Eut. surg. Res. 8: 51-60 (1976)
Metastasis Formation after Intravenous Tumour Cell Injection in Thrombocytopenic Rats1 La r s I varsson Department of Surgery I, Sahlgrenska sjukhuset, University of Göteborg, Göteborg
Key Words. Antiplatelet serum • Trauma ■Dextran 1000 • Metastasis formation Abstract. Induced thrombocytopenia has been found to decrease metastasis for mation after intravenous tumour cell injection which emphasizes the importance of platelets in metastasis formation. Using a syngeneic 20-methylcholanthrene-induced fibrosarcoma in rats, the effect of platelet reduction was investigated in combination with trauma and infusion of dextran 1000. It was found that platelets were impor tant for the increased formation of métastasés after trauma but not for the in creased formation of métastasés after infusion of dextran 1000. Thus, trauma and dextran 1000 stimulate metastasis formation by different mechanisms. Possible ex planations are discussed.
Remote trauma as well as infusion of dextran 1000 (average molecu lar weight 1,000,000) stimulate pulmonary metastasis formation after in travenous injection of tumour cells as shown in many experimental stud ies [8, 22, 23, 25, 30, 31, 37, 41-43]. Both these procedures induce rheo logical disturbances as aggregation of chylomicra, platelets and red cells with blockade of the microcirculation and signs of intravascular coagula tion [1-3, 9, 19-21, 34, 47]. Such changes can probably increase the pul monary lodgement of circulating tumour cells and might also stimulate the microthrombus formation around lodged tumour cells. According to the microthrombus theory, lodged tumour cells give rise to métastasés af ter the formation of a surrounding microthrombus, consisting of platelets, fibrin and a few leucocytes [45, 46].
Received: April 11, 1975; accepted: April 24, 1975.
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
1 This study was supported by grants from the Swedish Cancer Society.
52
IVARSSON
In 1968 G asic et al. [16] reported that induced platelet reduction de creased metastasis formation after intravenous tumour cell injection in mice. They suggested that this effect was related to ‘tumour cell attach ment to the host vascular endothelium either by clot formation, promo tion of surface stickiness or both’. The purpose of the present study was to investigate whether platelet reduction reduced the increased formation of pulmonary métastasés in duced by trauma and dextran 1000.
Animals. Inbred rats of the Wistar strain, resistant to the dextran anaphylactoid reaction, were used [32]. All experimental animals were 4-5 months of age. Ani mals of the same sex were always used in each experiment. They were fed on an ad libitum diet of rat pellets and water and were housed in plastic cages, 5-10 animals in each, and kept in an air-conditioned room at about 22 °C. Tumour. The tumour was a syngeneic transplantable 20-methylcholanthreneinduccd fibrosarcoma in its 34th transfer generation. A trypsin-disintegrated tumour cell suspension was used, which was prepared as described in an earlier paper [30], The cell viability was tested with nigrosin and was about 70°/o. Trauma. The trauma used was a bilateral femoral crush fracture delivered with pliers under light ether anaesthesia. This is a standardized type of trauma used in earlier experiments on trauma and metastasis formation [43]. Dextran. Dextran 1000 (average MW 1,000,000) was kindly supplied by Phar macia AB, Uppsala, Sweden, and a 10% solution in 5.5% glucose was prepared. Antiplatelet serum (APS). A rat APS was prepared in rabbits in the following way: 20 ml of arterial rat blood was collected in plastic tubes containing 3 ml of an acid ACD solution according to the formula of M orrison and Baldini [38] and centrifuged at 190 g for 15 min at room temperature. The platelet-rich plasma PRP was pooled in a polyvinylchloride bag (Fenwal TA-6) and centrifuged again at 190 g for 15 min to eliminate remaining erythrocytes and leucocytes. The PRP was transferred to a transfer pack (Fenwal transfer pack TA-3) and then centrifuged at 1,000 g for 20 min. The supernatant platelet-poor plasma was transferred to another transfer pack. The platelets were washed and re suspended in saline to a number of 10®/5 ml. 5 ml of this suspension was injected (i.v.) twice with an interval of 14 days into adult white domestic rabbits. Seven days after the last injection the rabbits were exsanguinated and serum was prepared. This serum was decomplemented at 56 °C for 30 min and absorbed for 1 h at 37 °C with an equal volume of packed rat red cells. At last the serum was sterile filtered (Millipore® filter, white plain, diameter 25 mm, pore size 0.22 wm) and frozen to -73 °C. The activity of the APS used in this experiment was tested in vivo in rats of the Wistar strain. The rats were given an injection of 0.2 ml APS i.v. The number of
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
Material and Methods
Metastasis and i.v. Tumour Cell Injection in Thrombocytopenic Rats
53
platelets was determined before and 24 h after the APS injection. The amount of APS given reduced the number of platelets by about 90%. The number of leuco cytes was determined simultaneously and reduced by up to 40%. A pilot study was also performed to test the effect of APS treatment on metas tasis formation after intravenous tumour cell injection. Now APS from another batch, 0.5 ml of which gave a 50% reduction of the platelet number in 24 h, was used. 16 animals of the same strain as used later on in the experiment were injected with 1 ml i.v. of this APS. 24 h later these animals, together with 14 control ani mals, were injected with 5X104 tumour cells i.v. from a syngeneic methylcholanthrene-induced fibrosarcoma in its 30th transfer generation. All animals were sacrificed 19 days later, and the pulmonary métastasés were registered in the way described below. The results are given in table I. As can be seen the formation of pulmonary métastasés was significantly reduced in the APS-treated animals. Platelet counts. These were done according to the method of Bjôrkman [4] and leucocytes were counted according to Ellerman’s method [40], Registration of métastasés. This was performed as described in a previous paper [30]. The total metastasis volume/cm3 pulmonary tissue, the number of metastases/cm3 pulmonary tissue and the average volume of a single metastasis in cm3 in animals with pulmonary métastasés were calculated according to Boeryd [5] and Boeryd et al. [6]. Since these values were distributed in a log-normal way, the sta tistical calculations were performed on their log values, and the median values are given in the tables where V = median total metastasis volume/cm3 pulmonary tis sue; N = median number of metastases/cm3 pulmonary tissue, and v = median metastasis volume in cm3. The number of animals with pulmonary métastasés per total number of animals in an experimental group was registered as the incidence of pulmonary métastasés. Some animals were unsuccessfully injected with tumour cells and were therefore excluded from the study. Statistical methods. Differences in incidence of animals with pulmonary métas tasés were tested with x 2 analysis according to F isher. Differences in volume and number of métastasés between two groups were ana lyzed with Student’s t test. For a simultaneous comparison between more than two groups a variance analysis was performed and differences between the groups were analyzed according to Scheffê . All statistical analyses were carried out at a 5% lev el of significance, i.e. there was a statistically significant difference when p0.05.
