The risk of venous thromboembolic disease in patients with monoclonal gammopathy of undetermined significance

Transcription

1 Original article Annals of Oncology 15: , 2004 doi: /annonc/mdh385 The risk of venous thromboembolic disease in patients with monoclonal gammopathy of undetermined significance S. Sallah 1 *, A. Husain 1, J. Wan 2, P. Vos 3 & N. P. Nguyen 4 1 Thrombosis and Hemostasis Program and Feist-Weiller Cancer, Louisiana State University Health Sciences Center, Shreveport, LA; 2 Department of Biostatistics, University of Tennessee, Memphis, TN; 3 Department of Epidemiology, East Carolina University, NC; 4 Radiation Oncology, Southwestern University, Dallas, TX, USA Received 10 March 2004; revised 11 May 2004; accepted 13 May 2004 Background: Recent evidence indicates that patients with multiple myeloma receiving combination chemotherapy containing thalidomide are at a significantly high risk for venous thromboembolic disease (VTD). However, information on the occurrence of VTD in a related disorder, benign monoclonal gammopathy of undetermined significance (MGUS), is limited. Patients and methods: We prospectively investigated patients with MGUS for the occurrence of VTD. The diagnosis of MGUS was based on well known criteria for the disorder. The variables examined were sex, age, race, presence of underlying conditions, level and type of immunoglobulin [serum monoclonal (M)-protein] platelet counts and level of fibrinogen. Results: Of a total of 310 patients with MGUS, 19 (6.1%) developed VTD after a median follow-up of 44 months (range months). In a univariate analysis, age >_65 years (P = 0.01), M-protein >_ 16 g/l (P = 0.001) and progression to plasma cell or lymphoproliferative disorders (28 of 310 patients; P = 0.001) were significant risk factors for VTD. In multivariate analysis, M-spike >_ 16 g/l [risk ratio (RR) = 6.3; 95% confidence interval (CI) ; P = 0.001] and future development of plasma cell or lymphoproliferative disorder (RR = 4.2; 95% CI ; P = 0.003) were variables strongly associated with the occurrence of VTD. A total of 46 patients (14.8%) died during the follow-up period of the study. Conclusion: This study demonstrates that patients with MGUS are at increased risk for VTD. Although a clear reason for the pre-thrombotic state in these patients is not currently evident, few risk factors were identified in the group of patients examined. Key words: hypercoagulable state, monoclonal gammopathy of undetermined significance, multiple myeloma, venous thrombosis Introduction Increasing evidence has recently emerged about the risk of venous thromboembolic disease (VTD) in patients with multiple myeloma [1, 2]. The hypercoagulable state associated with myeloma appears to be directly related to the use of thalidomide in combination with chemotherapy [1 3]. The incidence of VTD in patients with multiple myeloma has been estimated at 28% when using such combination regimens, compared with 7% with thalidomide and dexamethasone alone [1 3]. While the reason for the predisposition to thrombosis in this case is not clear, a recent report suggests that elevated levels of factor VIII may be a contributing factor in myeloma patients receiving thalidomide in combination with adriamycin and dexamethasone [3]. *Correspondence to: Prof. S. Sallah, LSU Health Sciences Center, 1501 Kings Highway, Shreveport, LA 71103, USA. Tel: ; Fax: ; reccos42000@yahoo.com Monoclonal gammopathy of undetermined significance (MGUS) is a well known condition that predisposes patients for the development of multiple myeloma, with risk approaching 25% at 20 years of follow-up [4]. Since patients with MGUS do not receive any specific treatment directed at the M-spike and because the condition is considered pre-malignant, we hypothesized that these patients may inherently be at high risk for venous thrombosis. The current study was aimed at defining the incidence of VTD in patients with MGUS. Patients and methods Definitions Between 1996 and 2003, newly diagnosed patients with MGUS were enrolled onto this prospective study. Diagnosis of MGUS was based on the presence of serum monoclonal (M)-protein at a concentration of <30 g/l, minimal or no M-protein in the urine, absence of any lytic lesions on bone survey, and <10% plasma cells upon examination of bone marrow [5]. Patients with evidence or history of multiple myeloma, amyloidosis or q 2004 European Society for Medical Oncology

