The Pathogenesis of Malignant Ascites

The Pathogenesis of Malignant AscitesJ TamsmaIntroductionPeritonitis carcinomatosa (PC) indicating the presence of malignant cells in the peritoneal cavity, is a well known complication of malignant disease. As a result, so-called malignant ascites develops. In this chapter, we will address the subject from a pathophysiologic perspective. First, we will review the complex micro-scopic anatomy and physiology of the normal peritoneal membrane. Secondly, characteristics of malignant ascites and its pathophysiology will be reviewed using Starling’s equation of capillary forces.Anatomical and Physiological ConsiderationsAnatomy of the Peritoneal MembraneThe microscopic anatomy of the peritoneal membrane reveals, apart from the cap-illary endothelium and basement membrane, three distinct barriers to prevent the loss of proteins into the peritoneal cavity: the interstitial stroma, the mesothelial basement membrane and the mesothelial cells lining the peritoneum [1]. Follow-ing the route from the intravascular to the intraperitoneal space, the endothelial cells are the first barrier encountered. Endothelial cells have an extracellular gly-cocalix with fixed anionic charges that is difficult to pass for anionic macromole-cules such as albumin, an important contributor to plasma oncotic pressure [2].Peritoneal endothelial cells are linked with tight junctions, and as a result transport is transendothelial using intracellular pores [3,4]. Endothelial cells are separated from the interstitial space by the endothelial basement membrane. The general plan for basement membranes is a core of collagen to which several different kinds of macromolecules are anchored. Proteoglycans present in the basement110 J Tamsmamembrane may constitute a net negative charge, which forms a selective barrier for anionic proteins.The interstitial space consists of loose connective tissue composed of fibro-blasts, collagen, hyaluronic acid and negatively charged macromolecules. Hyualu-ronic acid is able to bind a considerable amount of water, for instance edema during peritonitis. The interstitial space acts as a filter with significant resistance against diffusion of macromolecules.The submesothelial basement membrane normally appears as a continuous layer at the interstitial site of the mesothelial cells. Evidence is available that negatively charged glycosaminoglycans are also present at this site. Mesothelial cells are the last barrier to be passed. The mesothelium consists of a monolayer of flat cells with a total estimated surface of about 2 square meters. The mesothelial cells show some functional similarity to endothelial cells. They have a glycocalix containing anionic charges and transcellular channels for macromolecular transport [5].In short, the presence of tight junctions between the endothelial cells in the peri-toneal capillaries and the presence of negatively charged macromolecules at sev-eral extracellular sites produce an effective barrier against leakage of negatively charged molecules such as albumin from plasma to the peritoneal cavity. Thus, the anatomy of the peritoneal membrane is such that it constitutes a relative imperme-ability to proteins while fluid and solutes easily pass the membrane. In other words, basic requirements to prevent excessive fluid filtration from the capillaries to the peritoneum are met by the anatomic “construction” of the peritoneal mem-brane.The Peritoneal Lymphatic SystemThe lymphatic system collects fluid, proteins, other macromolecules and cells, to return them to the systemic circulation. The smallest lymphatics consist of one layer of endothelium and drain into lymphatic capillaries. A basement membrane may be present at this level but if so, it is interrupted. The lymphatic capillary net is organized as a plexus along the submesothelial surface and drains to lymph ves-sels. Lymph vessels have valves and spirally formed smooth muscle cells. They are innervated. Contractions of lymph vessels are generated by myogenic stimuli, and are at least influenced by activation of α-adrenoreceptors, temperature, cal-cium concentrations, and vasoactive peptides.A specialised, intriguing anatomic feature of the peritoneal lymphatic system are the so-called stomata. Stomata are open communications between the abdomi-nal cavity and the submesothelial diaphragmatic lymphatics. They are supposed to play a major role in peritoneal lymphatic drainage [1], as most of intraperitoneal fluid is absorbed at this site [6].The mechanisms involved in lymph formation