The use of drugs which can influence the status and progression of osteoarthritis is a relatively new concept. A major impediment to development in this area has been the multifactorial nature of the disorder, which renders definition of the pharmacological requirements for such agents somewhat difficult. Nevertheless, much is known about the pathology of osteoarthritis and in this review the prominent features have been identified and a profile of a suitable second-line agent suggested. It was concluded that potent anti-inflammatory or immunosuppressive activity was not sufficient for assignment of a drug to the classification as second-line agent in osteoarthritis. The drug should also be capable of supporting anabolic (repair) process in cartilage, bone and synovium which were essential for normalisation of joint function. Of particular importance was the need for the drug to improve subchondral and synovial circulation, which are impaired in the osteoarthritic joint. The agent pentosan polysulfate was examined in the light of the requirements postulated. While in vitro and in vivo studies in laboratory and domestic animals have provided clear evidence that this agent can fulfil most of criteria nominated for a DMAOD, clinical studies in man are presently limited. While long-term double-blind studies may provide a useful index of drug efficacy, the methods currently available to allow meaningful clinical assessment of these types of drugs are grossly inadequate. Most rely on improvement of symptoms which may, or may not, reflect a change in the underlying pathology. More objective measures of patient assessment are required and hopefully recent advances in x-ray microfocal techniques (Buckland-Wright et al., 1990; 1995a,b&c; Buckland-Wright, 1994), magnetic resonance imaging (Sabiston et al., 1987; Braunstein et al., 1990; Adams & Wallace, 1991; Chan et al., 1991; O’Byrne et al., 1993; Fernandez-Madrid et al.,1994; 1995; Bergman et al.,1994; Karvonen et al.,1994; Gahunia et al., 1995a&b; Hutton & Vennart, 1995; Kladny et al.,1996; Watson et al.,1996) and biochemical markers of cartilage breakdown (Heinegard et al., 1985; Lohmander et al., 1989; 1992; Shinmei et al., 1992; George et al., 1993; Thonar et al., 1993; 1995; Lohmander, 1995; Roos et al., 1995; Rørvik & Grøndahl, 1995; Sharif et al., 1995; Ratcliff et al., 1996; Georges et al., 1997; Kubota et al., 1997; Petersson et al., 1997; Lohmander & Felson, 1997) will provide a more useful means to evaluate the effectiveness of agents designed to modify the progress of osteoarthritis.

Of the drugs currently available, very few exhibit activities consistant with all of the above criteria. However, pentosan polysulfate fulfills many of these requirements and would appear to qualify as a disease-modifying antiosteoarthritis drug (DMAOD). Details of these pharmacodynamic activities are provided below.

The degree and positions of substitution of the sulphate esters and the ring conformation of pentosan polysulfate have been confirmed by 13^c-NMR spectroscopy (Lentini et al., 1988). Like all polymers, pentosan polysulfate is polydisperse and has a weight mean molecular weight (MW) of 5700 and a number mean molecular weight (Mn) of 4000 as determined by light scattering techniques (Jacobsson et al., 1986). This would indicate that the major population consists of 12 - 16 subunits and therefore n = 1 -1.5 in the structural formula.   

In contrast to other potential DMAODs, such as sodium hyaluronate (Hylartil®, Hylan®, Artz®, Hyalgan), glycosaminoglycan sulfate ester (GAGPS, Arteparon®), chondroitin sulfate (Chondrosulf®), pentosan polysulfate is not derived from animal or bacterial sources and is therefore free of contaminating prions, proteins or phospholipids.   
  
  
   
1. Anti-Inflammatory Activities   
The potent anti-inflammatory effects of pentosan polysulfate (as SP54) has been consistently demonstrated (Kalbhen et al., 1970; Kalbhen & van Heek, 1971; Bohn & Kalbhen, 1971; Kalbhen, 1972, 1973, 1978) using oedemas induced in rat paws by injection of dextran, formaldehyde, trypsin, hyaluronidase, carrageenan or kaolin. For all of these experimentally induced oedemas, a dose-response correlation was observed using pentosan polysulfate within the concentration range of 25 - 100mg/kg when the drug was administered subcutaneously. The optimal time for administration of the drug was found to be when it was given 60 minutes prior to oedema formation (Kalbhen, 1973). Pentosan polysulfate was also effective in reducing inflammation when given 60 minutes after induction of oedema in the rat paw by the injection of kaolin (Kalbhen, 1973). It was concluded ((Kalbhen, 1978) that unlike sodium salicylate, phenylbutazone or indomethacin, pentosan polysulfate was effective against all the types of inflammogens examined. The mechanism of action in this regard was attributed largely to stabilisation of the peripheral vascular system and improvement of the microcirculation in the inflamed tissues.   

According to Walb et al., (1971), pentosan polysulfate possesses marked anticomplementary properties. Using sensitised erythrocytes in vitro, it was found that pentosan polysulfate was ten times more potent than heparin in preventing the lysis by a complement preparation, when used over the concentration range 5.0 - 8.3µg/mL in vitro. Pentosan polysulfate also inhibited the complement C1-esterase in vitro, the ED50 being in the range of 7 - 8µg/mL.   

The anti-complementary effects of pentosan polysulfate have been demonstrated both in vitro and in vivo (Berthoux et al., 1977a, b & c). In vitro studies were conducted with human serum and the drug was effective at microgram levels in rapidly reducing total haemolytic complement activity. In vivo anticomplement activity was obtained with an intravenous dosage of 100mg per 8 hours, i.e. 300mg over 24 hours. In patients with already marked hypocomplementemia, the anticomplementary activity was difficult to establish, but in patients with normal complement activity or moderate hypocomplementemia, the decrease in CH50 levels in serums was marked. The dose response curve was sigmoidal with an ED50 of approximately 0.08 mg/mL in normal human serum. There was, however, no significant change in serum levels of C3, C4 or properdin factor B as determined by radial immunodiffusion techniques. C2, C3, and C4 were totally inhibited and C1 and C5 partially inhibited by pentosan polysulfate. It was concluded that the in vivo anticomplement effects of pentosan polysulfate were sufficiently demonstrated to suggest a decrease liberation of humoral mediators of inflammation during its clinical usage.   
  

2. In Vitro Effects On Chondrocyte Biosynthesis Of Proteoglycans   
When an in vitro model of lapine chondrocyte injury induced by briefly exposing the cells to sodium acetate was used, sodium pentosan polysulfate over the concentration range 0.01 - 0.2 mg/mL was found to improve proteoglycan incorporation into the matrix (Costeseque et al., 1986). The effects of pentosan polysulfate over the concentration range of 0.1 - 100µg/mL, on lapine chondrocyte biosynthesis of proteoglycans has also been examined (Collier & Ghosh, 1989). Arteparon®, diclofenac, indomethacin, ketoprofen and tiaprofenic acid were also evaluated under the same conditions. Cultures were maintained for up to eight days and drug effects monitored every two days. Unlike the NSAIDs examined, pentosan polysulfate and Arteparon® showed a stimulatory effect on proteoglycan biosynthesis at low concentrations (0.1, 1.0µg/mL). Pentosan polysulfate only showed evidence of slight inhibitory activity on chondrocytes at high concentration (50 and 100µg/mL). In contrast, none of the NSAIDs examined showed a stimulatory effect on proteoglycan biosynthesis and diclofenac and indomethacin significantly depressed synthesis at 50 and 100µg/mL. It should be noted that it would be impossible to attain the high concentrations of 50 and 100µg/mL of pentosan polysulfate in cartilage during normal therapeutic administration of this drug. The lower concentrations of 0.1 and 1.0µg/mL could be readily achieved in tissues following intra-articular injection, however (Burkhardt & Ghosh, 1986).   

Chondrocytes derived from bovine metacarpophalangeal joints grown in agarose culture according to the method of Aydelotte and Kuettner (1988) were used to evaluate the ability of pentosan polysulfate, tenidap, and RO31-9790 to stimulate proteoglycan synthesis in the absence and presence of interleukin-1 (Raiss et al., 1995). In the absence of interleukin-1, only pentosan polysulfate stimulated proteoglycan synthesis, whereas in the presence of this cytokine, all of these compounds were active.   

