Leukemia (1999) 13, 869–872  1999 Stockton Press All rights reserved 0887-6924/99 $12.00 http://www.stockton-press.co.uk/leu Deletions and losses in 5 or 7 in adult acute lymphocytic leukemia: incidence, associations and implications BS Dabaja1, S Faderl1, D Thomas1, J Cortes1, S O’Brien1, F Nasr1, S Pierce1, K Hayes2, A Glassman2, M Keating1 and HM Kantarjian1

Departments of 1Leukemia and 2Hematopathology, Section of , The University of Texas, MD Anderson Cancer Center, Houston, Texas, USA

Deletions or losses in chromosomes 5 or 7 are recurrent non- Table 1 Characteristics of the study group random abnormalities in (AML), and are associated with prior exposure to carcinogens or leukem- Characteristic Category No. of patients (%) ogenic agents, and with poor prognosis. Their occurrence and significance in adult acute lymphocytic leukemia (ALL) is not well described. The aim of the study was to evaluate the inci- 5 or P value dence, associations and implications of or 7 7 abnormality abnormalities in adult ALL. Patients with newly diagnosed ALL referred to MD Anderson Cancer Center between 1980 and 1996 Present Absent were analyzed. Characteristics and outcome of patients with or without chromosome 5 or 7 abnormalities were compared by Number 31 437 standard statistical methods. Thirty-one of 468 patients (6.6%) Age (years) у60 9 (29) 80 (18) 0.14 had chromosome 5 or 7 abnormalities. Loss of chromosome 5 Performance (Zubrod) у3 1 (3) 36 (8) 0.33 occurred in six cases, three of them had both chromosome 5 Hepatomegaly yes 3 (10) 96 (22) 0.1 and 7 abnormalities. or loss of Splenomegaly yes 9 (29) 141 (32) 0.71 occurred as a single abnormality in three patients; in 28 Hemoglobin (g/dl) Ͻ10 18 (58) 289 (66) 0.36 patients it was associated with other abnormalities. The most Platelets (×109/dl) Ͻ50 12 (39) 212 (48) 0.29 significant cytogenetic association was with the Philadelphia WBC (×109/dl) Ͼ50 22 (71) 357 (82) 0.14 chromosome (Ph) abnormality occurring in nine patients (29%). Marrow blasts % Ͼ50 30 (97) 427 (92) 0.38 Compared with patients without the abnormalities, patients with chromosome 5 or 7 abnormalities tended to express CD34 more frequently (74% vs 54% P = 0.07), to be older (age Ͼ60 years 29% vs 18% P = 0.14), and to be associated with Ph (29% vs 11% P = 0.004). With therapy, the complete response (CR) rate with chromosome 5 or 7 abnormalities was lower (64% vs Table 2 Immunophenotypic characteristics of patients with or 79% P = 0.038) but the survival rate was similar (3-year survival without chromosome 5 or 7 abnormalities rate 32% vs 36% P = 0.14). When the 22 patients without Ph were considered separately, the CR and survival rates were No. of patient/Total analyzed (%) similar among patients with or without chromosome 5 or 7 abnormalities. Abnormalities in chromosome 5 or 7 are not spe- Chromosome 5 or 7 P value cific for AML, and may occur in ALL. Unlike in AML, chromo- abnormalities some 5 or 7 abnormalities in ALL were not predictive of worse prognosis, which is accounted for mostly by the association Present Absent with Ph. Keywords: acute lymphocytic leukemia; cytogenetic abnormalities; chromosomes 5 and 7 (A) Specific marker CD 10 (CALLA) 17/24 (70) 209/305 (68) 0.81 CD 19 15/20 (75) 215/280 (76) 0.85 CD 34 17/23 (74) 152/280 (54) 0.07 Introduction T markers 5/24 (21) 73/347 (21) 0.98 Myeloid markers 12/23 (52) 131/289 (45) 0.52 Cytogenetic studies in de novo acute leukemia have identified nonrandom chromosomal abnormalities to be associated with (B) Immunophenotype significant differences in patient prognosis. Deletions or losses Mature B cell 1/24 (4) 41/347 (12) 0.25 Pre B 3/24 (12) 29/347 (8) 0.41 in chromosomes 5 or 7, alone or with other changes, have CALLA-positive 13/24 (54) 179/347 (52) 0.73 been found in a variety of hematological disorders, including T cell 5/24 (21) 73/347 (21) 0.98 acute myelogenous leukemia (AML) and myelodysplastic Null 2/24 (8) 25/347 (7) 0.84 syndromes (MDS), and are associated with prior exposure to carcinogens and with poor prognosis.1–6 Little is known about the prognostic significance of chromo- the incidence and implications of chromosome 5 or 7 some 5 or 7 abnormalities in ALL. Their presence may intuit- abnormalities in adult ALL. ively imply poor prognosis, as suggested by their prognostic effect in AML and MDS. However, such prognostic associ- ations have not been described in ALL. Herein, we report on Materials and methods

