LETTERS ANDCORRESPONDENCE

Letters and correspondence submitted for possible publication mustbe identified as such. Text length must not exceed 500 words andfive bibliographic references. A single concise figure or table may beincluded if it is essential to support the communication. Letters nottyped double-spaced will not be considered for publication. Letters notmeeting these specifications will not be returned to authors. Letters tothe Editor are utilized to communicate a single novel observation orfinding. Correspondence is to be used to supplement or constructivelycomment on the contents of a publication in the journal and cannotexceed the restrictions for Letters to the Editor. The Editor reservesthe right to shorten text, delete objectional comments, and makeother changes to comply with the style of the journal. Permission forpublication must be appended as a postscript. Submissions must besent to Paul Chervenick, M.D., Editor of Brief Reports/Letters toEditors, American Journal of Hematology, H. Lee Moffitt CancerCenter, University of South Florida, 12902 Magnolia Drive, Tampa,FL 33612 to permit rapid consideration for publication.

Repeated Efficacy of all-trans- Retinoic Acid in an AcutePromyelocytic Leukemia Patient

To the Editor:Differentiation therapy withall-trans-retinoic acid (ATRA)has marked a major advance and is now the first in the treatment of acutepromyelocytic leukemia (APL). However, patients who relapse afterATRA-induced complete remission (CR) have difficulty in obtaining asecond CR with a second course of ATRA therapy alone [1]. We report apatient diagnosed with relapsed APL who succeeded in achieving ATRA-induced CR four times.

A 56-year-old man was diagnosed with APL with chromosomal abnor-mality t(15;17)(q22;q11) in May of 1995. He had typical APL cells withmany Auer bodies and faggot cells and was complicated with disseminatedintravascular coagulation (DIC). He received 70 mg/day (45 mg/m2/day) oforal ATRA and 40 mg of daunorubicin (DNR) for 2 days on treatment days12 and 13, when the white blood cell (WBC) count was elevated to 30 ×109/l. He achieved CR on treatment day 39, when the chromosomal ab-normality had disappeared. He received consolidation chemotherapies[DNR, enocitabine (BHAC), 6-mercaptopurine (6MP), and prednisolone]twice, but after that he arbitrarily discontinued intensification therapy.

In December of 1996, he relapsed. He received oral ATRA (80 mg/day)and achieved CR again on treatment day without 40 using DNR. He re-ceived consolidation therapy consisting of DNR, BHAC, and 6MP fol-lowed by high-dose Ara-C. He subsequently received high-dose chemo-therapy (Ara-C, busulfan, and etoposide) with autologous peripheral bloodstem cell support.

In May of 1998, he relapsed again with the same chromosomal abnor-mality t(15;17)(q22;q11). He received oral ATRA (80 mg/day) and 300mgof intravenous G-CSF (treatment days 2–15 and 30–42) and 40 mg of DNRfor 4 days (treatment days 5, 6, 13, and 14), when the WBC count waselevated to 15 × 109/l [2]. He again achieved CR on treatment day 59. He

received consolidation chemotherapy (ICE; Ara-C, idarubicin, etoposide)and subsequent intensification chemotherapy with an alternating regimenof ICE and middle-dose Ara-C with mitoxantrone.

In November of 1999, a bone marrow study showed 46% atypical APLcells with rare Auer bodies and azurophilic granules. No additional chro-mosomal abnormality was found besides t(15;17)(q22;q11). Spinal taprevealed central nervous system invasion by APL cells. He was treatedwith oral ATRA (80 mg/day) and intrathecal administration of methotrex-ate. He achieved a fourth CR on treatment day 61.

In general, APL patients who have relapsed from ATRA-induced CRrarely achieve a second CR with ATRA therapy alone [1]. However, ourpresent patient achieved ATRA-induced CR four times. To our knowledge,there is no previous report of such a patient. The mechanisms of resistanceto second ATRA therapy have been studied in many laboratories [3] but arenot yet fully understood. The reason for the repeated efficacy of ATRA inour present case is unclear, but the duration of ATRA treatment necessaryto achieve CR became longer with each repetition of the therapy.

AYAKO WATANABE

KOITI INOKUCHI

TAROH MIZUKI

HIROKI YAMAGUCHI

NORIO YOKOSE

KAZUO DAN

Division of Hematology, Department of Internal Medicine, NipponMedical School, Tokyo, Japan

REFERENCES

1. Castaigne S, Chomienne C, Daniel MT, et al.all-trans-Retinoic acid as a differ-entiation therapy for acute promyelocytic leukemias. I. Clinical results. Blood1990;76:1704.

2. Nakajima K, Hatake K, Miyata T, et al. Acute promyelocytic leukemia, tretinoin,and granulocyte colony-stimulating factor. Lancet 1994;343:173–174.

3. Degos L, Dombret H, Chomienne C, et al.all-trans-Retinoic acid as a differen-

Fig. 1. Clinical course of the patient. Horizontal bars atupper column indicate chemotherapies.

American Journal of Hematology 65:87–89 (2000)

© 2000 Wiley-Liss, Inc.

tiating agent in the treatment of acute promyelocytic leukemia. Blood 1995;85:2643–2653.

Coexistence of Factor V 1691 G–A and Factor V 4070 A–GMutation in Turkish Thromboembolic Patients

To the Editor:A mutation in the factor V gene (1691 G–A) was identifiedthat formed the molecular explanation for the phenotype of APC-resistancein the majority of effected individuals. This mutation, which is associatedwith a significant increase in thrombotic risk, has been found in 30–50% ofselected families with thrombophilia and in almost 20% of consecutivepatients with venous thrombosis [1].

A4070G (FV1299 His-Arg) polymorphism in exon 13 of the factor Vgene (HR2 haplotype) was shown to influence circulating FV levels andcontribute to the activated protein C (APC) resistance phenotype [2–4].Although previous reports on the effect of A4070G on the occurrence ofdeep vein thrombosis were controverisal [5,6] double heterozygosity forFV1691A and FV4070G conferred a 3–4-fold increase in the relative riskof venous thromboembolism compared with FVR506Q alone [7].

We recently reported the frequency of FV1299G allele as 8.5% andFV1691A allele as 9.8% in Turkish population [6,8]. As both mutations iscommon in the Turkish population, we aimed to study the relative risk ofthe coexistence of these mutations in Turkish thromboembolic patients.

One hundred fourty-four healthy unrelated individuals from Ankarawithout any familial history of thrombosis and stroke were included in thestudy. One hundred twenty-nine patients with the diagnosis of thrombo-embolism were included. DNA was extracted by conventional methods,and polymerase chain reaction of exon 13 of the factor V gene was per-formed according to previously described method using primers58CAAGTCCTTCCCCACAGATATA38 and 58AGATCTG-CAAAGAGGGGCAT38. Amplication was performed for 35 cycles withannealing temperature of 57°C (Ericomp, USA). Amplified DNA was di-gested withRsaI enzyme (Promega, Madison WI) and 37°C and subjectedto 2% agarose gel electrophoresis [6]. FV1691 G–A and PT20210 G–Amutations were performed according to previously described methods[8,9].

The results of the FV mutations in healthy individuals and patients withthromboembolism are shown in Table I. Three of these six patients carriedan additional prothrombotic factor. FV4070G mutation did not have anyeffect on thromboembolism with an odds ratio of 1.2 (CI 95%, 0.56–2.56).On the other hand, it was 1.93 (CI 95%, 0.1–3.8) for FV1691A. When bothmutations taken together, it was 6.7 (CI 95%, 0.82–54.6).

Of the symptomatic 26 FV1691A carriers, 6 had the FV4070G mutation(23.0%). It is interesting that of these 6 patients, two patients carried theprothrombin 20210A mutation and one patient had protein C deficiency atthe same time. Two of the PT20210A carriers had mesenteric artery throm-bosis. The protein C-deficient patient was a four-year-old female child withthe diagnosis of cerebral thrombosis. The other three had clinical presen-tation of cerebral infarct, vascular graft thrombosis, and Budd-Chiari Syn-drome.

Our data revealed once more that carrying FV1691A is a risk factor forthromboembolism but FV4070G is not [6,8]. It is worth noting that the

frequency of FV4070G carriers is about one-fourth of the patients withFV1691A.

We conclude that the 4070G allele of the factor V gene is frequentlycoinherited in symptomatic FV1691A carriers. Thus a careful search forthe 4070G allele should be included in thrombophilia screening programsin FV1691A carriers, particularly in populations with high frequencies ofthis mutation.

NEJAT AKAR

ECE AKAR

ERKAN YILMAZ

Pediatric Molecular Genetic Department of Ankara University,Ankara, Turkey

REFERENCES

1. Bertina RM, Reitsma RH, Rosendaal FR, Vanderbroucke JP. Resistance to acti-vated protein C and factor V Leiden as risk factor for venous thrombosis. ThrombHaemost 1995;74:449–453.

2. Bernardi F, Faioni EM, Castoldi E, Lunghi B, Castaman G, Sacchi E, ManucciPM. A factor V genetic component differing from factor V R506Q contributes tothe activated protein C resistance phenotype. Blood 1997;90:1552–1557.

3. Castaman G, Lunghi B, Missiaghia E, Bernardi F, Rodeghiero F. Phenotypichomozygous activated protein C resistance associated with compound heterozy-gosity for Arg 506 Gln and His 1299 Arg substitutions in factor V. Br J Haematol1997;99:257–261.

4. Lunghi B, Iacoviello L, Gemmati D, di Iasio MG, Castoldi E, Pinotti M, CastamanG, Redaelli R, Mariani G, Marchetti G, Bernardi F. Detection of new polymorphicmarkers in the factor V gene: association with factor V levels in plasma ThrombHaemost 1996;75:45–48.

5. Alhenc-Gelas M, Nicaud V, Gandrille S, van Dreden P, Amiral J, Aubry ML,Fiessinger JN, Emmerich J, Aiach M. The factor V gene A4070G mutation and therisk of venous thrombosis. Thromb Haemost 1999;81(2):193–197.

6. Akar N, Akar E, Yılmaz E. Factor V (His 1299 Arg) in Turkish patients withvenous thromboembolism. Am J Hematol 2000;63(2):102–103.

7. Faioni EM, Franchi F, Bucciarelli P, Margaglione M, De Stefano V, Castaman G,Finazzi G, Mannucci PM. Coinheritance of the HR2 haplotype in the factor V geneconfers an increased risk of venous thromboembolism to carriers of factor VR506Q. Blood 1999;94(9):3062–3066.

8. Akar N, Akar E, Dalgın G, So¨zuoz A, Omurlu K, Cin S. Frequency of factor V(1691 G–A) mutation in Turkish Population. Thromb Haemost 1997;78:1527–1528.

9. Akar N, Mısırlıoglu M, Akar E, Avcu F, Yalcın A, Sozuoz A. Prothrombin gene20210 G–A mutation in the Turkish Population. Am J Hematol 1998;58:249.

Spontaneous Regression of Chronic LymphocyticLeukemia and Simultaneous Development of AutoimmuneHemolytic Anemia and Autoimmune Thrombocytopenia

To the Editor:Spontaneous remission of chronic lymphocytic leukemia(CLL) is extremely rare, although chemotherapy with alkylating agents,purine analogues, and corticosteroids often induces temporary remission.

We describe the case of a 77-year-old man who developed spontaneousregression of CLL without any treatment. Simultaneously, he developedautoimmune hemolytic anemia (AIHA) and autoimmune thrombocytope-

TABLE I. Distribution of FV1691A and FV4070G Mutations in Turkish Population

N FV1691A (%)FV1691Afrequency FV4070G (%)

FV4070Gfrequency

Bothmutations (%)

Normal controls 144 15 10.4 0.052 14 9.7 0.048 1 0.7Thromboembolic patients 129 26(2)a 20.1 0.1085 15 11.6 0.058 6 4.6

aTwo patients were homozygous FV1691A.

88 Letters and Correspondence

nia (AIT). Patient was asymptomatic when initially seen. There was nolymphadenopathy or hepatosplenomegaly. WBC was 35,000/ml with 80%lymphocytes, absolute lymphocyte count 28,480/ml, Hgb 14.2 g/dl, Hct41%, and platelets 212,000/ml. Bone marrow lymphocyte was >50%, andimmunophenotype of the bone marrow cells was CD20+, CD19+, CD5+,B-cell. He was diagnosed with CLL, stage 0. He did not receive anychemotherapy for CLL. During 3 years of follow-up, WBC and lympho-cyte counts declined gradually while anemia and thrombocytopenia devel-oped. After 3 years, WBC was 3,000/ml with 47% lymphocytes (absolutelymphocyte count 1,410/ml), 33% neutrophils, 17% monocytes, 2% eo-sinophils, and 1% basophils. Hgb was 8.3 g/dl, Hct 23.5%, platelets74,000/ml, MCV 105.5 fl, MCH 37.5 pg. Reticulocyte was 3.68%. Periph-eral blood smear showed large platelets, spherocytes, polychromatophiliccells, hypogranular neutrophils, and a few small lymphocytes.

Bone marrow aspiration and biopsy showed hypocellularity (15%), i.e.,decreased numbers of erythroid progenitor cells, myeloid precursor cells,and megakaryocytes. Residual foci of lymphocytes revealed CD19+/CD5+, B-cell. Cytogenetic examination was not done. Serum B12 and folicacid were normal. Blood chemistry and urinalysis were normal. Blood typewas O, Rh+. The patient developed shortness of breath and further anemia.Type and cross-match of red blood cells revealed IgG warm antibodies.Both direct and indirect Coomb’s tests were positive. He received themost compatible 2 units of available type-specific blood with prednisone(1 mg/kg/day). He responded to prednisone. His Hgb rose to 12.9/dl, Hctto 38%, and platelets to 235,000/ml. WBC also increased to 14.400/ml with74% lymphocytes (absolute lymphocyte count 10,700/ml), 18% neutro-phils, and 6% monocytes.

Regression of CLL with development of AIHA and AIT has neverreported in English written literature.

CLL is not only a malignant disease but also a complex immunologicdisease. The paradoxical findings of immune deficiency and autoimmunephenomena have been hallmarks of CLL. Autoimmune-associated phe-nomena are frequently observed in CLL. These autotoxic manifestationsare mainly directed against hematopoietic cells [1]. Spontaneous regressionof CLL is an extremely rare event [2–4]. The mechanism is poorly under-stood. The remission-associated event was infection, mainly viral, vacci-nation, and epithelial neoplasms before spontaneous remission was docu-

mented [2]. These suggest that spontaneous remission in CLL is the resultof an altered host–tumor relationship that seems to play a major role indisease regression.

The potential role of T-cell defects in inducing autoimmune complica-tions in B-cell CLL has been stressed by increased frequency of AIHA inpatients treated with purine nucleoside analogues like fludarabine and2-chlorodeoxyadenosine [5]. These drugs induce severe depletion of theCD4 cell subset and, to a lesser extent, the CD8 subset.

In this case regression was not related to chemotherapy, infection, orother neoplasm. The mechanism for the remission in the current case isunknown. It is hypothesized that spontaneous remission of CLL was theresult of an altered host–tumor relationship. In the absence of chemo-therapy in this patient, lymphocyte production may have been under thecontrol of a readjusted hematopoietic mechanism. CLL in bone marrowevidently altered stroma function and induced stromal abnormalities thatselectively suppressed lymphocyte production and induced AIHA and AIT.Prednisone altered this autoimmune phenomenon. CLL reappeared withimprovement of AIHA and AIT.

TAKESHI WAJIMA

Texas A&M University Health Science Center, College of Medicine andCentral Texas Veterans Health Care System, Temple, Texas

REFERENCES

1. Prisch O, Malourna K, Dighiero G. Basic biology of autoimmune phenomena inchronic lymphocytic leukemia. Semin Oncol 1998;25:34.

2. Ribera JM, Vinolas N, Urbano-Ispizua A, Montserrat E, Rozman C. Spontaneouscomplete remissions in chronic lymphocytic leukemia: report of three cases andreview of the literature. Blood Cells 1987;12:471–483.

3. Holmes JA, Whittaker JA. Spontaneous remission in chronic lymphocytic leuke-mia. Brit J Haematol 1988;69:97–98.

4. Denes AE, Shalhav AL, Kovacs G, Ralph V. Chronic lymphocytic leukemiaremission following extra corporeal shock wave lithotripsy for urinary calculi. AmJ Hematol 1998;58:239–240.

5. Myrnt H, Copplestone JA, Orchard J, Graig V, Curtis D, Prentice AG, HamonMD, Oscier DG, Hamblin J. Fludarabin–related autoimmune haemolytic anaemiain patients with chronic lymphocytic leukemia. Br J Haematol 1995;91:341.

Letters and Correspondence 89