Document dedROLgqv0kkV68gG3ZeBzr6

826 T H E NEW ENGL4ND JOURNAL O F MEDICINE Oct. 6: 1983 LYMPHOID BLAST CRISES OF CHRONIC MYELOGENOUS LEUKEMIA REPRESENT STAGES IN THE DEVELOPMENT OF B-CELL PRECURSORS AJAYBAKHSHI, M.B., B.S., J U N MINOWADA, M.D., ANDREW ARNOLD, M.D., JEFFREY COSSMANM,.D., JANE P. JENSEN, M.S., JACQUELINE WHANG-PENG, M.D., THOMAS A. WALDMANMN,.D., AND STANLEY J. KORSMEYER, M.D. Abstract The origin and stage of differentiation of the blast-crisis cells in chronic myelogenous leukemia have remained uncertain. Because immunoglobulin heavychain and light-chain genes must undergo a DNA rearrangement during B-cell development but rarely do so in human non-B-cell lineages, we examined these genes in 18 episodes of chronic myelogenous leukemia. In eight of nine episodes of lymphoid blast crisis, heavy-chain genes were rearranged, and in three, rearrangements in lightchain genes were also present. In contrast, cells from chronic myeloid, myeloid blast, and erythroid-like phases retained germ-like immunoglobulin genes. The observed phenotypic markers and gene configurations revealedthat most lymphoid blast crises represent stages of development of B-cell precursors. In two separate episodes of lymphoid crisis, cells from a single patient possessed identical heavy-chain but different light-chain-gene configurations. Thus, the precursor cells that monoclonall:. expand to produce a lymphoid crisis are capable of im munoglobulin-gene rearrangements and represent discrete steps in early B-cell maturation. (N Engl J Med 1983; 309~826-31.) CYTOGENETIC and isoenzyme studies have established that chronic myelogenous leukemia is a clonal proliferative disorder arising from a progenitor cell with a rather pluripotential capacity.1-6Studies of glucose-6-phosphate dehydrogenase isoenzymes in female heterozygotes with chronic myelogenous leukemia have revealed that involved granulocytes, erythrocytes, platelets, monocytes, and some lymphocytes displayed only a single glucose-6-phosphate dehydrogenase allele.3-6This finding indicated that in this disease, cells in multiple hematopoietic lineages could be of clonal origin. Because of this multipotential capacity of the clonally derived cell in chronic myelogenous leukemia, the exact cellular origin and indeed the state of differentiation of the cells comprising the blast-crisis phases of the disorder have remained uncertain. Occasionally, lymphoblasts from the acute phase have been classifiable as pre-B cells because of the presence of cytoplasmic immunoglob.~lin.'-H~ owever, most cases of lymphoid blast crisis lack both detectable cytoplasmic and surface immunoglobulin as well as definitive T-cell antigens that would permit their clear assignment to either B-cell or T-cell lineage.' Both heavy-chain and light-chain immunoglobulin genes must undergo a successful DNA rearrangement during B-cell development before a complete immunoglobulin molecule can be produced.lO.llThese rearrangements begin with an attempt to juxtapose the separated variable ( V H ) ,diversity (DH),and joining (JH) gene subsegments of the heavy-chain gene.12 Subsequently, germ-line light-chain genes must be similarly rearranged to recombine V, and JK or VA 'and Jh subsegments before a light chain can be pro- duced.".' Because these rearrangements are mandatory in B cells but only rarely occur within other From the Metabolism Branch, Laboratory of Patholozy. and Division of Cancer Treatment. National Cancer Institute. National 1nstiNter of Health. Bethesda. Md.. and Hines Veterans Administration Medical Center. Hines, Ill. Address reprint q u e s t s to Dr. Bakhshi at Bldg. IO. Rm. 4N104.National Institutes of Health. Bethcsda, MD 20205. Pan of this work has appeared in abstract form (Clin Res 1983: 31:308A). human hematopoietic lineages,I3 we examined the status of immunoglobulin genes in chronic myelogenous leukemia. In eight of nine episodes of lymphoid blast crisis. rearrangements of heavy-chain genes were observed and in three progression to light-chain-gene rear rangements had occurred as well. In contrast, cells from chronic myeloid and acute blastic myeloid phases had germ-line immunoglobulin genes. Thus, the lymphoid blast crises had initiated immunoglobulin-gene rearrangements and appear to represent monoclonal expansions of cells at B-cell-precursor stages ofdevelopment. Serial studies ofa single patient indicated