This article was downloaded by: [UNSW Library] On: 13 August 2015, At: 00:50 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: 5 Howick Place, London, SW1P 1WG
Leukemia & Lymphoma Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ilal20
Multiphenotypic Acute Leukemias: Clinicopathologic Correlations and Response to Therapy ac
b
c
c
Patrick J. M. Philip , Fabrice Monpoux , Isabelle Sudaka , Francoise Baudouin , Jacques c
d
Bayle , Marie C. Bene & Gilbert C. Faure
de
a
“Unité de Biologie Cellulaire et Tissulaire”, Hôpital Pasteur, France
b
Clinique Médicale Infantile, Hôpital de Cimiez, France
c
Laboratoire Central d'Hématologic, Hopital de Cimiez, CHR, Nice, France
d
Laboratoire d'Immunologie, Fac. Med. & CHRU, Nancy, France
e
Groupe d'Etude Immunologiques des Leucemies (GEIL)*, France Published online: 01 Jun 2015.
To cite this article: Patrick J. M. Philip, Fabrice Monpoux, Isabelle Sudaka, Francoise Baudouin, Jacques Bayle, Marie C. Bene & Gilbert C. Faure (1992) Multiphenotypic Acute Leukemias: Clinicopathologic Correlations and Response to Therapy, Leukemia & Lymphoma, 7:5-6, 489-495, DOI: 10.3109/10428199209049806 To link to this article: http://dx.doi.org/10.3109/10428199209049806
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
0 1992 Hanvood
Leukemia and Lymphoma, Vol. I , pp. 4 8 9 4 9 5 Reprints available directly from the publisher Photocopying permitted by license only
Academic Publishers GmbH Printed in the United Kingdom
Multiphenotypic Acute Leukemias : Clinicopathologic Correlations and Response to Therapy
Downloaded by [UNSW Library] at 00:50 13 August 2015
PATRICK J. M. PHILIP’v3, FABRICE MONPOUX’, ISABELLE SUDAKA3, FRANCOISE BAUDOUIN3, JACQUES BAYLE3, MARIE C. BENE4, GILBERT C. FAURE4 and the GEIL5 1. “Unite de Biologie Cellulaire et Tissulaire”, Hbpital Pasteur, 2. Clinique Medicale Infantile, Hbpital de Cimiez, 3. Laboratoire Central d’Htmatologie, Hbpital de Cimiez, CHR, Nice, 4. Laboratoire d’lmmunologie, Fac. Med. & CHRU, Nancy, and 5. Groupe d’Etude Immunologiques des Leucemies (GEIL)*. France.
(Received I6 September 1991; injinal form I6 February 1992)
Multiphenotypic acute leukemias (MAL), defined by the coexpression on most blast cells of antigens classically attributed to different lineages, remain a rare event. We isolated a series of 26 such cases from a cohort of 1565 leukemic patients whose cells were immunophenotyped at diagnosis. Markers of B and myeloid lineage (BM) were associated in 16 cases (62%0),3 coexpressed B and T markers (BT), and T-cell and myeloid antigens (TM) were found in 7 (27%). A tumoral syndrome was observed in 69% of the patients, without significant differences between the immunophenotypic subgroups. Median event free survivals in the three immunophenotypic subgroups as defined were respectively 24 months for BM-MAL, 4 months for TM-MAL and 7 months for BT-MAL respectively. The poorer prognosis of TM-MAL was significantly different from that of BM-MAL (p < 0.001). This concurred with the poorer prognosis associated with CD7 expression or absence of CD10, both characteristic features of TM-MAL. KEY WORDS:
Leukemia,
Lineage infidelity,
Address for correspondence: Dr P. J. M. Philip, Secretariat du GEIL, Laboratoire d’lmmunologie, Faculte de Medecine, 54500 Vandoeuvre les Nancy, France. * J. Gardais (Angers), M. C. Leglise, G. Lecalvez (Brest), M. Bernier, M. Massy (Bruxelles), A. M. Griveau, P. Boutard, C. Lebrun (Caen), J. Bonhomme, B. Chassagne, F. Demeocq, C. Rapatel (ClermontFerrand), H. Jouault (Creteil Mondor), F. Lelong (Creteil Intercommunal), E. Solary, 0. Casasnovas, M. Maynadit (Dijon), J. C. Bensa, M. Favre, M. C. Jacob (Grenoble), C. Sartiaux (Lille), M. Drouet (Limoges), L. Campos, A. Larese (Lyon), P. Poncelet, C. Brunet (Marseille), T. Lavabre-Bertrand, J. Taib (Montpellier), G . Kandel, G. Jung (Mulhouse), R. Garand (Nantes), P. Philip, F. Monpoux, F. Baudouin, 1. Sudaka, C. Caldani, G. Pomier, A. Bernard (Nice), M. Reguer, H. Schill (Paris Val de Grlce), G. Schaison, L. Boumsell (Paris Saint Louis), D. Vanhaecke, L. Douay (Paris Trousseau), A. Brizard (Poitiers), J. P. Roualet, T. Tabary (Reims), N. Genetet (Rennes), 0. Lees, B. Lenormand, J. P. Vannier (Rouen), J. C. Le Petit, S. Acquart, 0. Sabido (Saint-Etienne), A. Falkenrodt (Strasbourg), J. L. Bremond, A. M. Masy (Tours), F. Roubinet, E. Kuhlein (Toulouse), M. C. Bene, G. Faure, P. Bordigoni, F. Witz (Vandoeuvre les Nancy), C. Bayle, J. Antonini (Villejuif).
Immunophenotype,
Prognosis.
INTRODUCTION Several authors have stressed that lineage fidelity is maintained in leukemic blasts, and that their phenotypes are the counterparts of those expressed by normal cells during early stages of lineage-specific differentiati~nl-~. These observations have been used to support pathogenetic models of leukemic proliferation postulating that cell-differentiation in leukemia is not abnormal but blocked or abortive. To complement morphological classifications, immunological classifications based on differentiation patterns of each hematopoietic lineage have been proposed, allowing the identification of leukemia subsets which respond differently to the rap^^.^. However, in a minority of patients, blasts appear to co-express markers of more than one differentiation lineage, as
489
Downloaded by [UNSW Library] at 00:50 13 August 2015
490
P. J. M. PHILIP et al.
defined by the consecutive workshops on leukocyte differentiation6. Various terms have been used to designate these unusual acute leukemias showing both myeloid and/or lymphoid features: hybrid, biphenotypic, biclonal, bilineal, chimaeric, mixed lineage switch or infidelity lineage.. .7. We have preferred to use the term multiphenotypic acute leukemias (MAL), previously used by others'.' to describe both intralymphoid infidelity (coexpression of B and T cell markers) and interlineage infidelity (coexpression of myeloid and lymphoid markers). We investigated the characteristics of 26 cases of MAL observed among a collaborative series of more than 1500 leukemias collected by the Groupe d'Etude Immunologique des Leucemies (GEIL). In this report we describe the various phenotypes observed, the clinical and biological characteristics of the patients, and discuss the evolution of MAL.
