Immunosuppressive effects and clinical response of fludarabine in refractory chronic lymphocytic leukemia

Annals of Oncology 4: 371-375. 1993. O 1993 Klmver Academic Publishers. Printed in the Netherlands.

Original article Immunosuppressive effects and clinical response of fludarabine in refractory chronic lymphocytic leukemia L. Bergmann, K. Fenchel, B. Jahn, P. S. Mitrou & D. Hoelzer Division of Hematology, Department of Internal Medicine, J. W. Goethe University, Frankfurt/M., FRG

to fludarabine occurred within two treatment courses. Major toxic effects included infections in 11 patients and nausea in Background: Fludarabine monophosphate is a new adenine 8 (mainly grade 1). Meanwhile, three patients died of pronucleoside analogue with a promising efficacy in B-cell gressive disease and 8 of pneumonias or other infections. chronic lymphocytic leukemia (B-CLL) with response rates, Two patients had pneumocystis carinii pneumonias and one including hematological complete remissions, of 50%-60% an aspergillus pneumonia. The high infection rate may be due in previously treated and 75%-80% in previously untreated not only to hypogammaglobulinaemia or fludarabine-induced granulocytopenia but also to a remarkable decrease of patients. + Patients and methods: Here, the clinical experience with CD4 -cells during fludarabine therapy. In one case a tumor and side effects of fludarabine are reported in 19 patients lysis syndrome was observed. No CNS toxicity was noted. with refractory CLL (17 B-CLL, 2 T-CLL). All patients were Conclusion: It is concluded that fludarabine is effective pretreated with one to four different regimens and had pro- even in patients with advanced chronic lymphocytic leukemia gressive disease. Fludarabine was administered at a dosage of refractory to multiple chemotherapy regimens. However, 25 mg/m2 daily for 5 days as a 30-minute intravenous infu- fludarabine has a remarkable suppressive effect on T-lymsion. This course was repeated every fifth week. Dosage and phocytes, predominantly CD4+-lymphocytes. Long-term time course were adapted to toxicity. antibiotic prophylaxis is recommended. Results: 12/18 (67%) evaluable patients achieved partial remissions (PR), 1/18 (6%) had stable disease (SD) and 5/18 (28%) were progressive. The median duration of partial re- Key words: fludarabine, chronic lymphocytic leukemia, mission until relapse or death was 6 months. Most responses immunosuppression, infection rate Summary

Introduction

Chronic lymphocytic leukemia (CLL) is one of the most common forms of malignant lymphoma [1]. Whereas most clinical trials in lymphoproliferative disorders have focused on lymphomas of high-grade malignancy, a substantial amount of interest has recently shifted to chronic B- and T-cell leukemias because of an increased understanding of the biology and immunology of these leukemias, and of the availability of new promising drugs. With the standard first-line therapy, a combination of chlorambucil and prednisone, remission rates of 45% to 85% can be achieved in CLL [2|. More aggressive regimens used in other non-Hodgkin's-lymphomas, such as combination of cyclophosphamide, vincristine and prednisone (e.g., COP), or adriamycin (e.g., CHOP), have been reported to be of benefit in fairly advanced CLL [3, 4|. However, in advanced disease the therapy is frequently restricted due to sustained pancytopenia and recurrent, possibly life-threatening, infections |5]. As of now, no curative approaches are available and hematological complete remissions are rare.

Recently, promising results in progressive CLL resistant to standard regimens were reported with new cytostatic drugs [6—9]. These include purine nucleoside analogues such as fludarabine monophosphate (FAMP, F-ara-A), 2-deoxycoformycin, and 2-chlorodeoxyadenosine. Especially FAMP has been shown to be highly effective in B-CLL, with response rates of 50%—60% in previously treated and 75%-80% in previously untreated patients, with a remarkable proportion of hematological complete remissions [7, 10, 11). Fludarabine is converted intracellularly in its active metabolite (F-ara-ATP), which inhibits DNA synthesis as well as RNA and protein synthesis |12J. In phase I trials, the substance was used in the treatment of acute leukemias, lymphomas and solid tumors, and its major toxicity was myelosuppression [10, 11, 25]. Although severe, lethal demyelination was noted when high doses were used in phase I studies of acute leukemias, this toxicity was only rarely described for lower doses administered for other malignancies [26, 27]. Here we report our clinical results with fludarabine in refractory CLL and with fludarabine-related severe infectious complications resulting from its immunosuppressive activity.