Thrombospondin was identified in 4.0 M GuHCl extracts of human articular cartilage as a pentosan polysulfate binding protein (Ghosh & Hutadilok, 1996). Cultured human fibroblasts synthesise and secrete thrombospondin and incorporate it into extracellular matrix (Jaffe et al., 1983). Furthermore, synovial fibroblasts derived from osteoarthritic joints were shown to secrete thrombospondin and also bind pentosan polysulfate (Ghosh & Hutadilok, 1996). Using bovine erythrocytes conjugated with pentosan polysulfate a rosetting of the synovial fibroblast could be demonstrated. The level of rosetting was not affected by pre-incubating cultures with thrombospondin antibody, suggesting that pentosan polysulfate was interacting directly with the cells.   
  

3. Pentosan Polysulfate: Effects on the Biosynthesis of Matrix Components by Fibroblasts   
Cultures of human synovial fibroblasts derived from rheumatoid and osteoarthritic joints show a concentration dependent stimulation of hyaluronan when exposed to pentosan polysulfate (Hutadilok et al., 1988; Ghosh et al., 1990). In both types of cell line maximum stimulation occurred at drug concentrations of 0.25µg/mL. In the cells from rheumatoid joints at concentrations in excess of 1.0µg/mL, inhibition of hyaluronan synthesis was observed. However, this effect was not produced in the osteoarthritic cell line with concentrations up to 2.0µg/mL.   

An in vivo stimulatory effect of pentosan polysulfate on hyaluronan synthesis was found in fluids of the rat air pouch model treated with this drug (Francis et al., 1993). In these experiments, the drug at 2.5mg/kg was delivered daily for seven days into peptone inflamed air pouches in which cartilage had been implanted. Control animals received an equal volume of sterile saline for seven days into identically treated rat air pouches. Pouch fluids were analysed specifically for hyaluronan content using an enzyme linked immunoabsorbent-inhibition assay (ELISIA) and the results demonstrated a 200 - 300% increase in hyaluronan levels relative to controls.   

The influence of pentosan polysulfate on hyaluronan metabolism of the synovial lining cell was studied in vivo in human volunteers (Verbruggen & Veys, 1992). Significant increases in the mean degree of polymerisation of the hyaluronan chains were observed after a series of four to six intra-articular injections of this drug. No increases in hyaluronan synthesis rates were observed. Repeated administration of the drug did not cause any inflammation or bleeding in the joint cavity. More recently, it was shown that four intraarticular injections of pentosan polysulfate (50mg) significantly increased both molecular weight and the rheological indices (viscosity and elastic stiffness) of the synovial fliud hyaluronan (Adam et al., 1996).   
  

4. Abrogation Of The Suppressive Effects Of Hydrocortisone On Human Synovial Cell Metabolism By Pentosan Polysulfate   
As already discussed, hydrocortisone and other corticosteroids are frequently used in the treatment of arthritic disorders. The effect of hydrocortisone on the synthesis of hyaluronan and DNA by human synovial fibroblasts derived from rheumatoid and osteoarthritic joints was examined by Hutadilok et al. (1988). Inhibition of hyaluronan synthesis was evident at steroid concentrations as low as 10^-8 M, and this effect increased progressively with concentrations up to 10^-4 M compared to non-drug treated controls. DNA synthesis was also inhibited in these cell lines by concentrations of 10^-6 M hydrocortisone. From these in vitro results, it was reasoned that the stimulatory properties of pentosan polysulfate on the synthesis of hyaluronan and DNA in human synovial fibroblasts might be capable of overcoming the inhibitory effects of pentosan polysulfate if the two drugs were used in combination. The effect of combining hydrocortisone (10^-6 M and 10^-8 M) with various concentrations (0.1 - 1.0µg/mL) of pentosan polysulfate on hyaluronan synthesis in vitro by human synovial fibroblasts showed that only partial restoration of synthetic activity was possible when high concentrations of hydrocortisone were used. However, concentrations of pentosan polysulfate as low as 0.1µg/mL were effective in restoring hyaluronan synthesis in the presence of 10^-8 M hydrocortisone. A combination of 5µg/mL hydrocortisone (giving about 40% inhibition of DNA synthesis when incubated with the cells alone) and 10µg/mL of pentosan polysulfate (causing about 60% stimulation alone) restored DNA synthesis to within control levels. From these findings it may be concluded that pentosan polysulfate, when used in combination with corticosteroids for the treatment of inflammatory joint disease, may ameliorate the potentially deleterious effects that these steroids may exert on connective tissue cell metabolism. Subsequent in vivo studies using intra-articular injections of these drug combinations to rabbits supported this view (Kongtawelert et al., 1989).   
  

5. Enzyme Inhibitory (Anticatabolic) Effects Of Pentosan Polysulfate   
Pentosan polysulfate is a potent inhibitor of human neutrophil elastase (HNE) (Kruze et al., 1976a, 1977; Andrews et al., 1983). It acts as a simple hyperbolic, non-competitive inhibitor, the enzyme-inhibitor interaction being electrostatic in nature. and dissociated by 0.3M NaCl (Baici et al., 1981). It had been previously proposed that pentosan polysulfate interacted with the natural substrates of HNE to form complexes which were then resistant to degradation by the proteinase (Barg et al., 1979). This hypothesis was tested (Andrews et al., 1983) using an anion exchange competitive binding assay which demonstrated that the binding of pentosan polysulfate to HNE was very much greater than to gelatin or proteoglycans, which were good substrates for the enzyme (Lentini et al., 1988). These findings support the concept of a specific electrostatic interaction between pentosan polysulfate and HNE (Baici et al., 1981). An IC50 (inhibitor concentration at which 50% inhibition occurs) of 10^-8 M may be calculated for pentosan polysulfate when the concentration of HNE used was 6.5 x 10^-9 M and excess of the synthetic substrate succinyl-alanyl-alanyl-valyl-nitroanilide (SAAVN) was used (Andrews et al., 1983). This value is in good agreement with the results by Baici et al. (1981) who reported an IC50 of 15 x 10^-8 M for the drug at a measured enzyme concentration of 131 x 10^-9 M. Pentosan polysulfate also strongly inhibits PMN-derived cathepsin G (Steinmeyer & Kalbhen, 1991).   

Pentosan polysulfate has demonstrated a concentration dependent inhibition of testicular and aortic wall derived hyaluronidases in vitro (Buddecke & Platt, 1965; Mathies & Willms, 1965). This activity was found to be more potent than that of heparin or dermatan sulphate. Concentrations of 0.5 - 5µg/mL of the drug produced suppression of activity ranging from 65% to 95% (Kalbhen et al., 1970). Kinetic studies showed that this inhibition was of the competitive type. Other lysosomal enzymes inhibited by pentosan polysulfate include chondroitin-4-sulphatase and N-acetyl-glucosaminidase (Buddecke & Werries, 1965). Pentosan polysulfate as well as other drugs also inhibited purified cathepsin B1 (Kruze et al., 1976b). Using a synthetic substrate it was found NSAIDs such as diclofenac and indomethacin inhibited this lysosomal enzyme at concentrations of 10^-5 M, whereas pentosan polysulfate caused 72% inhibition at 10^-6 M. Significantly, sodium salicylate only showed inhibition of this enzyme at concentrations above 10^-2 M.   

Pentosan polysulfate produced concentration-dependent inhibition of the proteoglycan-degrading activity of both human fibroblast stromelysin (MMP-3) and rat tumour stromelysin (Nethery et al., 1992). Rat and human stromelysin activities were inhibited at drug concentrations above 0.005 microgram/mL. Fifty percent inhibition of rat stromelysin was produced by concentrations in the 0.5-5 microgram/mL range. The pattern of inhibition of human stromelysin was similar, except that drug concentrations in the 500-5000 micrograms/mL range produced 50% inhibition.   
  

6.Effects Of Pentosan Polysulfate In Animal Models Of Arthritis   

The atrophy arthritis model induced in the rabbit by immobilisation   

The atrophy model of arthritis in the rabbit (Langenskiold et al., 1979) has been used to evaluate the effects of pentosan polysulfate at two different dose levels on cartilage degradation (Golding & Ghosh, 1983). It was found that immobilisation of the knee joint of the rabbit in full extension for a period of four weeks produced significant changes in the composition of the joint articular cartilage. Both the proteoglycan content and the extractability were reduced significantly (p < 0.05) relative to control values. However, non-extractable proteoglycans and collagen levels in the residues were not significantly altered by joint immobilisation. The contralateral (free) joint of the immobilised group also showed a significant depression (p < 0.05) in proteoglycan levels of articular cartilage relative to the non-immobilised control group. All other parameters measured for these tissues were not statistically different to control values.   

