Clinical Significance of Immunophenotypic Markers in Pediatric T-cell Acute Lymphoblastic Leukemia

Transcription

1 Journal of the Egytian Nat. Cancer Inst., Vol. 2, No. 2, June: , 28 Clinical Significance of Immunohenotyic Markers in Pediatric T-cell Acute Lymhoblastic Leukemia IMAN SIDHOM, M.D. 1 ; KHALED SHAABAN, M.B.B.Ch. 2 ; SONYA SOLIMAN, M.D. 2 ; SAMEERA EZZAT, M.D. 3 ; WAFAA EL-ANWAR, B. PHARM 4 ; NAYERA HAMDY, M.D. 2 ; DINA YASSIN, M.D. 2 ; SHERINE SALEM, M.D. 2 ; HALA HASSANEIN, M.D. 2 and MOHAMED TAREK MANSOUR, Ph.D. 2 The Deartments of Pediatric Oncology, NCI 1 ; Clinical Pathology, NCI 2 ; Biostatistics & Research 3 and Research 4, Children Cancer Hosital, Cairo, Egyt. ABSTRACT Background: Cell-marker rofiling has led to conflicting conclusions about its rognostic significance in T-ALL. Aim: To investigate the revalence of the exression of CD34, CD1 and myeloid associated antigens (CD13/ CD33) in childhood T-ALL and to relate their resence to initial clinical and biologic features and early resonse to theray. Patients and Methods: This study included 67 consecutive atients with newly diagnosed T-ALL recruited from the Children's Cancer Hosital in Egyt during the time eriod from July 27 to June 28. Immunohenotyic markers and minimal residual disease (MRD) were studied by five-color flow cytometry. Results: The frequency of CD34 was 34.9%, CD1 33.3%, while CD13/CD33 was 18.8%. No significant association was encountered between CD34, CD1 or myeloid antigen ositivity and the resenting clinical features as age, sex, TLC and CNS leukemia. Only CD1 + exression had significant association with initial CNS involvement (=.39). CD34 and CD13/CD33 exression was significantly associated with T-cell maturation stages (<.5). No relationshi was observed for age, TLC, gender, NCI risk or CNS involvement with early resonse to theray illustrated by BM as well as MRD day 15 and day 42. CD34 +, CD13/CD33 + and early T-cell stage had high MRD levels on day 15 that was statistically highly significant (<.1), but CD1 + had statistically significant lower MRD level on day 15 (=.49). However, only CD34 retained its significance at an MRD cut-off level of.1%. Conclusion: CD34, CD1, CD13/CD33 exression, as well as T-cell maturation stages, may have rognostic Corresondence: Dr Iman Sidhom, National Cancer Institute, Cairo University, Pediatric Oncology Deartment, ImanSidhom@yahoo.com significance in ediatric T-ALL as they have a significant imact on early clearance of leukemic cells detected by MRD day 15. Key Words: Pediatric T-ALL CD34 CD1 Prognostic factors T-cell maturation stages. INTRODUCTION In contrast to childhood B-lineage ALL, for which the general outcome has imroved over the last decades to cure rates reaching nearly 85%, the rognosis for children with T-ALL remains inferior. About 3% of these atients relase on current treatment rotocols [1]. Methods to identify atients exected to fail on current rotocols are required, so that alternative theray can be introduced as early as ossible. Demograhic and clinical resenting features, such as older age and a high leucocyte count at diagnosis, are regarded as unreliable redictors of outcome in atients with T-ALL treated with intensive chemotheray [1,2]. More recent studies have rovided insights into the genetic abnormalities underlying T-ALL develoment; however, the rognostic associations of molecular abnormalities in T-ALL are not sufficiently comelling to justify their use in treatment lanning [3-5]. Cell-marker rofiling has led to conflicting conclusions about its rognostic significance. CD1-negative B-ALL atients were reorted to have a oor rognosis and may indicate an adverse rognosis in T-ALL [6]. Other results showed that the lack of CD1 exression is not an indeendent adverse rognostic factor either 111