372

Patients and methods Patient characteristics Between February 1990 and March 1992, 19 patients (13 male, 6 female) with advanced CLL were entered into the study and are evaluable for response or toxicity. The diagnosis of CLL was based on morphological examination of blood smears and bone marrow as well as on immunophenotypical analysis. Seventeen patients showed a CLL of B-cell and two of T-cell type, one of CD4 + - and one of CD8 + phenotype (Table I). The median age was 64 years (34-75 years). Fifteen patients were assigned a Binet stage C and 4 a Binet stage B [28], Four patients had been pretreated by 4 different regimens, 5 with three, 5 with two regimens and 5 with only one (Table 1). All patients had progressive disease, refractory to standard regimens.

Table I. Patient characteristics. Eligible Evaluable Age (years)0 Sex Binet stage B Binet stage C Immunological subtype B-CLL T-CLL Pretreatment 1 regimen 2 regimens 3 regimens 4 regimens Leukocytes/yl" Lymphocytes (%)° Platelets/yl" Hb g/dl11 Lymphomas Organomegaly (spleen/liver) Organ infiltration

19 patients 18 patients 34-75 (median 64) 13 male, 6 female 4 15 17 2

mg/ml and then mixed with 50 ml 0.9% NaCl. The standard dose level was 25 mg/m2 intravenous (I.V.) given during a 30-minute period each day for 5 days; this course was repeated every fifth week. Depending on toxicity, dose decreases were made in 5 mg/mVd increments for subsequent courses. The criteria for partial remission (PR) according to the National Cancer Institute Working Group were the reduction of circulating lymphocytes by more than 50% to baseline and reduction of lymph node size by more than 50%. Adherence to at least one of the following parameters was also required: granulocytes >1500/u.l or 50% improvement to baseline; platelets > l()0()00/|il or 50% improvement to baseline; hemoglobin > 11 g/dl or 50% improvement to baseline; reduction of hepato/splenomegaly to 50%; bone marrow infiltration by lymphocytes less than 30% for at least two months. The criteria for complete remission (CR) were: circulating lymphocytes <4000/(il; granulocytes >1500/ul; thrombocytes > 100000/nl; hemoglobin > 11 g/dl; bone marrow infiltration <30%; no palpable lymph nodes or hepato/splenomegaly. Progressive disease (PD) is characterized by at least one of the following parameters: > 50% increase in lymph node size; >50% increase in hepato/ splenomegaly; >50% increase in the number of circulating lymphocytes. Patients who have not achieved CR or PR and are without PD are considered as having stable disease (SD).

Immunophenotyping

6 5 5 3 171000 (7-595000) 98 (77-100) 40000 (5000-236000) 10.7(5.2-14.7) 17/19 patients 19/19 patients 1 lung

Median (range).

Study design The pretreatment evaluation included: physical examination, abdominal ultrasonography, x-ray of the chest, complete blood cell (CBC) count, differential white cell count, chemical survey, immunophenotyping, bone marrow histology and cytology, and serum levels of immunoglobulins. Follow-up tests included weekly CBC counts, and chemical profile before each treatment; immunophenotyping and other tests (such as CT scans, ultrasonic investigations, x-ray or bone marrow biopsies) demonstrating the initial presence of disease were repeated every two of three courses. Fludarabine was supplied by the National Cancer Institute (NCI), Bethesda MD, USA and by the Emmes Corp, Potomac MD, USA. The compound was initially diluted with distilled water to make a solution of 100

The surface markers on lymphocytes and leukemic cells were measured by flow cytometry using FITC- or PE-labeled antibodies (FACScan, Becton Dickinson Corp., Heidelberg, FRG). The following monoclonal antibodies were used: CD3, CD4, CD5, CD8, CD 19, CD20, anti-HLA-DR, and IgA, anti-IgG, anti-IgM, antikappa and anti-lambda.