icantly increased. However, when the effect of APS treatment combined with trauma was compared to trauma alone, the differences in total me tastasis volume and number of métastasés were highly significant. On the other hand there were no significant differences between APS treatment alone and APS treatment combined with trauma. Dextran 1000 alone increased the total metastasis volume, the num ber of métastasés and the median metastasis volume significantly. When APS treatment was combined with dextran 1000 the effect on metastasis formation was about the same as that of dextran 1000 alone. No extrapulmonary métastasés were observed in any of the groups.
This experiment shows that APS treatment decreases the formation of pulmonary metastases after intravenous tumour cell injection when given both alone and in combination with trauma. The results also indicate that the effect of APS might be more pronounced in traumatized animals than in non-traumatized. Trauma does not stimulate metastasis formation in APS-treated animals. It is also evident that APS treatment does not influ ence the effect of dextran 1000 on metastasis formation. W ood [45, 46] was the first to demonstrate the dynamic events of the early stage of metastasis formation using intravital microscopy. He found that cancer cells, attached to the vascular endothelium, were rapidly sur
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
Discussion
IVARSSON
rounded by a microthrombus consisting of platelets, fibrin and a few leu cocytes. This microthrombus seemed to be necessary for the origination of a metastasis from a lodged tumour cell. The importance of a microthrombus formation for metastasis formation from circulating tumour cells has been indirectly and directly supported by several studies [10, 11, 17, 18, 33, 35, 43, 44]. However, there are investigators who have ques tioned or denied that a microthrombus formation is of any importance for metastasis formation [5, 7,13, 27, 28]. If the formation of métastasés from circulating tumour cells depends on a microthrombus, procedures which inhibit the formation of micro thrombi would counteract metastasis formation. Induced thrombocyto penia ought to have such an effect. G asic et al. [16] were the first to find that platelet reduction was accompanied by decreased metastasis forma tion. They also demonstrated that platelet transfusion reversed the effect of neuraminidase-induced platelet reduction on metastasis formation. In this study, treatment with APS reduced the formation of pulmonary métastasés, and the stimulating effect of trauma was almost eliminated in APS-treated animals. G rogan [26] found that a single intravenous or intraperitoneal injec tion of antilymphocytic serum blocked the reticuloendothelial system (RES), while repeated injections caused a temporary but sharp increase in phagocytic activity. If APS treatment has a similar blocking effect on the RES function, this should favour metastasis formation. Therefore, it does not seem probable that the effect of APS treatment on metastasis forma tion works over the RES. F isher and F isher [14, 15] studied the effect of both reticuloendothelial blockade and stimulation on hepatic metastas is formation after intraportal tumour cell injection. They found no evi dence of a functional role of the RES in the process of metastasis produc tion in the tumour-host system which they used. These results prove that the platelets really are the important factor in the chain of events following APS treatment, and they support the mi crothrombus theory. The platelets undoubtedly take part in the process of metastasis formation after intravenous tumour cell injection, and one im portant mechanism behind the stimulating effect of trauma on metastasis formation might be an increased microthrombus formation around lodged tumour cells. Platelets could perhaps influence the formation of métastas és also by other mechanisms than a microthrombus formation. It is sug gested by some authors [12, 24, 39] that platelets, being important for the surface or interendothelial cementing substance, might contribute to the
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
56
Metastasis and i.v. Tumour Cell Injection in Thrombocytopenic Rats
57
quality of the vascular endothelium. This function of the platelets is, how ever, as yet too little known to be evaluated. The effect of dextran 1000 on metastasis formation was not changed in APS-treated animals. This indicates that trauma and dextran 1000 stimu late metastasis formation by different mechanisms. Platelets seem to be rather irrelevant for the effect of dextran 1000 on metastasis formation. The conception that coagulation disturbances with increased microthrom bus formation could be one important factor for the stimulating effect of dextran 1000 on metastasis formation, is not supported by these results. Other mechanisms must be sought to explain the increased metastasis for mation after treatment with dextran 1000 in this experiment. One such mechanism might be the disturbed microcirculation following infusion of dextran 1,000. This is brought about by aggregation of the blood cells and often leads to complete blockade of the capillaries and postcapillary venules [19, 20, 43]. Such changes might favour the lodgement of tumour cells. As dextran 1000 aggregates all formed elements of blood, it is also probable that circulating tumour cells would be aggregated. Aggregation of tumour cells could explain the increased median metastasis volume. Dextran 1000 might be able to change the stickiness of tumour cells. Kojima and Sakai [36] found that the stickiness of tumour cells was directly correlated to the metastatic frequency, and that the stickiness was de creased by agents such as trypsin and protamine sulphate. We found that incubation of tumour cells in dextran 1000 solution before intravenous injection increased the formation of métastasés [30]. This is an indication of a direct action of dextran 1000 on the tumour cell surface which could be of importance for the formation of métastasés. Another explanation might be that dextran 1000 stimulates the formation of métastasés by in ducing fibrin clot formation around trapped tumour cells without the par ticipation of platelets.