are still unclear. A hydraulic pressure theory has been proposed [7]. Normally, the interstitial pressure is nega-tive [8], and an increase in intraabdominal pressure will result in increased lymph production. A close correlation between fluid absorption and intra-abdominal pressure has been shown, in line with this theory [9]. Another hypothesis statesThe Pathogenesis of Malignant Ascites 111 that osmotic forces are dominant. This theory postulates a protein concentrating mechanism at the initial lymphatics [10]. Active transendothelial transport of al-bumin has been shown [11], which could create the necessary osmotic force.Characteristics of Malignant Ascites - Intraperitoneal Protein AccumulationMalignant ascites is characterised by positive cytology of malignant cells. Com-pared to ascites caused by cirrhosis more white blood cells and a higher lactate dehydrogenase (LDH) level are present [12,13]. Interestingly from the viewpoint of capillary hemodynamics is the observation that mean protein levels of ascites are high in patients with PC [13], as are the albumin concentrations [12-14]. The difference between serum- and ascites-albumin concentration is small. Thus, pro-tein and albumin accumulate intraperitoneally in malignant ascites, which will be reviewed further as we discuss Starling’s law of capillary hemodynamics. Impaired Drainage or Increased Production?Fluid accumulation will occur if lymphatic drainage of the peritoneal cavity is compromised or if net filtration is increased, overwhelming lymphatic capacity. Peritoneal fluid kinetics has been studied extensively in peritoneal dialysis. In di-alysis, fluid accumulation is possible only if net filtration exceeds net absorption. Net fluid filtration is the resultant of the osmolality of the dialysate. The higher the osmolality, the higher the force that attracts fluid from the intravascular compart-ment. Interestingly, the osmolality of the dialysate changes in time. Osmotically active molecules disappear through lymphatic transport and are diluted due to the attracted amount of fluid (water). An important consequence of this mechanism is a reduction in the rate of filtration in time [15]. In contrast, lymphatic drainage proceeds at a fairly constant rate of 40 ml/hr during dialysis. The above mentioned stomata located at the peritoneal membrane lining the diaphragm are the principle site of drainage [15]. Thus, the net effect on intraperitoneal fluid accumulation can be calculated from the combined effects of filtration and lymphatic transport. In malignant ascites, fluid accumulation can likewise be regarded as the resultant of filtration minus drainage.There is evidence for impaired lymphatic drainage in PC. This was studied in mice in which ascites was induced by injecting tumour cells intraperitoneally [16]. Alterations in diaphragmatic lymphatic absorption were determined radio-graphically. Diaphragmatic and retrosternal lymph vessels became occluded 5 days after tumour cell injection. Ascites formation was evident 5 to 7 days after injection of tumour cells. Comparable experimental data showing decreased lymphatic drainage have been produced by others [17]. Furthermore, lymphoscintigraphy showed decreased lymphatic drainage in humans [18,19].112 J TamsmaTogether, there is fair evidence for decreased lymphatic drainage as contributing factor in the pathogenesis of malignant ascites (Fig. 1) [19].In addition to impaired lymphatic drainage there is evidence for increased fluid production. Using radioactive isotopes, it was shown that the inflow rate of plasma into the peritoneal cavity was increased six- to sixteen-fold [20]. The patho-physiology of increased fluid production is described by Starling’s law of capillary hemodynamics.Starling’s law of Capillary HemodynamicsThe exchange of fluid between the plasma and the interstitium is determined by the hydraulic and oncotic pressures in each compartment. The relationship be-- P if cap -π if )]In this equation Lp is the unit permeability or porosity of the capillary wall, S is the surface area available for filtration, P cap and P if are the capillary and interstitial fluid hydraulic pressures, πcap and πif are the capillary and interstitial fluid oncotic pressures, and s represents the reflection coefficient of proteins across the capillary wall (with values ranging from 0 if completely permeable to 1 if completely impermeable) [21]. Increased capillary permeability (Lp), increased surface area (S) available for filtration, increased hydraulic pressure difference (P cap - P if ), a decreased oncotic pressure difference s(πcap -π if ) or a combination of these factors could account for an increase of net filtration.Figure 1. Schematic drawing of the proposed pathogenesis of malignant ascites. Normal physiology of fluid filtration and absorption results in a constant flow of fluid from the vascular compartment to the peritoneal space without significant fluid accumulationtween these parameters can be expressed by Starling’s law [21]:Net filtration = LpS (δ hydraulic pressure - δ oncotic pressure)= LpS [(P cap ) - s(πThe Pathogenesis of Malignant Ascites 113 Increased Capillary PermeabilityIn PC, increased permeability to proteins was observed in mice after intraperito-neal administration of tumour cells [16]. In another study, it was shown that a few days after intraperitoneal injection of Walker 256 carcinoma cells new capillaries were observed and bloody ascites developed. Inhibition of angiogenesis with lo-cally administered protamine prevented new capillaries to develop and also pre-vented the occurence of ascites [22]. The coincidence of increased permeability and new vessel formation is striking.It is now well understood that tumour growth is dependent on angiogenesis, the formation of new blood vessels [23]. Angiogenesis starts by stimulation of the endothelium, resulting in hyperpermeability of the endothelial membrane and degradation of the basement membrane and underlying stroma. The next step is the migration and proliferation of endothelial cells, and the formation of new blood vessels and capillaries [23]. Two important factors in angiogenesis are basic fibroblast growth factor (b-FGF) and vascular endothelial growth factor (VEGF) [24]. VEGF was discovered as a factor creating hyperpermeability and was initially named vascular permeability factor (VPF) [25].In a mouse model, it was shown that small blood vessels lining the peritoneal cavity (mesentery, peritoneal wall, diaphragm) became hyperpermeable several days after intraperitoneal tumour cell injection. The development of hyper-permeable microvessels correlated with ascites VPF concentration [26]. Most tumours express VEGF [27], including ovarian [28,29], gastric and colon carcinomas [30]. A study has been performed confirming the presence of high VEGF concentrations in malignant ascites [31,32]. Furthermore, malignant ascites production, but not tumour growth, was completely inhibited in mice when treated with function-blocking VEGF antibodies. When the treatment was stopped, all mice developed ascites within two weeks [33]. These experimental results have been confirmed by others using anti-VEGF antibodies [34,35], VEGF tyrosine kinase receptor inhibitors [35,36] or exogenous soluble human VEGF receptor [37].These data strongly suggest that increased capillary permeability due to production of locally active substances such as VEGF is an important factor in the pathophysiology of malignant ascites (Fig. 2). In words of the equation: the value of Lp become larger thus leading to increased net filtration.114 J TamsmaFigure 2. Presence of tumour cells results in the obliteration of lymphatic drainage as depicted with (X). This process seems particularly important at the diaphragm which is an important site of fluid drainage. Furthermore, production of locally active molecules such as VEGF and b-FGF results in changes in Starlings’ law of capillary hemodynamics (mostly at the level of LpS and s(πcap - πif), see text). Thus, the balance between fluid production and drainage is disturbed leading to the formation of ascitesIncreased Filtration Surface AreaIn mice, the size and number of peritoneal lining microvessels and subsequently cross sectional area increased after intraperitoneal tumour cell injection [38]. The site of production of malignant ascites has also been studied in patients using plas-tic rings with absorbent paper which were placed on peritoneal tumour and tu-mour-free surface. The rate of production of ascites of tumour-free omentum and small bowel surface was increased. The rate of fluid production from the tumour surface was also higher than the fluid production of peritoneum of control sub-jects, but less outspoken. The authors concluded that “undoubtedly fluid exuded from the tumour surface but the lion's share comes from the disease free perito-neum” [20]. Thus, in malignant ascites an increased cross sectional area of mi-crovessels lining the peritoneal cavity has been shown in an experimental setting. In addition, it seems that in human subjects the tumour free peritoneal surface is