progression of renal disease assessed by the slope of the reciprocal of serum creatinine concentration and changes in proteinuria.

 


Sample Size Planning  

 

Sample size estimation was based on the assumption that the 5-year doubling of serum creatinine/ESRD/death rate in the placebo group would be 58% and that this rate would be reduced to 46.4% (20% reduction) in the losartan group.  A total of 1320 to 1400 patients are needed to be able to detect this difference at 4.8% significance level (because of interim analysis described below) and with power 95%, assuming a minimum follow-up period of 3.5 years, 2-year enrollment period, proportional hazards for the treatments, and censoring at discontinuation time for a discontinuation rate of 13% per year.  The protocol originally planned for 1520 patients to be enrolled over a 1-year period but was amended to 1320 patients when it became clear that enrollment would take 2 years.  The study actually over-enrolled with 1513 patients due to a larger than anticipated number of screened patients successfully qualifying during the final months of the 2-year recruitment period.

 

Interim Analysis Plan  

 

The original plan allowed for a single efficacy interim analysis and a series of safety-only analyses. The preplanned interim analysis was to be performed when half of the expected events (284 of 568 events) associated with the sample size and event rate estimates had been observed or the last patient entered had been followed for 2 years, whichever came first. This interim analysis would use the O’Brien-Fleming stopping boundary as a guideline for any recommendation of early termination due to overwhelming efficacy (as measured by the composite for the hard endpoints of ESRD and death only). According to this boundary, the critical p-value at the interim analysis had to be 0.0035 for both the primary endpoint and the composite of the 2 hard endpoints, resulting in the final evaluation a of 0.049. However, at the routine DSMB meetings at which the safety data were unblinded, the committee also considered stopping the study if losartan was superior to placebo with respect to the primary endpoint and also with respect to the composite of the 2 “hard” endpoints, ESRD and death at a two-sided a–level of 0.0001. The DSMB planned 10 unblinded looks at death and ESRD during the course of the study. Each of these looks was provided at a a of 0.0001 to allow the DSMB to stop the study, if losartan was extremely superior to placebo. The total a adjustment for these evaluations was 0.001. This was subtracted from the original a of 0.049, to give the final evaluation a of 0.048, required for the primary hypothesis. For other outcomes, a p-value of <0.05 was considered to indicate statistical significance. All statistical tests were two-sided.

 

The Steering Committee could, at any point during the study, consider whether external circumstances (e.g., results from other diabetic studies, IRB issues with placebo control, etc.) had occurred that would make it difficult to keep patients in the study.  Any decision to end RENAAL for external reasons would be made independently of the unblinded interim data; therefore, for the end-of-study analyses, the only statistical adjustment would be due to the interim analysis by the DSMB.

 


Randomization

 

According to the sponsor’s response to this reviewer’ request, the date of randomization was not specifically collected for RENAAL. The sponsor used the date of the first dose of study drug as the date of the start of the study for the purpose of time to event analysis. The decision to use this date for endpoint analyses was made prior to unblinding the database. The next closest date to date of randomization is date of the “randomization” visit but there is no guarantee that randomization actually took place at that visit.. Among the 1513 randomized patients, the dates differ in only 12 patients, and in 11 of the 12 they differ by only 1 day and in the other case they differ by 12 days.

 

 

Analysis methods    

 

The statistical analysis for efficacy and safety will be based on intent-to-treat principle.  The primary analysis of the composite endpoint will be based on the time to the first event (i.e., doubling serum creatinine, ERD or death). The components of this endpoint will be analyzed in a similar fashion. For all time to event variables, the relative risk for each event, its confidence interval and the test for treatment difference will be based on the Cox regression model. According to the original protocol, the model will contain treatment group indicator, an indicator variable for the stratification factor of proteinuria at baseline and a factor for country. The treatment by country interaction will also be assessed using Cox regression (a=0.10). Life table event rates and mortality curves will be based on the product-limit estimates. In the protocol amendments, country was changed to region (North America [United States + Canada + Puerto Rico], Latin America, Europe [Eastern + Western], and Asia).

 

The renal function measured as the reciprocal of serum creatinine (1/Cr) across over time during the trial will be analyzed. The rate of decline in renal function will be assessed by the slope of the observed 1/Cr over the quarterly follow-up period. As the primary analysis, a linear random-effects model will be used to compare slopes in renal function. Region, strata of proteinuria at baseline and baseline serum creatinine values will be included in the model as fixed covariates.  Supportive analyses include an ANCOVA analysis to compare the treatment effect with the dependent variable of an individual chronic slope. If normality assumption is not appropriete, the same ANCOVA model will be performed with the dependent variable of the rank of an individual chronic slope.

 

Proteinuria which is the ratio of urine albumin to urine creatinine will be analyzed. The primary analysis will be based on the intention-to-treat approach. All the observed proteinuria values during the quarterly follow-up period from all patients will be included in the analysis according to the treatment group to which they are randomized, regardless of whether patients discontinue double-blind therapy or have the events of doubling of serum creatinine before the study is completed (3.5 years from last patient randomized). Note that the follow-up time for the secondary endpoint is defined as the period from randomization to the last measurement of proteinuria before ESRD during the study. The mixed-effects (longitudinal) model will be used with terms including treatment (losartan/placebo), time, region and baseline proteinuria levels. A natural logarithm transformation will be applied for the proteinuria data for the mixed model. The mean change profiles over time will be plotted to show the time course of the treatment effect on proteinuria. Interaction between treatment and time for the mean changes of the proteinuria levels will be explored.

 

Protocol Amendments

 

There appeared to be six protocol amendments. Only the following amendment may potentially have a substantive effect on statistical evaluation.