1 Bergentz, S.-E.; G elin , L.-E., and Rudenstam, C.-M.: Experimental studies on pulmonary fat embolism. Biblthca anat., vol. 1, pp. 290-294 (Karger, Basel 1961). 2 Bergentz, S.-E.: Studies on the genesis of post-traumatic fat embolism. Acta chir. scand., suppl. 282 (1961). 3 Bigelow, W. G.; H eimbecker, R. O., and H arrison, R. C : Intravascular agglu tination (sludged blood), vascular stasis and sedimentation rate of the blood in trauma. Archs Surg., Chicago 59: 667-693 (1949).
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
References
58
IVARSSON
4 Björkman, S. E.: A new method for enumeration of platelets. Acta haemat. 22: 377-379 (1959). 5 Boeryd, B.: Action of heparin and plasminogen inhibitor (EACA) on metastatic tumour spread in an isologous system. Acta path, microbiol. scand. 65: 395-404 (1965) . 6 Boeryd, B.; G anelius, T.; L undin , P., and M ellgren, J.: Counting and sizing of tumour metastases in experimental oncology. Int. J. Cancer 1: 497-502 (1966) . 7 Boeryd, B.: Effect of heparin and plasminogen inhibitor (EACA) in brief and prolonged treatment on intravenously injected tumour cells. Acta path, micro biol. scand. 68: 547-552 (1966). 8 Boeryd, B. and R udenstam, C.-M.: Effect of heparin, plasminogen inhibitor (EACA) and trauma on tumour metastases. Acta path, microbiol. scand. 69: 28-34 (1967). 9 Borgström, S.; G elin , L.-E., and Z ederfeldt, B.: The formation of vein thrombi following tissue injury. Acta chir. scand., suppl. 247 (1959). 10 Cuffton , E. E. and Agosttno, D.: Factors affecting the development of metas tatic cancer. Effect of alterations in clotting mechanism. Cancer 15: 276-283 (1962). 11 C liffton, E. E. and A gostino, D.: The effects of fibrin formation and altera tions in the clotting mechanism on the development of metastases. Vase. Dis. 2: 43-52 (1965). 12 Cronkite, E. P.; Bond, V. P.; F liedner, T. M.; P aglia, D. A., and Adamik, E. R.: Studies on the origin, production and destruction of platelets; in Blood plate lets, pp. 595-609 (Little, Brown, Boston 1961). 13 F isher, B. and F isher, E. R.: Anticoagulants and tumour cell lodgement. Can cer Res. 27: 421-425 (1967). 14 F isher, E. R. and F isher, B.: Experimental studies of factors influencing hepat ic metastases. VII. Effect of reticuloendothelial interference. Cancer Res. 21: 275-280 (1961). 15 F isher, E. R. and F isher, B.: Experimental studies of factors influencing hepat ic metastases. X. Effect of reticuloendothelial stimulation. Cancer Res. 22: 478-483 (1962). 16 G asic, J. G.; G asic, T. B., and Stewart, C. C.: Antimetastatic effects associated with platelet reduction. Pathology 61: 46-52 (1968). 17 G astfar, H.; G raeber, F.; H errmann, A. und L oebell, E.: Intravitalmi kroskopische Beobachtungen von Tumorzellen in der Blutbahn (Bayer Film). Arch. Ohr.- Nas.- KehlkHeilk. 78: 534 (1961). 18 G astpar, H.: Zur Frage der Tumorzellausschwemmung in den Kreislauf vor, während und nach ausgedehnten Tumoroperationen am Hals. Arch. Ohr.- Nas.KehlkHeilk. 188: 542-546 (1967). 19 G elin , L.-E.: Studies in anemia of injury. Acta chir. scand., suppl. 210 (1956). 113: 463-465 (1957). 21 G elin , L.-E.: The significance of intravascular aggregation following injury. Bull. Soc. int. Chir. 18: 4-19 (1959).