2 112 in B- or in T-ALL [7]. The exression of the CD34 was reorted as a favorable rognostic marker for children with B-lineage ALL but not for T-ALL [8]. A recent study by Van Grotel et al. [9] showed that CD34 ositive T-ALL had a relatively oor survival. The relevance of these markers in relation to outcome for ediatric T- ALL remains unclear. Multiarametric flow cytometric immunohenotying is useful in detecting minimal residual disease (MRD) following chemotheray [1]. In this resect, the use of henotyic markers of immaturity, including CD34 and CD1 as well as associated myeloid markers, in rognosis and theray of ALL is of articular imortance. The aim of this study was to investigate the revalence of the exression of the stem cell marker CD34, as well as CD1 and myeloid associated antigens (CD13 and/or CD33), in childhood T-ALL and to relate their resence to the initial clinical and biologic features as well as to early resonse to theray and minimal residual disease. PATIENTS AND METHODS Patients: From July 27 to June 28, 67 consecutive atients with newly diagnosed T-ALL were recruited from the Children's Cancer Hosital, Egyt (CCHE). They included 47 males and 2 females. Their ages ranged from 1 to 18 years. CD34 and CD1 exression were analyzed in 63 atients; myeloid associated markers in 64 atients. Written informed consents were obtained from the atients' arents and the rotocol was aroved by the Hosital Research Board of the CCHE. Diagnosis: Diagnosis was made according to standard methods including comlete blood icture, bone marrow (BM) asirate, cytochemistry and immunohenotying. Flow cytometry: Immunohenotyic analysis was erformed on bone marrow and eriheral blood samles freshly collected on a reservative-free sodium hearin tubes. The samles were rocessed by a Ficoll density gradient centrifugation to isolate Immunohenotyic Markers in Pediatric T-cell ALL mononuclear cells both at diagnosis and at the different minimal residual disease (MRD) follow-u checkoints (day 15 and day 42). Cells were stained using direct immunoflorescence mabs. The anel included B, T and myeloid lineage markers and early rimitive markers to identify the stage of maturation and detect leukemia-associated aberrant immunohenotyes (LAIPs) at diagnosis. Five-color flow cytometric immunohenotying was erformed on a FC5 Beckman Coulter machine (Beckman, USA). The markers used included CD45, anti Class II MHC, CD19, CD22, CD2, CD1, CD79a, Cyt µ, Kaa light chain, Lambda light chain, scd3, ccd3, CD7, CD5, CD2, CD1, CD4, CD8, Tdt, CD33, CD13, CD14 & CD34 (conjugated with the FITC, RPE, ECD, PC-5, PC-7). All antibodies were urchased from Beckmann Coulter/Immunotech, Becton Dickinson & DAKO (USA). T-Cell differentiation was determined as follows: T early henotye: leukemic cells exress ccd3 as well as any other an T marker (CD7, CD5, or CD2), T intermediate henotye: leukemic cells exress ccd3 or scd3 as well as CD1 and/or the co-exression of CD4 & CD8 lus any an T marker, T late henotye: leukemic cells exress scd3, CD4 or CD8 as well as any other an T marker. Some of our cases were considered unclassifiable and diagnosed as T henotye if they did not have the criteria mentioned earlier concerning the ccd3, scd3, CD1, and CD4 & CD8 attern of exression. The threshold of ositivity was considered 2% for all markers excet for CD1, CD34 and Tdt, in which a cut-off of 1% was used. After evaluation of the diagnostic samles, a combination of antibodies was selected and alied to the follow-u samles that best covered the LAIP. The resective combinations of antibodies were added to at least 1x16 mononuclear cells; multiarameter flow cytometric analysis was erformed. For samles at diagnosis, at least 2, events were acquired, while for follow-u samles at least.5x16 events were acquired. For MRD follow-u a 5-color tube was usually used combining CD45, CD3, CD5, Tdt, or CD34 or both lus an aberrant marker like DR, CD33 or CD19 and also CD1 if exressed by the malignant T cells [1]. Data were collected on LMD files and the analysis was erformed using Winlist 6. 3D software

3 Iman Sidhom, et al. 113 Treatment: All atients were treated according to the treatment rotocol adoted from the total theray study XV for standard/high-risk ALL of St. Jude Children s Research Hosital (SJCRH). It included 6 weeks of induction, 8 weeks of consolidation, and continuation theray of 12 weeks for girls and 146 for boys. Remission induction and consolidation: Remission-induction theray was instituted with rednisone, vincristine, doxorubicin, and asaraginase. Patients with a minimal residual disease of 1% on day 15 received three additional doses of asaraginase. Subsequent induction theray consisted of cyclohoshamide, mercatourine, and