MATERIALS AND METHODS Patients Information at presentation was collected for all patients from the 32 centers who work together in the Groupe d'Etude Immunologique des Leuckmies (GEIL). A total number of 1565 malignant hemopat hies immunophenotyped at diagnosis was entered in the collective file between November 1983 and July 1991. The group of patients reported here included 26 patients with MAL, diagnosed and phenotyped at first involvement. Multiphenotype was defined as the co-expression of at least one specific marker of two distinct hematopoietic lineages (Table 1). Only samples containing more than 60% of blasts and less than 20% lymphocytes were included in this selection. Individual leukemias were entered as MAL only if the coexpression of markers was observed on more than 50% of blasts, i.e. when the percentage of two individual markers added up to more than 100% of the cells. Positivity thresholds for other markers were 20% for myeloid lineage antigens (CDllb, CD13, CD14, CD15, CD33), and 40% for lymphoid lineage markers (sIg. c-mu, CD19, CD20, CD21, CD24, CD1, CD2, CD3, CD4, CD5, CD7, CD8). Previously reported lineage infidelities, considered to be compatible with normal counterparts, i.e. CD19/CD21° (n = 30) or CD7/myeloid markers'' (n = 35) were not entered in this series. Multiphenotypic leukemias, as classically defined by the presence of myeloid markers on more than 20% of
Table 1 Monoclonal antibodies used for MAL immunophenotyping T-lineage CDl CD2 CD3 CD4 CD5 CD7 CD8
OKT6, T6" OKTll, Tll", Leu5 OKT3, T3", Led, IOT3 OKT4, T4", Leu3,10T4 TI", Leul, lOTla 121, Leu9, IOT7 OKT8, TS", Leu2, IOT8
Myeloid lineage CDllb CD13 CD14 CD15 CD33
M01" MY7", MCS2 M02", MY4", 20.3 LeuM1, ION1 MY9"
B-lineage CD19 CD20 CD2 1 CD24 Non lineage specific CDlO HLA DR
B4"
B1" B2", lOBla BA 1
J5", lOT5 OKDR, OKIal, I2", IOT2a
OK series: Ortho Diagnostics, Raritan, NJ, USA; 'Coulterclone, Coultronics, Hialeah, FL, USA; Leu series: Becton Dickinson, Mountain View, CA, USA; IOT series: Immunotech. Luminy,France.
blasts, or of lymphoid markers on more than 40% were not taken into account since double-labelling had not been performed in this retrospective study. Only leukemias with indisputable co-expression based on added percentages were retained. The patients were treated with different therapeutic protocols, depending on the policy of the hospital to which they reported. The most frequent protocols used in children were the successive versions of the FRALLE12 and protocols similar to the BFM13. Adults were mostly treated according to LALA protocol^'^. Complete remission was evaluated according to the CALGB criteria: normal clinical features, normal blood cell counts and bone marrow smears containing less than 5% of blast cells for more than a month. Methods Bone marrow (70% of the cases) or peripheral blood (30%) was collected in heparin at the time of diagnosis, before therapy. Classical techniques were used in all laboratories. Each sample was centrifuged on density gradient (Ficoll). A differential count of collected leukemic cells was done on a cytospin smear. The FAB subtype was determined by standard
Downloaded by [UNSW Library] at 00:50 13 August 2015
MULTIPHENOTYPIC ACUTE LEUKEMIA
criteria”. In addition, cytological studies included staining for the following activities: myeloperoxidase, chloroacetate esterase, Periodic Acid Schiff (PAS), naphthol acetate esterase. Surface immunoglobulins (SIg +) were usually identified in direct immunofluorescence, using fluorescent F(ab); anti-human immunoglobulins. The presence of intracytoplasmic immunoglobulins (cIg+) was investigated in direct immunofluorescence on cytospin smears. Other membrane molecules were identified using the monoclonal antibodies listed in Table 1, in direct or indirect immunofluorescence using FITC-sheep or goat antimouse Ig serum as second step reagent. Microscopes equipped with Ploem systems of epi-illumination were used to enumerate positive cells in each sample until 1988; since then most centers have been using flow cytometry. Results were expressed as the percentage of fluorescent cells for each marker. Several internal workshops were performed within the GEIL, where the same frozen cells were tested by all participants. Specific instructions were given for thawing. After determination of the phenotypes, the codes and relevant data were collected and compared. An intergroup reproducibility score of 94% was obtained, allowing a multicenter analysis of patients’ data. Student’s t test and Fischer’s exact test were used for the statistical analyses performed, with the Myosotis (Coultronics France, Margency, France) software. Survival curves were plotted using the Logrank test.
RESULTS Immunological phenotypes Table 2 shows the individual phenotype of each of the 26 MAL analyzed. Three types of combinations between B, T and myeloid-specific markers were observed. Blasts from 16 MAL patients were of B and myeloid cell lineage (BM), 7 had markers of both T and myeloid cell lineage (TM), and 3 co-expressed immunological markers of B and T cell lineage (BT). In one of the TM patients, CD19 was expressed on blast cells as well as several T and myeloid markers. Addition of the percentages of cells labelled by markers from different lineages yielded values ranging between 130 and 181 (mean: 159.6).HLA class I1 DR antigens were expressed in 65% of the cases, mostly in BM-MAL (%YO),seldom in TM (29%) and never in BT-MAL.
49 I
Similarly, CDlO appeared to be mostly expressed in BM-MAL (81%), and was only found once in BT-MAL. Predictably, CD19 and CD24 were the most frequently found B-cell markers, CD7 the most frequent T-lineage antigen, and CD33 the predominant indicator of myeloid lineage. CD19 was the only B-lineage marker in 4 cases (all BM-MAL), CD7 the only T-lineage marker in 2 of the 3 cases of BT-MAL, and CD33 the only myeloid marker in 6 cases (5 BM-MAL, 1 TM-MAL). Clinical and biological characteristics at diagnosis The series included 13 children (age range 7 months-16 years) and 13 adults (age range 17-85). There were 10 females and 16 males. Table 3 provides the classical features observed in the three immunophenotypic groups of patients at diagnosis. Tumoral syndrome (69%) and anemia (62%) were frequent. Mediastinal enlargement was surprisingly infrequent both in TM (2/7) and BT (1/3) MAL cases. Hyperleukocytosis was a rare feature in all groups. Statistical comparisons of the frequency of each of these features showed no significant difference between sub-groups. The distribution of MAL in the morphological FAB subtypes showed the predominance of L1 (50%) and L2 (35%) ALL. None of the BM-MAL were classified as AML according to morphological criteria, while 3 of the 7 TM-MAL were. Prognostic features As in other studies published by our group, the variety of therapeutic protocols used enabled comparison of the influence of phenotypic features while minimizing the specific bias of a single protocol. Initial remission was obtained in 69% of the patients (12 children and 6 adults), and the death rate was 42%. The median follow-up for the entire group was 16 months (range 1-48). Survival analyses were thus performed for a period of 24 months after diagnosis, yielding a satisfactory confidence interval. Survival estimates were performed as disease free survival, without relapse or death in remission. There was no significant difference in the outcome of males and females, or, more surprisingly, adults and children. As seen in Figure 1, the prognosis was significantly worse for TM-MAL patients. Significance was also reached to demonstrate the good prognostic value of CD7- or CD10+ blasts. These features possibly are
LI LI L1 L1 L2 L2 L1 L1 L1 L2 L2 L1 L1 L2 L2
L1
L1 L2 L1
L1
BM BM BM BM BM BM BM BM BM BM BM BM BM BM BM BM
BT BT BT
TM TM TM TM TM TM TM
LO MI M4 L2 L2
M1
FAB
MAL
0
0 0
0 3
5
0 0
25 4 0
0 0
0 70 0
0
0 30
2
0 3 13
0 0
clg
0
slg
5 2 0 0 0
1
33 0 41
65 9
15 1
6 50 62 13 2
1
72 9 25 37 13 0
CDZO
0
0
0
1
65 4
1
0 89 98
97 90 36 87 82 77 89 94 97 74 74 58 88 89 60 83
CDI9
7 0
61 91 85
2
1
0
0
20
0
CDZI
Table 2 Individual immunophenotypic features
15 0
1
88 96
67 94 98 88 70 82
98
0 84
24 94 85
CDZ4
0 1 0 90 0
4
85 0 0
4
1 1
5
0
CDI
9
42 5 0 90 7
2
97
92 1 4
19
18
12 6 3 1 3 2 7 17 3 3 9
0 6
CDZ
1 6 80 6 90 0 0
0
1
18
4 0
3 88 0 0
3 26
4 4
1 1 10 12
1
2 2 12 17
1
4 0 8 0
0
0
2
CD4
6 2 13 5 4
0
CD3
5
0
97 58 43 7 8 0 70 0 9
4
81
15 17
4 2
15
7 2 15 12 2 1 3 2
CD5
9 0
0
0
2 0
79
4 7
1
1 0 1
8 2 4 3
CDB
95 3 69 70 44 6 2 0 80 90 0 0 6 6
93
2 1 2 1 13 15 3 2 9 16 20
8 2 14
CD7
20 0 0
80 88
4 0 0
92
8
77 62 2
5
CDIIh
Downloaded by [UNSW Library] at 00:50 13 August 2015
79 0 68 0 0 80
1
2 0 0
33 4 3
1
5
11
1
10 0 0 10 0
66 0
CD13
87 2 2 90 0
1
8 0 0
11
1
3
1
1 0
0
0
CDI4
0 0 0
77 10
0 0
96 68 1 4 2 0 75 65
92 1
69
CD33
1 65 0 2 93 70
3 0 0
8 53 84 50 0 70 6 4 83 0 44 8 80
CDIS
0
0 0
1
1
1
93 3 7
7 85 82 60 98 60 4 87 92 7 75 80
40
96 86 94
CDIO
77 83 87 85 39
94 35
5
CD34
2 69 35 86 0 9 0 0
43 2 0
65
1 4 4 19 0 0 30
95 96 69 87 47 33 84 98 97 70 80 88 92 91 66 80
DR
k
rp
-.