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
20 G elin , L.-E.: Intravascular aggregation and capillary flow. Acta chir. scand.
Metastasis and i.v. Tumour Cell Injection in Thrombocytopenic Rats
59
41 R ifi , K. el ; Bacon, B.; M ehigan, J.; H oppe, E., and Cole, W. H.: Increased in cidence of pulmonary metastases after celiotomy: counteraction by heparin. Archs Surg., Chicago 91: 626-629 (1965).
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
22 G elin, L.-E. and Rudenstam, C.-M.: The influence of intravascular aggregation on the spread and growth of experimental tumours. Biblthca anat. vol. 4. pp. 131-141 (Karger, Basel 1965). 23 G elin, L.-E. and R udenstam, C.-M.: Trauma, microcirculation and tumour spread. Swed. Cancer Soc. Yb. 4: 53-65 (1966). 24 G ore, I.; T akada, M., and A ustin, J.: The ultrastructural basis of experimental thrombocytopenia. Fed. Proc. Fed. Am. Socs exp. Biol. 26: 672 (1968). 25 G riffiths, J. D.: The dissemination of cancer cells during operative procedures. Ann. R. Coll. Surg. 27: 14-44 (1960). 26 G rogan, J. B.: Reticuloendothelial system function after single and multiple injections of antilymphocyte serum. J. reticuloendoth. Soc. 8: 561-570 (1970). 27 Hagmar, B. and Boeryd, B.: Distribution of intravenously induced metastases in heparin- and coumarin-treated mice. Pathol. Eur. 4: 103-111 (1969). 28 H agmar, B. and Boeryd, B.: Disseminating effect of heparin on experimental tumour metastases. Pathol. Eur. 4: 274-282 (1969). 29 H ald, A.: Statistical theory with engineering applications (Wiley, New York 1952). 30 Ivarsson, L. and R udenstam, C.-M.: Dextrans and the formation of pulmonary metastases after i.v. tumour cell injection in dextran nonsensitive rats. Eur. surg. Res. (accepted for publication). 31 I varsson, L. and R udenstam, C.-M.: Trauma, dextran 40 and metastasis forma tion after i.v. tumour cell injection into dextran non-sensitive rats (to be pub lished). 32 Ivarsson, L.; A ppelgren, L., and Rudenstam, C. M.: Plasma volume after dex tran infusion into rats sensitive and non-sensitive to dextran. Eur. surg. Res. (ac cepted for publication). 33 Ivarsson, L. and R udenstam, C.-M.: Heparin, dextran 1000 and metastasis for mation after i.v. tumour cell injection in dextran non-sensitive rats. Br. J. Can cer (in press, 1975). 34 Knisely, M. H.; E liot, T. S„ and Block, E. H.: Sludged blood in traumatic shock. Archs Surg., Chicago 51: 220-236 (1945). 35 Koike, A.: Mechanism of blood-borne metastases. I. Some factors affecting lodgement and growth of tumour cells in the lung. Cancer 17: 450-460 (1964). 36 Kojima, K. and Sakai, I.: On the role of stickiness of tumour cells in the forma tion of metastases. Cancer Res. 24: 1887-1891 (1964). 37 Moore, G. E. and K ondo, T.: Study of adjuvant cancer chemotherapy by model experiments. Surgery, St Louis 44: 199-209 (1958). 38 M orrison, F. S. and Baldini, M. G.: Antigenic relationship between blood and platelets and vascular endothelium. Blood 33: 46-56 (1969). 39 M ustard, J. F.; R owsell, H. C.; M urphy, E. A., and G lynn, M. F.: The plate let and endothelium. Blood 25: 613 (1965). 40 N ordenson, N. G.: Kliniska laborationsmetoder. Astra, Sodertalje 3: 15-21 (1954).
60
IVARSSON
Lars I varsson, MD, Department of Surgery I, Sahlgrenska sjukhuset, University of Göteborg, Göteborg (Sweden)
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
42 R udenstam, C.-M.: Effect of fibrinolytic, antifibrinolytic and flow-promoting agents on metastasis formation and tumour growth; in W issler, D ao and W ood Endogenous factors influencing host-tumour balance, pp. 288-298 (University of Chicago Press, Chicago 1967). 43 Rudenstam, C.-M.: Experimental studies on trauma and metastasis formation. Acta chir. scand., suppl. 391 (1968). 44 Suemasu, K. and I shikawa, S.: Inhibition effect of heparin and dextran sulfate on experimental pulmonary métastasés. Gann 61: 125-130 (1970). 45 W ood, S., jr.: Pathogenesis of metastasis formation observed in vivo in the rab bit ear chamber. Archs Path. 66: 550-568 (1958). 46 W ood, S., jr.: Experimental studies on the intravascular dissemination of ascitic V2 carcinoma cells in the rabbit, with special reference to fibrinogen and fibri nolytic agents. Bull. Schweiz. Akad. med. Wiss. 20: 92-121 (1964). 47 Z ederfeldt, B.: Studies on wound healing and trauma with special reference to intravascular aggregation of erythrocytes. Acta chir. scand., suppl. 224 (1957).