cytarabine. On hematooietic recovery on day 42, the minimal residual disease was assessed, and consolidation theray was started which consisted of daily mercatourine for 56 days and 4 courses of high dose methotrexate 5g/m 2 given every 2 weeks. Continuation theray: Patients received weekly asaraginase and daily mercatourine with ulses of doxorubicin lus vincristine lus dexamethasone. They also received two reinduction treatments between weeks 7 and 9 and between weeks 17 and 2. For the remaining continuation theray, atients received three rotating drug airs (mercatourine lus methotrexate, cyclohoshamide lus cytarabine, and dexamethasone lus vincristine). The total scheduled dosages of anthracyclines were limited to 23 mg/m 2, while the total scheduled dosages of cyclohoshamide were limited to 4.6 g/m 2. CNS-directed Theray: Intrathecal cytarabine was immediately instilled after a diagnostic lumbar uncture, and trile intrathecal chemotheray was used in all subsequent treatments. Deending on the resenting atient characteristics and the CNS status, atients received 16 to 25 intrathecal theray. Resonse criteria: Early resonse to treatment was measured by day 15 BM blast ercentage. Raid early resonse was defined as M1 (<5% blasts), slow early resonse was defined as M2 (5-25% blasts) or M3 (>25% blasts). Sixty-three atients were evaluable for BM day 15 and sixty-one for BM day 42. Minimal residual disease was measured in 4 atients by flow cytometry on day 15 of induction theray and day 42 (end of induction). Levels below.1% were considered as negative. Comlete remission (CR) was defined as the absence of leukemic blasts in eriheral blood and CSF and less than 5% blasts in BM, together with hematooietic regeneration and no evidence of extramedullary (localized) disease. Relase was defined as recurrence of more than 5% lymhoblasts or localized leukemic infiltrates at any site. Statistical analysis: Patient data were tabulated and rocessed using the SPSS (16) for Windows software. Qualitative data were exressed as frequency and ercentage, quantitative data as mean ± standard deviation and median. The Chi-square test was used for comarative analysis. The Fisher exact test was used for 2 x 2 contingency tables and if 25% of cells had exected counts less than 5. Differences were considered significant at a value of.5 and highly significant at a value of.1 [11]. RESULTS Patients characteristics: The study included 67 consecutive newly diagnosed T-cell ediatric ALL atients with a median age of 8 years (range 1-18 years). The male to female ratio was 2.4:1. The total leucocytic count (TLC) median was 11x1 9 /L with a range of x1 9 /L. Fifty-two ercent (52%, 35/67 atients) had initial TLC 1x 1 9 /L. CNS leukemia was resent in 12% (8/67 atients; 2 had CNS II and 6 had CNS III). Patients were classified according to NCI/Rome criteria [12] into standard risk which included atients with age ranges 1-9 years and initial WBC <5x1 9 /l; and high risk which included all other atients. Fifty atients (74.6%) were classified as high risk and 17/67 atients (25.4%) as standard risk.

4 114 According to surface antigen exression by leukemic cells, 22.4% of atients (15/67) had features of early T-cell maturation, 52.2% (35/67) of intermediate T-cell maturation, 13.4% (9/67) of late differentiation, while 11.9% (8/67) were classified as unsecified T-cell henotye. The frequency of CD34 was 34.9% (22/63), CD1 was 33.3% (21/63) and CD13 and/or CD33 was 18.8% (12/64). Immunohenotyic Markers in Pediatric T-cell ALL Table (1) summarizes the resenting clinical and biological features in relation to CD34, CD1 and CD13 and/or CD33 exression in the studied T-ALL atients. No significant association was encountered between CD34, CD1 or myeloid antigen ositivity and age, sex, white blood count and CNS leukemia aart from significant higher association of CD1 + cases with initial CNS involvement (=.39). Table (1): Presenting clinical and biological features in relation to CD34, CD1 and CD13/CD33 exression in the studied ediatric T-ALL atients. Clinical feature CD34 +ve n=22 CD34 ve n=41 CD1 +ve n=21 CD1 -ve n=42 CD13/CD33 +ve n=12 CD13/CD33 -ve n=52 Age: 1- <1 years 1 years 11 (5%) 11 (5%) 29 (7.7%) 12 (29.3%) (61.9%) 8 (38.1%) 27 (64.3%) 15 (35.7%) 1 4 (33.3%) 