Results

Nineteen patients were entered into the study. One patient died of anaphylactic shock after blood transfusion during the first course of fludarabine and was therefore considered unevaluable for response. The pretreatment characteristics are listed in Table J. The patients received a median of 4 (1-8) cycles of fludarabine. In 12/18 (67%) patients PR was achieved. Both of the patients with T-CLL were unresponsive to fludarabine except for a slight transient decrease of peripheral lymphocytes and both died in progression after two cycles each of fludarabine. The peripheral lymphocytosis responded rapidly to treatment followed by a regression of lymph node size and hepatosplenomegaly (Table 2). Peripheral remissions occurred in 10/12 patients after the 3rd cycle and in 2/12 after the 6th (Table 2). A response of lymph node enlargment was mostly observed

Table 2. Site of response (PR) and overall response in relation to therapy cycle. Site of response

Pre FAMP

After 1 cycle

3 cycles

Peripheral leucocytosis Lymph nodes Hepatosplenomegalie and/or lung infiltration Objective response (%)

17/18 16/18 15/18 18/18

6 1 0 0

4 9 8 I I

pts. — patients; FAMP - fludarabine.

pts. pts. pts. pts.

6 cycles

total

11/17(65%) 11/18(61%) 9/15 (60%) 12/18 (67%)

373

after at least 3 cycles of therapy (Table 2). In 4/12 patients in PR bone marrow cytology and histology revealed a decrease of bone marrow cellularity and infiltration by lymphocytes from 75%-95% to less than 30% after 6 cycles of therapy. Event free survival of patients with PR lasted a median of 6 months (2-12 months). 1/18 patients achieved stable disease (SD) by fludarabine and 5/18 patients (28%) were progressive (PD). The subjective toxic effects of fludarabine were mild. 8/18 (44%) patients complained of grade 1 WHO nausea. Objective toxic effects of fludarabine therapy were myelosuppression and infectious complications, and tumor lysis syndrome occurred in one patient, who also developed a hepatorenal syndrome in a succeeding cycle, probably due to a viral hepatitis. 17/18 (94%) patients developed myelosuppression, mainly grade 1-2 according to WHO criteria. In 11 patients (58%), all of whom had achieved partial remissions, severe infections were observed, three opportunistic infections, two pneumocystis carinii pneumonias (PCP) and one aspergillus pneumonia. Two patients were affected by bacterial pneumonia and six patients by life-threatening infections without localisation or isolation of agents. Eights of these patients (1 with PCP, 1 with aspergillus, 6 with severe infections without isolation of agents) succumbed to their infections. Four non-responders, two B-CLL and two T-CLLs, died due to progression of CLL.

During therapy, the number of circulating leucocytes dropped from a median of 238300 (7000-564000) before therapy to 39300 (2500-130500) after the 3rd cycle of therapy (Table 3); this corresponds to a cytoreduction of 84%. In differential blood cell count, the proportion of lymphocytes was reduced from a median 98% to 62% after therapy (Table 3, Fig. 1). Fludarabine therapy led not only to a decrease of leukemic B-cells but also to a reduction of T-cells, including CD4 + and CD8+-lymphocytes (Table 4, Fig. 2). The CD4+-cells dropped from a median 2935/ul to 1316/ul with less than 350 CD4+-cell/ul in 6/10 investigated cases (Fig. 2). In patients achieving PR, the reduction of CD4+-cells was more pronounced with a median of 276 cells/ul peripheral blood. CD8+-lymphocytes were also reduced by fludarabine therapy from a median Table 3. Pretreatment values of blood counts and their follow-up after FAMP therapy in all patients (median, range).

Leukocytes/|il (x 1 <>•') 171 (7-595) Lymphocytes (%) 98(77-100) Granulocytes (%) 5(1-20) Platelets/nl (x 10-') 40(5-236) Hbg/dl 10.7(5 2-14.7) FAMP-fludarabine.

lymphocytes/|il x 1000

600

500 -

400

300 -

200 -

100 -

b«fore FAMP

cyclt 1

cyole 3

cycle 6

Fig. I. Follow-up of lymphocytes during treatment with fludarabine (all values are just prior to start of the next therapy cycle).