The depletion of proteoglycans from articular cartilage of immobilised rabbit joints was prevented by intramuscular (i.m.) administration of pentosan polysulfate at 10mg/kg body weight every 48 hours over the 4-week immobilisation period. Similar results were obtained on analysis of the contralateral joints of drug-treated animals. When the concentration of pentosan polysulfate administered was reduced to 5mg/kg i.m. over the same period, the levels of proteoglycans and their extractability from joint cartilages was indistinguishable from the non-treated immobilised group.   

Proteoglycans mobilised from normal adult rabbit articular cartilage by 4M GuHCl were demonstrated by direct Sepharose CL-2B chromatography, in the presence of excess HA, to contain approximately 28% non-aggregating population. An identical result was obtained after isolation of the proteoglycan subunit by dissociative density gradient ultracentrifugation, followed by chromatography on Sepharose CL-2B using associative conditions.   

After 4-week joint immobilisation, the proportion of non-aggregatable proteoglycans extracted from articular cartilage increased to 60% of the total present. However, the proportion of non-aggregatable proteoglycans extracted from the corresponding tissues of the contralateral joints remained unchanged. Examination of extracts of articular cartilage from immobilised joints of drug-treated animals showed that the proportion of the non-aggregatable species present was comparable to controls (Golding & Ghosh, 1983).   

Sodium iodo-acetate induced degeneration in knee joint articular cartilage   

Sodium iodo-acetate induced arthritis in chicken knee joints was used to assess the disease-modifying effects of pentosan polysulfate (Kalbhen & Blum, 1977). Cartilage degeneration was assessed radiologically and histologically. It was found that weekly i.a. injections of 1 - 5µg of pentosan polysulfate over a six week experimental period prevented the breakdown of cartilage. Pentosan polysulfate when administered i.a. or systemically to these animals did not produce deleterious effects on joint cartilage. This contrasted with intra-articular injections of 1 - 5µg NSAIDs such as phenylbutazone and indomethacin, which caused significant damage.   

Lapine model of osteoarthritis induced by meniscectomy and co-lateral ligament severance   

The Moskowitz et al. (1979) rabbit osteoarthritis model was used to evaluate pentosan polysulfate and Arteparon® (GAGPE) at 1mg/kg (Muniz et al., 1986; Howell et al., 1987). The drugs were administered intra-articularly twice weekly to groups of rabbits in which medial partial meniscectomy had been performed; an equivalent dose of heparin was used as control. One week post meniscectomy, animals in the experimental groups were placed on these agents for 11 weeks. The animals were then killed, knee joints opened and femoral and tibial condyles processed for histological examination. Sections were then evaluated with respect to the severity of histological lesions in cartilages of the treated groups versus saline controls. The mean histological Mankin scale grade was 10.33 ± 1.21 (moderately severe) in the saline (positive) controls. A highly significant reduction in the grading scale was observed in the animals who received GAGPE or pentosan polysulfate (to 2.1 ± 1.2 and 3.2 ± 3.25 respectively). This reduction in cartilage damage was highly significant (p < 0.001). In contrast, heparin caused an insignificant decline in scale reduction to 8.67 ± 1.86 (p > 0.1). The aggregational properties of proteoglycans extracted from the cartilages of meniscectomised controls who received saline were poor, as assessed by ultracentrifugation levels. It was concluded that pentosan polysulfate had potential as a disease-modifying agent as assessed by this model of arthritis (Howell et al., 1987).   

Rat subcutaneous air pouch model   

The subcutaneous air pouch-cartilage model used in laboratory rodents has been employed by several groups to study the effects of NSAIDs and other drugs on inflammatory cell-cartilage interactions (Sedgwick et al., 1985; Bottomley et al., 1988). Rabbit articular cartilage was implanted into seven-day-old subcutaneous air pouches established beneath the dorsal facia of mature rats (Francis et al., 1989). The injection of a 3% (w/v) peptone solution into the pouches stimulated an acute inflammatory reaction with the influx of large numbers of PMN cells and the concomitant degradation of the cartilage implant. In this model it was shown that pentosan polysulfate at 10mg/kg was able to retard the loss of proteoglycans from articular cartilage (Francis et al., 1989).   

More recently, the effects of calcium pentosan polysulfate (CaPPS) on edema production, cellular infiltration and activity of inflammatory mediators in the rat air pouch were reported (Bansal et al., 1993). CaPPS or saline was injected subcutaneously daily for 7 days at 0.5 to 10 mg/kg into the abdomen of rats with preformed noninflamed or peptone-inflamed dorsal pouches. After sacrifice the next day, pouch fluid volume, white and differential cell counts, prostaglandin E2, TNF-alpha, IL-6 and IL-1ß levels were determined. In saline-treated control animals, injection of peptone into rat air pouches provoked a large influx of fluid and white cells, with a concomitant increase in levels of PGE2, TNF-alpha, IL-6 and IL-1ß. Although CaPPS treatment had no effect on fluid, white cell numbers or PGE2 levels, there was a 38% reduction in TNF-alpha activity after 2.5 to 10mg/kg drug treatment. IL-6 activity increased significantly over the same range, indicating that CaPPS could elicit specific effects on cytokine production by leucocytes and synovial cells.   

Protective effect of pentosan polysulfate in hydrocortisone-induced cartilage degradation in the rabbit   

Intra-articular administration of hydrocortisone succinate once a week for eight weeks into joints of mature New Zealand rabbits induced the loss of proteoglycan from articular cartilage (Oegema & Behrens, 1981). This observation was used to examine the disease-modifying activity of pentosan polysulfate (Kongtawelert et al., 1989). Hydrocortisone succinate (25mg) was administered intra-articularly to joints of a group of four mature New Zealand white rabbits for eight weeks. Joints of other groups of animals were injected for eight weeks with a solution containing the same amount of hydrocortisone but mixed with 5mg pentosan polysulfate, pentosan polysulfate (5.0 mg) alone, or with saline. The pentosan polysulfate/hydrocortisone combination reduced the loss of hexuronate (which is a marker for proteoglycan) from joint articular cartilage by maintaining tissue levels of hyaluronic acid which is necessary for aggregate formation. Some support for this conclusion was provided by monitoring serum keratan sulphate (KS) levels. In the HC treated animals, serum KS levels were elevated, but in joints treated with the HC/pentosan polysulfate combination KS values were held within the saline control range.   

It was concluded from these results that pentosan polysulfate possessed the ability to abrogate the suppressive effects of hydrocortisone on cartilage hyaluronic acid biosynthesis when administered intra-articularly in combination with the corticosteroid (Kongtawelert et al., 1989).   

Protective effect of pentosan polysulfate on cartilage degradation by anterior cruciate ligament transection in the dog   

The potential therapeutic effects of insulin-like growth factor-1 (IGF-1) and sodium pentosan polysulfate were evaluated in an anterior cruciate ligament-deficient canine model of osteoarthritis (Rogachefsky et al., 1993; 1994). A control group of animals received no treatment or surgery (N). Four other groups of animals received anterior cruciate transection and either no treatment (osteoarthritic), intra-articular IGF-1 (IGF-1), intra-muscular pentosan polysulfate (PPS), or a combination of intra-articular IGF-1 and intra-muscular pentosan polysulfate (IGF-1/PPS). All therapy was begun 3 weeks after surgery and continued for 3 weeks. At 6 weeks, articular cartilage from the femoral condyle was evaluated for anatomy, histology (Mankin grade) and biochemistry. Anatomically, only cartilage from dogs in the IGF-1/PPS group approximated that found in N. Mankin scores indicated less severe disease in both PPS and IGF-1/PPS groups compared with the osteoarthritis group. Consistent with histology, the level of active neutral metalloproteinase was lower in cartilage from the PPS group compared with the osteoarthritis group. Active and total neutral metalloproteinase, tissue inhibitor of metalloproteinases (TIMP), total collagenase, uronate and hydroxyproline contents were all near normal in the IGF-1/PPS group. Thus, in this model of mild osteoarthritis, therapeutic intervention with IGF-1 and PPS appeared to successfully maintain cartilage structure and biochemistry.   

Effect of pentosan polysulfate in an ovine meniscectomy model of osteoarthritis   

Medial meniscectomy was undertaken in adult merino sheep and after 16 weeks exercise each group was administered five weekly intra-articular injections of saline, pentosan polysulphate, hyaluronan or a combination of the two drugs (Ghosh et al., 1993). Gait analysis and x-rays were undertaken before and after drug treatment. At sacrifice (26-weeks), joints were examined for gross pathological and histochemical changes. Only the pentosan polysulphate-treated group showed an improvement in gait, with low radiological and histology scores.   

Effect of pentosan polysulfate in a polycation-induced inflammatory arthritis in the rabbit   

The antiinflammatory and cartilage-protecting activities of orally administered calcium pentosan polysulfate (CaPPS) were studied in a rabbit model of inflammatory arthritis (Smith et al., 1994). A single intraarticular injection of a preformed polycation complex (PC) of poly-D-lysine and hyaluronan was used to induce joint inflammation; saline was injected into the contralateral joint as a control. Animals were killed 1, 4, 7, or 10 days post-PC injection. CaPPS, at 5 mg/kg, 10 mg/kg, or 75 mg/kg, was given every 48 hours commencing 7 days prior to PC injection. Serum interleukin-6 (IL-6), synovial fluid prostaglandin E2, cell numbers, and cartilage proteoglycan content, composition, and biosynthesis were determined for PC- and saline-injected joints. In PC-injected, non-drug-treated animals, serum IL-6 activity, synovial fluid leucocyte numbers, and prostaglandin E2 levels were elevated, while cartilage proteoglycan content and biosynthesis were reduced. CaPPS at 10 mg/kg, but not at 5 mg/kg, decreased serum IL-6 levels but maintained cartilage proteoglycan concentration and biosynthesis. However, synovial fluid leukocyte counts and prostaglandin E2 levels (except on day 1) were not reduced. The ability of CaPPS to attenuate serum IL-6 levels and preserve cartilage proteoglycans in inflamed rabbit joints suggests that this substance could be of value as an effective orally administered drug.   
  

7.The Effect of Pentosan Polysulfate on Platelet Aggregation, Coagulation and Fibrinolysis   
Pentosan polysulfate has a profound effect on coagulation and fibrinolysis and, to a lesser degree, platelet aggregation. Most of the investigations into the mechanism by which this drug acts have been carried out are summarised in this review, however, it is now generally accepted that the overall action of pentosan polysulfate in vivo is to decrease coagulation and increase fibrinolysis. Platelet aggregation is said to be affected but the dosage and method of administration appear to be important in determining the intensity of the effect by the drug and there are conflicting reports on whether pentosan polysulfate increases or decreases platelet aggregation in vivo.  

The accompanying figure is a summary of the known pathways of blood coagulation and fibrinolysis. It is now apparent that the distinction made earlier between the intrinsic pathway (i.e. those reactions initiated by the contact of blood with a foreign surface) and the extrinsic pathway (i.e. the reactions initiated by the exposure of blood to injured tissues) has become less distinct. Many feedback mechanisms and interactions outside the classic schema have yet to be identified. Thus attempts to elucidate the mechanism of a drug such as pentosan polysulfate on the whole system is complicated by the inadequacies of our knowledge at the present time. In the diagram, the pathways which are known to be inhibited by pentosan polysulfate are shown as a solid bars, while pathways which are stimulated by pentosan polysulfate are marked with a star. The letters adjacent to these symbols refer to the sections below where these activities are discussed in detail.

A) Platelet aggregation  

The mechanisms of platelet adhesion and aggregation are complex and have been reviewed (Fuster et al., 1987). Platelet aggregation is a critical step for normal haemostasis. Platelet adhesion takes place when the normal endothelial cell layer is disrupted or disturbed. Additional platelets are recruited by degranulation of the primary platelets and a haemostatic plug is formed at the site of injury. A variety of compounds, such as adenosine diphosphate (ADP), thrombin or collagen can then stimulate the platelets to alter their shape and initiate a reversible aggregation. A fibrinogen receptor is then induced on the cell membrane and fibrinogen molecules can then bind platelets together. At this stage, platelets then release alpha-granule adhesive glycoproteins, such as thrombospondin, von Willebrand factor, fibrinogen and fibronectin. These proteins assemble on the platelet membrane and bring about a secondary phase of aggregation which is irreversible. At this stage, the clotting process may be initiated with the subsequent formation of thrombin, which further promotes platelet aggregation and results in the formation and polymerisation of fibrin.  

Aznar et al. (1967) reported that pentosan polysulfate exhibited significant inhibitory effect on bovine platelet aggregation ex vivo and shortly after Bicher (1970) showed that this drug caused a general decrease in platelet adhesiveness after in vivo administration. Aznar & Albenola (1970) confirmed this result in 104 normal and atherosclerotic individuals. A subsequent study by Frandoli et al. (1972) demonstrated under double-blind conditions, decreased platelet aggregation when pentosan polysulfate was administered both orally and parenterally to a group of thrombophilic patients. The experiments of Aznar et al. (1967) were subsequently repeated by Kindness et al. (1979) who found that pentosan polysulfate enhanced platelet aggregation ex vivo rather than inhibiting aggregation. These workers also found that this drug inhibited thrombin-induced platelet aggregation. Pentosan polysulfate increased the maximum collagen-induced platelet aggregation in human plasma ex vivo but had no effect on the lag-time of aggregation or the shape change of the platelets so treated (Vinazzer et al., 1980). Later studies (Messmore et al., 1989) found that pentosan polysulfate did show slight inhibitory activity against collagen aggregation and adhesion in vitro, and also interacted with the antibody induced by heparin therapy. Pentosan polysulfate also significantly increased the binding of fibrinogen to ADP-treated platelets in vitro (Dunn et al. 1983). A slight decrease in platelet numbers after a 3-day continuous infusion of pentosan polysulfate (4mg/kg body weight/24 hour) was observed in ten subjects (De Prost et al., 1985). Sie et al. (1986a) also reported a reduction of platelet numbers after administration of pentosan polysulfate either intramuscularly or subcutaneously to 30 volunteers for ten days.  

It appears therefore, that pentosan polysulfate after initial administration to man, induces a transient reduction in platelet count, probably due to some spontaneous aggregation of these cells. However, the platelet numbers recover with time and may even increase with further administration of the drug. From the published studies, pentosan polysulfate appears to have varying effects on platelet aggregation in isolated systems. Furthermore, results of these in vitro experiments appear to be highly dependent on the methods used, and do not always agree with in vivo observations. This conclusion may be explained by the many feedback mechanisms of haemostasis which are influenced by pentosan polysulfate. These feedback pathways which are operative in vivo may not be observable in ex vivo situations.  

B) Inhibition of Factor Xa  

Factor Xa is generated from Factor X by the action of Factor IXa complexed with activated Factor VIII, calcium ions and phospholipids (the Tenase complex, see Section D) in the intrinsic system. In the extrinsic pathway, Factor Xa is derived from activated Factor VII (also membrane bound) in the presence of calcium ions via the extrinsic pathway. Once Factor X is activated, it also is membrane bound and acts, in the presence of calcium ions and activated Factor V, on prothrombin, (see Figure). This complex is now often referred to as prothrombinase. It should be noted that both in purified systems and in plasma, Factor Xa is protected from the action of anti-thrombin III (ATIII) (and hence from the inhibitory action of heparin) by the phospholipids and Factor Va (Marciniak, 1973; Walker & Esmon, 1979; Ellis et al., 1984; Ofosu et al., 1984; Lindhout et al., 1986).  

Soria et al. (1980) were amongst the first to demonstrate that pentosan polysulfate had strong anti-Xa activity. This activity was detected both in vitro and in vivo. However, it was observed that higher concentrations of this drug were needed in vitro to inhibit Factor Xa than were required in vivo. The inhibitory activity of pentosan polysulfate was not demonstrated in the absence of ATIII and hence these workers concluded that the inhibition was ATIII-dependent. Similar findings were reported by Vinazzer et al. (1980) who showed strong inhibition of Factor Xa in the presence of ATIII in the same system and some inhibition of Factor Xa in plasma from patients treated with pentosan polysulfate. Pentosan polysulfate potentiated ATIII inhibition of Factor Xa in vitro but did not cause inhibition by binding strongly to ATIII (Czapek et al. 1980). Pentosan polysulfate thus exhibits a different mechanism of inhibition to that of heparin, which is recognised to bind strongly to ATIII. The effects of pentosan polysulfate on blood coagulation when administered subcutaneously to man demonstrated in vivo potentiation of Factor Xa inhibitory activity by this drug (Ryde et al., 1981). Pentosan polysulfate also produced a smaller interindividual response than did heparin, which was examined in the same study. From these results, it was concluded that the major effect of pentosan polysulfate on blood coagulation was at this site (see Section C). In contrast, Fischer et al. (1982a) could not demonstrate an effect of pentosan polysulfate on anti-Factor Xa activity after subcutaneous injection into man in either ex vivo or in vivo tests. Not only did pentosan polysulfate potentiate in vitro inhibition of Factor Xa by purified ATIII, but that prothrombin activation by Factor Xa and phospholipid was also suppressed by the drug in the absence of ATIII (Fischer et al., 1982b). Crossed immunoelectrophoretic data indicated that the ATIII-dependent effects of pentosan polysulfate were mediated by its binding to the enzyme (Factor Xa) rather than to ATIII, as occurs with heparin. The affinity of pentosan polysulfate for ATIII was weak but could still increase the inhibition of Factor Xa 240-fold (as compared to 4800-fold for ATIII-dependent inhibition of Factor Xa by heparin) (Scully & Kakkar, 1984). The apparent dissociation constant of pentosan polysulfate for human Factor Xa was 20µM. Ofosu et al. (1985) demonstrated that pentosan polysulfate was a potent inhibitor of thrombin generation but not a strong accelerator of Factor Xa inactivation and that pentosan polysulfate had both direct and indirect effects on the inhibition of Factor Xa (Ofosu et al., 1986). Not only did the drug directly inhibit Factor Xa enzyme activity, but it also disrupted the interaction of the constituents of the prothrombinase complex, as well as prothrombin, with coagulant surfaces. From these findings, they concluded that inhibition of the catalytic activity of the prothrombinase complex was one of the primary mechanisms by which sulphated polysaccharides, such as pentosan polysulfate act as anticoagulants.  

