Karyotype instability between diagnosis and relapse in 117 ... - Nature

Leukemia (2002) 16, 2084–2091  2002 Nature Publishing Group All rights reserved 0887-6924/02 $25.00 www.nature.com/leu

Karyotype instability between diagnosis and relapse in 117 patients with acute myeloid leukemia: implications for resistance against therapy W Kern1, T Haferlach1, S Schnittger1, WD Ludwig,2 W Hiddemann1 and C Schoch1 1

Ludwig-Maximilians-University, University Hospital Grosshadern, Dept. of Internal Medicine III, Muenchen, Germany; and 2Robert-Ro¨ssleKlinik, Charite´, Humbolt-University, Berlin, Germany

The instability of the karyotype may play a role in the development of refractoriness of acute myeloid leukemia (AML) to antileukemic therapy. Therefore, in the current study cytogenetic analyses were performed in 117 patients with AML both at diagnosis and at relapse. Changes in karyotype were observed in 38% (36% of initially normal karyotypes, 39% of initially aberrant karyotypes). An evolution of karyotype, ie the acquisition of further aberrations in addition to those present at diagnosis, occurred more frequently in patients with unfavorable karyotypes at diagnosis as compared to all others (60% vs 32%, P = 0.0095). The duration from initial diagnosis to relapse was significantly shorter in cases with an evolution of the aberrant karyotype as compared to cases with no changes in the aberrant karyotype between diagnosis and relapse or with solely regression of aberrations at relapse (9.2 ± 4.4 vs 14.0 ± 8.5 months, P = 0.0081). In an additional analysis, another cohort of 120 patients with refractory and relapsed AML who were treated uniformly within the respective trial of the German AML Cooperative Group was analyzed cytogenetically at diagnosis and at relapse to further prove the prognostic impact of karyotype aberrations at relapse. Karyotypes were prognostically favorable, intermediate, unfavorable and not available in 8%, 50%, 17% and 25% at diagnosis and in 8%, 49%, 21% and 22% at relapse, respectively. Karyotype aberrations at diagnosis had no impact on response to therapy (P = 0.32) but influenced survival and event-free survival significantly (P = 0.03 and P = 0.02). In contrast, karyotype aberrations at relapse strongly influenced response to therapy (P = 0.05), survival (P = 0.01), and event-free survival (P = 0.002). These data suggest that the instability of the karyotype between diagnosis and relapse and thus karyotype aberrations at relapse in particular contribute to the refractoriness of AML to anti-leukemic therapy. Leukemia (2002) 16, 2084–2091. doi:10.1038/sj.leu.2402654 Keywords: AML; cytogenetics; refractory; relapsed; prognosis

Introduction Acute myeloid leukemia (AML) in most cases is associated with alterations of the DNA detectable at the cytogenetic and at the molecular level which are considered the basis of the pathogenesis of the disease leading to a stop of differentiation and an uncontrolled proliferation of the malignant cells.1 In cases with balanced chromosome translocation specific fusion genes have been identified interfering with nuclear transcription factors which play an important role in cell proliferation and differentiation.2–4 The mechanisms of the pathogenesis of AML with unbalanced chromosome aberrations and of AML with complex aberrant karyotypes, in particular, are less well understood and are believed to be the result of a genetic instability which still is to be defined.5 Besides their role in the pathogenesis of AML karyotype

Correspondence: W Kern, Ludwig-Maximilians-University, University Hospital Grosshadern, Dept of Internal Medicine III, 81366 Muenchen, Germany; Fax: +49-89-7095-4971 Received 13 December 2001; accepted 29 April 2002