8 (66.7%) 36 (69.2%) 16 (3.8%).43 Sex: Males Females 16 (72.7%) 6 (27.3%) 29 (7.7%) 12 (29.3%) 1 13 (61.9%) 8 (38.1%) 31 (73.8%) 11 (26.2%) (75%) 3 (25%) 36 (69.2%) 16 (3.8%) 1 TLCx1 9 /L: < (45.5% 2 (9%) 1 (45.5%) 16 (39%) 3 (7.3%) 22 (53.7%).79 6 (28.6%) 3 (14.3%) 12 (57.1%) 2 (47.6%) 1 (2.4%) 21 (5%) (58.3%) 2 (16.7%) 3 (25%) 19 (36.5%) 3 (5.8%) 3 (57.7%).83 CNS: I II III 21 (95.5%) 1 (4.5%) 35 (87.5%) 1 (2.5%) 4 (1%) (76.2%) 1 (4.8%) 4 (19%) 39 (95.1%) 2 (4.9%).39* 12 (1%) 44 (86.3%) 1 (2%) 6 (11.8%).66 T-cell stages : T-early T-intermediate T-late 9 (47.4%) 7 (36.8%) 3 (15.8%) 5 (13.9%) 26 (72.2%) 5 (13.9%).15* 2 (11.8%) 12 (7.6%) 3 (17.6%) 11 (28.9%) 22 (57.9%) 5 (13.2%).38 8 (8%) 2 (2%) 6 (13%) 32 (69.6%) 8 (17.4%).1** = 8 Missed cases had unsecified T-henotye *<.5=Statistically significant **<.1=Statistically highly significant The correlation between CD34 exression and the stage of thymocyte maturation was statistically significant. CD34 ositive cases were mainly of the T-early henotye while the T intermediate stage was more revalent among the CD34 negative cases (=.15). However, there was no clear loss of the antigen with advancing thymocyte maturity as CD34 + was observed in 9/14 (64.3%) early T-cell atients, 7/33 (21.2%) intermediate T-stage and 3/8 (37.5%) of T-late stage. There was no statistical significance between CD1 exression and T- cell maturation stages (=.38). The association between myeloid antigen exression and T-cell maturation stages showed statistical significance, as 8% of the cases exressing myeloidassociated markers were of the T-early henotye and 2% were T-intermediate comared with 13% and 69.6% of the negative cases, resectively (=.1). CD34 showed a statistically significant correlation with myeloid-associated markers (=.2), but not with CD1 exression (=.57) (Table 2). In contrast, CD1 did not show a statistically significant correlation with myeloidassociated markers which were exressed in 3/21 (14.3%) of CD1 + cases vs 8/42 (19%) of CD1- cases (=.738).

5 Iman Sidhom, et al. 115 Table (2): CD34 exression in relation to CD1 and myeloidassociated marker exression in the studied ediatric T-ALL atients CD1+ CD1- CD13/33- CD13/33+ Positive 13/21 (61.9%) 8/21 (38.1%) 13/22 (59.1%) 9/22 (4.9%) CD34 *<.1=Statistically highly significant Negative 29/41 (7.7%) 12/41 (29.3%) 38/41 (92.7%) 3/41 (7.3%) value.57.2* Resonse to theray: Slow early resonse (SER), defined as M2 and/or M3 marrow at day 15, was observed in 23.8% of CD34 + vs 18.4% of CD34- cases, and in 1% of CD1 + vs 25.6% of CD1- cases. However, these differences were not statistically significant (=.73 and.19, resectively). CD34 as well as CD1 had no significant imact on comlete remission rate (=.59 and.58, resectively). Table (3) summarizes the effect of age, sex, TLC, NCI risk, CD34, CD1, myeloid markers and T-cell maturation stages on early resonse to theray as regard bone marrow day 15 and day 42. None had a significant effect on the bone marrow status at day 15 or day 42. Table (3): Prognostic factors in relation to bone marrow day 15 and day 42 in the studied ediatric T-ALL atients. Parameter BM day 15 n=63 BM day 42 n=61 z RER SER value RER SER value Age: 1- <1 years 1 years 33 (82.5%) 17 (73.9%) 7 (17.5%) 6 (26.1%) (95.1%) 18 (9%) 2 (4.9%) 2 (1%).591 Sex: Males Females 37 (84.1%) 13 (68.4%) 7 (15.9%) 6 (31.6%) (93%) 17 (94.4%) 3 (7%) 1 (5.6%) 1 TLCx1 9 /L: < (88%) 5 (1%) 23 (69.7%) 3 (12%) 1 (33.3%) (1%) 5 (1%) 27 (87.1%) 4 (12.9%).18 NCI Risk: Standard risk High risk 15 (93.8%) 35 (74.5%) 1 (6.2%) 12 (25.5%) (1%) 4 (9.9%) 4 (9.1%).569 CD34+ CD34-16 (76.2%) 31(81.6%) 5 (23.8%) 7 (18.4%) (89.5%) 37 (94.9%) 2 (1.5%) 2 (5.1%).591 CD1+ CD1-18 (9%) 29 (74.4%) 2 (1%) 1 (25.6%) (88.9%) 38 (95%) 2 (11.1%) 2 (5%).581 CD13/CD33+ CD13/CD33-7 (7%) 41 (82%) 3 (3%) 9 (18%).43 8 (8%) 47 (95.9%) 2 (2%) 2 (4.1%).13 T-cell stages: T-early T-intermediate T-late 1 (71.4%) 27 (81.8%) 9 (1%) 4 (28.6%) 6 (18.2%) (86.7%) 32 (97%) 7 (1%) 2 (13.3%) 1 (3%).349 RER= Raid early resonse (M1). SER= Slow early resonse (M2 or M3). Evaluable cases for BM day 15 were 59 cases for CD34 as well as CD1, 6 for CD13/CD33 and 56 for T-cell stages. z Evaluable cases for BM day 42 were 58 cases for CD34 as well as CD1, 59 for CD13/CD33 and 55 for T-cell stages.

