Danaparoid Sodium Lowers Proteinuria in Diabetic Nephropathy

JOHAN W. VAN DER PIJL,* FOKKO J. VAN DER WOUDE,* PETRONELLA H. L. M. GEELHOEDDUIJVESTIJN,t MARIJKE FROLICH, FELIX J. M. VAN DER MEER, HERMAN H. P.J. LEMKES,” and LEENDERT A. VAN ES,* *Department of Nephrology, Leiden University Hospital, P0 Box 9600, 2300 RC Leiden, The Netherlands; tDepartment of Internal Medicine, Den Haag, The Netherlands; *Department of Clinical Chemistry, Leiden

University Hospital, Leiden, The Netherlands; Department of Hematology, Leiden University Hospital, Leiden, The Netherlands; ‘Department of Endocrinology, Leiden University Hospital, Leiden, The Netherlands.

Abstract. Diabetic nephropathy is a progressive renal disease sodium, the albumin excretion ratio standardized for urinary with thickening of the gbomerular basement membrane and creatinine reduced with 17% in comparison with an increase of mesangial expansion and proliferation as histological hall- 23% after placebo (95% confidence interval of the difference, marks. The presence of the glycosaminoglycan side chains of -75.9-3.9%; P = 0.03). The percentage change of the urinary heparan sulfate proteoglycan, an important constituent of the protein excretion corrected for urinary creatinine differed at 8 glomerular basement membrane, is decreased in diabetic ne- wk significantly between both treatment arms (P 0.001). phropathy proportionally to the degree of proteinuria. Danap- Additional parameters for safety as hematobogical, hemostasis, aroid sodium is a mixture of sulfated glycosaminoglycans biochemical parameters, and fundusphotography did not show consisting mainly of heparan sulfate. The study presented here any clinically significant difference for both groups. Only two involved performing a randomized placebo-controlled cross- patients had minor skin hematomas at the injection site while over study with danaparoid sodium in diabetic patients with using danaparoid sodium. overt proteinuria. The aim of the study was to evaluate the In conclusion, the supplementation was found to be feasible effect on proteinuria and safety/tolerability. Nine patients com- and was not associated with side effects. A significant decline pleted the study, without major side effects; the crossover of proteinuria was found. More prospective dose-finding and study consisted of two 6-wk periods of treatment with 750 long-term studies must be performed to see whether danap- anti-Xa units danaparoid sodium subcutaneously once-daily or aroid sodium could not only induce a reduction of proteinuria placebo. Following danaparoid sodium, significant declines of but also halt the progression of renal disease. (J Am Soc both albuminuria and proteinuria were found. After danaparoid Nephrol 8: 456-462, 1997)

Diabetic nephropathy occurs in 30%-40% of patients with importance because mesangial expansion and proliferation are insulin-dependent diabetes melbitus (DM I). This type of renal both strongly correlated with decline of renal function (4,1 1). disease progresses over time and is strongly related to a high A reduced synthesis and sulfatation of heparan sulfate proteo- mortality and morbidity rate (1-4). Proteoglycans and espe- glycans by podocytes and mesangial cells, when cultured under cially the glycosaminoglycan (OAO) side chains of heparan high glucose conditions, has been reported (12,13). sulfate (HS) are considered to be involved in the pathogenesis The key role of HS for the normal function of the OBM is of diabetic nephropathy. Heparan sulfate is thought to be illustrated by the following findings: removal of HS by hepa- important for the gbomerular basement membrane (OBM) in rinitase (14) or shielding of HS by a monoclonal antibody (15) terms of structure and function. It prevents clogging of the generates albuminuria in vivo. In type I diabetic patients with membrane and determines the sieving-function (5,6). HS is nephropathy, staining for heparan sulfate OAOs in the OBM is mainly responsible for the negative-charge barrier in the OBM reduced in proportion to the amount of proteinuria (16,17). (7), which is of relevance for preventing the passage of albu- In the context of such a pivotal role of HS in diabetic mm through the capillary wall. HS may also slow down mes- nephropathy, several studies have investigated whether supple- angial expansion and proliferation (8-10). This is of utmost mentation of sulfated glycosaminoglycans could be of any benefit. Indeed, the development of diabetic nephropathy could be prevented by giving glycosaminoglycans to rats with strep- Received September 5, 1996 Accepted November 1 1, 1996 tozotocin-induced diabetes meblitus (18). Several clinical stud- Correspondence to Dr. J.W. van der Pijl, Department of Nephrobogy, C3-P, 1 Leiden University Hospital, P0 Box 9600, 2300 RC Leiden, The Netherlands. ies have reported preliminary results in type I diabetics with micro- and macroalbuminuria. Myrup et a!. found a reduction l046-6673/0803-0456$03.00/0 Journal of the American Society of Nephrology of microalbuminuria after treatment with low molecular weight Copyright (C) 1997 by the American Society of Nephrobogy heparin (LMWH) and with standard heparin (19). In mac- Danaparoid Sodium and Proteinuria 457