aberrations are the most important independent factors determining the prognosis of patients receiving chemotherapy or stem cell transplantation for front-line or salvage therapy of AML.6–9 Hence, karyotypes are devided into three groups associated with a favorable, intermediate, and unfavorable prognosis with balanced aberrations like t(8;21), t(15;17) and inv(16)/t(16;16) comprising the favorable group and unbalanced aberrations involving chromosomes 5 and 7 as well as complex aberrations being part of the unfavorable group. In all groups, cases with multiple chromosome abnormalities occur in which the additional aberrations are considered secondary aberrations. The occurrence of refractoriness of the disease to anti-leukemic therapy is the most important factor for the limitation of the prognosis in patients with AML. However, besides limited evidence for the involvement of specific genetic alterations interfering with intracellular drug transport10 there is no concept for the development of refractoriness which is applicable to all patients with AML. Therefore, the current study aimed at defining the changes in karyotype abnormalities occurring in patients with AML between diagnosis and relapse and to provide evidence for their involvement in the development of refractoriness of the disease and thus their prognostic impact.

Methods

Patients The analyses were performed in two cohorts of patients. The first cohort comprised prospectively analyzed patients with AML according to FAB criteria11–14 for whom cytogenetics were performed both at initial diagnosis and at relapse at the reference laboratory for leukemia diagnostics, Ludwig-Maximilians-University, Munich. The second cohort comprised patients being treated according to the current protocol of the German AML Cooperative Group for patients with refractory and relapsed AML.15 Therapy consisted of sequential highdose cytosine arabinoside and idarubicin (S-HAI) regimen comprising cytosine arabinoside (AraC) every 12 h by a 3 h infusion on days 1, 2, 8 and 9 and idarubicin 10 mg/m2/day as a 30 min infusion on days 3, 4, 10 and 11, respectively (Figure 1). Based on the results of a previous study comparing two dose levels of high-dose AraC,16 patients younger than 60 years with refractory disease, early relapses following a first CR of less than 6 months, or second and subsequent relapses received AraC at doses of 3.0 g/m2 per application while all other patients were treated with 1.0 g/m2 AraC per single dose. All patients were randomly assigned to receive fludarabine in addition to S-HAI or S-HAI alone. Patients randomized for fludarabine received the drug at 15 mg/m2 as a 30 min infusion 4 h prior to each AraC administration, ie twice daily on days 1, 2, 8 and 9.

Karyotype instability in AML W Kern et al

2085

Figure 1

Schedule of the S-HAI regimen.

Cytogenetic analyses Bone marrow or peripheral blood cells were cultured in RPMI1640 medium supplemented with 20% fetal calf serum at 37°C for 24 h and 48 h. Unstimulated as well as cytokine stimulated cultures as described elsewhere were performed in parallel.5,17 Standard cytogenetic preparations were made, a modified chromosome-banding technique (GAG = Giemsa bands by acetic saline-Giemsa) was used, 15 to 30 metaphases were analyzed and classified according to the ISCN (International System for Human Cytogenetic Nomenclature).

Molecular analyses Screening for an internal tandem duplication of the MLL gene and for length mutations of the FLT3 gene was performed as described previously.18,19

Cytogenetic classification Karyotype abnormalities were grouped according to definitions previously established for the German AML Cooperative Group first line trials.20–23 Aberrations classified prognostically favorable were t(8;21), t(15;17) and inv(16)/t(16;16), irrespective of additional aberrations. Unfavorable karyotypes included inv(3)/t(3;3), aberrations of chromosomes 5 and 7, t(6;9), aberrations of 11q23, aberrations of 17p, and complex aberrations (at least three structural and/or numeric aberrations). All other karyotypes including normal ones were classified intermediate.

Definition of change in karyotype Based on the cytogenetic analyses at diagnosis and at relapse, seven categories of changes in karyotype were defined: • normal→normal (normal karyotype both at diagnosis and at relapse); • normal→aberrant (normal karyotype at diagnosis, aberrant karyotype at relapse); • aberrant→same aberrant (aberrant karyotype at diagnosis, same aberrant karyotype at relapse); • aberrant→evolution (aberrant karyotype at diagnosis, same aberrant karyotype plus additional aberrations at relapse); • aberrant→regression (aberrant karyotype at diagnosis, less but same aberrations at relapse);

• aberrant→evolution and regression (aberrant karyotype at diagnosis, less but same aberrations plus additional new aberrations at relapse); • aberrant→unrelated aberrant (aberrant karyotype at diagnosis, aberrant karyotype at relapse with exclusively different aberrations).