Metastasis Formation after Intravenous Tumour Cell Injection in Thrombocytopenic Rats1 La r s I varsson Department of Surgery I, Sahlgrenska sjukhuset, University of Göteborg, Göteborg
Key Words. Antiplatelet serum • Trauma ■Dextran 1000 • Metastasis formation Abstract. Induced thrombocytopenia has been found to decrease metastasis for mation after intravenous tumour cell injection which emphasizes the importance of platelets in metastasis formation. Using a syngeneic 20-methylcholanthrene-induced fibrosarcoma in rats, the effect of platelet reduction was investigated in combination with trauma and infusion of dextran 1000. It was found that platelets were impor tant for the increased formation of métastasés after trauma but not for the in creased formation of métastasés after infusion of dextran 1000. Thus, trauma and dextran 1000 stimulate metastasis formation by different mechanisms. Possible ex planations are discussed.
Remote trauma as well as infusion of dextran 1000 (average molecu lar weight 1,000,000) stimulate pulmonary metastasis formation after in travenous injection of tumour cells as shown in many experimental stud ies [8, 22, 23, 25, 30, 31, 37, 41-43]. Both these procedures induce rheo logical disturbances as aggregation of chylomicra, platelets and red cells with blockade of the microcirculation and signs of intravascular coagula tion [1-3, 9, 19-21, 34, 47]. Such changes can probably increase the pul monary lodgement of circulating tumour cells and might also stimulate the microthrombus formation around lodged tumour cells. According to the microthrombus theory, lodged tumour cells give rise to métastasés af ter the formation of a surrounding microthrombus, consisting of platelets, fibrin and a few leucocytes [45, 46].
Received: April 11, 1975; accepted: April 24, 1975.
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
1 This study was supported by grants from the Swedish Cancer Society.
52
IVARSSON
In 1968 G asic et al. [16] reported that induced platelet reduction de creased metastasis formation after intravenous tumour cell injection in mice. They suggested that this effect was related to ‘tumour cell attach ment to the host vascular endothelium either by clot formation, promo tion of surface stickiness or both’. The purpose of the present study was to investigate whether platelet reduction reduced the increased formation of pulmonary métastasés in duced by trauma and dextran 1000.
Animals. Inbred rats of the Wistar strain, resistant to the dextran anaphylactoid reaction, were used [32]. All experimental animals were 4-5 months of age. Ani mals of the same sex were always used in each experiment. They were fed on an ad libitum diet of rat pellets and water and were housed in plastic cages, 5-10 animals in each, and kept in an air-conditioned room at about 22 °C. Tumour. The tumour was a syngeneic transplantable 20-methylcholanthreneinduccd fibrosarcoma in its 34th transfer generation. A trypsin-disintegrated tumour cell suspension was used, which was prepared as described in an earlier paper [30], The cell viability was tested with nigrosin and was about 70°/o. Trauma. The trauma used was a bilateral femoral crush fracture delivered with pliers under light ether anaesthesia. This is a standardized type of trauma used in earlier experiments on trauma and metastasis formation [43]. Dextran. Dextran 1000 (average MW 1,000,000) was kindly supplied by Phar macia AB, Uppsala, Sweden, and a 10% solution in 5.5% glucose was prepared. Antiplatelet serum (APS). A rat APS was prepared in rabbits in the following way: 20 ml of arterial rat blood was collected in plastic tubes containing 3 ml of an acid ACD solution according to the formula of M orrison and Baldini [38] and centrifuged at 190 g for 15 min at room temperature. The platelet-rich plasma PRP was pooled in a polyvinylchloride bag (Fenwal TA-6) and centrifuged again at 190 g for 15 min to eliminate remaining erythrocytes and leucocytes. The PRP was transferred to a transfer pack (Fenwal transfer pack TA-3) and then centrifuged at 1,000 g for 20 min. The supernatant platelet-poor plasma was transferred to another transfer pack. The platelets were washed and re suspended in saline to a number of 10®/5 ml. 5 ml of this suspension was injected (i.v.) twice with an interval of 14 days into adult white domestic rabbits. Seven days after the last injection the rabbits were exsanguinated and serum was prepared. This serum was decomplemented at 56 °C for 30 min and absorbed for 1 h at 37 °C with an equal volume of packed rat red cells. At last the serum was sterile filtered (Millipore® filter, white plain, diameter 25 mm, pore size 0.22 wm) and frozen to -73 °C. The activity of the APS used in this experiment was tested in vivo in rats of the Wistar strain. The rats were given an injection of 0.2 ml APS i.v. The number of
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
Material and Methods
Metastasis and i.v. Tumour Cell Injection in Thrombocytopenic Rats
53
platelets was determined before and 24 h after the APS injection. The amount of APS given reduced the number of platelets by about 90%. The number of leuco cytes was determined simultaneously and reduced by up to 40%. A pilot study was also performed to test the effect of APS treatment on metas tasis formation after intravenous tumour cell injection. Now APS from another batch, 0.5 ml of which gave a 50% reduction of the platelet number in 24 h, was used. 16 animals of the same strain as used later on in the experiment were injected with 1 ml i.v. of this APS. 24 h later these animals, together with 14 control ani mals, were injected with 5X104 tumour cells i.v. from a syngeneic methylcholanthrene-induced fibrosarcoma in its 30th transfer generation. All animals were sacrificed 19 days later, and the pulmonary métastasés were registered in the way described below. The results are given in table I. As can be seen the formation of pulmonary métastasés was significantly reduced in the APS-treated animals. Platelet counts. These were done according to the method of Bjôrkman [4] and leucocytes were counted according to Ellerman’s method [40], Registration of métastasés. This was performed as described in a previous paper [30]. The total metastasis volume/cm3 pulmonary tissue, the number of metastases/cm3 pulmonary tissue and the average volume of a single metastasis in cm3 in animals with pulmonary métastasés were calculated according to Boeryd [5] and Boeryd et al. [6]. Since these values were distributed in a log-normal way, the sta tistical calculations were performed on their log values, and the median values are given in the tables where V = median total metastasis volume/cm3 pulmonary tis sue; N = median number of metastases/cm3 pulmonary tissue, and v = median metastasis volume in cm3. The number of animals with pulmonary métastasés per total number of animals in an experimental group was registered as the incidence of pulmonary métastasés. Some animals were unsuccessfully injected with tumour cells and were therefore excluded from the study. Statistical methods. Differences in incidence of animals with pulmonary métas tasés were tested with x 2 analysis according to F isher. Differences in volume and number of métastasés between two groups were ana lyzed with Student’s t test. For a simultaneous comparison between more than two groups a variance analysis was performed and differences between the groups were analyzed according to Scheffê . All statistical analyses were carried out at a 5% lev el of significance, i.e. there was a statistically significant difference when p0.05.