6 116 Table (4) shows the studied rognostic factors in relation to minimal residual disease (MRD) day 15 and day 42. Age, sex, TLC and NCI Risk had no significant imact on MRD day 15 and day 42. CD34 had a highly statistically significant imact on MRD day15 as 1/16 (62.5%) of CD34 + cases had MRD 1% and none had MRD <.1% vs 5/23 (21.7%) and 12/23 (52.2%) for CD34- cases; resectively (=.2). A ositive case for CD34 with an MRD level of.16% on day15 is illustrated in Fig. (1). Similarly, T-cell maturation stages showed a highly statistically significant imact on MRD Immunohenotyic Markers in Pediatric T-cell ALL day 15 as all the 7 atients of T-early henotye had MRD 1% vs 7/26 (26.9%) of T- intermediate vs none of the late T-cell ALL (=.1). Also, myeloid-associated marker exression was statistically highly significantly associated with higher MRD day 15 as 87.5% of +ve cases had MRD 1% vs 25% of negative cases (=.4). CD1 exression had statistically significant lower levels of MRD day 15; 4% vs 29% had MRD <.1% and 13.3% vs 5% had MRD 1% for CD1 + and CD1- cases; resectively (=.49). None of these factors had a statistically significant imact on MRD day 42. Table (4): Prognostic factors in relation to minimal residual disease day 15 and day 42 in the studied ediatric T-ALL atients. <.1% MRD day 15 n=4 MRD day 42 n=4 z.1-1% 1% <.1%.1-1% 1% Age: 1- <1 years 1 years 7 (28%) 6 (4%) 8 (32%) 4 (26.7%) 1 (4%) 5 (33.3%) (4.7%) 6 (46.2%) 12 (44.4%) 3 (23.1%) 4 (14.8%) 4 (3.8%).312 Sex: Males Females 7 (26.9%) 6 (42.9%) 9 (34.6%) 3 (21.4%) 1 (38.5%) 5 (35.7%) (32%) 9 (6%) 11 (44%) 4 (26.7%) 6 (24%) 2 (13.3%).239 TLCx1 9 /L: < (26.7%) 2 (5%) 7 (33.3%) 5 (33.3%) 1 (25%) 6 (28.6%) 6 (4%) 1 (25%) 8 (38.1%).97 9 (5%) 8 (4%) 8 (44.4%) 2 (1%) 5 (25%) 1 (5.6%) 7 (35%).57 NCI Risk: Standard risk High risk 3 (27.3%) 1 (34.5%) 4 (36.4%) 8 (27.6%) 4 (36.4%) 11 (37.9%).97 7 (53.8%) 1 (37%) 6 (46.2%) 9 (33.3%) 8 (29.6%).83 CD34+ CD34-12 (52.2%) 6 (37.5%) 6 (26.1%) 1 (62.5%) 5 (21.7%).2 5 (38.5%) 11 (42.3%) 4 (3.8%) 11 (42.3%) 4 (3.8%) 4 (15.4%).514 CD1+ CD1-6 (4%) 7 (29.2%) 7 (46.7%) 5 (2.8%) 2 (13.3%) 12 (5%).49 6 (46.2%) 11 (4.7%) 6 (46.2%) 9 (33.3%) 1 (7.7%) 7 (25.9%).468 CD13/CD33+ CD13/CD33-1 (12.5%) 12 (37.5%) 12 (37.5%) 7 (87.5%) 8 (25%).4 1 (16.7%) 16 (47.1%) 3 (5%) 12 (35.3%) 2 (33.3%) 6 (17.6%).324 T-cell stages: T-early T-intermediate T-late 1 (38.5%) 2 (5%) 9 (34.6%) 2 (5%) 7 (1%) 7 (26.9%).1 3 (33.3%) 1 (41.7%) 3 (1%) 3 (33.3%) 9 (37.5%) 3 (33.3%) 5 (2.8%).58 *<.5= Statistically significant. **<.1= Statistically highly significant. Evaluable cases for MRD day 15 were 39 cases for CD34 as well as CD1, 4 for CD13/CD33 and 37 for T-cell stages. z Evaluable cases for MRD day 42 were 39 cases for CD34, 4 for CD1 as well as CD13/CD33 and 36 for T-cell stages.