that the separate cellular expansions during different crises can originate from distinct B-cell prccursors at discrete genetic stages of maturation. METHODS Cases Examined The leukemic cells were obtained from the peripheral blood or bone marrow during 14 episodes of chronic myelogenous leukemia and four representative spontaneous cell lines (RPMB, NALM-I, PTF, and K562) established during blast crisis of the disease were examined.s*'4.'5Eight patients were studied during nine episodes of lymphoid blast crisis (Cases I through 8 in Table 1). T w o patients were examined during a chronic granulocytic phase (Cases 2c and 9), a n d six during a myeloid biast crisis (Cases IO through 15) (n, listed in Table 1). T h e K562 erythroid-like cell line was examined the 16th case." Case 2 was studied on three occasions: during tv .I separate lymphoid blast crises (Cases 2a and 2b in Table 1) and during an intervening chronic granulocytic phase (Case 2c in Fig. 1B). Cells from 14 patients possessed a t(9;22) translocation, Phila- delphia chromosome, whereas the R P M B a n d PTF cell lines (Cases 6 and 8) had no t(9;22) translocation, representing the form of chronic myelogenous leukemia lacking the Philadelphia chromosome." None of the cells studied during the 18 episodes examined had cell-surface immunoglobulin, and none formed rosettes with sheep red cells. T h e presence of cytoplasmic immunoglobulin of classes 9. y, K, and A was assessed when possible either by cytoplasmic i: . munofluorescence or by measurements of the actual immunoglol lin present in cytoplasmic extracts of leukemic cells with a semiti. ~' double-antibody radioimmunoassay.13 Terminal deoxynucleotid! I- transferase (TdT)I6was present in all cases of lymphoid blast crisi, tested but was absent during the chronic and acute myeloid phases. ~ 111 +I983 \ ( , I . :309 No. I C H R O N I C ,L[YELOGENOUS LECKEXII;\ - B.4KHSHI E T ,AL, 827 ;ES Table 1. S t u d i e s of Nine Episodes of Lymphoid Blast Crisis in Eight Patients with Chronic Myelogenous Leukemia. PHENOTYPIC bfARKERE Ned that :lops of ;sed ;on- ially imdis383; the ige- sis, .ed, :arells oid IUS, bUen t .sor ent ing ire- I or mia 1-1, ere s of mts ind not I in :wo ind 'lg. ila- ses rm ro- lad :eP I P> m- 1u- ~ ~ 1 :ve I 1 isis . SURFACE \NTIGENF 5 6 RPhIB 7 N4L.M-I R FTF 15 TdT CALLA ++ ++ DA-2 BA-i HLA-DR pj0 7i -- BA-2 p ? l TAX MAg + - - +-- ++ + - +-- ++ r +--- ++ + +36-- + 17 NT - ++ ++ 20 17 + - 30 + NT NT - + 19 + - NT + +- - - ilC"I1. ,h," I Cp: 2 R e m JH: 2 Rem Cp: Z R e m JH: 1 R e m Cp: 1 Rem I,. 2 R e m Cp: 1 Rem JH: NT Cp: 2 R e m I,. 2 R e m Cp; Germ JH: Germ Cp: I Rem. I Del J,: 2 Rean Cg: 2 R e m JH: 2 R e m Cp: Del J,,: 2 R e m Germ Germ : Gem Germ Del Germ I Rem Gem Germ Germ I Ream Germ NT Germ Germ Germ Germ Germ ,uh ,!ens indicale that < 10 per cent of the cells macled wrh the munoclonal u u b c d y J5, whlch recupnize\ rhc common ~cuic-l)mphoblasric-leuk~m~niuagen ICALLAI. DA-2. EA-I. or .I 3 tluarrsccnce-acuvaled cell soner. or possessed T-cell xsaaciaied anugena I T A g UT myooid-arroclird inrigens IMAS).PI", ; i p s indnalc >50 px cent i e a c u v ~ y a. nd values k r w e e n .,,a I U p r cent m spccific~llyIisrcd. Tdt denores deorynucleoudyl tnnsiense. and NT nor resled. , H a r v - c h a m and light-cham gcnc parrerns wrrc determined with rhc C p . JH. C.. u d C, probo Rcon dcnoier remaneed. C m n . :erm-line: and Del. deleted :ticrm-ltne I( genes may have h e n conrnbuied by normal cells Immunoglobulin-Gene Analysis High-molecular-weight DNA prepared from leukemic cells or cell lincs was digested to completion with BamHI or EcoRI restriction rrtdonuclease, size-fractionated by agarose-gel electrophoresis. and red to nitrocellulose paper.'' These nitrocellulose blots were Led with nick-translated 3nP D N A probes of the human imI it.iohulin genes that were capable of detecting germ-line and ,...:iinnced alleles (Fig. 1.4).'"