za
2
a
3:
%
a
MULTIPHENOTYPIC ACUTE LEUKEMIA
493
Table 3 Clinical and biological features at diagnosis; incidences are reported as number of cases
Children Males Enlarged lymph nodes Splenomegaly Hepatomegaly Mediastinal mass Anemia (Hb < 100 g/L) High WBC ( > 109/L)
BM-MAL (n = 16)
BT-MAL (n = 3)
TM-MAL (n = 7)
11
1 2 1 1 1 1 2 0
2
10 10 8 8 0 10
2
4
3 2 4 2 4
30
2
20
CD7+ MllL 10. I
Downloaded by [UNSW Library] at 00:50 13 August 2015
redundant with the immunophenotypic partition, since CD7+/CD10- blasts were mostly observed in TM-MAL.
DISCUSSION The use of monoclonal antibodies against cell surface antigens has proven of great value in distinguishing various functional and/or differentiation stages of normal as well as malignant cells. According to the patterns of antigen expression, even morphologically uniform blast cell populations can now be subdivided into immunological subtypes according to their markers. In the vast majority of cases, all leukemic cells express the same phenotype. This is specially true for ALL, where cells are most likely to be frozen at a given stage of the maturation process. In acute myeloid leukemias (AML) the expression of myeloid
markers is more heterogeneous, and has led to the definition of lower positivity thresholds in order to establish the cell immunophenotype’6*’’. Apart from these “common” types of leukemias, more polymorphic proliferations have been described and in some cases, two distinct proliferating populations can be identified. These ALL are called double leukemias, and both the complementary percentages of positive labelling and the negativity of double-labelling allow one to define them. Other leukemias coexpress markers from several hematopoietic lineages. In these multiphenotypic leukemias, high percentages of blasts are stained by various markers, and double labelling, when applicable confirms their concomitant presence on the cells’ membrane”, 6-20.
’
im
CB lot
90
mi
80. ..-I-
70 bo..
54.. 40,.
.--t
pr0.003 ---!
A-,-.
XI C --.-r
M I
1 5
*
--r
-
c) --c
10-
mi
f
10
.
10
- - a _ -
1
15
20 HORCHS 25
Figure 1 Comparative event-free survival probability (Logrank test) of the three types of MAL described: BM (B-lineage and myeloid markers) ( x ), BT (B- and T-lineage markers) (full @), TM (T-lineage and myeloid markers) (open 0).
4
Downloaded by [UNSW Library] at 00:50 13 August 2015
494
P. J. M. PHILIP et a/.
In this study, the number of patients with leukemic blasts displaying multiphenotypic characteristics was low (2%), mostly because of the definition criteria we employed. Well recognized associations such as CD19 and CD2 in ALL, or CD7 in AML were not included. Moreover, the presence of a small subset of blast cells expressing myeloid antigens, usually considered as a good indicator of multilineage, was not regarded as acceptable in this study which dealt only with true co-expression of markers evident on the majority of blast cells. Nevertheless, the frequency of the immunological subtypes observed in our series is still very similar to that reported by Sobol et a/.1 8 . The B/myeloid subtype was confirmed as the most important, 62% in our study ( 5 5 % in Sobol’s series), while the incidence of the T myeloid subtype was higher than in Sobol’s series (27% and 19% respectively). The distribution of MAL subtypes within the FAB classification appears unclear in the literature. Mirro et a/. ” and Haas et a/.’O have reported MAL to occur mostly within the L1 FAB subtype, as we encountered in our study, while Sobol et a/.” reported a majority within the L2 FAB subclass. Our results also confirm that MAL may occur among AML, despite the fact that all BT cases had been classified as ALL. The investigation of clinical and biological parameters of MAL revealed no specific feature associated with a given MAL subtype. This is consistent with the observations reported by Foon rt al.’‘. MAL, however, do appear as acute leukemias with low WBC and low hemoglobin levels. The prognosis of MAL is better in our series than in that reported by Akashi et u / . ’ ~although , remission was not obtained in quite a substantial number of cases. Mirro et al. l 7 reported a high frequency of complete remission with conventional induction therapy in MAL (85% vs 90 in our patients), but this remission rate was much lower in the series reported by Smith et a/. (22%)16 and Sobol r t d.(49%)l8. We also observed a significant relationship between the cells’ surface phenotypes and the actuarial survival of MAL, associating CD7+ or CD10- MAL with poor survival. This result supports previous reports suggesting the bad prognosis associated with the CDl9+/CDlO- phen~-types’~-~’,and the more numerous studies demonstrating the positive value of CDlO expression on ALL4.24.In the study reported here, CDlO expression appeared more important than the unexpected presence of one or several myeloid markers on blasts also expressing B-lineage surface antigens. Conversely, the presence of T-lineage
markers was associated with a poorer prognosis in TM but not in BT patients, suggesting that the co-expression of myeloid markers on these cells did not improve the response to therapy. It is also worth noting the fact that the classical prognostic difference between adults and children was not confirmed in this series where the immunophenotype was more discriminant. In summary, this study stresses the importance of a complete immunophenotyping of leukemic cells, since such lineage infidelities may go unnoticed if too small a panel of markers is used. Acknowledgernent The GEIL is supported by a grant from the Association pour la Recherche contre le Cancer (ARC).
REFERENCES 1. Greaves. M. F. 11982) Target cells, cellular phenotypes, and
2. 3.
4.
5.
6.
7. 8.
9.