7 Iman Sidhom, et al. 117 Table (5) shows the rognostic significance of the studied immunohenotyic markers as well as T-cell maturation stages in relation to MRD day15 at cut-off levels of.1% and 1%. T-cell stages, CD34, CD1 and CD13/CD33 were all significant at an MRD level of 1%, but only CD34 retained its significance at an MRD cut-off level of.1%. Table (5): Prognostic significance at different minimal residual disease levels in the studied ediatric T-ALL atients. CD34+ CD34- MRD day 15 MRD day 15 <.1%.1% <1% 1% 12 (52.2%) 16 (1%) 11 (47.8%).1 6 (37.5%) 18 (78.3%) 1 (62.5%) 5 (21.7%).18 CD1+ CD1-6 (4%) 7 (29%) 9 (6%) 17 (7.8%) (86.7%) 12 (5%) 2 (13.3%) 12 (5%).38 CD13/CD33+ CD13/CD33-1 (12.5%) 12 (37.5%) 7 (87.5%) 2 (62.5%) (12.5%) 24 (75%) 7 (87.5%) 8 (25%).2 T-cell stages: T-early T-intermediate T-late 1 (38.5%) 2 (5%) 7 (1%) 16 (61.5%) 2 (5%) (73.1%) 4 (1%) 7 (1%) 7 (26.9%).1 *<.5= Statistically significant. **<.1= Statistically highly significant. Evaluable cases were 39 cases for CD34 as well as CD1, 4 for CD13/CD33 and 37 for T-cell stages R1: 97.9% R2: 95.82% R1: 61.59% R2: 99.24% 1 CYT 1 1 CD3 1 2 PC5Log CYT CD3 PC5Log R3: 39.27% CD34 PC7Log CYT CD3 PC5Log R5: 99.56% R4: 99.76% No gate G1 (R1) G2 (R1 & R2) G3 (R1,R2 & R3) G4 (R1,R2,R3 &R4) G5 (R5) Id1: 2_TDT-CD5-CD Panelldl Secimen R_1 2._MD Id2: 9 Aug 29 Percent %Gate geomeanx geomeany R1 R2 R3 R Initial T Intermediate, +ve for CD 34 Initial T Intermediate, +ve for CD34 D15 with.16% MRD ositivity Fig. (1): A T-ALL case ositive for CD34 with an MRD level of.16% on day CYT 1 1 CD3 1 2 PC5Log CYT CD3 PC5Log R3:.32% CD34 PC7Log CYT CD3 PC5Log R5: 93.1% R1 R2 R3 R R4: 9.87% No gate G1 (R1) G2 (R1 & R2) G3 (R1,R2 & R3) G4 (R1,R2,R3 &R4) G5 (R5) Id1: 1_TDT-CD5-CD 548 MRD BL Secimen R_1 1 LMD Id2: 9 Aug 29 Percent %Gate geomeanx geomeany

8 118 DISCUSSION In this study of 67 consecutive ediatric T- ALL, 34.9% exressed CD34, 33.3% CD1 and 18.8% CD13 and/or CD33. CD34 associated with CD1 or aberrant myeloid antigen is a good marker for the investigation of MRD in ALL [13]. Thus, assessment of these markers lay an imortant role in the context of MRD detection [14]. CD34 is a marker that is normally exressed on hematooietic stem cells and early thymic T-cell recursors; therefore, exression of CD34 could reflect early T-cell maturational arrest in T-ALL [8]. Although exression of CD34 in this study was statistically significantly correlated with early T-cell stage, CD34 was also associated with immature and less frequently with intermediate develoment stages. Pui et al. [8] reorted that among T-cell ALL, in contrast to B-lineage ALL, CD34 exression showed ositive correlations with initial CNS leukemia and CD1 negativity, but not with any good-risk resenting characteristics. In the resent study, CD34 exression was not significantly correlated to unfavorable resenting features or CD1 negativity, but was significantly associated with myeloid marker coexression. Pui et al. [8] did not observe a significant adverse rognostic effect for CD34 exression in 61 ediatric T-ALL atients. A recent study by Van Grotel et al. [9] conducted on 72 ediatric T-ALL showed that CD34 ositive T-ALL cases were associated with oor disease free (DFS) and overall survival (OS). Immunohenotyic Markers in Pediatric T-cell ALL For acute leukemia in general, exression of the stem-cell marker CD34 was reviously reorted to be associated with high exression of multidrug resistance (MDR) genes and to a oor resonse to chemotheray [15]. It was concluded that exression of MDR genes in these ALL subsets might rovide a ossible exlanation for oor outcome. Several studies showed an association between high P- glycorotein (P-g) exression, encoded by the MDR1 gene, with oor clinical outcome [16], whereas other studies contradicted these findings [17]. Recently, Van Grotel et al. [9] reorted an association of CD34 with