Study parameters Response to therapy was assessed according to CALGB criteria.24 CR was defined as a normal cellular bone marrow with normal erythroid and myeloid elements and less than 5% myeloblasts, and with peripheral blood counts of more than 100 000/␮l platelets and more than 1500/␮l granulocytes for at least 4 weeks. Patients with regenerated peripheral blood values but more than 5% and less than 25% myeloblasts were considered to be in partial remission (PR), as were patients fulfilling the bone marrow criteria of CR but without full recovery of peripheral blood platelet and/or white blood cell counts. Patients with persisting leukemic blasts in the bone marrow or blood or with leukemic regrowth within 4 weeks after initial response were considered as non-responders (NR). Patients dying within 6 weeks after the end of antileukemic therapy without evidence of leukemic regrowth were classified as early deaths (ED). Survival and event-free survival were measured by the time from the beginning of treatment to death, documentation of persisting leukemia, or relapse, respectively.

Statistics Numerical values were compared by the Fishe’s exact test and by the Student’s t-test. Response to antileukemic therapy was compared by an ordinal ␹2-test. Survival and event-free survival were calculated according to Kaplan–Meier estimates. Comparisons were carried out using the log-rank test.

Study conduct Prior to therapy all patients gave their informed consent for participation in the current evaluation after having been advised about the purpose and investigational nature of the study as well as of potential risks. The study design adhered to the Declaration of Helsinki and was approved by the ethics Leukemia


2086

committees of the participating institutions prior to its initiation. Results

Patients Within the first cohort, 117 evaluable cases with cytogenetic data both at diagnosis and at relapse were included into the first part of the current analysis. In all patients the initial diagnosis was de novo AML according to FAB criteria. The median age of the patients was 58 years (range, 19 to 80 years). The median duration from initial diagnosis to relapse amounted to 11.9 months (range, 2.7 months to 7.8 years). In most cases changes of karyotypes did not result in a switch into a prognostically different cytogenetic group. Thus, karyotypes were rated prognostically favorable in 13 (11%), intermediate in 73 (62%) and unfavorable in 25 (21%), respectively, both at diagnosis and at relapse. In six cases (5%), however, the karyotype at diagnosis was rated intermediate at diagnosis and unfavorable at relapse. Within the second cohort, 120 cases treated by S-HAI therapy for refractory or relapsed AML were evaluable for response to treatment and long-term outcome. In all patients the initial diagnosis was de novo AML according to FAB criteria. The median age of the patients was 55 years (range, 20 to 77 years). Seventeen percent had AML refractory to two induction courses, 21% had early relapses after a first CR of less than 6 months duration, while 41% and 18% had relapses after a first CR of 6 to 18 months duration and of more than 18 months duration, respectively. Two percent had second relapses. Karyotypes were prognostically favorable, intermediate, unfavorable and not available in 8%, 50%, 17% and 25% at diagnosis and in 8%, 49%, 21% and 22% at relapse, respectively.

Change in karyotype A change in the karyotype was observed in 44 of 117 cases (38%, cohort 1) while no changes occurred in 73 cases (62%, Table 1). The frequency of changes in karyotype were similar for patients with a normal karyotype at diagnosis as compared to those with an aberrant karyotype at diagnosis (20 of 56 cases (36%) vs 24 of 61 (39%), NS). Changes in the karyotype were significantly more frequent in patients with unfavorable aberrations at diagnosis as compared to other cases (15 of 25 (60%) vs 29 of 92 (32%), P = 0.0095). The duration from Table 1

Cases with same aberrations at diagnosis and at relapse Of the patients with the same chromosome aberrations both at diagnosis and at relapse, 21 had balanced abnormalities and 16 had unbalanced abnormalities including one case with complex aberrations (cohort 1). Within the group of 21 patients with balanced aberrations seven had an inv(16)/t(16;16), two had a t(8;21), and two had an inv(3). The distribution of aberrations within the group of 16 patients with unbalanced aberrations was heterogeneous.

