Cytogenetic Abnormalities in 42 Rhabdomyosarcoma: A United KingdomCancer Cytogenetics Group Study

Tony Gordon, PhD,1,2 Aidan McManus, PhD,1,2 John Anderson, MD, PhD,2,3

Toon Min, PhD,4 John Swansbury,4 Kathy Pritchard-Jones, MD, PhD,2 andJanet Shipley, PhD,1* on behalf of UKCCG and UKCCSG

Background. Rhabdomyosarcomas are themost common type of pediatric soft tissue sar-coma. The cytogenetic literature on RMS is bi-ased towards the less common alveolar subtype(ARMS), which is frequently associated withspecific translocations and the PAX3/7-FKHRfusion genes. Relatively few karyotypes are re-ported for the embryonal subtype (ERMS). Theaim of this study was to further cytogeneticknowledge of RMS subtypes. Procedure. Rep-resentative examples of all karyotypes fromUKCCG member laboratories were reexaminedand their histopathologies reviewed throughthe United Kingdom Children’s Cancer StudyGroup (UKCCSG). Molecular evidence for thePAX3/7-FKHR fusion genes was available forfive ERMS and seven ARMS cases and com-piled with the karyotypes. Results. Clonal chro-mosome aberrations were characterized for 25ERMS and 17 ARMS cases. Thirty-six percent of

the ERMS cases involved translocation break-points in the 1p11–q11 region. Ten of the sev-enteen cases of ARMS showed cytogenetic evi-dence for the t(2;13)(q35;q14), consistent withmolecular data available from four of these.Two further ARMS cases revealed a PAX3-FKHR and a variant PAX7-FKHR fusion geneproduct that were not detected cytogenetically.Conclusions.Many of the karyotypes from bothsubtypes were complex. The frequent involve-ment of the 1p11–1q11 region and gain ofchromosomes 2, 8, 12, and 13 in ERMS may befunctionally significant. There was no evidencefor involvement of the PAX3/7-FKHR genes inERMS, and cryptic involvement was found insome ARMS. There were no consistent chromo-somal rearrangements associated with appar-ently translocation negative ARMS cases. Med.Pediatr. Oncol. 36:259–267, 2001.© 2001 Wiley-Liss, Inc.

Key words: rhabdomyosarcoma; cytogenetics; karyotypes

INTRODUCTION

Rhabdomyosarcomas (RMS) are the most commontype of pediatric soft tissue sarcoma, accounting forabout 4% of all malignancies in children under 15 yearsold [1]. They exhibit a spectrum of differentiation towardskeletal muscle. Two main morphological subtypes arerecognized, the most common of which is the embryonalform (ERMS). This occurs predominantly in the genito-urinary tract, head, and neck and generally responds wellto chemotherapy, with a relatively good prognosis. Theless common alveolar form (ARMS) frequently presentsin the extremities, often with disseminated disease and isassociated with a poorer prognosis. It exhibits a bimodalage distribution around both 3 and 15 years. These clini-cal differences are mirrored by distinct genetic changes,which have been characterized to some extent at both thecytogenetic and the molecular genetic levels.

The literature on RMS cytogenetics is not extensiveand is biased towards the less common alveolar subtype.In total 114 RMS karyotypes have been reported [2–8].

Only 31 were ERMS by histology, whereas 70 wereclassified as ARMS. Twelve had no histological classi-fication beyond RMS, and one was pleomorphic RMS[3,4]. Although accurate histological classification of

these tumors can be difficult, some generalizations aboutrhabdomyosarcoma subtypes and their associated cyto-genetic changes can be made.

ERMS do not exhibit a single common chromosomalabnormality, although some consistent cytogenetic pat-terns have emerged. Complex structural and numericalchromosomal changes, including hyperdiploidy, typi-cally with gains of chromosomes 2, 8, 12, and 13, areconsistently described [9,10]. Rarely, the t(2;13)-(q35;q14), predominantly found in ARMS, has been de-

1Molecular Cytogenetics Laboratory, Section of Molecular Carcino-genesis, Institute of Cancer Research, Sutton, Surrey, United Kingdom2Section of Pediatric Oncology, Institute of Cancer Research, Sutton,Surrey, United Kingdom3Currently, Molecular Haematology, Institute of Child Health, Lon-don, United Kingdom4Department of Academic Haematology and Cytogenetics, RoyalMarsden NHSTrust, Sutton, Surrey, United Kingdom

*Correspondence to: Dr. Janet Shipley, Molecular Cytogenetics, Had-dow Laboratories, Institute of Cancer Research, 15 Cotswold Road,Sutton, Surrey, SM2 5NG United Kingdom. E-mail: shipley@icr.ac.uk

Received 14 February 2000; Accepted 16 July 2000

Medical and Pediatric Oncology 36:259–267 (2001)

© 2001 Wiley-Liss, Inc.

scribed in ERMS, and other translocations such ast(2;5)(q37.3;q31.3) and t(11;22)(q24;q12) have beenmentioned in the literature [11,12]. At the molecularlevel, loss of heterozygosity (LOH) at 11p15.5 has beendescribed in ERMS [13,14]. However, no consistent cy-togenetic changes in this region have been observed.Comparative genomic hybridization (CGH) studies ofERMS showed chromosomal gains in concordance withthe cytogenetic findings [8,10,15].