Study group Correspondence: HM Kantarjian, MD Anderson Cancer Center, Department of Leukemia, Box 61, 1515 Holcombe Blvd, Houston, TX Four hundred and sixty-eight patients with newly diagnosed 77030, USA; Fax: 1 713 792 2031 untreated ALL, 15 years and older, who were referred to our Received 18 November 1998; accepted 23 February 1999 institution between 1980 and 1996, were reviewed. The Insti- Chromosome 5 or 7 abnormalities in ALL BS Dabaja et al 870 Table 3 Detailed karyotypic analysis in the study group

Patient No.

1 47,XX,−7,t(9;22)(q34;q11), +17,−21,+der(22)t(9;22),+mar[13] 47,XX,del(7)(q11),t(9;22)(q34;q11),−21,+der(22)t(9;22),+2mar[7] 46,XX[2] 2 46,XX,inv(7)(p21q23),t(9;22)(q34;q11)[21] 46,XX,inv(7)(p23q23)[4] 3 67–69,XY,−3,−5,−5,−6,−7,+8,+8,+8,+9,−10,+12,−13,−14,−15,−16,−17,+18,−19,+21, +21,−22,+2−6mar[cp5] 46,XY[15] 4 45,XY,−7,add(16)(p13.3)[21] 45,X,−Y,−7,add(16)(p13.3),+mar[2] 5 45,XX,−7,t(9;22)(q34;q11)[25] 6 47,Y,−X,−5,−7,+13,−14,−15,+5−6mar[8] 84–91,YY,−X,−X,−4,−4,−5,−5,−7,−7,i(8)(q10),−14,−14,−15,del(22)(q13),+8,−10mar[cp11] 46,XY[1] 7 47,XY,−7,−15,+19,add(22)(p11),+2mar[10] 47,XY,−7,add(9)(p13),−15,+19,add(22)(p11),+2mar[2] 46–47,XY,−4,−6,−7,−15,+19,add(22)(p11),+2–3mar[cp7] 46,XY[1] 8 45,XX,−7[4] 46,XX[21] 9 45,XX,add(6)(p25),−7,dup(7)(q21q32),t(9;22)(q34;q11)[25] 10 46,XY,del(9)(p23)[25] 46,XY,−5,−7,der(7)t(7;?)(p22;?)del(7)(q31.3),+mar[19] 46,XY[15] 11 43–45,XX,−7,der(9)inv(9)(p11q12)c t(9;22)(q34;q11),add(19)(p13.3)[cp7] 12 45,XY,−7[11] 46,XY [14] 13 79–81,XXYY,−2,add(2)(q24)×2,−4,−5,−6,−7×2,−8,−9×2,−11x2,−12,−13,−15,−16,−17,−20×2,+22,+3−6mar[cp5] 46,XY[7] 14 47,XY,inv(7)(p13p22),−7,der(9)del(9)(p13)t(9;22)(q34;q11),−19,+3mar[22] 46,XY[3] 15 37,X,−Y,−2,−3,−4,−7,−9,−12,−13,−17[5] 46,XY[13] 16 41,XY,del(3)(p13),−5,−7,−16,−17,−20[12] 46,XY[13] 17 45,XY,−7,add(13)(p10)[17] 46,XY[2] 18 45,XX,−7[8] 46,XX[2] 19 46,X,−Y,add(1)(p36),del(9)(q12),del(7)(q32),−13,−13,add(14)(q32), add(16)(p13.3),−17,+add(22)(q13),+3mar[3] 46,XY[22] 20 46,XX,add(1)(p36),add(5)(q35),del(7)(q11.23),del(8)(q13),del(9)(p13), t(10;?)(p15;?),del(11)(q21),del(12)(p12),−13,add(15)(q26),add(19)(p13.3),+mar[9] 46,XX[41] 21 78,XXX,+X,+1,+2,−3,+4,+5,+6,−7,+8,del(9)(p13),+11,+12,+13,+14,−15,−16,−17,−19,+21,+22,+2mar[19] 22 46,XY,del(3)(q21),del(7)(q31),−9,del(9)(p21),del(12)(p11),add(20)(q13),+mar[25] 23 46,XX,+3,−5,−8,−10,t(14;18)(q32;q21),t(17;?)(p13;?),+2mar,ෂ42dmin[34] 24 46,XXYc,del(7)(q22),t(19;?)(p13;?),−20[19] 47,XXYc[2] 25 45,XY,−7,t(7;?)(q36;?),t(9;22)(q34;q11),del(20)(q13)[5] 46,XY[12] 26 39,XX,−3,−7,−13,−14,−15,−16,−17[18] 35–40,XX,−3,−7,−13,−14,−15,−16,−17,−18, +mar[cp6] 46,XX[1] 27 45,XY,−7,t(9;22)(q34;q11)[25] 28 43–44,XX,add(4)(p16),−5,t(10;11)(p15;q13),del(17)(p11.2),−19,+mar[cp27] 46,XX[6] 29 46,XX,−7,t(9;22)(q34;q11),−22,+2mar[18] 46–50,XX,−7,t(9;22)(q34;q11),−22,+2mar[cp7] 30 46,XX,del(7)(q21)[25] 31 44–46,XX,−5,+mar[cp4] 46,XX[3]