Changes in cases with initially normal karyotypes In seven patients with normal karyotypes at diagnosis, balanced abnormalities developed while in 13 cases unbalanced aberrations occurred at relapse (cohort 1). All balanced translocations encountered at relapse were abnormalities that do not resemble those frequently observed in patients with AML at initial diagnosis (Table 3). Thus, balanced aberrations occurring frequently at diagnosis like t(8;21), inv(16), and translocations involving 11q23 did not develop in patients with a normal karyotype at diagnosis. Within the 13 patients with development of unbalanced aberrations, five had a complex aberrant karyotype at relapse and four had a trisomy 8, while in the other cases aberrations were present that are usually detected rarely in patients with AML at diagnosis. In six of the total of 20 of these patients chromosome 6 was affected. With the exception of the cases with complex aberrant karyotypes, abnormalities usually associated with secondary AML at initial diagnosis, ie aberrations of chromosomes 5 and 7

Distribution of karyotypes according to prognostic classification and according to type of change

Karyotype/prognostic classification (at diagnosis → at relapse) favorable → favorable (n = 13) intermediate → intermediate (n = 73) unfavorable → unfavorable (n = 25) intermediate → unfavorable (n = 6)

Leukemia

initial diagnosis to relapse was significantly shorter in cases with an evolution of the aberrant karyotype or with an unrelated karyotype at relapse as compared to cases with no alteration of the aberrant karyotype between diagnosis and relapse or with solely regression of aberrations at relapse (9.2 ± 4.4 vs 14.0 ± 8.5 months, P = 0.0081; Table 2). There were no differences in the duration from initial diagnosis to relapse between cases with and without a regression of the karyotype aberrations. Also, in cases with normal karyotypes at diagnosis, this duration was not influenced by changes occurring at relapse (Table 2). In patients with a normal karyotype at diagnosis, however, relapses tended to occur earlier if balanced as compared to unbalanced translocations developed (9.7 ± 4.3 vs 13.9 ± 6.1 months, P = 0.13). There were no differences in the age distribution between the respective groups.

normal → normal

normal → aberrant

aberrant → same aberrant

aberrant → evolution

aberrant → regression

aberrant → regr. + evol.

aberrant → unrelated aberr.

n = 36

n = 20

n = 37

n=9

n = 10

n=4

n=1

–

–

9

2

2

–

–

36

14

18

1

3

–

1

–

–

10

6

5

4

–

–

6

–

–

–

–

–


Table 2

2087

Duration from initial diagnosis to relapse according to type of change of karyotype

Karyotype at diagnosis → at relapse

Duration from initial diagnosis to relapse (months)

Median

normal → normal

normal → aberrant

aberrant → same aberrant

aberrant → evolution

aberrant → regression

aberrant → regr. + evol.

n = 36

n = 20

n = 37

n=9

n = 10

n=4

aberrant → unrelated aberr. n=1

13.2*

11.3*

13.0*

8.4**

10.8*

7.3**

4.8**

3.2–27.1 12.4 ± 5.9

2.8–44.8 14.2 ± 8.4

5.9–17.8 10.5 ± 4.4

5.9–37.4 13.4 ± 8.9

2.7–13.1 7.6 ± 3.8

– –

Range 2.8–94.7 Mean ± SD 15.4 ± 15.3

t-test (comparison * vs **): P = 0.0081.