 

Amendment 147-03

 

Since patient enrollment had been slower than anticipated in the protocol, recruitment of 1520 patients would not be achieved in the projected timeframe. An enrollment period of approximately 2 years was estimated to allow recruitment of at least 1320 patients required to achieve 95% power. In addition, the interim analysis stopping rules were updated. For the purpose of the DSMB review only, an interim analysis (a=0.0035) would be performed by the Merck statistician assigned to the DSMB when half of the expected events associated with this sample size and event rate estimates have been observed or the last patient entered has been followed for 1.75 years, whichever came first.

 

 

Changes in the Conduct of the Study or Planned Analyses

 

The study was to be completed in March of 2002, 3.5 years from last patient in. However, the Steering Committee, whose obligation was to keep abreast of current research in the field and continually re-evaluate the ethical context of the trial, voted unanimously to end the study early, while blinded, for reasons unrelated to the study data. The reason for this decision was documented in the minutes of the Steering Committee meeting on 10-Feb-2001 and is described in the following paragraph taken from a letter that was sent to all investigators:

 

  “At its meeting on 10-Feb-2001, the RENAAL Steering Committee took this action due to increasing evidence that ACE inhibitors are effective in reducing cardiovascular events in patients with characteristics similar to RENAAL patients. This decision to discontinue was in part due to soon-to-be-published information showing that cardiovascular events are reduced by ACE inhibitors in diabetic patients with renal impairment. The action of the Steering Committee was taken on the basis of external evidence only and was therefore independent of any knowledge of the results of the trial. The Steering Committee was and remained blinded until the results of the trial were analyzed and presented. The Committee further recommended that physicians caring for patients in the RENAAL trial make this information available to their patients and strongly consider addition of therapy aimed at blockade of the renin-angiotensin-aldosterone system (RAAS). In the usual care arm of the RENAAL Study, patients were receiving antihypertensive therapy, excluding agents that block the RAAS.”

 

The “soon-to-be-published information” referred to data from the HOPE diabetic subpopulation, which demonstrated that in over 900 patients with or without renal insufficiency, there appeared to be significantly less cardiovascular events in patients receiving an ACE inhibitor. The endpoint cutoff date for this study was 10-Feb-2001, i.e., any endpoint occurring on or before February 10 was adjudicated. Endpoints occurring after this date were not collected with the exception of any possible silent myocardial infarctions detected by the core lab on the final visit ECGs.

 

 

Efficacy results

 

A total of 1513 patients at 250 centers from 28 countries were randomized. The total number of the primary endpoint events is 686, larger than the planned number of 568, despite the early ending of the trial by about 1-2 months.

 

Baseline

 

Two treatment groups were comparable with respect to baseline demographic characteristics, medical history, electrocardiographic changes, and drug therapy, as shown in Table 7 of the sponsor’s study report.

 

Primary endpoint (doubling serum creatinine, ESRD or death)

 

The sponsor’s analysis on the primary efficacy endpoint, doubling of serum creatinine, ESRD or death, was based on the Cox regression model using baseline proteinuria as a stratification factor and geographical region (Asia pacific, Europe, Latin America, North America) as the adjustment factor. As noted above, according to the sponsor, the date of randomization was not specifically collected for RENAAL. The sponsor used the date of the first dose of study drug as the date of the start of the study for the purpose of time to event analysis. The next closest date to date of randomization is date of the “randomization” visit but there is no guarantee that randomization actually took place at that visit. Among the 1513 randomized patients, the dates differ in only 12 patients, and in 11 of the 12 they differ by only 1 day and in the other case they differ by 12 days. Using the date of the “randomization” visit for analysis, the results changed little.

 

The censoring distribution with respect to the primary endpoint was comparable between the two treatment groups (Table A.1 in the Appendix). As shown in Table 1, the rate of the composite endpoint of doubling serum creatinine, ESRD or death was significantly lower in the losartan group than in the placebo group (hazard ratio of 0.84 with 95.2% CI of 0.72 to 0.97, p = 0.022). This treatment difference seemed to be largely attributed to the difference in doubling of serum creatinine. The hazard ratio of ESRD was 0.93.  The hazard ratio of all cause death was 0.98. ESRD or death constituted approximately 50% of the primary composite events as the first endpoint. However, on a whole, the losartan group had a significantly fewer ESRD events (hazard ratio of 0.71 with 95.2% CI of 0.57 to 0.89, p = 0.002). The mortality rate differed little between the two treatment groups.


Table 1.  Incidence of adjudicated primary events (Reviewer’s Analysis)

 

Losartan

(N=751)

Placebo

(N=762)

Hazard ratio

(95.2% CI)

p-value$

Primary event

Doubling serum creatinine, ESRD or death

  327 (43.5%)

Median time

= 1303 days

  359 (47.1%)

Median time

=1373 days

0.84 (0.72, 0.97)

0.022

Decomposition of the 1st primary event

   Doubling of SC

  162 (21.6%)

  198 (26.0%)

0.75 (0.61, 0.92)

0.006

   ESRD

    64 (  8.5%)

    65 (  8.5%)

0.93 (0.65, 1.31)

0.66

   Death

  101 (13.5%)

    96 (12.6%)

0.98 (0.74, 1.30)

0.91

 

1st component endpoint

   Doubling of SC

  162 (21.6%)

  198 (26.2%)

0.75 (0.61, 0.92)

0.006

   ESRD

  147 (19.5%)

  194 (25.5%)

0.71 (0.57, 0.89)

0.002

   Death

  158 (21.0%)

  155 (20.3%)

1.02 (0.81, 1.27)

0.88

   ESRD or death

  255 (34.0%)

  300 (39.4%)

0.80 (0.68, 0.95)

0.009

$ nominal p-value,  pre-specified primary analysis (Cox model using geographical region as covariate and baseline proteinuria as a stratification variable)

 

 

 

Time between components of primary endpoint

 

From Table 2, the proportion of  patients with ESRD who subsequently died seemed to be higher in the losartan group than in the placebo group. However, none of the comparisons in this table is a randomized comparison.