elevated MDR1 and MDR-associated rotein 1 (MRP1) mrna exression levels. However, the exression levels of these multile drug resistance genes were not associated with outcome, and therefore did not exlain the oor rognostic imact of CD34 exression. Moreover, in an Egytian study by Kamel et al. [18], P-g exression had no imact on survival in Egytian ediatric ALL and was not correlated to CD34 exression. In this study, no relationshi was observed for age, TLC, gender, NCI risk or CNS involvement with early resonse to theray illustrated by BM morhology as well as MRD days 15 and 42. CD34, CD13 and/or CD33 exression and early T-cell stage had high MRD levels on day 15 that were statistically highly significant (<.1) but this relationshi was not observed for MRD day 42. On the other hand, CD1 ositivity had statistically significant lower MRD level on day 15 (=.49). Patients who achieve morhologic remission and have minimal residual disease are associated with higher risk of relase [19]. Levels of MRD are roortional to the risk of relase [2,21]. The.1% threshold is commonly used to define MRD ositivity as it has roven to be clinically informative. St Jude studies for MRD found that atients who had MRD of.1% or higher in the bone marrow at any time oint during treatment had a significantly higher risk of relase [2]. Remission induction theray can roduce dramatic leukemia cytoreduction and undetectable MRD after only 2-3 weeks of theray in a roortion of atients. In a study by St Jude Children Research Hosital, 183/42 atients (45.5%) with B-lineage ALL were MRDnegative, defined as having less than.1% leukemic cells in the bone marrow, on day 19. Patients with early clearance of leukemic cells were found to have excellent rognosis with current treatment rotocols [22]. When analyzing the rognostic imact of the T-cell develoment stage for ediatric T- ALL, our revious study at the National Cancer Institute (NCI), Cairo University, [23] showed that early T-cell stage had significantly worse outcome than intermediate or late stages of maturation. Van Grotel et al. [24] did not observe an association for any differentiation stage according to EGIL or TCR classification criteria and outcome. Uon further extending this anal-

9 Iman Sidhom, et al. 119 ysis by investigating the rognostic imact of various individual CD markers, exression of CD34 was significantly associated with a oor 5-year DFS. In this study, when analyzing different cut-off levels of MRD on day 15, T-cell stages, CD34, CD1 and CD13/CD33 were all significant at an MRD cut-off level of 1%, but only CD34 was significant at an MRD level of.1%. In summary, we have illustrated the rognostic significance for CD34, CD1, CD13 and/or CD33 exression as well as T-cell maturation stages in childhood T-lineage ALL by their significant imact on early clearance of leukemic cells measured by minimal residual disease on day 15. CD34 retained its significance at an MRD level of.1%. A longer follow-u eriod is needed to confirm the rognostic significance of these variables on survival. REFERENCES 1- Pui CH, Evans WE. Treatment of acute lymhoblastic leukemia. N Engl J Med. 26, 354: Pullen J, Shuster JJ, Link M, Borowitz M, Amylon M, Carroll AJ, et al. Significance of commonly used rognostic factors differs for children with T-cell acute lymhocytic leukemia (ALL), as comared to those with B-recursor ALL. A Pediatric Oncology Grou (POG) study. Leukemia. 1999, 13: Cave H, Suciu S, Preudhomme C, Poe B, Robert A, Uyttebroeck A, et al. Clinical significance of HOX11L2 exression linked to t(5;14) (q35;q32), of HOX11 exression, and of SIL-TAL fusion in childhood T-cell malignancies: results of EORTC studies and Blood. 24, 13: Breit S, Stanulla M, Flohr T, Schrae M, Ludwig W, Tolle G, et al. Activating NOTCH1 mutations redict favorable early treatment resonse and long-term outcome in childhood recursor T-cell lymhoblastic leukemia. Blood. 