10.
lineage fidelity in human leukemia. In Cellulirr cmrl Molec,ulirr Biology of Heniopoietic Srem Cell D(fliwtiricrtioti. Hotiej, Hurbour Syniposiuni. edited by T. W. Mak and E. A. McCulloch, pp. 113-121. New-York: Alan R. Liss. Seligmann, M., Vogler, L. B., Preud’Homme, J. L., Guglielmi, P. and Brouet, J. C. (1981) Immunological phenotypes of human leukemias of the B-cell lineage. Bloocl C d s . 7 , 237 46. Schlossman, S. F., Chess, L., Hurnphreys, R. E.. and Strominger, J. L. (1976) Distribution of la-like molecules on the surface of normal and leukemic human cells. Proceeclings ?/’the ~ N l i O l l d Ac(rc1eniy qJ‘ Sciences USA. 73, 1288-92. Sallan, S. E.. Ritz, J.. Pesando. J.. Gelbcr, R., O’Brien, C., Hitchcock, S., Coral, F., and Schlossmann. S. F. (1980) Cell surface antigens: prognostic implications in childhood acute lymphoblastic leukemia. Blood. 55, 395-402. Garand. R., Vannier, J. P., Bent., M. C., Faure. G., and Bernard. A. (1989) Correlations between Acute Lymphoid Leukemia (ALL) lmmunophenotype and Clinical and Laboratory Data Presentation: a study of 350 patients. Cancer. 64. 1437-46. Bernard, A., Boumsell, L., and Hill. C. (1984) Joint report of the first international workshop on human leukocyte differentiation antigens by the investigators of the participating laboratories. In Leucocyte ryping. edited by A. Bernard, L. Boumsell, J. Dausset, C. Milstein and S. F. Schlossmann. pp. 9- 143, Berlin, Heidelberg, New York, Tokyo, Springer. Gale, R. P. and Ben Bassat, I. (1987) Hybrid acute leukemia. British Journal qf Haernatokq,. 65. 26 1-264. Yasukawa. M., Kanenitsu, N., Inatsuki, A,. Mise, H.. Iwamasa. K., and Fujita. S. (1989) Multiphenotypic lymphoma with rearrangements of the T cell receptor beta chain and gamma chain genes. Acto Haeniirrologictr ( B i r d ) . 81. 143-147. Uckun, F. M., Dibirdick. I., Smith, R., Tuel-Ahlgren. L.. Chandan-Langlie, M.. Schieven, G . L., Waddick. K. G.. Hanson. M.. and Ledbetter. J. A. (1991) Interleukin 7 receptor ligation stimulates tyrosine phosphorylation. inositol phospholipid turnover and clonal prolifaration of human B-cell precursors. Proceedings o/’ the Niriiotiid Ac~atletiiy of S ~ ~ i c w e s USA. 88, 3859-3893. Uckun, F. M., Muraguchi, A,, Ledbetter, J. A,, Kishimoto. T.. O’Brien, R. T.. RoloK J. S.. Gajal-Peczalska, K., and Provisor Koller. B. (1989) Biphenotypic leukemic lymphocyte precursors in CD2+/CD19- acute lymphoblastic leukemia and their
Downloaded by [UNSW Library] at 00:50 13 August 2015
MULTIPHENOTYPIC ACUTE LEUKEMIA putative normal counterparts in human fetal hematopoietic tissues. Blood, 73, 100&15. 1 I . Kurtzberg, J., Waldmann, T. A., Davey, M. P., Bigner, S. H., Moore, J. 0..Hershfield, M. S., and Haynes, B. F. (1989) CD7 +, CD4-, CD8- acute leukemia: a syndrome of malignant pluripotent lymphohematopoietic cells. Blood, 73, 381 -90. 12. Schaison, G., Leverger, G., Bancillon, A,, Marty, M., Olive, D., Cornu, G., Griscelli, C., Lemerle, S., Harrousseau, J. L., Bonnet, M., Freycon, F., Dufillot, D., Demeocq, F., Bauters, F., Lamagnere. J. P., and Taboureau, 0. (1987) Intermediate risk childhood acute lymphoblastic leukemias: ansacrine + vincristine versus monthly pulses for maintenance. Hiimarologie und Blurtransfusion, 30, 461-5. 13. Maurus, R., Boilletot, A,, Otten, J., Philippe, N., Bensit, Y., Behar. C., Casteels van-Daele, J., Chantraine, J. M., Delbecke, M. J., Gyselinck, H., Hainaut, H., Lutz. P..Plouvier, E., Robert, A., Sauveur. E., Solbu, G., Souillet, G., Sucui, S., and (EORTC Children’s Leukemia Cooperative Study Group) (1987) Treatment of acute lymphoblastic leukemia in children with the BFM protocol: a cooperative study and analysis of prognostic factors. Hanzatologie und Blurtransfusion, 30, 46670. 14. Fiere, D., Archambaud, E., Extra, J. M., Marty, M., David, B., Witz, F., Sotto, J. J., Rochant, H., Gastaut, J. A,, and LePrise, P. Y. (1987) Treatment of adult acute lymphoblastic leukemia. Preliminary results of a trial from the French Group. Hamatoloyie und Bluffransfusion.3, 125-9. 15. Bennet, J. M., Catovsky, D., Daniel, M. T., Flandrin, G., Galton, D. A. G., Gralnick, H. R., and Sultan, C. (1981) Proposals for the classification of the acute leukemias. French American British (FAB) cooperative group. British Journal of Haematology, 47, 553-61. 16. Smith, L. J., Curtis, J. E., Messner, H. A., Senn, J. S., Furthmayr, H., and McCulloch, E. A. (1983) Lineage infidelity in acute leukemia. Blood, 61, 1138-45. 17. Mirro, J., Zipf, T. F., Pui, C. H., Kitchingman, G., Williams, D., Melvin, S., Murphy, S. B., and Stass, S. (1985) Acute mixed
495
lineage leukemia: clinicopathological correlations and prognostic significance. Blood, 66, I 1 15-23. 18. Sobol, R. E., Mick, R., Royston, I., Davey, F. R.,Ellison, R. R., Newman, R.,Cuttner, J., Griffin, J. D., Collins, H., Nelson, D. A., and Bloomfield, C. D. (1987) Clinical importance of myeloid antigen expression in adult acute lymphoblastic leukemia. New England Journal of Medicine, 316, 11 11-7. 19. Pechansky, L., Kaplan, S. S., and Krause, J. R. (1990) Multiple lineage reactivity in childhood leukemia. Pediatric Pathology, 10, 217-29. 20. Haas, 0. A,, Bettelheim, P., Schmidtmeier, W., Gadner, H., Ludwig, H., Majdic, O., and Schultz, U. (1985) Sq-chromosome in acute leukemia with lymphoid morphology and expression of myeloid membrane determinants. Blood, 65, 1342-48. 21. Foon, K. A,, Billing, R.,and Terasaki, P. 1. (1980) Dual B and T markers in acute and chronic lymphocytic leukemia. Blood, 55, 1 6 2 0 . 22. Akashi, K., Harada, M., Shibuya, T., Morioka, E., Okamura, T., Asano, Y., Taniguchi, S., Teshima, T., Kikuchi, M., and Niho, Y. (1990) Clinical characteristics of hybrid leukemia: report of 5 cases. Leukemia Research, 14, 145-53. 23. Vannier, J. P., Bent, M. C., Faure, G . C., Bastard, C., Garand, R., and Bernard, A. (1989) Investigation of the CDlO (calla) negative acute lymphoblastic leukaemia: further description of a group with poor prognosis. British Journal of Haemarology, 72, 156-60. 24. Kersey, J., Goldman, A., Abramson, C., Nesbit, M., Perry, G., Gajli-Peczalska, K., and LeBien, T. (1982) Clinical usefulness of monoclonal antibody phenotyping in childhood lymphoblastic leukaemia. The Lancet, 2, 1419-23. 25. Crist, W., Pullen, J., Boyett, J., Faletta, J., VanEys, J., Borowitz, M., Jackson, J., Dowell, B., Russell, C., Quddus, F., Ragab, A,, and Vietti, T. (1988) Acute lymphoid leukemia in adolescent; clinical and biologic features predict a poor prognostic. A pediatric oncology group study. Journal of Clinicul Oncology, 6, 3 4 4 3 .
