Multiple myeloma (MM) is a genetically heterogeneous tumour of transformed plasma cells, terminally differentiated effectors of the B cell lineage, specialized in producing large amounts of monoclonal immunoglobulins (M-protein) [1]. Irrespectively of typical symptoms, MM patients present a broad spectrum of uncommon disease signs, among which coagulation abnormalities constitute an important clinical problem [2]. Unlike venous thromboembolism, which is relatively frequent, major bleeding remains uncommon, ranging from 0-7% in patients newly diagnosed with MM [3,4]. Due to its infrequency, the management of bleeding complications is often symptomatic and based on standard recommendations. Due to complex pathogenesis of the bleeding predisposition in MM, the lack of access to experienced hemostasis laboratory differential diagnosis remains challenging [5]. Until today, only a few cases of bleeding manifestation of relapsed MM have been reported [6,7].

Herein, we present a case of 53-years-old Caucasian male patient with symptomatic IgGλ multiple myeloma, stage IIIA according to Durie-Salmon classification. The initial treatment with thalidomide and dexamethasone (TD) resulted in a very good partial response (VGPR). At the time of autologous hematopoietic stem cell transplantation (autoHSCT), routine coagulation tests were normal. After autoHSCT he did not receive any maintenance therapy. The patient was readmitted to our department with malaise and abdominal pain of moderate intensity two years after autoHSCT. During hospitalization the pain deteriorated, peritoneal pressure symptoms became positive and nasal bleeding appeared. The abdominal ultrasonography (USG) showed a little amount of fluid in the right lower part of the peritoneal cavity and thickening of the intestinal wall. Computed tomography image showed enlargement of the right iliopsoas muscle. At the time of the admission, the laboratory tests revealed high concentration of monoclonal immunoglobulin G (32 g/l, 91.6% of total IgG) in the serum. The coagulation tests results showed prolongation of activated partial thromboplastin time (aPTT), prothrombin time (PT) and thrombin time (TT). The PT and the aPTT were not corrected with an equal volume plasma mixing test, which is indicative for circulating anticoagulant. Additionally performed coagulation tests confirmed significantly decreased factor V procoagulant activity and the presence of factor V inhibitor (5 Bethesda Units/ml). Screening tests for lupus-like inhibitor (LA) performed according to the manufacturer’s instruction (Lupus anticoagulant testing, Siemens) were negative (LA1/LA2-1.01; aPTT with Actin-32.7 sec.). Based on both the clinical signs of appendicitis, a decision to perform an appendectomy was made. During the procedure, a retroperitoneal hematoma was found and evacuated. There were no signs of appendicitis. After treatment with tranexamic acid and fresh frozen plasma the bleeding subsided. Thereafter, due to disease progression, six cycles of bortezomib, doxorubicin, and dexamethasone (PAD) were given. During the PAD treatment, the monoclonal IgG concentration in patient’s serum significantly decreased (up to 1.3 g/l). Finally, also normalization of the results of aPTT, PT and TT was stated. Therefore, the presence of factor V inhibitor in monoclonal immunoglobulin fraction was suspected. To confirm this, the retrospective, more detailed coagulation system analysis was undertaken. Activity of factor II, V, VIII and X was determined with the help of coagulometric methods and Helena BioSciences reagents. Coagulation system ex vivo studies were performed in normal human plasma after addition of lyophilized monoclonal IgG, isolated with the help of thiophilic adsorption method from 20 ml of the patient’s blood collected at the time admission [8-10]. Thereafter, the activity of factor II, factor V and aPTT, PT and TT in the prepared plasma samples (a series of dilutions of monoclonal IgG in standard human plasma at final concentration of 0, 0.468, 0.937, 1.875, 3.75, 7.5, 15 and 30 g/l, respectively) were measured. Results of the coagulation system evaluation before and during the treatment with PAD are presented in Table 1. The results of the performed ex vivo studies showed close relationship between the monoclonal immunoglobulin G plasma content and the degree of prolongation of PT and aPTT (Figure 1).

Table 1: The results of coagulation system evaluation in patient with recurrent multiple myeloma and retroperitoneal hematoma during treatment with bortezomib - based therapy. Bottom part of the Table shown the results of periodic monitoring of monoclonal protein concentration in the serum.

Figure 1: The results of ex vivo studies evaluating the relationships between the prolongation of activated partial thromboplastin time, prothrombin time and factor V procoagulant activity (Hemosil Instrumentation Laboratory reagents) and the monoclonal immunoglobulin G concentration (protein M isolated from multiple myeloma patient’s blood) in the control plasma samples.

Interestingly, it matched those obtained in vivo . This may prove the concept that the monoclonal immunoglobulin G produced by multiple myeloma cells may act as a factor V inhibitor.

The impact of isolated M protein on factor II procoagulant activity was also studied in the ex vivo assay. The decrease of factor II activity was only observed at very high concentration of M protein in the plasma sample. In our opinion, this phenomenon is the effect of inhibition of factor V by M protein, which is a very important player in the prothrombinase complex. To support this hypothesis, additional ex vivo studies were performed, examining the inhibitory effect of M protein on thrombin time. They confirmed only a slight prolongation of thrombin time at high concentration of M protein in the plasma sample (Supplementary Figure 1). Also 2D crossed electrophoresis studies did not confirm M protein and thrombin (bovine) cross reactivity (data not shown).

Patients with MM are a high-risk group as far as the emergence of unexpected events is concerned, and therefore there very frequently arises the need for immediate medical intervention [3]. In the presented case, USG imaging was not able to identify the bleeding episode, in contrast to angio-CT. However, it should be remembered that the diagnostic algorithm of hemoglobin and hematocrit drop in a patient with abdominal pain requires a customized procedure, including detailed laboratory evaluation of hemostasis

The pathogenesis of the observed clotting abnormalities in the presented patient is probably complex [11,12]. The increased fibrinogen concentration at the disease relapse is the result of disease activity and/or the coexistence of damaged muscle tissue. The high blood concentration of the M protein, with its impact on fibrin polymerization process, may be the cause of the prolongation of not only aPTT and PT, but also TT [13]. Interestingly, the results of equal volume plasma mixing test, before and during PAD treatment, showed no significant correction of the TT value in the beginning. However, later rapid normalization of abnormal test value was observed. The instant shortening of the TT right after the dilution of the M protein plasma content with fresh frozen plasma, supports the thesis that high concentration of monoclonal protein nonspecifically inhibits the fibrinogen polymerization. The value of PT and the aPTT was normalized when the factor V activity reached the level of 35%. The time for normalization of the PT and the aPTT was longer than in the case of TT. There was, however, close association between PT and aPTT values and the concentration of monoclonal immunoglobulin in the patient’s serum. Also, the results of periodically performed equal volume mix tests in case of PT and aPTT documented the normalization of abnormal coagulation tests results when chemotherapy related reduction of the monoclonal IgG serum content was noted. These facts, along with the result of the inhibitor titer measurement (Bethesda assay), suggested the presence of cross reacting factor V inhibitor in monoclonal immunoglobulin fraction. This was confirmed by our ex vivo studies.

To the best of our knowledge, the literature provides for only one reported case of factor V inhibitor in a MM patient with inherited factor V deficiency. Interestingly, the inhibitor completely disappeared after successful treatment with thalidomide [14]. This supports our thesis that protein M may act as a clotting factor inhibitor. Therefore, in our opinion it seems to be reasonable to recommend a detailed coagulation evaluation in MM patients with bleeding complications and prolongation of screening coagulation tests.

Szczepaniak Tomasz: SC in Clinical Trials ran by Janssen-Cilag, Amgen, Onyx; received a fee for hematological workshop in September 2015 from Celgene; Dytfeld Dominik: Co-Investigator in Clinical Trials ran by Janssen-Cilag, Amgen, Onyx; received a grant from Janssen, Advisory Board Member: Celgene, Amgen, Janssen, Novartis; Komarnicki Mieczyslaw: Principal Investigator in Clinical Trials ran by Janssen-Cilag, Amgen, Onyx

The other authors declare that they have no competing interests.

This study was performed in agreement with the Declaration of Helsinki and approved by the local Ethical Committee (Poznan University of Medical Sciences).