Chronic cold agglutinin disease (CAD) is a type of autoimmune hemolytic anemia (anemia due to destruction of red blood cells by abnormal antibodies). Almost all patients also suffer from cold-induced disturbances of blood circulation. The purpose of this study is to assess the efficacy and safety of combination therapy with rituximab (an antibody against B lymphocytes) and fludarabine (a cytotoxic drug) for CAD. Another aim is to try to assess whether these agents in combination are better than single agent therapy with rituximab.

1. Background

Chronic cold agglutinin disease (CAD) is mediated by monoclonal cold-reactive autoantibodies that bind to erythrocyte surface antigens, causing haemagglutination and complement-mediated haemolysis. Anaemia is severe (Hb 8.0 g/dL or lower) in one-third of patients and complement-induced exacerbation during febrile illness occurs frequently 1-3. Cold-induced circulatory symptoms are present in more than 90% of patients and may be disabling 1. CAD not associated with overt lymphoma or other disease has traditionally been classified as primary or idiopathic. However, a lymphoproliferative bone marrow disorder can be demonstrated by flow cytometry in 90% and by histology in approximately 75% of these patients, characterized by clonal proliferation of CD20+,κ+ B-cells 1,4,5. The histological features are those of lymphoplasmacytic lymphoma in about 50% of the patients 1.

Many standard therapies used in other autoimmune diseases or indolent lymphomas are inefficient, e.g. corticosteroids, alkylating agents, interferon-α and, probably, purine analogue single agent therapy 1,6-8. Treatment with the chimeric monoclonal anti-CD20 antibody rituximab has been shown in prospective studies to induce remission in more than half of patients 9,10. Almost all responses were partial, however, and the median response duration was less than one year. Further studies are warranted, therefore, in order to explore the possibilities for increasing the response rate and duration.

The purine analogues, cladribine and fludarabine, are not cyclus dependent, and they have yielded partial response rates of 30-75% and complete response rates of 3-10% in lymphoplasmacytic lymphoma 11,12. Although probably clinically inefficient as monotherapy in most CAD patients, clinical effect of fludarabine has been reported in a single case 13, and cladribine has been shown to induce tumour reduction even in this condition 8. In lymphoplasmacytic lymphoma, purine analogue and rituximab combination therapy has resulted in higher response rates and more prolonged remissions as compared to purine analogue single agent therapy 12. The combination has produced favourable results in some patients with the related condition cryoglobulinaemia type I 14. Fludarabine has induced autoimmune haemolytic anaemia in patients with chronic lymphocytic leukaemia, but such events have not been reported in CAD patients 13,15. Moreover, there are reasons to assume that rituximab therapy will further reduce the risk of this adverse effect of fludarabine 16.

2 Clinical study

The clinical study is a prospective, non-randomized multicentre study in order to investigate the efficacy and safety of rituximab and fludarabine combination therapy in patients with CAD. The protocol has approved by the Regional Medical Research Ethics Committee of Southern Norway, Norwegian Medicines Agency (EudraCT nr: 2004-002936-25), and Norwegian Social Science Data Services (Privacy Issue Unit).

2.1 Study objectives

The major study objective is to assess efficacy of rituximab and fludarabine in combination for patients with CAD.

The second objective is to assess safety of rituximab and fludarabine in combination for patients with CAD.

The third objective is to try to assess whether rituximab and fludarabine in combination are superior to rituximab monotherapy comparing patients who have received both therapies for CAD.

2.2 Study design

A prospective, non-randomized international multicentre study.

Registration

Treatment:

Day 1: Rituximab; 375 mg/m2 Day 1-5: Fludarabine orally; 40 mg/m2

Day 28: Rituximab; 375 mg/m2 Day 28-33: Fludarabine orally; 40 mg/m2

Day 56: Rituximab; 375 mg/m2 Day 56-60: Fludarabine orally; 40 mg/m2

Day 84: Rituximab; 375 mg/m2 Day 84-88: Fludarabine orally; 40 mg/m2

Evaluation

2.3 Dose adjustments

Doses of fludarabine will be adjusted in case of haematological toxicity or renal insufficiency. For details, see chapter 4.3.