." Such blots were washed at [he .Ippropriate strinqencies a n d were visualized o n autoradiograms. .I iic~pulationof normal, polyclonal B cells contains many diKerentlv 9ized imniunoglohuiin-gene rearransements and when collectivelv .inal)ized reveals no detectable individual rearrangement. In contrast a B-cell-lineage leukemia is a monoclonal expansion in which ~ 3celil has a n identical immunoglobulin-,qene recombination and displays an identifiable rearranged immunoglobulin-gene fragment. h c h reaxingemcnts are a sensitive marker capable or detecting WYI minority populations (5 per cent) of clonal B cells when ad~ ' trith non-B cells or polyclonal B cells. L,.: :ace-Antigen Characterization \\IICII available, leukemic cells were reacted with monoclonal ,Illttbodies or directlv Huoresceinated antiserum and were analyzed I)\. How microfluorometry as previouslv described.'' b'hen more I1l.m 50 per cent o f t h e cells o f a case reacted with a n antibody i t was m r e d as positive, less t h a n 10 per cent was scored as negative, a n d vdues between 10 a n d 50 per cent a r e specifically listed for the I\mphoid crises in Table I . T h e antibodies used include the J5 nllinoclonal antibodv (kindly provided by Dr. Jerome Ritz. Har', ,Ird L'niversirv), which recognizes the common acute-lymphohlas' : I ~'-5ll;emia ntigen," the DA-2 monoclonal antibody (a kind gift ' . Peter Parham and Jack Strorninger, Harvard University). , : :, :dentilies a nonpolymorphic H L A - D R determinant." the [monoclonal antibodv (courtesy of Dr. Tucker LeBien. Lni"l'lsIt) of Slinnesota), which detects a p30 antigen that is expresscd on cells at multiple stages of B-cell development," and the BA-2 monoclonal antibody I provided by Dr. Tucker LeBien, University of Xlinnesota), which identifies a 24,000-dalton leukemia and B-cell-associated antigen'3 (Table I ) . T h e presence of T-cell-associ- aced antigens was assessed with either the O K T 3 . O K T 4 , O K T 6 , and OKT8 monoclonal antibodies (Orrho Pharmaceuticals, Rari- tan. S.J.)"or an absorbed heteroantiserum specific for human T-cell-surface antigen^.^' Results Tor the lymphoid cases are listed under T.Ag in T a b l e I . T h e existence of myeloid antigens was deter- mined with the OKX11'6 ( O r t h o Pharmaceuticals) and MCS-'? monoclonal antibodies2' and another absorbed heteroanriserum recoTnizing myeloid Results are listed under .Ll;\g in Table I . RESULTS Immunoglobulin-Gene Rearrangements Cells from seven of the eight patients examined during a lymphoid blast crisis possessed rearranged hea\.y-chain immunoqlobulin genes, and three of these also had rearranged light-chain genes (Table 1 and Fig. 1B ) . These rearrangements result from attempts at recombinational joining of the separated immunoglobulin-gene subseqments VH, DH, and JH, which comprise the final heavy-chain variable region, or VL and JL, which comprise the final light-chain variable region. These DN.4 reorganizations within B cells occur in a sequential order in which rearrangements of heavy-chain genes precede those of the light-chain genes. 19.29In contrast, human leukemic T cells representing various s t q e s of maturation have not shown light-chain rearrangements and have usually (18 of 20 828 THE NEW ENGLAND J O U R N A L OF M E D I C I N E Occ. 6, 1983 Figure 1. Immunoglobulin-Gene Re- arrangements in Chronic Myelogenous Leukemia. Part A shows the human irnmunoglobulin-gene probes used in this study to detect germ-line and rearranged genes. The 1.3-kb germ-line EcoRl probe containing the C, region hybridized to a 17kb BamHl fragment in germ-lineDNAand was capable of detecting rearrangements. Such rearrangements alter the size of this fragment by introducinga new 5' BamHl site when a VH and DH seg- ment combine with a JH region. The 2.z- kb Sau3a JH probe recognizedthe same 17-kb BamHl germ-line fragment and a 16-kb germ-line EcoRl fragment. The 2.5-kb EcoRl C, probe recognized a 12.0-kb germ-line BamHl fragment. We MM~~CC-~~ L.rnbd. G- corn Ecom EcoRl EmPlECofn + -14hbd1+ +78kb+J+ 3 IaW 4-i6kbT II CombifmdC,RDbaM m I EcoRI 1 1 A - I-i-have not observed any polymorphism with the above three probes in over 50 subjects examined. The combined C, pErcoobRel usedconsistedof a 0.8-kb BamHC fragment containing the CAI gene caonndtaain1in.2g-kthbeBCamAH2gI-eEnceo.RIThfirsagpmroebnet case IZC Chronic Granul&c phase *c, E 'CA ~~~~ G S ~1% - ~1s1Lymphoid Blesl Crisis - -'C, *cA P 1st BC P 1st BC caw 1% I--lly~2nd Lymphoid Blast Cibis - -'C, *cA P 2nd BC P 2nd BC .1--- could discern the several polymorphic patterns of C, genes in human beings." Panel 0 shows the results of examinations of Case 2 onthree occasions(aster- ,,Barn - E m RI Barn HI Eco R I - 21- Barn HI Eco RI isks indicate radiolabeled probe). DNA genomic blotsduring the chronic granulocytic phase (Gr, Case 2c) revealedgermline (dashmarks) Cp