tutional Review Board approved the studies at MD Anderson evaluated using cytochemical staining procedures including Cancer Center. Samples were obtained with the informed myeloperoxidase (MPO), terminal deoxynucleotidyl transfer- consent of patients. ase (TdT), non-specific esterase, and periodic-acid schiff The diagnosis was based on morphological analysis of bone (PAS). The diagnostic evaluations were complemented by marrow aspirates and biopsies according to the French–Amer- immunophenotypic and flow cytometric studies. The cut-off ican–British (FAB) guidelines.7 Bone marrow specimens were point for positivity was at 20%. ALL was diagnosed if blasts Chromosome 5 or 7 abnormalities in ALL BS Dabaja et al 871

Figure 1 Survival of patients with or without chromosome 5 or 7 abnormalities.

Figure 2 Survival of patients with or without chromosome 5 or 7 abnormalities, excluding patients with the Philadelphia .

were morphologically lymphoid, TdT-positive, MPO-negative, Results and were positive for lymphoid markers. Cytogenetic tech- niques utilized on the bone marrow and peripheral blood have been previously described.8 Karyotypic classification Characteristics of patients with chromosomes 5 or 7 was according to the International System for Cyto- abnormalities genetic Nomenclature (ISCN).9

Thirty-one of 468 patients analyzed (6.6%) had abnormalities Statistical methods of chromosomes 5 or 7. Compared with patients not exhibit- ing the changes, patients with chromosome 5 or 7 abnormali- Survival was estimated by the method of Kaplan and Meier.10 ties did not have significantly different characteristics, except Differences in survival curves were analyzed by the log rank for a trend for older age (age у60 years 29% vs 18%, test.11 Variables were compared by chi-square tests. P = 0.14) (Table 1). Chromosome 5 or 7 abnormalities in ALL BS Dabaja et al 872 Phenotypic and cytogenetic characteristics (Table 1). There was a tendency for patients with chromosome 5 or 7 abnormalities to be older and to express CD34 more Among patients who had imunophenotypic analysis, positive often. They had also a significant association with Ph expression of CD10 (CALLA) was detected in 17 of 24 patients (P = 0.004). The CR rate with chromosome 5 or 7 abnormali- analyzed (70%), positive expression of CD19 in 15 of 20 ties was lower than that of adult ALL patients without these patients (75%) and positive expression of T cell markers (CD2, abnormalities (64% vs 79%; P = 0.038), but the estimated 3- CD5, CD7) in five patients (21%). Myeloid marker positivity year survival rates were not statistically different (32% vs 36% (CD13, CD14, CD33) was found in 12 of 23 patients (52%). P = 0.14) (Figure 1). When the nine patients who had both Compared with patients without chromosome 5 or 7 abnor- chromosome 5 or 7 abnormalities and Ph were excluded, the malities, these patients had more frequent expression of CD34 CR rate of the remaining 22 with chromosome 5 or 7 abnor- positivity (17 of 23 patients = 74% vs 54%, P = 0.07; Table 2). malities was comparable to adult ALL without these abnor- Cytogenetic abnormalities detected in the 31 patients were malities (68% vs 81%, P = 0.11) and the estimated 3-year deletion in chromosome 7 in seven patients, loss of chromo- survival rates were similar (Figure 2). some 7 in 19 patients, and both abnormalities occurred in two In summary, chromosome 5 or 7 abnormalities can be asso- patients. Loss of chromosome 5 was found in six cases; three ciated with ALL, and should not be used as a diagnostic clue of them had both chromosome 5 and 7 abnormalities. Three for AML in ambiguous cases. In contrast to what was observed patients had deletion of chromosome 7 as the sole abnor- with myeloid malignancies, they were not associated with mality. The remaining 28 had other chromosomal abnormali- poor prognosis in ALL, except when occurring together with ties, the most significant being the presence of the Philadel- Ph. phia chromosome (Ph) in nine patients (29%). The incidence of this association was higher than in patients without chro- References mosome 5 or 7 abnormalities (29% vs 11%, P = 0.004). The detailed chromosomal abnormalities associated with chromo- 1 Keating MJ, Cork A, Broach Y et al. Toward a clinically relevant some 5 or 7 abnormalities among the 31 patients are shown cytogenetic classification of acute myelogenous leukemia. Leuke- mia Res 1987; 11: 119–133. in Table 3. 