Marginal zone lymphoma-associated antiphospholipid antibodies successfully treated with bendamustine rituximab

Abstract

A 46-year-old man presented with splenomegaly, abdominal adenopathy and profoundly elevated prothrombin time and partial thromboplastin time. He was diagnosed with marginal zone lymphoma (MZL) and small lymphocytic lymphoma, and the abnormal coagulation studies were secondary to the presence of a lupus anticoagulant. Optimal upfront therapy for MZL has not been established, and the incidence of antiphospholipid antibodies (APLA) in this patient population is rare. Following treatment with six cycles of bendamustine and rituximab with 2 years of rituximab maintenance, our patient remained in remission and his coagulation studies normalised. This report describes a case of successful treatment of APLA associated with MZL that resolved after treatment of the lymphoma.

Background

Antiphospholipid antibodies (APLA), including lupus anticoagulant (LA), anti-B2-glycoprotein I (B2GPI) and anticardiolipin (aCL) antibodies, are a heterogeneous group of autoantibodies directed against phospholipid binding proteins.¹ They are well known as risk factors for thrombophilia.² However, in vitro LA interferes with the binding of clotting factors on the phospholipid surface in the test tube, thus prolonging the activated partial thromboplastin time (aPTT) without correction on mixing studies.³ The association of APLA and malignancy is a well-documented phenomenon.⁴ The prevalence of APLA was increased in cancer patients in one case-control study,⁵ three prospective cohort studies^6–8 and three retrospective studies.^9–11 Most of these studies also demonstrated an increase in thromboembolic events in this patient population, and several studies have shown that treatment of the underlying malignancy often leads to normalisation of APLA.¹² We present a case of newly diagnosed symptomatic marginal zone lymphoma (MZL) and indolent small lymphocytic lymphoma (SLL) with profoundly elevated prothrombin time (PT) and PTT with strongly positive LA, aCL IgM and anti-beta 2 glycoprotein IgM.

Case presentation

A 46-year-old African–American man with no significant medical history was admitted for progressive left upper abdominal pain. He reported intermittent abdominal pain and fullness for the past several years. One month prior to admission, he developed severe pain and a 10-pound weight loss. He denied fever, chills or night sweats. CT confirmed a 21 cm spleen and adenopathy of the right hepatic hilum and retroperitoneum, the largest nodal mass measured 6.7×3.9 cm. He also presented with profoundly prolonged PT of 42 s (reference range: 11.9–15.0 s) and aPTT of 122 s (reference range: 23–38 s).

Differential diagnosis

Prolongation of aPTT occurs due to some anticoagulants, an inhibitor to coagulation factors, and deficiency of coagulation factors. To differentiate between these diagnoses, we proceeded with a 1:1 mixing study, which confirmed incomplete correction of the aPTT. This was consistent with the presence of an inhibitor.

With the above results and a clinical picture inconsistent with coagulopathy, there was a strong suspicion for LA causing prolongation of the aPTT. This was confirmed with prolonged dilute Russell viper venom test at 67 s (reference range: 33–44 s), which did not correct with the 1:1 mixing study. The anti-beta 2 glycoprotein IgM and aCL IgM were also elevated (table 1). Factor XII was normal, while factors II and IX were low. There was no clinical evidence of DVT or PE, and he had no history of venous or arterial thromboembolic events.

Table 1.

Admission laboratory results of a marginal zone lymphoma patient with elevated antiphospholipid antibodies

Value	On admission	Reference range
WBC	5.9	5–10×109/L
Haemoglobin	81	135–175 g/L
Platelet	110	150–450×109/L
PT	42	11.9–15.0 s
International Normalized Ratio	4.2
aPTT	122	23–38 s
PTT 1:1 mixing study	48	23–38 s
Fibrinogen	320	232–450 mg/dL
DRVVT	67	33–44 s
DRVVT 1:1 mixing study	106	33–44 s
Anti-B2 glycoprotein IgG	2.6	<20.0 U/mL
Anti-B2 glycoprotein IgM	105	<10.0 U/mL
Anticardiolipin IgG	8.7	<11.0 U/mL
Anticardiolipin IgM	214.3	<10.0 U/mL

aPTT, activated partial thromboplastin time; DRVVT, diluted Russell Viper Venom time; PT, prothrombin time; WBC, white blood cell.

Excisional axillary lymph node biopsy was consistent with a composite B-cell and MZL. Bone marrow aspirate and biopsy revealed two abnormal populations of B-cells identified as CLL/SLL and MZL involvement with no evidence of large cell transformation. Flow cytometry was positive for CD20, CD19, CD45 and CD11c in both populations, and only one population was also positive for CD23 and CD5. Neoplastic B cells comprised approximately 10%–20% of the marrow cellularity. Cytogenetic analysis was positive for del(13q).

Treatment

The chemotherapy regimen of bendamustine–rituximab (BR) was chosen to treat MZL, which was felt to be the disease responsible for his coagulation abnormalities. He received six cycles of bendamustine (90 mg/m²) and rituximab (375 mg/m²) given every 28 days with no significant complications. He demonstrated complete response on PET/CT, and his spleen decreased to 13 cm in size. No evidence of clonal B cells were seen in peripheral blood or bone marrow biopsy. He received maintenance rituximab 375 mg/m² intravenously every 3 months for a total of eight doses over 2 years. His CBC normalised during maintenance therapy.