 

Table 2. Time from Doubling of Serum Creatinine to ESRD and From ESRD to death (Sponsor’s analysis, not confirmed by Reviewer)

 

                Losartan

                   Placebo

events/sample size (%)

Mean

follow-up

(days)

events/sample size (%)

Mean follow-up

(days)

DSC to ESRD

 83/162 (51.2%)

 279.5

129/198 (65.2%)

252.5

ESRD to death

 50/147 (34.0%)

 363.4

  49/194 (25.3%)

400.3

 

 

Losartan effect on primary endpoint over time

 

    The effect of losartan relative to placebo in terms of the hazard of the primary endpoint did not appear to be constant over the duration of the trial (see Figure 1, in the Appendix). The hazard curves appeared to be converging.

 

 

Subgroup results

 

     There was no noticeable inconsistency in the results over the subgroups by demographic variables or baseline factors (Table 3).

 

Table 3. Subgroup results on primary endpoint (Reviewer’s analysis)

 

Losartan

         (N=751)  

Placebo

(N=762)

Hazard ratio

(95% CI)

Female

Male

138 / 289 ( 47.8 %)

189 / 462 ( 40.9 %)

145 / 268 ( 54.1 %)

214 / 494 ( 43.3 %)

0.80 ( 0.64 , 1.01 )

0.90 ( 0.74 , 1.09 )

Age

   < 65 yrs

   ³ 65 yrs

 

222 / 503 ( 44.1 %)

105 / 248 ( 42.3 %)

 

246 / 502 ( 49.0 %)

113 / 260 ( 43.5 %)

 

0.83 ( 0.69 , 1.00 )

0.95 ( 0.73 , 1.23 )

Asian

Hispanic

black

white

  49 / 117 ( 41.9 %)

  77 / 140 ( 55.0 %)

  50 / 125 ( 40.0 %)

145 / 358 ( 40.5 %)

  74 / 135 ( 54.8 %)

  74 / 137 ( 54.0 %)

  41 / 105 ( 39.0 %)

163 / 377 ( 43.2 %)

0.69 ( 0.48 , 0.99 )

1.01 ( 0.74 , 1.39 )

0.97 ( 0.64 , 1.47 )

0.87 ( 0.69 , 1.09 )

Proteinuria

   < 2000 mg/g

   ³ 2000 mg/g

 

150 / 501 ( 29.9 %)

177 / 250 ( 70.8 %)

 

161 / 511 ( 31.5 %)

198 / 251 ( 78.9 %)

 

0.91 ( 0.73,   1.14)

0.78 ( 0.64,   0.96)

BMI

   < 30 kg/m2

   ³ 30 kg/m2

 

195 / 437 ( 44.6 %)

132 / 314 ( 42.0 %)

 

226 / 471 ( 48.0 %)

133 / 291 ( 45.7 %)

 

0.87 ( 0.72 , 1.06 )

0.88 ( 0.69 , 1.11 )

Duration of hypertension

   < 10 yrs

   ³ 10 yrs

 

178 / 387 ( 46.0 %)

149 / 364 ( 40.9 %)

 

204 / 409 ( 49.9 %)

155 / 353 ( 43.9 %)

 

0.88 ( 0.72 , 1.08 )

0.86 ( 0.69 , 1.08 )

Total Cholesterol

  < 240 mg/dL

  ³ 240 mg/dL

 

187 / 496 ( 37.7 %)

140 / 255 ( 54.9 %)

 

205 / 489 ( 41.9 %)

154 / 273 ( 56.4 %)

 

0.84 ( 0.69 , 1.03 )

0.93 ( 0.74 , 1.17 )

Serum Creatinine

  < 2 mg/dL

  ³ 2 mg/dL

 

174 / 482 ( 36.1 %)

153 / 269 ( 56.9 %)

 

173 / 483 ( 35.8 %)

186 / 279 ( 66.7 %)

 

0.97 ( 0.78 , 1.20 )

0.77 ( 0.62 , 0.95 )

Serum Albumin

  < 3.6 mg/dL

  ³ 3.6 mg/dL

 

143 / 207 ( 69.1 %)

184 / 544 ( 33.8 %)

 

145 / 202 ( 71.8 %)

214 / 560 ( 38.2 %)

 

0.87 ( 0.69 , 1.10 )

0.84 ( 0.69 , 1.02 )

Serum Uric Acid

  < 7 mg/dL

  ³ 7 mg/dL

 

197 / 459 ( 42.9 %)

130 / 292 ( 44.5 %)

 

203 / 448 ( 45.3 %)

156 / 314 ( 49.7 %)

 

0.90 ( 0.74 , 1.09 )

0.83 ( 0.66 , 1.05 )

HbA1c

  < 10%

  ³ 10%

 

    5 /   9   ( 55.6 %)

322 / 742 ( 43.4 %)

 

    2 /   8   ( 25.0 %)

357 / 754 ( 47.3 %)

 

2.73 ( 0.53 , 14.1 )

0.86 ( 0.74 , 1.00 )

Hemoglobin

  < 12 mg/dL

  ³ 12 mg/dL

 

163 / 315 ( 51.7 %)

164 / 436 ( 37.6 %)

 

178 / 310 ( 57.4 %)

181 / 452 ( 40.0 %)

 

0.81 ( 0.66 , 1.01 )

0.90 ( 0.73 , 1.11 )

Nonsmoker

Smoker

  69 / 147 ( 46.9 %)

258 / 604 ( 42.7 %)

  61 / 130 ( 46.9 %)

298 / 632 ( 47.2 %)

0.98 ( 0.70 , 1.39 )

0.84 ( 0.71 , 0.99 )

Sitting SBP

  < 140 mmHg

  ³ 140 mmHg

 

  60 / 191 ( 31.4 %)

267 / 560 ( 47.7 %)

 

  66 / 187 ( 35.3 %)

293 / 575 ( 51.0 %)

 

0.86 ( 0.61 , 1.22 )

0.87 ( 0.74 , 1.03 )