and C, genes in comparisonto control placental(P) DNA (a source of non-B-cell DNAthat hasgerm-line, non-rearrangedimmunoglobulingenes). Duringthe first lymphoid blast crisis (1st 0C, Case 2a) both heavy-chain genes had rearranged (arrows), as had one C, allele. In the second lymphoid blast crisis (2nd BC. Case 2b) 113-19--II;Case#6 - - -'JH 'C r 'Cl Bc Bc Bc the cells displayed identical heavy-chain-gene rearrangements as in the first crisis but then had germ-line CA genes. c In Case 6, rearrangements of both heavy-chain genes were detected in EcoRl digests with the JH probe. One C, allele had rearranged, and the A genes were germ-line. Eco RI Barn Ill Eco RI examined) had germ-line heaw-chain genes as well.I3 Similarly, in both cases of chronic myelogenous leukemia examined during a chronic granulocytic phase, all six cases of myeloid blast crisis, and the early erythroid-like K562 cell line, germ-line heavy-chain and light-chain immunoglobulin genes were retained (data not shown). These contrasting findings offer strong evidence that the vast majority of the episodes of lymphoid blast crisis in chronic myelogenous leukemia represent monoclonal expansions of cells that are genetically committed to the B-cell lineage. The only potential exception (Case 5 in Table l ) , with germ-line immunoglobulin genes, also had detectable myeloid surface antigens and had B-cell-associated antigens on only a minority of the cells examined. Therefore, the cells from this patient appeared to represent a mixed population, perhaps accounting for the lack of demonstrably rearranged immunoglobulin genes. T h e demonstration of rearranged immunoglobulin genes is of particular importance in establishing the B-cell origin of lymphoid crisis in chronic mvelogenous leukemia, since each of the other markers that were tested (pSO/BA-l, p24/BA-2, HLA-DR, the common acute-lymphoblastic-leukemia antigen, and T d T ) can.be found on cells of lineages other than B cell^.'^^'^,^^ In fact, HLA-DR and p24/BA-2 were detected on some of the cells from myeloid phases of chronic myelogenous leukemia, even though these same cells also displayed myeloid antigens. Of note. the common acute-lymphoblastic-leukemia and HLADR antigens were present during all lymphoid crises. whereas the p30/BA- 1 and p24/BA-2 B-cell-associa: ed antigens were more variable in their presenc (Table 1). Thus, when such markers are present the. are consistent with a B-cell lineage, but their presence alone is insufficient for assigning a B-cell origin to a cell. Of the seven cases with rearranged heaq-chain genes, five (Cases 1,2b, 3, 7, and 8) retained germ-line light-chain genes, indicating that they represented an early B-cell-precursor stage of development. Tu.0 cases had progressed to K-gene recombination (boll1 alleles were deleted in Case 4 and a rearrangemel : had occurred in Case 6; Table 1 and Fig. 1B). Thes cells retained their A genes in the germ-line form; this has been observed in K-producing B - c e l l ~ 'a~nd is consistent with an ordered usage of light chains i n ~ k =d - - - __ - __ - \',>I 309 u o I I CHRONIC 4 1 Y E L O G E N O C S LEC'KEblIa\ - BAKHSHI ET \ L 829 which K generally precedes A. Case la had moved on lo A-gene rearrangement (Table 1 and Fig. IB). We 'i;ive previously noted the loss of germ-line K genes in .-producing B-cells. The status of'germ-line K genes in Case 2a, however, could not be accurately assessed because of the admixture of some normal cells (approximately 25 per cent of the total), which contributed their own germ-line K genes. The precise sene-rearrangement patterns demonstrated during the chronic pranulocytic phase and two lymphoid crises of Case 2 and during the lymphoid crisis o f c a s e ij .ire shown in Figure IB. Although seven of eight cases I $lymphoid blast crises had immunoglobulin-gene re- -rangements. only two had detectable cytoplasmic ,111nunoglobulin(Table I ) . This may reflect the erroriirone nature of immunoglobulin rearrangements, in Lvhich many rearrangements are nonproductive, and is similar to our findings in the B-cell-precursor form of acute lymphoblastic leukemia, in which heavychain genes were always rearranged but were productive of p chain in only 25 per cent of cases.I3 Serial Examinations in a Single Patient In one patient (Case 2) leukemic cells were exam.G.d during an initial lymphoid blast crisis (Case l a j , ; intervening chronic granulocytic phase (Case 2c), .