2 1491 lymphoproliferative disorders were excluded. The diagnosis of VTD was confirmed using duplex ultrasound and/or venography. A mismatched defect on ventilation/perfusion scan and/or positive spiral tomogram and/or positive angiogram were required for the diagnosis of pulmonary embolism. All underlying conditions were recorded. All patients signed a consent form and Institutional Review Board approval was obtained. Blood samples The presence of serum M-protein was detected using agarose gel electropheresis. Immunofixation was used to determine the type of M-protein [6]. Both electropheresis and immunofixation were performed upon diagnosis on a 24-h collection of urine specimens. Complete blood count, chemistry and fibrinogen were obtained at the time of inclusion. All measurements were repeated every 3 months. Statistical consideration The variables examined in this study included sex, age, race, presence of underlying conditions, level and type of M-protein, platelet counts and level of fibrinogen. The cut-off values for these variables were chosen arbitrarily. Chi-square test or Fisher s exact test was used to compare variables such as race and size of immunoglobulin between patients with and without VTD. Factors found to be associated with the occurrence of VTD in univariate analysis were included in multivariate Cox regression analysis. The Kaplan Meier method was used to calculate time to VTD. Results Characteristics of patients The median age of the 166 men and 144 women enrolled onto this study was 67.5 years (range years). The median follow-up for all patients was 44 months (range months). The diagnosis of MGUS was established in 41 patients while they were hospitalized or undergoing surgical procedures. The work-up for MGUS was performed in the outpatient setting for 269 patients. While the majority of patients were referred to the hematology service from other specialties, all patients were followed at our institution. A total of 19 patients (6.1%) with MGUS developed VTD during the study period (1.6 per 100 patient-years). The diagnosis of VTD in all patients was established in the outpatient setting. The characteristics of the patients with and without VTD are shown in Table 1. The median time from enrollment until the diagnosis of VTD was 23 months (range months). The Kaplan Meier curve for the time to development of VTD is depicted in Figure 1. There were 23 episodes of lower extremity deep venous thrombosis. Two of these patients were also diagnosed with pulmonary embolism. All the thrombotic episodes were diagnosed in the outpatient setting. None of these events were fatal. Recurrence of VTD was diagnosed in two patients (10.5%). A circumstantial risk factor was identified in four patients: VTD followed a motor vehicle accident in one patient, cholecystectomy in one patient, and congestive heart failure and pneumonia in two patients. No recognizable precipitating factor for VTD was found in the remaining 15 patients. None of the patients diagnosed with VDT was Table 1. Characteristics of 310 patients with MGUS included in this study Total number of patients (%) Patients with VTD Men 166 (53.5) 12 Women 144 (46.5) 7 Age (years) < (38) 2 (10.5) >_ (62) 17 (89.5) Race White 135 (43.5) 8 (42) Black 166 (53.5) 11 (58) Others 9 (3) 0 Underlying conditions None 231 (74.5) 14 (73.7) Autoimmune 24 (8) 2 (10.5) Liver disease 19 (6) 1 (5.3) HIV 9 (2) 0 Others 27 (8.5) 2 (10.5) Level of Ig (g/l) < (73) 6 (36.9) >_ (27) 13 (63.1) Type of Ig IgG 219 (70.5) 13 (68.4) IgM 47 (15.5) 4 (21) IgA 39 (12.5) 1 (5) (5.3) Biclonal 5 (1.5) 1 (5) (5.3) Platelet count (per ml) < (76) 14 (73.7) >_ (24) 5 (26.3) Fibrinogen (mg/dl) < (79) 15 (79) >_ (21) 4 (21) MGUS, benign monoclonal gammopathy of undetermined significance; VTD, venous thromboembolic disease; HIV, human immunodeficiency virus; Ig, immunoglobulin. Figure 1. Kaplan Meier estimate of time to development of venous thromboembolic disease in 310 patients with monoclonal gammopathy of undetermined significance.

3 1492 receiving cytotoxic treatment, corticosteroids or immunosuppressent therapy. All patients received anticoagulation with heparin or low weight molecular heparin followed by 6 months of warfarin targeted at an International Normalized Ratio of 2 3. The two patients with recurrence of VTD were placed on long-term anticoagulation with low molecular weight heparin, without subsequent recurrence. Univariate and multivariate analyses Table 2 shows the results of a comparison between the patients with and without VTD. Predominantly, the patients that were diagnosed with VTD were >65 years of age (P = 0.01). Although more African-Americans than Caucasians were identified with MGUS, race was not a significant risk factor for the development of VTD in this study. There were 24 patients with autoimmune disorders, including systemic lupus erythematosous (five patients), rheumatoid arthritis (five patients), fibromyalgia (five patients), vasculitis (two patients), scleroderma (two patients), thyroiditis (one patient) and polyarteritis (one patient). The etiology of liver disease in Table 2. Comparison of patients with and without VTD VTD P-value Relative risk Yes No (95% confidence interval) Number of patients Age (years) < ( ) >_ Race White ( ) Black Underlying condition Non identified ( ) Known disease 5 74 Type of Ig IgG ( ) IgA or IgM 6 85 Level of Ig (g/l) < ( ) >_ Platelet count (per ml) < ( ) >_ Fibrinogen (mg/dl) < ( ) >_ Progression to malignancy Yes ( ) No VTD, venous thromboembolic disease; Ig, immunoglobulin. 19 patients was cirrhosis in 10, hepatitis C in five, chronic active hepatitis in two and unknown in two. Other conditions included sarcoidosis (seven patients), Helicobacter pylori (seven patients), chronic cytomegalovirus infection (five patients), dermatologic disease (four patients) and chronic granulomatous disease (four patients). The presence of an underlying disorder was not a significant risk factor for the occurrence of VTD. At multivariate analysis, serum level >_ 16 g/l [risk ratio (RR) 6.3; 95% confidence interval (CI) ; P = 0.001] and future progression to lymphoproliferative disorders including myeloma and amyloidosis (RR 4.2; 95% CI ; P = 0.003) were associated with increased risk for VTD. Progression of MGUS Increase in the initial level of M-protein during the study period was observed in nine patients (3%). The level of M-protein in these patients remained at <30 g/l and none of them had evidence of a lymphoproliferative disorder or amyloidosis. The median time for the development of a lymphoproliferative disease was 36.4 months (range months). Among the 310 patients with MGUS, 28 patients (9%) developed lymphoproliferative disorders or amyloidosis. The breakdown of these conditions included multiple myeloma in 17 patients (61%), Waldenstrom s macroglobulinemia in three (11%), amyloidosis in three (11%) and lymphoma in five (17%). The median time for progression to these disorders was 36.4 months (range months). A total of 46 patients (14.8%) died during the follow-up period of this study. Of these patients, 24 (52%) died of progressive cardiovascular disorders including congestive heart failure and coronary artery disease, and 13 (28%) died of cerebrovascular disease. Renal failure was the cause of death in three patients (6.5%), while sepsis occurred in two patients and multiple myeloma and lymphoma were considered the cause of death in four patients (8.5%). The cause of death was not known in two patients. Discussion Malignant conditions have long been recognized as a strong risk factor for VTD [7, 8]. Although the predisposition for thrombosis varies among malignancies, an estimated incidence of 7.8% VTD in patients with solid tumors has been reported [9]. Several authors have recently reported a much higher incidence of VTD in patients with multiple myeloma. In studies by Zangari et al., the incidence of VTD in myeloma patients receiving thalidomide, adriamycin and dexamethasone was 28% [1, 2]. An incidence of 35% VTD was observed by Minnema and colleagues in a cross-sectional study of 20 patients with refractory or relapsed multiple myeloma [3]. Although the risk of VTD appears to be directly related to the use of thalidomide in combination with other cytotoxic agents (mainly adriamycin), the latter study demonstrated that this risk was present, although to a lesser degree, in patients receiving thalidomide with dexamethasone only [3]. Advanced