able to produce the surplus of fluid in malignant ascites [20]. These findings are in line with an increased S (experimental) and S or LpS (human) in Starlings’ equa-tion.Increased Hydraulic Pressure DifferenceThe same paper [20] reported on portal pressure in controls and in patients with ovarian cancer with or without ascites. A minor increase of portal vein pressure was observed in patients with ascites. (P cap - P if) will probably not change dramati-cally.The Pathogenesis of Malignant Ascites 115 Decreased Oncotic Pressure DifferenceIn normal physiology, albumin is known to be an effective osmol which contributes to intravascular oncotic pressure necessary to reabsorb fluid from the interstitial space. If the oncotic pressure difference decreases, reabsorbtion decreases and interstitial fluid accumulation results. In PC, protein accumulates intraperitoneally [12]. These intraperitoneal proteins may be partly degraded to smaller peptides and amino acids, which could contribute to intraabdominal oncotic pressure. This situation is comparable to peritoneal dialysis solutions containing a 5% amino acid concentration, which are very effective in forcing ultrafiltration [15]. As the plasma to peritoneal oncotic pressure difference decreases and even becomes negative reabsorption into the intravascular compartment will diminish and fluid may even be “filtrated” into the peritoneal cavity (Fig. 1).Thus, regarding Starling’s law of capillary hemodynamics we propose that the increased capillary permeability is essential in the pathophysiology of malignant ascites. The resultant decreased or negative oncotic pressure difference attracts fluid into the peritoneal cavity. In the equation: s(πcap -πif) will approximate zero or attain a negative value thus increasing net filtration. Overall the equation will become:Net filtration ↑↑ = LpS↑ [(P cap - P if) - s(πcap -πif) ↓]ConclusionThe pathogenesis of malignant ascites is beginning to be elucidated. Decreased lymphatic absorption and increased fluid production can be identified as contribut-ing features of ascites formation. The increased net capillary fluid production is due to an increase of capillary permeability and surface area, and a subsequent in-crease of intraperitoneal protein concentration leading to increased intraperitoneal oncotic pressure. This sequence might be the result of biologically active peptides produced by tumour cells such as VEGF and b-FGF. Interference with these me-diators may serve as target in future therapeutic strategies.References1. Gotloib L, Shostak A (1990) The functional anatomy of the peritoneum as adialyzing membrane. In: Twardowski ZJ, Nolph KD, Khanna R, editors. Con-tempory issues in Nephrology vol 22. Peritoneal Dialysis: new concepts and applications. New York: Churchill Livingstone, 1-29116 J Tamsma2. Nolph KD, Miller F, Rubin J, Popovich R (1980) New directions in peritonealdialysis concepts and applications. Kidney Int Suppl 10:S111-S1163. Renkin EM (1986) Some consequences of capillary permeability to macro-molecules: Starling’s hypothesis reconsidered. Am J Physiol 250 (5Pt2): H706-H7104. Grega GJ, Adamski SW, Dobbins DE (1986) Physiological and pharmacol-ogical evidence for the regulation of permeability. Fed Proc 45(2):96-1005. Gotloib L, Digenis GE, Rabinovich S, et al (1983) Ultrastructure of normalrabbit mesentery. Nephron 34(4):248-2556. Lifshitz S (1982) Ascites, pathophysiology and control measures. Int J RadiatOncol Biol Phys 8(8):1423-14267. Allen L (1931) Volume and pressure changes in terminal lymphatics. AmJ Physiol 123:38. Guyton AC (1963) A concept of negative interstitial pressure based on pres-sures in implanted perforated capsules. Circ Res 12:3999. Zink J, Greenway CV (1977) Control of ascites absorption in anesthetizedcats: effects of intraperitoneal pressure, protein, and furosemide diuresis. Gas-troenterology 73(5):1119-112410. Casley-Smith JR (1978) A fine ultrastructural study of variations in proteinconcentration in lacteals during compression and relaxation. Lymphology 12:59-6511. Shasby DM, Shasby SS (1985) Active transendothelial transport of albumin.Interstitium to lumen. Circ Res 57(6):903-90812. Runyon BA, Hoefs JC, Morgan TR (1988) Ascitic fluid analysis in malig-nancy-related ascites. Hepatology 8(5):1104-110913. Salerno F, Restelli B, Incerti P, et al (1990) Utility of ascitic fluid analysis inpatients with malignancy-related ascites. Scand J Gastroenterol 25(3):251-256 14. Jungst D, Xie Y, Gerbes AL (1992) Pathophysiology of elevated ascites fluidcholesterol in malignant