Cyclosporin A for the Treatment of CytopeniaAssociated with Chronic Lymphocytic Leukemia

Jorge Cortes, M.D.

Susan O’Brien, M.D.

Javier Loscertales, M.D.

Hagop Kantarjian, M.D.

Francis Giles, M.D.

Deborah Thomas, M.D.

Charles Koller, M.D.

Michael Keating, M.D.

Department of Leukemia, The University of Texas,M. D. Anderson Cancer Center, Houston, Texas.

Presented in part at the 40th Annual Meeting of theAmerican Society of Hematology, Miami Beach,Florida, December 1998.

Jorge Cortes, M.D., is a Clinical Research Scholarfor the Leukemia and Lymphoma Society.

Address for reprints: Jorge Cortes, M.D., Depart-ment of Leukemia, 1515 Holcombe Blvd., Box 428,Houston, TX 77030, Fax: (713) 794-4297; E-mail:jcortes@mdanderson.org

Received June 12, 2001; revision received July 28,2001; accepted August 1, 2001.

BACKGROUND. Autoimmune cytopenias are a frequent complication in patients

with chronic lymphocytic leukemia (CLL). Anecdotal reports suggest that cyclo-

sporin A (CsA) may be beneficial for patients with CLL-associated pure red cell

aplasia. In the current study, the authors investigated the use of CsA in the

management of anemia or thrombocytopenia of presumed autoimmune etiology

associated with CLL.

METHODS. Thirty-one patients with CLL and anemia or thrombocytopenia of

presumed autoimmune etiology were treated with CsA at a dose of 300 mg/day.

Sixteen patients (52%) had anemia (hemoglobin # 11 g/dL) and 29 patients (94%)

had thrombocytopenia (platelet count # 100 3 109/L). Seventeen patients (55%)

had cytopenia that developed during the course of treatment with fludarabine-

based regimens. Nineteen patients (61%) had received prior therapy for this

complication using steroids, intravenous immunoglobulin, and/or splenectomy.

RESULTS. Eighteen patients (62%) with thrombocytopenia and 10 patients (63%)

with anemia had a major response defined as an increase in the platelet count $ 50

3 109/L or an increase in hemoglobin $ 3 g/dL. The median time to initial

response was 3 weeks (range, 1–13 weeks) and the median time to best response

was 10.5 weeks (range, 1– 48 weeks). The median duration of response was 10

months (range, 11–391 months). Three patients with fludarabine-associated cy-

topenias were able to receive further therapy with fludarabine with a lesser de-

crease in the platelet count. A modest decrease in the tumor burden was observed

in six patients. The most common toxicity was # Grade 2 (according to the

National Cancer Institute’s Common Toxicity Criteria) elevation of creatinine,

which was observed in 6 patients (19%). Three patients developed opportunistic

infections.

CONCLUSIONS. CsA is an effective alternative for the treatment of anemia or

thrombocytopenia of suspected autoimmune etiology, including those cases oc-

curring in the course of treatment with fludarabine. A modest antileukemic effect

was observed in some patients. Cancer 2001;92:2016 –22.

© 2001 American Cancer Society.

KEYWORDS: cyclosporine A, chronic lymphocytic leukemia, anemia, thrombocyto-penia.

Patients with chronic lymphocytic leukemia (CLL) frequently de-velop anemia and/or thrombocytopenia that often is immune-

mediated (IM).1 Approximately 10 –70% of patients develop autoan-tibodies at some time during the course of their disease, most oftenagainst hematopoietic cells, and the incidence of a positive Coombstest increases significantly with disease stage.2 In addition, treatmentwith purine analogues causes immune deregulation, which has beenproposed as a mechanism of secondary hemolytic anemia in patientstreated with these agents.3 Several case reports and small series of

2016

© 2001 American Cancer Society

patients developing autoimmune hemolytic anemia orthrombocytopenia after exposure to fludarabine havebeen published to date.4 –9 Approximately 75% of pa-tients with IM cytopenias respond to therapy withsteroids, but there are few effective alternative thera-pies for patients whose disease is resistant or refrac-tory to steroids. Some of these alternatives includeintravenous inmunoglobulin (IVIG) and splenecto-my.10

