- Email: [email protected]
Prognostic impact of an activation of coagulation in lung cancer*
Annals of Oncology 8: 781-784, 1997. © 1997 Kluwer Academic Publishers. Primed in the Netherlands.
Original article Prognostic impact of an activation of coagulation in lung cancer* R. Seitz,1 H.-H. Heidtmann, M.Wolf, A. Immel & R. Egbring Philipps University Hospitals, Department of Haematology/Oncology, Marburg, Germany; ^Present address: Department of Haematology and Transfusion Medicine, Paul-Ehrlich-Institute, Langen, Germany
Introduction A growing body of evidence demonstrates that activation of coagulation is involved in fundamental processes during the evolution of cancer tissues, and is of major importance for invasive tumour growth and metastatic spread [1]. The activation of the clotting cascade can be due to activities expressed by the tumour cells themselves, or the stimulation of tumour-associated inflammatory cells, such as macrophages [2, 3]. Several activities in cultured human lung tumour cells [4] may contribute to activation of the tissue factor pathway, or of factors II or X directly. The substrates of coagulation (fibrinogen, clotting factors) are thought to reach the tumour tissue by diffusion after extravasation of plasma, facilitated by vascular permeability factor [1]. Also, platelet activation may be triggered by lung cancer cells via thrombin generation [5]. Well-documented clinical observations [6, 7] have shown an increased risk of thromboembolic complications early in the course of cancer. Formation of a fibrin matrix appears to support tumour growth by promoting neoangiogenesis, and by shielding tumour cells against the attack of immunocompetent cells [8]. Thrombin may even be a growth * Dedicated to Professor KJaus Havemann on the occasion of his 65th anniversary.
factor for tumour cells [9]. In two controlled therapy trials, anticoagulation with warfarin as an adjuvant to chemotherapy was found to prolong survival of small cell lung cancer (SCLC) patients [10,11]. With sensitive molecular markers, in many lung cancer patients sustained activation of haemostasis can be demonstrated [12-14]. We could show [12] that the degree of activation of coagulation is related to initial tumour stage. The present study addresses the question, whether the extent of activation of coagulation, as assessed by pretherapeutic activation marker levels, has an impact on the response to subsequent standard antineoplastic treatment. Patients and methods In this prospective diagnostic study, 99 patients with newly diagnosed, histologically proven lung cancer, or relapse of lung cancer after previously documented complete remission, were evaluated. Patients with malignancies other than lung cancer, with pre-existing coagulation disturbances of other causes, on oral anticoagulant therapy, or who did not give their consent to participate, were not included. The study was approved by the local ethics committee. The histologic type was small cell lung cancer (SCLC) in 41 patients, and non-small cell lung cancer (NSCLQ in 58 patients (29 squamous cell, 12 adeno, 17 large cell and mixed cell type). Complete tumour staging was carried out using standard procedures. In SCLC, the categories of tumour stage were adapted from criteria approved in a previous multicentre chemotherapy trial [15], Extensive disease (ED)
Downloaded from https://academic.oup.com/annonc/article-abstract/8/8/781/304832 by guest on 19 September 2019