2.4 Study population

2.4.1 Inclusion criteria

1. CAD diagnosis defined by the combination of -

   1. Chronic haemolysis
   2. Cold agglutinin titre > 64
   3. Positive direct antiglobulin test when performed with polyspecific antiserum, negative (or only weakly positive) with anti-IgG, and strongly positive with anti-C3d

2. The presence of a clonal B-cell lymphoproliferative disorder defined by -

   1. Monoclonal IgMκ band by serum electrophoresis and immunofixation, and
   2. Lymphocyte phenotype with κ/λ-ratio > 3.5 and CD20+,κ+ co-expression, using flowcytometric immunophenotyping of bone marrow aspirates

3. Clinical symptoms requiring treatment, such as anaemia or Raynaud-like symptoms

4. Informed consent

2.4.2 Exclusion criteria

1. An aggressive lymphoma
2. Blood lymphocyte count > 50 . 109/L
3. Non-lymphatic malignant disease other than basal cell carcinoma
4. Contra-indications to rituximab or fludarabine therapy
5. Inability to cooperate

2.5. Response criteria

Responses will be assessed using the following, previously published definitions 8,9,17:

Complete response (CR), Absence of anemia, no signs of hemolysis, no clinical symptoms of CAD, undetectable serum monoclonal protein, and no signs of clonal lymphoproliferation by bone marrow histology, immunohistochemistry and flow cytometry.

Partial response (PR), Stable increase in Hb levels by at least 2.0 g/dL or to the normal range, combined with a reduction of serum IgM concentrations by at least 50% or to the normal range, improvement of clinical symptoms, and transfusion independency.

Non-response (NR), Patients not meeting the criteria for CR or PR.

3 Patient examination at inclusion

3.1 History. Clinical and radiological examination

Year of first occurrence of clinical symptoms is registered along with data on haemolytic anaemia, circulatory symptoms, cold- or fever-induced exacerbation, previous therapies, lymph node enlargement, and spleen size (clinical assessment). Chest radiograph and abdominal ultrasonography should be done if not performed already during the last four months.

3.2 Blood tests

Haemolysis is detected and quantified based on Hb, reticulocyte count (x 109/L), LDH, bilirubin and haptoglobin. These measurements should be done twice during the last two months before treatment.

The following haematological, biochemical and immunological assessments should be done once at inclusion:

• WBC, leukocyte differential count, platelet count
• Iron, transferrin (or TIBC), ferritin, cobalamine and folate
• CRP
• Quantification of IgM, IgG and IgA
• Serum electrophoreses with immunofixation (*) (Immunofixation must be performed even if visual assessment of agarose electrophoreses does not show any monoclonal band.)
• Cold agglutinin titre (*)
• Specific direct antiglobulin test (DAT, direct Coombs' test), i.e. using polyspecific antiserum, anti-C3d and anti-IgG)
• Complement assessments (C3, C4 and CH50) (*, **)
• CMV and VZV antibodies
• Freezing of 5 ml serum (*, **)
• Freezing of 5 ml EDTA-blood for possible later DNA-based studies (**)

Remarks:See protocol text.

3.3 Bone marrow examinations

Centres outside Norway should follow the guidelines listed below. Norwegian centres should refer to the Norwegian protocol version.

I.) Morphologic assessment of bone marrow aspirate is performed according to the routines of the participating centre.

II.) A bone marrow trephine biopsy should be obtained according to the routines of the department. Morphological and immunohistochemical assessments are to be done at a university pathology laboratory by an experienced haematopathologist. If possible, a fresh biopsy specimen should be preferred.

III.) Flow-cytometric immunophenotyping of bone marrow aspirate should be done at a university hospital laboratory. Cells should be washed at 37oC in order to remove cold agglutinins, using the previously published procedure 5 (Appendix 1) or an equally satisfactory method. The antibody panel should comprise CD19, CD20, CD5, kappa, lambda and preferably also IgM and IgG. The kappa/lambda ratio should be calculated. CD20 expression is registered semi-quantitatively as 0, + or ++.