Hematological parameters

Before FAMP (n-18)

5281 cells/(il to 1131 cells/|il, resulting in a nearly unchanged CD4/8-ratio. A significant increase in platelet counts was only occasionally observed (Table 4). However, despite per-

After cycle 3 (n-14)

After cycle 6 (n-7)

24 (4-93) 65(17-98) 21 (2-74) 47 (8-59) 9.9(7-14.8)

16(2-65) 62 (7-95) 34 (3-88) 62(16-172) 9.5(8-11.3)

CD 4+ lymphocytes/u,l

100000

10000

r i

1000 :

100 z

before FAMP

alter FAMP

Fig. 2. Change of CD4+-lymphocytes in relation to fludarabine cycles administered (all values are just prior to start of the next therapy cycle).

374 Table 4. Hematological parameters and CD4+-lymphocytes before and after last administration of fludarabine (FAMP) in relation to response. Patient

1 B-CLL 2 B-CLL 3 B-CLL 4 B-CLL 5 T-CLL 6 B-CLL 7 B-CLL 8 B-CLL 9 B-CLL 10 B-CLL 1 1 B-CLL 12 B-CLL 13 T-CLL 14 B-CLL 15 B-CLL 16 B-CLL 17 B-CLL 18 B-CLL

Administered FAMP-cycles

Response

6 2 7 6 2 3 7 5 8 4 12 6 2 5 2 4 3 4

SD PD PR PR PD PR PR PR PR PR PR PR PD

PD PR PD PR PR

Lymphocytes/u.1

Hb g %

Thrombocytes/|il

CD4Vul

Before FAMP

End of therapy

Before FAMP

End of therapy

Before FAMP

End of therapy

Before FAMP

115000 63000 322000 171000 564000 450000 285500 131700 7000 288600 224400 159000 331500 147000 31 160 138320 161920 54900

65000 88200 19900 34000 234000 102300 4000 7900 2400 58700 2800 2200 460000 230000 3293 63012 24985 17000

14.7 10.2

15.4

9.6 8.1 9.2

10.0 10.2

236000 14000 48000 5000 18000 113000 20000 14000 50000 4000 22000 32000 57000 156000 21000 93000 113000 92000

175000 13000 19000 18000 51000 46000 80000 12000 141000 24000 19000 24000 29000 78000 19000 6000 47000 13000

11.0 12.6 10.8

7.9

7.3 9.0 8.1

10.4

8.9

9.1

10.7

10.3 11.2

5.2

10.3 10.9 12.1 10.7 9.9

10.9 10.1

9.8

10.1 1 1.1 9.2 7.9 8.5

1 1.1

3105

End of therapy 100

539

n.d.

3059

248 794

851

n.d. 4365 5866 3295 346

11082 4443 6260 174646 4104

n.d. 5204 295 335 252

3451 142 19

313898

604

817 292

5831 3365

4104 1900

658

364

n.d. - not done, PR — partial remission, SD — stable disease, PD - progressive disease, FAMP - fludarabine monophosphate. 'end of therapy' indicates values four weeks after last FAMP administration.

sisting thrombocytopenia, the frequency of platelet substitution was reduced in responding patients.