30 of 66 (45%) NRSP, but only 4 of 33 (12%) RSP. 88% of patients with TAT <6 ug/1 achieved remission, and 45% with Background: There is evidence that activation of coagulation TAT > 6 ug/1 (P = 0.0014). In the subgroup of 46 patients with by influencing tumour biology may have impact on clinical advanced disease, the six RSP showed lower TAT than the 40 course of lung cancer. NRSP: 4.65 ± 0.94 vs. 11.92 ± 2.49 ug/1 (P < 0.01); one of six Patients and methods: We measured the activation markers (17%) RSP, but 21 of 40 (53%) NRSP showed TAT > 6 ug/1. thrombin-antithrombin complex (TAT) and prothrombin fragThese data suggest that in lung cancer the activation of ment F 1+2 in 99 lung cancer patients immediately after diag- coagulation is an independent prognostic factor, since TAT nosis, before antineoplastic treatment. Outcome was assessed levels were different between RSP and NRSP, also within the at the end of appropriate standard primary therapy (four to six homogeneously unfavourable metastatic subgroup. It should courses of chemotherapy, surgery or radiation). be further studied, whether TAT can identify patients, whose Results and conclusions: The activation markers (means ± prognosis could be improved by anticoagulation as an adjunct SEM) were lower in the 33 responders (RSP; complete or to standard antineoplastic therapy. partial remission) than in the 66 non-responders (NRSP): TAT 3.96 ± 0.48 vs. 9.69 ± 1.57 ug/1 (P < 0.001), and F1+21.09 ± 0.09 Key words: lung cancer, prognosis, thrombin-antithrombin vs. 1.64 ± 0.25 nmol/1 (P < 0.05). TAT levels were > 6 ug/1 in complex Summary
Table 1. Activation marker levels (means ± SEM) of the patients acheiving remission (RSP, n = 33) and the non-responders (NRSP, 7i = 66); for comparison, values of 32 healthy volunteers. TAT RSP
NRSP Volunteers
3.96 ± 0.48 P = 0.0008 9.69 ± 1.57 2.1210.09
1.09 ±0.09 /> = 0.0415 1.64 ±0.25 0.71 ±0.04
Methods
Blood samples were obtained using 10 ml syringes containing one part 3.8% sodium citrate to nine parts blood The citrated blood samples were immediately centrifuged at 1500 g for 15 min at room temperature. Aliquots were frozen at —40 °C until assay. The activation markers thrombin-antithrombin complex (TAT) and prothrombin fragment Fi +2 (F 1+2 ) were determined by ELISA, using kits from Behringwerke (Marburg, Germany).
RSP
Figure 1. Distribution of the entire study patients according to their TAT levels (cut-off 6 ug/1) into the outcome categories RSP and NRSP.
(data not shown). The analysis of the entire study population showed that the values of RSP and NRSP were significantly different, as shown in Table 1. In addition, the values of 32 healthy volunteers are displayed in order to show a reference range of the parameters in our laboratory; this group was not sex- and age-matched with the study population. The distribution of RSP and NRSP patients according to their TAT levels (cutoff 6 ug/1) is shown in Figure 1; the odds ratio was found to be 8.7 (95% confidence interval (CI) 2.7-27.6; P = 0.0014) for patients with TAT «S6 ug/1 to reach remission. The TAT values were significantly higher in the patients with ED than in those with LD (Table 2). Within the ED subgroup, as shown in Table 3, the TAT values were lower in the RSP than in the NRSP patients; this difference was statistically significant despite the low number of RSP in the ED collective. The distribution of RSP and NRSP patients according to their TAT levels Table 2. Activation marker levels (means ± SEM) of the patients with localized disease (LD, n = 53) and those with extensive disease (ED,
Statistical evaluation
The mean values and standard errors of means were calculated for groups according to the histologic types, the tumour stage at the time of diagnosis, the outcome in the entire study population and in the subgroup with ED, as well as for the healthy volunteer group. Differences between the groups were tested for significance by means of the alternate Welch's (-test for different variances, and significance was assumed at P values below 0.05. The distribution of patients with TAT levels above 6 ug/1 into the groups RSP and NRSP was analysed using contingency tables and Fisher's exact test.
Results There was no difference in activation marker levels between patients with different histologic tumour types
NRSP
LD ED
TAT
F1+2
5.26 ± 0.69 /> = 0.0182 10.75 ±2.15
1.28 ±0.09 NS 1.67 ±0.35
Table 3. Activation marker levels (means ± SEM) within the subgroup with extensive disease (ED); patients achieving remission (RSP, n = 6) compared with the non-responders (NRSP, n - 40).