roalbuminuric patients treated with enoxaparin (a LMWH), a dialysis. The placebo had the same appearance as the danaparoid significant decline of the urinary albumin excretion rate (AER) sodium containing ampoubes; it contained isotonic sodium chloride was found, but significance was not reached in comparison and sodium sulfite. Danaparoid sodium (0.6 ml, 750 anti-Xa units) with the placebo treated group (20). and the placebo were administered once-daily by subcutaneous (SC) injection for 6 wk, followed by a wash-out period of at least 4 wk. The Danaparoid sodium is a new, commercially available, hep- patients were trained to self-administer the study medication in an armnoid. It is a mixture of sulfated OAOs, consisting mainly of abdominal skinfold. The date and time of each injection was recorded HS with a small subfraction, which is highly sulfated. It is used daily by the patient on a list and checked by the investigator at the end both for the prevention and treatment of venous thromboem- of each treatment period, together with the number of returned unused bolism, where it has been shown to be efficacious and rela- ampoules. Protamine-containing insulin preparations were not al- lively safe. It functions by augmenting the inhibitory action of bowed to be injected at the same site, and study medication was not antithrombin on activated clotting factors Xa and ha. The high allowed to be mixed with insulin. ratio of anti-Xa/anti-IIa activity is thought to be responsible for a good efficacy/safety profile. It is seldom associated with heparin-induced thrombocytopenia. Clearance takes place Efficacy Assessments mainly by the kidneys (21-23). Therefore, and because dan- In diabetic nephropathy, both proteinuria and macroalbuminuria are associated with a steady decline of renal function. Taking the aparoid sodium consists to a large extent of heparan sulfate, we relative short duration of treatment into account, we decided to use decided to study the effect of danaparoid sodium on proteinuna proteinuria and albuminuria as surrogate markers. These parameters in type I diabetic patients. were expressed per mmol of creatinine to correct for incomplete urinary collections. The main parameters for efficacy were urinary Materials and Methods albumin excretion rate (AER) and urinary protein excretion rate Design (PER) corrected for urinary creatinine excretion rate. Therefore, total The study was designed as a phase II, randomized, double-blind, protein, albumin, and creatinine contents in 2 X 24-h urine samples placebo-controlled, crossover study. The medical ethical committee of (collected by the subject at home) were assessed. Baseline measure- the Leiden University Hospital had approved the protocol of the study, ments were taken before both treatment periods. At 0, 4, 6, and 8 wk, and all patients gave informed consent. Eligible patients were ran- AER, PER, and creatinine clearance were measured twice, and the domly allocated to one of the two treatment groups: 750 anti-Xa units means were compared with baseline (0 wk). danaparoid sodium (Orgaran#{174}, NV Organon, Oss, the Netherlands), subcutaneously (SC) administered once daily for 6 wk; a wash-out period of at least 4 wk; placebo (saline with sulfite), SC administered Safety Assessments once-daily for 6 wk, or the same scheme in the opposite direction. A general medical history (including duration of patients’ DM I),

