Antiproteinuric Therapy and Fabry Nephropathy

This report describes an open-label, nonrandomized, prospective evaluation of the effects of angiotensin-converting enzyme inhibitor and angiotensin receptor blocker therapy on patients who have Fabry disease and also received enzyme replacement therapy with agalsidase-β, given at 1 mg/kg body wt every 2 wk. Previous placebo-controlled phase III and phase IV trials with agalsidase-β demonstrated clearing of globotriaosylceramide from vascular endothelia but little effect on proteinuria or progressive loss of kidney function in patients with Fabry disease and severe chronic kidney disease marked by overt proteinuria and/or estimated GFR <60 ml/min per 1.73 m2. Angiotensin-converting enzyme inhibitor and/or angiotensin receptor blocker therapy is the standard of care for patients with proteinuric kidney diseases, but their use is challenging in patients with Fabry disease and low or low-normal baseline systemic BP. A group of patients with Fabry disease were treated with antiproteinuric therapy, in conjunction with agalsidase-β; sustained reductions in proteinuria with stabilization of kidney function were achieved in a group of six patients who had severe Fabry nephropathy; the progression rate was −0.23 ± 1.12 ml/min per 1.73 m2 per yr with 30 mo of follow-up. Fabry disease is an X-linked disorder caused by lysosomal α-galactosidase A deficiency, with resulting accumulation of glycosphingolipids, and progressive kidney, cardiac, and neurologic involvement that can cause death in the fifth decade.1,2 Women can also have serious disease manifestations.3 Enzyme replacement therapy (ERT) with agalsidase-β (recombinant human α-galactosidase A) clears globotriaosylceramide from vascular endothelial cells of patients with Fabry disease.4,5 We use the term “Fabry nephropathy” for this spectrum of kidney involvement: Focal, segmental, and global sclerosis; mesangial widening; epithelial deposits; ischemic changes; tubulointerstitial fibrosis; and vascular changes including GL-3 deposits and hyalinosis.6 Studies have demonstrated clinical benefit of ERT in patients with mild Fabry disease7,8 but relatively little impact on patients with initial estimated GFR (eGFR) <60 ml/min per 1.73 m2 and baseline proteinuria >1 g/d. ERT had no evident impact on proteinuria in an open-label study9; placebo-controlled, double blind phase III4,5 and phase IV studies8; and an open-label extension study of the original phase III cohort.10 ERT may have dosage-related effects on eGFR progression11,12 but, per se, does not reduce urinary protein excretion.13 The recommended BP target in chronic kidney disease (CKD) is <130/80 mmHg, with the use of angiotensin-converting enzyme inhibitor (ACEI) or angiotensin receptor blocker (ARB) therapy as the treatment of choice for overt proteinuria, which is an important factor in the progression of CKD. Systolic BP <130 mmHg is recommended if the urine protein/creatinine ratio is >0.5 g/g in CKD14 or the urine albumin/creatinine ratio is >500 mg/g in diabetic kidney disease.15 These clinical practice recommendations are challenging in Fabry nephropathy, because many patients have relatively normal or low BP in the early stages of their disease.1,16 The objective of our study was to describe the effects of ACEI/ARB therapy on urinary protein excretion in patients who had Fabry disease and were treated with ERT. The hypothesis was that reduction of proteinuria with ACEI/ARB therapy in conjunction with ERT would be associated with slowing of the progressive decline of eGFR, especially in high-risk patients who had proteinuria and initial eGFR rates <60 ml/min per 1.73 m2. Our patients tolerated ACEI/ARB therapy and had sustained reductions in BP and proteinuria and eGFR stabilization compared with previous experiences with similar patients who were treated with ERT alone.7–10,12 RESULTS Patient Characteristics The median age for the eight mens and three women was 37.1 yr (range 18.3 to 56.7) at the time ERT was begun. Nine families with seven distinct mutations were represented, and there were two relative pairs: Mother (patient 10) and son (patient 8), and brother (patient 6) and brother (patient 7; Table 1). The clinical manifestations were diverse, with multiple organ system involvement in addition to Fabry nephropathy. Some patients had hyperlipidemia, coronary artery disease, and hypertension, in addition their Fabry disease (Table 1). Timing of ACEI/ARB and Agalsidase-β Therapy Most patients were begun on ACEI/ARB therapy before starting ERT. The subsequent duration of ERT is shown in Table 1 (mean 30.3 ± 9.8 mo; median 30.1 mo; range 16 to 43; excludes patient 11). Some patients received both ACEI and ARB; the dosages were adjusted empirically, depending on the baseline proteinuria, the response to antiproteinuric therapy, and tolerance of the antihypertensive effects of ACEI/ARB therapy. Other agents that could affect kidney function and proteinuria are also listed in Table 1. Two of the patients participated in the previous phase IV study of agalsidase-β8; patient 9 was randomly assigned to the active treatment arm throughout the phase IV study, whereas patient 10 received placebo during the blinded phase and went on to open-label treatment with agalsidase-β after having sustained ventricular tachycardia. Three patients were previously treated with agalsidase-α (Replagal; Shire, Cambridge, MA) at 0.2 mg/kg every 2 wk at the National Institutes of Health and switched to agalsidase-β at the University of Alabama at Birmingham (UAB). These patients received agalsidase-α for 60 mo (patient 7), 16 mo (patient 6), and 24 mo (patient 8) before switching to agalsidase-β therapy. Two patients (9 and 12) started ACEI/ARB therapy at outside clinics; their referring physicians provided their baseline information when these patients were transferred to UAB. Clinical Events during the Observation Period The clinical manifestations are listed in Table 1. Eight patients had kidney biopsies, and all had Fabry nephropathy.6 Transient ischemic attacks and strokes did not occur during the observation period despite fairly aggressive lowering of systolic BP with ACEI/ARB therapy. Two patients received pacemakers before the start of ERT, and one patient (8) developed atrial fibrillation 5 mo after the completion of the observation period. Two female and 1 male patient developed ventricular ectopy after starting ERT and now have intracardiac defibrillator devices. Four patients developed hyperkalemia with serum K+ >6.0, but these episodes were minor and managed with reductions in dietary K+ intake. One male patient (11) with severe Fabry nephropathy (initial eGFR 25.3 ml/min per 1.73 m2 and urine protein/creatinine ratio 3.54 g/g) developed acute pancreatitis as a result of pancreas divisum, acute cholecystitis with cholecystectomy, and subacute bacterial endocarditis and required acute dialysis soon after the start of ERT. His kidney function slightly improved, but his urine protein excretion remained elevated and chronic dialysis was needed within 6 mo of starting ERT. His data are presented in the figures but are not included in the summary statistics for evaluation of the effects of ACEI/ARB therapy on urine protein/creatinine ratio or the eGFR progression rate. Two other male patients also had laparoscopic cholecystectomy during the observation period, but these events were deemed to be coincidental and not related to ERT. Systemic BP, Urine Protein/Creatinine Ratios, and eGFR The four patients with stages 1 and 2 CKD were younger (mean 33.8 ± 11.8 yr) than the seven patients with stages 3 and 4 CKD (mean 42.2 ± 11.4 yr; Table 1). As shown in Table 2, the baseline BP were higher for the patients with stages 3 and 4 CKD than for the patients with stages 1 and 2 CKD (135/82 and 98/63 mmHg, respectively). ERT was started 7.3 ± 6.4 mo (median 4.9; range 0.3 to 20.6) after initiation of ACEI/ARB therapy. At that point, the BP were decreased by ACEI/ARB therapy, especially for patients with stages 3 and 4 CKD, and these effects were sustained throughout the period of follow-up (Table 2). The average BP measurements relative to the start of ERT are shown in Figure 1. Despite more aggressive ACEI/ARB therapy, the BP in the patients with stages 3 and 4 remained higher than the BP in patients with stages 1 and 2 CKD. The urinary protein/creatinine ratios are shown in Table 2; three of the patients with stages 1 and 2 CKD had overt proteinuria (≥0.3 g/g) at baseline. The median urinary protein/creatinine ratio was 0.36 (range 0.17 to 0.46) for the patients with stages 1 and 2 CKD and 1.24 (range 0.23 to 6.3) for patients with stages 3 and 4 CKD, excluding patient 11. ACEI/ARB treatment reduced the median urinary protein/creatinine ratio for patients with stages 3 and 4 CKD to 0.21 (range 0.1 to 1.6) and for patients with stages 1 and 2 CKD to 0.22 (range 0.07 to 0.37; Table 2). The individual urinary protein/creatinine ratios at baseline, at initiation of ERT, and throughout the observation period are shown in Figure 2. Higher baseline urine protein/creatinine ratios and sustained reductions were more evident in patients with stages 3 and 4 CKD (Figure 2B) than in patients with stages 1 and 2 CKD (Figure 2A). Two of these patients (5 and 8 in Figure 2A) had initial decreases in their urine protein/creatinine ratios that were not sustained because they chose to stop or reduce their ACEI/ARB therapy during the observation period. The regressions of urine protein/creatinine ratios over time are presented in Table 3. The second column represents the baseline ratio for each patient obtained at the time that ACEI/ARB therapy was started. The third column represents the intercept value, corresponding to the point at which ERT was started in each patient. The duration of follow-up on ERT and the number of complete evaluations for each patient are also presented in Table 3. The baseline urine protein/creatinine ratio was 0.34 ± 0.13 (median 0.36) for patients with stages 1 and 2 CKD, and this ratio was significantly different from the regression intercept (0.22 ± 0.12; median 0.22). The baseline urine protein/creatinine ratio was 1.91 ± 2.29 (median 1.24) for patients with stages 3 and 4 CKD, and this ratio was also significantly different from the regression intercept (0.30 ± 0.38; median 0.15), which reflects the antiproteinuric effects of ACEI/ARB therapy. The regression slopes for both groups were not significantly different from zero, indicating that the reductions in urine protein/creatinine ratio were maintained throughout the observation period. Urine protein and urine albumin measurements are shown in Table 4. Albumin excretion accounted for approximately half of the total urinary protein excretion and was reduced in parallel with total protein by ACEI/ARB therapy. The individual regressions of eGFR for each patient are presented in Figure 3, and the results are presented in Table 5. The second column presents the baseline eGFR obtained when ACEI/ARB therapy was started, and the third column presents the intercept, corresponding to the point at which ERT was started in each patient. The initial eGFR was 108 ± 10.8 ml/min per 1.73 m2 for patients with stages 1 and 2 CKD and was reduced to 96.9 ± 17.2 ml/min per 1.73 m2 by ACEI/ARB therapy (Table 2). The initial eGFR was 52.0 ± 16.5 ml/min per 1.73 m2 for patients with stages 3 and 4 CKD and was significantly reduced to 39.4 ± 10.9 ml/min per 1.73 m2 by ACEI/ARB therapy (Table 2). As was seen in other forms of proteinuric kidney disease,17 eGFR was initially reduced by ACEI/ARB therapy as BP was lowered. Excluding the initial change in eGFR before ERT was begun, the average progression rate for patients with stages 1 and 2 CKD was 1.18 ± 2.78 ml/min per 1.73 m2 per yr and −0.23 ± 1.12 ml/min per 1.73 m2 per yr for patients with stages 3 and 4 CKD. The individual regressions are presented in Table 5, and the averaged values for both groups are presented in Figure 4. The slopes for both groups were not significantly different from zero, indicating that eGFR was stabilized in Fabry nephropathy that was treated with ACEI/ARB therapy and ERT during the 30-mo observation period. DISCUSSION Our results demonstrate the feasibility of using ACEI/ARB therapy in patients with Fabry nephropathy, most of whom are not overtly hypertensive. The patients tolerated ACEI/ARB therapy and had sustained reductions of urinary protein excretion (Table 1, Figure 2). The progression rate for patients with stages 3 and 4 CKD (0.22 ± 1.47 ml/min per 1.73 m2 per yr) is better than previously described in patients who had moderately severe Fabry nephropathy and who were treated ERT but did not receive antiproteinuric therapy.7–10,12 Agalsidase-β given at 1 mg/kg had no evident effect on urinary protein excretion in patients with Fabry disease in an open-label study,9 in placebo-controlled double blind phase III4,5 and phase IV trials,8 and in an open-label extension of the phase III trial.10 Agalsidase-α, at the usual dosage of 0.2 mg/kg, did not reduce proteinuria.7,12 Although weekly infusions may have slowed progression compared with infusions every 2 wk, agalsidase-α still did not affect proteinuria.12 Relatively low systemic BP was described early as a feature of Fabry disease.16 Nevertheless, the use of antiproteinuric therapy has been recommended for treatment of Fabry disease with kidney involvement.18,19 ACEI/ARB therapy reduced systemic BP in our patients with Fabry nephropathy, but our approach is consistent with the Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines for proteinuric kidney disease,14,15 in which the goal of reducing urine protein excretion to <500 mg/d is