RSP NRSP
TAT
F ,1+2 +
4.65 ± 0.94 P = 0.0092 11.92 ±2.49
0.97 ±0.18 NS 1.78 ±0.41
Downloaded from https://academic.oup.com/annonc/article-abstract/8/8/781/304832 by guest on 19 September 2019
SCLC was defined as tumour with involvement of contralateral lymph nodes and/or metastases to other structures of the thorax and/or distant sites. In NSCLC, the non-resectable stages HIB and IV according to IUCC, were defined as ED. Altogether, the study included 46 ED patients (24 SCLC, six squamous cell, eight adeno, and eight large or mixed cell type). The other patients were assigned to the group of localised disease (LD). All patients received primary standard antineoplastic treatment (typically four to six courses of chemotherapy, surgery or radiation) as appropriate for their histologic tumour type, stage, and their general condition. The outcome was documented by standard restaging procedures [15]; for the purpose of this study, response (RSP) to treatment was defined as achievement of complete or partial remission, and nonresponse (NRSP) as no change, progressive disease or death before therapy could be completed. None of the patients showed clinical signs or symptoms suggestive of recent thromboembolic disorders at the tune of inclusion into the study and blood sampling; diagnostic procedures such as Doppler sonography or phlebography to exclude occult thrombosis were not performed. Blood samples were obtained before the start of antineoplastic treatment; the measured values of coagulation activation did not have any influence on decisions about modalities of tumour therapy. In order to determine in our laboratory reference ranges of the parameters in normal individuals, citrated blood samples were obtained from 32 healthy volunteers.
783
Downloaded from https://academic.oup.com/annonc/article-abstract/8/8/781/304832 by guest on 19 September 2019
ciated with worse outcome after subsequent treatment. However, this finding might have been due to the fact that among the patients with poor outcome there was a higher proportion with metastatic tumour disease and thus higher TAT levels. Therefore, we decided to analyse the subgroup of patients with prognostically unfavourable extensive disease separately. Also in this more homogenous patient group, there was a considerable difference of the TAT values between patients with effective therapy and the others. Remission was achieved byfiveof 24 ED patients with TAT ^ 6 ug/1, but only by 1 of 22 patients with TAT > 6 ug/1. Though this difference was not statistically significant in Fisher's exact test, probably due to the low number of remissions in the ED group, it appears to be remarkable (Figure 2). NRSP RSP Since the tumour extension as could be assessed by Figure 2. Distribution of the patients with extensive disease according standard diagnostic procedures and according to acto their TAT levels (cut-off 6 ug/1) into the outcome categories RSP and cepted staging criteria was comparable within the ED NRSP. group, the higher levels in NRSP can not be explained merely by a more advanced tumour stage at the time of diagnosis and TAT determination. The data suggest (cut-off 6 ug/1) is displayed in Figure 2. The analysis that, in patients with lung cancer, the activation of shows that remission of their metastatic disease was coagulation is an independent prognostic factor. This achieved by 5 of 24 (20%) of patients with TAT ^ 6 ug/1, interpretation is supported by experimental evidence but only by 1 of 22 (4.5%) of patients with TAT > 6 ug/1. pointing to several advantages for the tumour brought The odds ratio was 5.5 (95% CI 0.6-51.7); however, the about by activated coagulation [1, 8, 9]. difference was not significant in Fisher's exact test. It would be a major step forward in the so far often frustrating treatment of lung cancer, if a new, easy-tomeasure prognostic marker could support therapeutic decisions. In two controlled therapy trials, low grade Discussion anticoagulation with warfarin as an adjunct to regular This study confirmed previous results of this and other chemotherapy has been found to prolong survival of groups that activation of coagulation takes place in a small-cell lung cancer (SCLC) patients [10, 11]. This large part of patient with lung cancer of any histologic effect was statistically significant in both trials. Howtype. In a previous paper [12], we had been able to ever, it was not strong enough to prompt the adoption of show that the plasma levels of markers demonstrating this approach as part of the standard treatment of lung thrombin (TAT) and fibrin formations (D-dimers) are cancer. Two aspects might be considered in the interpresignificantly higher in metastatic lung cancer. Since the tation of the two trials: 1) since unselected patients were difference was particularly pronounced for TAT, we treated, it appears likely that also patients without sigdecided to use the markers of thrombin generation TAT nificantly increased coagulation received anticoagulaand Fi+2 for assessment of the prognostic impact of an tion; 2) the warfarin doses were relatively low, possibly activation of coagulation. In this study, 40% of patients due to the fear of severe bleeding in those patients. The showed elevated TAT levels; the TAT levels were much results of the present study raise the possibility that the higher in patients with metastases. This might not only determination of TAT could identify patients, who might be due to their higher tumour mass, leading to a more have an advantage from anticoagulation as an adjunct to pronounced activation of coagulation. It might also standard antineoplastic therapy. If this could be verified indicate that thrombin is mainly formed in the tumour by further studies, TAT and other activation markers tissue and 'leaks out' into the circulation, when the might also be helpful to monitor anticoagulation and tumour is spreading. In this direction points also our find the appropriate individual therapeutic levels. finding that the difference between localised and metaFurther research should evaluate the potential value static disease is much more pronounced for TAT than of diagnostic tests and therapeutic interventions directed for Fi +2 . This suggests that thrombin generation with at the activation of coagulation in the ongoing struggle cleavage of the large peptide prothrombin fragment Fi +2 to improve the outcome of lung cancer patients. might take place predominantly in a different compartment, i.e., tumour tissue, than the formation of complexes of thrombin with its inhibitor antithrombin in Acknowledgement plasma. Already in our previous study [12], we had found that This study was supported by a grant from Deutsche elevated TAT values at the time of diagnosis were asso- Krebshilfe, M 73/92/Se 2.