\rid a clearly separate later lymphoid blast crisis (Case 2 h ) (Table 1 and Fig. 1B). T h e myeloid cells from the intervening chronic phase had the expected germ-line configuration of heavy-chain, K ? and h immunoglobulin genes and displayed a 46,XY t(9;22) karyotype. In both episodes of lymphoid blast crisis, however, the cells possessed a new unique cytogenetic marker with :L4 5 , X Y - i karyotype but retained the t(9;22) Phila, ' -1phia chromosomal translocation. Since blast crisis i chronic myelogenous leukemia is thought to reprec n t the evolution of an altered clone of cells, the crisis cells not infrequently display additional chromosomal changes.3' In this particular case a single chromosome i had been lost in the lymphoid crises. I n addition, the cells from the first and second lymphoid hlast crises shared identical rearrangements of both alleles of their heavy-chain genes (Cases 2a and 2b in Figure IB). T h e individual cells that enter the E-cell pathway rearrange their immunoglobulin genes a unique fashion to create a certain antibody specific':..10-i2 Yet, there were identically sized rearrangements of both heavy-chain gene alleles in the two sepJrate clinical blast crises. This finding argues strongly that both these episodes can ultimately be traced to a common lymphoid progenitor cell that had undergone heavy-chain but not light-chain gene rearrangements (Fig. 2). The two lymphoid crises did have a distinguishing immunoglobulin-gene marker, however. The firstlymphoid crisis possessed a light-chain rearrange' x m t that was A in type (Case 2a), whereas the sec.ad lymphoid crisis had totally germ-line light-chain acnes (Case 2b, Fig. IB). This indicates that the immediate precursor cells that gave rise to the separate lvmphoid blast crises in this patient were distinctly "Pluriporent" Clonigenic Cell Myeloid Lineage H-Genes Germline L-Genes Germline I i'I "Blast Crisis" Monoclonal Expansion @45XY-7 @45XY-7 2nd Lymphoid Crisis 1st Lymphoid Crisis Both H-Genes Rearranged Both H-Genes Rearranged L-Genes Germline L-Gene Rearranged Figure2. A DifferentiationScheme in Chronic MyelogenousLeukemia Suggested by Serial Analysis of Case 2. The pluripotent stem cell bearing the Philadelphia chromosome t(9;22) may pursue a chronic myelogenousor acute myeloblastic course, in which it would retain germ-line immunoglobulingenes. When the affected cells pursue 8-cell differentiation they are capable of sequential rearrangementsof their heavy-chain (H) and then their light-chain (L) genes. Separate episodes of lymphoid blast crisis are monoclonal expansionsof affected lymphoid precursor cells, which represent discrete genetic steps of 8-cell maturation. different. I n the first ins-5tance, the precursor cell had advanced to A-light-chain-gene rearrangement. whereas in the second, the source of the blast-crisis expansion was a less mature cell that had retained germ-line h genes (Fig. 2). DISCUSSION The patterns of immunoglobulin-gene rearrangement observed in this study indicate that the cells that comprise most episodes of lymphoid blast crisis in chronic myelogenous leukemia are genetically committed B-cell precursors. Consistent with these findings, Ford et al. have also observed the presence of heavy-chain-gene rearrangements in lymphoid but not myeloid ph tses of this disease.32 T h e determination of immunoglobulin-gene rearrangements in other human cancers has allowed cases of "non-T,non-B" acute lymphocytic leukemia,13 hairy-cell leukemia,j3 and certain lymphomas3' to be assigned to the B-cell lineage. The genetic information we now report is particularly useful in classifying the lymphoid blast crises of chronic myelogenous leukemia, because of their morphologic variation and because many of the phenotypic markers expressed, such as T d T , Ia antigen, and the common acute-lymphoblastic-leukemia antigen, are known to be present on multiple different cellular lineage^.'