Further in vitro work (Sie et al., 1986b) provided evidence that the main anticoagulant effects of pentosan polysulfate in "normal" plasma were both ATIII and heparin co-factor II (HCII) independent. This would suggest that pentosan polysulfate acts by directly preventing thrombin generation by Factor Xa rather than by enhancing inhibition of thrombin. In summary, although earlier work suggested that pentosan polysulfate acted against factor Xa in a similar manner to the way in which heparin acts against thrombin, i.e. by potentiating the inhibition by ATIII, later studies showed that pentosan polysulfate interferes with the binding of factor Xa to the enzyme via an ATIII independent mechanism. At the present time, this process is considered to be one of the primary sites where pentosan polysulfate can influence the blood coagulation system but its activity in this regard is less potent than heparin.  

C) Inhibition of Thrombin  

Thrombin (or Factor IIa) is formed from prothrombin (Factor II) by the action of Factor Xa, Factor Va and calcium ions in the presence of phospholipids (the prothrombinase complex). Once formed, thrombin catalyses a number of reactions including the conversion of fibrinogen to fibrin and the activation of Factors V, VII, VIII, XII and XIII. Thus inhibition of thrombin will not only prevent the direct formation of fibrin but will also cause feedback inhibitions to earlier steps in the coagulation cascade.  

The anti-thrombin effect of pentosan polysulfate was first measured as negligible, being only 4% of the effect of an equivalent amount of commercial beef-lung heparin (Vinazzer et al., 1980). However, Soria et al. (1980) demonstrated that pentosan polysulfate inhibited thrombin in vitro and that this inhibition was ATIII dependent. In vivo, however, the effect of this drug appeared to be partially ATIII independent, so it was clear that more than one mechanism was operative. Further evidence that thrombin inhibition by ATIII was potentiated by pentosan polysulfate was provided by Czapek et al. (1980). In this study, heparin was found to be ten times more potent as an ATIII accelerator than pentosan polysulfate on a weight for weight basis. A potentiation of inhibition of thrombin by pentosan polysulfate occurred in the presence of ATIII and it bound to thrombin as well as factor Xa (Fischer et al., 1982a&b). However, Scully et al. (1983) reported that pentosan polysulfate at high concentrations (> 40µg/mL) acted by directly inhibiting thrombin without the intervention of ATIII. These workers then demonstrated that the affinity of pentosan polysulfate for ATIII was weak but could increase the inhibition of thrombin 40-fold, as compared to a 50,000-fold increase with ATIII dependent inhibition of thrombin by heparin (Scully & Kakkar, 1984a). In the presence of pentosan polysulfate, the dissociation constant for the initial complex of ATIII and thrombin was shown to be reduced from approximately 2mM to 61µM without any change in the limiting rate constant. Though this acceleration of ATIII inhibition was much lower than that achieved by heparin, it approached physiologically significant levels. Heparin and pentosan polysulfate were further investigated for their abilities (a) to bind antithrombin, (b) to induce conformational change of the inhibitor and (c) to potentiate antithrombin inhibition of thrombin (Sun & Chang, 1989). The binding capacity was characterized by the extent to which a polysaccharide could protect chemical modification of Lys-125 and Lys-136, two lysyl residues of antithrombin which have been implicated in heparin binding. Pentosan polysulfate protected only Lys-125 and caused no appreciable conformational change, although it was also capable of enhancing the antithrombin-thrombin interaction. These data clearly demonstrated that the heparin and pentosan polysulfate binding sites of antithrombin overlap (at Lys-125) but are not identical.  

A breakthrough then occurred after it was realised that the pentosan polysulfate-dependent inhibitory activity co-eluted with heparin co-factor II (HCII) on gel filtration of normal plasma (Scully & Kakkar, 1984b). A maximal second order rate constant of 2.5 x 10⁸ M^-1 min^-1 (with apparent Kd of 1.8µM) was found for the potentiation of the thrombin-HCII interaction by pentosan polysulfate. In contrast, Sie et al. (1985) found that the effect of this drug on coagulation was both ATIII and HCII independent.  

Although the effect of pentosan polysulfate on the inhibition of thrombin by ATIII was found to be negligible, the drug strongly prevented heparin potentiation of ATIII dependent inhibition of thrombin by binding to the thrombin molecule itself (Pletcher et al., 1985). Pentosan polysulfate has a greater affinity for the thrombin molecule than heparin, as evidenced by its ability to displace thrombin from a heparin-Sepharose affinity column more efficiently than heparin.  

Ofosu et al. (1986) confirmed the earlier result of Scully and Kakkar (1984b) that pentosan polysulfate catalyses thrombin inhibition by HCII, and poorly inhibits the interaction of thrombin and ATIII. Significantly, on fractionating pentosan polysulfate, species of differing molecular weight were found which exhibited varying thrombin inhibitory activities (Scully & Kakkar, 1984b). There was a greater electrostatic interaction between pentosan polysulfate and HCII than between pentosan polysulfate and ATIII, or between HCII and heparin or dermatan sulphate. It was concluded that the control of coagulation was principally mediated through inhibition of thrombin and 80% of this was due to the action of HCII. However, further studies (Simmons et al., 1995) revealed that this statement was probably oversimplifying the true in vivo situation.  

[125I]-Labeled thrombin was incubated with human plasma and its interactions with AT-III or HC-II were investigated in the presence of sodium pentosan polysulfate (Simmons et al., 1995). The complexation of thrombin with HC-II or with both AT-III and HC-II were enhanced depending upon the concentration and the duration of the interactions of pentosan polysulfate with plasma. Within a 10 second interaction, thrombin-HCII was enhanced, but after 30 seconds both thrombin complexations were enhanced in the presence of sodium pentosan polysulfate.  

D) Inhibition of Factor IXa  

Factor IXa is generated from Factor IX by the action of Factor XIa and calcium ions. The activated Factor VII complex can also generate Factor IXa. Factor IXa forms a complex with Factor VIIIa (generated by thrombin), calcium ions and phospholipid and is called the "Tenase" complex due to its ability to activate Factor X.  

Although relatively high concentrations of pentosan polysulfate were required for the inhibition of thrombin and factor Xa in plasma, at concentrations less than 2µg/mL this drug markedly suppressed the intrinsic activation of Factor X (Fischer et al. 1982a). This effect was shown to be independent of ATIII and was due largely to the inhibition of Factor IXa. Further studies (Fischer et al. 1982b) showed that pentosan polysulfate impaired the generation of Factor Xa when administered subcutaneously in man. It is possible that the impairment of Factor Xa generation was due to the pentosan polysulfate inhibition of thrombin thereby preventing the activation of factor VIII; a vital component of the "Tenase" complex. A later study (Ofosu et al, 1986) also found that pentosan polysulfate directly inhibited the formation of Factor Xa. E) Inhibition of the Activation of Factor V Factor V is activated by thrombin to become part of the "Tenase" complex. When 25 - 100mg of pentosan polysulfate was administered intravenously to patients, it extended the prothrombin time with a decrease in Factor V levels (Sampol et al., 1983). Ofosu et al. (1986) confirmed this observation and also reported reduced Factor V activation by thrombin in the presence of this drug. These workers later concluded that the anti-coagulant effect of pentosan polysulfate was mediated primarily through its ability to inhibit the thrombin-dependent activation of Factor V, thereby inhibiting the formation of the prothrombinase complex, the physiological activator of prothrombin (Ofosu et al., 1987).  