IV.) If possible, 5 ml bone marrow aspirate should be frozen at -70oC for the purpose of later DNA-based examinations.

4 Therapy

4.1 Treatment schedule

Day 1: Rituximab; 375 mg/m2 Day 1-5: Fludarabine orally; 40 mg/m2

Day 28: Rituximab; 375 mg/m2 Day 28-33: Fludarabine orally; 40 mg/m2

Day 56: Rituximab; 375 mg/m2 Day 56-60: Fludarabine orally; 40 mg/m2

Day 84: Rituximab; 375 mg/m2 Day 84-88: Fludarabine orally; 40 mg/m2

4.2 Administration and precautions

Administration and monitoring of rituximab and fludarabine therapy should be in accordance with the manufacturers' recommendations, the official regulations that apply in the participating country, and the routine procedures of the participating unit.

4.3 Dose adjustments

4.3.1 Haematological toxicity If ANC<1,0x109/L and/or platelets<75x109/L at day 1 of any cycle treatment should be delayed for up to 2 weeks and fludarabine dose should be reduced to 30 mg/m2 in the remaining cycles. Myelosuppression of the same degree in the subsequent cycles should result in a reduction of the fludarabine dose to 20 mg/m2.

Adjustments of the rituximab dose should not be done due to myelosuppression.

4.3.2 Renal insufficiency In patients with a creatinine clearance between 30-60 ml/min the fludarabine dose should be reduced to 20 mg/m2. Patients with a creatinine clearance below 30 ml/min are not eligible for the study.

5 Evaluation

5.1 Follow-up during treatment

The parameters listed below must be recorded before each therapy cycle:

Clinical condition, possible adverse effects, need for transfusion, Hb, reticulocytes count, WBC including differential count, platelet count, CRP, LDH, ASAT, ALAT, alkaline phosphatase, urea, creatinine, bilirubin and haptoglobin.

Any adverse effects of fludarabine are registered in CRF 2 (Appendix 3). Any adverse effects of rituximab are registered in CRF 3 (Appendix 4). In case of death or other serious events, these should be reported without delay to S. Berentsen or G. Tjønnfjord as well as the relevant national authorities according to regulations that apply in the participating country.

5.2 Follow-up after treatment (first six months)

A.) The following measurements must be done monthly during the first six months after the last cycle of therapy:

I.) All measurements listed in Paragraph 5.1. II.) Quantification of immunoglobulin classes. In case of reduction of a previously elevated IgM level to the normal range, serum electrophoresis with immunofixation should be performed at the next visit.

III.) Number of blood transfusions after the previous registration.

B.) At the first visit after cessation of therapy, a new bone marrow biopsy is performed in those patients where histological or immunohistochemical signs of a lymphoproliferative disorder were present at baseline. If a lymphoproliferative bone marrow involvement could be detected at baseline by flow cytometry only, flow-cytometric immunophenotyping is also repeated at this visit.

C.) Bone marrow examinations should also be repeated four months after the last therapy cycle if the examinations three months before showed signs of lymphoma.

5.3 Long-term follow-up

When more than six months have passed since cessation of treatment, patients should be followed every third month for three years or until they need treatment again.

Inclusion Criteria:

1. CAD diagnosis defined by the combination of:

   • Chronic haemolysis
   • Cold agglutinin titre > 64
   • Positive direct antiglobulin test when performed with polyspecific antiserum, negative (or only weakly positive) with anti-IgG, and strongly positive with anti-C3d

2. The presence of a clonal B-cell lymphoproliferative disorder defined by:

   • Monoclonal IgMκ band by serum electrophoresis and immunofixation, and
   • Lymphocyte phenotype with κ/λ-ratio > 3.5 and CD20+,κ+ co-expression, using flowcytometric immunophenotyping of bone marrow aspirates

3. Clinical symptoms requiring treatment, such as anaemia or Raynaud-like symptoms

4. Informed consent

Exclusion Criteria:

1. An aggressive lymphoma
2. Blood lymphocyte count > 50 . 109/L
3. Non-lymphatic malignant disease other than basal cell carcinoma
4. Contra-indications to rituximab or fludarabine therapy
5. Inability to cooperate