Cyclosporin A (CsA) has been reported to be ef-fective in the treatment of pure red cell aplasia asso-ciated with CLL,11–17 a complication that occurs in, 1% of patients with CLL.10 Isolated case reports havesuggested that CsA might be effective in the treatmentof other autoimmune complications of CLL, includingthose refractory to steroids.18 –21 An antileukemic ef-fect also has been observed in some patients, with amoderate decrease in the tumor burden and in Bsymptoms.17,18,20,21

In the current study, 31 patients with autoim-mune complications associated with CLL, includingIM anemia and/or thrombocytopenia or fludarabine-associated cytopenias, were treated with CsA. The ob-jective of the current was to evaluate the efficacy andtoxicity of CsA for the treatment of IM complications.A secondary objective was to determine the possibleantileukemic activity of CsA.

MATERIALS AND METHODSBetween August 1998 and October 1999, 31 patientswith CLL and IM cytopenias were treated with CsA atThe University of Texas M. D. Anderson Cancer Cen-ter. The diagnosis of CLL was based on the NationalCancer Institute (NCI) working group criteria.22 Beforetreatment, the following parameters were investigat-ed: age, gender, serum albumin level, complete bloodcount (CBC), bone marrow aspiration, serum b2-mi-croglobulin levels, Coombs test, Rai stage,23 numberof prior therapies, time from diagnosis, Zubrod per-formance status, and prior therapy for cytopenias.

Eligibility criteria included: 1) adults with a diag-nosis of CLL at any stage of the disease, 2) cytopenia ofsuspected immune ethiology or that associated withprevious fludarabine treatment (i.e., occurring duringthe course of therapy with fludarabine-based regi-mens and not recovering after at least 4 weeks), 3)adequate hepatic (bilirubin # 2.0 mg/dL, alanine ami-notransferase , 300) and renal (creatinine , 2.0mg/dL or creatinine clearance $ 60 mL/minute) func-tion, and 4) signed informed consent according toinstitutional guidelines.

TreatmentPatients received CsA at a dose of 300 mg orally daily.Patients were followed with CBC, differential, andplatelets every 1–2 weeks, and sequential multipleanalysis–12-channel biochemical profile (SMA-12) ev-ery week until a stable dose of CsA was found, andthen every 2– 4 weeks. CsA levels were not measuredroutinely unless clinically indicated. The dose of CsAwas adjusted for toxicity according to the followingguidelines: 1) in patients whose serum creatinine levelrose to 2–3 mg/dL, CsA was withdrawn until the cre-atinine returned to normal and then was readminis-tered at two-thirds of the initial dose; and 2) in pa-tients whose serum creatinine levels rose to . 3 mg/dL, CsA was withdrawn until the creatinine returnedto normal, and then was readministered at 50% of theinitial dose. Patients did not receive any chemother-apy while receiving CsA until their response could beassessed.

Response CriteriaThe response criteria reported by Thiruvengadam etal.24 were used to assess response to CsA and arepresented in Table 1.

Statistical AnalysisThe data were analyzed as of December 1999. Survivalwas calculated from the date of the initiation of treat-ment with CsA and analyzed using the method re-ported by Kaplan and Meier.25 Patients who were aliveat the most recent follow-up visit were censored as ofthat date. Categoric data were compared using thechi-square test.26

RESULTSThirty-one patients were included in the current studyand their clinical characteristics are presented in Ta-ble 2. Their median age was 61 years (range, 49 – 81years). All but 3 patients had received prior therapy forCLL; 15 patients (48%) had received $ 2 prior treat-ment regimens (range, 0 – 8 treatment regimens). Six-

TABLE 1Hematologic Response Criteriaa

Response Hb (g/dL) Plts (3 109)

None 2, or 1 , 1 2, or 1 , 20Fair 1 , 3 1 , 50Good 1 $ 3, but Hb , 12 1 $ 50, but Plts , 150Excellent 1 $ 3, and Hb $ 12 1 $ 50, and Plts . 150

Hb: hemoglobin; Plts: platelets.a Based on the response criteria of Thiruvengadam et al.24