Table 3 Aberrant karyotypes at relapse in cases with initially normal karyotypes at diagnosis

Changes in cases with regression of initially aberrant karyotypes

46,XY,t(2;6)(p13;p22) [15]; 46,XY [1] 46,XX,t(3;6)(p36;p11) [16] 46,XY,t(6;12)(q25;q11) [4]; 46,XY [24] 46,XX,t(4;12)(q11;p13) [7]; 46,XX [3] 48,XY,t(2;3)(q31;p21),+8,+8,t(10;17)(q22;q22),del(13)(q14q31) [13]; 46,XY [12] 46,X,del(X)(q11),t(7;18)(q22;p11.2) [8]; 46,XX [7] 46,XY,t(1;13)(q21;q14),t(2;22)(q11;q13) [9]; 46,XY [7] 46,XX,der(19)t(5;19)(?;p13.3) [19]; 46,XX [1] 46,XY,del(20)(q11.2) [17]; 46,XY,t(10;14)(q22;q32),del(20)(q11.2) [4]; 46,XY [1] 46,XY,add(16)(p13.3) [11]; 46,XY [9] 46,XY,dup(2)(p13p25) [13]; 46,XY [12] 46,XX,del(9)(q22) [7]; 46,XX [9] 46,XX,add(6)(p22) [3]; 46,XX [17] 48,XY,+8,+13 [2]; 48,XY,del(6)(p21p23) [8]; 46,XY [10] 46,XX,add(1)(p34),−4−6,+2xmar [cp4]; 46,XX,add(3)(q27),−4,−5,−6,+5–6xmar [cp5]; 46,XX multiple sporadic aberrations 47,XY,+13 [17]; 47,XY,dup(1)(q23q42),+13 [5]; 46,XY [2] 45,XY,−7,add(17)(p11) [2];45,idem,add(11)(q23)[6]; 45,idem,del(9)(p13),add(11)(q23) [6]; 45,idem,del(3) (q25),add(11)(q23) [4]; 46,XY [7] 49,XY,+8,+13,+19 [20] 47,XX,+8 [6]; 46,XX [10] 45,X,−Y,t(10;12)(p13;q13),-17,+mar1,+mar2 [13]; 46,XY [5]

In 10 patients with an aberrant karyotype at diagnosis, a regression of aberrations was observed at relapse (cohort 1, Table 5). Among these there were five cases with complex aberrations, in two cases inv(16) was diagnosed, and in the remaining three cases aberrations mainly comprised gains of whole chromosomes. The aberrations disappearing between diagnosis and relapse were mostly gains or losses of whole chromosomes or parts of them. In one case one of an initially doubled set of chromosomes had disappeared. There was no case with disappearance of a balanced translocation.

Cases with balanced translocations only are shown first followed by those with unbalanced abnormalities and complex aberrant karyotypes.

and aberrations involving 11q23, were not observed in these patients.

Changes in cases with evolution of initially aberrant karyotypes In nine patients one or more aberrations were present at relapse in addition to the aberrations detected at diagnosis (cohort 1). Of these patients, three had balanced translocations only and four had complex aberrations at diagnosis (Table 4). There was no consistent pattern of the type of aberrations acquired at relapse. However, in contrast to the cases with an initially normal karyotype, in the cases with an evolution of karyotype, acquisition of a trisomy was observed only once and in no case was a monosomy acquired.

Changes in cases with both evolution and regression of initially aberrant karyotypes There were four patients in whom both evolution and regression of initially aberrant karyotypes were observed (cohort 1). All of them had complex aberrations both at diagnosis and at relapse. There was no consistent pattern in the types of changes.

Cases with length mutations of the FLT3 gene or with internal tandem duplications of the MLL gene In 16 patients length mutations of the FLT3 gene (n = 15) and/or internal tandem duplications of the MLL gene (n = 4) were detected at relapse (cohort 1). In all of them the same molecular alterations had been present at diagnosis. Nine of these patients had a normal karyotype at diagnosis which was normal (n = 6) or aberrant (n = 3) at relapse. In the other patients, the same cytogenetic aberrations were observed at diagnosis and at relapse (n = 3), or an evolution (n = 2) or a regression (n = 2) of the karyotype had occurred at relapse.