Medical and laboratory assessments were performed every 2 wk (0 = height, weight, blood pressure, and heart rate after 3 mm in sitting baseline; 2, 4, 6, and 8 wk after start of the respective treatments). position were taken, and a medical examination was performed. Fundusphotography (including staging of observed retinopathy) was performed before the inclusion and at the end of the study period. Patients Sodium, potassium, urea, creatinine, glucose, total protein, albumin, The study population consisted of nine patients who had nephrop- serum glutamic oxaboacetic transaminase (SOOT), serum glutamic athy due to DBM and fulfilled the following selection criteria: they pyruvic transaminase (SOPT), alkaline phosphatase, yGT, HbA , had to be > 18 years old; they had to have insulin-dependent diabetes hemoglobin, hematocrit, leukocytes, thrombocytes, prothrombin time, mellitus type I (DM I) with macroabbuminuria (AER >300 mg/24 h) antithrombin III, activated partial thromboplastin time (APTE’), and because of diabetic nephropathy and not because of other renal plasma anti-Xa activity were tested regularly as safety and tolerability disease; and they had to give informed consent. The use of angiotensin parameters. I converting enzyme (ACE)-inhibitors and other antihypertensive medications was only allowed if given at a stable dosage at least 6 wk prior to start of the study. Laboratory Assessments Patients complying with the inclusion criteria were eligible if none Urinary albumin was assessed using immunonephebometry on an of the following exclusion criteria were present: pregnancy or wish to autoanalyser (Array Protein system, Beckman Instruments, Brea, CA) become pregnant; hemorrhagic diathesis; use of medications known to with specific antibodies. Urinary protein, serum sodium, potassium, interfere with hemostasis and/or platelet function; use of corticoste- urea, creatinine, protein, albumin, SGOT, SGPT, alkaline phospha- roids; uncontrolled hypertension; active proliferative retinopathy; his- tase, and yGT levels were assessed on a Hitachi 747 or 91 1 autoanal- tory of heparin-induced thrombocytopenia; severe hepatic failure; yser (Boehnnger, Mannheim, Germany). The Technicon H-l system creatinine clearance of <40 mL/min; hypersensivity to sulfite. (Bayer Diagnostics, MUnchen, Germany) was used for the analysis of hematobogical parameters. Prothrombin time was determined using Medication the Thromborel S reagent (Behringwerke AG, Marburg, Germany), Danaparoid sodium is a mixture of sulfated GAGs, consisting of amd APTT was determined using the Cephotest reagent (Nyegaard, 84% heparan sulfate (HS), 12% dermatan sulfate, and 4% chondroitin Oslo, Norway), both on an Electra 1000C coagulometer (MLA, Pleas- sulfate isolated from porcine intestinal mucosa. It has an average antville, NY). Antithrombin and anti-Xa activity were assessed on an molecular of weight 4000-7000 d, a specific anti-Xa activity of ACL 200 (Instrumentation Laboratory, Milan, Italy) with Coamatic 11.0-17.0 U/mg, and an anti-Xa/anti-IIa ratio of >22. The elimina- Antithrombin and Coatest LMWHeparin as reagents (Chromogenix, tion half-life of anti-Xa activity is -25 h. The half-life time is only M#{246}lndal, Sweden). HbAIC was assessed on high-performance liquid affected by severe renal failure, as seen in patients on chronic hemo- chromatography (HPLC) after hemolysis. 458 Journal of the American Society of Nephrology

Table 1. Demographic dataa

Patient Age Sex DM BMI MAP HbA1C AER AER/Creat PER/Creat CrCb ACEi no. (y) (y) (kg/rn2) (mmHg) (%) (mg/24 h) (mg/mmol) (mg/mmol) (mi/mm)

1 43 M 20 No 32.8 99 8.3 1000 64 90 119 2 52 M 28 Yes 29.8 100 7.5 1032 91 122 70 3 32 F 20 Yes 22.4 107 7.6 3044 250 368 62 4 33 M 24 No 27.1 107 7.4 968 63 97 127

5 35 M 31 No 27.6 113 10.8 715 95 103 54 6 44 M 25 Yes 28.7 103 7.7 912 64 86 141 7 29 M 16 No 22.1 105 7.4 670 52 72 68 8 29 M 26 Yes 23.9 103 7.4 258 21 35 96 9 25 M 14 No 28.7 107 9.8 403 46 70 79

a DM, diabetes mellitus type I; ACEi, yes denotes the use of ACE inhibitor, and no denotes no ACE inhibition; BMI, body mass index; BP, blood pressure; MAP, mean arterial pressure; AER/creat, urinary albumin excretion rate standardized for urinary creatinine excretion rate; PER/creat, urinary protein excretion rate standardized for urinary creatinine excretion rate; CrCl, creatinine clearance.

Statistical Analyses tients were treated with lasercoagulation for proliferative reti- The mean change of the primary and secondary parameters during nopathy for ophthalmologic reasons only. Demographic data the two treatment periods were analyzed using a mixed-model anal- and characteristics are given in Table 1 . In general, the HbAlc ysis of variance (ANOVA), with random patient factor to account for reflected an acceptable metabolic control of the diabetes. In the correlations between the repeated measurement, and fixed treat- contrast, the blood pressure was rather high for these patients. ment and time factors. In addition, single degree-of-freedom (paired t) The albuminuna and proteinuria ranged from borderline tests were used to evaluate differences between and within the two treatment periods. Because the distribution of the primary parameters “overt” (patient 8) to proteinuna in the nephrotic range (patient was extremely skewed, a logarithmic transformation was applied. 3). Four patients were on ACE-inhibitors, and one of these also values (relative change) for AERlcreat and PER/creat were calculated used amlodipine. Table 2 shows the primary parameters per as follows: group and treatment, Tables 3a and 3b show individual results. (y-x/x)* 100% = 6, where y is the nth week value for AERicreat or HbAlc was 8.5% before the treatment period with danap- PER/creat and x is the 0 week value. Significance was accepted at the aroid sodium and 8.3% before placebo (P = 0.4). No major P-value level of 0.05. The calculations were performed with the SPSS changes were made for the insulin dosage during both periods. package release 6.0 for Windows. The baseline values of the MAP did not differ between the two Results groups (103 mmHg and 105 mmHg; P = 0.3). No major Patients adverse events occurred. One patient needed antibiotic treat- Nine patients were enrolled in the study, and all completed ment during the placebo period because of sinusitis without both crossover treatment periods. Prior to the study, two pa- fever. Subcutaneous injections were performed once daily dur-