Insulin Use

  No

  Yes

 

113 / 290 ( 39.0 %)

214 / 461 ( 46.4 %)

 

128 / 313 ( 40.9 %)

231 / 449 ( 51.4 %)

 

0.92 ( 0.71 , 1.18 )

0.83 ( 0.69 , 1.00 )

Dihydropyridine

  No

  Yes

 

128 / 345 ( 37.1 %)

199 / 406 ( 49.0 %)

 

148 / 351 ( 42.2 %)

211 / 411 ( 51.3 %)

 

0.84 ( 0.66 , 1.06 )

0.89 ( 0.74 , 1.08 )

ACEI or AIIA Use

  No

  Yes

 

146 / 351 ( 41.6 %)

181 / 400 ( 45.3 %)

 

180 / 386 ( 46.6 %)

179 / 376 ( 47.6 %)

 

0.85 ( 0.68 , 1.06 )

0.88 ( 0.72 , 1.08 )

Beta Blocker Use

  No

  Yes

 

265 / 614 ( 43.2 %)

  62 / 137 ( 45.3 %)

 

298 / 622 ( 47.9 %)

  61 / 140 ( 43.6 %)

 

0.84 ( 0.72 , 1.00 )

0.99 ( 0.69 , 1.41 )

Calcium Blocker Use

  No

  Yes

 

  82 / 219 ( 37.4 %)

245 / 532 ( 46.1 %)

 

  89 / 216 ( 41.2 %)

270 / 546 ( 49.5 %)

 

0.84 ( 0.62 , 1.14 )

0.88 ( 0.74 , 1.05 )

 

 

Results of primary endpoint by geographical region

 

As suggested in the sponsor’s Figure 9, the hazard ratio for the primary endpoint appeared to be quite different between Asia and other regions. The Asia region appeared to show a big effect with losartan (45% reduction in hazard) but other regions showed little or no effect; see Table 4.

 

Table 4. Primary endpoint by geographical region (Reviewer’s analysis)

 

Losartan       

Placebo

Hazard ratio (95% CI)

All regions

327/751 (43.5%) 

359/762 (47.1%)

0.84 (0.72, 0.98)

Asia

  49/125 (39.2%)

  78/132 (59.1%)

0.55 (0.39, 0.79)

Europe

  58/151 (38.4%)

  51/144 (35.4%)

1.05 (0.72, 1.53)

Latin America

  78/137 (56.9%)

  80/137 (58.4%)

0.93 (0.68, 1.27)

North America

142/338 (42.0%)

150/349 (43.0%)

0.94 (0.75, 1.19)

 

Table 5 summarizes by-country results of the primary endpoint. In Asia, four countries showed a big effect with losartan (hazard reduction ranging from 17% to 78%); Singapore having very small and the smallest sample size in the region showed a reversed trend. In Europe, Spain had the largest sample size and gave a hazard reduction of 19%; United Kingdom having the next largest sample size showed no effect or a bit reversed trend. Other countries had very small sample size and showed mixed trends. In Latin America, Mexico and Brazil having the largest sample size showed different trends; the former showed a bit opposite trend but the latter gave a hazard reduction of 22%. Other countries also showed mixed trends. In North America, United States showed little effect with losartan; other countries contributed few events. Overall, only Asia seemed to show a consistent favorable effect with losartan. However, Israel contributed a small number of patients and gave the biggest effect with losartan. By putting Israel in Europe, the hazard ratio changed to 0.64 with 95% CI of 0.44 to 0.93 for Asia and to 0.87 with 95% CI of 0.61 to 1.24 for Europe. A substantial numerical difference in hazard ratio between Asia and non-Asia remained.

 

Table 5. Primary endpoint by country (Reviewer’s analysis)

 

 

Losartan       

Placebo

Hazard ratio

(95% CI)

All regions

 

327/751 (43.5%) 

359/762 (47.1%)

0.84 (0.72, 0.98)

Asia

 

 

 

 

Hong Kong

  19/46 (41.3%)

  27/46 (58.7%)

0.57 (0.32, 1.04)

Israel

   4/19 (21.1%)

  12/18 (66.7%)

0.22 (0.07, 0.70)

Japan

 22/44 (50.0%)

  34/52 (65.4%)

0.73 (0.43, 1.26)

Malaysia

   2/11 (18.2%)

   4/10 (40.0%)

0.42 (0.08, 2.31)

Singapore

   2/5 (40.0%)

   1/6 (16.7%)

2.12 (0.19, 23.51)

Europe

 

 

 

 

 

 

 

 

 

 

 

 

 

Austria

   5/8 (62.5%)

   4/7 (57.1%)

1.17 (0.31, 4.38)

Czech Republic

   9/17 (52.9%)

   6/16 (37.5%)

1.63( 0.58, 4.59)

Denmark

   3/8 (37.5%)

   3/8 (37.5%)

0.76 (0.15, 3.83)

France

   5/5 (100%)

   7/7 (100%)

1

Germany

   3/6 (50.0%)

   2/6 (33.3%)

1.15 (0.19, 6.97)

Hungary

   3/5 (60.0%)

   2/5 (40.0%)

1.43 (0.24, 8.61)

Italy

   2/13 (15.4%)

   3/13 (23.1%)

0.55 (0.09, 3.33)

Netherlands

   4/4 (100%)

   3/3 (100%)

1

New Zealand

   1/1 (100%)

   1/2 (50.0%)

2

Portugal

   3/5 (60.0%)

   2/5 (40.0%)

1.23 (0.20, 7.40)

Russian Federation

    6/14 (42.9%)

   4/12 (33.3%)

1.34 (0.38, 4.77)

Slovakia

    1/1 (100%)

   1/1 (100%)

1

Spain

  14/36 (38.9%)

 14/31 (45.2%)

0.81 (0.38, 1.69)

United Kingdom

  10/28 (35.7%)

 10/28 (35.7%)

1.02 (0.42, 2.45)

Latin America

 

 

 

 

 

 

Argentina

   2/9 (22.2%)

   5/8 (62.5%)

0.32 (0.06, 1.65)

Brazil

 17/28 (60.7%)

 20/30 (66.7%)

0.78 (0.41, 1.49)

Chile

   7/13 (53.8%)

  8/13 (61.5%)

0.69 (0.25, 1.91)

Costa Rica

  12/17 (70.6%)

   8/16 (50.0%)

1.36 (0.55, 3.34)

Mexico

  19/33 (57.6%)

  18/34 (52.9%)

1.08 (0.57, 2.07)

Peru

  10/21 (47.6%)

  13/21 (61.9%)

0.69 (0.30, 1.58)

Venezuela

  11/16 (68.8%)

    8/15 (53.3%)

2.24 (0.82, 6.08)

North America

 

 

Canada

         .

    1/1 (100%)

          .

Puerto Rico

    2/2 (100%)

    1/3 (33.3%)

3

United States

142/336 (42.3%)

148/345 (42.9%)

0.95 (0.76, 1.20)

 

 

The differences in hazard ratio of the primary endpoint between Asia and non-Asia regions seemed to appear in most of the subgroups by baseline characteristics (Tables A.2 and A.3 in the Appendix). The incidence of the primary composite endpoint in the placebo group appeared to be higher in Asia than in non-Asia region whereas the incidence in the losartan group appeared to be in the same ballpark between the regions. Table A.4 in the Appendix exhibits the potential differences in baseline characteristics between Asia and Non-Asia regions.  The two regions appeared to be comparable with respect to most of the baseline characteristics, except possibly on BMI, Dihydropyridine use, and proteinuria. Asia region seemed to have a larger proportion of the patients with proteinuria ³ 2000 mg/g. In this study, proteinuria seemed to be a strong predictor of the primary composite endpoint, as reported by the sponsor and in Table 9 of this review. Thus, the apparent difference in hazard ratio between Asia and non-Asia regions might be attributed, at least partly, to a higher baseline proteinuria level (Table A.5 in the Appendix), or to a higher incidence rate of the primary endpoint in the placebo group in Asia region.

 

 

Secondary endpoints

 

Losartan seemed to be associated with a larger reduction of hospitalization for HF. There was no clear suggestion for  beneficial effect on any of other secondary endpoints with losartan (Table 6).

 

Table 6.  Incidence of secondary endpoints (Reviewer’s analysis)

 

Losartan

(N=751)

Placebo

(N=762)

Hazard ratio (95% CI)

p-value$

Cardiovascular mortality/morbidity

  247 (32.9%)

  268 (35.2%)

0.90 (0.76, 1.08)

 0.25

Hospitalized for HF

    89 (11.9%)

  126 (16.5%)

0.68 (0.52, 0.90)

 0.006

MI

    50 (  6.7%)

    68 (  8.9%)

0.72 (0.50, 1.04)

 0.079

Stroke

    47 (  6.3%)

    50 (  6.6%)

0.85 (0.64, 1.41)

 0.78

Cardiovascular death

    90 (12.0%)

    79 (10.4%)

1.12 (0.83, 1.52)

 0.45

Revascularization

    69 (  9.2%)

    60 (  7.9%)

1.19 (0.84, 1.68)

 0.34

Hospitalized for Unstable angina

    42 (  5.6%)

    41 (  5.4%)

1.03 (0.67, 1.59)

 0.89

$ using the same Cox regression model as for the primary endpoint

 

 

Progression of Renal Disease

 

The rate of decline in renal function was smaller in the losartan group than in the placebo group

 

Table 7. Summary of slopes of reciprocal of serum creatinine at quartiles (Sponsor’s results confirmed by Reviewer’s analysis)

 

Losartan

(N=751)

Placebo

(N=762)

Est. renal

loss reduction

p-value$

Chronic slope (dL/mg/yr)

(Month 3 and after)

Quartiles:

   25th

   50th

   75th

 

N = 693

 

-0.10

-0.052

-0.020

 

N=678

 

-0.13

-0.070

-0.028

 

 

 

 

25.5%

 

 

 

 

< 0.0001

Overall slope (dL/mg/yr)

(all phases)

Quartiles:

   25th

   50th

   75th

 

N = 745

 

-0.11

-0.057

-0.025

 

N=754

 

-0.14

-0.070

-0.025

 

 

 

 

18.5%

 

 

 

 

   0.011

Negative slope indicates a loss of renal function

Est. loss reduction is estimated using a linear random effects model adjusted for region, startum, and baseline serum creatinine

$ based on two-sample median score nonparametric test

 

 

Table 8. Comparison of mean slopes for renal progression

 

Analysis of slope (dL/mg/yr)

Losartan

(N=751)

Placebo

(N=762)

Est. renal

loss reduction

p-value$

Chronic phase

 -0.060

 -0.070

    13.9%

 0.0033

All phases

 -0.067

 -0.077

    12.7%

 0.0091

Negative slope indicates a loss of renal function

Est. loss reduction is estimated using a linear random effects model adjusted for region, startum, and baseline serum creatinine

$ based on two-sample median score nonparametric test

 

 

Change in Proteinuria over Time

 

Table 14, Figures 6 and 7 of the sponsor’s results (not yet confirmed by reviewer’s analysis) appeared to indicate that the Losartan group had a much greater percent decrease in proteinuria in terms of geometric mean and median than the placebo group.

 

Predictibility of Proteinuria for Treatment Effect on Primary Endpoint

 

When proteinuria (on the log scale) was adjusted based on the Month 6 or on the values over the entire study, prior to the primary composite endpoint, the treatment effect of losartan diminished. This seems to suggest that proteinuria is a strong predictor for the primary composite endpoint in this study.