^^'^^^^ T h e cells comprising these cri- ses are similar in genotype to the B-cell-precursor type 830 THE NEW ENGLAND JOURNAL OF MEDICINE Oct. 6, 1983 Vol. 309 in acute lymphocytic 1 e ~ k e m i a . IT~he fact that they represent similar stages of differentiation may account for the known responsiveness of some lymphoid crises in chronic myelogenous leukemia to chemotherapeutic regimens that are similar to those successfully usedin acute lymphocytic leukemia. Our findings show no evidence for involvement of the T-cell lineage within the lymphoid blast crises examined here. However, a rare T-cell-type lymphoid blast crisis has been reported.35 We have demonstrated that the clonally abnormal cell of chronic myelogenous leukemia can advance along the B-cell lineage and is capable of undergoing the normal genetic process of DNA rearrangement of the immunoglobulin loci (Fig. 2). Even though this may include both heavy-chain and light-chain-gene rearrangements, the cells in these lymphoid blast crises lack surface immunoglobulin and thus appear to be a population of B-cell precursors that have not advanced to more mature stages of development. The mechanism that arrests these malignant cells at a certain point of B-cell differentiation is unclear. Considerable attention has focused on the normal cellular oncogenes as elements that may effect the uncoupling of growth and differentiation capacities. Klein has suggested that chromosomal translocations may alter the expression of a cellular transforming gene by placing it into another locus.36 For example cAbl (the Abelson murine leukemia virus cellular oncogene analogue), which is normally on chromosome 9, has been translocated to chromosome 22 in some cases of chronic myelogenous l e ~ k e m i a . ~T'he A immunoglobulin genes are located at the very band, 22ql1, involved in the Philadelphia translocation (Kirsch I: unpublished data). This area is also the breakpoint ofsome variant translocations found in Burkitt's lymphoma, suggesting that it is a vulnerable site for interchromosomal re corn bin at ion^.^^ However, the lack of rearrange- ment of the A-gene locus in all the myeloid episodes of chronic myelogenous leukemia and in all but one of the lymphoid episodes suggests that this translocation does not occur directly within this portion of the A immunoglobulin-gene locus. Our serial examination of one patient demonstrated that the monoclonal expansions recognized clinically as separate lymphoid blast crises arose from precursor cells at slightly different stages of genetic maturation. Thus, even though all affected lymphoid cells in chronic myelogenous leukemia may ultimately be the progeny of a common lymphoid progenitor cell, blastcrisis expansions appear to represent unique subclones at different stages of differentiation. Consistent with this, Martin et aL3' have shown that both K-producing and A-producing B cells bearing a Philadelphia chromosomal marker can be observed after the transformation of lymphocytes in chronic myelogenous leukemia with the Epstein-Barr virus. This indicates that clonally affected B-cell precursors in chronic myelogenous leukemia can be induced to dif- ferentiate to mature B-cell stages of development. The immunoglobulin-gene analysis reported in the present study has shown that the episodes of lymphoid blast crisis in chronic myelogenous leukemia involve clonal proliferations of cells at an earlier, B-cell-precursor stage of differentiation. REFERENCEs -I. K a f a a HP. Golde DW. Chronic myelogenousleukemia new conceps. N En@ J Med 1981; 304:1201-9. 1269-74. 2. Nowell PC. Hungerford DA. Chromosome studies on normal and leukemic human leuk-. JNCI 1960, 25:85-109. 3. Walkow PJ. lacubson RI. Papaymopoulou T. Chronic myelocytic leuke- mia: clonal origin in a stem cell common to the granulocyte erythrocyte. platelet and monocytelmacrophage. Am J Med 1977;63:125-30. 4. Beurla E, Yeh M. Fairbanks VF. The normal human female as a msaic of X-chromosomc activity: studies using the gene for G-6-PD-deficiencyas a marker. Roc Natl Acad Sci USA 1962;48:9-16. 5 . Barr RD. Adkow PI. Clonal origin ofchmnic myelocytic leukemia. N Engl J Med 1973;289:307-9. 6. Fialkow PJ. Dcman AM, Jacobson RJ. Lowenthal MN. Chronic myelocyt- ic leukemia:origin of some lymphocytes from leukemic stem cells. I Clin lnvesf 1978;62:815-23. 7. LeBien TW.Hozier I, Minowada 1.KerseyJH. Originof chronic myelocyt- ic leukemia in a precursor of pre-B lymphocytes. N Engl J Med 1979 301:1467. 8. Minowada J. Tsubota T,Greaves MF, Walters TR.A non-T.non-B human leukemia cell lm (NALM-I): establishment of the cell line and presence of leukemia-associated antigens. JNCI 1977;5983-7. 9. Vogler LB. Crist WM, Vinson PC.Sanif A, BraKain MG. Coleman MS. Philadclphia-chromosome~~itivpere-B-cell leukemia presenting as blast crisis of chronic myelogenous leukemia. Blood 1979; 541 164-70. 