784 References 11.
12. 13. 14 15.
Received 24 March; accepted 12 June 1997.
Correspondence to: R. Seitz, MD Head, Dept. of Haematology and Transfusion Medicine Paul-Ehrlich-Institute Paul-Ehrlich-Street 51-59 D-63225 Langen Germany
Downloaded from https://academic.oup.com/annonc/article-abstract/8/8/781/304832 by guest on 19 September 2019
1. Dvorak HM. Tumours: Wounds that do not heaL New Engl J Med 1986; 315: 1650-9. 2. Zacharski LR, Wojtukiewicz MZ, Constantini V et al. Pathways of coagulation/fibrinolysis activation in malignancy. Semin Thromb Hemost 1992; 18: 104-16. 3. Adany R, Kappelmayer J, Berenyi E et al. Factors of the extrinsic pathway of blood coagulation in tumor associated macrophages. Thromb Heamostas 1989; 62: 850-5. 4. Seitz R, Heidtmann H-H, Maasberg M et al. Activators of coagulation in cultured human lung tumor cells. Int J Cancer 1993: 53: 514-20. 5. Heinmdller E, Weinel RJ, Heidtmann H-H et al. Studies on tumor-cell-induced platelet aggregation in human lung cancer cell lines. J Cancer Res Clin Oncol 1996; 122: 735-44. 6. Rickles FR, Edwards RL. Activation of blood coagulation in cancer: Trousseau's syndrome revisited. Blood 1983; 62: 14-31. 7. Prandoni P, Lensing AWA, Bailer HR et al. Deep-vein thrombosis and the incidence of subsequent symptomatic cancer. N Engl J Med 1992; 327: 1128-33. 8. Gunji Y, Lewis J, Gorehk E. Fibrin formation inhibits the in vilro cytotoxic activity of human natural and lymphokine-activated killer cells. Blood Coag Fibrinol 1990; 1: 663-72. 9. Bruhn HD, Zuborn KH. Influence of clotting factors (thrombin, factor XIII) and fibronectin on the growth of tumor cells and leukemic cells in vitro. Blut 1983; 46: 85-8. 10. Zacharski LR, Henderson WG, Rickles FR et al. Effect of
warfarin anticoagulation on survival in carcinoma of the lung, colon, head and neck, and prostata. Cancer 1984; 53: 2046-52. Chahinian AP, Propert KJ, Ware JH et al. A randomized trial of anticoagulation with warfarin and of alternating chemotharapy in extensive small-cell lung cancer by the Cancer and Leukemia Group B. J Clin Oncol 1989; 7: 993-1002. Seitz R, Rappe N, Kxaus M et al Activation of coagulation and fibrinolysis in patients with lung cancer: Relation to tumour stage and prognosis. Blood Coag Fibrinol 1993; 4: 249-54. Van Wersch JWJ, Tjwa MKT. Coagulation/fibrinolysis balance and lung cancer. Haemostasis 1991; 21: 117-23. Kempkes-Matthes B, Bleyl H. Factor IXi-antithrombin (IXiAT) and thrombin-antithrombin (TAT) complexes in lung cancer patients. Ann Hematol 1992; 64: 35-9. Havemann K, Wolf M, Holle R. Alternating versus sequential chemotherapy in small cell lung cancer: A randomized German multicenter trial. Cancer 1987; 59: 1072-82.