Impact of karyotype aberrations on prognosis The impact of karyotype aberrations both at diagnosis and at relapse on the prognosis of the patients was analyzed in the second cohort comprising 120 cases. Separations of response rates according to karyotype aberrations at relapse resulted in significant differences while separation according to karyotype at diagnosis did not (Table 6). Rates of complete remission were highest in cases with favorable karyotypes and Leukemia


2088

Table 4

Evolution of aberrations from diagnosis to relapse (changes in bold)

Karyotype at diagnosis

Karyotype at relapse

46,XY,t(8;21)(q22;q22) [25] 46,XX,t(8;21)(q22;q22) [20]

46,XX,t(9;11)(p22;q23) [25] 47,XX,t(11;17)(q23;q25),+20 [18]; 51,XX,+5,+6,t(11;17)(q23;q25),+19,+20,+21 [4]; 46,XX [3] 47,XX,del(5)(q15q33),+11 [20] 46,XY,del(6)(q21q23) [18]; 46,XY [12] 44,XY,der(1)t(1;5)(p36;q13),−5,−21,+mar [3] 46,XX,del(5)(q13q31),−17,der(20)t(17;20)(q21;q13),+mar1[2]; 45, idem,del(4)(q23q31),−12[5]; 43,idem, del(4)(q23q31),−7,−12,−16 [7];44,idem,del(4)(q23q31),−7,−12,−16,+mar2 [10] 42,XY,der(5)del(5)(p15)del(5)(q13),−7,der(12)t(12;16)(p11;p11), ins(12;5)(p13;q?q?),−16,−18, der(22;22)(p10;p10) [10]; 42,idem,i(11)(q10) [7]; 46,XY [3]

46,XY,t(1;5)(q21;q33),t(8;21)(q22;q22)[6]; 46,XY,t(1;5)(q21;q33),t (8;21)(q22; q22), del(11)(p13)[4] 46,XX,del(7)(q22q32),t(8;21)(q22;q22) [6]; 46,XX,der(1)t(1;15)(p36;q22),t(8;21)(q22;q22),t(10;18) (q22;q12)[10]; 46,XX,inv(1)(p35q21),t(1;7)(p34;q35),t(8;21) (q22;q22)[2]; 46,XX, t(5;17)(q22;p13),del(6)(q23),t(8;21)(q22;q22),dup(15)(q11q26),der (16)t(5;16)(?;p13.1)[2] 46,XX,der(9)t(9;11)(p22;q23),der(11)t(9;11)(p22;q23)del(11) (p11p13)[15]; 46,XX [5] 51,XX,+5,+6,t(11;17)(q23;q25),+19,+20,+21 [17]; 52,XX,+5,+6,+8,t(11;17)(q23;q25),+19,+20,+21[3] 47,XX,del(5)(q15q33),+11 [9]; 47,idem,del(17)(q23)[9]; 47,idem, add(7)(q3?4) [2] 46,XY,dup(2)(q31q35),del(6)(q21q23) [15]; 46,XY [5] 44,XY,der(1)t(1;5)(p36;q13),−5,del(7)(q22),der(16)t(16;21) (q12.1;q11.2),add (21)(q22) [10]; 46,XY [10] 43–46,XX,del(3)(q24),del(4)(q23q31), del(5)(q13q31),−7,−12,−16,−17,der(20)t(17;20)(q21;q13),2xmar[2] 42,XY,der(5)del(5)(p15)del(5)(q13),−7,i(11)(q10),der(12)t(12;16) (p11;p11), ins(12;5)(p13;q?q?),−16,−18, der(22;22)(p10;p10) [15]; 42,idem,t(1;11)(p36;p11) [4]; 46,XY [1]


Table 5

Regression of aberrations from diagnosis to relapse (changes in bold)