F) Inhibition of Factor VIII Activation  

Factor VIII is activated by thrombin to become part of the "Tenase" complex. Pentosan polysulfate has been reported to prevent Factor VIII activation by thrombin independently of HCII (Wagenvoord et al., 1985). This suggestion was at variance with the most recent reports on HCII-dependent thrombin inhibition (see Section C). It is highly likely, however, that Factor VIII activation is impaired by pentosan polysulfate due to its action on thrombin. It has been shown that the activation of factor V and factor VIII by thrombin are necessary for the efficient formation of the prothrombinase complex (Ofosu et al., 1986).   

G) Stimulation of Fibrin Polymerisation   

Once thrombin has cleaved fibrinogen to fibrin monomers, spontaneous polymerisation to soluble fibrin occurs. An unpublished observation described the potentiation of fibrin polymerisation by pentosan polysulfate at concentrations as low as 5µg/mL (Scully & Kakkar, 1983). This observation has yet to be substantiated.  

H) Inhibition of Plasmin  

Plasmin degrades both fibrinogen and insoluble fibrin to soluble products. It is thus the ultimate reaction in the fibrinolytic process. Pentosan polysulfate is reported to enhance the inhibition of plasmin by ATIII (Czapek et al., 1980). However, as this only occurs at high concentrations (ca. 400µg/mL) it is unlikely that this inhibition is of any physiological significance during normal administration of the drug.  

I) Stimulation of Fibrinolysis  

Fibrinolysis is the series of events and reactions which culminates in the degradation of fibrin. There are numerous pathways for the activation of plasminogen to plasmin, the enzyme responsible for the degradation of fibrin. The ability of pentosan polysulfate to stimulate fibrinolysis is well documented (Halse, 1962; Frandoli et al., 1972; Kostering et al., 1973; Vinazzer et al., 1982; Gaffney & Marsh, 1986; Sie et al., 1986a), however, pentosan polysulfate has also been reported to have no effect on fibrinolysis in healthy volunteers (Bergqvist and Nilsson, 1981; de Prost et al., 1985).  

The shortened euglobulin clot lysis times (ECLT) induced by pentosan polysulfate may reflect the release of plasminogen activator from vascular endothelium (Marsh & Gaffney, 1980). Shorter ECLTs were also observed after subcutaneous injection in man (Fischer et al., 1982a&b). Pentosan polysulfate also induced the release of lipoprotein lipase to a greater extent than after heparin administration, suggesting that pentosan polysulfate had a marked effect on blood vessel walls. It was subsequently shown that pentosan polysulfate increased the cell-associated plasminogen activator activity in dividing bovine endothelial cells (Delvos et al., 1985). After twelve months oral administration (300mg/day) to patients, the enhancement of fibrinolysis by pentosan polysulfate was not diminished and pentosan polysulfate appears to act in vitro on fibrinolysis via Factor XII and prekallikrein as well as increasing the availability of tissue plasminogen activator (Vinazzer, 1986). Pentosan polysulfate in patients with a diminished fibrinolytic capacity resulted in a 60% reduction of thromboembolic manifestations (Vinazzer, 1989).  

Pentosan polysulfate was demonstrated to release plasminogen activator from the isolated pig ear and rat lung preparation (Klocking & Markwardt, 1985; 1986). In isolated rabbit ear and whole-body studies in rabbits, the plasminogen activator was demonstrated to be of the tissue-type. The regulation of tissue-plasminogen activator (tPA) and plasminogen activator inhibitor 1 (PAI-1) synthesis was studied in cultured human mesothelial cells derived from omentum (Grulich-Henn et al., 1990). Pentosan polysulfate (300 µgrams/mL) stimulated tPA synthesis by these cells 2.9 - 4.5-fold but had no effect on PAI-1 synthesis. Klocking (1992) showed that calcium pentosan polysulfate when given orally to rats at 10mg/kg released more tPA into the plasma than sodium pentosan polysulfate given at the same dose. However, after both intramuscular and subcutaneous administration to human volunteers, pentosan polysulfate was reported to enhance fibrinolytic activity without a concomitant increase in antigen levels to tPA (Sie et al., 1986a). These pharmacological effects were at their maximum between two and four hours after drug administration.  

From the above studies, it would appear that pentosan polysulfate exhibits its anticoagulant effects primary by catalysing the formation of the thrombin-heparin cofactor II complex, whereas heparin acts primarily by catalysing the formation of the thrombin-antithrombin III and thrombin-heparin cofactor II complexes. Pentosan polysulfate also shows much weaker effects than heparin on prothrombin activation and catalysis of Factor Xa inhibition by ATIII. The net result is that in vivo, heparin is approximately six times more potent as an anticoagulant than pentosan polysulfate but the antithrombotic activity of the drug is almost equivalent to that of heparin.  

Pentosan polysulfate is thus an extremely potent activator of the fibrinolytic system, thereby efficiently lysing vascular thrombi. The fibrinolytic activity of pentosan polysulfate appears to be achieved by an endogenous pathway as well as by the stimulation of the release of tissue type plasminogen activator from vascular endothelium. The weak anticoagulant but strong fibrinolytic effects of pentosan polysulfate would facilitate the clearing of vascular occlusions in all blood vessels perfused by the drug. By this means, thrombotic emboli deposited in the subchondral vasculature and synovium of arthritic joints should be mobilised thereby redressing the imbalance in blood supply to cells of these tissues.  
  

8. Antilipidæmic Effects  
The effects of pentosan polysulfate on fat metabolism have been extensively characterised with particular respect to the increase of the clearing factor and the reduction in blood total lipids, triglycerides, cholesterol and lipoprotein ratio. The effect of the drug on these systems has been observed during and after cessation of treatment. These data have been obtained both in experimentally induced hyperlipemias and hypercholesterolemias as well as in human patients with atherosclerotic hyper- and dyslipidemias (Paramelle, 1962; Wolff, 1962; Schwartz et al., 1962; Schon & Sauer, 1963; Zeller et al., 1964; Brunaud et al., 1967; Tinti et al., 1967; Raveux et al., 1967).  

The clearing effect of pentosan polysulfate and heparin on plasma lipids has been examined in rats (Brunaud et al., 1967). Groups of 10 animals were used and blood collected from the aorta 15 minutes after the intravenous administration of 0.25, 0.5, 1.0, 3.0 and 5.0mg/kg heparin or pentosan polysulfate. With both drugs, nearly optimal results were obtained in the low dose group. The maximal effect produced by administration of 1 - 2 mg/kg could not be improved by using higher doses. Some minutes after the pentosan polysulfate injection, clearing activity was already increased. Within 1 hour, the effect reached optimal values, and decreased progressively during the following 3 hours. Heparin, given at dose to achieve full activity, also caused an inhibition of the coagulation or even complete acoagulemia, while equivalent doses of pentosan polysulfate did not alter blood clotting. In the dog, clearing activity was measured prior to the injections and 15 minutes, 30 minutes, 1 hour, 2 hours and 6 hours after intravenous (i.v.) or i.m. administration. Equivalent i.m. doses of pentosan polysulfate gave effects superior to those of heparin with regard to the degree and duration of activity. In rats, 10mg/kg heparin or 6mg/kg of pentosan polysulfate diminished by about 50% of hypercholesterolemia caused by 300mg/kg Triton WR 1339 administration. These workers also fed rats a high-fat and high-protein diet over 7 months that produced increased plasma levels of total fats and a change in lipoprotein pattern in favour of the B-fractions. Heparin or pentosan polysulfate was continuously administered to these animals while the fat diet was maintained. A group of 10 animals was given 5mL saline intraperitoneally (ip; control animals); group 1 was given 5mg/kg heparin i.p. daily for three weeks; group 2 was given 5mg/kg of pentosan polysulfate i.p. daily for three weeks; group 3 was given 3mg/kg of pentosan polysulfate i.p. daily for three weeks and group 4 was given 5mL saline i.p. Pentosan polysulfate as well as heparin caused a regression of the blood changes induced by the 7 month high fat diet. A three week therapy with pentosan polysulfate almost normalised the total lipids and the lipoprotein pattern, while control animals remained unchanged. The effects of heparin and pentosan polysulfate were qualitatively equal but pentosan polysulfate developed more pronounced activity. "Under these test conditions, pentosan polysulfate was clearly superior to heparin” (Brunaud et al., 1967). Additionally the disintegration of histologically and chemically detectable deposits of cholesterol and neutral fats in the organs was favoured by pentosan polysulfate to a higher degree than by heparin. After a treatment with pentosan polysulfate the cholesterol level of the liver was analysed chemically and proved to be lower than that in the liver of animals treated with heparin. This observation coincided with the histological findings. The tissues of the animals treated with pentosan polysulfate contained less lipids as identified by their double refraction and colorimetric assessment.  