Due to concern for thrombotic storm related to cytokine release during the initial chemotherapy administration, he was treated with fondaparinux 5 mg subcutaneously daily for a total of 14 days.

Outcome and follow-up

His PT and aPTT gradually decreased during his treatment. The PT normalised by the end of the sixth cycle of BR, and aPTT normalised during the first year of rituximab maintenance. He remains in remission with normal coagulation studies 6 years from his initial diagnosis at the time of this publication (figure 1).

Figure 1.

Prothrombin time and activated partial thromboplastin time before and after treatment. aPTT, activated partial thromboplastin time; BR, bendamustine–rituximab; MZL, marginal zone lymphoma; PT, prothrombin time.

Discussion

MZL is a rare indolent B-cell non-Hodgkin’s lymphoma that warrants treatment in symptomatic patients.¹³ We present a patient with MZL and indolent SLL who exhibited symptomatic splenomegaly, anaemia, thrombocytopenia, diffuse lymphadenopathy and abnormal coagulation studies. His profoundly elevated PT and aPTT and lack of history or clinical evidence of coagulopathy prompted investigation for APLA.¹⁴

Our patient demonstrated APLAs, specifically LA, aCL IgM and anti-beta 2 glycoprotein IgM. His coagulation studies, including aPTT and PT, were elevated, and aPTT did not correct on mixing study. The diagnosis of LA is a three part process, which consists of a screening test with diluted Russell Viper Venom time followed by sensitive aPTT. It is then followed by a mixing study and a confirmatory test.¹⁵ LAs causing prolonged aPTT, and more rarely, PT is a known phenomenon, which is thought to be caused by the LA interference with phospholipid-depdendent clot-based tests, such as aPTT.¹⁶

MZL is associated with autoimmune processes in 20% of cases.¹⁷ Prevalence of elevated APLA in non-Hodgkin’s lymphoma (NHL) has been explored in two prospective cohort studies.^{7 8} These studies showed that APLAs were seen in 41% and 26.6% of the NHL patients by Bairey et al and Genvresse et al. More specifically, a retrospective study investigating the prevalence of LA in MZL patients only showed elevated LA in 9 of 70 patients¹¹ and 4 of those patients developed DVT. However, patients with asymptomatic or transient LA positivity may have been missed.

No optimal upfront therapy has been established due to the rarity of the MZL and lack of clinical trials. Treatment ranges from splenectomy, rituximab and chemotherapy with or without rituximab.^{18 19} In the restrospective study by Gebhart et al,¹¹ out of the nine cases with MZL, patients treated with rituximab–bendamustine achieved complete remission of LA whereas none of the patients treated with splenectomy alone had achieved complete remission. Although it is an underpowered study, we used BR immunochemotherapy for our patient with similar results. His MZL and SLL are in remission. His PT had normalised by the end of six cycles of BR, and aPTT had normalised by the first year of rituximab maintenance.

While most studies have noted increased thromboembolic events in cancer patients with elevated APLAs,5 6 9–11 a prospective cohort study by Genvresse et al ⁸ following 90 non-Hodgkin’s lymphoma patient found elevated APLAs in 24 patients. None of the 24 patients developed thromboembolic events in a median follow-up period of 14 months, while 12 patients with negative APLAs had thromboembolic events. We chose to place the patient on 14 days of intermediate dosing fondaparinux due to concern for thrombotic storm related to cytokine release during initial therapy²⁰ as well as being triple positive for LA, aCL and B2GPI, a condition that is associated with high thrombotic potential.²¹ Thrombotic storm is diagnosed clinically, and the concern for thrombotic storm in our patient was due to several high risk clinical features to include hypercoagulable state secondary to both malignancy and positive APLAs and the possibility of cytokine release immediately after the treatment of his malignancy. Our patient did not develop any thromboembolic events off anticoagulation for the remainder of his treatment, but the use of anticoagulation for cancer patients with elevated APLA and no evidence of thromboembolic events remains controversial.

In conclusion, in patients with known malignancy who present with profoundly abnormal coagulation studies, APLA should be considered as part of the diagnostic evaluation. In patients with MZL and positive APLA, we recommend rituximab–bendamustine over splenectomy based on retrospective data and the outcome of our case. The use of anticoagulation in cancer patients with APLA but without thromboembolic events remain controversial, but an initial course of prophylactic anticoagulation may be warranted to prevent thrombotic storm.

Learning points.

• Prolonged prothrombin time and activated partial thromboplastin time without clinical evidence of coagulopathy should prompt investigation for lupus anticoagulant antibody.

• Bendamustine–rituximab is a reasonable first-line therapy for patients with marginal zone lymphoma with positive antiphospholipid antibodies.

• Intermediate or prophylactic dosing anticoagulation should be considered during initial therapy for cancer patients with positive antiphospholipid antibodies to prevent thromboembolic events.