26, 18: Moussa H, Sidhom I. HOX11L2 Exression in Egytian Pediatric T-cell Acute Lymhoblastic Leukemia. J Egyt Soc Hematol Res. 26, 2 (2): Pui CH, Rivera GK, Hancock ML, Raimondi SC, Sandlund JT, Mahmoud HH, et al. Clinical significance of CD1 exression in childhood acute lymhoblastic leukemia. Leukemia. 1993, 7: Consolini R, Legitimo A, Rondelli R, Guguelmi C, Barisone E, Lii A, et al. Clinical relevance of CD1 exression in childhood ALL. Haematologica. 1998, 83: Pui CH, Hancock ML, Head DR, Rivera GK, Look AT, Sandlund JT, et al. Clinical significance of CD34 exression in childhood acute lymhoblastic leukemia. Blood. 1993, 82: Van Grotel M, Van den Heuvel-Eibrink M, Van Wering E, van Noesel MM, Kams WA, Veerman AJ, et al. CD34 Exression is Associated with Poor Survival in Pediatric T-Cell Acute Lymhoblastic Leukemia. Pediatr. Blood Cancer. 28, 51: Vidriales MB, San-Miguel JF, Orfao A, Coustan-Smith E, Camana D. Minimal residual disease monitoring by flow cytometry. Best Pract Res Clin Haematol. 23, 16: Saunders DB, Tra GR. Basic and clinical biostatistics, 3 rd edition. Connecticut, Aleton and Lang, Smith M, Arthur D, Camitta B, Carroll Crist W, Gaynon P, Gelber R, et al. Uniform aroach to risk classification and treatment assignment for children with acute lymhoblastic leukemia. J. Clin. Oncol. 1996, 14: Coustan-Smith E, Ribeiro RC, Stow P, Zhou Y, Pui CH, Rivera GK, et al. A simlified flow cytometric assay identifies children with acute lymhoblastic leukemia who have a suerior clinical outcome. Blood. 26, 18: Camana D, Coustain Smith E. Detection of minimal residual disease in acute leukemia by flow cytometry. Cytometry. 1999, 38: Plasschaert SL, Vellenga E, de Bont ES, van der Kolk DM, Veerman AJ, Sluiter WJ, et al. High functional -glycorotein activity is more often resent in T-cell acute lymhoblastic leukaemic cells in adults than in children. Leuk Lymhoma. 23, 44: De Moerloose B, Swerts K, Benoit Y, Laureys G, Loeys T, Philié J, et al. The combined analysis of -glycorotein exression and activity redicts outcome in childhood acute lymhoblastic leukemia. Pediatr Hematol Oncol. 23, 2: Wuchter C, Leonid K, Ruert V, Schrae M, Büchner T, Schoch C, et al. Clinical significance of - glycorotein exression and function for resonse to induction chemotheray, relase rate and overall survival in acute leukemia. Haematologica. 2, 85: Kamel A, Sharkawy N, Yassin D, Shaaban K, Hussein H, Sidhom I, et al. P-g Exression and Rh 123 Efflux Assay have no Imact on Survival in Egytian Pediatric Acute Lymhoblastic Leukemia Patients. J. Egyt Natl. Cancer Inst. 25, 17 (3): Camana D. Determination of minimal residual disease in leukemia atients. Br J Haematol. 23, 121: Coustan-Smith E, Sancho J, Hancock ML, Boyett JM, Behm FG, Raimondi SC, et al. Clinical imortance of minimal residual disease in childhood acute lymhoblastic leukemia. Blood. 2, 96: Dworzak MN, Fr_schl G, Printz D, Mann G, Pötschger U, Mühlegger N, et al.: Prognostic significance and modalities of flow cytometric minimal residual disease detection in childhood acute lymhoblastic leukemia. Blood. 22, 99:

10 12 Immunohenotyic Markers in Pediatric T-cell ALL 22- Coustan-Smith E, Sancho J, Behm FG, Hancock ML, Razzouk BI, Ribeiro RC, et al. Prognostic imortance of measuring early clearance of leukemic cells by flow cytometry in childhood acute lymhoblastic leukemia. Blood. 22, 1: Hussein H, Aboul Naga Sh, Sidhom I, Kamel A, Hattab O, Hamza R. Heterogeneity of resenting features and their relation to treatment outcome in children with T-cell acute lymhoblastic leukemia in NCI, Egyt. J. Egyt. Natl. Cancer Inst. 1998, 1: Van Grotel M, Meijerink JP, Van Wering ER, Langerak AW, Beverloo HB, Buijs-Gladdines JG, et al. Prognostic significance of molecular-cytogenetic abnormalities in ediatric T-ALL is not exlained by immunohenotyic differences. Leukemia. 28, 22: