Document BvXqOx270b4yeMm6LgeBjw8Y4

AMERICAN- JOURNAL OF Oficial Publication of American Society of Clinical Pathologists CLINICAL PATHOLOGY M. Desmond Burke Associate Editor Clinical Pathology Lynn Olson Managing Editor Mark R. Wick Editor-in-Chief Paul E. Swanson Assistant Editor Department of Pathology Barnes Hospital Washington University Medical Center St. Louis, Missouri 63 1 10 Stacey E. Mills Associate Editor Anatomic Pathology Tonya Brookshire Editorial Assistant BOARD OF EDITORS David N. Bailey John C. Baenziger Peter M. Banks Asa Barnes, Jr. Hector Battifora Debra A. Bell Rodger L. Bick Thomas A. Bonfiglio John T. Brandt Richard D. Brunning Peter Burger Darryl Carter Paul G. Catrou Wing C. Chan P. Joanne Cornbleet John R. Craig Michael J. Deegan Louis P. Dehner C. Terrence Dolan J. Roger Edson Diane C. Farhi Robert E. Fechner Henry Frierson, Jr. Glauco Fnzzera John R. Goellner Curtis Hanson Nancy Hams Charles L. Hitchcock George C. Hoffman Charles A. Horwitz Joan H. Howanitz Andrew G. Huvos Brooks Jackson Peter I. Jatlow J. Charles Jennette William W. Johnston Michael T. Kelly John A. Koepke Chin-Yang Li James Linder Virginia A. Livolsi Fred V. Lucas, Jr. Bruce Mackay J. Carlos Manivel Geoffrey Mendelsohn Paul D. Mintz S . Breanndan Moore Mehrdad Nadji William P. Newman Harold A. Oberman Nelson G. Ord6fiez Lance R. Peterson LoAnn C. Peterson Herbert F. Polesky Robert Robinson Jeffrey Saffitz Cosimo G. Sciotto Gene P. Siegal Dale C . Snover Mark Stoler Calvin L. Strand Jerome B. Taxy Barbara Tindle Douglas A. Triplett John E. True11 Patrick C. J. Ward John C . Watts Louis H. Weiland - Mark Weiss Sharon W. Weiss James W. Winkelman Washington C. Winn, Jr. Barbara C. Wolf Margaret M. Yungbluth Richard J. Zarbo Published for the Society by J. B. Lippinwtt Company Philadelphia The AMERICAN JOURNAL OF CLINICAL PATHOLOGY (ISSN 0002-9173) is published monthly for the American Society of Clinical Pathologists by J. B. Lippincon Company, P.O. Box 1550, Hagerstown, MD 21740. Business officesare located at 227 East Washington !Square, Philadelphia, PA 19106. Printed in the U.S.A. 0 Copyright 1993 by the American Society of Clinical Pathologists. Second class postage paid at Hagerstown, MD, and at additional mailing offices. Subscription information, orders or changes of address: ASCP members contact AJCP Circulation Dept., 2 100 W. Hamson Street, Chicago, Ill. 60612. Nonmembers contact (except in Japan, India, Nepal, Bangladesh, or Sri Lanka) AJCP, Lippincott Subscriber Services, I2 107 Insurance Way, Hagerstown, MD 2 1740, or call 1-800-638-3030:in Maryland, call collect 301-824-7300. In Japan, contact USACO Corporation, 13-12, Shimbashi I-chome, Minato-ku, Tokyo 105, Japan. In India, Nepal, Bangladesh, and Sri Lanka contact Universal Subscription Agency Pvt. Ltd., 101-102 Community Centre (F.F.)Saket, New Delhi-I 10017, India. Annual subscription rates: US. $135.00 individual, $210.00 institution; all other countries except Japan, India, Nepal, Bangladesh, and Sri Lanka $220.00 individual, $295.00 institution. ResidenWStudents (US. only) $50.00.Canada $ 195.00individual; $270 institution. Single copies$23.00. Rates for airmail delivery available upon request. Subscription rates in Japan: 49,200 yen individual, 64,200 yen institution (includes airmail postage). Copies will be replaced without charge if the publisher receives a request within 60 days of the mailing date in the U.S. or within 5 months in all other countries. Authorization to photocopy items for internal or personal use,or the internal or personal useof specificclients,is granted by the American Society of Clinical Pathologists for libraries and other users registered with the Copyright Clearance Center (CCC) provided that the base fee of $3.00 per copy is paid directly to: CCC, 21 CongressSt., Salem, MA 01970. ISSN 0002-9173/1993/$3.00. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes. for creating new collective works, or for resale. Advertising inquiries should be sent to Advertising Sales Representative, AJCP, J. B. Lippincott Company, Fast Washington Square, Philadelphia, PA 19105, (215) 2384283. Postmaster: Send address changes to AJCP, P.O. Box 1550, Hagerstown, MD 21740. Volume 99, Number 4 April 1993 AMERICAN JOURNAL OF CLINICAL PATHOLOGY Contents Special Issue A TRIBUTE TO HENRY RAPPAPORT Introduction JAMES W. VARDlMAN AND GLAUCO FRIZZERA Foreword ROYALD F. DORFMAN 359 360 HEMATOPATHOLOGY Rwiew Articles - Immunophenotyping of Lymphoma and Leukemias in Paraffin-Embedded Tissues SHERRIE L. PERKINS AND CARL R. KJELDSBERG 362 Flow-CytometricDNA Analysis in the Diagnosis and Prognosis of Lymphoma RPIUL C. BRAYLAN 374 The Uses of Molecular Biology in Hematopathology: Oncogenes as Diagnostic Markers in Non-Hodgkin's 38 1 Malignant Lymphoma HOWARD RATECH Reactive Lymphadenopathies and Atypical Lymphoproliferative Disorders JAYASHREE KRISHNAN, ARDlS D. DANON, AND GLAUCO FRIZZERA 385 The Lymphadenopathy of HIV Infection CARLO D. BARON1 AND STEFANIA UCClNl 397 Malignant Disorders of Small Lymphoc>.tes:Small Lymphocytic Lymphoma, Lymphoplasmacytic Lymphoma. and Chronic Lymphocytic Leukemia. Their Clinical and Laboratory Relationship GERASSIMOS 4. PANGALIS. VIKl A. BOUSSIOTIS. AND CHRISTOS KITTAS 402 Lymphomas of Follicles: Mantle Cell and Follicle Center Cell Lymphomas DENNIS D. WElSENBURGER AND WING C. CHAN 409 Monocytoid B-Cell Lymphoma SUNG SlK SHIN AND KHALIL SHEIBANI 42 1 Peripheral T-cell Lymphoma: Morphologic and Immunologic Observations CARL D. WINBERG 426 Unresolved lssues Concerning Hodgkin's Disease and Its Relationship to Non-Hodgkin's Lymphoma DAlNA VARlAKOJlS AND JOHN ANASTASI 436 Composite Lymphoma and Related Disorders HUN KIM 445 Mycosis Fungoides: Diagnosis and Pathogenesis MAURICE BARCOS 452 Reactive Histiocytic Disorders BRUCE A. WODA AND JOHN L. SULLIVAN 459 Langerhans' Cell Histiocytosis and Malignancies of the M-PIRE System JONATHAN M. BEN-EZRA AND CHAE H. KOO 464 Extranodal Lymphoid Disorders KEVIN E. SALHANY AND GIUSEPPE G. PIETRA 472 Splenic Lymphoid Hyperplasias Versus Lymphomas/Leukemias: A Diagnostic Guide JEROME S. BURKE 486 Chronic Lymphoproliferative Disorders Involving Blood and Bone Marrow DANIEL A. ARBER, JEAN R. LOPATEGUI, AND RUSSELL K. BRYNES 494 Immunophenotypic Analysis of Acute Leukemia S. THOMAS TRAWEEK 504 Issues in the Pathology and Diagnosis of the Chronic Myeloproliferative Disorders and the Myelodysplastic 5 13 Syndromes JEROME I. DICKSTEIN AND JAMES W. VARDIMAN Immunophenotypic Analysis of Acute Leukemia S. THOMAS TRAM'EEK,MD Since the introduction of monoclonal antibodies for the study of hematopoietic disease in the early 1980s. tech- niques to aid i n the identification of cell lineage haye proliferated. Immunofluoresceiit flow-cytonietric stud- ies are particularly helpful in tlie characterization of acute leukemia, and when used in conjunction \4,itli light-microscopic findings, can niake a significant con- tribution to tlie final diagnosis i n 20-30% of cases.' Tlie introduction of immunoplicnot\rpic studies in tlie char- acterization of acute leukemia is not without cost. how- ever. because tlie dcgree of hcterogencity in these dis- eases has dramatically increased. In particular. tlie existence of multilineage antigen expression i n acute leu- kcmia is now wcll-acccpted, and when present, ma\'con- * siderably frustrate the classification process. Despite this marked heterogeneity. accurate diagnosis of acute leuke- mia is extremely important. because therapy and prog- nosis depend primarily on propcr charactei-ization. The most widely accepted and applied acutc leukeniia classifi- cation scheme, developed by thc French-American-Bri- tish (FAB) Cooperative Group in 1976. initially used only morphologic and cytochcmical findings.2 Since thcn, the diagnostic criteria has been re\ ised and ex- pandcd to include iiiimunoplienotypic and ultrastruc- tural studies for certain acute Tlie in- creased use of inmunophenotypic analyses in acute leukemia also prccipitated a need for classification of Froiii ~ I i cJotiiev Irviiic Ccww /iv / / I C S / u l / v o / ' , ! ! w k c i i i i u uiid l-j.111p l i o i m . Diitsioii cfPui/io/ogy. C`i/j. (!fIlopc "?uiimu/ . \ f ( ~ / i uC/ ctir(*r. DllNrlf. C~d!/ilrlliu Keccivcd Oclobcr 26. 1992: acccpted for publication Ociobcr 30. 1992. Dr. Trawcck is a nienihcr of thc City or Hope Canccr Kcsearch Cenlcr. which is supportcd hy Public Hcalth Scrvicc Grant CA-33571. and is also thc rccipient oran Anicrican Canccr Socicty Clinical Oncology Career Development Award. Addrcss corrcspondcnce and rcprini rcqucst to Dr. Trawcck: Dcpanmcnt of Anatomic Pathology. City of Hope Naiional hlcdical Ccnicr. 1500 Easl D u a ~ i cRoad. Duarlr. CA 9 I O IO. niixcd lineage leukemia. and se\feralproposals have been but none havc been universally accepted. I n an atteinpt to further define the range of heniatopoietic cell antigen reactivity i n acute leukeniia. this study was undertaken to establish immunoplienot\~picprofiles on samples referred to The City of Hope National Medical Ccnter. Thcsc cases u ~ r scpecifically submitted for characterization and diagnosis of acute leukemia, and were all studied in a relatively uniform manner in one department. The findings on this referral-based population ivill be compared \villi a number of pre\.ious immunophenotypic studies of acute leukemia. METHODS AND MATERIALS S ~ i ~ d j y o iT~lipc.findings on pcriplieral blood or bone n iarrow samples submittrd to tlie James 1n.ine Ccnter for the Study of Leuhcmia and Lymphoma at Tlie City of Hope National Medical Ccnter bebyeen December I . 1987 and August I . 1992 were rc\kwed. The accrual start date was determined b y the initiation of flo\v-c\ tometric studies in the analysis of acute leukemia at this institution. The patient population was primarily adult. Inclusion criteria for the study were: 1. That a n unequivocal diagnosis of acute Ieuhemia was made. based on examination of a peripheral blood or bone marrow specimen. I n most cases. tlie acute leukcmia \vas assigned an F A B category. based on the \Yi-igIit-Giemsa-staincd morphnlogic appcarancc of the blasts and on tlie rcsults ofcytocheniical stains for sudan black-B. niyeloperoxidase, nonspecific esterase (either a-naphthyl acetate esterase performed \villi and without fluoride inhibition or butyrate esterase). acid phosphatase. and periodic acid-Schiff. \vliich were pcrformed according to standard methods. A n accuratc FAB designation could not be reliably deduced in a few cascs, however, due to either poor tcchnical quality oftlie submitted slides. lack o f a sufficicnt sample for a full q.tochemica1pancl, or avail- 504 TIiA W EEL 505 TABLE I. MONOCLONAL ANTIBODIES Littcgrp Associrifiorr T-I!;m plioid T-lymphoid T hclpcr/induccr T-I! niphoid 1. supprcswr T-lymphoid B-Iy niphoid Myloid/rnonocytic Monoc>-tic Mycloid/nionocytic B-lyniplioid U-lymphoid Mycloid/monocytic SlCill ccll Lcu k ocytc coni nion 3n tigcn I-ILA-DR -N i l l fICIS0urcc Leu S/BD Lcu 3/HD Lcu-~/BD LCU- I /BD L~IJ-Z/BD Lcu-9iBD JS/Coulicr My7/Coultcr My4/Coul1cr LCU-MI/BD Brt/Coultcr B I /Coultcr h4\9/Coultcr IiI'CA- I/BD Lctikocylc/ BD 12/Coultcr ability only of cytocentrifuge preparations for morphologic evaluation. 2. That iminunophenotypic studies were performed using flow cytonietry on either peripheral blood or bone marrow containing at least 30%blast cells. If the phenotypic results of bone marrow were not cliaracleristic of the disease as classified by FAB, ut least 3070 blasts were also required in the peripheral blood to cliniinatc the possibility of spurious phenotype results secondary to hemodilution. 3. That enough viable blast cclls were rcco\ ered to perform an immunophenotypic pancl that included at least two lineageassociated antigens each from the B-lymphoid, T-lymphoid, and myeloid cell lines. Cell prepuratiurt arid i~ii~iiiiiiuplic~i~a~ii~u~l~~~siics. After mononuclear cell enrichment by centrifugation over Ficoll-Paque (Pharmacia, Piscataway, NJ), the pcriplieral blood and bone marrow samples were studied for surface antigen expression with direct and indirect ininiunofluorescent techniques using a variety of commercially a\r&ible monoclonal antibodies. (Table 1) In most cascs. greater than 50%blast cells were present after mononuclear-cell enrichment, as determined by a Wright-Giemsa-staincd cytocciitr-i fugc prepara ti on. The antibody panel used to evaluate specimens submitted for acute leukemia characterization varied only slightly over the 4.5-year study period. Specifically, CD45 and CD34 antibodies were added, whereas CD3 and CD5 were dropped during this time. In addition, a limitcd nunibcr of cells i n some cases resulted in inconiplete antibody pancls. Analysis of the stained cells was performcd initially with an Epics 5 flow cytometer (Coulter, Hialeah, FL), and later with a FACScan (Becton Dickinson, Mountain View, CA). Only cells within a niononuclear ccll gate, individually created for each acute IcuLcmia. w r c considcrcd i n thc final tally. Matched conccntrations of immunoglobulin isot\pe controls \iJcrcr u n concurrently with all cascs. Terminal dcosynuclcotid\ I transrerase (Tdt ) dctcrniinations werc not consistent 1y done, and are reported here only for sclccted cascs. Ciircvirr,fin.mi/igm c~.qsllr.e.uioitP.ositivity for an i nd ividual antigen was defined as more than 20T0 fluorescent cells above the isotype control for all antibodies except CD34. CD34 was considered to be positivc if there were 10%of fluorescent cells above the isotope control. h4ultilineage antigen expression was considered to be present if: ( I ) the sum of the most reliable lineage marker for the t!pe of acute leukemia and the inappropriately ex-. prcsscd antigen was greater than 12070.or (2) coexprcssion of multilineage antigens was demonstrated in more than 20% of the gated cell population by dual-color imin u no fl u orcscem e . Clussificariori qfucirtc. kcrikcruiu. The final classification of each acute leukemia was based on an integrated assessment of the morphologic, cytochemical. and immunophenotypic data. The phenotypic findings were chiefly confirmatory in the classification of AML-M 1 through AML-M6, as the diagnosis was based priinarily on morphologic and cytocliemical fi11di ngs. I inmu n ophenotyping was instrumental however. i n thc final diagnosis and classification of acute lymphoblastic leukcinia, minimal Iy diKerentiated acute niyclogc nous leukcniia (AML-MO), acute undifferentiated leuhernia. and mixed lineage leukemia, because morphologic and cytochemical studies provide little information about the lineage of thcse diseases. After classification, imniunophenotypic profiles were generated for each subtype of acute leukemia. RESULTS Acute Myelogetious Leukenria Two hundred sevcii cases of AML werc studied; the results are given in Table 2. Almost all cascs (99%) es- prcsscd CD45.The I-lLA-DR antigcn was present i n the majority of AML, cxcluding AML-M3. CD33 was thc myeloid lineage antigen most commonly present, followed in frequency by CD 13 and CD 15. The monocyteassociated antigen CD14 was detected in about one half of thc AMLs with a monocytic coniponcnt (as detccted by nonspecific esterase activity), but was also occasionally prcscnt in subgroups of AML with no rnonoc\tic features. CD7 expression was primarily sccn i n AMLMO; AML-M1; and AML-MSa. the least iiiaturc AML: but also was detected sporadically i n other types of Vol. 99. No. 4 506 I IEMA T 0 PATI i 0LOG Y TABLE 2. ACUTE MYELOGENOUS LEUKEMIA PHENOTYPES. Pl1rtlor)~pc Leukenria T y p (No. of Pafirrifs)t CLM5 CD2 CD3 CD5 CD7 CD19 CD20 CD13 CD14 CD15 CD33 CP10 /!LA-Dr' CP3.1 AML-M I (n = 40) AM L-M2 (n = 69) AML-M3 (n = 19) A h4L-M4 (n = 39) AML-M5 ( n = 18) AML-M6 (n = 3 ) AML-M7 (n = 2) AML-MO (n = 9 ) AML-NOS (n = 8) AML Total 18/19 0/40 0/18 (95%) 22/22 0/68 0149 ( loo%) I1/1I 6/19 018 ( 100%) (31%) 14/14 01.79 0125 ( loo%) 515 0/18 0113 0115 0143 0110 0125 0113 111 013 012 012 111 012 313 0i8 011 0,11 017 015 313 018 013 012 9/37 (241) 1/62 (11%) 0/19 1/40 (3%) 4/67 (68) 0119 0139 1/68 (1%) 0119 3/38 0139 (R1) 8/18 0117 (44%) 113 013 0/38 0/18 0/3 112 012 012 317 019 019 014 0/7 018 29/38 (76%) 43/66 (67';) 16/18 (891,) 16/35 (46%) 12/17 (71%) 213 I /z 0127 4/67 ( 6 %1 1/19 (5%) 11/38 (457) 8/18 (44%) 013 14/26 (54'7 ) I XI.36 (SOC,) 5/12 (42%) 15/20 (75%) 518 ( 6 3 5) 111 012 0/ 1 3.?/40 1/38 (8.3';) (3'7) 54/69 1/61 (785) (25) 16119 o/ I9 (837) 36/39 2/39 (92T) ( 5 % ) 15/18 1/18 (835) (6%) 013 013 I/:! 012 23/27 (85%) 51/67 (815) 1/18 (6%) 33/38 (87'7) 15/18 (83'7) '1.7 01 I 719 119 216 519 019 919 617 0/8 316 718 016 717 78/79 6/205 01126 0/l I6 321190 5/203 11204 1331195 351191 6311 I6 167/207 51198 1431169 (99%') ( 3 %)$ (17%) (170) ( < I % ) (68%) ( 18%) (54%) (80%) ( 3 % ) (85%,$ NOS * not olhcrwir spcci6cd: A M L : acute m y l q c n o u s Icukcmia. Results arc crprcsscd Tor cach antigcn as numher ofposilivc cascs/numhcr o f c a r s stud- -ied. t Total AML studied 207. Threc CBSCS. all MI. failrd tocaprcss mycloid lincagc antigens: thisrquals 3/69 (4'E.)oTM?or 3/207 i l . 5 ' i ) d a l l .AML siudlcd t CD? prcxnt only in M3. 3 Exchrdcs h13 from m a l . AML. Other than CD7, single 1)irnplioid-associatedantigen positivity was present in 3% or less of the AMLs, and was restricted to CD2, CDIO, CD19, andCD20. The proportion of CD34-positive cases was 67% for all AML, and there was no significant difference i n expression of this antigen between the AML subtypes. .4A4L-A4l. All 40 cases of AML-MI had detectable myeloperoxidase activity (as required by FAB criteria), and all expressed at least one myeloid lineage antigen. This subtype conipriscd 19% of AMLs and 10% of all acute leukemias. The single AML in this study that was not positive for CD45 was from the AML-M 1 subgroup. One case was positive for the B-cell antigen CD 19, and also contained a t(8:2 1 ) chromosomal translocation. One case expressed CDIO, with no other lymphoid antigens. AAfL-Af2. Sixty-nine cases. representing 33% of AMLs and 1870 of acute leukemias wcrc studicd. This subgroup had the highest rate of aberrant B-cell antigen expression among the AMLs; either CD 19 or CD20 was expressed in 5 cases (7% of AML-M2). In no case was there more than one lymphoid antigen expressed. Cytogenetic studies were performed on three of the four CD 19-positiveAML-M2; all contained a t(8;2 1)chromosomal translocation. Significantly, three cases of clearcut AML-M2 by morphologic and cytochemical evaluation wcre devoid of the myeloid antigens CDI 3, CD14, or CD33. CD 15, howevcr, was not examined on these cases. One of these phenotypically null AML-M? e\ en contained a t(R2 1) translocation, as demonstrated by cytogenetic examination. AAIL-A43.Nineteen cases of acute promyeloc~ticleukemia were studied, representing 9%of AMLs and about 5% of all acute leukemias. The T-ccll antigen CD2 \vas expressed in nearly one third of the cases. and AblL-h13 was thc only subtype of AML to be positile for this antigen. The results of cytogcnctic studies wcre a\~ailablein nine cases; all displayed the pathognomonic t( 15: 17). i n cluding all six CD2-positive cases ofAML-M3. No cascs were CD7- or C D IO-positive. but one sample expressed CD14 with no evidence of nonspecific esterase acti\ it!. I n marked contrast to the othcr types of AMI-. only one case ofAML-M3 was positive for HLA-DR. The degree of positivity for this antigen was not great. howevcr. only 26%, and contributions from nonblasts cannot be cntirely excluded. The stern cell antigen CD33 \\as detected i n I O of 16 cases (63%).a level ofpositivity similar to the other AML subtypcs. A.44L-Al4 a n d A f L - A 4 5 .The 39 cases of acute myelomonocytic leukemia (19% of AMLs and 10% of acute leukemias) and 18 cases of acute monocytic leuhcmia (9% of AMLs and 5% of acute leukemias) were similar immunophet~otypically. Almost onc half of the cascs from each subtype cxprcsscd CD14, a much higher pcrcentage than was dctected in any other group of AMLs. The proportion of cases that were positivc for CD7 was .4.J.C.P.* April IY93 TR A W E E I< TABLE 3. ACUTE LYMPHOBLASTIC LEUKEMIA PHENOTYPES .4LL-LI* (I1 = 27) ALL-L2* (n = 101) ALL-NOS* (I1 = I I ) Toial ALL-L3+ (n = 6 ) ALL-T CCII$ (n= 11) b/Y 0/26 27/32 l / l O l (84%) ( 1 % ) l/l 0/ll 31/42 1/138 (81%) (<I%) 1/1 0/6 313 7/11 (61%) 0/23 O/Y 1 019 27/27 ( I 00% ) IO'/ IO1 ( I 00%) 11/11 i/ 7 (19%) 37/ I 0I ( 3 7 %) ?/I I o/ I24 016 9/11 (81%) 1-to/1-10 ( 100% ) h/h ( I [KK ) O/lI ?h/l 3Y ( 3 3 %) 016 ( loo'; ) 0/1 1 507 especially high in AML-MS, but positivity was also evident to a lesser degree in AML-M4. Both types of Ah4L had tlie highest percentage of CD 10-positive cases, but no other B-cell markers were detected. ,4.VL-M6 arid AML-,!f7.Only a few cases of erythroleukemia and megakaryocytic leukemia were studiedloo few for nican i ngful imm u n ophcnotypic analjsis. The cases studied, however, showed no obvious dwiation phcnotypically from other types of AML. Ad4IL-Ai0. Nine cases of minimally differcntiated AML were studied. representing 4% of AMLs and 2% of all acute leukemias. By definition. these acute leukemias were niyeloperoxidase- and Sudan black-B-negative, but expressed at least one myeloid associated antigen. CD7 was positive in nearly one half of the cases examined for this antigen. CD 14was detected in one case. but nonspecific esterase activity was lacking. All ninc cases were positive for HLA-DR and CD34. 9ciuc / j n i p hoblusric Icirkcmiu. One h undred fifty-se\*en cases of ALL were studied: the results are giwn in Table 3 . Becausc immuuophenot~~pdcefincs important prognostic and therapeutic subgroups in this tl'pc of acute Icuheniia. tlie ALLs were divided into thrcc groups on tlie basis of lineage, as determined by surhce antigen expression. Yrc~cto:vor-U-ccl.l-ILL. One hundred folly precursorB-cell ALL were studied. Included were 27 ALL-L1 ( 1 770of ALLs and 7% of all acutc Icukeniias), 102 ALLL2 (65% of ALLs and 2770 of all acute leukemias). and 1 1 cases (7% of ALLs and 370of all acute leukemias) in which a morphologic subtype could not bc assigned. By definition, all 140 cases were positive for CD 19. In addition, all expressed HLA-DR. CD2O-positivity was de- tected in one third of the cascs. and was morc prevalent in the ALL-L2 than in the ALL-L1 subtype. Only 6'70 of the cases failed to express CD IO. and the majority, 81%. were positive for CD34. Prccursor-B-cell ALL was the acute leukemia type most prone to mixed-lineage antigen expression. A single myeloid antigen was present in 14% of these ALLs, whereas two myeloid antigens were evident i n 9%. Myeloid antigen positivity \vas morc Trcquent in ALL-LI than i n ALL-LI. Precursor-B-cell ALL was the most likcly acute leukemia in this study to be CD45-negative, with 19% of cases failing to express the coninion leukocyte antigen. A single T-cell antigen w a s detected i n two cases. One ofthe ALL-L~hs ad monoclonal surfacc iiiiniunoglobulin light and heavy chains. and was also Tdt- and CD34positi\.e. B-cdl .1LL. All six cases of B-cell ALL expressed a mature B-cell phenotype (CDl.9- and CD'O-positive, C D W and Tdt-negative) \\pith monoclonal surface imniunoglobulin light and heavy chains. Four of the six cases were positive for CD 10. and onc case expressed the pan-T-cell antigcn CD5. Four of the six cases displayed typical FAB-L3 niorpholog!~. Dcspitc a mature B-ccll phcnot\pc. houwcr. t\vo displa!.ed non-L3 blast morphology. T-ccll.ILL. Eleven cases of ALL \vith a T-ccll phcnot!pc wcrc studicd (7% of ALLs). CD5 and CD7 wcrc the T-ccll antigens niost commonly expressed. An atypical T-ccll phcnotypc, with abcrrant expression of at lcast one pan-T-cell antigen. was detected in all but one case. CD3 and CD8 cxprcssion were analyzcd in six cases; fivc were negative for both antigens and one was positive for both. CDlO was dctectcd i n four cases, but none of the nine T-cell ALLs studied for HLA-DR expression were positi\,c. singlc nigcloid antigen \vas prcscnt in two cascs. .4cwrc ~r,~di~~~~/.c,iriIccrurkcc.rn/tirr. Four cases ( 1%) of the 382 acute leukcniia studied oKercd no cvidcncc of lineage coinmitmcn t, cit Iicr cytochcm ically or inim u nophenotypically. N o Iyniphoid or mycloid-associatcd antigens were dctccted. M!dopcroxidasc was not dctcctcd i n any of the cascs. Terminal deoxynuclcotidc transferasc was positive in two of tlie three cases tcstcd. HLADR was cxprcssed by three cascs of thc acutc undifl'crcntiatcd leukemia, but CD34 was prescnt i n only one case. I~r~class~firrahcilrre Icirkcniiri. Fourteen cases (4%) of acute lcukcmia wcrc difficult to classify. primarily duc to tlie prcscnce of a confusing array of niultilincage-associated antigcns and no distinguishing morphologic or cptochcmical fealurcs. Included i n thcsc w'crc I2 caws of n7yclopcroxidasc-iicgative acute lcukcmia u.it1i myeloid antigens (AML-MO) and coexprcssion of one or more Bor T-lymphoid antigens. DISCUSSION Inimunoplienotypic studies have an established place in the diagnosis and classification of acute leukemia.' The availability of murine iiionoclorial antibodies that are reactive with lymphoid- and myeloid-associated surface epitopes, coupled tvith relatively user-friendly and inexpensive flow cytoineters, has made tlie use of immunophcqotypic analyses in acutc lcukcmia casily acccssible to many laboratorics. Nunicrous reports on the phcnotypic characteristics of acute leukemia have been generated. Most ofthe studies have expanded our understanding of these diseases, and have facilitated the recognition of certain types of acute leukemia, such as minimally differcntiated AML and B- and T-cell ALL. Applying extensive panels of antibodies to large nuiiibers of acute leukemias. however. also has dramatically increased the heterogeneity of acute leukemia. At times this has created confusion and uncertainty about the significance oftlic relationship between phcnotypc and biologic bchavior. and has complicated classification in a small proportion ofcascs.8This study gcncratcs inimunophenotypic profiles that may hclp to bcttcr dcfine thc range of antibody rcacti\rity for many typcs of acute leukemia. Thc data were collected from samplcs rcfcrrcd for characterization and diagnosis of acute lcukcmia. Tlic relatively uniform diagnostic approach and consistent antibody panel applied to thc acute leukcniias creatcd a unique opportunity for an analysis of this sort. Acute Myelogeirous Leukemia I n this study, as i n previous AML was a moderately heterogeneous group of diseases. The ini- munophenotypic signature of these acute leukemias is cxprcssion of one or niorc myeloid antigcns, which was seen i n the nearly all cascs. ,2 lack of myeloid Lintigcn positii'ity docs not absoltilcl\~c\cludc AML. ho\\.c\-cr. hccausc thrcc cascs failed to mark as myloid despite thc prcscnce of niycloperoxidasc activity and morphologic fcaturcs that w r c consistent \vith AML-h42. ,4ltliou_rli unusual, niycloid antigcn negativity in h o ~ ~ c i , f i rA/ Mc I- 'has been repotlcd prc\.ioLISI y.9.12. I n general. little niixcd-lincagc antigcn cxprcssion by a singlc Iyniphoidassociated antigen was sccn in this series of AML. Whcn prcsciit , thcse smalI exprcssions of 1incage proiniscuity should not be construed as sulticicnt c\,idenccto v'arrant a diagnosis of nii xed4in cage leu kcmia. l4 Initially thought to bc a T-cell antigen. CD7 is now kno\vn to be present in A M L.I5 CD7 expression appears to bc associated with blast ccll irnmaturity,15 because tlie perccntage of positive cascs \vas greatest i n AML-MO. AML-M I , and AML-R4Sa. CD2 was the only othcr Tccll antigen dctccted. B-ccll antigcn expression was almost cxclusivcly CD19. and \vas predominantly sccn i n .AML-h42. This association has recently been reported by Kit3 and colleagues. and also lias bcen linkcd \vith a specific cytogenetic abnormality. the t(S:2 1) translocation." This translocation \tas dcmonstratcd in all four of the CD 19-positive cases of AML that \ve studied karyotypically. CD 19 has been associated previously with AML-h45," but none u'crc detected in this series. Alniost onc half of the cascs of AML-M4 and AMLA45 cxprcsscd CD14. This antigcn is not rcstrictcd to acute lcukcmia with monocytic diffcrcntiation. howlever. but was also present in 6% of AML-MZ and one case ofAML-M3. Up to 25% of M l and M 2 AMLs have been previously rcported to express this antigen."," but CD14 positivity in AML-M3 is apparently rare. Acute promyclocytic leukemia (AhlL-M3) is an unusual and distinctive disease clinically. morphologically. ininiunoplienotypically. and ei'cn genotypically." As previously reported, the majority of cases of AML CNpress HLA-DR.s.9.'1The clear exception to this rule is AML-M3, in which only rarc cases are reported to bc p o ~ i t i v c . ~I .n' ~this study. all CD3-positivc AML were AML-M3 types. which \vas confirmed by cytogciietic findings. Positivity for this T-ccll marker has becn u:clldocunicnted in AML-M3. and the prcscnce of this antigcn lias c w i becn associated \villi specific breakpoints within the t( 15: 17) translocation.20Othcr studies clearly indicate that exprcssion of this antigen niay occur in other subtypes of AML." For this group of acutc leukemia, ho\twer. a CD'-positivc/HLA-DR-negati\.e myeloid phenotype was spccific for AML-M3. Acute Lympliocytic Lerikeniirl The separation of ALL into B- and T-cell typcs is clinically important, with both therapcutic and prognostic implication^.'^^^' The cases of ALL i n this study were TR A W EEK 509 thercfore divided into thrcc subtypes, which were based primarill on immunoplienotypic findings. The Precur- sor-B-cell ALL all expressed CD19, by definition. I n ad- dition, all were H LA-DR-positive. Otherwise, tlicse acute leuhemias were the most hcterogcncous leukemias in this study. The majority of cases expressed CD10, which is a marker of longstanding and wcll-known prog- nostic value.24On the other hand. C D IO-negativity has been associated with a poor'progn~sis.'A~s has been prc- viously reported, CD20 was more comnion on ALL-L2 than on ALL-L1 .'6 In marked contrast to AML, a signifi- cant minority of precursor-B-cell ALLs failed to express the comnion leukocyte antigen CD45, a finding of possi- ble prognostic sig~iificance.~M' yeloid antigens were commonly present, however, and were seen either singly or in pairs in 23% of cases. As in AML, the presence of these aberrantly expressed antigens should not preclude an othenvisc straightforward diagnosis of precursor-B- cell ALL. The expression of T-cell antigens is infre- qucntly detected in this type of B-cell ALL is a rare disease, and is the least conimon subtype of ALL. Classic FAB-L3 morphologic features are associated only with a B-cell phenotype. Morpho- logic features that are characteristic of non-L3 blasts how- ever, also can be seen in this subtype of acute leukemia.'* Two cases in this study expressed typical B-cell ALLplie- notypes, but did not display an ALL-L3 appearance. Cy- togenetic study was performed on one of these cases, and revealed a t(8:22) chromosomal abnormality. Inimuno- phenotypic studies are but most cases previ- ously reported and all of the cases of B-cell ALL in this stydy expressed nionoclonai surface immunoglobulins and were Tdt- and CD34-negative. One case expressed CD5. an unusual finding, but none were positive for other T-cell or myeloid markers. CDlO was detected i n the majority of cases, which is t y p i ~ a l . ~ * * ' ~ The phenotypic heterogeneity of T-cell neoplasms is well- documented, and covers the range of phenotypes expressed during thymic differentiati~n.~'-M'~ost cases will express more than one T- lineage marker. Aberrant deletion of one or more pan-T-cell antigens is comnion in this disease, however, and may be a helpful diagnostic finding.34 All but one of the I 1 cases of T-cell A L L showed deletion of one or more of thc 4 pan-T-cell anti- gens used. CD7 was the pan-T- cell antigen most often expressed by the T-cell ALL in this scnes and in CD7 is also the antigen that is most commonly deleted in these acute leukemias.36No CD7-positivity was detected in any pre-B- or B-cell ALL, which is typical.35 The HLA-DR antigen often is not expressed in T-cell ALL,32 which is another phenotypic feature that distinguishes this acute leukemia from pre-B-cell ALL and B-cell ALL. None of tlic cases of T-cell ALL in this study were HLA-DR-positivc. CD 10 has l m n rcpoi-ted i n 20-3070 of T-cell A L L s . ~ 'a frcqucncy approximately similar to that sccn in this series. Myeloid antigen expression also has bccn reported.-" as was seen in one CBSC in this study. Expression of myeloid antigens may bc a potential sourcc of confusion considering thc occasional presencc of CD2 and CD7 in AML. Urrd#kcnliated and Mixed-Lineage Leukemias A null phenotype or an indeterminate phenotype, 3s seen in acute undiffcrentiated leukemia and niixcd-lin- cage leukemia, respectively, comprised approximately 570of the cases studied. All four cases of acute undifferentiated leukemia expressed no myeloid or I!.mphoid * surface antigens, and were characterized only by variable expression of HLA-DR and CD34. The clinical conse- quence of this null phenotype is ~ n c l e a r . ~ *T*h'e~mixed- lineage leukemia conipriscd a diverse group with pheno- types specifically excluded from classification by FAB criteria, such as AML-MO Lvitli single B- or T-lymphoid antigen expression. or with confusing multilineage pkc- notypes and no convincing morphologic or cytochemi- cal evidence of lineage preference. The clinical signifi- cance of multilineage phcnotypcs in acute leukemia is also unknown, and because the definition ofthis condi- tion varies, direct comparison \vith existing studies is dif- ficult. CD34 is an antigen normally expressed by immature hematopoietic cells. Normal bone marrow contains less than 3% CD34-positive cells. making this an excellent niarkcr for monitoring blast cell populations. However, not all acute leukemias express this antigen. Approxi- mately 45-6570 of AMLs404' and about 75% of pre-B- cell ALLs~"'~have been reponed to be CD-34-positive. Using a level of positivity that is similar to pre\ious rc- port^,^'.^^ a slightly greater proportion of CD34-positive acute leukemias were dctected in this study. both for AML and ALL. Previous studies suggest CD34-positive AMLs are more likely to be AML-M1 or AML-M? type^,^'.^' but this was not the case in our study. The level of CD34 cxpression was approsiniately equal in all sub- types of AML except AML-MO, u.hich was consistently positive. Even AML-M3, with relatively differentiated morphologic and imniunophenotypic features. ex- pressed CD34 at Icvels similar to other AMLs. This intu- itively unusual finding is similar to that reported by Merle-Beral and and may be an atavistic fcature of leukemic cells, because CD34 expression in normal promyelocytes has not been documented. Simi- lar to a large group of ALLs studied by Borowitz and c o - w ~ r k e r s C, ~D~34 expression in ALL was associated with fcaturcs of inimaturc blasts, such as ALL-LI nior- VOl. 99 * No.4 510 I l I< M ,\ T(11' AT 1 1 0 1.0 G Y phology,abscncc of CD20 positi\ ity, and ni!doid anti- gen coexprcssion. Only cascs of ALL-L3. composcd of nearly terniinally differentiated B cells. consistcntly failed to mark for this antigcn. Largc studics arc fcw.41-43 but the clinical significance of CD33 cxprcssion i n acutc Icuhcniia appears to bc a function of cell lincagc. with positivity portending a poor prognosis in adult AhlL, and a good prognosis i n pediatric ALL. Thc utility of ininiunophcnot! pic analysis i n acutc leukemia is not liniitcd to tlie charactcrization of cell lineagc. but also may yield important prognostic infor- mation. For cxample. CDIO-positivity in pre-B-ccll ALL is clearly associated with a favorable o u l ~ o n i ew. ~he~reas lack of this antigen portends a poor progn~sis.'C~ D7 expression in AML has been linked with an immature phenotgpc in AML. and conscquently, a worse out- c o ~ i i c .I' n~ many instances, howcvcr, conflicting rcpoi-ts makc evaluation of the significancc of antigen cxprcs- sion difficult. For example. the significance of HLA-DR in AML is unclear. bccausc discrcpant results Iiavc becn r e p ~ r t e d .T~h~e .s~ig~nificance of mixed-lineage antigen expression is particularly tortuous. Myeloid antigen ex- pression in cases of ALL has becn reported i n numerous previous studies to have an adversc clinical influ- e n ~ e . 4 ~b.u~t *evidence also exists that intensive niulti- agent chemotherapy negates this Lymphocyte surface inackers in AML has bcen rcpoi-ted to infer a poor response to therapy,'O have no ~ft'ect,o~r' ha\c fa- vorable prognostic significa~iceT.~h~e reason for thcsc conflicting findings is unclcar, but may be related to dif- ferenccs in patient populations (adult i'ersiI.y pcdiatric, refcrral wrsus community-based). i n therapy, i n antibod- ies used, and i n definitions of what constitutes atypical lineage antigen expression. I n general, mixcd-lineage an- tigen expression is often, but not always. associated with other immunoplienot\pic and morphologic fcaturcs of immaturity in both AML and ALL. Like CD34, this ap- pears to confer a worse prognosis in AML, and either have no effect or predict a better response to tlicrapy i n ALL. cspecially in pediatric patients. Funhcr studies us- ing readily available reagents and uniform patient inclu- sion/responsc critcria will be nccded before valid conclu- sions can be drawn. I n tlie nieantinie. caution is urged in the application of inimunoplienotypic findings with pur- ported clinical significance. A few studies on tlie phenotypic changcs associated with recurrent acute leukemia have becn publishcd. For AML. rclapsc of disease is associated with a tendency toward a more inimaturc phcnotypc nianifcst as in- crcascd expression of CD33 and CD34 with dccrcascd exprcssion of CD15, and is associated with poor out- Only small changcs in phcnotypc liavc becn re- ported i n the fcw studics pcrformcd on rclapscd ALL.26.54 In both AML and ALL. thc con\m-sion from a CD34-negative phcnot\pe at initial prcscntation to CD34-positi\~at rclapsc has heen reported,s3 but thc convcrsc change. from CD31-positi\.ityto CD34-ncgati\,ity. is apparently ram. I n addition to singlc aiitigcn changcs. dramatic shifts in lincagc also may occur with the rccurrcncc of acutc lcukeniia.''~56 Examples of a \vli olesaIc li ncage s\vitch arc u n coninion .a11d t11 e cli ii i cal significancc of this occurrencc is unclcar. The motivation to institute ininiunoplieiiot!pic stud- ies on acutc Icukcmia was fuelcd initially by the dcsil-cto improve tlie charactcrization of this hetcrogcncous, group of neoplasms. Clcarly this objccti\,e has bccn met. as imniunoplieiiotypic analyses have been shown repcatcdly to improvc tlie rccognition of lincage in acutc leukcinia. Frcnch-A m crican-Rri t isli criteria now rcq u i rc plicnotypic studics for some subsets of .4ML.4.' Thcsc studies are also an cssential component in the diagnosis of ALL.' I~ui-thermore,thc existcncc of mixed-lincagc lcukeniia is esscntially dependent on phenotypic studies. I n addition to identifying cell lineage in acutc leukemia. certain antigenic markers have becn associated with patient outcome, and will undoubtedly play a bigger role in prognostication as more cases are studied and clinically significant parameters become clearer. Finally. immunophcnotJping can play a role in the evaluation of treatmcnt cfficacy. by monitoring blast-associated markcrs such as CD34. and may also prove to bc an important tool in assessing niinimal residual . ~ l ~ ~ ~ - / i ~ ~ i i . / ~ ,S~p/e~c(ia, l/ ~th~a~n,krs~to/ .J~ohn Zhang for c ~ c c l l c ntechnical assistance. and to Drs. Russell Brynes and Daniel .Arlxr for helpful discussions. REFERENCES I . Neame PB. Soamhoonsrup P. Bro\~mianGP. et al. Classif!ing acute leukemia by ininiunoplienot!ping: A combined FAB-immunologic classification of .AhlL. Blood 19S6:68: 1-755-1 362. 2. Bennett J M . Catovsky D. Daniel M-T. et al. Proposals for tlie classification of acutc leukemias: French-Amcrican-British (I'AB) co-operaliic group. Br 1 Hacmatol 1976:.33:15 I-45X. 3. Rcnnclt J M. Catovsky D. Daniel h4-T. et 31. f'roposcd rc\.iscd (Tilt..ria for the classification of acuic leukemia. Ann Intcrn bled 1985: lO3:62O-625. 4. Bcnnctl J M , Catovsky D. Daniel M-T. CI al. Criteria for the diagnosis of acute leukemia of megakanocytic lineage (M7). .Ann Intern h4cd 1985: 103:460-361. 5 . Bennett J M . Catovsky D. Daniel hl-T. et al. Proposal for the recognition of minimally difercntiatcd acute myeloid leuheniia (AML-hl0). Br J Hacniatol 1991:78:325-379. 6. Gale RP. Ben Bassat I.Annotalion: Hybrid acute leukemia. Br J Hacmatol 1987:h5:261-161. 7. Foon KA. Todd KI-.Immunologic classification of leukemia and Iynphoma. Blood 1986:bPS: I-?I . 8. Bradstock KF. Kirk J.Grimslcy l'G.ct 11. Unusual inimunophcnot! pes i n acute leukemias: lncidcncc and clinical corrclations. Br J Hacniatol 19X9:72:512-5IX. 9. Hanson CA. Gajl-Pcczalsha tiJ, Parhin JL. et 31. Immunophcnot!ping of acutc niycloid leukemia using monoclonal antibodics and the alhalinc pliospliatasr-aiitialkalinc phaspliat;isc tcchniquc. Blood 1987:70:83-89. IO. Tcrstappcn LWhihf. Safibrd hl. Koncmann S. et al. Ha\cytonici- ric charactcrization o f acute mycloid Icuhcniia. Leukemia I Y9 I :S:757-767. 1 I . Rohcns GI-. Spcncc DG. Padmos M A . et al. hlorphologic iininunophcnotyic and cytogcnctic patterns of adult acute mycloid Icuhcniia in Saudi Arahia. L e d Rcs l902:16:l81-1~~0, I ? . San hligucl JF. Gonzales hl. Canizo MC. et al. Surhcc niarkcr is in ;~cutcni!cloiJ Icuhciiiia and correlation \vitli FAB cation. Br J Hacmatol 19Sb:61:547-560. 13. Scslii B. Kashyap A. Bcnnctt J M . Aculc niycloid Ieukcmia with an unusual phcnotypc: h4yclopcroiidasc (+). CDI 3 (--). CDiJ (-) and CD33 (-). Br J Ilacniatol 1992:81:374-377. 13. Grcavcs MI-. C'han LC. F u h y AJW. CI ;II. Lineage proniiscuit!, in hcniatopoietic differentiation and Icukcmia. Blood 198(,:07: I I I. 15. Lo Coco I:. Dc Rossi G. Pasqualetti 1). ct al. CL17 positive acutc mycloid lcukcinia: A suht!yc associated with ccll immaturity. Br J Ilacmatol 1989:73:380-415. 16. Kita K. Nakasc ti. h4iu.a 1-1. ct al. Phenotyl1ical char.actciistics o f acute myclocytic leuhcmia associatcd u.iLli thc t(8:l I )(q??:qX) chromosomal abnormality: Frcqucnt chprcssion of immature B-ccll antigcn CD19 together with stcm ccll antigen CD34. Blood 1992:80:470-377. 17. Campos L. Guyotat D. Larcsc A. ct al. Exprcssion orCD 19 antigcn on aculc monohlastic leukcniia cells at diagnosis and after TPAinduced differentiation. Leukemia Res 1988: I2:369-372. 18. Avvisati G. Annotation: Acutc proniyelocytic Icukcmia. Br J Hacmatol 1992:81:315-330. 19. Lo Coco F. Avvisali C. Divcrio D. et ai. Rcarrangcnicnts of the RAR-n gcnc in acutc promycloc!.tic Icukcmia: Correlation \vith morphology and iinmunophcnotypc. Br J liaematol I99 I : 78~494-499. 20. Claiton DF. Rcading CL. Nagrajan 1. et ai. Corrclation o f CDZ cxprcssion with Phl L gcnc breakpoints in paticnts with acute promycloc!zic lcukcmia. Blood 1992:80:582-586. 21. Cross AH, Goorha R M . Suss R. et ai. Acutc niycloid Icukcmia with T-lymphoid features: .4 distinct biologic and clinical entity. Blood 1988:72:579-587. 22. Sobol RE. Royston 1. LcBien TU'. ct ai. .Adult acutc Iympoblastic leukcmia phenotypes defined by monoclonal antibodies. Blood 1985:65:730-735. 23. Thicl E. Rodt H. H u h D. et al. Multiparameter classification of acutc lymphoblastic leukemia: Evidcncc for further T subgroups and evaluation of tlieir clinical significance. Blood 1980;56:759-7772, 24. Grcavcs MI:. Haiiri G . Newman R A . ct al. Sclectivc ckprcssion of thc coninion x u t c lymphoblastic Icuhcmia (gp l@O)antigcn on imniaturc lymphoid cclls and their nialignaiit counterparts. Blood 1983;61:6?8-639. 25. Vannicr. JP. Bcnc MC. FaurcGC, ct ai. Invcstigaiion ofthc CDlO (CALLA) ncgative acute lymphoblastic Icukcniia: Further dcsription of a group with a poor prognosis. Br J Hacmatol 1989:72: 156-160. 26. Borowitz MJ. Immunologic niarkcrs in childhood acute lyniphoblastic Icukcmia. I-lcmatol Oncol Clin North Am 1990:4:743765. 27. Caldwvll CW. Patterson WP. Hakami N. Altcrations of HLc-I (T?OO)fluorcsccncc intcnsity on acutc lymphoblastic Icukcniia cclls may rclatc to tlicrapcutic outcoinc. Lcukcmia RCS 1987~11:103-106. 28. Kosanda C . Cantu-Rajnoidi A. Invcrniui R. ct al. R-cell acuic lymphoblastic lcuhcmia (B-?\LLI:A report of I7 pcdiatric casrs. f-lacniatolagica 1992:77: I 5 1-1 5 5 . 29. \!ccchio LD. Fasanaro A. Schiavonc EM. Fcri-ara I-.B-ccll acute lymphoblastic lcuhcmia (B-ALL) hctcrogcncit! (Icttcr). Br J I lacmatol 198'):72:29 I . 30. Walk A. AI-Katib A. W'ong (3)'. el al. Multiparamctcr charactrrization of L3 Icukcmia cell populations. Lcuk Res 1957: 1 l:7.;83. 3 I . Do\\dl BL. Borowitz MJ. Bnyctt JM. et al. Iminunologic and clinicopathologic features of coninion acute I! mphohlastic Icuhcniia antigen-positivc childhood T ccll Icuhcmia: .4 Pediatric Oncology Group study. Cancer 1087: j9:2020-?@26. .32. Fcllcr AC. Pnrnarcsch MR. Stcin H. et al. Inini~rnoplicnot~poicf T-lymplioblastic Icukcmia/l! mphoma: Corrclations \\ i t h normal I`ccll maturation. I.cuh Res 1986: 10:IO25-IU3I. 33. Crist U'M. Shustcr JJ. Fallctta J. et al. Clinical fcaturcs and outcome in childhood T-ccll leuhernia-lyniphoina according to , stage of thymocytc dilti.rentiarion: A pediairic oncolog! group study. Blood I9SR7l: 189 I - IS97. 3-4. Kno\vlcs Dhl. Tlic human Tccll leukemias: Clinical. c!tomorpliologic. ininiunophcnot!pic. and _eenot!pic characteristics. I turn l';~thol 1956: 17:13-3.7. 35. Vodinclich W. Tax M'.Bai 1'.rt 31. 4 monoclonal antihod! (\\`.l.l I for dctccting Icukcmia of T cell precursors (T-ALL). Blood 19833:62: 1 108-1 I 13. 36. Knowlcs DM.lmmunophcnotypic and antigcn receptor y e w rcarrangcmcnt analj.sis i n T ccll ncoplasia. /\in J Pathol 1989: 134:761-785. 3 7 . Drcxlcr HG. Sagawa K, Menon hl. Minowada J. I'hcnot!.ping of nialignant hematopoietic cells: 11. Reactivity pattern ofm!cloid monoclonal antibodics \vir11 emphasis on M C S - I . Leuk Kes 1986: 10:17-23. 38. Conipana D, Hanscn-Hage TE. Matutes E. ct al. Phenotypic. gcnotypic. cytochcniical and ultrastructural characterization of acu te u ndi ffcrcnt iated Icukenlia. Lcukcmia I 990: 1:620-621. 39. Brito-Babapullc F. Pullon H. Layton Dh4. et al. Clinicopathologic features of acute undifferentiated leukcmia with a stem ccll phenotypc. Br J Hacmatol I990:76:110-2 I . 10. Tindle RW, Nichols RAB. Chan L. CI al. PI novel monoclonal antibody BI-3CS rccugnircs myelohlasts and non-B non-T I! niplwblasts in acutc leukemia and CGL hlast crisis. and rcacts with immature cells in normal hone niarro\\. Lcuk Rcs 198539:1-9. 31. Borou,itz MJ. Gockcrnian JP. h4oorc JO. et ai. Clinicopathologic and cytogenetic featurcs of CD34 ( M y I O)-positivc acute nonlymphocytic Ieukcmia. Am J Clin Pathol I959:9 1:265-270. 12. Gcllcr RB. Zahurak M. Hunvitz CA. et al. Prognostic iniponancc of iiiiiiiunOphcnot!ping in adults with acute m!eloc!tic leukemia: The significance of the stem-cell glycoprotein CD33 ( h o IO). Br J Hacmatol 1990;76:350-337. 13. Borowitz MJ. ShustcrJJ. Civin CI. et al. Prognostic significancc o f CD33 cxprcssion in childhood B-prcsursor acute i! mphnc! tic leukemia: A Pediatric Oiicol~gyC r o u p study. J Clin Oncol 1990~5:13S9-1398. 11. Mcrlc-Beral 1-1. Duc LXC. Leblond V. et al. Diagnostic and prognostic significance or niyclomonocytic antigens in acute myeloid leukemia. Br J Haematol 1989: 73:3?3-?3O. 3 5 . Griffin JD, Davis R. Nelson DA. ct al. Use ofsurface marhcranalysis to predict outconic of adult acute myelogenous Icuhcniia. Blood 1986~68:1232-1241. 36. Swirsky IIM.Crcavcs hlF. Gray RG. Rces JKH. Terminal dcoxynuclcotidyl transferase and HLA-DR cxprcssion appear unrclatcd to prognosis of acutc ni!,cloid Icuhcniia. Br J I-lacniatol 1988;70:193-198. 37. Sobol RE. hlick R. Royston I. et 31. Clinical importance of mycloid antigcn cxprcssion in adult acute Iyniphoblastic Icuhcmia. N Engl J hlcd 1987:316:1111-I 117. 48. Wiersnia SR. Ortega J, Sobel E, Weinberg KI. Clinical importance of myeloid antigen exprcssion in acute lymphoblastic leukemia of childhood. N Engl J hled 1991:324:800-808. 49. Pui C-H. Behni FG. Singh B. e( al. Myeloid anligcn expression lachs prognostic value in childhood acute lymphoblastic lcuke- niia treated wit11 intensive muhiagent chemotherapy. Blood 1990:75:198-202. 50. Mirro J M , ZiprTF. Pui C-H, et al. Acute mixed lineage leukemia: Clinicopathologic correlations and prognostic significance. Blood 1985;66:1 115-1 123. 51. Smith FO. Lampkin RC. Versteeg C , et al. Expression of lymphoid-associated cell surface antigens by childhood acute myelogenous leukemia lacks prognostic significance. Blood 1992:79:24 15-2422. 52. Ball ED. Davis RB. Grilfin JD. et al. Prognostic value of lynipho- cyte surface markers in acute niycloid leukemia. Blood 199 I ;77:2242-2250. 53. Thomas X. Campos L. Archinihaud E. el 31. Surface marker expression in acute myeloid leukemia at first relapse. Br J Haeina- 101 IYY2;HI:40-44. 54. Raghavachar A. Thiel E. Bartrani CR. Analyses of phenotype and genotype in acute Iymphohlaskic leukemia at first presentalion and in relapse. Blood 1987:?0:1079-1083. 55. Stass S. Mirro J. Unexpected hclerogeneity in acute leukemia: Mixed lineages and lineage switch. Human Pathol 19SS; 16:564-866. 56. Neame PB. Soaniboonsrup f'. Browman G. et al. Simultaneous or sequential expression of lymphoid and myeloid phenotypes in acute leukemia. Blood 1985;65:142-148. 57. Canipana D. Couaan-Smith E. Janossy G. The immunologic detection of minimal residual disease in acute leukemia. Blood 1990:76: 163-17 I . A.J.C.P. * April 1993