The effect of pentosan polysulfate on altered tissue lipid levels, was studied in hyperoxaluric condition during glycollate feeding in rats (Subha & Varalakshmi, 1993). Elevated cholesterol with reduced phospholipid levels in both liver and kidney tissues, were the significant observations in the experimental animals. In addition, total lipids were increased in the kidney. Administration of pentosan polysulfate to hyperoxaluric rats reduced tissue cholesterol and triglyceride levels significantly and raised the phospholipid levels in the tissues. Cholesterol ester synthetase (CES) and cholesterol ester hydrolase (CEH) activities were also assayed in the liver and glycollate fed rats exhibited increased CES activity. Pentosan polysulfate treatment had a lowering effect on CES but enhanced the CEH activity in both control and glycollate administered rats.  

Chromatographical analyses of serum were used after i.v. administration of 100mg heparin or pentosan polysulfate to humans (Tinti et al., 1967). As a consequence of the activation of the lipoprotein system, the concentration of free fatty acids temporarily increased. Short and long chain fatty acids were proportionally involved. In the course of 20 days treatment (100mg pentosan polysulfate i.m. per day) the unsaturated/polyunsaturated fatty acids in the cholesterol and phospholipid fractions decreased at the expense of the saturated fatty acids. A qualitatively and quantitatively analogous shift of this relationship occurred also after oral pentosan polysulfate treatment (100mg per day during 20 days). The authors concluded that "the advantageous influence on the quotient unsaturated/saturated fatty acids in the phospholipid or cholesterol fractions of the serum was evident and conclusive for the evaluation of the antiatherogenous properties of pentosan polysulfate” (Tinti et al., 1967).  

The ability of pentosan polysulfate to increase blood lipids was used to develop a model for the pre-exercise elevation of plasma free fatty acids in the horse, with a view to its future use in investigations of fat metabolism during exercise (Orme & Harris, 1997). A comparison of the lipase releasing and anticoagulative effects and the ability of heparin and pentosan polysulphate to affect an increase in plasma free fatty acid concentration, when co-administered with-a triglyceride emulsion, was quantified. Doses of 0.39 and 1.3 mg kg-1 body weight of heparin and pentosan polysulphate respectively, administered intravenously, resulted in a significant increase in plasma total lipase activity (P < 0.001). There was, however, no significant difference in plasma lipase activity between treatments. Heparin resulted in a mean 14 ± 6.5-fold increase in activated partial thromboplastin times (aPTT) compared with a mean 1.6 ± 0.1-fold increase with pentosan polysulphate. Both heparin and pentosan polysulphate when coadministered with a triglyceride emulsion (lverlip 20) resulted in a significant increase in plasma free fatty acid concentration (P < 0.001). Thus, whereas a higher dose of pentosan polysulphate elicited a comparable lipolytic effect to heparin, including significant elevation of plasma free fatty acids, this was associated with a much reduced effect upon clotting function. Pentosan polysulphate, therefore, represents a suitable alternative to heparin for the elevation of plasma free fatty acids before exercise when used in conjunction with a triglyceride emulsion. A significant decrease in postprandial plasma triacylglycerol concentration was associated with a mean 50% increase in plasma total lipase activity following pentosan polysulfate administration (Orme et al., 1997). The increase in plasma total lipase activity following pentosan polysulfate administration may reflect an increase in muscle lipoprotein lipase (EC 3.1.1.34) activity, which would increase the capacity of muscle for free fatty acid uptake from circulating triacylglycerol-rich plasma lipoproteins.  

The mode of action of pentosan polysulfate in accelerating intravascular lipolysis appears to be by activating the glyceryl-ester hydrolase system (tri, di and monoesterases) to improve utilisation of exogenous and endogenous neutral fats. This was clearly demonstrated by measurements of the levels of triesterases, monoesterases and free fatty acids in human serum 1, 3, 5, 7, and 9 hours after subcutaneous administration of pentosan polysulfate. The levels of the esterase increased considerably within one hour of drug treatment. The optimum levels were achieved between 1 to 3 hours and declined progressively over the 9 hour period.  

The effective antilipid effects produced by pentosan polysulfate have been used with success in the clinical treatment of arteriosclerosis (Leblois et al., 1962; Moinade & Moinade, 1963; Cloarec et al., 1964; Colson, 1964; Martin-Noel & Grunwald, 1964; Broustet et al., 1965) and in cerebral insufficiency (Milbled & Delmas-Marsalet, 1965; Tricot et al., 1965; Micheletti et al., 1965; Siedek, 1967).  

Pentosan polysulfate could also assist in the prevention of atherogenesis by enhancing endothelial regeneration (Herbert et al., 1989). Endothelium was removed from rabbit aortae using a balloon catheter and the endothelial outgrowth was measured. In the controls, after initial regrowth of approximately 4 weeks, endothelial regeneration stopped before cellular regrowth was complete. However, when treated with daily subcutaneous injections of pentosan polysulphate (8 mg/kg/day), endothelium fully repopulated the denuded areas 16 weeks after beginning of the treatment.  
  

9. Summary Of Clinical Pharmacokinetics  
Pentosan polysulfate is, as described above, a polysulfated oligosaccharide which, because of its strongly anionic character and molecular size, exhibits unusual pharmacokinetics. Pentosan polysulfate thus behaves quite differently than small, easily diffusable molecules of lesser charge, such as the conventional non-steroidal anti-inflammatory drugs (NSAIDs). The highly charged nature of the drug confers strong binding to serum proteins as well as those present in connective tissues. Strong interactions also occur with glycoproteins of cell surfaces and in some cases this process may influence cell motility, communications and metabolism. The pharmacokinetics of pentosan polysulfate is for these reasons complex and is very much influenced by the dose and the route of administration used.  

The pharmacokinetics of pentosan sulfate has been studied using iodinated derivatives (MacGregor et al., 1985a; Dawes & Pepper, 1992) and a competitive binding assay (MacGregor et al., 1985b). These data indicated that pentosan polysulfate was rapidly cleared from the circulation when administered parenterally returning later in a desulfated form. Organ distribution studies suggested that the liver and spleen were the major sites of metabolism. As the amount of drug administered was raised, however, the percentage of desulfated metabolites appearing in the plasma decreased, indicating a limited capacity of the reticuloendothelial system to desulfate this molecule. Since no low-molecular-weight metabolites were detected in plasma it was concluded that the depolymerisation of the drug occurred in the kidneys, and/or the breakdown products were rapidly excreted. Again saturation of the sites of metabolism in the kidney was evident as undegraded pentosan was excreted in the urine when high dosages were administered. Whole-body autoradiography using tritiated sodium pentosan polysulfate given to rats intravenously showed a preferential deposition of the drug in the urinary tract after 4 hours (Odlind et al., 1987). However, high radioactivity and thus drug, was also found in the upper intestine, liver and lymph nodes. When given orally, tritiated pentosan polysulfate was largely localised to the large intestine, although radioactivity was also detected in the liver, kidney, ureter and bladder lining.  

A calcium derivative of pentosan polysulfate has been developed which is more effectively absorbed after oral administration than the sodium salt. The relative oral absorption has been studied in the rat using activated partial thromboplastin times (aPTT) and release of tissue plasminogen activator (Klocking & Markwardt, 1985; 1986; Klocking, 1993). Although the absorption of the sodium salt was in the order of 0.5 - 1% after a single oral dose of 5mg/kg, the blood levels of the calcium salt of pentosan polysulfate were maintained for more than 4 hours, indicating 10 to 20% absorption (Klocking & Markwardt, 1986).  

Pentosan polysulfate was radiolabelled with ¹²⁵I and its catabolism by human vascular endothelial cells in culture was studied (Dawes & Pepper, 1992). Pentosan polysulfate was associated with the cellular fraction and incorporated into the subendothelial matrix. High molecular weight, fully desulphated carbohydrate chains were major catabolic products. Thus the first step in catabolism of pentosan polysulfate by human vascular endothelial cells appears to be complete desulphation, a process which is dependent upon binding of the molecule to the cell.  

The published pharmacokinetic studies of pentosan polysulfate indicate that when the drug is administered parenterally it is rapidly cleared from the blood, being taken up by connective tissues and the reticuloendothelial system. The affinity of these tissues for polysulfated polysaccharides has been attributed to strong binding to specific matrix proteins such as thromboplastin (Andrews et al., 1983; Hutadilok & Ghosh, 1989; Ghosh & Hutadilok, 1996). 
 

 
1. Canine Osteoarthritis Clinical Studies
The efficacy of pentosan polysulfate was studied in 40 dogs diagnosed with chronic osteoarthritis in a double-blind placebo controlled study (Read et al., 1996). Animals randomly received nil (placebo), 1, 3 or 5 mg/kg sodium pentosan polysulfate intramuscularly for 4 weeks and were assessed weekly for stiffness, mobility, pain on joint manipulation and overall response using a validated scoring system for up to 4 weeks after the last injection. The 3 mg/kg dose but not the 1 or 5 mg/kg, was shown to afford significant improvement in the parameters defined. 

In a study conducted with ten elderly dogs with osteoarthritis given calcium pentosan polysulfate (3 mg/kg intramuscularly) once weekly for 4 weeks, the improvement in symptoms was found to correlate with plasma indicies of fibrinolytic activity and lipid profiles (Cheras et al., 1993). Baseline (time zero) measures of fibrinolytic activity were compared with control values in dogs that did not have osteoarthritis. Further measurements were made in treated dogs at weeks 1, 2, 3 and 7. The euglobin clot lysis time, which is a measure of fibrinolytic activity, was abnormally long in osteoarthritic dogs before treatment and was shortened with the administration of CaPPS. This effect lasted for up to 4 weeks after CaPPS treatment. Baseline triglyceride levels were significantly increased in the osteoarthritic animals and these levels decreased during the treatment period to be within control levels at week 7. Clinical imrprovement in symptoms was correlated with the change in laboratory parameters determined, ie., increased fibrinolytic activity and normalisation of triglyceride levels. 

Intramuscular sodium pentosan polysulfate (3 mg/kg) has also been evaluated in a randomised prospective study to determine its applicability in the treatment of fragmented coronoid process (FCP) and osteochondritis dissecans (OCD) of the elbow compared with conventional surgical management of these disorders in dogs (Bouck et al., 1995). In the animals treated with sodium pentosan polysulfate, a more rapid improvement in limb function was observed relative to the surgically treated group as determined by force plate analysis. Significantly, a 9 month follow-up showed no differences between groups in any of the gait parameters studied. This finding demonstrated that pentosan polysulfate injections as a medical treatment are a valid alternative to conventional surgery for the management of FCP and OCD in the dog. 
 

2. Human Osteoarthritis Clinical Studies
Although the number of clinical studies undertaken with pentosan polysulfate for conditions other than disorders of the musculoskeletal system are considerable (see preceding section), the reports of its clinical usage as an antiarthritic agent are at present limited. 

In an open study with 23 osteoarthritis patients, CaPPS at 2mg/kg intramuscularly was found to be effective in improving symptoms of pain (by visual analogue score, VAS), functional scores and reduction in the comsumption of NSAIDs for up to 12 weeks posttreatment (Verbruggen et al., 1994). This positive clinical response was accompanied by changes in the serological parameters: plasma free fatty acids, triglycerides, tissue plasminogen activator, its inhibitor (PAI-1) and alpha-antitrypsin inhibitor activites all showed correlations with patient improvement, suggesting that they might be useful surrogate markers of disease activity in osteoarthritis. 

A doubleblind, placebo-controlled clinical study in 114 patients with osteoarthritis of the knee has been performed in Perth, Australia (Edelman et al., 1994) where patients either received a salt solution or sodium pentosan polysulphate at 3mg/kg as an intramuscular injection once weekly for 4 weeks. Before entry into the study, a 2 week NSAID washout period was enforced. Outcome measures were performed at baseline, at the time of each injection and then monthly for 6 months. Clinical assessments included physician-determined step time, pain at rest (VAS); pain on walking (VAS); early morning stiffness using duration and VAS, patients’ self-assessed treatment effectiveness; global pain and a number of lifestyle function scores. Analysis of the 60 placebo and 54 actively treated pateints at the completion of the study showed they were matched for age, grade of osteoarthritis, body mass index and pain on entry into the study. Data was analysed both as change from baseline for each parameter and as frequency of response. In the pentosan polysulfate-treated patients, stiffness significantly improved after the first week and pain on walking or at rest, time to walk upstairs and overall pain significantly improved after the first month. Pain at rest and step time were still significantly better than the placebo group after 2 months. The patient satisfaction with treatment and survival rate for the 6 month study was also higher for the pentosan polysulfate-treated patients. As found with the Verbruggen et al. (1994) study, administration of pentosan polysulfate normalised the fibrinolytic and lipolytic abnormalities that existed in patients’ plasma before treatment. 

A preliminary report has been published (Rasaratnam et al., 1996) of a double-blind placebo controlled study of intra-articular sodium pentosan polysulphate (0 or 50mg per joint once weekly for 4 weeks) in 50 patients suffering osteoarthritis of the knee. Even with 31 of the 50 patients evaluated, there were significant improvements in pain and mobility for up to 2 months after completing the treatment. 

Results of a double-blind study of oral CaPPS in 49 patients with osteoarthritis of the hand (finger joints) are soon to be published (Verbruggen et al., 1999). Twenty-four patients received two courses of treatment of 6 weeks duration (20 mg/kg CaPPS twice weekly) over an observation period of 6 months, with 25 patients receiving placebo. Left-hand grip strength and night pain markedly improved in the CaPPS-treated group over the duration of the study, while global pain and morning stiffness were significantly improved after week 15. The placebo response as determined by VAS scores for pain or stiffness was found to be strong in these patients but in no instance did this group reach a 20% change from baseline. In contrast, the same assessments for the CaPPS-treated group all demonstrated a 20% or greater improvement relative to baseline by the end of the trial, that is, for 6 weeks after the last course of therapy, with no apparent side effects. These data strongly suggest that multiple courses of treatment with CaPPS may be more effective in eliciting symptomatic relief in patients with osteoarthritis of the hands than a single loading dose. 
 

 
In acute and chronic animal studies, pentosan polysulfate has exhibited low toxicity with LD₅₀s between 0.5 - 6.0 g/kg depending on the salt, species and route of administration. In the mouse and rat the LD₅₀ of both sodium and calcium pentosan polysulfate given intravenously is more than 600mg/kg (Klocking et al., 1991). In the same species, the LD₅₀ of both drugs administered orally exceeds 6 g/kg. An important, but rare, side-effect reported during human applications has been thrombocytopenia. Thrombocytopenia is known to occur occasionally with heparin and heparinoids (Warkenyin & Kelton, 1989), as well as other drug preparations (Kosty et al., 1989). Pentosan polysulfate may induce a transient thrombocytopenia in some individuals but this rapidly returns to the normality range within a few days of terminating treatment. 

The haemorrhagicological changes induced in healthy elderly subjects (81 ± 10 years) treated i.m. twice daily with 50mg pentosan polysulfate, were examined over a seven day period (Freyburger et al., 1987). Blood viscosity, filterability and triglycerides were improved without concomitant changes in hematocrit, fibrinogen, haptoglobin, ß-2-globulins and plasma lipid levels. At the commencement of the study the mean platelet count was 266,000 ± 47,000µl and after continuous daily treatment declined to 171,000 ± 35,000µl. The depression in platelet recovered to the pre-drug range in the elderly patients 24 - 48 hours after cessation of drug administration. In rare cases (25 cases in 25 years) there may occur an immunoallergic thrombocytopenia similar to that sometimes observed with heparin (Warkenyin & Kelton, 1989). 

This adverse reaction needs a sensitisation period of several days and therefore the severe thrombocytopenia (platelet number below 100,000µl or more than 40% below the original value) will be found only 6 - 14 days after the initiation of drug treatment. Pentosan polysulfate may also cross-react with individuals sensitive to heparin and heparin-like substances. It is considered that these allergic-type thrombocytopenias are due to the consumption of platelets by a pentosan polysulfate primed platelet aggregating antibody (Follea et al., 1986). The reaction is often combined with thrombosis or haemorrhages, and in these cases medication must be stopped immediately. 

Since pentosan polysulfate is approved for marketing for human use in most EEC countries, Australia, South Africa and the Scandinavian countries, the rigorous requirements for chronic toxicity, mutagenicity and teratogenicity have been satisfied.

HOME | REFERENCES