Original article Prognostic impact of an activation of coagulation in lung cancer* R. Seitz,1 H.-H. Heidtmann, M.Wolf, A. Immel & R. Egbring Philipps University Hospitals, Department of Haematology/Oncology, Marburg, Germany; ^Present address: Department of Haematology and Transfusion Medicine, Paul-Ehrlich-Institute, Langen, Germany
Introduction A growing body of evidence demonstrates that activation of coagulation is involved in fundamental processes during the evolution of cancer tissues, and is of major importance for invasive tumour growth and metastatic spread [1]. The activation of the clotting cascade can be due to activities expressed by the tumour cells themselves, or the stimulation of tumour-associated inflammatory cells, such as macrophages [2, 3]. Several activities in cultured human lung tumour cells [4] may contribute to activation of the tissue factor pathway, or of factors II or X directly. The substrates of coagulation (fibrinogen, clotting factors) are thought to reach the tumour tissue by diffusion after extravasation of plasma, facilitated by vascular permeability factor [1]. Also, platelet activation may be triggered by lung cancer cells via thrombin generation [5]. Well-documented clinical observations [6, 7] have shown an increased risk of thromboembolic complications early in the course of cancer. Formation of a fibrin matrix appears to support tumour growth by promoting neoangiogenesis, and by shielding tumour cells against the attack of immunocompetent cells [8]. Thrombin may even be a growth * Dedicated to Professor KJaus Havemann on the occasion of his 65th anniversary.
factor for tumour cells [9]. In two controlled therapy trials, anticoagulation with warfarin as an adjuvant to chemotherapy was found to prolong survival of small cell lung cancer (SCLC) patients [10,11]. With sensitive molecular markers, in many lung cancer patients sustained activation of haemostasis can be demonstrated [12-14]. We could show [12] that the degree of activation of coagulation is related to initial tumour stage. The present study addresses the question, whether the extent of activation of coagulation, as assessed by pretherapeutic activation marker levels, has an impact on the response to subsequent standard antineoplastic treatment. Patients and methods In this prospective diagnostic study, 99 patients with newly diagnosed, histologically proven lung cancer, or relapse of lung cancer after previously documented complete remission, were evaluated. Patients with malignancies other than lung cancer, with pre-existing coagulation disturbances of other causes, on oral anticoagulant therapy, or who did not give their consent to participate, were not included. The study was approved by the local ethics committee. The histologic type was small cell lung cancer (SCLC) in 41 patients, and non-small cell lung cancer (NSCLQ in 58 patients (29 squamous cell, 12 adeno, 17 large cell and mixed cell type). Complete tumour staging was carried out using standard procedures. In SCLC, the categories of tumour stage were adapted from criteria approved in a previous multicentre chemotherapy trial [15], Extensive disease (ED)
Downloaded from https://academic.oup.com/annonc/article-abstract/8/8/781/304832 by guest on 19 September 2019
30 of 66 (45%) NRSP, but only 4 of 33 (12%) RSP. 88% of patients with TAT <6 ug/1 achieved remission, and 45% with Background: There is evidence that activation of coagulation TAT > 6 ug/1 (P = 0.0014). In the subgroup of 46 patients with by influencing tumour biology may have impact on clinical advanced disease, the six RSP showed lower TAT than the 40 course of lung cancer. NRSP: 4.65 ± 0.94 vs. 11.92 ± 2.49 ug/1 (P < 0.01); one of six Patients and methods: We measured the activation markers (17%) RSP, but 21 of 40 (53%) NRSP showed TAT > 6 ug/1. thrombin-antithrombin complex (TAT) and prothrombin fragThese data suggest that in lung cancer the activation