Evidence of gene amplification associated with doubleminutes (dmin) and homogeneously staining regions(hsr) is rare but has been noted in about 10% of ERMScases [8–10].

Cytogenetic studies of ARMS have identified the fre-quent presence of t(2;13)(q35;q14) and the variantt(1;13)(p36;q14). The molecular consequences of thesetranslocations are the fusion between thePAX3 andPAX7 genes, respectively to theFKHR gene [16]. Pos-sible other variant translocations with breakpointsinvolving 2q35 and 13q14 with other partner chromo-somes have been reported [17–19]. The translocationder(16)t(1;16)(q21;q13) has been described and has beensuggested to be a consistent secondary feature in ARMS[20,21]. Sixteen percent of ARMS show cytogenetic evi-dence for gene amplification, although the incidence isconsiderably higher by CGH analysis [10,15,22]. Theaim of the current study was to further our cytogeneticknowledge of RMS subtypes, particularly the more com-mon embryonal subtype, for which data is relativelysparse, and indicate new regions of the genome that maybe significant in their development.

MATERIALS AND METHODS

The cytogenetic analysis was performed in UnitedKingdom Cancer Cytogenetics Group (UKCCG) mem-ber laboratories during the period 1985–1999. Meta-phases were obtained from either solid primary materialor bone marrow. Various standard methods of tissue dis-aggregation, short-term cell culture, harvesting, andchromosome banding were employed.

Detailed cytogenetic information on all cases carryinga provisional diagnosis of rhabdomyosarcoma were pro-vided, including failed cultures and normal karyotypes.A total of 149 cytogenetic reports from 125 cases ofrhabdomyosarcoma were contributed from ten cytoge-netics laboratories. Details from histopathology reviewwere obtained largely from the United Kingdom Chil-dren’s Cancer Study Group (UKCCSG) database, repre-senting patients entering into the Malignant Mesenchy-mal Tumours (MMT)89 and MMT95 clinical trials of theInternational Society of Paediatric Oncology (SIOP). Therelevant details were provided by local pathologists inthe few cases not entered into the trials. Representativekaryotypes from all cases included in the study were

reviewed by A. McManus, J. Shipley, T. Min, and J.Swansbury. The karyotypes were described according tothe International System for Human Cytogenetic Nomen-clature [23].

Frozen material from five ERMS and seven ARMScases was used to determine thePAX3-FKHRandPAX7-FKHR fusion gene status as part of a separate study on alarger number of cases to assess the value of identifyingfusion genes as a prognostic indicator in rhabdomyosar-coma (Anderson et al., unpublished). Detection ofPAX-FKHR fusion gene transcripts was undertaken by reversetranscription PCR (RT-PCR) and interphase FISH detec-tion of FKHRgene disruption using previously describedmethods [24,25].

RESULTS

From karyotype data submitted from the 125 cases, 83were excluded in total. Twelve cases were excluded onthe basis of their reviewed pathology and included a finaldiagnosis of alveolar soft part sarcoma, schwannoma,primitive neuroectodermal tumor, T-cell non-Hodgkinlymphoma, desmoplastic small round-cell tumor, ecto-mesenchymoma, synovial sarcoma, lipoblastoma, lipo-sarcoma, malignant rhabdoid tumor, and spindle-cell sar-coma. From a total of 59 cases, 34 and 42 reports wereexcluded on the basis of failed cytogenetic analysis andthe presence of only normal metaphases, respectively(certain cases were multiply reported). A further fivecases had only nonclonal abnormalities. Two ERMS andfive ARMS karyotypes had been previously reported andare therefore not included here [21,26–29].

Table I shows the cytogenetic results from 42 cases ofrhabdomyosarcoma included in the study. These in-cluded 25 ERMS patients, whose ages ranged from 1month to 15 years, and 17 ARMS cases, whose agesranged from 2 weeks to 20 years. Corresponding molecu-lar data relating to the translocation status were availablefor 12 of these cases, five ERMS and seven ARMS. Infive cases both pre- and post-treatment cytogeneticanalysis had been successful. Summaries of the positionsof the breakpoints and chromosomal imbalances identi-fied from the karyotypes are shown in Figures 1 and 2.Figure 3 shows a representative karyotype from ERMScase 71.

DISCUSSION

This study describes the acquired chromosomechanges characterized by G-banding analysis in 42 pre-viously unreported RMS and represents the largest groupof rhabdomyosarcoma karyotypes reported in a singlestudy. Importantly, the study includes an independentpathological diagnosis. Molecular data on the transloca-tion status were available for 12 cases as part of a sepa-

260 Gordon et al.

rate study (Anderson et al., unpublished) and providedsignificant additional information for two cases.

There have been 70 ARMS and 31 ERMS karyotypesdescribed in the literature, accounting for 61% and 27%,respectively, of the total number of karyotypes for RMSreported [9]. This is in contrast to only 40% of cases herediagnosed as ARMS, whereas 60% were ERMS. Thevariation between the number of published karyotypesfor ARMS vs. ERMS and the respective numbers de-scribed here probably reflects a reporting bias towardsARMS, the cytogenetically and molecularly better de-fined tumor subtype. The ARMS/ERMS split of cases inthis study more realistically reflects the clinical incidenceof RMS subtypes.

Significantly, the cytogenetic analysis of 25 cases ofERMS presented here nearly doubles the number ofkaryotypes reported for this subtype. These karyotypesshow that breakpoints involving 1p11–q11 are a commonfinding in ERMS, potentially representing a consistentmolecular event (Fig. 1). Nine cases in this study (36%)and four previously published cases (13%) are describedwith translocation breakpoints involving this region, al-though no consistent partner chromosomes were noted.Breakpoints at 1p11–q11 have been widely described ina diverse number of cancers [30]. These breakpointscould lead to unbalanced loss of material from 1p11-ptercontaining a tumor suppressor gene(s), gain of chromo-some 1q material containing an oncogene(s), or produc-tion of novel fusion genes. No other chromosomal re-gions were as frequently involved in rearrangements.However, three ERMS cases have been previously re-ported with breakpoints at 12q13 [26], and here a furthercase involving this region was found in the form of anadd(12)(q13). None of the cases of ERMS in this studyexhibited evidence of a t(2;13)(q35;q14).

Only three cases of ERMS in this study showed cy-togenetic evidence for genomic amplification, either asdmins or hsrs. This is concordant with a similarly lowfrequency in previously reported cytogenetic and CGHanalyses. The cytogenetic gain/loss profiles were verysimilar to those of the previously published karyotypesand CGH analysis [8–10]. Multiple copies of chromo-some 8 in particular were frequently found and also gainof chromosomes 2, 12, 13, and 20 (Fig. 1) [8–10] .

The translocation statuses of ARMS in our study andthe previously published studies were broadly similar. Asdetermined here by cytogenetics alone, 10 of 17 cases(59%) carried the t(2;13)(q35;q14), and none showed thevariant t(1;13)(p36;q14). Forty-nine of the seventy (70%)previously published cases exhibited either at(2;13)(q35;q14) or a t(1;13)(p36;q14) [9]. Amplificationof the PAX7-FKHR fusion gene and more rarely thePAX3-FKHR fusion gene has been shown to occur[29,31]. The karyotypes of cases 28 and 34 were notconsistent with the presence of the classic translocations,

but they did have dmins that may be associated with anamplified fusion gene, although it was not possible toinvestigate this. Case 43 had a complex karyotype withbreakpoints at both 2q35 and 13q14 and containeddmins. This case was positive for thePAX3-FKHRfusiongene transcript and showed disruption of theFKHRgene.A further case (case 31) was found to have aPAX7-FKHR fusion gene product, although no evidence of am-plification was identified.

The overall number of fusion gene-positive cases was12 of 17 (71%). Therefore, a proportion of ARMS casesin this study, and also in the literature, do not appear tohave have either of the known translocations or resultantfusion gene products using standard analyses [9]. Noconsistent rearrangements or variant rearrangements in-volving PAX or FKHR genes were indicated in the cur-rent study.

Breakpoints around the centromere of chromosome 1were noted in three ARMS cases and two cases showedbreakpoints at 5q31, which has been previously reportedin a single case [11]. The add(16)(q2?) described in onecase may correspond to the secondary change ofder(16)t(1;16)(q21;q13) reported in ARMS [20,21]. Thegains and losses of chromosomal material were similar tothose reported previously by CGH and cytogenetic stud-ies of ARMS, gain of chromosome 2 being one of themost commonly seen events (Fig. 2).

Twenty-four percent of the 17 ARMS cases here and16% of published cases showed cytogenetic evidence ofamplification (dmins and hsrs). In contrast to the cyto-genetic evidence, CGH studies of ARMS have shown avery high frequency of genomic amplification. In addi-tion to amplification of the fusion genes [29,31], severalamplicons are associated with this tumor type most fre-quently involving chromosomes 2, 12, and 13 [10,15,22].The genes amplified include MYCN at 2p24 and canincludeCDK4, SAS, GLI1, or MDM2 in the 12q13–q15region [32–38]. In other regions that are amplified, suchas 1q21 and 13q31, genes have not yet been implicated[9,22]. Interestingly, the regions amplified in ARMS arederived from some of the chromosomes most frequentlygained in ERMS (Fig. 1). It is not known whether thesame genes are involved potentially through subtle dos-age effects.

This study has added to the cytogenetic knowledge ofrhabdomyosarcoma, particularly for the embryonal sub-type in which we have identified that the chromosomeregion 1p11–q11 is consistently involved. This and therearrangements associated with ARMS cases apparentlylacking the known translocations require further investi-gation.

ACKNOWLEDGMENTS

We are grateful to all the laboratories that contributedto this study and are indebted to the UKCCSG for their

Rhabdomyosarcoma Cytogenetics 261

TABLE I. Clinico-pathological data with PAX-FKHR fusion gene status and karyotypes of ERMS and ARMS cases*

Caseno.

Age atdiagnosis(years)

Histologicalsub-type

Pre-/Post-treatment

Fusiongenestatus Karyotype

2 2 ERMS n.d. 54–56,XX,+2,+3,+5,+8,+8,+8,+8,+8,+11,+15,t(17;19)(q2?1;p13.3),+19[cp4]4 0.6 ERMS - RT-PCR 46,XX,t(7;18)(q36;q1?),add(17)(q21)[6]6 4 ERMS - RT-PCR 47–49,XY,+6,+add(6)(p?23),+8,+12,inc[cp13]

10 5 ERMS pre n.d. 90–96,XXY,−Y,add(1)(q21)x2,add(1)(q31)x2,add(1)(q32)x2,−3,−3,t(5;7)(q12;p11)x2,del(6)(q14)x2,i(6)(p10)x2,−8,−8,−9,−9,−14,−14,−17,−17,mar1x2,mar2x2,+14mar,dmin

post n.d. 45–49,X,−Y,add(1)(q32),t(1;14)(p10;q10),+add(1)(q21),−2,add(2)(q23),−3,del(3)(q12),del(4)(q22q25),t(5;7)(q12;p11),−6,+add(8)(q22),−9,del(10)(q26),add(12)(q21),−13,add(13)(p11),−14,add(14)(q32),−19,−20,−21,+r,+10mar48−49,X,−Y,add(1)(q21),−2,−3,del(3)(q24),−4,t(5;7)(q12;p11),add(6)(q14),−8,add(8)(q13),−9,del(10)(q26),del(11)(p14),add(12)(q21),−13,−14,+der(?)t(?;1)(q21),+6mar,inc

12 3 ERMS pre n.d. 55,XY,+2,+7,+8,+8,+11,+12,+12,+17,add(18)(q23),+2055,XY,+2,+7,+add(8)(q24),del(9)(p22),del(10)(p12),+11,+12,+12,+17,add(18)(q23),+2055,XY,der(1;4)(q10;q10),+2,+7,+8,+8,add(9)(p12),+ins(11;?)(p13;?),+12,+12,+17,+2055,XY,add(2)(q21),+7,+8,+8,+der(11)t(1;11)(q25;p11),+12,+12,+17,add(18)(q23),

der(19)t(2;19)(q24;p13),+20post n.d. 53,XY,+2,+7,+8,+8,+12,+2mar

53,XY,+2,+7,+8,+12,+20,+2mar13 1.6 ERMS n.d. 56,XY,+2,+7,+8,+12,+13,+13,+13,+15,+19,+20

58,XY,idem,+10,+1061,XY,idem,+add(1)(p11),+3,+10,+10,+12

19 4 ERMS - FISH 52,XX,der(2)t(2;13)(q1;q1)x2,−2,+8,+11,+12,+12,hsr(16)(p1?),+20,+mar1,+mar220 0.6 ERMS n.d. 52,XX,+2,+5,+8,+8,+11,−14,+15,+2026 0.1 ERMS n.d. 45,XX,−9,add(11)(p15),add(17)(p11)27 6 ERMS n.d. 70–72,XXX,+add(X)(p2),−3,−3,−4,−5,add(11)(q2),−13,−14,−15,−17,−17,−18,+19,+19,−21,+10−12mar,inc[5]33 2 ERMS n.d. 47,XX,+8,der(15)t(1;15)(q11;p11.1)[13]

47,XX,idem,add(12)(q13)[5]36 2 ERMS - RT-PCR 56–60,XY,−X,−Y,+der(1)t(1;19)(p11;q11),−3,−4,−6,−7,+8,−9,−10,+12,−15,−17,−18,−19,−22[cp7]39 1 ERMS pre n.d. 69,XX,−X,−1,+add(1)(p?)x2,−2,−4,−5,+8,+8,+8,+8,−9,−10,+13,+13,+13,−15,−16,−18,−20,+mar1,+mar2,

post n.d. 44–54,X,−X,−1,del(1)(q21−qter),−2,add(3)(q27),−4,−5,−6,−7,−8,−9,−10,del(11)(p11.2),del(12)(p11.2),−13,−16,−16,−17,−17,−18,+19,−21,+3–6mar,3–4dmin[cp10]

40 3 ERMS pre n.d. 57,XY,add(1)(q?),+2,+3,+5,+8,+10,+12,+13,+13,+16,+19,+20post n.d. 58,XY,add(1)(q?),+der(1),+2,+3,+5,+10,+12,+13,+13,+16,+19,+20,+i(21)(q10)

42 14 ERMS n.d. 45,XX,der(5)t(5;18)(p13;q11.2),r(13),−1849 13 ERMS n.d. 53–56,XX,add(7)(p?),inc50 3 ERMS n.d. 53–63,XX,+2,+6,+19,+20,+3mar,inc52 13 ERMS n.d. 55,XY,+2,+5,+6,+8,+10,+13,+13,der(9)(9qter−>9p1::?::1q2−>1qter),der(9)(9qter−>9p1::?::hsr::?),der(12)

(12qter−>12p11::?::hsr::1q2−>1qter),der(14)(14qter−>14p11::?),+der(19)(?19pter−>?19q13::1q21−>1qter),+mar53 15 ERMS n.d. 60–74,XY,+del(1)(p11p22),add(2)(q37),+5,+8,+12,+del(12)(p12),+18,+18,+21,+22,+5mar,inc55 ERMS n.d. 100–111,XXYY,−1,+2,+4,+5,+5,+5,+6,+6,−7,+12,−13,−14,−15,+20,+20,+21,+21,+mar,dmin,inc58 4 ERMS n.d. 60,XY,+2,+2,+5,+del(6)(q21),+7,+8,+8,+8,+add(8)(p22),+11,+14,+19,+mar65 1 ERMS n.d. 50,XY,+2,add(6)(q13),+8,+12,add(15)(p11),+20[6]

TABLE I. ( Continued)

Caseno.

Age atdiagnosis(years)

Histologicalsub-type

Pre-/Post-treatment

Fusiongenestatus Karyotype

66 3 ERMS n.d. 43–46,X,−Y,+X[3],del(1)(q31),add(2)(p21)[5],del(4)(q21),add(5)(q11)[6],del(5)(q)[2],+8[3],add(8)(q24)[2],add(9)(p)[3],+11,add(11)(p)[4],+13[4],−14,−16,−17,add(17p)(13),−22[cp7]

70 9 ERMS n.d. 61–70<2n>,add(X)(p22),Y,+Y,+1,idic(1)(p1?),+2,+2,+5,+6,+7,+7,+7,+8,+8,+8,+10,+11,+12,+12,+14,+15,+18,+18,+19,+20,+21,+mar[cp6]

71 6 ERMS - FISH 48,XX,+2,+der(8)t(3;8)(p?21−?25;q22)[13]49,idem,+12[2]49,idem,+r[13]

11 3 ARMS n.d. 86–112,XXX,der(1),der(10),mar,inc21 ARMS n.d. Hyperdiploid,t(2;13)(q35;q14)28 0.04 ARMS n.d. 70,XYY,−3,+5,+8,+13,+20,−21,inc[18]

71,idem,+2[10]140,complex,dmin[1]

30/44 5.1 ARMS pre - RT-PCR 47,XY,del(15)(q15–q22),der(19)t(1;19)(q?11;p13.3),+20post - RT-PCR 91–94,XXY,−Y,t(1;19)(q21;p13.3),+der(19)t(1;19)(q21;p13.3),−4,+5,+6,+7,+7,−9,add(9)(q34),−12,−14,+20,+21,3dmin[cp5]

31 2 ARMS post PAX7-FKHR 46,add(X)(q22),Y,t(1;1)(q12;p13),add(4)(p12),add(9)(q?),add(17)(p13)RT-PCR 46,XY,t(2;16)(p25;p13.1),inv(5)(p14q33),t(10;11)(q26;q13)

46,XY,t(1;9)(p?;q?),add(2)(p?),add(3)(q?),t(3;7)(p?;q?),add(4)(q?),add(8)(q?),t(9;15)(q?;q?),add(11)(p?),del(11q?),add(13)(q?),add(17)(q?)

34 2.2 ARMS n.d. 46,XX,dmin41b 1 ARMS PAX3-FKHR 46,XY,t(2;13)(q35;q14)

FISH,RT-PCR43 4 ARMS post PAX3-FKHR 41–91<4n>,XYY,−X,add(1)(p13),+der(1;12)(p10;p10),−2,der(2)t(2;2;?)(p21;q35;?)x2,der(2)t(2;3)(q14.2;q12),−3,add(3)

RT-PCR (q25.2),add(4)(q35),del(4)(q27)x2,add(5)(q31),−6,add(6)(p21.3),del(7)(q21q34)x2,der(9;12)(q10;q10),dic(9;14)(p22;q32),−10,del(10)(p15)x2,del(11)(p11.1),−12,add(13)(q14)x2,add(14)(q32)x2,der(14)t(7;14)(q11.2;q32)x2,del(15)(q15)x2,−16,−16,+add(17)(p11.2),add(17)(q23),−19,add(19)(q13.3),−20,−21,+22,+22,der(22;22)(q10;q10)x2,+5mar,3–20dmin[cp5]

46 1 ARMS n.d. 89,XXYY,+Y,−1,+7,+14,+18[cp3]47 14 ARMS n.d. 96,XXYY,t(2;13)(q36;q14.1)x2,+2,−3,+der(13)t(2;13)(q36;q14.1),add(16)(q2?)x2,+17,+19,+19STS ARMS PAX3-FKHR 46,XX,t(2;13)(q36.1;q14.1)354a RT-PCR51 18 ARMS PAX3-FKHR 81,XYY,−X,t(2;13)(q36;q14),der(2)t(2;13)(q36;q14),−3−4,−6,−8,−9,−10,+12,−14,−16,−18,−19

FISH,RT-PCR54 ARMS n.d. 49,XY,t(2;13)(q36.1;q14.1),+12,der(13)t(2;13)(q36.1;q14.1),+20

48,idem,−Y96,idemx2,−Y,−Y

56 ARMS n.d. 93,XXYY,del(1)(p3?),del(1)(q2?;q32),t(2;13)(q36.1;q14.1)x2,+2,del(5)(q1?q3?),+del(5)(q1?q3?),+7,−9,−10,−16,+20,−2264 15 ARMS PAX3-FKHR 81,XXYY,t(2;13)(q35;q14),+der(13)t(2;13)(q35;q14),inc[1]

RT-PCR68 20 ARMS n.d. 79,XXY,−Y,−1,t(2;13)(q35;q14)x2,−3,−4,−6,−8,−9,−10,−10,−13,−16,−18,

−21,−22,inc[cp3]69 ARMS n.d. 64,XX,−X,del(1)(p3?),t(2;13)(q35;q14)x2,−3,der(4)t(3;4)(q?;q13),−6,−10,

−16,+20,−21[6]

*RT-PCR, reverse transcription PCR analysis forPAX3/7-FKHRfusion gene product; FISH, fluorescence in situ hybridization analysis using probesflanking the FKHR gene; -, no evidence for presence of either PAX3-FKHR or PAX7-FKHR fusion gene product; n.d., translocation status notdetermined. Comparative genomic hybridization results published on cases.aWeber-Hall et al., 1996.bGordon et al., 2000.

Fig. 1. Chromosomal breakpoints and gains/loss of chromosomal material in 25 embryonal rhabdomyosarcoma cases. Circles represent breakpoints of either reciprocal or unbalanced translo-cations. Vertical lines to the left of chromosome ideograms represent losses of material, and vertical lines to the right of chromosome ideograms represent gains of material. Thick lines representmultiple gains or losses in single cases.

Fig. 2. Summary of chromosomal breakpoints and gains/loss of chromosomal material in 17 alveolar rhabdomyosarcoma cases. Circles represent breakpoints of either reciprocal or unbalancedtranslocations. Vertical lines to the left of chromosome ideograms represent losses of material, and vertical lines to the right of chromosome ideograms represent gains of material. Thick linesrepresent multiple gains or losses in single cases.

help and cooperation. We also acknowledge supportfrom the Medical Research Council of the United King-dom and the Royal Marsden Hospital Children’s CancerUnit Fund.

REFERENCES

1. Enzinger FM, Weiss SW. Soft tissue tumors, 3rd ed. St. Louis:Mosby-Year Book, Inc.; 1995.

2. Kenet G, Sharon N, Rosner E, et al. Chromosomal translocation(1;13) in a case of alveolar rhabdomyosarcoma. J Pediatr HematolOncol 1998;20:86–87.

3. Kullendorff C, Donner M, Mertens F, et al. Chromosomal aber-rations in a consecutive series of childhood rhabdomyosarcoma.Med Pediatr Oncol 1998;30:156–159.

4. Mitelman F, ed. Catalogue of chromosome aberrations in cancer1998, 1 ed. New York: Wiley-Liss; 1998.

5. Savasan S, Lorenzana A, Williams J, et al. Constitutional balancedtranslocations in alveolar rhabdomyosarcoma. Cancer Genet Cy-togenet 1998;105:50–54.

6. Gil-Benso R, Carda-Batalla C, Navarro-Fos S, et al. Cytogeneticstudy of a spindle-cell rhabdomyosarcoma of the parotid gland.Cancer Genet Cytogenet 1999;109:150–153.

7. Polito P, Dal Cin P, Sciot R, et al. Embryonal rhabdomyosarcoma

with only numerical chromosome changes. Cancer Genet Cyto-genet 1999;109:161–165.

8. Bridge J, Liu J, Weibolt V, et al. Novel genomic imbalances inembryonal rhabdomyosarcoma revealed by comparative genomichybridization and fluorescence in situ hybridization: an Intergrouprhabdomyosarcoma study. Genes Chrom Cancer 2000;27:337–344.

9. Anderson J, Gordon A, Pritchard-Jones K, et al. Genes, chromo-somes and rhabdomyosarcoma. Genes Chrom Cancer 1999;26:275–285.

10. Pandita A, Zielenskam M, Thorner P, et al. Application of com-parative genomic hybridization, spectral karyotyping and micro-array analysis in the identification of subtype-specific patterns ofgenomic changes in rhabdomyosarcoma. Neoplasia 1999;1:262–275.

11. Moriyama M, Shuin T, Kubota Y, et al. A case of rhabdomyo-sarcoma of the bladder with a (2;5) chromosomal translocation inperipheral lymphocytes. Cancer Genet Cytogenet 1986;22:177–181.

12. Thorner P, Squire J, Chilton-MacNeil S, et al. Is the EWS/FLI-1fusion transcript specific for Ewing sarcoma and peripheral primi-tive neuroectodermal tumor? A report of four cases showing thistranscript in a wider range of tumor types. Am J Pathol 1996;148:1125–1138.

Fig. 3. Representative karyotype from an embryonal rhabdomyosarcoma. Case 71: 48,XX,+2,+der(8)t(3;8)(p?21–?25;q22),+r.

266 Gordon et al.

13. Scrable H, Witte D, Shimada H, et al. Molecular differential pa-thology of rhabdomyosarcoma. Genes Chrom Cancer 1989;1:23–35.

14. Anderson J, Gordon A, McManus A, et al. Disruption of imprintedgenes at chromosome region 11p15.5 in paediatric rhabdomyo-sarcoma. Neoplasia 1999;1:340–348.

15. Weber-Hall S, Anderson J, McManus A, et al. Gains, losses andamplification of genomic material in alveolar rhabdomyosarcomaanalyzed by comparative genomic hybridisation. Cancer Res1996;56:3220–3224.

16. Davis R, Barr F. Fusion genes resulting from alternative chromo-somal translocations are overexpressed by gene-specific mecha-nisms in alveolar rhabdomyosarcoma. Proc Natl Acad Sci USA1997;95:8047–8051.

17. Hayashi Y, Inaba T, Hanada R, et al. Translocation 2;8 in a con-genital rhabdomyosarcoma. Cancer Genet Cytogenet 1987;30:343–345.

18. Douglass EC, Valentine M, Etcubanas E, et al. A specific chro-mosome abnormality in rhabdomyosarcoma. Cytogenet CellGenet 1987;45:148–155.

19. Sawyer J, Crussi F, and Kletzel M. Pericentric inversion(2)(p15q35) in an alveolar rhabdomyosarcoma. Cancer Genet Cy-togenet 1994;78:214–218.

20. Mrozek K, Arthur D, Karakousis C, et al. Der(16)t(1;16) is anonrandom secondry chromosome aberration in many types ofhuman neoplasia. including, myxoid liposarcoma, rhabdomyosar-coma and Philadelphia chromosome positive acute lymphoblasticleukemia. Int J Oncol 1998;6:531–538.

21. McManus A, Min T, Swansbury G, et al. der(16)t(1;16)(q21;q13)as a secondry change in alveolar rhabdomyosarcoma. A case re-port and review of the literature. Cancer Genet Cytogenet 1996;87:179–181.

22. Gordon A, Brinkschmidt C, Anderson J, et al. A novel and con-sistent amplicon at 13q31 associated with alveolar rhabdomyosar-coma. Genes Chrom Cancer 2000;28:220–226.

23. Mitelman F, ed. ISCN: an international system for human cyto-genetic nomenclature. Basel: S. Karger; 1995.

24. McManus A, O’Reilly M, Pritchard-Jones K, et al. Interphasefluorescence in situ hybridization detection of t(2;13)(q35;q14)—a diagnostic tool in minimally invasive biopsies. J Pathol 1996;178:410–414.

25. Anderson J, Renshaw J, McManus A, et al. Amplification of thet(2;13) and t(1;13) translocations of alveolar rhabdomyosarcomain small formalin-fixed biopsies using a modified reverse tran-scriptase polymerase chain reaction. Am J Pathol 1997;150:477–482.

26. Roberts P, Browne C, Lewis I, et al. 12q13 abnormality in rhab-domyosarcoma. A nonrandom occurence? Cancer Genet Cyto-genet 1992;60:135–140.

27. Bown N, Reid M, Malcolm A, et al. Cytogenetic abnormalities ofsmall round cell tumors. Med Pediatr Oncol 1994;23:124–129.

28. Yule S, Bown N, Malcolm A, et al. Solid alveolar rhabdomyo-sarcoma with a t(2;13). Cancer Genet Cytogenet 1995;80:107–109.

29. Weber-Hall S, McManus A, Anderson J, et al. Novel formation

and amplification of thePAX7-FKHR fusion gene in a caseof alveolar rhabdomyosarcoma. Genes Chrom Cancer 1996;17:7–13.

30. Mertens F, Johnasson B, Hoglund M, et al. Chromosomal imbal-ance maps of malignant solid tumors: a cytogenetic survey of3185 neoplasms. Cancer Res 1997;57:2765–2780.

31. Barr F, Nauta L, Davis R, et al. In vivo amplification of thePAX3-FKHRand PAX7-FKHR fusion genes in alveolar rhabdo-myosarcoma. Hum Mol Genet 1996;5:15–21.

32. Garson J, Clayton J, McIntyre P, et al. N-myc oncogene amplifi-cation in rhabdomyosarcoma at release. Lancet 1986;1:1496.

33. Dais P, Kumar P, Marsden H, et al. N-myc is amplified in alveolarrhabdomyosarcoma (RMS) but not in embryonal RMS. Int J Can-cer 1990;45:593–596.

34. Driman D, Thorner P, Greenberg M, et al. MYCN gene amplifi-cation in rhabdomyosarcoma. Cancer 1994;73:2231–2237.

35. Boman F, Brel D, Autunes L, et al. Gene alterations and apoptosisin rhabdomyosarcoma. Pediatr Pathol Lab Med 1997;17:233–247.

36. Roberts M, Douglass E, Peiper S, et al. Amplification of the gligene in childhood sarcomas. Cancer Res 1989;49:5407–5413.

37. Forus A, Florenes V, Maelandsmo G, et al. Mapping of amplifi-cation units in the q13–q14 region of chromosome 12 in humansarcomas: some amplica do not include MDM2. Cell Growth Dif-fer 1993;4:1065–1070.

38. Keleti J, Quezado M, Abaza M, et al. The MDM2 oncoprotein isoverexpressed in Rhabdomyosarcoma cell lines and stabilizeswild-type p53 protein. Am J Pathol 1996;149:143–151.

APPENDIX

United Kingdom Cancer Cytogenetics Group(UKCCG) Chairman: C. Harrison; Secretary: M. Martin-eau; Department of Haematology, Royal Free Hospitaland University College School of Medicine, London. Afull list of members is available from the Secretary. TheUKCCG laboratories that contributed to this study wereas follows.

Merseyside and Cheshire Genetics Laboratories,Liverpool Women’s Hospital; Regional CytogeneticsCentre, Southmead Hospital, Bristol; Duncan Guthrie In-stitute of Medical Genetics, Yorkhill NHS Trust Hospi-tals, Glasgow; West Midlands Regional Genetics Ser-vices, Birmingham Women’s Hospital; Department ofHuman Genetics, University of Newcastle; The GeneticsUnit, Westlakes Research Unit, Cumbria; North TrentCytogenetics Service, Sheffield Children’s Hospital NHSTrust; Cytogenetics Department, St. James UniversityHospital Trust, Leeds; Wessex Regional Genetics Labo-ratory, Salisbury District Hospital; ICRF Department ofMedical Oncology, St. Bartholomew’s Hospital, London.

Rhabdomyosarcoma Cytogenetics 267