4 1493 age and the presence of chromosome 11 abnormalities were found to be the two leading factors strongly associated with thrombosis in patients with multiple myeloma and VTD [2]. While one could only speculate on the reasons for the intense hypercoagulable state in patients with myeloma after the administration of thalidomide and adriamycin, several factors have been implicated. For example, Minnema and colleagues have observed elevated levels of plasma factor VIII in the majority of these patients [3]. Also, angiogenic stimuli, cytokine release and activated protein C resistance were proposed as possible risk factors for thrombosis in myeloma patients [1, 2, 10 12]. The hallmark of patients with MGUS is the presence of a clone of plasma cells that produce one specific type of protein. By definition, at the time of diagnosis of MGUS, no clinical, biochemical or histological evidence of a malignant process can be identified. Because MGUS is potentially a premalignant condition, we hypothesized that an increased risk for thrombosis may exist regardless of whether a transformation to myeloma occurs or not. In the current investigation, an incidence of 6.1% of VTD was observed in 310 patients with MGUS followed for a median of 44 months. The size of immunoglobulin >16 g/l and future progression to multiple myeloma or other lymphoproliferative disorders emerged as risk factors for VTD in the multivariate analysis. The relative merits of these findings can be best addressed in the context of the natural history of patients with MGUS. For example, it has been shown by Kyle et al. that the size of the M-protein is the most important factor in predicting the transformation to a plasma cell disorder [4]. Therefore, a high concentration of M-protein is clearly linked to both an increased risk for VTD and further progression of the disease. We recognize that in the absence of a controlled arm in the current study, it is not possible to draw a definitive conclusion on the magnitude of risk of VTD in MGUS or, specifically, the relative risk of thrombosis compared with the general population. On the other hand, comparative observations may help to illustrate this risk. For example, the risk of VTD in the general population in our institution is similar to the risk reported for the general population in the United States and approximates 1/1000/year [13]. The observed incidence of VTD in individuals >_65 years of age referred or diagnosed in our institution is 0.9% per year. This is not dissimilar to the rate of VTD reported in persons between 65 and 75 years old of % per year [13]. Finally, the rate of VTD in the current study is slightly lower than the 7.8% (1.8% per year) incidence of VTD in cancer patients that we have recently reported [9]. It is well known that angiogenesis, an active process in myeloma patients, is present, albeit to a lesser degree, in MGUS [11, 12]. The increased secretion of pro-inflammatory cytokines such as interleukin-6 (IL-6) and tissue necrosis factor (TNF) has also been demonstrated in patients with MGUS [11]. While IL-6 is a major growth factor for myeloma cells, both IL-6 and TNF can activate the coagulation and fibrinolytic pathways. Also, chromosome 11 abnormalities and translocation (11;14) in particular can be found in up to 20% of patients with MGUS [14, 15]. These abnormalities have been strongly associated with an increased risk for VTD in multiple myeloma [2]. Finally, it has long been recognized that monoclonal proteins may interfere with fibrin structure leading to the formation of thin gel fibers and an alteration in clot retraction [16, 17]. These changes may cause the clots that are formed in patients with MGUS to be resistant to lysis. While these observations do not necessarily provide an explanation for the increase in thrombotic complications in patients with MGUS, one may speculate that a host of factors are responsible in this case. It is quite conceivable that abnormalities in the stromal cell environment coupled with the effect of the monoclonal protein on fibrin formation may potentiate a thrombogenic state in MGUS. The role of known hereditary