recommended even if systolic BP falls below 130 mmHg. BP reductions in patients with stages 3 and 4 CKD did limit the total dosing of ACEI/ARB therapy but was generally well tolerated. No serious adverse events were observed during this study other than occasional hypotensive symptoms that quickly responded to reductions in the dosage of the ACEI/ARB therapy and mild episodes of hyperkalemia. Whereas the reduction in BP in patients with stages 3 and 4 CKD was evident, the systolic pressures were not reduced below that observed in patients with stages 1 and 2 CKD before ACEI/ARB therapy was started (Table 2, Figure 1). Previous reports documented the progressive course of Fabry nephropathy. Branton et al.1 reported that 14 male patients developed “chronic renal insufficiency” with a progression rate of −12.2 ± 9.1 ml/min per yr. Breunig et al.9 described 26 patients who were treated with agalsidase-β. Eight patients with stage 2 or 3 CKD (baseline GFR 71 ± 17 ml/min per 1.73 m2) showed progressive deterioration of their GFR to 60 ± 23 ml/min per 1.73 m2 during a 26.4-mo follow-up period, with an apparent progression rate of −4.7 ± 5.4 ml/min per 1.73 m2 per yr. Fourteen patients received ACEI/ARB therapy, but there were no effects on BP or proteinuria with the dosages that were used. Banikazemi et al.8 reported a prospective, placebo-controlled trial of agalsidase-β in patients with Fabry disease and mild to moderate kidney disease. The baseline eGFR in the treatment group was 53 ± 18 ml/min per 1.73 m2, and BP was 126 ± 16 mmHg (systolic) and 77 ± 10 mmHg (diastolic). The baseline urine protein/creatinine ratios were 1.5 ± 1.0, similar to what we observed for patients with Fabry disease and stages 3 and 4 CKD, before the institution of ACEI/ARB therapy (Table 2). Some of the patients did received ACEI/ARB therapy, but there was no systematic effort to reduce urinary protein excretion. The final BP were 121 ± 15 mmHg (systolic) and 72 ± 10 mmHg (diastolic), and the final urine protein/creatinine ratios were 1.4 ± 1.6,8 values that are much higher than we report for our patients with stages 3 and 4 CKD and Fabry disease at the end of the 30-mo follow-up period (Table 2). A phase III extension study with agalsidase-β has been reported10; minimal rates of progression during a 5-yr follow-up period were seen patients who did not initially have overt proteinuria. Of note, a subset of patients who had normal initial eGFR but had glomerulosclerosis in 50% of their glomeruli and/or proteinuria >1 g/d had rapid decline of their eGFR.10 Schiffmann et al.12 described 14 patients with overt baseline proteinuria; 11 were switched from bimonthly to weekly infusions of agalsidase-α and followed for 24 mo. At the switch, mean eGFR was 53.7 ± 20.9 ml/min per 1.73 m2, and mean rate of progression was −8.0 ± 2.7 ml/min per 1.73 m2 per yr. With weekly dosing of agalsidase-α, the progression rate was −3.3 ± 4.6 ml/min per 1.73 m2/yr, but urinary protein excretion remained the same. The improved progression rate suggests an effect of increasing the dosage of ERT, as has previously been noted,11,13 but these progression rates are greater than reported herein with agalsidase-β and ACEI/ARB therapy. The importance of BP control17,20,21 and the renal protective effects of antiproteinuric therapy have received well-deserved attention.22–26 The primary effect is undoubtedly due to reductions in BP, but other, nonhemodynamic effects may account for improved outcomes with ACEI/ARB compared other antihypertensive agents. We have applied this approach for reducing proteinuria to patients who had Fabry nephropathy and were also being treated with agalsidase-β. Our earlier experience suggested that the beneficial effects of ACEI/ARB therapy on eGFR progression requires optimal dosing of ERT; we would not expect ACEI/ARB therapy to slow eGFR progression in the absence or with suboptimal dosages of ERT.11 Schiffmann et al.12 recently confirmed this conclusion that 0.2 mg/kg body wt agalsidase-α given every 2 wk is not an effective dosage of ERT for treating moderately severe Fabry nephropathy. Although ACEI/ARB therapy reduced systemic BP, the urinary protein/creatinine ratios were reduced to ≤0.5 g/g. The achieved systemic BP levels were similar in our patients with stages 3 and 4 CKD to patients with stages 1 and 2 CKD before or after institution of ACEI/ARB therapy and were reasonably well tolerated without any serious adverse events. This approach is similar to that recently proposed for treating diabetic nephropathy,26 with the goal of reducing urinary