10. Seidman JG. hiax EE. Leder P. A K-immunoglobulin gene is formed by site-spcific recombination without funher somatic mutation. Nature 1979; 28Q370-5. 11. Brack 8 , Hirama R. Lenhard-Schuller R. Tonegawa S . A complete i m munoglobulin gene is created by somatic recombination. Cell 1978; 15:l- 14. 12. Early P. Huaog H. Davis M. Calame K. Hood L. An immunoglobulin heavy chain variable region gene is generated from three segments of DNA VH.D and JH. Cell 1980;19:981-92. 13. Korsmeyer SJ. Arnold A, Bakhshi A, et al. Immunoglobulin gene re- arrangement and cell surface antigen expression in acute lymphocytic leu- kemias o f f cell and B cell precursor origins. I Clin Invest 1983;71:301- -13. 14. Anderson LC. Nilsson K. Gahmbcrg GG. K562 a human erythroleuke- mic cell line. Int J Cancer 1979 23:143-7. 15. MohankumarT. Pauly JL.Sokal JE. MeKgar RS.Human Ieukemia-associ- atcd antigens: detection on cells of established lymphoblastoid lines. J Im- munol 1975; 1151542-8. 16. Sarin PS.Gallo RC. Terminal deoxynucleotidyluansfeme in chronic mye- logenous leukemia. J Biol Chem 1974: 2498051-3. 17. Southern E. Gel electrophoresis of restriction fragments. Methods Enzymol 1979; 68:352-76. 18. Ravetch N. Siebcnlist U,Korsmeyer SI WaldmannT, Lcder P. Smcturr of Ihe human immunoglobulin p locus: characterization of embryonic and rearranged J and D genes. Cell 1981;27583-91. 19. Hietet PA, Konmeyer SI, Waldmann TA, M e r P. Human immunoglobu lin K light-chain genes arc deleted of m g e d in A-producing B cells Nature 1981;290:368-72. 20. Ria 1. Pesarido JM, N a i s - M c C o n q I. Lazarus H. Schlossman SF. A monoclonal antibody to human acute lymphoblastic leukaemia antigen. Na- ture 1980; 283:583-5. 21. Brodsky FM.F'arham P. Barnstable CJ,Cmmpton MJ.Bodmer WF. Mono- clonal antibodies for analysis of the HLA system. Immunol Rev 1979;473- 61. 22. Abnmson CS, Kemy JH. LcBien TW.A monoclonal antibody (BA-I) Racdve with cells of human B lymphocyte lineage. J lmmunol 1981: 12683-8. 23. Kency JH, LeBien TW,Abramson CS. Newman R. Sutherland R, Greaves M. p24: a human leukemia-associated and lymphohcmopictic progenitor cell surfacestructurr identified with monoclonal antibody. J Exp Med 198!. 153:72631. 24. Reinhen EL.Kung PC, Goldstein G. Levey RH.Schlossman SF. Discre. stages of human innthymic differentiation: analysis of normal thymocytes and leukemic lymphoblasts of T-cell lineage. Roc Narl Acad Sei USA 1980; 77~1588-92. 3. Tsub of ce 5984 26. BE= clona muno n. Tam in hui Res I 28. Mino kcmi 100. 29. Korsi P. D hum2 7100 33. COSS neuu J Ex1 31. ROW chow 32. Ford Abstra opsy fo tion-en1 PrWa ibility o pies w: five pr cases, termin: AS gel been b SI:; about and K on tec c),pa y.30 using tion h; tcchni in Fra risk fc dctic obrair: car hull -=iuilt +h-b, c.91. Hf hppo (=Eli. vel. 309 NO. 14 -P R E N A T A L D I A G N O S I S OF S I C K L E - C E L L A N E M I A GOOSSENS ET AL. 83 1 5. Tsubota T. Minowada J. Nakuawa S , et al. Correlation of surface markers of cells of h u m lymphanc leukemias with disease type. JNCI 1977: 59845-50. -6 B r e d J, Reinhen EL. Kung PC, Goldstein G. Schlossman SF. A mono- clonal antibody reactive with human peripheral b l d monocytes. J Im- munol 1980; 124194343. 27. Tatsufi E, Sagawa K. M i m J, et al. Immunologic membrane phenotypes in human myeloid leukemia by monoclonal antibodies. Proc Am Soc Cancer Res Am Soc Clin Oncol 1981; 22:181. absuact. 2s. Minowada J, Koshiba H,Sagawa K, et al. Marker profiles of human leukemia and Lymphoma cell lines. J Cancer Res Clin Oncol 1981; 101:91- 100. 29. Konmeyer SJ. Hieter PA, Ravetch JV, Poplack DG. Waldmann TA, Leder P. Developmental hierarchy of immunoglobulin gene remgements in human leukemic pre-Esells. Roc Natl Acad Sci USA 1981; 78:7096- 7100. 30. Cossman J, Necken LM. Leonard WJ. Greene WC. Polymorphonuclear neuaophils express the common acute lymphoblastic leukemia antigen. I Exp Med 1983; I57:1064-9. 31. Xowley JD. Chromosome abnormalities in the acute phase of CML. Virihows Arch [Cell Pathol] 1978: 2957-63. 32. Ford AM. Molgaard HV. Greaves MF. Could HJ. Immunoglobulin gene organization and expression in hematopoietic stem cell leukemia. Embo J 1983; 2:997-1001. 33. Konmeyer SJ, G e n e WC, Cossman J, et al. Reaxrangemerit and expres- sion of immunoglobulin genes and expression of Tac antigen in hairy cell leukemia. Proc Nail Acad Sci USA 1983; 80:4522-6. 