of ment F 1+2 in 99 lung cancer patients immediately after diag- coagulation is an independent prognostic factor, since TAT nosis, before antineoplastic treatment. Outcome was assessed levels were different between RSP and NRSP, also within the at the end of appropriate standard primary therapy (four to six homogeneously unfavourable metastatic subgroup. It should courses of chemotherapy, surgery or radiation). be further studied, whether TAT can identify patients, whose Results and conclusions: The activation markers (means ± prognosis could be improved by anticoagulation as an adjunct SEM) were lower in the 33 responders (RSP; complete or to standard antineoplastic therapy. partial remission) than in the 66 non-responders (NRSP): TAT 3.96 ± 0.48 vs. 9.69 ± 1.57 ug/1 (P < 0.001), and F1+21.09 ± 0.09 Key words: lung cancer, prognosis, thrombin-antithrombin vs. 1.64 ± 0.25 nmol/1 (P < 0.05). TAT levels were > 6 ug/1 in complex Summary
Table 1. Activation marker levels (means ± SEM) of the patients acheiving remission (RSP, n = 33) and the non-responders (NRSP, 7i = 66); for comparison, values of 32 healthy volunteers. TAT RSP
NRSP Volunteers
3.96 ± 0.48 P = 0.0008 9.69 ± 1.57 2.1210.09
1.09 ±0.09 /> = 0.0415 1.64 ±0.25 0.71 ±0.04
Methods
Blood samples were obtained using 10 ml syringes containing one part 3.8% sodium citrate to nine parts blood The citrated blood samples were immediately centrifuged at 1500 g for 15 min at room temperature. Aliquots were frozen at —40 °C until assay. The activation markers thrombin-antithrombin complex (TAT) and prothrombin fragment Fi +2 (F 1+2 ) were determined by ELISA, using kits from Behringwerke (Marburg, Germany).
RSP
Figure 1. Distribution of the entire study patients according to their TAT levels (cut-off 6 ug/1) into the outcome categories RSP and NRSP.
(data not shown). The analysis of the entire study population showed that the values of RSP and NRSP were significantly different, as shown in Table 1. In addition, the values of 32 healthy volunteers are displayed in order to show a reference range of the parameters in our laboratory; this group was not sex- and age-matched with the study population. The distribution of RSP and NRSP patients according to their TAT levels (cutoff 6 ug/1) is shown in Figure 1; the odds ratio was found to be 8.7 (95% confidence interval (CI) 2.7-27.6; P = 0.0014) for patients with TAT «S6 ug/1 to reach remission. The TAT values were significantly higher in the patients with ED than in those with LD (Table 2). Within the ED subgroup, as shown in Table 3, the TAT values were lower in the RSP than in the NRSP patients; this difference was statistically significant despite the low number of RSP in the ED collective. The distribution of RSP and NRSP patients according to their TAT levels Table 2. Activation marker levels (means ± SEM) of the patients with localized disease (LD, n = 53) and those with extensive disease (ED,
Statistical evaluation
The mean values and standard errors of means were calculated for groups according to the histologic types, the tumour stage at the time of diagnosis, the outcome in the entire study population and in the subgroup with ED, as well as for the healthy volunteer group. Differences between the groups were tested for significance by means of the alternate Welch's (-test for different variances, and significance was assumed at P values below 0.05. The distribution of patients with TAT levels above 6 ug/1 into the groups RSP and NRSP was analysed using contingency tables and Fisher's exact test.
Results There was no difference in activation marker levels between patients with different histologic tumour types
NRSP
LD ED
TAT
F1+2
5.26 ± 0.69 /> = 0.0182 10.75 ±2.15
1.28 ±0.09 NS 1.67 ±0.35
Table 3. Activation marker levels (means ± SEM) within the subgroup with extensive disease (ED); patients achieving remission (RSP, n = 6) compared with the non-responders (NRSP, n - 40).