icantly increased. However, when the effect of APS treatment combined with trauma was compared to trauma alone, the differences in total me tastasis volume and number of métastasés were highly significant. On the other hand there were no significant differences between APS treatment alone and APS treatment combined with trauma. Dextran 1000 alone increased the total metastasis volume, the num ber of métastasés and the median metastasis volume significantly. When APS treatment was combined with dextran 1000 the effect on metastasis formation was about the same as that of dextran 1000 alone. No extrapulmonary métastasés were observed in any of the groups.
This experiment shows that APS treatment decreases the formation of pulmonary metastases after intravenous tumour cell injection when given both alone and in combination with trauma. The results also indicate that the effect of APS might be more pronounced in traumatized animals than in non-traumatized. Trauma does not stimulate metastasis formation in APS-treated animals. It is also evident that APS treatment does not influ ence the effect of dextran 1000 on metastasis formation. W ood [45, 46] was the first to demonstrate the dynamic events of the early stage of metastasis formation using intravital microscopy. He found that cancer cells, attached to the vascular endothelium, were rapidly sur
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
Discussion
IVARSSON
rounded by a microthrombus consisting of platelets, fibrin and a few leu cocytes. This microthrombus seemed to be necessary for the origination of a metastasis from a lodged tumour cell. The importance of a microthrombus formation for metastasis formation from circulating tumour cells has been indirectly and directly supported by several studies [10, 11, 17, 18, 33, 35, 43, 44]. However, there are investigators who have ques tioned or denied that a microthrombus formation is of any importance for metastasis formation [5, 7,13, 27, 28]. If the formation of métastasés from circulating tumour cells depends on a microthrombus, procedures which inhibit the formation of micro thrombi would counteract metastasis formation. Induced thrombocyto penia ought to have such an effect. G asic et al. [16] were the first to find that platelet reduction was accompanied by decreased metastasis forma tion. They also demonstrated that platelet transfusion reversed the effect of neuraminidase-induced platelet reduction on metastasis formation. In this study, treatment with APS reduced the formation of pulmonary métastasés, and the stimulating effect of trauma was almost eliminated in APS-treated animals. G rogan [26] found that a single intravenous or intraperitoneal injec tion of antilymphocytic serum blocked the reticuloendothelial system (RES), while repeated injections caused a temporary but sharp increase in phagocytic activity. If APS treatment has a similar blocking effect on the RES function, this should favour metastasis formation. Therefore, it does not seem probable that the effect of APS treatment on metastasis forma tion works over the RES. F isher and F isher [14, 15] studied the effect of both reticuloendothelial blockade and stimulation on hepatic metastas is formation after intraportal tumour cell injection. They found no evi dence of a functional role of the RES in the process of metastasis produc tion in the tumour-host system which they used. These results prove that the platelets really are the important factor in the chain of events following APS treatment, and they support the mi crothrombus theory. The platelets undoubtedly take part in the process of metastasis formation after intravenous tumour cell injection, and one im portant mechanism behind the stimulating effect of trauma on metastasis formation might be an increased microthrombus formation around lodged tumour cells. Platelets could perhaps influence the formation of métastas és also by other mechanisms than a microthrombus formation. It is sug gested by some authors [12, 24, 39] that platelets, being important for the surface or interendothelial cementing substance, might contribute to the
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
56
Metastasis and i.v. Tumour Cell Injection in Thrombocytopenic Rats
57
quality of the vascular endothelium. This function of the platelets is, how ever, as yet too little known to be evaluated. The effect of dextran 1000 on metastasis formation was not changed in APS-treated animals. This indicates that trauma and dextran 1000 stimu late metastasis formation by different mechanisms. Platelets seem to be rather irrelevant for the effect of dextran 1000 on metastasis formation. The conception that coagulation disturbances with increased microthrom bus formation could be one important factor for the stimulating effect of dextran 1000 on metastasis formation, is not supported by these results. Other mechanisms must be sought to explain the increased metastasis for mation after treatment with dextran 1000 in this experiment. One such mechanism might be the disturbed microcirculation following infusion of dextran 1,000. This is brought about by aggregation of the blood cells and often leads to complete blockade of the capillaries and postcapillary venules [19, 20, 43]. Such changes might favour the lodgement of tumour cells. As dextran 1000 aggregates all formed elements of blood, it is also probable that circulating tumour cells would be aggregated. Aggregation of tumour cells could explain the increased median metastasis volume. Dextran 1000 might be able to change the stickiness of tumour cells. Kojima and Sakai [36] found that the stickiness of tumour cells was directly correlated to the metastatic frequency, and that the stickiness was de creased by agents such as trypsin and protamine sulphate. We found that incubation of tumour cells in dextran 1000 solution before intravenous injection increased the formation of métastasés [30]. This is an indication of a direct action of dextran 1000 on the tumour cell surface which could be of importance for the formation of métastasés. Another explanation might be that dextran 1000 stimulates the formation of métastasés by in ducing fibrin clot formation around trapped tumour cells without the par ticipation of platelets.