Leukemia & Lymphoma Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ilal20
Multiphenotypic Acute Leukemias: Clinicopathologic Correlations and Response to Therapy ac
b
c
c
Patrick J. M. Philip , Fabrice Monpoux , Isabelle Sudaka , Francoise Baudouin , Jacques c
d
Bayle , Marie C. Bene & Gilbert C. Faure
de
a
“Unité de Biologie Cellulaire et Tissulaire”, Hôpital Pasteur, France
b
Clinique Médicale Infantile, Hôpital de Cimiez, France
c
Laboratoire Central d'Hématologic, Hopital de Cimiez, CHR, Nice, France
d
Laboratoire d'Immunologie, Fac. Med. & CHRU, Nancy, France
e
Groupe d'Etude Immunologiques des Leucemies (GEIL)*, France Published online: 01 Jun 2015.
To cite this article: Patrick J. M. Philip, Fabrice Monpoux, Isabelle Sudaka, Francoise Baudouin, Jacques Bayle, Marie C. Bene & Gilbert C. Faure (1992) Multiphenotypic Acute Leukemias: Clinicopathologic Correlations and Response to Therapy, Leukemia & Lymphoma, 7:5-6, 489-495, DOI: 10.3109/10428199209049806 To link to this article: http://dx.doi.org/10.3109/10428199209049806
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
0 1992 Hanvood
Leukemia and Lymphoma, Vol. I , pp. 4 8 9 4 9 5 Reprints available directly from the publisher Photocopying permitted by license only
Academic Publishers GmbH Printed in the United Kingdom
Multiphenotypic Acute Leukemias : Clinicopathologic Correlations and Response to Therapy
Downloaded by [UNSW Library] at 00:50 13 August 2015
PATRICK J. M. PHILIP’v3, FABRICE MONPOUX’, ISABELLE SUDAKA3, FRANCOISE BAUDOUIN3, JACQUES BAYLE3, MARIE C. BENE4, GILBERT C. FAURE4 and the GEIL5 1. “Unite de Biologie Cellulaire et Tissulaire”, Hbpital Pasteur, 2. Clinique Medicale Infantile, Hbpital de Cimiez, 3. Laboratoire Central d’Htmatologie, Hbpital de Cimiez, CHR, Nice, 4. Laboratoire d’lmmunologie, Fac. Med. & CHRU, Nancy, and 5. Groupe d’Etude Immunologiques des Leucemies (GEIL)*. France.
(Received I6 September 1991; injinal form I6 February 1992)
Multiphenotypic acute leukemias (MAL), defined by the coexpression on most blast cells of antigens classically attributed to different lineages, remain a rare event. We isolated a series of 26 such cases from a cohort of 1565 leukemic patients whose cells were immunophenotyped at diagnosis. Markers of B and myeloid lineage (BM) were associated in 16 cases (62%0),3 coexpressed B and T markers (BT), and T-cell and myeloid antigens (TM) were found in 7 (27%). A tumoral syndrome was observed in 69% of the patients, without significant differences between the immunophenotypic subgroups. Median event free survivals in the three immunophenotypic subgroups as defined were respectively 24 months for BM-MAL, 4 months for TM-MAL and 7 months for BT-MAL respectively. The poorer prognosis of TM-MAL was significantly different from that of BM-MAL (p < 0.001). This concurred with the poorer prognosis associated with CD7 expression or absence of CD10, both characteristic features of TM-MAL. KEY WORDS:
Leukemia,
Lineage infidelity,
Address for correspondence: Dr P. J. M. Philip, Secretariat du GEIL, Laboratoire d’lmmunologie, Faculte de Medecine, 54500 Vandoeuvre les Nancy, France. * J. Gardais (Angers), M. C. Leglise, G. Lecalvez (Brest), M. Bernier, M. Massy (Bruxelles), A. M. Griveau, P. Boutard, C. Lebrun (Caen), J. Bonhomme, B. Chassagne, F. Demeocq, C. Rapatel (ClermontFerrand), H. Jouault (Creteil Mondor), F. Lelong (Creteil Intercommunal), E. Solary, 0. Casasnovas, M. Maynadit (Dijon), J. C. Bensa, M. Favre, M. C. Jacob (Grenoble), C. Sartiaux (Lille), M. Drouet (Limoges), L. Campos, A. Larese (Lyon), P. Poncelet, C. Brunet (Marseille), T. Lavabre-Bertrand, J. Taib (Montpellier), G . Kandel, G. Jung (Mulhouse), R. Garand (Nantes), P. Philip, F. Monpoux, F. Baudouin, 1. Sudaka, C. Caldani, G. Pomier, A. Bernard (Nice), M. Reguer, H. Schill (Paris Val de Grlce), G. Schaison, L. Boumsell (Paris Saint Louis), D. Vanhaecke, L. Douay (Paris Trousseau), A. Brizard (Poitiers), J. P. Roualet, T. Tabary (Reims), N. Genetet (Rennes), 0. Lees, B. Lenormand, J. P. Vannier (Rouen), J. C. Le Petit, S. Acquart, 0. Sabido (Saint-Etienne), A. Falkenrodt (Strasbourg), J. L. Bremond, A. M. Masy (Tours), F. Roubinet, E. Kuhlein (Toulouse), M. C. Bene, G. Faure, P. Bordigoni, F. Witz (Vandoeuvre les Nancy), C. Bayle, J. Antonini (Villejuif).
Immunophenotype,
Prognosis.
INTRODUCTION Several authors have stressed that lineage fidelity is maintained in leukemic blasts, and that their phenotypes are the counterparts of those expressed by normal cells during early stages of lineage-specific differentiati~nl-~. These observations have been used to support pathogenetic models of leukemic proliferation postulating that cell-differentiation in leukemia is not abnormal but blocked or abortive. To complement morphological classifications, immunological classifications based on differentiation patterns of each hematopoietic lineage have been proposed, allowing the identification of leukemia subsets which respond differently to the rap^^.^. However, in a minority of patients, blasts appear to co-express markers of more than one differentiation lineage, as
489
Downloaded by [UNSW Library] at 00:50 13 August 2015
490
P. J. M. PHILIP et al.
defined by the consecutive workshops on leukocyte differentiation6. Various terms have been used to designate these unusual acute leukemias showing both myeloid and/or lymphoid features: hybrid, biphenotypic, biclonal, bilineal, chimaeric, mixed lineage switch or infidelity lineage.. .7. We have preferred to use the term multiphenotypic acute leukemias (MAL), previously used by others'.' to describe both intralymphoid infidelity (coexpression of B and T cell markers) and interlineage infidelity (coexpression of myeloid and lymphoid markers). We investigated the characteristics of 26 cases of MAL observed among a collaborative series of more than 1500 leukemias collected by the Groupe d'Etude Immunologique des Leucemies (GEIL). In this report we describe the various phenotypes observed, the clinical and biological characteristics of the patients, and discuss the evolution of MAL.
MATERIALS AND METHODS Patients Information at presentation was collected for all patients from the 32 centers who work together in the Groupe d'Etude Immunologique des Leuckmies (GEIL). A total number of 1565 malignant hemopat hies immunophenotyped at diagnosis was entered in the collective file between November 1983 and July 1991. The group of patients reported here included 26 patients with MAL, diagnosed and phenotyped at first involvement. Multiphenotype was defined as the co-expression of at least one specific marker of two distinct hematopoietic lineages (Table 1). Only samples containing more than 60% of blasts and less than 20% lymphocytes were included in this selection. Individual leukemias were entered as MAL only if the coexpression of markers was observed on more than 50% of blasts, i.e. when the percentage of two individual markers added up to more than 100% of the cells. Positivity thresholds for other markers were 20% for myeloid lineage antigens (CDllb, CD13, CD14, CD15, CD33), and 40% for lymphoid lineage markers (sIg. c-mu, CD19, CD20, CD21, CD24, CD1, CD2, CD3, CD4, CD5, CD7, CD8). Previously reported lineage infidelities, considered to be compatible with normal counterparts, i.e. CD19/CD21° (n = 30) or CD7/myeloid markers'' (n = 35) were not entered in this series. Multiphenotypic leukemias, as classically defined by the presence of myeloid markers on more than 20% of
Table 1 Monoclonal antibodies used for MAL immunophenotyping T-lineage CDl CD2 CD3 CD4 CD5 CD7 CD8
OKT6, T6" OKTll, Tll", Leu5 OKT3, T3", Led, IOT3 OKT4, T4", Leu3,10T4 TI", Leul, lOTla 121, Leu9, IOT7 OKT8, TS", Leu2, IOT8
Myeloid lineage CDllb CD13 CD14 CD15 CD33
M01" MY7", MCS2 M02", MY4", 20.3 LeuM1, ION1 MY9"
B-lineage CD19 CD20 CD2 1 CD24 Non lineage specific CDlO HLA DR
B4"
B1" B2", lOBla BA 1
J5", lOT5 OKDR, OKIal, I2", IOT2a
OK series: Ortho Diagnostics, Raritan, NJ, USA; 'Coulterclone, Coultronics, Hialeah, FL, USA; Leu series: Becton Dickinson, Mountain View, CA, USA; IOT series: Immunotech. Luminy,France.
blasts, or of lymphoid markers on more than 40% were not taken into account since double-labelling had not been performed in this retrospective study. Only leukemias with indisputable co-expression based on added percentages were retained. The patients were treated with different therapeutic protocols, depending on the policy of the hospital to which they reported. The most frequent protocols used in children were the successive versions of the FRALLE12 and protocols similar to the BFM13. Adults were mostly treated according to LALA protocol^'^. Complete remission was evaluated according to the CALGB criteria: normal clinical features, normal blood cell counts and bone marrow smears containing less than 5% of blast cells for more than a month. Methods Bone marrow (70% of the cases) or peripheral blood (30%) was collected in heparin at the time of diagnosis, before therapy. Classical techniques were used in all laboratories. Each sample was centrifuged on density gradient (Ficoll). A differential count of collected leukemic cells was done on a cytospin smear. The FAB subtype was determined by standard
Downloaded by [UNSW Library] at 00:50 13 August 2015
MULTIPHENOTYPIC ACUTE LEUKEMIA
criteria”. In addition, cytological studies included staining for the following activities: myeloperoxidase, chloroacetate esterase, Periodic Acid Schiff (PAS), naphthol acetate esterase. Surface immunoglobulins (SIg +) were usually identified in direct immunofluorescence, using fluorescent F(ab); anti-human immunoglobulins. The presence of intracytoplasmic immunoglobulins (cIg+) was investigated in direct immunofluorescence on cytospin smears. Other membrane molecules were identified using the monoclonal antibodies listed in Table 1, in direct or indirect immunofluorescence using FITC-sheep or goat antimouse Ig serum as second step reagent. Microscopes equipped with Ploem systems of epi-illumination were used to enumerate positive cells in each sample until 1988; since then most centers have been using flow cytometry. Results were expressed as the percentage of fluorescent cells for each marker. Several internal workshops were performed within the GEIL, where the same frozen cells were tested by all participants. Specific instructions were given for thawing. After determination of the phenotypes, the codes and relevant data were collected and compared. An intergroup reproducibility score of 94% was obtained, allowing a multicenter analysis of patients’ data. Student’s t test and Fischer’s exact test were used for the statistical analyses performed, with the Myosotis (Coultronics France, Margency, France) software. Survival curves were plotted using the Logrank test.
RESULTS Immunological phenotypes Table 2 shows the individual phenotype of each of the 26 MAL analyzed. Three types of combinations between B, T and myeloid-specific markers were observed. Blasts from 16 MAL patients were of B and myeloid cell lineage (BM), 7 had markers of both T and myeloid cell lineage (TM), and 3 co-expressed immunological markers of B and T cell lineage (BT). In one of the TM patients, CD19 was expressed on blast cells as well as several T and myeloid markers. Addition of the percentages of cells labelled by markers from different lineages yielded values ranging between 130 and 181 (mean: 159.6).HLA class I1 DR antigens were expressed in 65% of the cases, mostly in BM-MAL (%YO),seldom in TM (29%) and never in BT-MAL.
49 I
Similarly, CDlO appeared to be mostly expressed in BM-MAL (81%), and was only found once in BT-MAL. Predictably, CD19 and CD24 were the most frequently found B-cell markers, CD7 the most frequent T-lineage antigen, and CD33 the predominant indicator of myeloid lineage. CD19 was the only B-lineage marker in 4 cases (all BM-MAL), CD7 the only T-lineage marker in 2 of the 3 cases of BT-MAL, and CD33 the only myeloid marker in 6 cases (5 BM-MAL, 1 TM-MAL). Clinical and biological characteristics at diagnosis The series included 13 children (age range 7 months-16 years) and 13 adults (age range 17-85). There were 10 females and 16 males. Table 3 provides the classical features observed in the three immunophenotypic groups of patients at diagnosis. Tumoral syndrome (69%) and anemia (62%) were frequent. Mediastinal enlargement was surprisingly infrequent both in TM (2/7) and BT (1/3) MAL cases. Hyperleukocytosis was a rare feature in all groups. Statistical comparisons of the frequency of each of these features showed no significant difference between sub-groups. The distribution of MAL in the morphological FAB subtypes showed the predominance of L1 (50%) and L2 (35%) ALL. None of the BM-MAL were classified as AML according to morphological criteria, while 3 of the 7 TM-MAL were. Prognostic features As in other studies published by our group, the variety of therapeutic protocols used enabled comparison of the influence of phenotypic features while minimizing the specific bias of a single protocol. Initial remission was obtained in 69% of the patients (12 children and 6 adults), and the death rate was 42%. The median follow-up for the entire group was 16 months (range 1-48). Survival analyses were thus performed for a period of 24 months after diagnosis, yielding a satisfactory confidence interval. Survival estimates were performed as disease free survival, without relapse or death in remission. There was no significant difference in the outcome of males and females, or, more surprisingly, adults and children. As seen in Figure 1, the prognosis was significantly worse for TM-MAL patients. Significance was also reached to demonstrate the good prognostic value of CD7- or CD10+ blasts. These features possibly are
LI LI L1 L1 L2 L2 L1 L1 L1 L2 L2 L1 L1 L2 L2
L1
L1 L2 L1
L1
BM BM BM BM BM BM BM BM BM BM BM BM BM BM BM BM
BT BT BT
TM TM TM TM TM TM TM
LO MI M4 L2 L2
M1
FAB
MAL
0
0 0
0 3
5
0 0
25 4 0
0 0
0 70 0
0
0 30
2
0 3 13
0 0
clg
0
slg
5 2 0 0 0
1
33 0 41
65 9
15 1
6 50 62 13 2
1
72 9 25 37 13 0
CDZO
0
0
0
1
65 4
1
0 89 98
97 90 36 87 82 77 89 94 97 74 74 58 88 89 60 83
CDI9
7 0
61 91 85
2
1
0
0
20
0
CDZI
Table 2 Individual immunophenotypic features
15 0
1
88 96
67 94 98 88 70 82
98
0 84
24 94 85
CDZ4
0 1 0 90 0
4
85 0 0
4
1 1
5
0
CDI
9
42 5 0 90 7
2
97
92 1 4
19
18
12 6 3 1 3 2 7 17 3 3 9
0 6
CDZ
1 6 80 6 90 0 0
0
1
18
4 0
3 88 0 0
3 26
4 4
1 1 10 12
1
2 2 12 17
1
4 0 8 0
0
0
2
CD4
6 2 13 5 4
0
CD3
5
0
97 58 43 7 8 0 70 0 9
4
81
15 17
4 2
15
7 2 15 12 2 1 3 2
CD5
9 0
0
0
2 0
79
4 7
1
1 0 1
8 2 4 3
CDB
95 3 69 70 44 6 2 0 80 90 0 0 6 6
93
2 1 2 1 13 15 3 2 9 16 20
8 2 14
CD7
20 0 0
80 88
4 0 0
92
8
77 62 2
5
CDIIh
Downloaded by [UNSW Library] at 00:50 13 August 2015
79 0 68 0 0 80
1
2 0 0
33 4 3
1
5
11
1
10 0 0 10 0
66 0
CD13
87 2 2 90 0
1
8 0 0
11
1
3
1
1 0
0
0
CDI4
0 0 0
77 10
0 0
96 68 1 4 2 0 75 65
92 1
69
CD33
1 65 0 2 93 70
3 0 0
8 53 84 50 0 70 6 4 83 0 44 8 80
CDIS
0
0 0
1
1
1
93 3 7
7 85 82 60 98 60 4 87 92 7 75 80
40
96 86 94
CDIO
77 83 87 85 39
94 35
5
CD34
2 69 35 86 0 9 0 0
43 2 0
65
1 4 4 19 0 0 30
95 96 69 87 47 33 84 98 97 70 80 88 92 91 66 80
DR
k
rp
-.
za
2
a
3:
%
a
MULTIPHENOTYPIC ACUTE LEUKEMIA
493
Table 3 Clinical and biological features at diagnosis; incidences are reported as number of cases
Children Males Enlarged lymph nodes Splenomegaly Hepatomegaly Mediastinal mass Anemia (Hb < 100 g/L) High WBC ( > 109/L)
BM-MAL (n = 16)
BT-MAL (n = 3)
TM-MAL (n = 7)
11
1 2 1 1 1 1 2 0
2
10 10 8 8 0 10
2
4
3 2 4 2 4
30
2
20
CD7+ MllL 10. I
Downloaded by [UNSW Library] at 00:50 13 August 2015
redundant with the immunophenotypic partition, since CD7+/CD10- blasts were mostly observed in TM-MAL.
DISCUSSION The use of monoclonal antibodies against cell surface antigens has proven of great value in distinguishing various functional and/or differentiation stages of normal as well as malignant cells. According to the patterns of antigen expression, even morphologically uniform blast cell populations can now be subdivided into immunological subtypes according to their markers. In the vast majority of cases, all leukemic cells express the same phenotype. This is specially true for ALL, where cells are most likely to be frozen at a given stage of the maturation process. In acute myeloid leukemias (AML) the expression of myeloid
markers is more heterogeneous, and has led to the definition of lower positivity thresholds in order to establish the cell immunophenotype’6*’’. Apart from these “common” types of leukemias, more polymorphic proliferations have been described and in some cases, two distinct proliferating populations can be identified. These ALL are called double leukemias, and both the complementary percentages of positive labelling and the negativity of double-labelling allow one to define them. Other leukemias coexpress markers from several hematopoietic lineages. In these multiphenotypic leukemias, high percentages of blasts are stained by various markers, and double labelling, when applicable confirms their concomitant presence on the cells’ membrane”, 6-20.
’
im
CB lot
90
mi
80. ..-I-
70 bo..
54.. 40,.
.--t
pr0.003 ---!
A-,-.
XI C --.-r
M I
1 5
*
--r
-
c) --c
10-
mi
f
10
.
10
- - a _ -
1
15
20 HORCHS 25
Figure 1 Comparative event-free survival probability (Logrank test) of the three types of MAL described: BM (B-lineage and myeloid markers) ( x ), BT (B- and T-lineage markers) (full @), TM (T-lineage and myeloid markers) (open 0).
4
Downloaded by [UNSW Library] at 00:50 13 August 2015
494
P. J. M. PHILIP et a/.
In this study, the number of patients with leukemic blasts displaying multiphenotypic characteristics was low (2%), mostly because of the definition criteria we employed. Well recognized associations such as CD19 and CD2 in ALL, or CD7 in AML were not included. Moreover, the presence of a small subset of blast cells expressing myeloid antigens, usually considered as a good indicator of multilineage, was not regarded as acceptable in this study which dealt only with true co-expression of markers evident on the majority of blast cells. Nevertheless, the frequency of the immunological subtypes observed in our series is still very similar to that reported by Sobol et a/.1 8 . The B/myeloid subtype was confirmed as the most important, 62% in our study ( 5 5 % in Sobol’s series), while the incidence of the T myeloid subtype was higher than in Sobol’s series (27% and 19% respectively). The distribution of MAL subtypes within the FAB classification appears unclear in the literature. Mirro et a/. ” and Haas et a/.’O have reported MAL to occur mostly within the L1 FAB subtype, as we encountered in our study, while Sobol et a/.” reported a majority within the L2 FAB subclass. Our results also confirm that MAL may occur among AML, despite the fact that all BT cases had been classified as ALL. The investigation of clinical and biological parameters of MAL revealed no specific feature associated with a given MAL subtype. This is consistent with the observations reported by Foon rt al.’‘. MAL, however, do appear as acute leukemias with low WBC and low hemoglobin levels. The prognosis of MAL is better in our series than in that reported by Akashi et u / . ’ ~although , remission was not obtained in quite a substantial number of cases. Mirro et al. l 7 reported a high frequency of complete remission with conventional induction therapy in MAL (85% vs 90 in our patients), but this remission rate was much lower in the series reported by Smith et a/. (22%)16 and Sobol r t d.(49%)l8. We also observed a significant relationship between the cells’ surface phenotypes and the actuarial survival of MAL, associating CD7+ or CD10- MAL with poor survival. This result supports previous reports suggesting the bad prognosis associated with the CDl9+/CDlO- phen~-types’~-~’,and the more numerous studies demonstrating the positive value of CDlO expression on ALL4.24.In the study reported here, CDlO expression appeared more important than the unexpected presence of one or several myeloid markers on blasts also expressing B-lineage surface antigens. Conversely, the presence of T-lineage
markers was associated with a poorer prognosis in TM but not in BT patients, suggesting that the co-expression of myeloid markers on these cells did not improve the response to therapy. It is also worth noting the fact that the classical prognostic difference between adults and children was not confirmed in this series where the immunophenotype was more discriminant. In summary, this study stresses the importance of a complete immunophenotyping of leukemic cells, since such lineage infidelities may go unnoticed if too small a panel of markers is used. Acknowledgernent The GEIL is supported by a grant from the Association pour la Recherche contre le Cancer (ARC).
REFERENCES 1. Greaves. M. F. 11982) Target cells, cellular phenotypes, and
2. 3.
4.
5.
6.
7. 8.
9.