RSP NRSP
TAT
F ,1+2 +
4.65 ± 0.94 P = 0.0092 11.92 ±2.49
0.97 ±0.18 NS 1.78 ±0.41
Downloaded from https://academic.oup.com/annonc/article-abstract/8/8/781/304832 by guest on 19 September 2019
SCLC was defined as tumour with involvement of contralateral lymph nodes and/or metastases to other structures of the thorax and/or distant sites. In NSCLC, the non-resectable stages HIB and IV according to IUCC, were defined as ED. Altogether, the study included 46 ED patients (24 SCLC, six squamous cell, eight adeno, and eight large or mixed cell type). The other patients were assigned to the group of localised disease (LD). All patients received primary standard antineoplastic treatment (typically four to six courses of chemotherapy, surgery or radiation) as appropriate for their histologic tumour type, stage, and their general condition. The outcome was documented by standard restaging procedures [15]; for the purpose of this study, response (RSP) to treatment was defined as achievement of complete or partial remission, and nonresponse (NRSP) as no change, progressive disease or death before therapy could be completed. None of the patients showed clinical signs or symptoms suggestive of recent thromboembolic disorders at the tune of inclusion into the study and blood sampling; diagnostic procedures such as Doppler sonography or phlebography to exclude occult thrombosis were not performed. Blood samples were obtained before the start of antineoplastic treatment; the measured values of coagulation activation did not have any influence on decisions about modalities of tumour therapy. In order to determine in our laboratory reference ranges of the parameters in normal individuals, citrated blood samples were obtained from 32 healthy volunteers.
783
Downloaded from https://academic.oup.com/annonc/article-abstract/8/8/781/304832 by guest on 19 September 2019
ciated with worse outcome after subsequent treatment. However, this finding might have been due to the fact that among the patients with poor outcome there was a higher proportion with metastatic tumour disease and thus higher TAT levels. Therefore, we decided to analyse the subgroup of patients with prognostically unfavourable extensive disease separately. Also in this more homogenous patient group, there was a considerable difference of the TAT values between patients with effective therapy and the others. Remission was achieved byfiveof 24 ED patients with TAT ^ 6 ug/1, but only by 1 of 22 patients with TAT > 6 ug/1. Though this difference was not statistically significant in Fisher's exact test, probably due to the low number of remissions in the ED group, it appears to be remarkable (Figure 2). NRSP RSP Since the tumour extension as could be assessed by Figure 2. Distribution of the patients with extensive disease according standard diagnostic procedures and according to acto their TAT levels (cut-off 6 ug/1) into the outcome categories RSP and cepted staging criteria was comparable within the ED NRSP. group, the higher levels in NRSP can not be explained merely by a more advanced tumour stage at the time of diagnosis and TAT determination. The data suggest (cut-off 6 ug/1) is displayed in Figure 2. The analysis that, in patients with lung cancer, the activation of shows that remission of their metastatic disease was coagulation is an independent prognostic factor. This achieved by 5 of 24 (20%) of patients with TAT ^ 6 ug/1, interpretation is supported by experimental evidence but only by 1 of 22 (4.5%) of patients with TAT > 6 ug/1. pointing to several advantages for the tumour brought The odds ratio was 5.5 (95% CI 0.6-51.7); however, the about by activated coagulation [1, 8, 9]. difference was not significant in Fisher's exact test. It would be a major step forward in the so far often frustrating treatment of lung cancer, if a new, easy-tomeasure prognostic marker could support therapeutic decisions. In two controlled therapy trials, low grade Discussion anticoagulation with warfarin as an adjunct to regular This study confirmed previous results of this and other chemotherapy has been found to prolong survival of groups that activation of coagulation takes place in a small-cell lung cancer (SCLC) patients [10, 11]. This large part of patient with lung cancer of any histologic effect was statistically significant in both trials. Howtype. In a previous paper [12], we had been able to ever, it was not strong enough to prompt the adoption of show that the plasma levels of markers demonstrating this approach as part of the standard treatment of lung thrombin (TAT) and fibrin formations (D-dimers) are cancer. Two aspects might be considered in the interpresignificantly higher in metastatic lung cancer. Since the tation of the two trials: 1) since unselected patients were difference was particularly pronounced for TAT, we treated, it appears likely that also patients without sigdecided to use the markers of thrombin generation TAT nificantly increased coagulation received anticoagulaand Fi+2 for assessment of the prognostic impact of an tion; 2) the warfarin doses were relatively low, possibly activation of coagulation. In this study, 40% of patients due to the fear of severe bleeding in those patients. The showed elevated TAT levels; the TAT levels were much results of the present study raise the possibility that the higher in patients with metastases. This might not only determination of TAT could identify patients, who might be due to their higher tumour mass, leading to a more have an advantage from anticoagulation as an adjunct to pronounced activation of coagulation. It might also standard antineoplastic therapy. If this could be verified indicate that thrombin is mainly formed in the tumour by further studies, TAT and other activation markers tissue and 'leaks out' into the circulation, when the might also be helpful to monitor anticoagulation and tumour is spreading. In this direction points also our find the appropriate individual therapeutic levels. finding that the difference between localised and metaFurther research should evaluate the potential value static disease is much more pronounced for TAT than of diagnostic tests and therapeutic interventions directed for Fi +2 . This suggests that thrombin generation with at the activation of coagulation in the ongoing struggle cleavage of the large peptide prothrombin fragment Fi +2 to improve the outcome of lung cancer patients. might take place predominantly in a different compartment, i.e., tumour tissue, than the formation of complexes of thrombin with its inhibitor antithrombin in Acknowledgement plasma. Already in our previous study [12], we had found that This study was supported by a grant from Deutsche elevated TAT values at the time of diagnosis were asso- Krebshilfe, M 73/92/Se 2.