and acquired defects predisposing to hypercoagulability is still being examined in this population of patients. None of the other variables examined in the univariate analysis, such as type of underlying illness, platelet count or fibrinogen level, were associated with an increase risk for VTD in this study. These findings do not favor a role for chronic inflammation as a precipitating event for clotting in this group of patients. Collectively, our results indicate that, not unlike patients with multiple myeloma, patients with MGUS are at increased risk for VTD. Also not unlike myeloma, the thrombotic episodes observed in patients with MGUS are very well managed with anticoagulant therapy. A key concept to keep in mind when examining the findings of our study is the view of MGUS and multiple myeloma as a continuum rather than as separate entities. This is true in at least 25% of patients with MGUS who progress to a plasma cell disorder. The implication here is two-fold. First, some patients with MGUS, e.g. those with a high concentration of M-protein, are at increased risk both for VTD and progression to myeloma. Secondly, and most importantly, it is possible that a diagnosis of VTD in a patient with MGUS may in fact portend a higher risk for transformation to myeloma in the future. In summary, this is the first study that provides evidence for the presence of a hypercoagulable state in MGUS. Future long-term studies are important to confirm our findings. In particular, research efforts should be directed toward resolving the mechanism for the predisposition to thrombosis in both MGUS and multiple myeloma. References 1. Zangari M, Anaissie E, Barlogie B et al. Increased risk of deep vein thrombosis in patients with multiple myeloma receiving thalidomide and chemotherapy. Blood 2001; 98: Zangari M, Siegel E, Barlogie B et al. Thrombogenic activity of doxorubicin in myeloma patients receiving thalidomide: implications for therapy. Blood 2002; 100: Minnema MC, Fijnheer R, De Groot PG, Lokhorst HM. Extremely high levels of von Willebrand factor antigen and of procoagulant factor VIII found in multiple myeloma patients are associated with

5 1494 activity status but not with thalidomide treatment. J Thromb Haemost 2003; 1: Kyle RA, Therneau TM, Rajkumar SV et al. A long-term study of prognosis in monoclonal gammopathy of undetermined significance. N Engl J Med 2002; 346: Keren DF, Alexanian R, Goeken JA et al. Guidelines for clinical and laboratory evaluation of patients with monoclonal gammopathies. Arch Pathol Lab Med 1999; 123: Kyle RA. Sequence of testing for monoclonal gammopathics. Arch Pathol Lab Med 1999; 123: Trousseau A. Phlegmasia alba dolens. Clinique Medicale de L Hotel- Dieu de Paris. London: New Sydenham Society Donati MB, Possi A. Malignancy and haemostasis. Br J Haematol 1983; 44: Sallah S, Wan JY, Nguyen NP. Venous thrombosis in patients with solid tumors: determination of frequency and characteristics. Thromb Haemost 2002; 87: Zangari M, Saghafifar F, Anaissie E et al. Activated protein C resistance in the absence of factor V Leiden mutation is a common finding in multiple myeloma and is associated with an increased risk of thrombotic complications. Blood Coagul Fibrinolysis 2003; 13: Rajkumar SV, Mesa RA, Fonesca R et al. Bone marrow angiogenesis in 400 patients with monoclonal gammopathy of undetermined significance, multiple myeloma, and primary amyloidosis. Clin Cancer Res 2002; 8: Tricot G. New insights into role of microenvironment in multiple myeloma. Lancet 2000; 355: Silverstein MD, Heit JA, Mohr DN et al. Trends in the incidence of deep vein thrombosis and pulmonary embolism: a 25-year populationbased study. Arch Intern Med 1998; 158: Fonesca R, Bailey RJ, Ahmann GJ et al. Genomic abnormalities in monoclonal gammopathy of undetermined significance. Blood 2003; 100: Avet-Loiseau H, Li JY, Facon T et al. High incidence of translocations t(11;14) (q13;q32) and (4;14) (p16;q32) in patients with plasma cell malignancies. Cancer Res 1998; 58: Gabriel DA, Smith LA, Folds JD et al. The influence of immunoglobulin (IgG) on the assembly of fibrin gels. J Lab Clin Med 1983; 101: Carr ME, Zebkert SL. Abnormal clot retraction, altered fibrin structure and normal platelet function in multiple myeloma. Am J Physiol 1994; 266: