p15ink4b is not mutated in adult familial myelodysplastic syndromes

Correspondence

POSITION PAPER ON IMATINIB MESYLATE IN CHRONIC MYELOID LEUKAEMIA*

This document summarizes the position of the BritishSociety for Haematology, representing haematologists inthe UK, on the use of the new ABL tyrosine kinaseinhibitor imatinib mesylate in the treatment of chronicmyeloid leukaemia (CML). It is not our intention to reviewcomprehensively all clinical data on the use of imatinib assuch data are available from other sources (Druker et al,2001a,b, 2002; Kantarjian et al, 2002; Ottmann et al,2002; Sawyers et al, 2002; Talpaz et al, 2002). We arenot recommending guidelines for the overall managementof CML – such a document will be produced by the BCSHlater this year. We do summarize the new data onimatinib that were presented at the recent meeting of theAmerican Society for Clinical Oncology on 20 May 2002.It should be emphasized that allografting remains theonly known ‘curative’ therapy for CML and this shouldbe considered in all younger patients with a suitabledonor.

The BSH believes that imatinib should be broadlyavailable for prescription by UK Haematologists as it is ahighly effective agent in the treatment of CML. It is easy toadminister in an outpatient setting, offers patients improvedquality of life and has fewer side-effects than the currentstandard treatment for non-transplanted patients, inter-feron alpha (IFN) (+ ⁄ – cytarabine, Ara-C). Given theoutstanding cytogenetic responses achieved, it seems rea-sonable to expect that imatinib will improve survival. Inlight of current evidence, imatinib is indicated in thetreatment of CML in the following settings:1. At presentation. Haematologists in the UK should have

the option to treat any newly diagnosed patient withimatinib. Ideally, all appropriate ⁄ eligible patients shouldbe entered into the multinational SPIRIT (STI571 Pros-pective International Randomized Trial) protocol (section4) or other suitable prospective study. The MRC declinedfunding for SPIRIT on 24 July 2002 due to insufficientfunds. Alternative funding possibilities are currentlybeing explored to allow SPIRIT to proceed in the UK.

2. Patients currently treated with IFN who are intolerant ofthis treatment should be offered imatinib. Any definitionsof intolerance would be arbitrary but clinician andpatient should make reasonable attempts to deliver IFNtherapy.

3. Patients currently treated with IFN who achieve haem-atological control but without a major cytogeneticresponse after 1 year on treatment should be offeredimatinib.

4. Patients currently treated with IFN who have a majorcytogenetic response (< 35% Ph-positive by G-bandingor fluorescence in situ hybridization) on IFN and aretolerating this treatment should not be offered imatinibunless or until an increase in Ph-positive metaphases isobserved.

5. Patients who enter accelerated phase or blast transfor-mation should be offered imatinib until furtherprogression.

BACKGROUND

1. CML is rare: the incidence is approximately 1:100 000(estimated 500–800 new cases per year in the UK). Theestimated prevalence is 3000–4000 patients.

2. The only known ‘curative’ therapy for CML is allo-geneic haemopoietic stem cell transplantation (Goldman& Druker, 2001). IFN has been shown to prolongsurvival compared with hydroxyurea or busulphan(The Italian Cooperative Study Group on ChronicMyeloid Leukaemia, 1994; Allan et al, 1995; ChronicMyeloid Leukaemia Trialists’ Collaborative Group,1997; Benelux CML Study Group, 1998). The additionof Ara-C to IFN may further improve survival (Guilhotet al, 1997; Baccarani et al, 2002).

3. Formerly known as CGP57148 or STI571, imatinibmesylate (Glivec�, Novartis, Basel) is a 2¢ phenylamino-pyrimidine derivative (C30H35N7SO4, molecular weight,589Æ7) that specifically inhibits the SH1 domaintyrosine kinase activity of the p160ABL and p190 ⁄p210BCR–ABL proteins with an IC50 of 0Æ1 lmol ⁄ l(Buchdunger et al, 1996; Druker et al, 1996). Imatinibis known to also inhibit the tyrosine kinase activityassociated with KIT (Heinrich et al, 2000), ARG (Okudaet al, 2001 and PDGF (Buchdunger et al, 2000).

4. Imatinib has been given to human subjects with Ph-positive leukaemias for almost 4 years. The first patientwas treated in June 1998 (Druker et al, 2001a). Firstpatient in UK: September 1999. There are now over1000 patients in the UK and 12 000 globally on thedrug.

5. In the treatment of chronic-phase CML, imatinibproduces much better haematological and cytogeneticresponses than have historically been achieved withIFN (Kantarjian et al, 2002) with most patientssustaining these responses. In newly diagnosed CML

*Produced on behalf of the Haematological Oncology Task Force of

the British Committee for Standards in Haematology.

British Journal of Haematology, 2002, 119, 268–288

268 � 2002 Blackwell Publishing Ltd

the major cytogenetic response on imatinib therapy is83% with 68% complete responses (compared with20% and 7%, respectively, with the IFN + Ara-C arm –see section 3) (Druker et al, 2002). Initial licensedindication in chronic phase is for those who have failedtherapy with IFN. An extension of the licence to newlydiagnosed patients will be sought later this year.

6. For certain younger patients with chronic-phase CML,allogeneic bone marrow transplant will remain thetreatment of choice but, increasingly, patients will betreated, perhaps for a trial period, with imatinib ratherthan undergoing transplantation (Goldman & Druker,2001).

7. In accelerated phase CML response rates are inferior tothose seen in chronic phase, but still many of theseresponses are sustained (Talpaz et al, 2002). Imatinib isjustified in such patients for as long as haematologicalresponse is maintained.

8. In blast crisis of CML (both myeloid and lymphoid) onlya minority of patients durably respond to imatinib andthe response is usually short lived – a matter of months(Ottmann et al, 2002; Sawyers et al, 2002). In suchpatients, use of imatinib may be justified in order toachieve transient control of disease with minimal side-effects mostly as an outpatient. If a transplant option isbeing explored, imatinib can ‘buy time’ while thepracticalities are being arranged.

9. Imatinib is given as a once daily oral dose and isgenerally well tolerated (chronic phase – 400 mg ⁄ d;more advanced disease – 600 mg ⁄ d). Patients takingthe drug usually have few side-effects and it is clearthat imatinib produces fewer side-effects than IFN.Most patients prefer imatinib to IFN and formal quality-of-life comparisons within the IRIS (InternationalRandomized Study of Interferon + Ara-c Vs. STI571)study (Novartis study 0106) are now available – seesection 3.

10. There are as yet no data to conclusively demonstratethat imatinib improves long-term survival for CMLpatients compared with an IFN-containing regimen.This question is currently being addressed by theNovartis 0106 phase III study (see section 3). It isproven that improvements in cytogenetic response

rates on IFN are associated with longer survival: thismay also be the case with imatinib and it seemsreasonable to expect that the drug will improvesurvival.

11. Imatinib was licenced by the Food and Drug Admin-istration (FDA); for use in the United States on 10 May2001. The European Medicines Evaluation Agency(EMEA) granted a licence covering the UK on 7November 2001. The licenced indications are: chro-nic-phase CML in patients who have failed IFN therapy(lack of cytogenetic response or intolerance); acceler-ated-phase CML; blast crisis of CML.

LATEST DATA IN NEWLY DIAGNOSEDCHRONIC-PHASE CML PATIENTS

Data on late chronic-phase patients have recently beenpublished (Kantarjian et al, 2002). This position paper istimed to coincide with the presentation of data at theAmerican Society for Clinical Oncology (ASCO – 20 May2002) on the ‘IRIS’ study (Druker et al, 2002). IRIS (alsoknown as Novartis study 0106) is a direct comparison ofimatinib against the current ‘gold standard’ – interferonplus Ara-C (Guilhot et al, 1997; Baccarani et al, 2002) – innewly diagnosed (less than 6 months) patients (Table I). Thisstudy is being conducted in 178 hospitals throughout theworld (13 in the UK) and closed to recruitment on 31January 2001.

One thousand one hundred and six patients wererecruited in 7 months (106 in the UK), making it one ofthe fastest recruiting leukaemia trials ever. The differencesin the two arms of the study were so compelling that, inDecember 2001, the Independent Data Monitoring Boardrecommended early disclosure of the results. The majorcytogenetic response on imatinib therapy is 83% with 68%complete responses compared with 20% and 7%, respect-ively, in the IFN + Ara-C arm (Druker et al, 2002). Rates ofprogression to accelerated phase or blast crisis were alsosignificantly different between the two arms (Table I). TheIRIS study has also compared quality of life in patientsreceiving imatinib or IFN and Ara-C. These data haverecently been analysed for the first time and will bepresented at the European Haematology Association

Table I. Response rates in imatinib studies to date.

Study 0110 (Kantarjian

et al, 2002) n ¼ 532

Late chronic phase

IRIS (imatinib arm) (Druker

et al, 2002) n ¼ 553 (of 1106)

Newly diagnosed chronic phase

IRIS

(IFN & Ara-C arm)

n ¼ 553 (of 1106)

Complete haematological response rate 95% 94%* 55%*

Complete cytogenetic response 41% 68%* 7%*Partial cytogenetic response 19% 15% 13%

Major cytogenetic response 60% 83%* 20%*

Progression to accelerated phase or blast crisis NA 1Æ5%* 6Æ9%*Intolerance of therapy requiring cross-over NA 1%* 23%*

Response and progression rates for the IRIS (0106) study are based on minimum of 12 months follow-up on all patients. All comparative

values indicated by an asterisk (*) are statistically significantly different. NA, not available.

Correspondence 269

� 2002 Blackwell Publishing Ltd, British Journal of Haematology 119: 268–288

Meeting in June 2002. Taken overall, the data form the IRISstudy offer compelling evidence that imatinib is superior toIFN and Ara-C in terms of cytogenetic response, progressionrates, tolerability and quality of life in patients with chronic-phase CML. Data on survival of patients in the study iscurrently being analysed and will be presented at theAmerican Society for Hematology in December 2002.

FUTURE STUDY FOR NEWLY DIAGNOSEDCHRONIC-PHASE PATIENTS: SPIRIT

SPIRIT (STI571 Prospective International RandomizedTrial) is a proposed multinational trial involvingco-operative groups in the USA, UK and France withpotential partnership with many other countries(http: ⁄ ⁄www.spirit-cml.org). The current schema (whichmay still be subject to change) is shown in Fig 1. It is acrucial vehicle for extending our understanding of the roleof imatinib in the therapy of CML. As well as assessingsurvival and other efficacy endpoints, SPIRIT has beenspecifically designed to stringently evaluate health econo-mic and quality of life considerations. The MRC hadinsufficient funds for this alpha A-rated project asannounced on 24 July 2002 and alternative funding isbeing sought. Recruitment could commence at the begin-ning of 2003 in the UK.

REGULATORY STATUS ⁄ COST ⁄ NICE (NATIONALINSTITUTE FOR CLINICAL EXCELLENCE) APPROVAL

Imatinib was licenced by the FDA on 10 May 2001 and bythe EMEA on 7 November 2001 (approval covers UK).

Imatinib costs approximately £18 000 per year at a doseof 400 mg ⁄ d. In their Final Appraisal Determinationpublished on 12 August 2002, NICE have made thefollowing summary recommendation:• Imatinib is recommended as a treatment option for the

management of Philadelphia-chromosome-positive chronicmyeloid leukaemia (CML) in chronic phase in adults whoare intolerant of interferon-alpha (IFN-a) therapy or inwhom IFN-a is deemed to have failed to control thedisease.

• IFN-a failure is defined as either:• failure to achieve complete haematological response after

3 months of IFN-a treatment as monotherapy or incombination with hydroxyurea, or

• failure to achieve major cytogenetic response after1 year of IFN-a treatment despite haematologicalresponse.

IFN-a intolerance is defined as the presence of documentedGrade 3 non-haematological toxicity, persisting for morethan 2 weeks, in a patient receiving a regimen thatcontains IFN-a.