34. Arnold A. Cossman J. Jaff.. E. Waldmann T. Korsmeyer SJ. Immunoglobulin gene reamngement as a marker for monoclonality in human lymphoid neoplasms. Clin Rcs 1983: 31:402A. abstract. 35. Griffin JD, Tanuavahi R. Cmnellos GP.et al. T cell surface antigens in a patient with blast crisis of chronic myelogenous leukemia. Blood 1983; 61:6K)-4. 36. Klein G. The role of gene dosage and genetic transpositions in cminogenesis. Nature 1981; 294:313-8. 37. de Klein A, Geum van Kessel A. Grosveld A. et al. A cellularoncogcne is translocated to the Philadelphia chromosome in chronic myelocytic leukaema. Nature 1982: 300.765-7. 38. Lenoir GM, Preud'homme JL. Bernheim A. Berger R. Correlation between immunoglobulin light chain expression and variant nanslocation in Burkitt's lymphoma. Nature 1982: 298:474-6. 39. Martin PJ. Najfeld V. Hansen IA, Penfold GK. Jacobson RJ. Fialkow PJ. Involvement of the B-lymphoid system in chronic myelogenous leukaemia. Nature 1980; ?87:49-50. PRENATAL DIAGNOSIS OF SICKLE-CELL ANEAMIAI N THE FIRST TRIMESTER OF PREGNANCY 3 t bfrcHEL GOOSSENSM, .D., YVESDUMEZM, .D., LILIANAKAPLANM, .D., MIEKELUPKER,B.S., 3C WUDE CHABRET, B.S., ROGER HENRION, M.D., AND JEAN ROSA, M.D. ' Abstract To investigate the usefulness of chorionic bi- ly. In one case in which fetal death was observed three opsy for prenatal diagnosis of sickle-cellanemia by restric- weeks after sampling, placental abnormalities found on tion-endonuclease analysis of fetal DNA, we studied 30 pregnancies before elective abortion. When the reproduc- ibility of the technique for obtaining adequate D N A sam- histologic examination were compatible with a chromoso- mal aberration. t Our study shows that chorionic biop,sy is feasible for the ples was established, we successfully applied the test to prenatal diagnosis of sickle-cell disease before the 10th five pregnancies at risk for sickle-cell anemia. In two gestational week. If subsequent expeGence demonstrates cases, sickle-cell disease of the fetus led to a decision to this technique to be safe enough for $other and fetus, the terminate the pregnancy. In three other cases, a normai or ability to test in early pregnancy rnaypake prenatal diag- AS genotype was demonstrated. One normal infant has nosis acceptable to more couples at r@kfor serious genet- been born, and one other pregnancy is continuing norrnal- ic disorders. (N Engl J Med 1983; 309:831-3.) d ymol 4,' .A-I) "^ I ! S N C E restriction-enzyme techniques for prenatal diagnosis of sickle-cell anemia by amniocentesis at about 17 weeks ofgestation were developed by Chang and Kan' and by Orkin et al.,' emphasis has focused on techniques to permit diagnosis earlier in pregnancy, particularly with methods using trophoblast biopS Y . ~One report of three instances ofprenatal diagnosis using DNA from trophoblast tissue obtained by SUCtion has recently appeared.' We have been developing techniques for the diaposis of fetal sickle-cell disease in France. We report here our results in five fetuses at risk for sickle-cell anemia, for which we used the restriction enzyme Con1 to analyze trophoblast DNA obtained at six to eight weeks ofpregnancy. The biopSy can be performed during outpatient consultation Fmm the Service de Biochimie and INSERM,U.91.Cenm Hospitalier UniverJit& H c f i Mondor. 94010 Cretd. France, and the ~ h k p Uc~VerSiLairC Pom-Royal and INSERM.U.166,Maternit6 de Port-Royal,75014 Paris. France. Address reprint rcquesu to Dr.Goossens. at the Service de Biochimic, INSERM 91. Hapital Henri Mondor. 94010 Creteil, France. without anesthesia, and sufficient&uantities of DNA are routinely obtained to permi the sensitive assay recently des Trophoblast Biopsy As a prelim~na,,, tajned (after informed frc 9 . .gestation, before elective abortion. T h e . iopsies were performed using a Dyonics biopsy forceps '' (1.7. m. m- n diameter), introduEcd ~s~~~~~ an area adjacent , L L c ',,L~"P,., mens were taken from the placcn t m u e obtained were c r a r n i ~ e dunder a$sscc&g microscope, and chorionic w e r c ~ e p a r a t e dfron :'-d c ~ c n b e d . ~ Once the reproducibility of the t&iqu?yyas' established,c;he of using it wPe FlrcFIlllv rl;&errt Gfh'm,dr. It Gsk fnrf h a ~ n ga sicl was notknc tal . test, . which wa.s p.m i ~ I ,:.:I de la Recherche MCdicale. had already terminated a cause )hefetus had the dis- I' 2- -- ;