784 References 11.
12. 13. 14 15.
Received 24 March; accepted 12 June 1997.
Correspondence to: R. Seitz, MD Head, Dept. of Haematology and Transfusion Medicine Paul-Ehrlich-Institute Paul-Ehrlich-Street 51-59 D-63225 Langen Germany
Downloaded from https://academic.oup.com/annonc/article-abstract/8/8/781/304832 by guest on 19 September 2019
1. Dvorak HM. Tumours: Wounds that do not heaL New Engl J Med 1986; 315: 1650-9. 2. Zacharski LR, Wojtukiewicz MZ, Constantini V et al. Pathways of coagulation/fibrinolysis activation in malignancy. Semin Thromb Hemost 1992; 18: 104-16. 3. Adany R, Kappelmayer J, Berenyi E et al. Factors of the extrinsic pathway of blood coagulation in tumor associated macrophages. Thromb Heamostas 1989; 62: 850-5. 4. Seitz R, Heidtmann H-H, Maasberg M et al. Activators of coagulation in cultured human lung tumor cells. Int J Cancer 1993: 53: 514-20. 5. Heinmdller E, Weinel RJ, Heidtmann H-H et al. Studies on tumor-cell-induced platelet aggregation in human lung cancer cell lines. J Cancer Res Clin Oncol 1996; 122: 735-44. 6. Rickles FR, Edwards RL. Activation of blood coagulation in cancer: Trousseau's syndrome revisited. Blood 1983; 62: 14-31. 7. Prandoni P, Lensing AWA, Bailer HR et al. Deep-vein thrombosis and the incidence of subsequent symptomatic cancer. N Engl J Med 1992; 327: 1128-33. 8. Gunji Y, Lewis J, Gorehk E. Fibrin formation inhibits the in vilro cytotoxic activity of human natural and lymphokine-activated killer cells. Blood Coag Fibrinol 1990; 1: 663-72. 9. Bruhn HD, Zuborn KH. Influence of clotting factors (thrombin, factor XIII) and fibronectin on the growth of tumor cells and leukemic cells in vitro. Blut 1983; 46: 85-8. 10. Zacharski LR, Henderson WG, Rickles FR et al. Effect of
warfarin anticoagulation on survival in carcinoma of the lung, colon, head and neck, and prostata. Cancer 1984; 53: 2046-52. Chahinian AP, Propert KJ, Ware JH et al. A randomized trial of anticoagulation with warfarin and of alternating chemotharapy in extensive small-cell lung cancer by the Cancer and Leukemia Group B. J Clin Oncol 1989; 7: 993-1002. Seitz R, Rappe N, Kxaus M et al Activation of coagulation and fibrinolysis in patients with lung cancer: Relation to tumour stage and prognosis. Blood Coag Fibrinol 1993; 4: 249-54. Van Wersch JWJ, Tjwa MKT. Coagulation/fibrinolysis balance and lung cancer. Haemostasis 1991; 21: 117-23. Kempkes-Matthes B, Bleyl H. Factor IXi-antithrombin (IXiAT) and thrombin-antithrombin (TAT) complexes in lung cancer patients. Ann Hematol 1992; 64: 35-9. Havemann K, Wolf M, Holle R. Alternating versus sequential chemotherapy in small cell lung cancer: A randomized German multicenter trial. Cancer 1987; 59: 1072-82.