1 Bergentz, S.-E.; G elin , L.-E., and Rudenstam, C.-M.: Experimental studies on pulmonary fat embolism. Biblthca anat., vol. 1, pp. 290-294 (Karger, Basel 1961). 2 Bergentz, S.-E.: Studies on the genesis of post-traumatic fat embolism. Acta chir. scand., suppl. 282 (1961). 3 Bigelow, W. G.; H eimbecker, R. O., and H arrison, R. C : Intravascular agglu tination (sludged blood), vascular stasis and sedimentation rate of the blood in trauma. Archs Surg., Chicago 59: 667-693 (1949).
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
References
58
IVARSSON
4 Björkman, S. E.: A new method for enumeration of platelets. Acta haemat. 22: 377-379 (1959). 5 Boeryd, B.: Action of heparin and plasminogen inhibitor (EACA) on metastatic tumour spread in an isologous system. Acta path, microbiol. scand. 65: 395-404 (1965) . 6 Boeryd, B.; G anelius, T.; L undin , P., and M ellgren, J.: Counting and sizing of tumour metastases in experimental oncology. Int. J. Cancer 1: 497-502 (1966) . 7 Boeryd, B.: Effect of heparin and plasminogen inhibitor (EACA) in brief and prolonged treatment on intravenously injected tumour cells. Acta path, micro biol. scand. 68: 547-552 (1966). 8 Boeryd, B. and R udenstam, C.-M.: Effect of heparin, plasminogen inhibitor (EACA) and trauma on tumour metastases. Acta path, microbiol. scand. 69: 28-34 (1967). 9 Borgström, S.; G elin , L.-E., and Z ederfeldt, B.: The formation of vein thrombi following tissue injury. Acta chir. scand., suppl. 247 (1959). 10 Cuffton , E. E. and Agosttno, D.: Factors affecting the development of metas tatic cancer. Effect of alterations in clotting mechanism. Cancer 15: 276-283 (1962). 11 C liffton, E. E. and A gostino, D.: The effects of fibrin formation and altera tions in the clotting mechanism on the development of metastases. Vase. Dis. 2: 43-52 (1965). 12 Cronkite, E. P.; Bond, V. P.; F liedner, T. M.; P aglia, D. A., and Adamik, E. R.: Studies on the origin, production and destruction of platelets; in Blood plate lets, pp. 595-609 (Little, Brown, Boston 1961). 13 F isher, B. and F isher, E. R.: Anticoagulants and tumour cell lodgement. Can cer Res. 27: 421-425 (1967). 14 F isher, E. R. and F isher, B.: Experimental studies of factors influencing hepat ic metastases. VII. Effect of reticuloendothelial interference. Cancer Res. 21: 275-280 (1961). 15 F isher, E. R. and F isher, B.: Experimental studies of factors influencing hepat ic metastases. X. Effect of reticuloendothelial stimulation. Cancer Res. 22: 478-483 (1962). 16 G asic, J. G.; G asic, T. B., and Stewart, C. C.: Antimetastatic effects associated with platelet reduction. Pathology 61: 46-52 (1968). 17 G astfar, H.; G raeber, F.; H errmann, A. und L oebell, E.: Intravitalmi kroskopische Beobachtungen von Tumorzellen in der Blutbahn (Bayer Film). Arch. Ohr.- Nas.- KehlkHeilk. 78: 534 (1961). 18 G astpar, H.: Zur Frage der Tumorzellausschwemmung in den Kreislauf vor, während und nach ausgedehnten Tumoroperationen am Hals. Arch. Ohr.- Nas.KehlkHeilk. 188: 542-546 (1967). 19 G elin , L.-E.: Studies in anemia of injury. Acta chir. scand., suppl. 210 (1956). 113: 463-465 (1957). 21 G elin , L.-E.: The significance of intravascular aggregation following injury. Bull. Soc. int. Chir. 18: 4-19 (1959).