has been encountered and no neurotoxicity become evident in our patients. The myelosuppression was mild. However, in the meanwhile there have been several reports of infectious complications including opporDiscussion tunistic infections such as pneumocystis carinii pneuFludarabine has been shown to be highly effective in monia and cytomegalovirus infections [21, 30]. In this pretreated and previously untreated CLL with re- study 11/18 patients developed severe infectious comsponse rates (CR + PR) of about 55% and 80%, re- plications including opportunistic infections such as spectively [6, 7, 11, 12, 29]. The present study confirms pneumocystis carinii and aspergillus pneumonias. CLL the high activity of fludarabine in a partially severely patients are known to have impaired humoral defense pretreated patient group. In the study reported here, mechanisms with reduced immunoglobulin levels. there were no CRs to fludarabine. This fact may be due Additionally, remarkable long-lasting depletions of to a very unfavourable patient selection. No benefit was CD4+-lymphocytes, similar to the values in advanced observed in two patients with T-CLL except a transient acquired immunodeficiencies, were observed in the reduction of peripheral lymphocyte counts. So far, present study. This loss of CD4+-cells may thus contrithere are only a few reports with fludarabine in periph- bute to a similar spectrum of opportunistic infections. eral T-cell lymphomas or leukemias and these describe Hence, we performed and recommend infection preonly a moderate activity [21, 24]. In responding pa- vention with low-dose co-trimoxazole for patients tients, there was a dissociation between rapid periph- treated with fludarabine according to Ruskin and La eral cytoreduction and regression of involved lymph Riviere |31 ]. The value of this supportive therapy, hownodes during the first three cycles of therapy. An in- ever, still needs to be elucidated. crease of platelets in initially thrombopenic patients Fludarabine is a highly effective and well-tolerated occurred after only three to six cycles of fludarabine drug in the treatment of refractory CLL. In addition, (Table 4). the results in previously untreated patients with CLL The subjective toxicity of fludarabine was mild. In and low-grade non-Hodgkin's lymphomas are encourprevious studies using high doses of fludarabine in aging [11, 21]. Randomized trials comparing fludaraacute leukemias, severe, even life-threatening CNS bine with standard regimens in these patient groups toxicities, including cortical blindness, were described are warranted. Fludarabine-induced immunodefi[23, 26, 27]. Merkel et al. [27] reported on pulmonary ciency, however, is accompanied by a high rate of infectoxicities using high-dose fludarabine. Such drug-relat- tions, including opportunistic ones requiring antibiotic ed side effects have rarely been observed using low- prophylaxis. dose regimens [6, 11, 27]. Thus far, no organ toxicity

375 Acknowledgements

The following investigators from participating institutions contributed patients to this study: Dr. Wilhelm, Div. of Hematology/Oncology, Municipal Hospital, Karlsruhe, FRG; Dr. C. Gorg and Dr. G. Gorg, Division of Hematology, University Hospital, Marburg, FRG.

16. 17. 18. 19.

References 1. Lennert K. Malignant lymphomas. Springer Verlag: Berlin 1978; 106-52. 2. Knospe WH, Loeb V, Huguley CM. Bi-weekly chloranbucil treatment of chronic lymphocytic leukemia. Cancer 1974; 31: 502-8. 3. French Cooperative Group on Chronic Lymphocytic Leukemia. Effectiveness of 'CHOP' regimen in advanced chronic lymphocytic leukemia. Lancet 1986; I: 1346-9. 4. Raphael B, Andersen JW, Silber R et al. Comparison of chlorambucile and prednisone versus cyclophosphamide, vincristine, and prednisone as initial treatment for chronic lymphocytic leukemia: Long-term follow-up of an Eastern Cooperative Oncology Group randomized trial. J Clin Oncol 1991; 9:770-6. 5. Monserrat E, Alcala A, Parody R et al. Treatment of chronic lymphocytic leukemia in advanced stages. Cancer 1985; 56: 2369-75. 6. Grever MR, Coltman CA, Files JC et al. Fludarabine monophosphate in chronic lymphocytic leukemia. Proc Am Soc Hematol Blood 1986; 68: 223a (abstr). 7. Keating MJ, Kantarjian H, Talpaz M et al. Fludarabine: A new agent with major activity against chronic lymphocytic leukemia. Blood 1989; 74: 19-24. 8. O'Dwyer PJ, Wagner B, Leyland-Jones B et al. 2'-Deoxycoformycin for lymphoid malignancies - rational development of an active new drug. Ann Intern Med 1988; 108: 733-43. 9. Piro LD, Carrera CJ, Beutter E et al. 2'-Chlorodeoxyadenosine: An effective new agent for the treatment of chronic lymphocytic leukemia. Blood 1988; 72: 1069-73. 10. Keating MJ, Kantarjian H, Talpaz M et al. Fludarabine therapy in chronic lymphocytic leukemia (CLL). Nouv Rev Fr Hematol 1988; 30:457-9. 11. Keating MJ, Kantarjian H, O'Brien S et al. Fludarabine: A new agent with marked cytoreductive activity in untreated chronic lymphocytic leukemia. J Clin Oncol 1991; 9:44-9. 12. Noker PE, Duncan GF, El Dareer SM et al. Disposition of 9-p"-arabinofuranosyl-2-fluoroadenine 5'-phasphate in mice and dogs. Cancer Treat Rep 1983; 67:445-56. 13. Carpenter JT Jr, Vogel CL, Wang G et al. Phase II evaluation of fludarabine in patients with metastatic breast cancer: A Southeastern Cancer Study Group trial. Cancer Treat Rep 1986; 70: 1235-6. 14. Harvey WH, Fleming TR, Von Hoff DD et al. Phase II study of fludarabine phosphate in previously untreated patients with colorectal carcinoma: A Southwest Oncology Group Study. Cancer Treat Rep 1987; 71: 1319-20. 15. Harvey WH, Fleming TR, Beltran TG et al. Phase II study of fludarabine phosphate in previously untreated patients with