10.
lineage fidelity in human leukemia. In Cellulirr cmrl Molec,ulirr Biology of Heniopoietic Srem Cell D(fliwtiricrtioti. Hotiej, Hurbour Syniposiuni. edited by T. W. Mak and E. A. McCulloch, pp. 113-121. New-York: Alan R. Liss. Seligmann, M., Vogler, L. B., Preud’Homme, J. L., Guglielmi, P. and Brouet, J. C. (1981) Immunological phenotypes of human leukemias of the B-cell lineage. Bloocl C d s . 7 , 237 46. Schlossman, S. F., Chess, L., Hurnphreys, R. E.. and Strominger, J. L. (1976) Distribution of la-like molecules on the surface of normal and leukemic human cells. Proceeclings ?/’the ~ N l i O l l d Ac(rc1eniy qJ‘ Sciences USA. 73, 1288-92. Sallan, S. E.. Ritz, J.. Pesando. J.. Gelbcr, R., O’Brien, C., Hitchcock, S., Coral, F., and Schlossmann. S. F. (1980) Cell surface antigens: prognostic implications in childhood acute lymphoblastic leukemia. Blood. 55, 395-402. Garand. R., Vannier, J. P., Bent., M. C., Faure. G., and Bernard. A. (1989) Correlations between Acute Lymphoid Leukemia (ALL) lmmunophenotype and Clinical and Laboratory Data Presentation: a study of 350 patients. Cancer. 64. 1437-46. Bernard, A., Boumsell, L., and Hill. C. (1984) Joint report of the first international workshop on human leukocyte differentiation antigens by the investigators of the participating laboratories. In Leucocyte ryping. edited by A. Bernard, L. Boumsell, J. Dausset, C. Milstein and S. F. Schlossmann. pp. 9- 143, Berlin, Heidelberg, New York, Tokyo, Springer. Gale, R. P. and Ben Bassat, I. (1987) Hybrid acute leukemia. British Journal qf Haernatokq,. 65. 26 1-264. Yasukawa. M., Kanenitsu, N., Inatsuki, A,. Mise, H.. Iwamasa. K., and Fujita. S. (1989) Multiphenotypic lymphoma with rearrangements of the T cell receptor beta chain and gamma chain genes. Acto Haeniirrologictr ( B i r d ) . 81. 143-147. Uckun, F. M., Dibirdick. I., Smith, R., Tuel-Ahlgren. L.. Chandan-Langlie, M.. Schieven, G . L., Waddick. K. G.. Hanson. M.. and Ledbetter. J. A. (1991) Interleukin 7 receptor ligation stimulates tyrosine phosphorylation. inositol phospholipid turnover and clonal prolifaration of human B-cell precursors. Proceedings o/’ the Niriiotiid Ac~atletiiy of S ~ ~ i c w e s USA. 88, 3859-3893. Uckun, F. M., Muraguchi, A,, Ledbetter, J. A,, Kishimoto. T.. O’Brien, R. T.. RoloK J. S.. Gajal-Peczalska, K., and Provisor Koller. B. (1989) Biphenotypic leukemic lymphocyte precursors in CD2+/CD19- acute lymphoblastic leukemia and their
Downloaded by [UNSW Library] at 00:50 13 August 2015
MULTIPHENOTYPIC ACUTE LEUKEMIA putative normal counterparts in human fetal hematopoietic tissues. Blood, 73, 100&15. 1 I . Kurtzberg, J., Waldmann, T. A., Davey, M. P., Bigner, S. H., Moore, J. 0..Hershfield, M. S., and Haynes, B. F. (1989) CD7 +, CD4-, CD8- acute leukemia: a syndrome of malignant pluripotent lymphohematopoietic cells. Blood, 73, 381 -90. 12. Schaison, G., Leverger, G., Bancillon, A,, Marty, M., Olive, D., Cornu, G., Griscelli, C., Lemerle, S., Harrousseau, J. L., Bonnet, M., Freycon, F., Dufillot, D., Demeocq, F., Bauters, F., Lamagnere. J. P., and Taboureau, 0. (1987) Intermediate risk childhood acute lymphoblastic leukemias: ansacrine + vincristine versus monthly pulses for maintenance. Hiimarologie und Blurtransfusion, 30, 461-5. 13. Maurus, R., Boilletot, A,, Otten, J., Philippe, N., Bensit, Y., Behar. C., Casteels van-Daele, J., Chantraine, J. M., Delbecke, M. J., Gyselinck, H., Hainaut, H., Lutz. P..Plouvier, E., Robert, A., Sauveur. E., Solbu, G., Souillet, G., Sucui, S., and (EORTC Children’s Leukemia Cooperative Study Group) (1987) Treatment of acute lymphoblastic leukemia in children with the BFM protocol: a cooperative study and analysis of prognostic factors. Hanzatologie und Blurtransfusion, 30, 46670. 14. Fiere, D., Archambaud, E., Extra, J. M., Marty, M., David, B., Witz, F., Sotto, J. J., Rochant, H., Gastaut, J. A,, and LePrise, P. Y. (1987) Treatment of adult acute lymphoblastic leukemia. Preliminary results of a trial from the French Group. Hamatoloyie und Bluffransfusion.3, 125-9. 15. Bennet, J. M., Catovsky, D., Daniel, M. T., Flandrin, G., Galton, D. A. G., Gralnick, H. R., and Sultan, C. (1981) Proposals for the classification of the acute leukemias. French American British (FAB) cooperative group. British Journal of Haematology, 47, 553-61. 16. Smith, L. J., Curtis, J. E., Messner, H. A., Senn, J. S., Furthmayr, H., and McCulloch, E. A. (1983) Lineage infidelity in acute leukemia. Blood, 61, 1138-45. 17. Mirro, J., Zipf, T. F., Pui, C. H., Kitchingman, G., Williams, D., Melvin, S., Murphy, S. B., and Stass, S. (1985) Acute mixed
495
lineage leukemia: clinicopathological correlations and prognostic significance. Blood, 66, I 1 15-23. 18. Sobol, R. E., Mick, R., Royston, I., Davey, F. R.,Ellison, R. R., Newman, R.,Cuttner, J., Griffin, J. D., Collins, H., Nelson, D. A., and Bloomfield, C. D. (1987) Clinical importance of myeloid antigen expression in adult acute lymphoblastic leukemia. New England Journal of Medicine, 316, 11 11-7. 19. Pechansky, L., Kaplan, S. S., and Krause, J. R. (1990) Multiple lineage reactivity in childhood leukemia. Pediatric Pathology, 10, 217-29. 20. Haas, 0. A,, Bettelheim, P., Schmidtmeier, W., Gadner, H., Ludwig, H., Majdic, O., and Schultz, U. (1985) Sq-chromosome in acute leukemia with lymphoid morphology and expression of myeloid membrane determinants. Blood, 65, 1342-48. 21. Foon, K. A,, Billing, R.,and Terasaki, P. 1. (1980) Dual B and T markers in acute and chronic lymphocytic leukemia. Blood, 55, 1 6 2 0 . 22. Akashi, K., Harada, M., Shibuya, T., Morioka, E., Okamura, T., Asano, Y., Taniguchi, S., Teshima, T., Kikuchi, M., and Niho, Y. (1990) Clinical characteristics of hybrid leukemia: report of 5 cases. Leukemia Research, 14, 145-53. 23. Vannier, J. P., Bent, M. C., Faure, G . C., Bastard, C., Garand, R., and Bernard, A. (1989) Investigation of the CDlO (calla) negative acute lymphoblastic leukaemia: further description of a group with poor prognosis. British Journal of Haemarology, 72, 156-60. 24. Kersey, J., Goldman, A., Abramson, C., Nesbit, M., Perry, G., Gajli-Peczalska, K., and LeBien, T. (1982) Clinical usefulness of monoclonal antibody phenotyping in childhood lymphoblastic leukaemia. The Lancet, 2, 1419-23. 25. Crist, W., Pullen, J., Boyett, J., Faletta, J., VanEys, J., Borowitz, M., Jackson, J., Dowell, B., Russell, C., Quddus, F., Ragab, A,, and Vietti, T. (1988) Acute lymphoid leukemia in adolescent; clinical and biologic features predict a poor prognostic. A pediatric oncology group study. Journal of Clinicul Oncology, 6, 3 4 4 3 .
![[Hemostatic disorders in acute and chronic leukemias. Pathogenesis and therapy].](/assets/img/hold.png)