protein excretion to ≤0.5 g/d, rather than simply lowering systemic BP to a preset target level. Confirmation of our single-center experience with a larger group of patients in a multicenter study is under way, using combined agalsidase-β and ACEI/ARB therapy in which the primary treatment effect will be sustained reduction in urinary protein/creatinine ratio ≤0.5 g/g, with the primary outcome measure being the rate of progression of eGFR.13,27 CONCISE METHODS Patient Selection and Treatment We describe the first group of patients who had Fabry disease and were treated at UAB with agalsidase-β (Fabrazyme; Genzyme Corp., Cambridge MA), given intravenously at 1 mg/kg body wt every 2 wk. Four male patients who had kidney transplants before starting ERT and were not expected to have any effect of ERT on their kidney function are excluded from this The 30-mo observation period was a to mo the median follow-up period of the phase IV study of agalsidase-β The patients were seen on a in our and received standard clinical care for patients with initial the patients were started on ACEI/ARB therapy for proteinuria. of their ERT treatment was begun with agalsidase-β. were all of which were from the UAB Fabry and urine were in urine at the initial evaluation and at all subsequent patients by the UAB and were in the Fabry This open-label study was at and The primary goal was to the of ACEI/ARB therapy in dosages that would the urinary protein/creatinine ratio in Fabry nephropathy. The eGFR rate of decline was the outcome The of in Disease eGFR serum and age were all of the were at the UAB Clinical the has been The progression rate per 1.73 m2 per yr) was as the regression of the eGFR obtained at each follow-up The data obtained before initiation of ACEI/ARB therapy, at the to the start of agalsidase-β therapy, at the to 6 18 and at the for each patient. The figures individual patient regressions for urine protein/creatinine ratio and The first patient was seen on and the patient the study on The final on The patients were treated with ACEI/ARB therapy for 7.3 ± 6.4 mo (median mo; range to before starting ERT. The average duration of follow-up was 30.3 ± 9.8 mo (median 30.1 mo; range 16 to after starting ERT for the 10 patients who could be with the phase IV we the patients two groups on the of the initial Four patients with eGFR ml/min per 1.73 m2 1 and 2 and seven patients with eGFR ml/min per 1.73 m2 3 and 4 were for of and regression for were for BP and eGFR, and these were compared with Urine protein excretion is not in Fabry these were with urinary protein/creatinine The of proteinuria was in the phase IV we the because we had patients to the regression was for individual patients as well as The of these were compared with the of urine protein/creatinine ratio and eGFR obtained before ACEI/ARB therapy was started. The regression slopes were to the of urine protein/creatinine ratio and eGFR over time after ERT was begun. was when the values were data are presented as ± and as a for Genzyme and on and the on Fabry disease. has received from Genzyme and participated in the phase IV agalsidase-β has been by from Genzyme The patients described in this study all have been in the Fabry at of the and of the were of Genzyme Corp., including any and all and data has no of of mean BP measurements with initial GFR ml/min per 1.73 m2. with initial GFR ml/min per 1.73 m2. The time was at which agalsidase-β therapy was Systolic BP. angiotensin-converting enzyme receptor ERT, enzyme replacement protein/creatinine ratio for individual with initial GFR ml/min per 1.73 m2. with initial GFR ml/min per 1.73 m2. patient is shown as an individual regression starting with the time at which agalsidase-β therapy was the in the A and of in Disease GFR during the are shown for each patient and to their individual progression rates per 1.73 m2 per The point for each patient was at the start of agalsidase-β GFR and progression rates during the values for the four patients with severe Fabry nephropathy; mean values for the six patients with more Fabry nephropathy. The mean progression slopes are shown for both groups and are not significantly different from the average regression intercept with average initial eGFR when therapy was and clinical BP, urine and patient regression of ratios over Urine and ratios and urine protein and albumin patient regression of eGFR over our patients for and in this We also for of and National Institutes of and for patient and baseline clinical and results from this study were presented at the of the of 8 and 14 and in and