20. 21. 22. 23.

24.

25. 26. 27. 28. 29. 30. 31.

hepatoma: A Southwest Oncology Group Study. Cancer Treat Rep 1987; 71: 1111-2. Kantarjian HM, Alexanian R, Koller CA et al. Fludarabine therapy in macroglobulinemic lymphoma. Blood 1990; 75: 1928-31. Kavanagh JJ, Stringer CA, Copeland LJ et al. Phase II trial of fludarabine in patients with epithelial ovarian cancer. Cancer Treat Rep 1986; 70:425-6. Kraut EH, Crowley JJ, Grever MR et al. Phase II study of fludarabine phosphate in multiple myeloma. A Southwest Oncology Group Study. Invest New Drugs 1990; 8: 199-200. Mittelman A, Savona S, Puccio C et al. Phase II trial of fludarabine phosphate (F-Ara-AMP) in patients with advanced head and neck cancer. Invest New Drugs 1990; 8: 865-7. Pazdur R, Samson MK, Baker LH. Fludarabine phosphate. Phase II evaluation in advanced soft-tissue sarcomas. Am J Clin Oncol 1987; 10:341-3. Redman JR, Cababillas F, Velasquez WS et al. Phase II trial of fludarabine phosphate in lymphoma: An effective new agent in low-grade lymphoma. J Clin Oncol 1992; 10:790-4. Smith OP, Mehta AB. Fludarabine monophosphate for prolymphocytic leukemia. Lancet 1990; 336:820. Spriggs DR, Stopa E, Mayer RJ et al. Fludarabine phosphate (NSC 312878) infusions for the treatment of acute leukemia: Phase I and neuropathological study. Cancer Res 1986; 46: 5953-8. Von Hoff DD, Dahlberg S, Hartstock RJ et al. Activity of fludarabine monophosphale in patients with advanced mycosis fungoides: A Southwest Oncology Group Study. J Natl Cancer Inst 1990; 82: 1353-5. Hurst PG, Habib MP, Garewal H, Bluestein M et al. Pulmonary toxicity associated with fludarabine monophosphate. Invest New Drugs 1987; 5: 207-10. Chun HG, Leyland-Jones BR, Caryk SM et al. Central nervous system toxicity of fludarabine phosphate. Cancer Treat Rep 1986; 70: 1225-88. Merkel DE, Griffin NL, Kagan-Hallet K et al. Central nervous toxicity with fludarabine. Cancer Treat Rep 1986; 70: 1449-50. Binet JL, Auquier A, Dighiero G et al. A new prognostic classification of chronic lymphocytic leukemia derived from a multivariate survival analysis. Cancer 1981; 48: 198-201. Hiddemann W, Rottmann R, Wbrmann B et al. Treatment of advanced chronic lymphocytic leukemia by fludarabine. Results of a clinical phase II study. Ann Hematol 1991; 63: 1—4. Schilling PJ, Vadhan-Raj S. Concurrent cytomegalovirus and pneumocystis pneumonia after fludarabine therapy for chronic lymphocytic leukemia. N Engl J Med 1990; 323: 833-4. Ruskin J, La Riviere M. Low-dose co-trimoxazole for prevention of pneumocystis carinii pneumonia in human immunodeficiency virus disease. Lancet I 1991; 337:468-71.

Received 14 September 1992; accepted 20 January 1993. Correspondence to: L. Bergmann, M.D. Div. of Hematology Dept. of Internal Medicine University Hospital Theodor-Stern-Kai 7 D-6000 Frankfurt/M, FRG