• Imatinib is recommended as an option for the treatment ofadults with Philadelphia-chromosome-positive CML in accel-erated phase or blast crisis provided they have not receivedimatinib treatment at an earlier stage.The BSH are acutely aware that imatinib is expensive and

may place a significant financial burden on the Departmentof Health (DoH) and postcode prescribing must be avoided.Although not initially licenced for use in newly diagnosedCML patients, there will inevitably be prescribing in thiscontext. In view of this, the society would wish to explorethe following mechanism for the use of the drug in newly

Fig 1. The current schema for SPIRIT.

270 Correspondence


diagnosed patients. This would be in the context of theSPIRIT trial (see section 4).1. High quality national ⁄ international trials should be

conducted to further understand the role of imatiniband should be supported by government.

2. Clinicians should be encouraged to use imatinib fornewly diagnosed CML patients within an approvednational ⁄ international clinical trial programme (suchas SPIRIT) and there should be an expectation that datawill be contributed. It is crucial that as many patients aspossible are entered into properly formulated clinicalstudies or we learn nothing and have no monitoring ofcost effectiveness.

3. A single price negotiating ⁄ purchasing ⁄ distributionmechanism for imatinib should be explored for newlydiagnosed CML patients with central government fund-ing such that trial drugs could be purchased anddistributed from one location.

The advantages of this model are: (i) best possible pricefor drugs could be negotiated because of bulk purchase; (ii)the distribution of drugs could be carefully controlled ⁄administered to ensure the highest standards of compli-ance ⁄ data collection for clinical studies; (iii) such a systemwould enable precise monitoring of drug costs for theNational Health Service ⁄DoH which is impossible usingcurrent local mechanisms; (iv) ‘postcode prescribing’ wouldbe avoided and there would be no need for innumerablelocal negotiations ⁄ battles. Drugs would be shipped directlyto involved hospitals with equal availability across the UK.This proposal has been put to NICE in anticipation of itsguidance in August 2002.

DECLARATION OF INTERESTS

Drs O’Brien and Rule have both acted as consultants toNovartis and have received research funding from thecompany. Dr O’Brien is a member of the StudyManagement Committee (together with Professors Dru-ker, Guilhot and Larson) of the current Novartis 0106study. Neither author owns stock in Novartis. Dr O’Brienis the Principal applicant for MRC funding for the SPIRITproject.

Stephen G. O’Brien1

Simon A. J. Rule2

1Department of Haematology,University of Newcastle, and2Department of Haematology,Derriford Hospital, Plymouth, UK.E-mail: s.g.o’[email protected] [email protected]

ACKNOWLEDGMENTS

The following individuals have also reviewed and ⁄ orcontributed to this position paper:

Dr Donald W. Milligan (Chair, BCSH HaematologicalOncology Task Force), Dr Stephen Johnson (Secretary,BCSH Haematological Oncology Task Force), Dr RobertMarcus (Clinical Director, BCSH).

National Cancer Research Network CML Working Group*Professor John M. Goldman (Chairman)

Members of the NCRN WG: Dr Jane Apperley, Dr RichardClark, Dr Charles Craddock, Professor A. R. Green, Dr TessaHolyoake, Dr Anne Lennard, Dr Guy Lucas, Dr StephenMackinnon, Dr Eduardo Olavarria, Dr Pat Shepherd, DrGraeme Smith.

*Drs O’Brien and Rule are also members of the NCRNCML WG.

REFERENCES

Allan, N.C., Richards, S.M. & Shepherd, P.C. (1995) UK Medical

Research Council randomized, multicentre trial of interferon-

alpha n1 for chronic myeloid leukaemia: improved survivalirrespective of cytogenetic response. The UK Medical Research

Council’s Working Parties for Therapeutic Trials in Adult Leuk-

aemia. Lancet, 345, 1392–1397.

Baccarani, M., Rosti, G., de Vivo, A., Bonifazi, F., Russo, D.,Martinelli, G., Testoni, N., Amabile, M., Fiacchini, M.,

Montefusco, E., Saglio, G. & Tura, S. (2002) A randomized study

of interferon-alpha versus interferon-alpha and low-dose arabi-nosyl cytosine in chronic myeloid leukemia. Blood, 99,

1527–1535.

Buchdunger, E., Cioffi, C.L., Law, N., Stover, D., Ohno-Jones, S.,

Druker, B.J. & Lydon, N.B. (2000) Abl protein-tyrosine kinaseinhibitor STI571 inhibits in vitro signal transduction mediated

by C-kit and platelet-derived growth factor receptors. Journal of

Pharmacological Experimental Therapy, 295, 139–145.

Buchdunger, E., Zimmerman, J., Mett, H., Meyer, T., Muller, M.,Druker, B.J. & Lydon, N.B. (1996) Inhibition of the Abl protein-

tyrosine kinase in vitro and in vivo by a 2-phenylaminopyr-

imidine derivative. Cancer Research, 56, 100–104.Chronic Myeloid Leukemia Trialists’ Collaborative Group (1997)

Interferon alfa versus chemotherapy for chronic myeloid leuk-

aemia: a meta-analysis of seven randomized trials. Journal of the

National Cancer Institute, 89, 1616–1620.Druker, B.J., Tamura, S., Buchdunger, E., Ohno, S., Segal, G.M.,

Fanning, S., Zimmermann, J. & Lydon, N.B. (1996) Effects of a

selective inhibitor of the Abl tyrosine kinase on the growth of

Bcr-Abl positive cells. Nature Medicine, 2, 561–566.Druker, B.J., Talpaz, M., Resta, D.J., Peng, B., Buchdunger, E., Ford,

J.M., Lydon, N.B., Kantarjian, H., Capdeville, R., Ohno-Jones, S. &

Sawyers, C.L. (2001a) Efficacy and safety of a specific inhibitor of

the BCR-ABL tyrosine kinase in chronic myeloid leukemia. NewEngland Journal of Medicine, 344, 1031–1037.

Druker, B.J., Sawyers, C.L., Kantarjian, H., Resta, D.J., Fernandes

Reese, S., Ford, J.M., Capdeville, R. & Talpaz, M. (2001b) Activityof a specific inhibitor of the BCR-ABL tyrosine kinase in the blast

crisis of chronic myeloid leukemia and acute lymphoblastic leu-

kemia with the Philadelphia chromosome. New England Journal

of Medicine, 344, 1038–1042.Druker, B.J. & The IRIS Study – International Randomized Study of

Interferon + Ara-C Vs. STI571 (Imatinib ⁄Glivec ⁄Gleevec) in

Chronic Myeloid Leukemia (2002) (ASCO abstract 1) (WWW

document) URL http: ⁄ ⁄www.asco.org/asco/ascoMainConstruc-tor/1,47468,_12|002324|00–29 |00A| 00–18|002002|

00–19|001,00.asp.

Goldman, J.M. & Druker, B.J. (2001) Chronic myeloid leukemia:current treatment options. Blood, 98, 2039–2042.

Guilhot, F., Chastang, C., Michallet, M., Guerci, A., Harousseau,

J.L., Maloisel, F., Bouabdallah, R., Guyotat, D., Cheron, N.,

Correspondence 271


Nicolini, F., Abgrall, J.F. & Tanzer, J. (1997) Interferon alfa-2b

combined with cytarabine versus interferon alone in chronic

myelogenous leukemia. French Chronic Myeloid Leukemia StudyGroup. New England Journal of Medicine, 337, 223–229.

Heinrich, M.C., Griffith, D.J., Druker, B.J., Wait, C.L., Ott, K.A. &

Zigler, A.J. (2000) Inhibition of C-kit receptor tyrosine kinase

activity by STI 571, a selective tyrosine kinase inhibitor. Blood,96, 925–932.

Kantarjian, H., Sawyers, C.L., Hochhaus, A., Guilhot, F., Schiffer, C.,

Gambacorti-Passerini, C., Niederwieser, D., Stone, R., Goldman,

J.M., Fischer, T., Cony-Makhoul, P., O’Brien, S.G., Miller, C., Tall-man, M., Brown, R., Schuster, M., Gratwohl, A., Loughran, T.,

Mandelli, F., Saglio, G., Ottmann, O.G., Lazzarino, M., Russo, D.,

Tura, S., Facon, T., Morra, E., Russell, N., Zoellner, U., Resta, R.,

Capdeville, R., Talpaz, M. & Druker, B.J. (2002) Hematologic andcytogenetic responses to imatinib mesylate in chronic myelogen-

ous leukemia. New England Journal of Medicine, 346, 645–652.

Okuda, K., Weisberg, E., Gilliland, D.G. & Griffin, J.D. (2001) ARGtyrosine kinase activity is inhibited by STI571. Blood, 97, 2440–

2448.

Ottmann, O.G., Druker, B.J., Sawyers, C.L., Goldman, J.M., Mahon,

F.X., Silver, R.T., Tura, S., Fischer, T., Deininger, M., Schiffer,C.A., Baccarani, M., Gratwohl, A., Hochhaus, A., Hoelzer, D.,

Fernandes-Reese, S., Gathmann, I., Capdeville, R. & O’Brien, S.G.

(2002) Glivec (imatinib mesylate) induces hematologic and

cytogenetic responses in patients with relapsed or refractoryPhiladelphia chromosome-positive acute leukemias: results of a

phase II study. Blood (in press).

Sawyers, C.L., Hochaus, A., Feldman, E., Goldman, J.M., Miller,C.M., Ottman, O.G., Schiffer, C.A., Talpaz, M., Guilhot, F.,

Deininger, M.W.N., Fischer, T., O’Brien, S.G., Stone, R., Gam-

bacorti-Passerini, C., Russel, N., Reiffers, J., Shea, T., Chapuis,

B., Coutre, S., Tura, S., Morra, E., Larson, R.A., Saven, A.,Peschel, C., Gratwohl, A., Mandelli, F., Ben-Am, M., Ganth-

mann, I., Capdeville, R., Paquette, R.L. & Druker, B.J. (2002)

Glivec (imatinib mesylate) induces hematologic and cytogenetic

responses in patients with chronic myeloid leukemia in mye-loid blast crisis. Results of a Phase II Study. Blood, 99, 3530–

3539.

Talpaz, M., Silver, R.T., Druker, B.J., Goldman, J.M., Gambacorti-

Passerini, C., Guilhot, F., Schiffer, C.A., Fischer, T., Deininger,M.W., Lennard, A.L., Hochhaus, A., Ottmann, O.G., Gratwohl,

A., Baccarani, M., Stone, R., Tura, S., Mahon, F.X., Fernandes-

Reese, S., Gathmann, I., Capdeville, R., Kantarjian, H.M. &

Sawyers, C.L. (2002) Imatinib induces durable hematologic andcytogenetic responses in patients with accelerated phase chronic

myeloid leukemia: results of a phase 2 Study. Blood, 99, 1928–

1937.The Benelux CML Study Group (1998) Randomized study on

hydroxyurea alone versus hydroxyurea combined with low-dose

interferon-alpha 2b for chronic myeloid leukemia. Blood, 91,

2713–2721.The Italian Cooperative Study Group on Chronic Myeloid

Leukemia (1994) Interferon alfa-2a as compared with con-

ventional chemotherapy for the treatment of chronic

myeloid leukemia. New England Journal of Medicine, 330, 820–825.

Keywords: CML, STI-571, imatinib mesylate.

BCR-ABL FISH MONITORING OF CML: A SURVEY OF CURRENT UK PRACTICE

Fluorescence in situ hybridization (FISH), performed onperipheral blood (PB) or bone marrow (BM), has beenadvocated as an effective technique for the diagnosis andmonitoring of chronic myeloid leukaemia (CML) (Buno et al,1998). Relative to conventional cytogenetic analysis (CCA),the advantages of FISH include increased sensitivity, reducedcost and the ability to detect masked BCR-ABL transloca-tions. In particular, analysis of interphase (IP) cells avoidsthe need for metaphase (MP) BM cells and the technique cantherefore be applied to peripheral blood (Buno et al, 1998).Furthermore, dual fusion (DF) probes can detect deletions ofthe derivative chromosome 9, recently shown to conferconsiderable prognostic significance independent of standardrisk-scoring systems (Sinclair et al, 2000; Huntly et al,2001). Anecdotally, however, our impression has been thatFISH practice varies considerably throughout the UK. Withthe likelihood that patient numbers on non-transplanttreatment regimens will increase (i.e. incorporating imatinibmesylate alone or with additional agents) and that risk-adapted treatment strategies will develop (Goldman &Druker, 2001), it is important to address issues surroundingthe routine use of FISH and standardization of procedurebetween laboratories. For these reasons we carried out asurvey of current FISH practice in the UK.

Of the 27 genetics laboratories that we identified asoffering a routine haematological malignancy service in theUK (according to the British Society for Human GeneticsMembership Directory, April 2000), 25 (93%) replied to apostal or electronic questionnaire sent out during August2001. Twenty-three responders (92%) performed BCR-ABLFISH and this number forms the denominator for all furtherdata. Responses are summarized in Table I.

These results demonstrated considerable variationthroughout the UK in the FISH scoring procedure andin its application to the diagnosis and monitoring of CMLpatients. Obvious applications, such as metaphase failuresand cryptic translocation monitoring, were practicedalmost universally and, in general, it was viewed as acomplementary technique to CCA, rather than as areplacement. However, there were some surprising results.Only 13% of laboratories routinely performed FISH atdiagnosis and, although it may be performed later andchromosome 9 deletions identified, this information maybe crucial to treatment planning from the outset (e.g.early allogeneic BM transplantation or experimentalprotocols). Such objective assessment of individual risk isof particular value, given (at least in Scottish centresrecently surveyed) the limited use of the Sokal or New

272 Correspondence


Score risk systems outside centres with an interest in CML(Harrison et al, 2002). With the widespread use of DFprobes, routine reporting of these deletions is possible.There was a clear variation in the scoring of neutrophils,presumably because of difficulty in interpreting theirnuclear morphology. Excluding neutrophils from theanalysis (48% of laboratories) may skew results in favourof lymphoid (hence BCR-ABL negative) cell types, partic-ularly once normal PB leucocyte counts have beenachieved. In two comparative studies (Buno et al, 1998;Muhlmann et al, 1998) PB FISH gave lower percentagepositive values than BM CCA, partly, perhaps, due tooverrepresentation of lymphoid cells in the FISH analysis.Indeed one group demonstrated successful correction ofPB FISH results using lymphocyte counts (Muhlmannet al, 1998), suggesting that the accuracy of FISH inestablishing the size of the leukaemic clone can beimproved by using PB FISH in conjunction with thewhite blood cell differential count.

The recent dramatic results with imatinib willundoubtedly alter the approach to disease responsemonitoring. Patients achieving complete cytogeneticresponses may require routine use of sensitive PBquantitative reverse transcription polymerase chain reac-tion in place of routine CCA or FISH, a strategycurrently under investigation. We would therefore sup-port assessment of the various monitoring proceduresavailable within large-scale clinical trials, with standard-ization of techniques, consensus opinion and subsequentguideline publication.

Mark W. Drummond1

Lynn M. Hendry2

Edward Maher2

Tessa L. Holyoake1

1Academic Transfusion MedicineUnit, Glasgow University, and2Cytogenetics Department,Western General Hospital,Edinburgh, UK. E-mail:[email protected]

ACKNOWLEDGMENT

M. W. Drummond is supported by the Leukaemia ResearchFund UK (grant number 9965) and Glasgow University.

REFERENCES

Buno, I., Wyatt, W.A., Zinsmeister, A.R., Dietz-Band, J., Silver, R.T.

& Dewald, G.W. (1998) A special fluorescent in situ hybridizationtechnique to study peripheral blood and assess the effectiveness

of interferon therapy in chronic myeloid leukemia. Blood, 92,

2315–2321.

Goldman, J.M. & Druker, B.J. (2001) Chronic myeloid leukemia:current treatment options. Blood, 98, 2039–2042.

Harrison, S.J., Johnson, P.R.E. & Holyoake, T.L. (2002) Scottish

Leukaemia Registry CML audit. British Journal of Haematology,

117, Abstract 48.Huntly, B.J., Reid, A.G., Bench, A.J., Campbell, L.J., Telford, N.,

Shepherd, P., Szer, J., Prince, H.M., Turner, P., Grace, C.,

Nacheva, E.P. & Green, A.R. (2001) Deletions of the derivativechromosome 9 occur at the time of the Philadelphia

translocation and provide a powerful and independent prog-

nostic indicator in chronic myeloid leukemia. Blood, 98,

1732–1738.Muhlmann, J., Thaler, J., Hilbe, W., Bechter, O., Erdel, M., Uter-

mann, G. & Duba, H.C. (1998) Fluorescence In Situ Hybridistion

(FISH) on peripheral blood smears for monitoring Philadelphia

chromosome-positive chronic myeloid leukaemia (CML): a newstrategy for remission assessment. Genes Chromosomes and Cancer,

21, 90–100.

Sinclair, P.B.N.E., Leversha, M., Telford, N., Chang, J., Reid, A.,

Bench, A., Champion, K., Huntly, B. & Green, A.R. (2000) Largedeletions at the t (9; 22) breakpoint are common and may

identify a poor-prognosis subgroup of patients with chronic

myeloid leukaemia. Blood, 95, 738–744.

Keywords: chronic myeloid leukaemia, cytogenetics, bcr-abl, FISH.

Table I. Summary of responses to a questionnaire on FISH practice

in CML.

Criteria

Percentage

of laboratories

Application of FISH

All diagnostic BM specimens 13%

All BM samples (monitoring) 9%

Complementary to CCA 96%Alternative for MP failure 96%

Monitoring of cryptic translocations 100%

Types of probe usedCommercial single fusion 4%

Commercial dual fusion 52%

Commercial extra signal 17%

Extra signal plus dual fusion 26%Scoring procedure

Number of nuclei scored

50 4%

100 56%100–200 13%

200 13%

other 13%

Neutrophils scored 52%Controls

Negative control scored simultaneously 9%

Checking procedure (independent operator) 78%False-positive cut-off

unnecessary 26%

0Æ5% 4%

1% 22%2% 4%

5% 26%

10% 17%

Correspondence 273


STI571-INDUCED STEVENS–JOHNSON SYNDROME

STI571 is a specific inhibitor of the BCR-ABL tyrosinekinase. The efficacy of this drug has been demonstrated inpatients with Philadelphia-positive (Ph+) chronic myeloge-nous leukaemia (CML) resistant or intolerant to alpha-interferon (IFN) and in accelerated and blastic phases ofCML (Druker et al, 2001a,b). We present, to our knowledge,the first reported case of STI571-induced SJS.

A 58-year-old man was diagnosed with Ph+ CML in1987. The patient was treated with busulphan and laterswitched to IFN and hydroxyurea. In March 2001, thepatient developed asthenia, mucocutaneous pallor andhepatosplenomegaly. Abnormal blood values were leuco-cytes 3Æ38 · 109 ⁄ l (8% blast cells), haemoglobin 10Æ6 g ⁄ dl,erythrocyte sedimentation rate 100 mm ⁄h, alkaline phos-phatase 305 IU ⁄ l and lactate dehydrogenase 490 IU ⁄ l.Bone marrow smear and biopsy demonstrated an infiltra-tion of 70% blast cells expressing myeloid antigens. Thepatient was diagnosed with CML in myeloid blast crisis,STI571 600 mg ⁄ d was started and a complete haemato-logical remission was achieved 2 weeks later. On d +28 thepatient developed an itching symmetrical macular eruptionon both arms. Despite topical corticosteroids and oralantihistamines, on d +34 the eruption evolved to over70% of the total body surface (Fig 1). Blisters and ulcer-ations appeared, Nikolsky’s sign was present and 10% of the

total skin surface was denuded. The patient’s generalcondition deteriorated rapidly, but he had no fever, lymphnode enlargement or mucosal involvement. The skin biopsyshowed pronounced dermal papillary oedema with subepi-dermal vesiculation, mild inflammatory infiltrate and a focalinfiltrate of neutrophils and blast cells inside a follicularinfundibulum. These features were consistent with toxicdrug reaction. No apoptotic keratinocyte could be demon-strated by using the terminal deoxynucleotidyl transferase-mediated dUTP-biotin end-labelling (TUNEL) technique.Results of direct immunofluorescence studies on involvedskin were negative for all antiserums. The patient washospitalized with the diagnosis of STI571-induced Stevens–Johnson syndrome (SJS); systemic corticosteroids werebegun and STI571 was discontinued. Seven days afteradmission, all the ulcerated lesions were completelyre-epithelized, the patient was discharged 3 d later andcorticosteroids therapy was tapered gradually. A bonemarrow smear performed in June 2001 showed a reversionto second chronic phase CML and the patient is well after6 months follow-up.

STI571 is administered orally in an outpatient settingand it has a fairly good tolerance profile. Severe cutaneouseruptions due to STI571 are not frequent. From 141patients treated with STI571 (Druker et al, 2001a,b), 26(18%) developed cutaneous eruptions, but it was severeonly in three (2%) patients. These three patients werereceiving STI571 doses between 600 and 1000 mg ⁄ d. Arecent case reported an acute generalized exanthematicpustulosis in a CML patient 1 month after startingSTI571 600 mg ⁄ d (Brouard et al, 2001). Brouard & Saurat(2001) pointed out that all five patients reported in theliterature presenting serious cutaneous reactions hadfollowed high STI571 doses. These authors suggest thatcutaneous adverse effects due to STI571 appear to be dosedependent and related to a pharmacological effect of thedrug. In contrast, drug-induced SJS is usually an idiosyn-cratic adverse reaction of hypersensitivity and early with-drawal of causative drug is imperative (Garcia-Doval et al,2000). In our case, SJS occurred after 4 weeks of therapywith 600 mg ⁄ d of STI571. The negative results of the directimmunofluorescence and TUNEL studies suggest that thepathogenesis of STI571-induced SJS is not related toimmune deposition or apoptosis. Although the developmentof the case presented here was with high doses of STI571,the pattern of reaction and the scarcity of similar publica-tions may suggest a hypersensitivity mechanism. Weconclude that due to the severity of STI571-induced SJS, ahigh suspicion index should enable early diagnosis andprompt withdrawal of STI571 treatment.

D. Vidal

L. Puig

A. Sureda

A. Alomar

Departments of Dermatology andClinical Haematology,Hospital de la Santa Creu i SantPau, Universitat Autonoma deBarcelona, Barcelona, Spain.E-mail: [email protected]

Fig 1. Day +34 of STI571 therapy. Top panel: generalized erythema.Bottom panel: Nikolsky’s sign on patient’s arm.

274 Correspondence


REFERENCES

Brouard, M. & Saurat, J.H. (2001) Cutaneous reactions to STI571.

New England Journal of Medicine, 345, 618–619.

Brouard, M.C., Prins, C., Mach-Pascual, S. & Saurat, J.H. (2001)

Acute generalized exanthematous pustulosis associated withsti571 in a patient with chronic myeloid leukemia. Dermatology,

203, 57–59.

Druker, B.J., Talpaz, M., Resta, D.J., Peng, B., Buchdunger, E.,

Ford, J.M., Lydon, N.B., Kantarjian, H., Capdeville, R., Ohno-Jones, S. & Sawyers, C.L. (2001a) Efficacy and safety of a

specific inhibitor of the BCR-ABL tyrosine kinase in chronic

myeloid leukemia. New England Journal of Medicine, 344, 1031–1037.

Druker, B.J., Sawyers, C.L., Kantarjian, H., Resta, D.J., Reese, S.F.,

Ford, J.M., Capdeville, R. & Talpaz, M. (2001b) Activity of a

specific inhibitor of the BCR-ABL tyrosine kinase in the blast

crisis of chronic myeloid leukemia and acute lymphoblastic leu-kemia with the Philadelphia chromosome. New England Journal of

Medicine, 344, 1038–1042.

Garcia-Doval, I., LeCleach, L., Bocquet, H., Otero, X.L. & Roujeau,J.C. (2000) Toxic epidermal necrolysis and Stevens–Johnson

syndrome: does early withdrawal of causative drugs decrease the

risk of death? Archives of Dermatology, 136, 323–327.

Keywords: STI571, cutaneous adverse reaction, Stevens–Johnson syndrome.

THALIDOMIDE TREATMENT OF RELAPSED MULTIPLE MYELOMA PATIENTS AND CHANGES

IN CIRCULATING VEGF AND bFGF

We read with interest the report by Neben et al (2001a) inwhich they reported the serial measurements of plasma levelsof vascular endothelial growth factor (VEGF), basic-fibroblastgrowth factor (bFGF) and hepatocyte growth factor (HGF), ina group of 51 patients with progressive multiple myeloma(MM) treated with thalidomide (Thal). After 6 months oftreatment, only HGF in the group of responsive patientsshowed a statistical change. The authors concluded that themechanism of action of Thal in MM was not caused by aspecific inhibition of angiogenic cytokine secretion.

These results and conclusions must be interpreted withcaution. In MM, several lines of evidence suggested a role forangiogenic factors, such as VEGF, in the regulation oftumour cell growth and disease activity. Firstly, purifiedmyeloma cell fractions of MM marrow were found to secretesignificantly more VEGF than the corresponding non-tumour fractions. Sezer et al (2001) have recently reportedthat VEGF and bFGF were found to decrease in patientsresponding to chemotherapy but not in patients who did notachieve a remission. Secondly, myeloma cells express bothangiogenic activity and angiogenic cytokines.

Recently, Bertolini et al (2001) and Kakimoto et al(2001) have performed serial assessments of plasma VEGFand bFGF levels in myeloma patients treated with Thal. Atthe time of the best clinical response, both VEGF and b-FGFwere significantly decreased.

In our experience (18 MM patients), at the time of thebest response, both VEGF and bFGF were significantlydecreased when compared with pretreatment values. Infact, Neben et al (2001b) have recently reported that whensummary measures were used to correlate the changes ofcytokine levels over time with response, decreasing values ofbFGF and VEGF were found to be associated with response.As these data are from various sources with differingmethods of VEGF detection, different lengths of follow-upand diverse treatments administered, a true analysis is notfeasible. Nevertheless, all these findings support the hypo-thesis that the action of Thal may be due, at least in part, to

the inhibition of cytokine signalling among stromal, endo-thelial and neoplastic cells.

V. Pitini1

D. Teti2

C. Arrigo1

G. Aloi1

M. Righi2

1Department of Oncology and2Pathology, University of Messina,Messina, Italy. E-mail:[email protected]

REFERENCES

Bertolini, F., Mingrone, W., Alietti, A., Ferrucci, P.F., Cocorocchio,

E., Peccatori, F., Cineri, S., Mancuso, P., Corsini, C., Burlini, A.,

Zucca, E. & Martinelli, G. (2001) Thalidomide in multiple mye-

loma, myelodysplastic syndromes and histiocytosis. Analysis ofclinical results and surrogate angiogenesis markers. Annals of

Oncology, 12, 987–990.

Kakimoto, T., Hattori, Y., Okamoto, S., Sato, N., Yamaguchi, M.,

Morita, K., Yamada, T., Takayama, N., Shimada, N., Tanigawara,Y. & Ikeda, Y. (2001) Clinical significance of serial monitoring of

plasma concentration of thalidomide and angiogenic growth

factors in patients with refractory myeloma. Proceeding ofAmerican Society of Hematology (abstract 682). Blood, 98, 682.

Neben, K., Moehler, T., Kraemer, A., Benner, A., Egerer, G., Ho, A.D.

& Goldschmidt, H. (2001a) Response to thalidomide in progressive

multiple myeloma is not mediated by inhibition of angiogeniccytokine secretion. British Journal of Haematology, 115, 605–608.

Neben, K., Moehler, T., Egerer, G., Kraemer, A., Hillengass, J.,

Benner, A., Ho, A.D. & Goldschmidt, H. (2001b) High plasma

basic Fibroblast Growth Factor concentration is associated withresponse to thalidomide in progressive multiple myeloma. Clinical

Cancer Research, 7, 2675–2681.

Sezer, O., Jacob, C., Euker, J., Wernecke, K.D. & Possinger, K. (2001)

Serum levels of the angiogenic cytokines basic Fibroblast GrowthFactor (bFGF), Vascular Endothelial Growth Factor (VFGF) and

Hepatocyte Growth Factor (HGF) in multiple myeloma. European

Journal of Haematology, 66, 83–88.

Keywords: thalidomide, vascular endothelial growth factor,basic fibroblast growth factor.

Correspondence 275


SERUM LEVELS OF VEGF DO NOT CORRELATE WITH THE SEVERITY OF MULTIPLE MYELOMA

We read the study by Iwasaki et al (2002) with interest.Although the authors concluded that their results dem-onstrated that serum levels of vascular endothelial growthfactor (VEGF) correlated with the severity of multiplemyeloma, this conclusion was not supported by the dataof their study. In contrast, their results clearly show thatserum VEGF levels do not correlate with disease severityin multiple myeloma. First, no significant differences inserum VEGF levels were found between the stage IIIgroup and the stage I or II groups, as has been published(Sezer et al, 2001a). Second, to study the correlationbetween serum VEGF and markers of disease severity, theauthors divided the study population into dichotomizedgroups according to the presence or absence of anaemia,hypercalcaemia, renal dysfunction and bone lesions.Serum VEGF levels were not statistically different in anyof these dichotomized groups according to levels ofhaemoglobin, serum calcium, serum creatinine, or thepresence or absence of bone lesions. Third, serum VEGFlevels in myeloma patients with amyloidosis were signi-ficantly lower than those in patients without amyloidosis.Therefore, we conclude that this study did not show thathigh serum VEGF levels correlate with increased diseaseseverity in multiple myeloma.

Multiple myeloma is the only haematological malig-nancy in which the prognostic relevance of bone marrowangiogenesis has been shown in multivariate analysis(Rajkumar et al, 2000; Sezer et al, 2000). VEGF is knownto be released by several types of blood cells and inmultiple myeloma we could not find a correlation betweenbone marrow angiogenesis and serum VEGF levels.Although a significant correlation between bone marrowangiogenesis and some other disease characteristics, b2-microglobulin levels (r ¼ 0Æ749, P < 0Æ0005) for example,has been shown (Sezer et al, 2001b), Iwasaki et al (2002)found no correlation between serum VEGF and b2-micro-globulin levels or any other myeloma-related parameterinvestigated. For these reasons serum VEGF levels shouldbe interpreted with caution and should not be regarded asa valid surrogate marker of angiogenesis in multiplemyeloma. This issue may be important in the context offuture antiangiogenic treatment strategies, which shouldbe aimed against tumour angiogenesis rather than circu-lating VEGF levels.

The study has further shortcomings. Although 49 of52 patients were in stage II or III, only 27 patients werereported to have bone disease, i.e. 55% of patients instage II ⁄ III. This proportion appears to be much lower

than reported in other studies (Fonseca et al, 2000).Although all 52 patients were treated, the type of thetreatment given remains unclear. Last, but not least, wethink it is not acceptable that the authors included only39 of the 52 patients in the survival analysis. Unfortu-nately, the reason for this was not given in the sectionPatients and methods.

Christian Jakob

Ivana Zavrski

Ulrike Heider

Corinna Langelotz

Jan Eucker

Kurt Possinger

Orhan Sezer

Department of Haematology andOncology, UniversitatsklinikumCharite, Humboldt-Universitatzu Berlin, Berlin, Germany.E-mail: [email protected]

REFERENCES

Fonseca, R., Trendle, M.C., Leong, T., Kyle, R.A., Oken, M.M.,

Kay, N.E., Van Ness, B. & Greipp, P.R. (2000) Prognostic

value of serum markers of bone metabolism in untreated

multiple myeloma patients. British Journal of Haematology,109, 24–29.

Iwasaki, T., Hamano, T., Ogata, A., Hashimoto, N., Kitano, M. &

Kakishita, E. (2002) Clinical significance of vascular endothelial

growth factor and hepatocyte growth factor in multiple myeloma.British Journal of Haematology, 116, 796–802.

Rajkumar, S.V., Leong, T., Roche, P.C., Fonseca, R., Dispenzieri, A.,

Lacy, M.Q., Lust, J.A., Witzig, T.E., Kyle, R.A., Gertz, M.A. &Greipp, P.R. (2000) Prognostic value of bone marrow angio-

genesis in multiple myeloma. Clinical Cancer Research, 6, 3111–

3116.

Sezer, O., Niemoller, K., Eucker, J., Jakob, C., Kaufmann, O., Zavrski,I., Dietel, M. & Possinger, K. (2000) Bone marrow microvessel

density is a prognostic factor for survival in patients with mul-

tiple myeloma. Annals of Hematology, 79, 574–577.

Sezer, O., Jakob, C., Eucker, J., Niemoller, K., Gatz, F., Wernecke,K.D. & Possinger, K. (2001a) Serum levels of the angiogenic

cytokines basic fibroblast growth factor (bFGF), vascular

endothelial growth factor (VEGF) and hepatocyte growth factor(HGF) in multiple myeloma. European Journal of Haematology, 66,

83–88.

Sezer, O., Niemoller, K., Jakob, C., Zavrski, I., Heider, U., Eucker, J.,

Kaufmann, O. & Possinger, K. (2001b) Relationship betweenbone marrow angiogenesis and plasma cell infiltration and serum

b2-microglobulin levels in patients with multiple myeloma.

Annals of Hematology, 80, 598–601.

Keywords: angiogenesis, multiple myeloma, vascular endo-thelial growth factor (VEGF).

REPLY TO SEZER ET AL

We thank Dr Sezer and colleagues for their comments onour report (Iwasaki et al, 2002). Their comments documen-ted there was no correlation of serum vascular endothelial

growth factor (VEGF) levels with the severity of multiplemyeloma (MM). We drew the following conclusions fromour study: (1) Effective chemotherapy was accompanied by

276 Correspondence


significant decreases in serum hepatocyte growth factor(HGF) and VEGF, while no decrease in these factor levelswas found in non-responders. (2) Serum HGF levelscorrelated with disease severity of MM, as indicated by thestage of disease and by b2-microglobulin levels. (3) SerumVEGF levels correlated with responses to chemotherapy andwith accompanying aggressive types of extramedullaryplasmacytoma.

We are in agreement with Dr Sezer and colleagues (Sezeret al, 2001) that serum VEGF levels do not correlate withdisease severity of MM, as indicated by the stage of diseaseand by b2-microglobulin levels. However, we would like toreiterate that our paper documented a correlation of serumVEGF levels with poor response to chemotherapy andextramedullary expansion of tumour cells. Increased angio-genesis has been found to correlate with metastatic diseasein a variety of solid tumours, such as in patients withinvasive breast cancer (Acenero et al, 1998). As VEGF haslong been known to play a major role in growing andmetastasizing tumours (Martiny-Baron & Marme, 1995),this factor may contribute to extramedullary expansion oftumour cells and a poor response to chemotherapy. Themechanisms behind the increased angiogenesis in MM arenot fully understood. There is strong evidence for VEGFexpression by myeloma cells and a mutual stimulation ofmyeloma and marrow stromal cells triggered by VEGF andinterleukin 6 (IL-6) (Dankbar et al, 2000). A paracrineco-operation between VEGF and IL-6 may contribute totumour growth and protection of myeloma cells fromtreatment-induced apoptosis. We also agree that serumVEGF should not be regarded as a valid marker ofbone marrow angiogenesis in MM. However, directly orindirectly, VEGF is still likely to be involved in theneovascularization of bone marrow infiltrated by myelomacells. Our data indicate that two angiogenic factors (HGFand VEGF) may be important markers of different aspects of

angiogenesis in MM. However, their use as indicators ofprogression in MM awaits larger studies and longer periodsof follow-up analysis.

Tsuyoshi IwasakiDivision of Rheumatology andClinical Immunology, Departmentof Internal Medicine, Hyogo Collegeof medicine, Nishinomiya, Hyogo,Japan. E-mail: [email protected]

REFERENCES

Acenero, M.J., Gonzales, J.F., Gallego, G.M. & Ballesteros, P.A.(1998) Vascular enumeration as a significant prognosticator for

invasive breast carcinoma. Journal of Clinical Oncology, 16, 1684–

1688.

Dankbar, B., Padro, T., Leo, R., Feldmann, B., Kropff, M., Mesters,R.M., Serve, H., Berdel, W.E. & Kienast, J. (2000) Vascular

endothelial growth factor and interleukin-6 in paracrine tumor–

stromal cell interactions in multiple myeloma. Blood, 95, 2630–2636.

Iwasaki, T., Hamano, T., Ogata, A., Hashimoto, N., Kitano, M. &

Kakishita, E. (2002) Clinical significance of vascular endothelial

growth factor and hepatocyte growth factor in multiple myelo-ma. British Journal of Haematology, 116, 796–802.

Martiny-Baron, G. & Marme, D. (1995) VEGF-mediated tumor

angiogenesis: a new target for cancer therapy. Current Opinion of

Biotechnology, 6, 675–680.Sezer, O., Jakob, C., Eucker, J., Niemoller, K., Gatz, F., Wernecke, K.

& Possinger, K. (2001) Serum levels of the angiogenic cyto-

kines basic fibroblast growth factor (bFGF), vascular endo-

thelial growth factor (VEGF) and hepatocyte growth factor(HGF) in multiple myeloma. European Journal of Haematology,

66, 83–88.

Keywords: HGF, VEGF, multiple myeloma angiogenesis.

P15I N K 4 B IS NOT MUTATED IN ADULT FAMILIAL MYELODYSPLASTIC SYNDROMES

The myelodysplastic syndromes (MDS) comprise a groupof clonal haemopoietic stem cell disorders characterized byineffective haematopoeisis with an increased propensity toacute myeloid leukaemic (AML) transformation (Mufti,1992). Although some insight into the genetics of MDShas been gained from the analysis of children withneurofibromatosis type I, and ‘familial monsomy 7’,little is known about adult familial MDS. We haveidentified 13 kindreds in which two or more first-degreerelatives developed MDS in adult life (Mijovic & Mufti,1998). These families provide valuable research materialfor the identification of candidate genes involved infamilial MDS.

In cancers and haematological malignancies, alteredregulation of the cyclin D ⁄ pRB pathway is often observed.However, in haematological malignancies, pRB inactivation

is infrequent, suggesting the loss of other genes involved incell cycle regulation. The P16INK4A and P15INK4B genes,located at chromosome 9p21, are lost through deletion inmany malignancies, including the leukaemias (Kamb et al,1994). It is now well established that promoter regionhypermethylation, resulting in transcriptional silencing ofP15INK4B and P16INK4A, is an alternative mechanism forinactivation of these tumour suppressor genes (Hermanet al, 1997). This is a frequent mode for inactivation of p16in many tumour types, and, recently, the promoter ofP15INK4B has been shown to be hypermethylated in up to70% of advanced MDS cases, suggesting that it may be animportant gene in the multistep pathogenesis of MDS(Nakamaki et al, 1997).

Genes which are hypermethylated in sporadic tumourscan, in the familial setting, have germline mutations, which

Correspondence 277


are the basis for the inherited forms of these cancers. Suchexamples include the p16 gene in familial melanomasyndromes, the VHL gene in familial renal cancer and theMLH1 gene in inherited colon cancer (Hussussian et al,1994). In this study, we have sought to determine whetherthis may be true for the p15 gene in familial MDS, i.e.whether germline mutations of this gene are present inaffected members of MDS families (Mijovic & Mufti, 1998).To study the P15INK4B gene, bone marrow mononuclearcells were obtained from eight affected and an unaffectedmember of five MDS families (Fig 1), and DNA wasextracted by standard procedures. Polymerase chain reac-tion (PCR) primers were designed to amplify the 2 exonsof P15INK4B, and these are as follows: Exon 1 Forwardprimer: 5¢-CGATCACAGCGGACAGGGG-3¢, Reverse primer:5¢-ATCGGCGATCTAGGTTCCAGC-3¢, giving a fragmentlength of 669 bp. Exon 2 Primers: Forward: 5¢-GAAGG-TGGGGTGGGAAAGTGG-3¢, Reverse: 5¢-CTTCATGGTGAGT-GTCGAGGGC-3¢, giving a fragment length of 668 bp. PCRproducts were column purified and subcloned into the pCR�

2Æ1 vector using the TA Cloning kit (Invitrogen, San Diego,CA, USA), as directed. PCR-amplified DNA from five

individual colonies was subjected to automated sequenceanalysis of both DNA strands, using an Applied Biosystem377 automated sequencer, and analysed for mutations,using the MACVECTOR program.

Analysis of both p15 exons 1 and 2 from multiple affectedindividuals of five MDS families demonstrated no apparentdisruptions of the gene, including the coding sequence, andno deviations in sequence from that published in theliterature or changes from the sequence of an unaffectedmember of one of the families (Fig 1). These data suggestthat mutation of P15INK4B is not a mechanism for geneinactivation in familial MDS.

Elizabeth Cameron1

Aleksandar Mijovic2

James G. Herman1

Stephen B. Baylin1

Anjala Pradhan2

Ghulam J. Mufti2

Feyruz V. Rassool2

1The Oncology Center,The Johns Hopkins UniversitySchool of Medicine, Baltimore,USA, and 2Leukaemic ScienceLaboratories, Department ofHaematological Medicine, TheRayne Institute, GKT Schoolof Medicine, London, UK.E-mail: [email protected]

Fig 1. Pedigrees of MDS families. The individuals tested for P15 mutations are denoted with asterisks. Age at diagnosis is given for family

members. A bar across the patient symbol indicates deceased individual. The proband is indicated with an arrow. Family 1: The proband (II-7)

was diagnosed aged 40 years with refractory anaemia (RA) and died aged 49 years of acute myeloid leukaemia (AML) with complexkaryotypic abnormalities. One sister died of AML aged 35 years (II-1). Two sisters (II-5 ⁄ II-6) have been alive with MDS for over 10 years,

despite one of them (II-5) having unbalanced translocation t(7;15) detected in 1990. Family 2: The father of nine (I-1) developed sideroblastic

anaemia at age 60 years and died 8 years later. The proband (II-5) was diagnosed with RA with excess blasts (RAEB) at age 76 years, with anormal karyotype. A brother (II-1) also developed MDS. One sister (II-3) died of myeloma while two other brothers had monoclonal

gammopathy of unknown significance. Family 3: The proband developed RA with normal cytogenetics aged 43 years. Her sister had been

previously diagnosed with MDS-RA. The mother developed sideroblastic anaemia at age 63 years and died of AML aged 69 years. Family 4:

The proband (II-1) was diagnosed with RAEB in 1988. The proband’s mother developed RAEB in transformation (RAEBT) in 1997, aged90 years. Family 5: The proband was one of dizygotic twins, diagnosed with hypoplastic MDS in 1996, at age 18 years. During routine

screening, his human leucocyte antigen (HLA)-identical brother was found to be pancytopenic and was subsequently diagnosed with MDS. No

history of in utero exposure to carcinogens was found.

278 Correspondence


REFERENCES

Herman, J.G., Civin, C., Issa, J.-P.J., Collector, M., Sharkus, S.J. &

Baylin, S.B. (1997) Distinct patterns of inactivation of P15INK4b

and P16INK4a characterize the major types of hematologic

malignancies. Cancer Research, 57, 837–841.Hussussian, C.J., Struewing, J.P., Goldstein, A.M., Higgins, P.A.,

Ally, D.S., Sheahan, M.D., Clark, W.H., Tucker, M.A. & Draco-

poli, N.C. (1994) Germline mutations in familial melanoma.

Nature Genetics, 8, 15–21.Kamb, A., Gruis, N.A., Weaver-Feldhaus, J., Liu, Q., Harshman, K.,

Tavtigian, S.V., Stockert, E., Day, III, R.S., Johnson, B.E. &

Skolnick, M.H. (1994) A cell cycle regulator potentially involvedin the genesis of many tumor types. Science, 264, 436–440.

Mijovic, A. & Mufti, G.J. (1998) Familial syndromes (f-MDS):

towards the genetic basis of leukaemia. British Journal of

Haematology, 102, 1354.

Mufti, G.J. (1992) A guide to risk assessment in the primary mye-lodysplastic syndrome. Hematology Oncology Clinics of North

America, 6, 587–590.

Nakamaki, T., Bartram, C., Seriu, T., Kahan, J., Fukuchi, K.,Tsuruoka, N., Janssen, J.W., Miller, C.W. & Koeffler, H.P. (1997)

Molecular analysis of the cyclin-dependent kinase inhibitor

genes, p15, 16, 18 and p19 in the myelodysplastic syndromes.

Leukemia Research, 3, 235–240.

Keywords: P15INK4B gene, adult familial MDS.

RED CELL APLASIA DUE TO PARVOVIRUS B19 IN A PATIENT TREATED WITH ALEMTUZUMAB

Alemtuzumab is a humanized monoclonal antibody directedagainst the CD52 antigen, expressed on most normal andmalignant peripheral blood B and T cells. It was recentlyshown to give excellent disease control in T-cell prolympho-cytic leukaemia (Deardon et al, 2001). Alemtuzumab causesprofound and durable B- and T-cell depletion, and thereforeinduces significant cellular and humoral immunodeficiency.Opportunistic infections associated with cellular immuno-deficiency have been reported with this drug, includingPneumocystis carinii pneumonia, cytomegalovirus infection,aspergillosis and herpes simplex viral infections (Lundinet al, 1998). To our knowledge, we report the first case ofparvovirus B19-induced red cell aplasia associated withalemtuzumab therapy.

A 73-year-old woman was investigated for unsteady gaitand poor balance. She was noted to have palpable lymphglands in the right axilla. A blood count revealed lympho-cytosis of 6 · 109 ⁄ l and mild anaemia (Hb 10Æ2 g ⁄ dl). Thelymphocytes were medium-sized with irregular nuclei,indistinct nucleolous and basophilic cytoplasm. They werepositive for CD3, CD4, CD38, CD52 and CD59, and negativefor CD8, CD19, CD20, CD22, CD23 and CD79b. A probederived from the T-cell receptor (TCR) ba chain locushybridized to bands in addition to germ line hybridizationwith both of the restriction enzymes used, indicating T-cellclonality. A diagnosis of T-cell prolymphocytic leukaemiawas established.

Seven months later, the patient’s white cell count was82 · 109 ⁄ l and she began deoxycoformycin therapy. Thewhite cell count fell to 27 · 109 ⁄ l over the next 3 monthsbut then began to rise. Treatment with alemtuzumab wasstarted 14 months after initial diagnosis. She received3 mg on d 1 and 2, 10 mg on d 3 and 30 mg on d 4.Thereafter, she received 30 mg three times a week for4 weeks and entered complete remission with T cellscomprising 0Æ35% of the bone marrow. Following treat-ment, she became progressively anaemic (Hb 8Æ6 g ⁄ dl) andrequired transfusion. At this point, there were no reticu-

locytes in her blood and the direct Coombes test wasnegative, while liver function tests and serum haptoglobinswere normal. A bone marrow aspirate revealed completeabsence of erythroid precursors consistent with red cellaplasia. Myelopoiesis and thrombopoiesis were normal.These findings prompted testing for parvovirus B19 infec-tion. Quantitative DNA dot blot hybridization using serum(109 genome copies ⁄ml) confirmed parvovirus B19 vira-emia. Immune electron microscopy (Fig 1), which involvedthe addition to the patient’s sample of parvovirus B19-specific immune serum to agglutinate parvovirus B19particles making them easier to detect, further confirmedactive parvovirus B19 infection. She had no history ofcontact with any individual with a rash in the previous3 months. She received daily infusions of immunoglobulin(0Æ4 g ⁄ kg ⁄ d i.v. for 5 d) after which her reticulocyte countrose to 33 · 109 ⁄ l and 52 · 109 ⁄ l on d 7 and 14respectively. Parvovirus B19 DNA dot blot analysis onserum collected on d 14 was negative. Further tests forparvovirus B19 DNA at 3, 6 and 18 months werenegative, while haematological markers, including reticu-locyte count, have remained normal since.

Alemtuzumab impairs humoral immunity, and was themost likely explanation for parvovirus B19-induced red cellaplasia in our patient. Although bone marrow examinationrevealed erythroid hypoplasia, giant pronormoblasts con-sidered pathognomic of this infection were not seen; afinding noted previously (Abkowitz et al, 1993). Destructionof erythroid precursors constitutes the main source ofmorbidity and mortality associated with this infection(Harris, 1992). Clearance of parvovirus B19 viraemia andhence recovery of erythropoiesis are dependent on anadequate antibody response, otherwise a chronic transfu-sion-dependent anaemic state might develop. Commercialintravenous immunoglobulin contains IgG antibodies toparvovirus B19 and thereby can control it (Kurtzman et al,1987). However, monitoring for evidence of relapse isneeded, by measurement of reticulocyte counts, and assays

Correspondence 279


for parvovirus B19 viraemia until humoral immunityrecovers from the effects of alemtuzumab.

Parvovirus B19 is a remediable cause of red cell aplasia inpatients receiving alemtuzumab.

Brendan Crowley1

Barrie Woodcock2

1Liverpool Public HealthLaboratory, and 2HaematologyDepartment, University HospitalAintree, Liverpool, UK.E-mail: [email protected]

REFERENCES

Abkowitz, J.L., Brown, K.E., Wood, R.W., Cohen, B., Kovach, N.L. &

Young, N.S. (1993) Human parvovirus B19 (HPV B19) is not arare cause of anemia in HIV-seropositive individuals. Clinical

Research, 41, 393A.

Deardon, C.E., Matutes, E. & Cazin, B. (2001) High remission ratesin T-cell prolymphocytic leukemia with CAMPATH-1H. Blood,

98, 1721–1726.

Harris, J.W. (1992) Parvovirus B19 for the hematologist. American

Journal of Hematology, 39, 119–130.Kurtzman, G.J., Ozawa, K., Cohen, B., Hanson, G., Oseas, R. &

Young, N.S. (1987) Chronic bone marrow failure due to persis-

tent B19 parvovirus infection. New England Journal of Medicine,317, 287–294.

Lundin, J., Osterborg, A. & Brittinger, G. (1998) CAMPATH-1H

monoclonal antibody for previously treated low-grade non-

Hodgkin’s lymphomas: a phase 2 multicenter study. Journal ofClinical Oncology, 16, 3257–3263.

Keywords: alemtuzumab, humoral immunodeficiency,parvovirus B19, red cell aplasia, intravenous immuno-globulin.

PARVOVIRUS B19-ASSOCIATED PURE RED CELL APLASIA IN CHRONIC

GRAFT-VERSUS-HOST DISEASE

Although thrombocytopenia is well reported in chronicgraft-versus-host disease (GVHD), the finding of severeanaemia is less common. GHVD can reduce haemopoeiticprogenitor cell function (Atkinson et al, 1986) and causeautoimmune haemolytic anaemia. Other aetiologies ofanaemia include relapse, graft rejection and infection.Parvovirus B19 is a well-known cause of pure red cellaplasia in patients with chronic immunosuppressionand ⁄ or high erythrocyte turnover. The prevalence of B19parvovirus-positive serologies in adults is 40–60% (Azziet al, 1993), but the virus does not cause clinicallysignificant anaemia in normal hosts. The incidence ofpure red cell aplasia from parvovirus B19 in the post-

allogeneic transplant population is approximately 1Æ4%(Gallinella et al, 1999). All of these cases were describedwithin the first year of transplantation. We present a caseof parvovirus B19-induced pure red cell aplasia in apatient with chronic GVHD 3 years after receiving histransplant.

A 47-year-old man with mantle cell lymphoma receivedan allogeneic bone marrow transplantation from his humanleucocyte antigen (HLA)-identical brother in June 1997.GVHD prophylaxis consisted of prednisone, tacrolimus andmycophenolate mofetil. His post-transplant course wasuncomplicated until May 1998 when he developed exten-sive chronic GVHD of the skin. He was treated with steroids,

Fig 1. Immune electron microscopy showing

aggregates of parvovirus B19 particles inserum, confirming viraemia during the

patient’s illness.

280 Correspondence


psoralen plus Ultraviolet A (UVA) Light (PUVA), thalido-mide and hydroxychloroquine without benefit. A trial ofpentostatin 4 mg ⁄m2 every other week for 6 months led toimprovement in his GVHD. Infliximab 10 mg ⁄ kg weeklywas initiated and it was noted that his haemoglobin droppedfrom 10Æ3 g ⁄ dl to 8Æ2 g ⁄ dl after the first dose. His whiteblood cell and platelet counts were normal, but hisreticulocyte count was 0. His anaemia continued toprogress, and infliximab and hydroxychloroquine werediscontinued. A bone marrow aspirate and biopsy revealed< 1% erythroid precursors and giant pronormoblasts withviral inclusion bodies consistent with parvovirus B19(Fig 1). He was treated with intravenous immunoglobulinfor 5 d with normalization of his haemoglobin. Infliximaband hydroxychloroquine were restarted with no impact onhis haemoglobin levels.

It is not surprising that parvovirus B19 pure red cellaplasia occurs in bone marrow transplant recipients. TotalB-cell numbers return to normal approximately 1 monthafter transplant while total immunoglobulin concentrationscan take up to 1 year to return to normal levels (Atkinson,1990). The development of GVHD can considerably delayimmunological recovery. Most prior case series of parvo-virus B19-induced pure red cell aplasia occurred during thistime period. Our case is unusual in that the pure red cellaplasia occurred 3 years after bone marrow transplanta-tion. As far as we are aware, this is the first documentedcase of parvovirus B19-induced anaemia associated withchronic GVHD.

It is known that parvovirus B19 can produce pure red cellaplasia in the human immunodeficiency virus (HIV) popu-lation (Abkowitz et al, 1997). It is thought that eitherreduced production of immunoglobulins or the productionof ineffective immunoglobulins results in persistent B19infection from the inability to clear the virus and, therefore,

clinically significant anaemia. Our patient was severelyimmunosuppressed for over 1 year in the attempt to controlhis GVHD.

The development of pure red cell aplasia in the context ofchronic GVHD is an uncommon finding. Chronic GVHD hasbeen reported to inhibit haematopoiesis and has also beenassociated with autoimmune haemolytic anaemia. Addi-tionally, anaemia may be the harbinger of more seriousmarrow dysfunction such as graft rejection or relapse. Thiscase illustrates that it is important to perform a thoroughinvestigation into the cause of anaemia as other easilyreversible aetiologies, such as parvovirus B19 infection,may be responsible.

Jack W. Hsu

Magdalena Czader

Viki Anders

Georgia Vogelsang

Robert A. Brodsky

Sidney Kimmel ComprehensiveCancer Center at Johns Hopkins,Baltimore, MD, USA.E-mail: [email protected]

REFERENCES

Abkowitz, J.L., Brown, E., Wood, R.W., Kovach, N.L., Green, S.W. &

Young, N.S. (1997) Clinical Relevance of Parvovirus B19 as

a Course of Anemia in Patients with Human Immuno-

deficiency Virus Infection. Journal of Infectious Diseases, 176,269–273.

Atkinson, K. (1990) Reconstruction of the Haemopoietic and

Immune systems After Marrow Transplantation. Bone Marrow

Transplantation, 5, 209–226.Atkinson, K., Norrie, S., Chan, P., Zehnwirth, B., Downs, K. & Biggs,

J. (1986) Hemopoietic Progenitor Cell Function After HLA-

Identical Sibling Bone Marrow Transplantation: Influence of

Chronic Graft-Versus-Host Disease. International Journal of CellCloning, 4, 203–220.

Fig 1. (A) Bone marrow aspirate smear showing a giant pronormoblast with a large intranuclear inclusion. Note the size of this erythroid

precursor in comparison with accompanying erythrocytes and lymphocyte (·1250). (B) Immunohistochemical stain of the bone marrow

biopsy, demonstrating numerous pronormoblasts positive for parvovirus B19 (·250).

Correspondence 281


Azzi, A., Fanci, R., Ciappi, S., Zakrzewska, K. & Bosi, A. (1993)

Human Parvovirus B19 Infection in Bone Marrow Trans-

plantation Patients. American Journal of Hematology, 44, 207–209.

Gallinella, G., Manaresi, E., Benturoli, S., Grazi, G.L., Musiani, M. &

Zerbinin, M. (1999) Occurrence and Clinical Role of Active

Parvovirus B19 Infection in Transplant Recipients. European

Journal of Clinical Microbiological Infection and Disease, 18, 811–

813.

Keywords: parvovirus B19, pure red cell aplasia, GVHD.

CONGENITAL AFIBRINOGENAEMIA IN A NEWBORN INFANT DUE TO A NOVEL MUTATION

IN THE FIBRINOGEN Aa GENE

Congenital afibrinogenaemia is a rare haemorragic disorderthat is occasionally diagnosed in the neonatal period. Wedescribe haemorrhagic diathesis in a newborn infant,reflecting a novel mutation of afibrinogenaemia.

A full-term female neonate was born after an uneventfulpregnancy and delivery. She was the first child of consan-guine parents (first cousins). Following her birth, shereceived 1 mg of vitamin K orally. Two hours after birth,the child was admitted to the neonatology departmentunder suspicion of a perinatal infection and treated withantibiotics intravenously.

After admission, the child showed persistent bleeding andformation of large haematomas, following blood samplingand the insertion of intravenous lines. A haemorrhage ofthe umbilical stump required infusion of fresh-frozen plasmaand packed cells. Repeated ultrasound studies did not revealany internal haemorrhage.

Laboratory tests showed normal platelet counts, normalliver functions, partial thromboplastin time (PTT) > 120 s[normal range (N): 8–13 s], activated PTT > 300 s (N: 25–40 s), thrombin time > 300 s (N: 15–20 s), reptilase time> 300 s (N: 15–20 s). Fibrinogen (activity and antigen)< 0Æ1 g ⁄ l (N: 2–4 g ⁄ l), D-dimer < 0Æ5 mg ⁄ l (normal), andnormal activity of coagulation factors II, V and X.

Plasma fibrinogen levels of the father and the motherwere 1Æ6 g ⁄ l and 2Æ8 g ⁄ l respectively.

Subsequent DNA sequence analysis of the genes that codefor the three polypeptide chains of fibrinogen, FGA (Aachain), FGB (Bb chain) and FGG (c chain), revealed theinsertion of an extra thymine between nucleotides 3983and 3986 in exon 5 of FGA. The patient was homozygousfor this novel mutation.

Congenital afibrinogenaemia is an autosomal recessivedisorder (incidence 1–2:1 000 000), characterized by defi-cient synthesis of fibrinogen in the hepatocytes, resulting inabsent or very low levels of fibrinogen. As in our patient,

consanguine parents have been described in more than halfof the cases.

To date, mutations that are associated with afibrinogena-emia have been identified in all three fibrinogen genes with arelatively high frequency of a mutation in intron 4 of FGA(Neerman-Arbez, 2001). The novel insertion of a thymine inexon 5 of FGA that was found in our patient leads to a shift inthe reading frame of Aa with a subsequent premature end oftranslation at residue 219 (Table I). As a consequence, theAa chain would lack, if synthesized, two thirds of itspolypeptide chain and would no longer be able to interactwith the Bb and c chain in the formation of fibrinogen.

Phenotypically, congenital afibrinogenaemia is charac-terized by a mild to severe haemorrhagic diathesis(Al-Mondhiry & Ehmann, 1994; Leeners et al, 1995).Laboratory clotting tests show impaired clot formation,whereas functional and immunological tests indicateabsence of fibrinogen (£0Æ2 g ⁄ l). Signs of haemorrhagicdiathesis led to the diagnosis in our patient, althoughsymptoms only occasionally occur in neonates. A review ofthe literature of the last 16 years revealed 35 case reports,of which only two were neonates. Serious haemorrhagicdiathesis may occur at an older age. Haemorrhages mainlyoccur after minor trauma or after surgical procedures butspontaneous diathesis is rare (Al-Mondhiry & Ehmann,1994; Leeners et al, 1995). Spontaneous rupture of thespleen is a relatively frequent finding in older childrenduring the growth spurt in puberty (Ehmann &Al-Mondhiry, 1994; Shima et al, 1995).

Supplementation of fibrinogen with fresh-frozen plasma,cryoprecipitate or fibrinogen concentrate is only requiredduring periods of bleeding or before surgery (Al-Mondhiry &Ehmann, 1994; Leeners et al, 1995). Several infusions offresh-frozen plasma have been necessary in our patientbecause she frequently shows serious haemorrhages spon-taneously or after minor trauma.

Table I. Fragment of nucleotide- and amino-acid sequence of the Aa chain in normal indi-

viduals and in the patient.

Aa chain (normal) ccagacttggttcccggaaattttaagagccagcttcagaaggtacccccagagtgg

ProAspLeuValProGlyAsnPheLysSerGlnLeuGlnLysValProProGluTrpAa chain (mutation) ccagacttggttcccggaaattttTaagagccagcttcagaaggtacccccagagtg

ProAspLeuValProGlyAsnPhe****

282 Correspondence


In conclusion, a haemorrhagic diathesis in a neonate wasfound to reflect a case of congenital afibrinogenaemia dueto a novel mutation in FGA that abolishes Aa chainproduction. As the patient frequently shows haemorrhagesspontaneously or after minor trauma, the novel mutationprobably causes a relatively more severe form of congenitalafibrinogenaemia.

Jose J. Vlietman1

Jan Verhage1

Hans L. Vos2

Richard van Wijk3

Jasper A. Remijn4

Wouter W. van Solinge3

Frank Brus1

1Department of Paediatrics,Rijnstate Hospital, Arnhem,2Department ofHaematology, Haemostasisand Thrombosis ResearchCentre, Leiden UniversityMedical Centre,3Department of ClinicalChemistry and4Department of Haematology,University Medical CentreUtrecht, The Netherlands.E-mail: [email protected]

REFERENCES

Al-Mondhiry, H. & Ehmann, C. (1994) Congenital afibrinogenemia.

American Journal of Hematology, 46, 343–347.

Ehmann, W.C. & Al-Mondhiry, H. (1994) Congenital afibrino-

genemia and splenic rupture. The American Journal of Medicine,96, 92–94.

Leeners, J.V., Mossakowski, J. & Kayser, S. (1995) Fallbeispiel

einer Kongenitalen Afibrinogenamie. Klinische Padiatrie, 207,

34–35.Neerman-Arbez, M. (2001) The molecular basis of

inherited afibrinogenemia. Thrombosis and Hemostasis, 86,

154–163.Shima, M., Tanaka, I., Sawamoto, Y., Kanehiro, H., Matsuo, N.,

Nishimura, A., Giddings, J.C. & Yoshioka, A. (1995) Successful

treatment of two brothers with congenital afibrinogenemia for

splenic rupture using heat- and solvent detergent-treatedfibrinogen concentrates. Journal of Pediatric Hematology and

Oncology, 5, 462–465.

Keywords: congenital afibrinogenaemia, newborn infant,mutation, Aa gene.

RELAPSE OF SEVERE APLASTIC ANAEMIA AFTER INFLUENZA IMMUNIZATION

Following successful immunosuppressive treatment forsevere aplastic anaemia (SAA) using antithymocyte globulin(ATG) and cyclosporine, approximately one-third of patientswill relapse. There is a higher incidence of relapse in thosewho respond early to the treatment and in those who waitlonger between diagnosis and treatment (Schrezenmeier et al,1993). Relapse may occur on withdrawal of cyclosporinetherapy (Frickhofen & Rosenfield, 2000), after viral infectionsand during pregnancy (Meletis et al, 1998). It is thought thatthe immunological challenge brought about by these eventstriggers the relapse. Furthermore, up to 10% of cases ofacquired aplastic anaemia are preceded by a viral hepatitis,most often non-A, non-B, non-C hepatitis (Young & Alter,1994). As a consequence, it has been postulated thatvaccination might precipitate relapse by a similar mechanismand so clinicians often advise against vaccination in thissituation. There has, however, been no published cases ofrelapse post vaccination, although Viallard et al (2000)recently reported a case of transient bone marrow aplasiaoccurring after administration of hepatitis B vaccine. Wereport a case of relapse in SAA post-influenza vaccination.

A 60-year-old Caucasian woman was diagnosed as havingSAA in January 1996 when she presented to the hospitalwith profound pancytopenia (Hb 3Æ9 g ⁄ dl, total white cellcount 2Æ2 · 109 ⁄ l, neutrophil count of 0Æ1 · 109 ⁄ l andplatelet count 4 · 109 ⁄ l). Her illness was preceded by aninfluenza-like condition 2 weeks previously. A bone marrowbiopsy confirmed the diagnosis of SAA.

She was initially treated with ATG (Merieux UK Ltd), andsubsequently with cyclosporine. She achieved a completeremission and remained well with normal blood counts (Hb

13Æ8 g ⁄ dl, WBC 5Æ5 · 109 ⁄ l, neutrophil count of 3Æ0 ·109 ⁄ l and platelet count 177 · 109 ⁄ l) until November1998 when she received an influenza vaccine (Fluvirin;Medeva). Her last documented full blood count prior to thiswas in September 1998 (Hb 13Æ7 g ⁄ dl, WBC 5Æ9 · 109,neutrophil count 3Æ8 · 109 and platelets 175 · 109 ⁄ l.

Within 1 week of vaccination, her counts started falling,reaching a nadir in January 1999 (Hb 6Æ8 g ⁄ dl, WBC2Æ9 · 109 ⁄ l, neutrophil count of 0Æ9 · 109 ⁄ l and plateletcount 5 · 109 ⁄ l). The patient declined further treatmentwith ATG. She was therefore treated with high-dose methylprednisolone and cyclosporine, with appropriate bloodproduct support and antibiotic prophylaxis. She has beenmaintained on reducing doses of prednisolone and cyclosp-orine. Apart from requiring regular blood transfusions, sheis currently well with stable neutrophil and platelet counts(3Æ3 · 109 ⁄ l and 30 · 109 ⁄ l respectively).

The temporal relationship suggests that vaccination mayhave precipitated relapse of aplastic anaemia in this patient.Although there have been no previously published cases ofvaccination leading to relapse of SAA, there has been arecent case report of transient bone marrow aplasia postvaccination. This occurred in a previously well patient whodeveloped a fever, arthritis, cutaneous vasculitis and severepancytopenia with bone marrow aplasia 3 weeks after athird vaccination boost with a recombinant hepatitis Bvaccine (Viallard et al, 2000). It is also generally recognizedthat an acute viral infection may trigger a temporary orsometimes permanent relapse of aplastic anaemia. Thislends support to our theory that the influenza vaccinationmay have caused our patient’s relapse.

Correspondence 283


It is our current practice to advise patients with aplasticanaemia against having influenza vaccination and only togive other vaccinations if medically indicated.

C. L. Hendry1

M. Sivakumaran2

J. C. W. Marsh1

E. C. Gordon-Smith1

1Department of Haematology,St George’s Hospital MedicalSchool, London, and2Department of Haematology,Peterborough District Hospital,Peterborough, UK.E-mail: [email protected]

REFERENCES

Frickhofen, N. & Rosenfield, S.J. (2000) Immunosuppressive ther-

apy of aplastic anaemia with anti lymphocyte globulin andcyclosporine. Seminars in Haematology, 37, 56–68.

Meletis, J., Samarkos, M., Mesogitis, S., Melitis, C., Mougiou, A.,

Terpos, E., Tsimberidou, A., Andreopoulos, A., Konstantopoulos,

K. & Loukopoulos, D. (1998) Severe aplastic anaemia relapsing

during a pregnancy; spontaneous remission following a termi-

nation. Haematologica, 29, 147–151.Schrezenmeier, H., Marin, P., Raghavachar, A., McCann, S.,

Hows, Gluckmann, E., Nissen, C., van’t Veer-Korthof, E.T.,

Ljungman, P. & Hinterberger (1993) Relapse of aplastic

anaemia after immunosuppressive treatment, a report fromthe European Bone Marrow Transplantation Group

SAA Working Party. British Journal of Haematology, 85, 371–

377.

Viallard, J.-F., Boiron, J.-M., Parrens, M., Moreau, J.-F., Ranchin, V.,Reiffers, J., Leng, B. & Pellegrin, J.-L. (2000) Severe pancytopenia

triggered by recombinant hepatitis B vaccine. British Journal of

Haematology, 110, 230–233.

Young, N.S. & Alter, B.P. (eds) (1994) Aplastic Anaemia, Acquiredand Inherited. pp. 133–158. W.B. Saunders.

Keywords: aplastic anaemia, influenza vaccine, immuni-sation, pancytopenia, relapse.

ACQUIRED INHIBITOR OF THE INTRINSIC PATHWAY IN A NON-HAEMOPHILIC PATIENT.

CONTROL OF BLEEDING BY RECOMBINANT FACTOR VIIa

The incidence of acquired inhibitors against clotting factorsis rare in previously healthy subjects. Most of them arerelated to underlying autoimmune diseases or postpartumstate. We report a case of acquired postpartum bleedingdisorder in a non-haemophilic patient caused by aninhibitor of the intrinsic pathway that affected factor IXand the response to treatment with recombinant factor VIIa(rFVIIa; NovoSeven, Novo Nordisk, Denmark) and predni-sone.

A 39-year-old woman without previous history of bleed-ing disorders was admitted as a result of the recentappearance of skin haematomas in upper and lowerextremities. The patient had delivered her third child1 month before without significant bleeding, requiringanalgesic treatment with Diclofenac and Aspirin for 10 d.Initial laboratory data showed normal blood counts andbiochemistry. A plasma coagulation study disclosed aprolonged activated partial thromboplastin time (aPTT) of76Æ8 s (ratio 2Æ7) and normal prothrombin time, thrombintime, fibrinogen level and platelet aggregation. While othercoagulation tests were being carried out, the patientdeveloped an haemarthrosis at her right elbow with signsof compartment syndrome.

Clotting factors assay revealed a decreased factor IX(40 l ⁄ dl), factor XI to be on the lower limit of normal range(61 l ⁄ dl), and normal factors II, VIII, X, XII, XIII and vonWillebrand factor. No alterations in D-dimer levels wereobserved (173 ng ⁄ml), and functional tissue type plasmi-nogen activator (t-PA) and a2-anti-plasmin were barelyincreased. Screening and confirmation tests for lupusanticoagulant were negative. The aPTT and factor IX levels

were hardly corrected after mixing with control plasma,indicating the presence of an inhibitor that showed no timedependence. New mixtures with normal pooled plasmashowed the presence of an inhibitor against factor IX(2Æ5 Bethesda units).

Treatment with rFVIIa 90 lg ⁄ kg ⁄4 h and prednisone1Æ5 mg ⁄ kg ⁄24 h was started, resulting in rapid recovery offactor IX levels. However, the prolongation of aPTT persis-ted, requiring 15 infusions of rFVIIa to achieve the completedisappearance of the clinical picture. Table I shows thepatient’s coagulation data at admission and during thetreatment period.

A prospective study of 100 consecutive cases withprolonged aPTT of unknown aetiology identified the pres-ence of lupus anticoagulant as the most common cause(Kitchens, 1988). Alterations of clotting factors in theintrinsic pathway may also be responsible for the prolon-gation of the aPTT. In the present report, a postpartum-acquired inhibitor of the intrinsic pathway affecting factorIX was detected. The presence of an acquired factor IXinhibitor is infrequent in haemophilia B patients and isextremely rare in non-haemophilic patients (Feinstein,1995). A literature search identified reports of acquiredfactor IX inhibitors in a patient with autoimmune disease,two children and a patient with adenocarcinoma of thecolon.

Bleeding in patients with acquired haemophilia is oftensevere, even with low titres of inhibitor, especially duringthe first weeks after presentation. However, the presence ofa functional factor IX of 40 l ⁄ dl seems insufficient to causesuch prolongation of the aPTT. This suggests that other

284 Correspondence


components of the intrinsic pathway may be affected,although the determination of other clotting factors showedno significant alterations.

The aims in the treatment of acquired haemophilia arecontrol of the haemorrhage and elimination of the auto-antibody. rFVIIa has proven effective in patients withalloantibodies and autoantibodies against factor VIII or IX(Lusher, 1996; Negrier & Hay, 2000). In a NovoSevencompassionate-use programme in 38 patients withacquired autoimmune depletion of factor VIII (Hay et al,1997), a median of 28 doses were administered perepisode. In this case, 15 infusions were required to achievea complete clinical response. Remarkably, D-dimer did notincrease significantly during the treatment with rFVIIa.Persistence of the prolongation of the aPTT, despiterecovery of normal factor IX levels during treatment withrFVIIa, supports the hypothesis of an inhibitory actionagainst other factors involved in the intrinsic pathway.Another explanation could be interference between thepresence of large amounts of factor VIIa and the deter-mination of factor IX.

Spontaneous disappearance of postpartum-acquiredinhibitors has been reported. However, owing to the severityof the bleeding, our patient started treatment with predni-sone, which was maintained with tapered doses for4 months. After discontinuation of immunosuppresivetherapy, factor IX levels and aPTT remain between normalranges.

Ramon Lecumberri

Carlos Panizo

Jose Antonio Paramo

Marta Perez Salazar

Eduardo Rocha

Servicio de Hematologıa yHemoterapia, ClınicaUniversitaria de Navarra,Universidad de Navarra,Pamplona, Spain.E-mail: [email protected]

REFERENCES

Feinstein, D.I. (1995) Inhibitors of blood coagulation. In: Hematol-

ogy, Basic Principles and Practice, pp. 1742–1758. 2nd edn.

Churchill Livingstone.Hay, C.R.M., Negrier, C. & Ludlam, C.A. (1997) The treatment of

bleeding in acquired haemophilia with recombinant factor

VIIa: a multicentre study. Thrombosis and Hemostasis, 78,

1463–1467.Kitchens, C. (1988) Prolonged activated partial thromboplastine

time of unknown etiology: a prospective study of 100 consecu-

tive cases referred for consultation. American Journal of Hemat-

ology, 27, 38–45.Lusher, J.M. (1996) Recombinant factor VIIa (NovoSeven) in the

treatment of internal bleeding in patients with factor VIII and IX

inhibitors. Haemostasis, 26, 124–130.Negrier, C. & Hay, C. (2000) The treatment of bleeding in hemo-

philic patients with inhibitors with recombinant factor VIIa.

Seminars in Thrombosis and Hemostasis, 26, 407–412.

Keywords: acquired inhibitor, intrinsic pathway, factorIX, recombinant factor VIIa, prednisone.

Ta

ble

I.a

PT

T,

fun

ctio

na

lfa

cto

rsIX

an

dV

II,

an

dD

-dim

erp

erfo

rmed

at

ad

mis

sio

n,

1h

aft

erev

ery

rFV

IIa

infu

sio

n,

at

dis

cha

rge

an

d1

wee

ka

fter

dis

cha

rge.

Ad

mis

sio

n1

st2

nd

3rd

4th

5th

6th

7th

8th

9th

10

th1

1th

12

th1

3th

14

th1

5th

Dis

cha

rge

1w

eek

late

r

aP

TT

(s)

76Æ8

60

60Æ4

57Æ9

57Æ5

53Æ9

61Æ7

57Æ9

56Æ4

57Æ9

56Æ4

51Æ3

55Æ1

55Æ7

55Æ1

50Æ7

53Æ1

34Æ7

rati

o2Æ7

2Æ1

2Æ2

22

1Æ9

2Æ2

22

22

1Æ8

1Æ9

1Æ9

1Æ9

1Æ8

1Æ9

1Æ2

F.

IX(u

⁄dl)

40

14

11

51

19

51

72

21

31

30

13

41

36

14

81

52

18

41

43

16

61

43

16

01

88

97

F.

VII

(u⁄d

l)7

71

96

91

91

52

20

61

96

92

08

32

02

52

14

32

20

62

14

32

27

12

55

52

14

32

20

62

12

32

48

01

00

10

0

D-d

imer

(ng⁄m

l)1

73

18

01

61

29

22

13

19

72

63

23

51

94

28

12

01

17

52

39

15

52

29

15

84

29

25

0

Correspondence 285


THE USE OF RECOMBINANT FACTOR VIIA IN A JEHOVAH’S WITNESS WITH AUTO-IMMUNE

THROMBOCYTOPENIA AND POST-SPLENECTOMY HAEMORRHAGE

We report a 68-year-old female Jehovah’s Witness whopresented with a widespread purpuric rash, blood blisters inthe mouth and a platelet count of 3 · 109 ⁄ l. The blood

count was otherwise normal and bone marrow examinationwas consistent with idiopathic auto-immune thrombo-cytopenia.

Fig 1. (A and B) Serial coagulation tests, haemoglobin and platelet counts.

286 Correspondence


She was commenced on prednisolone 60 mg ⁄ d. On d 11after presentation, the platelet count was only 10 · 109 ⁄ l.It was confirmed that immunoglobulins were an accept-able treatment and the patient received a 3 d course ofintravenous immunoglobulin (IVIgG) 25 g ⁄ d. The maxi-mum platelet count achieved was 31 · 109 ⁄ l. She there-fore received methyl prednisolone 500 mg i.v. for 3 d, butthe platelet count remained at 14 · 109 ⁄ l. On d 40, thepatient was again given IVIgG, 30 g (0Æ4 g ⁄ kg) daily for4 d. On d 44, the platelet count had risen to 52 · 109 ⁄ lwith Hb 14Æ4 g ⁄ dl; this platelet count was consideredsufficient to allow splenectomy safely to be performed andsurgery was carried out on d 47. The preoperativecoagulation screen was prothrombin time (PT) 12 s(normal range, 11–16 s), activated partial thromboplastintime (APTT) 21 s (22–35 s), fibrinogen 3Æ3 g ⁄ l (1Æ5–4Æ5 g ⁄ l).

Several hours after splenectomy, her abdomen was foundto be tense and distended; the haemoglobin (Hb) was8Æ0 g ⁄ dl with platelets 46 · 109 ⁄ l. Gelatin solution(1500 ml) was given as a non-blood-derived volumeexpander and packing inserted to try to contain the bleeding;aprotinin and tranexamic acid were commenced. Thepatient was transferred to the intensive therapy unit(ITU) with Hb 4Æ5 g ⁄ dl, platelet count 17 · 109 ⁄ l, PT21 s, APTT 74 s and fibrinogen 1Æ6 g ⁄ l. Cryoprecipitatewas acceptable to the patient and she received 8 units ontwo occasions.

Despite continuing life-threatening haemorrhage and inthe face of significant thrombocytopenia and coagulopa-thy, the patient continued to refuse most plasma products.Recombinant factor VIIa (Novosen, Novo Nordisk) wasthen administered at a dose of 90 lg ⁄ kg. Haemostasis wasachieved, following a single dose, and after two dosescoagulation was normal (PT 11 s, APTT 28 s, fibrinogen4Æ4 g ⁄ l). A third dose was given after the patient returnedto theatre for removal of packs, when the coagulationtests deteriorated (see Fig 1A). Factor VIIa has beenreported to promote haemostasis in intractable haemor-rhage, e.g. following severe trauma (Kenet et al, 1999;Martinowitz et al, 2001). As it is not derived from blood,the product is acceptable to Jehovah’s Witnesses (Majum-dar & Savidge, 1993; Papatheodoridis et al, 1999). Herthrombocytopenia was managed by continuing steroidtherapy and IVIgG.

Erythropoeitin (10 000 units three times per week) wascommenced on the second day in the ITU (Koestner et al,1990); vitamin B12, folate and intravenous iron sucrosewere also administered. Further blood loss was minimizedby use of paediatric microsampling. Human or bovinehaemoglobin solutions (natural or recombinant) or per-fluorocarbons (Marelli, 1994; Doyle, 2001) were found notto be available outside the context of clinical trials in theUSA and Canada. Hyperbaric O2 was also considered, but itwas thought that the risk of transferring the patient wouldoutweigh the benefit.

On d 51, the patient suffered an episode of fast atrialfibrillation and the following day she developed Candidaalbicans pneumonia. On d 62, there was evidence of auto-

immune haemolytic anaemia (Hb 6Æ4 g ⁄ dl, positive directantiglobulin test, reticulocytes 9Æ1%, lactate dehydrogenase1488 iu ⁄ l) and further IVIgG was given. By d 72, theHb had risen to 10Æ4 g ⁄ dl, with a platelet count of131 · 109 ⁄ l (Fig 1B). Coagulation was normal as follows:PT 14 s, APTT 25 s, fibrinogen 4Æ2 g ⁄ l. Erythropoeitin wasreduced to 3000 units three times per week. Inspiredoxygen had been reduced to 40% and it was hoped towean her off the ventilator. However, on d 73, the patientsuffered two rectal bleeds, despite a platelet count of145 · 109 ⁄ l and normal coagulation. She continued tobleed and her Hb fell to 6Æ5 g ⁄ dl. It was thought that thepatient would not survive total colectomy and she died ond 75. At post mortem, there was evidence of ischaemiccolitis, a relatively common complication of long-termtreatment on the ITU, as well as adult respiratory distresssyndrome and multiorgan failure.

The treatment of this patient using non-blood methodsresulted in a significantly higher cost. The major drug costsand an estimate of the costs of keeping the patient on ITUtotalled almost £44 000. The cost of recombinant FVIIawas £10 250 but, although the final outcome in thispatient was unsuccessful, recombinant factor VIIa clearlyarrested bleeding in the face of life-threatening haemor-rhage and coagulopathy and would be of value again insimilar circumstances.

Dominic P. Waddington1

Frank T. McAuley2

John P. Hanley3

Geoffrey P. Summerfield2

1University of Newcastleupon Tyne, Newcastle uponTyne, 2Queen ElizabethHospital, Gateshead, and3Haemophilia Centre,Royal Victoria Infirmary,Newcastle upon Tyne, UK.E-mail: [email protected]

REFERENCES

Doyle, D.J. (2001) Hemoglobin-based blood substitutes forJehovah’s Witnesses. Journal of Clinical Anesthesiology, 13,

472.

Kenet, G., Walden, R., Eldad, A. & Martinowitz, U. (1999) Treat-ment of traumatic bleeding with recombinant factor VIIa. Lancet,

354, 1879.

Koestner, J.A., Nelson, L.D., Morris, Jr, J.A. & Safcsak, K. (1990) Use

of recombinant human erythropoietin (R-HuEPO) in a Jehovah’sWitness patient refusing transfusion of blood products: case

report. Journal of Trauma-Injury Infection and Critical Care, 30,

1406–1408.

Majumdar, G. & Savidge, G.F. (1993) Recombinant factor VIIa forintracranial haemorrhage in a Jehovah’s Witness with severe

haemophilia A and factor VIII inhibitors. Blood Coagulation and

Fibrinolysis, 4, 1031–1033.

Marelli, T.R. (1994) Use of a hemoglobin substitute in the anemicJehovah’s Witness patient. Critical Care Nurse, 14, 31–38.

Martinowitz, U., Kenet, G., Segal, E., Luboshitz, J., Lubetsky, A.,

Ingerslev, J. & Lynn, M. (2001) Recombinant activated factor VIIfor adjunctive hemorrhage control in trauma. Journal of Trauma-

Injury Infection and Critical Care, 51, 431–438.

Correspondence 287


Papatheodoridis, G.V., Chung, S., Keshav, S., Pasi, J. & Burroughs,

A.K. (1999) Correction of both prothrombin time and primary

haemostasis by recombinant factor VII during therapeuticalcohol injection of hepatocellular carcinoma in liver cirrhosis.

Journal of Hepatology, 31, 747–750.

Keywords: Jehovah’s Witness, recombinant FVIIa, auto-immune thrombocytopenia, post splenectomy, hae-morrhage.

288 Correspondence


p15ink4b is not mutated in adult familial myelodysplastic syndromes

Documents