2 Johansson B, Mertens F, Mitelman F. Cytogenetic deletion maps of hematologic neoplasms: circumstantial evidence for tumor sup- pressor loci. Chromosomes Cancer 1993; 8: 205–218. Response and survival 3 Le Beau MM, Albain KS, Larson RA et al. Clinical and cytogenetic correlations in 63 patients with therapy-related myelodysplastic syndromes and acute nonlymphocytic leukemia: further evidence The complete remission (CR) rate was 64% among patients for characteristic abnormalities of chromosomes no. 5 and 7. J with chromosomes 5 or 7 abnormalities vs 79% for other ALL Clin Oncol 1986; 4: 325–345. patients (P = 0.038). The estimated 3-year survival rate was 4 Heim S, Mitelman F. Chromosome abnormalities in the myelod- 32% vs 36% (P = 0.14) (Figure 1). When Ph-positive cases ysplastic syndromes. Clin Haematol 1986; 15: 1003–1021. were excluded, the CR rates were comparable among patients 5 Van den Berghe H, Vermaelen K, Mecucci C, Barbieri D, Tricot G. The 5q− anomaly. Cancer Genet Cytogenet 1985; 17: 189–255. with or without chromosome 5 or 7 abnormalities (68% vs = 6 Ruutu P, Ruutu T, Vuopie P, Kosunen TU, de la Chapelle A. Defec- 81% P 0.11), and the estimated 3-year survival rates were tive chemotaxis in -7. Nature 1977; 265: 146–147. similar (36%) (Figure 2). 7 Bennett JM, Catovsky D, Daniel MT et al. Proposed revised criteria for the classification of acute myeloid leukemia. A report of the French–American–British Cooperative Group. Ann Intern Med 1985; 103: 620–625. Discussion 8 Trujillo JM, Cork A, Ahearn MJ, Youness EL, McCredie KB. Hema- tologic and cytologic characterization of 8/21 translocation acute Since chromosome 5 or 7 abnormalities predominantly affect granulocytic leukemia. Blood 1979; 53: 695–706. the myeloid lineage, it has been assumed, in suspicious mor- 9 ISCN (1995). An International System for Human Cytogenetics phology instances, that they would favor myeloid rather than Nomenclature. Felix Mitelman (ed). S Karger: Basel, 1995. 10 Kaplan EL, Meier P. Nonparametric estimation from incomplete lymphoid lineage, and would implicate poor prognosis. How- observations. J Am Stat Assoc 1958; 53: 457–481. ever, they have been also rarely described in lymphoid dis- 11 Mantel N. Evaluation of survival data and two new rank order orders.12,13 Van den Berghe and Michaux12 reported 5q− as a statistics arising in its consideration. Cancer Chemother Rep 1966; sole anomaly in six of 52 cases with B cell ALL, and in two 50: 163–170. − of 10 cases with T cell ALL. Deletion of 7q was also reported 12 Van den Berghe H, Michaux L. 5q , twenty-five years later: a syn- in association with chronic lymphoid malignancy.13 opsis. Cancer Genet Cytogenet 1997; 94: 1–7. 13 Hernandez JM, Mecucci C, Michaux L et al. del(7q) in chronic B- Little is known about the significance of chromosome 5 or cell lymphoid malignancies. Cancer Genet Cytogenet 1997; 93: 14,15 7 abnormalities in ALL where chromosomal abnormalities 147–151. have a major prognostic impact.16 14 Diez-Martin JL, Dewald GW, Pierre RV. Possible cytogenetic dis- This study attempted to define the associations of chromo- tinction between lymphoid and myeloid blast crisis in chronic some 5 or 7 abnormalities in adult ALL, in relation to host granulocytic leukemia. Am J Hematol 1988; 27: 194–203. and disease features, and to prognosis. In our analysis in 15 Mitelman F, Heim S. Quantitative acute leukemia cytogenetics. Genes Chromosomes Cancer 1992; 5: 57–66. adults with ALL, no significant differences were found 16 Faderl S, Kantarjian HM, Talpaz M, Estrov Z. Clinical significance between patients with and without chromosomes 5 or 7 of cytogenetic a normalities in adult acute lymphoblastic leuke- abnormalities with regard to clinical or hematologic features mia. Blood 1998; 91: 3995–4019.