Cyclosporin A in CLL/Cortes et al. 2017

teen patients (52%) had anemia with a hemoglobinlevel # 11 g/dL and 29 patients (94%) had thrombo-cytopenia with a platelet count # 100 3 109/L at theinitiation of therapy. Seventeen patients had receivedfludarabine-based therapy (fludarabine alone [n 5 11]and fludarabine plus cyclophosphamide [n 5 6]).These patients had received a median of 4 courses offludarabine-based therapy prior to the initiation ofCsA therapy (range, 1– 6 courses); the median timefrom the last course was 8 weeks (range, 4 –12 weeks).Fourteen patients (45%) had cytopenias considered tobe autoimmune and not associated with fludarabinetherapy. The Coombs test was positive in 8 of 31patients (26%); 6 patients (19%) had low haptoglobinlevels. Antiplatelet antibodies were present in two ofthe nine patients in whom they were investigated.

Nineteen of 29 patients (66%) with thrombocyto-penia and 11 of 16 patients (69%) with anemia had afair to excellent response according to the criteriashown in Table 1. Of these, 12 patients (41%) and 9patients (56%), respectively, achieved an excellent re-sponse (Table 3). There was no significant differencein the overall response rate for the patients who hadnot received prior therapy for the cytopenias (6 of 12patients; 50%) compared with those patients who had

failed prior therapy with steroids, IVIG, and/or sple-nectomy (14 of 19 patients; 74%) (P 5 0.25). Thirteenof the 17 patients (76%) with postfludarabine cytope-nia had achieved some response, compared with 8 of14 patients (57%) whose cytopenia was not found tobe associated with fludarabine therapy (P 5 0.44).Four of the nonresponders (unrelated to fludarabine)had clear evidence of progressive CLL prior to theinitiation of treatment with CsA that required therapyshortly thereafter. The response per patient is shownin Table 4.

The median time to initial response was 3 weeks(range, 1–13 weeks), although the median time to thebest response was 10.5 weeks (range, 1– 48 weeks). Themedian duration of response was 10 months (range,11–391 months). After a median follow-up of 16months (range, 3–31 months), 23 patients were alive.Among the eight patients who died, seven died ofprogressive disease and one of postsplenectomy com-plications.

A major consequence of cytopenia developing inpatients treated with fludarabine is that often theplanned treatment cannot be delivered. Three of thepatients with fludarabine-associated cytopenia re-ceived further therapy with fludarabine while still re-ceiving CsA. In all three patients the platelet nadir wassignificantly better than before treatment with CsAwas instituted. A representative example is shown inFigure 1.

Treatment with CsA appeared to be well tolerated.The most common toxicities were elevation of creati-nine in 6 patients (19%), hypertension in 3 patients(10%), headache in 3 patients (10%), nausea in 2 pa-tients (6%), and gum hypertrophy, weight loss, diar-rhea, blurred vision, hematuria, and fluid retention in1 patient each (3%) (Table 5). In all cases toxicity was# Grade 2 according to the NCI Common ToxicityCriteria, version 2.0. Fourteen patients had the dose ofCsA adjusted due to toxicity and/or elevated plasmalevels of CsA. Three patients had infectious complica-tions while receiving CsA: two patients developedPneumocystis carinii pneumonia and the other patientdeveloped a pneumonia of an unknown pathogen. Allthree cases resolved with appropriate therapy.

For six patients treated with CsA in the currentstudy, some reduction in tumor burden was observed.

TABLE 2Patient Characteristics (n 5 31)

Characteristic Median Range

Age (yrs) 61 49–81Hemoglobin (g/dL) 10.5 6.5–16.1Leukocytes (3 109/L) 11 0.2–145Platelets (3 109/L) 48 4–433Bone marrow lymphocytes (%) 53 2.5–88b2-microglobulin (mg/dL) 3.8 1.4–8.2Time from diagnosis (mos) 36 1–160No. of prior therapies 1 0–8ECOG performance status

0 10 (32%)1 16 (52%)2 4 (13%)

Coombs direct test positive 8 (26%)Coombs indirect test positive 5 (16%)Prior therapy for cytopenias

Prednisone 17 (55%)IVIG 8 (26%)Splenectomy 5 (16%)None 12 (39%)

Indication for treatmentIM anemia 9 (29%)IM thrombocytopenia 12 (39%)PF anemia 7 (23%)PF thrombocytopenia 17 (55%)

ECOG: Eastern Cooperative Oncology Group; IVIG: intravenous immunoglobulin; IM: immune-medi-

ated; PF: postfludarabine.

TABLE 3Hematologic Response

Response No. evaluable Excellent Good Fair None

Platelet response 29 12 (41%) 6 (21%) 1 (3%) 10 (34%)Hemoglobin response 16 9 (56%) 1 (6%) 1 (6%) 5 (31%)

2018 CANCER October 15, 2001 / Volume 92 / Number 8

Five patients showed a mild to moderate decrease inthe leukocyte count from a median of 47 3 109/L(range, 14 –145 3 109/L) to 15 3 109/L (5– 80 109/L). Intwo of these patients the response was transient andthe leukocyte count started to increase after 5 monthswhile the patients still were receiving CsA. One patienthad a response lasting 3 months when CsA was with-drawn; the leukocyte count increased shortly after. Atlast follow-up, the fourth patient had had a sustainedresponse for 81 months. Two patients had a decreasein the size of their lymph nodes; one patient hadcervical lymphadenopathy measuring 2 cm that re-solved while the patient was receiving CsA; and asecond patient had cervical lymphadenopathy mea-suring 4 3 4 cm, axillary lymphadenopathy measuring6 3 6 cm, and inguinal lymphodenopathy measuring 23 2 cm that decreased to 2 3 2 cm, 4 3 4 cm, and 1 3 1

cm, respectively. In both patients the response wastransient.

DISCUSSIONDuring the natural history of CLL, some patients de-velop anemia and/or thrombocytopenia. This can beIM or a manifestation of disease progression.1 Someauthors have suggested that immune dysregulationfrom treatment with fludarabine may induce autoim-mune anemia and/or thrombocytopenia.3 CsA hasbeen used with good results in patients with CLL-associated pure red cell aplasia,11–17 and both a T-cell-dependent and a T-cell-independent effect have beenproposed.14 We believe the results of the current studyconfirm the efficacy of CsA in the treatment of patientswith autoimmune complications of CLL, as had beensuggested by small series and isolated case re-ports.18 –21 In the current study, we treated 31 patientswith anemia or thrombocytopenia of a suspected im-mune mechanism. Nineteen of 29 patients (66%) withthrombocytopenia and 11 of 16 patients (69%) withanemia responded, with the majority of the responsesconsidered to be good or excellent according to thecriteria presented in Table 1. The mechanism of actionwas unclear, because the mechanism of the autoim-mune manifestations of CLL is not completely under-stood. Mouzaki et al. reported one patient with B-CLLin whom polyclonal immunoglobulin (IG) G was se-creted by normal B cells that in turn were stimulatedby interleukin-2 produced by leukemic B cells. UsingCsA treatment, both Ig secretion and the B-cell countdecreased, resulting in an improvement in the auto-immune complications.27

Steroids are the standard treatment for patientswith CLL and IM cytopenia. Approximately 67% ofpatients with IM thrombocytopenia or hemolytic ane-mia will respond to steroids, most frequently pred-nisone. In patients with refractory disease, IVIG andsplenectomy are among the few available alterna-tives.10 IVIG may induce responses in some patientswith refractory disease, and might have an additionalbenefit for patients with severe hypogammaglobuline-mia associated with recurrent infections requiringhospitalizations. However, the benefits of IVIG fre-quently are transient. The majority of patients in-cluded in the current study (61%) had received andfailed prior therapy with steroids and/or IVIG or sple-nectomy. The response rate was similar between thosepatients who had not received prior therapy for theircytopenia and those who had failed prior interven-tions (overall response rate of 50% vs. 74%; P 5 0.25),a finding that suggests that the high response rate wasnot due to the inclusion of previously untreated pa-tients. Furthermore, four patients included in the trial

TABLE 4Response Per Patient Showing Baseline and Posttreatment(Best) Values

Patient

Prior therapyHemoglobin(g/dL)

Platelets(3 109/L)

Splenectomy Prednisone IVIG Pre Post Pre Post

1 No No No 14.2 — 59 1522 No Yes No 13.2 — 29 1763 No No Yes 10.5 12.5 51 1554 No No Yes 7.5 12.4 9 2225 No Yes No 14.2 — 39 1486 No Yes Yes 10.4 13.5 99 1837 Yes Yes Yes 8.8 14 82 1888 No No No 11.6 — 59 829 No No No 12.9 — 47 4510 Yes Yes No 9.6 7.7 12 1311 No No No 9.3 8.8 47 11412 Yes Yes No 8.3 8.4 433 —13 No Yes No 6.7 13.1 92 15814 No Yes No 10.5 14.8 67 11315 No Yes No 8.3 12.3 190 —16 No No No 8 8.2 9 1217 No No No 8.1 14.6 12 20418 No No No 6.5 13.2 67 14019 No Yes Yes 11.5 — 46 7720 No Yes Yes 13.7 — 9 921 No No No 11.6 — 52 6922 No No No 11.9 — 30 9323 No No No 10 15.7 61 25024 No Yes Yes 8.5 12.3 30 20425 No No No 7.2 7.7 91 10926 Yes Yes Yes 13.9 — 21 62727 No Yes No 15.9 — 56 5728 No Yes No 14.1 — 45 37229 No No No 13 — 48 5530 Yes Yes No 13.2 — 4 431 No Yes No 16.1 — 56 131

IVIG: intravenous immunoglobulin; Pre: pretreatment; Post: posttreatment.

Cyclosporin A in CLL/Cortes et al. 2019

whose cytopenias did not respond to CsA had cytope-nias that most likely were related to progression ofdisease (i.e., no autoantibodies present, and progres-sive disease evident before the initiation of CsA, whichdid not appear to alter the course of their disease).Splenectomy also has been used successfully in themanagement of CLL-associated cytopenias.24,28 Thiru-vengadam et al.24 reported a series of 30 patients withCLL and 3 patients with prolymphocytic leukemiawho underwent splenectomy; 20 of the 23 patients(87%) with thrombocytopenia and 12 of the 17 pa-tients (71%) with anemia achieved a good or excellentresponse.24 In a series of 55 patients with CLL who hadundergone splenectomy reported by Seymour et al.,28

24 patients (44%) had a hemoglobin level , 11 g/dLand 27 patients (49%) had a platelet count , 1003 109/L. Using the same response criteria, the majorresponse rate (good plus excellent) was 38% and 81%,respectively.28 Other series have reported a plateletimprovement to $ 50 3 109/L in 53–91% of patients,and hemoglobin improvements . 2–3 g/dL in 56 – 80%of patients.29 –33 In the current series we report a majorresponse rate (i.e., an increase of $ 50 3 109/L in theplatelet count and $ 3 g/dL in the hemoglobin level)of 62% for patients with thrombocytopenia and 63%for patients with anemia. Five patients had undergonesplenectomy previously; 2 of 4 patients evaluable forthrombocytopenia and 1 of 3 patients evaluable foranemia achieved an excellent response. Therefore,CsA appears to be a good alternative to splenectomy.

Response to CsA occurred after a median of 3weeks, but the median time to best response was 10.5weeks. This is a pattern similar to that observed withresponse to steroids. Mauro et al.34 reported a majorresponse rate of 84% among 50 evaluable patientswith CLL and autoimmune hemolytic anemia (all ofwhom were Coombs test positive). The median time toa hemoglobin level $ 12 g/dL was 4.5 months, and thedisappearance of autoantibodies occurred after a me-dian of 6 months, including 1 patient in whom thisoccurred after 24 months of therapy.34 After splenec-tomy, a significant change in the platelet count usuallyis observed as early as 7 days after the procedure,although a change in hemoglobin levels is slower in

TABLE 5Toxicity of CsA Therapya

Event No. (%)

Creatinine elevation 6 (19)Hypertension 3 (10)Headache 3 (10)Nausea 2 (6)Gum hypertrophy 1 (3)Hematuria 1 (3)Weight loss 1 (3)Diarrhea 1 (3)Blurred vision 1 (3)Fluid retention 1 (3)

CsA: cyclosporin A.a All events were # Grade 2 according to the National Cancer Institute’s Common Toxicity Criteria.