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
20 G elin , L.-E.: Intravascular aggregation and capillary flow. Acta chir. scand.
Metastasis and i.v. Tumour Cell Injection in Thrombocytopenic Rats
59
41 R ifi , K. el ; Bacon, B.; M ehigan, J.; H oppe, E., and Cole, W. H.: Increased in cidence of pulmonary metastases after celiotomy: counteraction by heparin. Archs Surg., Chicago 91: 626-629 (1965).
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
22 G elin, L.-E. and Rudenstam, C.-M.: The influence of intravascular aggregation on the spread and growth of experimental tumours. Biblthca anat. vol. 4. pp. 131-141 (Karger, Basel 1965). 23 G elin, L.-E. and R udenstam, C.-M.: Trauma, microcirculation and tumour spread. Swed. Cancer Soc. Yb. 4: 53-65 (1966). 24 G ore, I.; T akada, M., and A ustin, J.: The ultrastructural basis of experimental thrombocytopenia. Fed. Proc. Fed. Am. Socs exp. Biol. 26: 672 (1968). 25 G riffiths, J. D.: The dissemination of cancer cells during operative procedures. Ann. R. Coll. Surg. 27: 14-44 (1960). 26 G rogan, J. B.: Reticuloendothelial system function after single and multiple injections of antilymphocyte serum. J. reticuloendoth. Soc. 8: 561-570 (1970). 27 Hagmar, B. and Boeryd, B.: Distribution of intravenously induced metastases in heparin- and coumarin-treated mice. Pathol. Eur. 4: 103-111 (1969). 28 H agmar, B. and Boeryd, B.: Disseminating effect of heparin on experimental tumour metastases. Pathol. Eur. 4: 274-282 (1969). 29 H ald, A.: Statistical theory with engineering applications (Wiley, New York 1952). 30 Ivarsson, L. and R udenstam, C.-M.: Dextrans and the formation of pulmonary metastases after i.v. tumour cell injection in dextran nonsensitive rats. Eur. surg. Res. (accepted for publication). 31 I varsson, L. and R udenstam, C.-M.: Trauma, dextran 40 and metastasis forma tion after i.v. tumour cell injection into dextran non-sensitive rats (to be pub lished). 32 Ivarsson, L.; A ppelgren, L., and Rudenstam, C. M.: Plasma volume after dex tran infusion into rats sensitive and non-sensitive to dextran. Eur. surg. Res. (ac cepted for publication). 33 Ivarsson, L. and R udenstam, C.-M.: Heparin, dextran 1000 and metastasis for mation after i.v. tumour cell injection in dextran non-sensitive rats. Br. J. Can cer (in press, 1975). 34 Knisely, M. H.; E liot, T. S„ and Block, E. H.: Sludged blood in traumatic shock. Archs Surg., Chicago 51: 220-236 (1945). 35 Koike, A.: Mechanism of blood-borne metastases. I. Some factors affecting lodgement and growth of tumour cells in the lung. Cancer 17: 450-460 (1964). 36 Kojima, K. and Sakai, I.: On the role of stickiness of tumour cells in the forma tion of metastases. Cancer Res. 24: 1887-1891 (1964). 37 Moore, G. E. and K ondo, T.: Study of adjuvant cancer chemotherapy by model experiments. Surgery, St Louis 44: 199-209 (1958). 38 M orrison, F. S. and Baldini, M. G.: Antigenic relationship between blood and platelets and vascular endothelium. Blood 33: 46-56 (1969). 39 M ustard, J. F.; R owsell, H. C.; M urphy, E. A., and G lynn, M. F.: The plate let and endothelium. Blood 25: 613 (1965). 40 N ordenson, N. G.: Kliniska laborationsmetoder. Astra, Sodertalje 3: 15-21 (1954).
60
IVARSSON
Lars I varsson, MD, Department of Surgery I, Sahlgrenska sjukhuset, University of Göteborg, Göteborg (Sweden)
Downloaded by: King's College London 137.73.144.138 - 1/14/2019 10:50:49 AM
42 R udenstam, C.-M.: Effect of fibrinolytic, antifibrinolytic and flow-promoting agents on metastasis formation and tumour growth; in W issler, D ao and W ood Endogenous factors influencing host-tumour balance, pp. 288-298 (University of Chicago Press, Chicago 1967). 43 Rudenstam, C.-M.: Experimental studies on trauma and metastasis formation. Acta chir. scand., suppl. 391 (1968). 44 Suemasu, K. and I shikawa, S.: Inhibition effect of heparin and dextran sulfate on experimental pulmonary métastasés. Gann 61: 125-130 (1970). 45 W ood, S., jr.: Pathogenesis of metastasis formation observed in vivo in the rab bit ear chamber. Archs Path. 66: 550-568 (1958). 46 W ood, S., jr.: Experimental studies on the intravascular dissemination of ascitic V2 carcinoma cells in the rabbit, with special reference to fibrinogen and fibri nolytic agents. Bull. Schweiz. Akad. med. Wiss. 20: 92-121 (1964). 47 Z ederfeldt, B.: Studies on wound healing and trauma with special reference to intravascular aggregation of erythrocytes. Acta chir. scand., suppl. 224 (1957).