Table 2. Creatinine clearance, blood pressure, AER/creatinine, and PER/creatinine before and after treatment with placebo and danaparoid sodiuma

Danaparoid Sodium Placebo

0 wk 8 wk P value 0 wk 8 wk P value

CrC1 (mi/mm) 102.9 96.9 0.37 90.4 100.5 0.20 (75.5-1 30.4) (68.4-1 25.4) (74.0-106.8) (69.7-131.4) MAP (mm Hg) 103.0 102.2 0.68 105.4 98.5 0.26 (100.6-105.4) (98.5-106.0) (98.9-1 1 1.8) (92.7-104) AERlcreat (mg/mmol) 74.8 60.6 0.03 64.8 75.8 0.18 (42.7-1 3 1 .2) (33. 1-1 10.9) (36.7-1 14.7) (37.2-154.3) PER/creat (mg/mmol) 111.0 89.3 0.008 87.0 110.6 0.12 (65.5-1 87.9) (5 1 .9-1 53.8) (5 1 .9-146.0) (57.4-213.1)

a All values are expressed as mean and (95% confidence interval of the mean), except for AER/creat and PER/creat, which are expressed as geometric mean (antilog 95% CI). CrCb, creatinine clearance; MAP, mean arterial pressure; AER/creat, urinary albumin excretion rate standardized for urinary creatinine excretion rate; PER/creat, urinary protein excretion rate standardized for urinary creatinine excretion rate. P values are obtained with paired t tests (within each treatment block). Danaparoid Sodium and Proteinuria 459

Table 3a. Individual results: AER/creat

. Danaparoid Sodium Placebo Patient R No. wk 4 () wk 6 (6) wk 8 (3) wk 0 wk 4 () wk 6 () wk 8 () wk

1 D 64 83(+29%) 75(+17%) 53(-18%) 78 63(-20%) 77(-l%) 77(-l%) 2 P 133 1l4(-14%) lll(-l7%) 107(-l9%) 90 74(-19%) 138(+52%) l24(+36%) 3 D 250 205(-l8%) 238(-5%) 217(-13%) 280 292(+4%) 223(-20%) 337(+2l%) 4 P 75 78(+4%) 90(+l9%) 77(+3%) 63 87(+37%) 75(+18%) 73(+l5%) 5 P 144 b86(+29%) l29(-ll%) l13(-22%) 96 132(+38%) l36(+42%) ll8(+24%) 6 D 64 33(-49%) 42(-35%) 29(-54%) 39 44(+l4%) 55(+4l%) 47 (+21%) 7 D 52 56(+8%) 43(-l8%) 54(+3%) 60 38(-36%) 43(-28%) 42 (-30%) 8 P 21 l9(-lb%) 22(+4%) 15(-26%) 21 24(+13%) l3(-40%) 20(-6%) 9 D 46 26(-44%) 40(-l2%) 43(-7%) 34 nd 72(+112%) 78(+l27%)

Table 3b. Individual results: PER/creat’

. Danaparoid Sodium Placebo Patient R No. wk 4 () wk 6 () wk 8 (t) wk 0 wk 4 () wk 6 () wk 8 () wk

1 D 90 l31(+46%) l0l(+l3%) 97(+8%) 119 105(-12%) l23(+3%) l23(+3%) 2 P 178 l9l(+8%) l52(-15%) l32(-26%) 122 9l(-26%) l7l(+40%) l63(+33%) 3 D 368 302(-l8%) 347(-6%) 323(-12%) 365 526(+44%) 351 (-4%) 459(+26%) 4 P 127 126(-l%) l30(+2%) l08(-l5%) 97 l30(+35%) llO(+14%) ll2(+l6%) 5 P 211 235(+ll%) l90(-bO%) l40(-34%) 52 195(+277%) l75(+237%) 15l(+l92%) 6 D 86 64(-25%) 62(-27%) 52(-40%) 62 73(+l8%) 80(+29%) 79(+28%) 7 D 72 67(-6%) 63(-l2%) 69(-4%) 83 50(-40%) 58(-29%) 52 (-37%) 8 P 38 34(-ll%) 39(+2%) 27(-29%) 35 43(+2b%) 29(-l9%) 35(-2%) 9 D 70 37(-48%) 58(-l7%) 60(-14%) 60 nd 89(+48%) l24(+105%)

a Note. values (relative change) for AER/creat and PER/creat were calculated as follows: (y - x/x*lOO%) = 5, where y is the net’ week value for AER/creat or PER/creat and x is the 0 wk value. R (randomization) denotes whether danaparoid sodium (D) or placebo (P) was the first treatment block; AER/creat (and PER/creat), urinary albumin (protein) excretion rate standardized for urinary creatinine excretion rate; nd, no data.

ing both treatment periods without clinically important side 150. effects. In two patients, small skin hematomas at the injection 8) 100. site occurred during treatment with danaparoid sodium. Ac- ‘3 50. cording to the number of returned unused ampoules and the AA& compliance checklist, the compliance was excellent; only two a) 0) C Tyvy, patients skipped one injection. The duration of treatment was Cs -50. 42 days (median) for danaparoid sodium (range, 41-47) and 42 ‘3 -100. days (range, 41-45) for the placebo treatment. -150. danaparoid placebo

Efficacy Parameters Figure 1. A scatter diagram showing the percentage difference () of A significant decline of both albuminuria and proteinuria the albumin excretion rate standardized for the urinary creatinine was found after treatment with danaparoid sodium compared excretion comparing 8-wk with baseline in both treatment arms.

with baseline and with the placebo treatment. Figure 1 shows values for AER/creat and PER/creat were calculated as follows: (y - the percentage reduction of albuminuria for both groups. Com- x/x)*l00% = & where y is the 8-wk value for AER/creat and x is the pared with baseline, the AER/creatinine at the end of the 0-wk value. The difference is significant (P < 0.03). treatment period was - I 7% for the danaparoid sodium group and +23% for the placebo group (95% confidence interval (CI) ofthe difference -75.9-3.9%, P = 0.03). Even after omitting 40% was seen in the placebo group (95% CI of the difference

the outlier of 126% increase during placebo, a significant - 1 15.8-1.0%, P <0.05). Figure 2 shows the mean percentage difference was found (P <0.05). The protein excretion rate change of the urinary protein/creatinine ratios during both standardized for creatinine excretion showed a decrease of treatment regimens. The difference of the overall change pat- 18% in the danaparoid sodium group, whereas an increase of tern in both treatment groups was highly significant (P = 460 Journal of the American Society of Nephrology

‘I) treatment, antihypertensive treatment in general, and especially > ACE inhibition have been shown to slow the progression of w diabetic nephropathy (24-27). We decided to evaluate in a

25 clinical setting the applicability of proteoglycans as an addi- tional treatment modality. All of our patients had macroabbu- 0).O 0 C minuria, and they had, therefore, passed the point of revers- Cs ibility of this specific renal disease. The Danish group has U -25 already shown a benefit of LMWH and heparin for microalbu- minuric patients (19). A previous study with enoxaparin in weeks macroalbuminuric patients from our unit failed to demonstrate

Figure 2. The percentage change () of the protein excretion rate a statistically significant difference between placebo and standardized for urinary creatinine excretion from baseline in both LMWH (20), although a significant decrease in proteinuria was treatment arms. The overall difference is P = 0.001 . The relative seen in the treated group when pretreatment values were taken change at the respective timepoints 4, 6, and 8 wk in comparison with into consideration. The difference between placebo and baseline was as follows: P = 0.22 (95% CI of the difference -118.3- LMWH may not have been found because of the small number 29.2%), P = 0.13 (95% CI ofthe difference - 100.1-14.1%), and P = of patients studied and the short duration of the study. In the 0.02 (95% CI of the difference - 108.3-9.5%). The bars are standard study presented here, we observed a clear reduction for the errors of the mean. Triangles denote placebo, and squares denote the AER and PER after treatment with danaparoid sodium for 6 danaparoid sodium treatment. wk. Only one woman participated in this study (with a good response on therapy); we acknowledge more women need to be studied before our results may be applied to both sexes. Al- 0.001). The relative change at the respective timepoints of 4, 6, though not clinically or statistically significant, MAP showed a and 8 wk in comparison with baseline was P = 0.41 (95% CI trend to decline in the placebo group. This might be explained of the difference -39.6-17.2%), P = 0.15 (95% CI of the by regression to the mean. We found no effects on blood difference -61.7-10.1%), and P = 0.02 (95% CI of the pressure during danaparoid sodium treatment. difference -73.2-6.6%). Similar results were found for the The most important result of our study is that the goal to overall change pattern for urinary albumin/creatinine ratios reduce AER and PER by HS therapy may be feasible. The (P = 0.002). Especially Figure 2 demonstrates that the effects favorable chemical and clinical profile of danaparoid sodium in on proteinuria (and albuminuria-figure not shown) become comparison to heparin and LMWHs made us decide to study more apparent after 2 months than after the first month. A this medication (28). Heparmn-induced thrombocytopenia and small and not significant difference was found comparing the thrombosis is a feared clinical picture with a dismal outcome. pre- and post-treatment values for the MAP: after placebo, a Only danaparoid sodium in comparison to heparin and LM- decline of 4.7% was found, and after danaparoid sodium, a WHs is associated with a very low incidence of this compli- decline of 0.9% was found (P = 0.4). Creatinine clearances did cation (28). Osteoporosis is another drawback associated with not differ significantly, either: -5. 1% after danaparoid sodium long-term heparin use. Limited experience with LMWHs and + 16.6% after placebo (P = 0.2). Subanalysis respective- showed a more favorable outcome in this respect (29,30). lyof the use of ACE inhibition and the order of treatment Heparins and HS both exert anti-inflammatory actions as (danaparoid sodium/placebo or placebo/danaparoid sodium) well as antiproliferative effects. An excellent short review on was not possible due to low figures this topic was recently published by Wardle (3 1). Heparin prevents basic fibroblast growth factor induced proliferation Safety Parameters and may also stop smooth muscle cell proliferation by releas- No changes in visual acuity occurred during treatment. Fun- ing transforming growth factor 3 from its binding protein dusphotography did not show hemorrhages, and no progression (32,33). It also inhibits the synthesis and release of endothelin-I of retinopathy was found in any of the patients. Anti-Xa values by endothelial cells (34). The negative charge at the N position were detectable only once in two patients (respectively, 0.12 is thought to be required for the antiproliferative action (35). and 0.10 U/mL; bower limit of detection <0.10 U/mL); this Danaparoid sodium is not a LMWH but a heparinoid. It was not accompanied by any change of coagulation test results consists of 84% HS. Only a small fraction (4%-5%) of this HS (APTT). No changes occurred for hematological parameters, is highly sulfated and is considered to be important for the and values within each patient during both treatment periods anti-Xa activity. The fraction of HS with a low affinity for remained about the same for hemoglobin, leukocytes and antithrombin III does not affect coagulation factors Xa and lIa thrombocytes. Liver enzymes (including alkaline phosphatase) but contributes substantially to the antithrombotic activity, did not change, either. probably through an effect on endothelial cells (23). The dose- related response to danaparoid sodium remains gradual and Discussion linear over a wide range, which may contribute to its safety. It Optimizing the treatment of DM I with the goal of slowing has a 100% bioavailibility after subcutaneous administration. down the progression of vascular and renal damage has been Studies in humans have been using heparin or LMWH. We the subject of many studies. In the past , intensified insulin thought that danaparoid sodium could have a more favorable Danaparoid Sodium and Proteinuria 461

benefit/risk profile in case of long-term prescription in com- diabetic renal disease: With emphasis on the stage of incipient parison to LMWH. The theoretical arguments of a bower sub- diabetic nephropathy. Diabetes 32: 564-578, 1983 fate content and, hence, less effect on proteinuria were not 3. Deckert T, Feldt-Rasmussen B, Borch-lohnsen K, Kofoed- Enevoldsen A: Abbuminuria reflects widespread vascular dam- validated by our findings. Whether only the 4%-5% high- age: The Steno hypothesis. Diabetologia 32: 219-226, 1989 affinity HS fraction has been of relevance for the antiprotein- 4. Osterby R, Gundersen HI, Horlyck A, Kroustrup IP, Nyberg 0, uric effect or the low sulfated fraction has been of a comple- Westberg 0: Diabetic gbomerubopathy: Structural characteristics mentary benefit has yet to be established. Based on the half-life of the early and advanced stages. Diabetes 32: 579-582, 1983 time of danaparoid sodium (25 h for the anti-Xa activity), a 5. Kanwar YS, Rosenzweig I, lakubowski ML: Distribution of de 4-wk wash-out period was thought-prior to the study-to be novo synthesized sulfated glycosaminoglycans in the glomerular reasonable and long enough. The results and especially the basement membrane and the mesangial matrix. Lab Invest 49: increase of proteinuria during the placebo period suggest in 216-225, 1983 retrospect that this increase might, in part, be explained by a 6. Kanwar YS, Rosenzweig U: Clogging of the gbomerubar base- carry-over effect. However, in the Danish study, the LMWH ment membrane. J Cell Biol 93: 489-494, 1982 and the placebo groups also showed 3 months after the study an 7. Kanwar YS, Hascall VC, Farquhar MG: Partial characterization increase of albuminuria of 30% and 125% relative change, of synthesized proteoglycans isolated from the glomerular base- ment membrane. J Cell Biol 90: 527-532, 1981 respectively (19). For further crossover studies, not only the 8. Groggel OC, Marinides ON, Hovingh P, Hammond E, Linker A: drug half-life time, but also an additional lag-time because of Inhibition of rat mesangiab cell growth by heparan sulfate. Am J the mediated effects of danaparoid sodium should be taken in Physiol 258: F259-F260, 1990 consideration to prevent this putative confounding effect. 9. Striker LI, Peten EP, Elliot SI, Doi T, Striker GE: Mesangiab cell Our results are in agreement with previously reported results turnover: Effect of heparin and peptide growth factors. Lab of proteoglycans on proteinuria in diabetic nephropathy lnvest 64: 446-456, 1991 (19,20). This short-term study demonstrates, as already seen in 10. Coffey AK, Karnovsky Ml: Heparin inhibits mesangial cell animal models, that proteoglycans can exert a reversible effect proliferation in habu-venom-induced gbomerular injury. Am J also at a late stage of nephropathy ( 18). Because we have not Pathol 120: 248-255, 1985 taken any renal biopsies to substantiate the level of changes on 1 1. Mauer SM, Steffes MW, Ellis EN, Sutherland DES, Brown DM, histological examinations, we can only hypothesize about the Ooetz FC: Structural-functional relationships in diabetic ne- phropathy. J Clin Invest 74: 1 143-1 155, 1984 mechanism of action. As danaparoid sodium is mainly cleared 12. van Det NF, van den Born IT, Tamsma IT, Verhagen NAM, van by the kidneys, both replacement of “faulty” HS molecules in den Heuvel LPWI, Berden IHM, Bruijn JA, Daha MR. van der the OBM as well, as a beneficial effect on smooth muscle cell Woude Fl: Proteoglycan production by human gbomerular vis- and mesangial cell proliferation, could have accounted for the ceral epithebial cells and mesangial cells in vitro. Biochem J 307: clinically observed effects. In previous studies using sulfated 759-768, 1995 glycosaminoglycans, no differences were found for gbomerular 13. van Det NF, van den Born I, Tamsma IT, Verhagen NAM, filtration rate, renal plasma flow, and filtration fraction Berden IHM, Bruijn IA, Daha MR. van der Woude Fl: Effects of (19,20,36). This suggests that the antiproteinuric effect of HS high glucose on the production of heparan sulfate proteoglycan is not related to altered renal hemodynamics. by human mesangial and gbomerular visceral epithebiab cells in In conclusion, the once-daily SC administration of 750 an- vitro. Kidney Int 49: 1079-1089, 1996 ti-Xa u of danaparoid sodium resulted in a significant reduction 14. Rosenzweig LI, Kanwar YS: Removal of sulfated (heparan sulfate) of proteinuria in patients with overt diabetic nephropathy. or nonsulfated (hyaluronic acid) glycosaminoglycans results in in- creased permeability of the gbomerular basement membrane to ‘ More clinical studies are needed to confirm and extend our bovine serum albumin. Lab Invest 47: 177-184, 1982 findings. A dose-finding study should also be performed. It is 15. van den Born I, van den Heuvel LP, Bakker MA, Veerkamp IH, obvious that only long-term studies will prove that the reduc- Assmann IU, Berden IH: A monocbonab antibody against OBM tion of proteinuria will be associated with a slower decline or heparan sulfate induces an acute selective proteinuria in rats. even maintenance of renal function in diabetic nephropathy. Kidney mt 4 1: 1 15-1 23, 1992 16. Vernier RL, Steffes MW, Sisson-Ross 5, Mauer SM: Heparan sulfate proteoglycan in the gbomerubar basement membrane in Acknowledgments type I diabetes mellitus. Kidney mt 41: 1070-1080, 1992 Ian Egberts (NV Organon, Oss, the Netherlands) has contributed 17. Tamsma IT, van den Born I, Bruijn IA, Assmann KJM, Weening with fruitful discussions and provided us with medication and pla- IJ, Berden IHM, Wieslander I, Schrama E, Hermans I, cebo. We are indebted to Koos Zwinderman (Department of Medical Veerkamp IH, Lemkes HHPJ, van der Woude Fl: Expression of Statistics, University of Leiden, The Netherlands), who assisted in the glomerular extracelbular matrix components in human nephrop- statistical analysis. athy: Decrease of heparan sulphate in the gbomerular basement membrane. Diabetologia 37: 313-320, 1994 18. Oambaro 0, Cavazzana AO, Luzzi P, Piccoli A, Borsatti A, References Crepaldi 0, Marchi E, Venturini AP, Baggio B: Glycosamino- 1. Andersen AR, Christiansen IS, Andersen 1K, Kreiner S. Deckert glycans prevent morphological renal alterations and albuminuria T: Diabetic nephropathy in type I (insulin-dependent) diabetes: in diabetic rats. Kidney mt 42: 285-291, 1992 An epidemiobogical study. Diabetologia 25: 496-501, 1983 19. Myrup B, Hansen PM, Iensen T, Kofoed-Enevoldsen A, Feldt- 2. Mogensen CE, Christensen CK, Vittinghus E: The stages in Rasmussen B, Gram I, KIuft C, Iespersen I, Deckert T: Effect of 462 Journal of the American Society of Nephrology

low-dose heparmn on urinary albumin excretion in insulin-depen- 27. Lewis El, Hunsicker LO, Bain RP, Rohde RD, for the Collabo- dent diabetes mellitus. Lancet 345: 421-422, 1995 rative Study Group: The effect of angiotensin-converting-en- 20. Tamsma IT, van der Woude FJ, Lemkes HHPI: Effect of sub- zyme inhibition on diabetic nephropathy. N Engi J Med 329: phated glycosaminoglycans on albumiuria in patients with overt 1456-1462, 1993 (type I) nephropathy. Nephrol Dial Transplant 1 1 : 182-185, 28. Magnani HN: Heparin-induced thrombocytopenia (HIT): An 1996 overview of 230 patients treated with orgaran (Org 10172). 21. de Valk HW, Banga ID, Wester 1W, Brouwer CB, van Hessen Thromb Haemostasis 70: 554-561, 1993 MW, Meuwissen 01, Hart HC, Sixma II, Nieuwenhuis HK: 29. Shaughnessy SO, Young E, Deschamps P. Hirsh I: The effects of Comparing subcutaneous danaparoid with intravenous unfrac- low molecular weight and standard heparin on calcium loss from tionated heparin for the treatment of venous thromboembobism: fetal rat calvaria. Blood 86: 1368-1373, 1995 A randomized controlled trial. Ann Intern Med 123: 1-9, 1995 30. Monreal M, Olive A, Lafoz E, del Rio L: Heparmns, coumarin, 22. Bradbrook ID, Magnani HN, Moelker HCT, Morrison P1, Rob- and bone density [Letter]. Lancet 338: 706, 1991 inson I, Rogers HJ, Spector RO, van Dinther T, Wijnand H: 31. Wardle EN: Heparmns for proliferative nephritides: Short review ORG 10172: A low molecular weight heparinoid anticoagulant on an advancing topic. Nephron 73: 515-519, 1996 with a long half-life in man. Br J Clin Pharmacol 23: 667-675, 32. McCaffrey TA, Falcone DF, Brayton CF. Agarwal LA: TOF- 1987 beta activity is potentiated by heparin via dissociation of the 23. Nurmohamed MT. Fareed I, Hoppensteadt D, Wabenga IM, ten TOF-beta-a2-macrogbobubin inactive complex. J Cell Biol 109: Cate 1W: Pharmacological and clinical studies with bomoparan, a 441-448, 1989 low molecular weight glycosaminoglycan. Semin Thromb He- 33. Lindner V, Olson NE, Clowes AW, Reid MA: Inhibition of most 17 52: 205-213, 1991 smooth muscle cell proliferation in injured rat arteries. J Clin 24. The Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development Invest 90: 2044-2049, 1992 and progression of long-term complications in insulin-dependent 34. Imai T, Hirata Y, Emori T, Marumo F: Heparin has an inhibitory diabetes melbitus. N Engi J Med 329: 977-986, 1993 effect on endothelin 1 synthesis and release by endothelial cells. 25. Viberti OC, Mogensen CE, Oroop L, Pauls IF, for the European Hypertension 21: 353-358, 1993 Microalbuminuria Captopril Study Group: Effect of captopril on 35. Tiozzo R, Cingi MR, Reggiani D, Andreoli T, Calandra 5, Milani progression to clinical proteinuria in patients with insulin-depen- MR, Piani 5, Marchi E, Barbanti M: Effect of desulfatation of dent diabetes mellitus and microalbuminuria. JAMA 271: 275- heparin on its anticoagulant and anti-proliferative effect. Thromb 279, 1994 Res 70: 99-106, 1993 26. Parving H-H, Rossing P: The use of antihypertensive agents in 36. Ichikawa I, Yoshida Y, Fogo A, Purkerson ML, Klahr 5: Effect prevention and treatment of diabetic nephropathy. Curr Opin of heparin on the gbomerular structure and function of remnant Nephrol Hypertension 3: 292-300, 1994 nephrons. Kidney mt 34: 638-644, 1988