FIGURE 1. Treatment with cyclosporin

A in a patient with fludarabine-associ-

ated thrombocytopenia. Plts: platelets;

WBC: white blood cell (leukocyte) count.

2020 CANCER October 15, 2001 / Volume 92 / Number 8

onset.28 The responses to CsA are durable, with amedian duration of 10 months. The possibility of aspontaneous recovery from myelosuppression has tobe considered because some patients had receivedprior chemotherapy. However, the median time fromprior chemotherapy was 8 months and the regimensused are not known to be highly myelosuppressive.Several investigators have suggested that cytopeniasassociated with fludarabine frequently are IM.3–9

In the current study, treatment with CsA was welltolerated, with elevation of creatinine being the mostcommon side effect. However, all side effects were# Grade 2 and transient. Three patients (10%) devel-oped infectious complications while receiving CsA (all3 had pneumonia). This compares favorably with a52% incidence of infections (40% incidence of pneu-monia) with the use of prednisone in patients withautoimmune hemolytic anemia associated with CLL.34

In some patients, a direct antileukemic effect wasnoted as measured by a decrease in the absolute lym-phocyte count and lymph nodes. Similar results havebeen reported previously,17,18,20,21 but others have dis-puted this effect19,35 and even suggested that CsA mayinduce proliferation of CLL.36 The results presented inthe current study suggest that a small percentage ofpatients may indeed experience an antileukemic ef-fect, which usually is transient and modest. Possibleexplanations for this effect have been proposed. CsAhas a significant effect against T-cells, suppressing theproduction of a number of cytokines including IL-2,IL-4, tumor necrosis factor (TNF)-a, and interferon(IFN)-g.37 Dancescu et al. reported that IL-4 preventedCLL cells from undergoing spontaneous and hydro-cortisone-induced apoptosis.38 This was reflected byan increased expression of bcl-2.38 Buschle et al.showed a similar effect with IFN-g.39 Schmid et al.reported that CsA could inhibit the IL-4- or TNF-a-induced proliferation of CLL cells in vitro, and in somecases their spontaneous proliferation.40 Thus, it is pos-sible that CsA might have some antileukemic effect inCLL through suppression of these cytokines and cyto-kine-induced growth. A nonimmunosuppressive ana-logue of CsA, PSC-833, has been shown to have simi-lar, although weaker, cytotoxic activity against CLL ornon-Hodgkin lymphoma cells in vitro.41

We conclude that CsA is an effective therapy forCLL patients with IM cytopenias. Response to CsAfrequently translated into improved tolerance to fur-ther chemotherapy for CLL, and in some cases a mod-est antileukemic effect can be noted. CsA should beconsidered for the treatment of these complications,and further studies into the mechanisms of the anti-leukemic effects of CsA might provide some insightinto the role of cytokines in CLL.

REFERENCES1. Keating MJ. Chronic lymphocytic leukemia. Semin Oncol

1999;26(5 Suppl 14):107–14.2. De Rossi G, Granati L, Girelli G, Gandolfo G, Arista MC,

Martelli M, et al. Incidence and prognostic significance ofautoantibodies against erythrocytes and platelets in chroniclymphocytic leukemia (CLL). Nouv Rev Fr Hematol 1988;30(5– 6):403– 6.

3. Juliusson G. Complications in the treatment of CLL withpurine analogues. Hematol Cell Ther 1997;39(Suppl 1):S41– 4.

4. Sen K, Kalaycio M. Evan’s syndrome precipitated by fludara-bine therapy in a case of CLL [letter]. Am J Hematol 1999;61(3):219.

5. Gonzalez H, Leblond V, Azar N, Sutton L, Gabarre J, Binet JL,et al. Severe autoimmune hemolytic anemia in eight pa-tients treated with fludarabine. Hematol Cell Ther 1998;40(3):113– 8.

6. Shvidel L, Shtarlid M, Klepfish A, Sigler E, Berrebi A. Evanssyndrome complicating fludarabine treatment for advancedB-CLL [letter]. Br J Haematol 1997;99(3):706.

7. Montillo M, Tedeschi A, Centurioni R, Leoni P. Treatment ofrelapsed adult acute lymphoblastic leukemia with fludara-bine and cytosine arabinoside followed by granulocyte col-ony-stimulating factor (FLAG-GCSF). Leuk Lymphoma 1997;25(5– 6):579 – 83.

8. Bay JO, Fouassier M, Beal D, Alcaraz L, Cure H, Chollet P, etal. Autoimmune thrombocytopenia after six cycles of flu-darabine phosphate in a patient with chronic lymphocyticleukemia. Hematol Cell Ther 1997;39(4):209 –12.

9. Woodcock BE. Cytopenia and fludarabine [letter; com-ment]. Lancet 1993;342(8878):1049.

10. O’Brien S, del Giglio A, Keating M. Advances in the biologyand treatment of B-cell chronic lymphocytic leukemia.Blood 1995;85(2):307–18.

11. Cesana C, Carlo-Stella C, Mangoni L, Almici C, Giachetti R,Zappa M, et al. Response to cyclosporin A and recombinanthuman erythropoietin in a case of B cell chronic lympho-cytic leukemia and pure red cell aplasia. Leukemia 1996;10(8):1400 –1.

12. Tura S, Finelli C, Bandini G, Cavo M, Gobbi M. CyclosporinA in the treatment of CLL associated PRCA and bone mar-row hypoplasia. Nouv Rev Fr Hematol 1988;30(5– 6):479 – 81.

13. Finelli C, Bandini G, Ricci P, Giudice V, Vianelli N, Rabbi C,et al. Ciclosporin A in idiopathic and CLL-associated purered cell aplasia. Haematologica 1987;72(6):537– 40.

14. Yamada O, Yun-Hua W, Motoji T, Mizoguchi H. ClonalT-cell proliferation causing pure red cell aplasia in chronicB-cell lymphocytic leukaemia: successful treatment with cy-closporine following in vitro abrogation of erythroid colony-suppressing activity. Br J Haematol 1998;101(2):335–7.

15. Gotic M, Basara N, Rolovic Z, Boskovic D, Antunovic P,Marisavljevic D, et al. Successful treatment of refractorypure red cell aplasia secondary to chronic lymphocytic leu-kaemia with cyclosporine A: correlation between clinicaland in vitro effects. Nouv Rev Fr Hematologie 1994;36(4):307–9.

16. Sivakumaran M, Bhavnani M. Cyclosporin therapy for purered cell aplasia in patients with chronic lymphocytic leuke-mia [letter; comment]. Am J Hematol 1994;45(2):192.

17. Chikkappa G, Pasquale D, Zarrabi MH, Weiler RJ, DivakaraM, Tsan MF. Cyclosporine and prednisone therapy for purered cell aplasia in patients with chronic lymphocytic leuke-mia. Am J Hematol 1992;41(1):5–12.

Cyclosporin A in CLL/Cortes et al. 2021

18. Ruess-Borst MA, Waller HD, Muller CA. Successful treat-ment of steroid-resistant hemolysis in chronic lymphocyticleukemia with cyclosporine A [letter]. Am J Hematol 1994;46(4):375– 6.

19. Emilia G, Messora C, Bensi L. The use of cyclosporin-A inthe treatment of B-chronic lymphocytic leukemia [letter;comment]. Leukemia 1995;9(2):357–9.

20. Wright SJ, Keating MJ. Cyclosporine A in chronic lympho-cytic leukemia: dual anti-leukemic and immunosuppressiverole? Leuk Lymphoma 1995;20(1–2):131– 6.

21. Bergui L, Gregoretti MG, Caligaris-Cappio F. Cyclosporin Ain the treatment of B-chronic lymphocytic leukemia (B-CLL)[letter]. Leukemia 1994;8(7):1245– 6.

22. Cheson BD, Bennett JM, Grever M, Kay N, Keating MJ,O’Brien S, et al. National Cancer Institute-sponsored Work-ing Group guidelines for chronic lymphocytic leukemia: re-vised guidelines for diagnosis and treatment. Blood 1996;87(12):4990 –7.

23. Rai KR, Sawitsky A, Cronkite EP, Chanana AD, Levy RN,Pasternack BS. Clinical staging of chronic lymphocytic leu-kemia. Blood 1975;46(2):219 –34.

24. Thiruvengadam R, Piedmonte M, Barcos M, Han T, Hender-son ES. Splenectomy in advanced chronic lymphocytic leu-kemia. Leukemia 1990;4(11):758 – 60.

25. Kaplan EL, Meier P. Non-parametric estimation from in-complete observations. J Am Stat Assoc 1953;53:457– 81.

26. Snedcor GW, Cochran WG. Statistical methods. Ames, IA:Iowa State University Press, 1980.

27. Mouzaki A, Matthes T, Miescher PA, Beris P. Polyclonalhypergammaglobulinaemia in a case of B-cell chronic lym-phocytic leukaemia: the result of IL-2 production by theproliferating monoclonal B cells? Br J Haematol 1995;91(2):345–9.

28. Seymour JF, Cusack JD, Lerner SA, Pollock RE, Keating MJ.Case/control study of the role of splenectomy in chroniclymphocytic leukemia. J Clin Oncol 1997;15(1):52– 60.

29. Ferrant A, Michaux JL, Sokal G. Splenectomy in advancedchronic lymphocytic leukemia. Cancer 1986;58(9):2130 –5.

30. Merl SA, Theodorakis ME, Goldberg J, Gottlieb AJ. Splenec-tomy for thrombocytopenia in chronic lymphocytic leuke-mia. Am J Hematol 1983;15(3):253–9.

31. Adler S, Stutzman L, Sokal J, Mittelman A. Splenectomy for

hematologic depression in lymphocytic lymphoma and leu-kemia. Cancer 1975;35(2):521– 8.

32. Strumia MM, Strumia PV, Bassert D. Splenectomy in leuke-mia: hematologic and clinical effects on 34 patients andreview of 299 published cases. Cancer Res 1966;26(3):519 –28.

33. Neal TF Jr., Tefferi A, Witzig TE, Su J, Phyliky RL, NagorneyDM. Splenectomy in advanced chronic lymphocytic leuke-mia: a single institution experience with 50 patients. Am JMed 1992;93(4):435– 40.

34. Mauro FR, Foa R, Cerretti R, Giannarelli D, Coluzzi S, Man-delli F, et al. Autoimmune hemolytic anemia in chroniclymphocytic leukemia: clinical, therapeutic, and prognosticfeatures. Blood 2000;95(9):2786 –92.

35. Van Den Neste E, Maerevoet M, Martiat P, Ferrant A, Del-annoy A, Michaux JL. Cyclosporin A in the treatment ofrefractory B chronic lymphocytic leukemia (B-CLL). Leuke-mia 1995;9(6):1102–3.

36. Nishii K, Katayama N, Mitani H, Matsumoto T, Miwa H, KitaK, et al. Effects of cyclosporin A on refractory B-cell chroniclymphocytic leukemia. Int J Hematol 2000;71(1):59 – 65.

37. Andersson J, Nagy S, Groth CG, Andersson U. FK 506 andcyclosporine inhibit antigen- or mitogen-induced monokineand lymphokine production in vitro. Transplant Proc 1992;24(1):321–5.

38. Dancescu M, Rubio-Trujillo M, Biron G, Bron D, DelespesseG, Sarfati M. Interleukin 4 protects chronic lymphocyticleukemic B cells from death by apoptosis and upregulatesBcl-2 expression. J Exp Med 1992;176(5):1319 –26.

39. Buschle M, Campana D, Carding SR, Richard C, HoffbrandAV, Brenner MK. Interferon gamma inhibits apoptotic celldeath in B cell chronic lymphocytic leukemia. J Exp Med1993;177(1):213– 8.

40. Schmid M, Merk B, Porzsolt F. Cyclosporin A inhibits cyto-kine-induced proliferation in B-chronic lymphocytic leuke-mia. Leuk Lymphoma 1994;15(3– 4):317–25.

41. Martinsson P, Nygren P, Fridborg H, Nilsson K, Kristensen J,Larsson R. Cytotoxic activity of cyclosporin A and [3-keto-Bmt1]-[Val2]-cyclosporin (SDZ PSC 833) on tumour cellsfrom patients with haematological malignancies. Eur J ClinPharmacol 1997;52(3):199 –203.

2022 CANCER October 15, 2001 / Volume 92 / Number 8