ascites. Increased ascites to serum relation of pro-teins and lipoproteins in patients with peritoneal carcinomatosis as compared to patients with cirrhosis of the liver. J Hepatol 14(2-3):244-24815. Mactier RA (1990) Kinetics of ultrafiltration with glucose and alternativeosmotic agents. In: Twardowski ZJ, Nolph KD, Khanna R, editors. Contempo-rary issues in nephrology. Peritoneal dialysis, new concepts and applications.New York: Churchill Livingstone, 29-5216. Fastaia J, Dumont AE (1976) Pathogenesis of ascites in mice with peritonealcarcinomatosis. J Natl Cancer Inst 56(3):547-55017. Feldman GB, Knapp RC, Order SE, Hellman S (1972) The role of lymphaticobstruction in the formation of ascites in a murine ovarian carcinoma. Cancer Res 32(8):1663-166618. Coates G, Bush RS, Aspin N (1973) A study of ascites using lymphoscinti-graphy with 99m Tc-sulfur colloid. Radiology 107(3):577-58319. Bronskill MJ, Bush RS, Ege GN (1977) A quantitative measurement of peri-toneal drainage in malignant ascites. Cancer 40(5):2375-238020. Hirabayashi K, Graham J (1970) Genesis of ascites in ovarian cancer. AmJ Obstet Gynecol 106(4):492-497The Pathogenesis of Malignant Ascites 117 21. Rose BD, Post TW (2001) Edematous states. In: Rose BD, Post TW (eds).Clinical physiology of acid-base and electrolyte disorders. New York, NY: McGraw-Hill, 478-53422. Heuser LS, Taylor SH, Folkman J (1984) Prevention of carcinomatosis andbloody malignant ascites in the rat by an inhibitor of angiogenesis. J Surg Res 36(3):244-25023. Hanahan D, Folkman J (1996) Patterns and emerging mechanisms of the an-giogenic switch during tumourigenesis. Cell 86(3):353-36424. Neufeld G, Cohen T, Gengrinovitch S, Poltorak Z (1999) Vascular endothe-lial growth factor (VEGF) and its receptors. FASEB J 13(1):9-2225. Senger DR, Galli SJ, Dvorak AM, et al (1983) Tumour cells secrete a vascu-lar permeability factor that promotes accumulation of ascites fluid. Science 219(4587):983-98526. Nagy JA, Masse EM, Herzberg KT, et al (1995) Pathogenesis of ascites tu-mour growth: vascular permeability factor, vascular hyperpermeability, and ascites fluid accumulation. Cancer Res 55(2):360-36827. Senger DR, Perruzzi CA, Feder J, Dvorak HF (1986) A highly conserved vas-cular permeability factor secreted by a variety of human and rodent tumour cell lines. Cancer Res 46(11):5629-563228. Yamamoto S, Konishi I, Mandai M, et al (1997) Expression of vascular endo-thelial growth factor (VEGF) in epithelial ovarian neoplasms: correlation with clinicopathology and patient survival, and analysis of serum VEGF levels. Br J Cancer 76(9):1221-122729. Barton DP, Cai A, Wendt K, Young M, et al (1997) Angiogenic protein ex-pression in advanced epithelial ovarian cancer. Clin Cancer Res 3(9):1579-158630. Zebrowski BK, Liu W, Ramirez K, et al (1999) Markedly elevated levels ofvascular endothelial growth factor in malignant ascites. Ann Surg Oncol 6(4):373-37831. Kraft A, Weindel K, Ochs A, et al (1999) Vascular endothelial growth factorin the sera and effusions of patients with malignant and nonmalignant disease.Cancer 85(1):178-18732. Santin AD, Hermonat PL, Ravaggi A, et al (1999) Secretion of vascular endo-thelial growth factor in ovarian cancer. Eur J Gynaecol Oncol 20(3):177-181 33. Mesiano S, Ferrara N, Jaffe RB (1998) Role of vascular endothelial growthfactor in ovarian cancer: inhibition of ascites formation by immunoneutraliza-tion. Am J Pathol 153(4):1249-125634. Luo JC, Toyoda M, Shibuya M (1998) Differential inhibition of fluid accu-mulation and tumour growth in two mouse ascites tumours by an antivascular endothelial growth factor/permeability factor neutralizing antibody. Cancer Res 58(12):2594-260035. Schlaeppi JM, Wood JM (1999) Targeting vascular endothelial growth factor(VEGF) for anti-tumour therapy, by anti-VEGF neutralizing monoclonal anti-bodies or by VEGF receptor tyrosine-kinase inhibitors. Cancer Metastasis Rev 18(4):473-481118 J Tamsma36. Xu L, Yoneda J, Herrera C, et al (2000) Inhibition of malignant ascites andgrowth of human ovarian carcinoma by oral administration of a potent inhibi-tor of the vascular endothelial growth factor receptor tyrosine kinases. Int J Oncol 16(3):445-45437. Stoelcker B, Echtenacher B, Weich HA, et al (2000) VEGF/Flk-1 interaction,a requirement for malignant ascites recurrence. J Interferon Cytokine Res20(5):511-51738. Nagy JA, Morgan ES, Herzberg KT, Manseau EJ, Dvorak AM, Dvorak HF(1995) Pathogenesis of ascites tumour growth: angiogenesis, vascular remod-eling, and stroma formation in the peritoneal lining. Cancer Res 55(2):376-385。