 

Table 9. Adjustment of the Primary endpoint for proteinuria at Month 6 and as a time-varying covariate (Sponsor’s results confirmed by reviewer’s analysis)

 

 

Losartan

(N=751)

Placebo

(N=762)

Est. risk

reduction

p-value$

Primary endpoint

  Adj. for Month-6 Proteinuria

  Adj. for overall proteinuria changes

 

   43.3%

   43.5%

 

   46.1%

   47.1%

 

  -2.9%

    1.7%

 

  0.73

  0.83

ESRD

  Adj. for overall proteinuria changes

 

   19.6%

 

   25.5%

 

  14.1%

 

  0.17

ESRD or death

  Adj. for overall proteinuria changes

 

   34.0%

 

   39.4%

 

     9.5%

 

  0.25

Est. risk reduction using a proportional hazards regression model with adjustment for region and stratum

 


4. CONCLUSIONS

 

         There was some evidence that losartan might reduce the incidence of the primary composite endpoint, doubling of serum creatinine, ESRD or death, (reduction of risk = 16% with 95% CI of 3% to 28%, p = 0.022). The strength of evidence did not meet the usual standard of statistical evidence, at least lower by an order of magnitude in p-value. The effect of losartan in terms of the hazard of the primary endpoint did not appear to be constant over the duration of the trial (Figure 1, the hazard curves appeared to be converging). Approximately a half of the primary events were ESRD or death. The treatment difference seemed to be largely attributed to the difference in doubling of serum creatinine. However, in the patients with doubling serum creatinine, 51% developed ESRD in the losartan group and 65% in the placebo group; mean time from doubling serum creatinine to ESRD was about 30 days longer in the losartan group. In addition, 19.5% of the losartan patients and 25.5% of the placebo patients developed ESRD; thus, losartan seemed to be associated with a 29% reduction in risk of having an ESRD (p = 0.002). There is little difference in death rate between the losartan group and the placebo group.

 

Proteinuria seemed to be a strong predictor of the primary composite event in this study. When proteinuria was adjusted based on the Month 6 value or on the values over time, prior to the primary composite endpoint, the effect of losartan diminished (Table 9).

 

The data seemed to suggest that there might be a difference in the effect of losartan in terms of hazard reduction of the primary endpoint between Asia and other regions. The Asia region appeared to show a big effect with losartan whereas other regions as a whole showed little effect (Table 4). Futher exploration seemed to suggest that this apparent difference, if real, might be attributed to a higher incidence rate of the primary endpoint in the placebo group in Asia, or to a higher baseline proteinuria level in Asia (Table A.5).

 

The rate of decline in renal function appeared to be smaller in the losartan group than in the placebo group. Losartan seemed to be associated with a reduction of hospitalizations for heart failure. No beneficial effect was found with losartan on cardiovascular mortality or morbidity, myocardial infarction, stroke, revascularization, or hospitalization for unstable angina. This seemed to make it difficult to interpret the apparent benefit with losartan in reduction of hospitalization for heart failure.
5. APPENDIX

 

Table A.1  Distribution of time (days) to non-event censoring (Reviewer’s analysis)

 

Event=doubling serum creatinine, ESRD, or death

 

    Losartan

   (N=751)

 

     Placebo

     (N=762)

 

# of censored cases

  327 (43.5%)

  359 (47.1%)

 

Max

 

         1615

 

         1613

 

99th %tile

 

         1580

 

         1579

 

95th

 

         1522

 

         1504

 

90th

 

         1475

 

         1445

 

75th

 

         1376

 

         1376

 

50th

 

         1199

 

         1196

 

Mean

 

         1212

 

         1201

 

25th

 

         1040

 

         1026

 

10th

 

         949

 

         940

 

5th

 

         914

 

         906

 

1st

 

         878

 

         885

 

Min

 

         865

 

         873

 


Table A.2. Primary endpoint by subgroups in Asia region excluding Israel (Reviewer’s analysis)

 

      Losartan

      (N=106)

        Placebo

        (N=114)

 HR

Female

12 /  29 ( 41.4 %)

28 /  39 ( 71.8 %)

0.38

Male

33 /  77 ( 42.9 %)

38 /  75 ( 50.7 %)

0.82

Age < 65 yrs

32 /  69 ( 46.4 %)

45 /  79 ( 57.0 %)

0.70

Age >= 65 yrs

13 /  37 ( 35.1 %)

21 /  35 ( 60.0 %)

0.51

Asian

45 / 106 ( 42.5 %)

66 / 114 ( 57.9 %)

0.64

BMI < 30 kg/m2

41 /  98 ( 41.8 %)

64 / 109 ( 58.7 %)

0.62

BMI >= 30 kg/m2

4 /   8 ( 50.0 %)

2 /   5 ( 40.0 %)

1.04

Duration of hypertension < 10 yrs

31 /  69 ( 44.9 %)

45 /  78 ( 57.7 %)

0.70

Duration of hypertension >= 10 yrs

14 /  37 ( 37.8 %)

21 /  36 ( 58.3 %)

0.52

Total Cholesterol < 240 mg/dL

27 /  70 ( 38.6 %)

38 /  75 ( 50.7 %)

0.71

Total Cholesterol >= 240 mg/dL

18 /  36 ( 50.0 %)

28 /  39 ( 71.8 %)

0.55

Serum Creatinine < 2 mg/dL

19 /  60 ( 31.7 %)

32 /  63 ( 50.8 %)

0.52

Serum Creatinine >= 2 mg/dL

26 /  46 ( 56.5 %)

34 /  51 ( 66.7 %)

0.79

Serum Albumin < 3.6 mg/dL

20 /  29 ( 69.0 %)

29 /  34 ( 85.3 %)

0.63

Serum Albumin >= 3.6 mg/dL

25 /  77 ( 32.5 %)

37 /  80 ( 46.3 %)

0.62

Serum Uric Acid < 7 mg/dL

28 /  73 ( 38.4 %)

36 /  60 ( 60.0 %)

0.55

Serum Uric Acid >= 7 mg/dL

17 /  33 ( 51.5 %)

30 /  54 ( 55.6 %)

0.81

HbA1c < 10%

1 /   1 (  100 %)

1 /   1 (  100 %)

 

HbA1c >= 10%

44 / 105 ( 41.9 %)

65 / 113 ( 57.5 %)

0.63

Hemoglobin < 12 mg/dL

28 /  50 ( 56.0 %)

43 /  60 ( 71.7 %)

0.61

Hemoglobin >= 12 mg/dL

17 /  56 ( 30.4 %)

23 /  54 ( 42.6 %)

0.69

Nonsmoker

12 /  29 ( 41.4 %)

14 /  23 ( 60.9 %)

0.59

Smoker

33 /  77 ( 42.9 %)

52 /  91 ( 57.1 %)

0.64

Sitting SBP < 140 mmHg

5 /  23 ( 21.7 %)

12 /  34 ( 35.3 %)

0.61

Sitting SBP >=140 mmHg

40 /  83 ( 48.2 %)

54 /  80 ( 67.5 %)

0.57

Insulin Use No

19 /  48 ( 39.6 %)

29 /  56 ( 51.8 %)

0.71

Insulin Use Yes

26 /  58 ( 44.8 %)

37 /  58 ( 63.8 %)

0.56

Dihydropyridine No

7 /  24 ( 29.2 %)

11 /  23 ( 47.8 %)

0.61

Dihydropyridine Yes

38 /  82 ( 46.3 %)

55 /  91 ( 60.4 %)

0.65

ACEI or AIIA Use No

13 /  42 ( 31.0 %)

25 /  53 ( 47.2 %)

0.59

ACEI or AIIA Use Yes

32 /  64 ( 50.0 %)

41 /  61 ( 67.2 %)

0.61

Beta Blocker Use No

39 /  88 ( 44.3 %)

60 /  99 ( 60.6 %)

0.64

Beta Blocker Use Yes

6 /  18 ( 33.3 %)

6 /  15 ( 40.0 %)

0.71

Calcium Blocker Use No

6 /  22 ( 27.3 %)

11 /  23 ( 47.8 %)

0.56

Calcium Blocker Use Yes

39 /  84 ( 46.4 %)

55 /  91 ( 60.4 %)

0.65

Proteinuria < 2000 mg/g

15 /  62 ( 24.2 %)

22 /  64 ( 34.4 %)

0.66

Proteinuria >=2000 mg/g

30 /  44 ( 68.2 %)

44 /  50 ( 88.0 %)

0.58

 


 

Table A.3. Primary endpoint by subgroups in non-Asia region including Israel (Reviewer’s analysis)

 

         Losartan

         (N=645)

          Placebo

          (N=648)

  HR

Female

126 / 260 ( 48.5 %)

117 / 229 ( 51.1 %)

0.90

Male

156 / 385 ( 40.5 %)

176 / 419 ( 42.0 %)

0.91

Age < 65 yrs

190 / 434 ( 43.8 %)

201 / 423 ( 47.5 %)

0.86

Age >= 65 yrs

92 / 211 ( 43.6 %)

92 / 225 ( 40.9 %)

1.04

Asian

4 /  11 ( 36.4 %)

8 /  21 ( 38.1 %)

1.21

Hispanic

77 / 140 ( 55.0 %)

74 / 137 ( 54.0 %)

1.01

black

50 / 125 ( 40.0 %)

41 / 105 ( 39.0 %)

0.97

white

145 / 358 ( 40.5 %)

163 / 377 ( 43.2 %)

0.87

BMI < 30 kg/m2

154 / 339 ( 45.4 %)

162 / 362 ( 44.8 %)

0.97

BMI >= 30 kg/m2

128 / 306 ( 41.8 %)

131 / 286 ( 45.8 %)

0.87

Duration of hypertension < 10 yrs

147 / 318 ( 46.2 %)

159 / 331 ( 48.0 %)

0.94

Duration of hypertension >= 10 yrs

135 / 327 ( 41.3 %)

134 / 317 ( 42.3 %)

0.91

Total Cholesterol < 240 mg/dL

160 / 426 ( 37.6 %)

167 / 414 ( 40.3 %)

0.87

Total Cholesterol >= 240 mg/dL

122 / 219 ( 55.7 %)

126 / 234 ( 53.8 %)

1.02

Serum Creatinine < 2 mg/dL

155 / 422 ( 36.7 %)

141 / 420 ( 33.6 %)

1.07

Serum Creatinine >= 2 mg/dL

127 / 223 ( 57.0 %)

152 / 228 ( 66.7 %)

0.76

Serum Albumin < 3.6 mg/dL

123 / 178 ( 69.1 %)

116 / 168 ( 69.0 %)

0.93

Serum Albumin >= 3.6 mg/dL

159 / 467 ( 34.0 %)

177 / 480 ( 36.9 %)

0.88

Serum Uric Acid < 7 mg/dL

169 / 386 ( 43.8 %)

167 / 388 ( 43.0 %)

0.98

Serum Uric Acid >= 7 mg/dL

113 / 259 ( 43.6 %)

126 / 260 ( 48.5 %)

0.85

HbA1c < 10%

4 /   8 ( 50.0 %)

1 /   7 ( 14.3 %)

4.15

HbA1c >= 10%

278 / 637 ( 43.6 %)

292 / 641 ( 45.6 %)

0.91

Hemoglobin < 12 mg/dL

135 / 265 ( 50.9 %)

135 / 250 ( 54.0 %)

0.88

Hemoglobin >= 12 mg/dL

147 / 380 ( 38.7 %)

158 / 398 ( 39.7 %)

0.93

Nonsmoker

57 / 118 ( 48.3 %)

47 / 107 ( 43.9 %)

1.10

Smoker

225 / 527 ( 42.7 %)

246 / 541 ( 45.5 %)

0.88

Sitting SBP < 140 mmHg

55 / 168 ( 32.7 %)

54 / 153 ( 35.3 %)

0.90

Sitting SBP >=140 mmHg

227 / 477 ( 47.6 %)

239 / 495 ( 48.3 %)

0.93

Insulin Use No

94 / 242 ( 38.8 %)

99 / 257 ( 38.5 %)

0.98

Insulin Use Yes

188 / 403 ( 46.7 %)

194 / 391 ( 49.6 %)

0.88

Dihydropyridine No

121 / 321 ( 37.7 %)

137 / 328 ( 41.8 %)

0.86

Dihydropyridine Yes

161 / 324 ( 49.7 %)

156 / 320 ( 48.8 %)

0.98

ACEI or AIIA Use No

133 / 309 ( 43.0 %)

155 / 333 ( 46.5 %)

0.89

ACEI or AIIA Use Yes

149 / 336 ( 44.3 %)

138 / 315 ( 43.8 %)

0.96

Beta Blocker Use No

226 / 526 ( 43.0 %)

238 / 523 ( 45.5 %)

0.9

Beta Blocker Use Yes

56 / 119 ( 47.1 %)

55 / 125 ( 44.0 %)

1.03

Calcium Blocker Use No

76 / 197 ( 38.6 %)

78 / 193 ( 40.4 %)

0.88

Calcium Blocker Use Yes

206 / 448 ( 46.0 %)

215 / 455 ( 47.3 %)

0.94

Proteinuria < 2000 mg/g

135 / 439 ( 30.8 %)

139 / 447 ( 31.1 %)

0.95

Proteinuria >=2000 mg/g

147 / 206 ( 71.4 %)

154 / 201 ( 76.6 %)

0.84

 


Table A.4  Baseline characteristics for Asia versus Non-Asia regions (Reviewer’s analysis)

 

         Asia*

       (N=220)

         Non-Asia*

         (N=1293)

Female

Male

  68   (    30.9   %)

152   (    69.1   %)

489   (    37.8   %)

804   (    62.2   %)

Age < 65 yrs

Age >= 65 yrs

148   (    67.3   %)

  72   (    32.7   %)

857   (    66.3   %)

436   (    33.7   %)

BMI < 30 kg/m2

BMI >= 30 kg/m2

207   (    94.1   %)

  13   (      5.9   %)

701   (    54.2   %)

592   (    45.8   %)

Duration of hypertension < 10 yrs

Duration of hypertension >= 10 yrs

147   (    66.8   %)

  73   (    33.2   %)

649   (    50.2   %)

644   (    49.8   %)

Total Cholesterol < 240 mg/dL

Total Cholesterol >= 240 mg/dL

145   (    65.9   %)

  75   (    34.1   %)

840   (    65.0   %)

453   (    35.0   %)

Serum Creatinine < 2 mg/dL

Serum Creatinine >= 2 mg/dL

123   (    55.9   %)

  97   (    44.1   %)

842   (    65.1   %)

451   (    34.9   %)

Serum Albumin < 3.6 mg/dL

Serum Albumin >= 3.6 mg/dL

  63   (    28.6   %)

157   (    71.4   %)

346   (    26.8   %)

947   (    73.2   %)

Serum Uric Acid < 7 mg/dL

Serum Uric Acid >= 7 mg/dL

133   (    60.5   %)

  87   (    39.5   %)

774   (    59.9   %)

519   (    40.1   %)

HbA1c < 10%

HbA1c >= 10%

   2    (      0.9   %)

218   (    99.1   %)

  15   (     1.2   %)

1278 (    98.8   %)

Hemoglobin < 12 mg/dL

Hemoglobin >= 12 mg/dL

110   (    50.0   %)

110   (    50.0   %)

515   (    39.8   %)

778   (    60.2   %)

Nonsmoker

Smoker

  52   (    23.6   %)

168   (    76.4   %)

225   (    17.4   %)

1068 (    82.6   %)

Sitting SBP < 140 mmHg

Sitting SBP >=140 mmHg

  57   (    25.9   %)

163   (    74.1   %)

321   (    24.8   %)

972   (    75.2   %)

Insulin Use No

Insulin Use Yes

104   (    47.3   %)

116   (    52.7   %)

499   (    38.6   %)

794   (    61.4   %)

Dihydropyridine No

Dihydropyridine Yes

  47   (    21.4   %)

173   (    78.6   %)

649   (    50.2   %)

644   (    49.8   %)

ACEI or AIIA Use No

ACEI or AIIA Use Yes

  95   (    43.2   %)

125   (    56.8   %)

642   (    49.7   %)

651   (    50.3   %)

Beta Blocker Use No

Beta Blocker Use Yes

187   (    85.0   %)

  33   (    15.0   %)

1049 (    81.1   %)

244   (    18.9   %)

Calcium Blocker Use No

Calcium Blocker Use Yes

  45   (    20.5   %)

175   (    79.5   %)

390   (    30.2   %)

903   (    69.8   %)

Proteinuria < 2000 mg/g

Proteinuria >=2000 mg/g

126   (    57.3   %)

  94   (    42.7   %)

886   (    68.5   %)

407   (    31.5   %)

* Israel is put in Non-Asia region


Table A.5  Distribution of proteinuria level (mg/g) [Reviewer’s analysis]

 

 

      Asia

       Other$

 

Max

 

       10150

 

       12208

 

99th %tile

 

         7670

 

         7537

 

95th

 

         5248

 

         5102

 

90th

 

         4602

 

         4045

 

75th

 

         2976

 

         2433

 

50th

 

         1672

 

         1161

 

Mean

 

         2151

 

         1749

 

25th

 

           767

 

           538

 

10th

 

           422

 

           294

 

5th

 

           319

 

           203

 

1st

 

           122

 

             75

 

Min

 

             45

 

             31

 $ North America, Latin America, Europe


Figure 1. log(-log(survival)) versus log(days) for the primary endpont, doubling serum creatlinie, ESRD, or death