Karyotype at diagnosis

Karyotype at relapse

46–49,XY,del(5)(q13q3?3),del(6)(q13),−16,add(19)(p13), +add(19)(p13),+3-5xmar,+r [cp20] 90-92,XXYY,der(5)t(5;7)(q13;?),−7,+8; 92,XXYY [8]; 46,XY [7] 42,XX,add(1)(p13),−5,−13,−14,der(17)t(5;17)(p11;p11),−18,−22, +mar [17]; 46,XX [3] 49–52,XY,+1,+add(6)(q15),+8,+11,−14,−17,−17,+22,+2-5mar [cp18]; 46,XY [3] 87,XXYY,del(1)(p13)x2,−2,add(3)(p25)x2,del(5)(q13q31)x2, add(6)(q27)x2,−7,−7,−16,−17 [cp13]; 46,XY [5] 46,XY,der(10)t(10;16)(q26;p12)inv(16)(p13q22),der(16)t(10;16) (q26;p12) [9]; 47,idem,+8 [2]; 48,idem,+8,+21 [3]; 46,XY [12] 46,XY,inv(16)(p13q22) [2]; 47,XY,+8,inv(16)(p13q22) [1]; 49,XY,del(1)(q21),+8,+13,inv(16)(p13q22),+21 [4]; 46,XY [8] 48,XY,+8,+10 [10] 48,XY,+8,+11 [10]; 46,XY [5] 46,XX,dup(4)(q31.3q35) [3]; 47,XX,dup(4)(q31.3q35),+21 [4]

46–48,XY,del(5)(q13q3?3),del(6)(q13),−16, add(19)(p13),+3– 5xmar,+r [cp8] 45,XY,−7 [18]; 46,XY [2] 45,XX,−5,der(17)t(5;17)(p11;p11) [17] 49–52,XY,+1,+8,+11,−14,−17,−17,+22,+2-5mar [cp12]; 46,XY [5] 45,XY,del(1)(p13),add(3)(p25),del(5)(q13q31), add(6)(q27),−7 [4]; 90,idem x2 [6]; 46,XY [5] 46,XY,der(10)t(10;16)(q26;p12)inv(16)(p13q22), der(16)t(10;16)(q26;p12) [12]; 47,idem,+8 [1]; 46,XY [2] 46,XY,inv(16)(p13q22) [1]; 47,XY,+8,inv(16)(p13q22) [14] 46,XY [25] 47,XY,+11 [16]; 46,XY [4] 46,XX,dup(4)(q31.3q35) [4]; 46,XX [5]


Table 6

Response to S-HAI chemotherapy according to karyotype at diagnosis and at relapse

Karyotype At diagnosis*

Complete remission Partial remission Non-response Early death Ordinal ␹2-test: *P = 0.32; **P = 0.05. Leukemia

At relapse**

Favorable n = 10

Intermediate n = 61

Unfavorable n = 20

Favorable n = 10

Intermediate n = 59

Unfavorable n = 25

80% 0% 10% 10%

49% 3% 34% 13%

30% 5% 45% 20%

70% 0% 10% 20%

53% 2% 36% 10%

24% 8% 40% 28%


2089

Figure 2 diagnosis.

Overall survival according to karyotype aberrations at

Figure 3 Event-free survival according to karyotype aberrations at diagnosis.

lowest in cases with unfavorable karyotypes, mainly due to higher rates of non-response in the latter group of patients. Patients with unfavorable karyotypes at diagnosis had an inferior overall survival and event-free survival as compared to cases with prognostically intermediate or favorable karyotypes (7.8 months vs 5.2 months, P = 0.03; 3.2 months vs 1.4 months, P = 0.02; Figures 2 and 3). There were stronger differences for the respective comparison according to aberrations at relapse. Patients with unfavorable karyotypes at relapse had an inferior overall survival and event-free survival as compared to cases with prognostically intermediate or favorable karyotypes (7.8 months vs 4.1 months, P = 0.01; 3.1 months vs 1.2 months, P = 0.002; Figures 4 and 5). There were no differences in the prognosis for patients with prognostically intermediate karyotypes as compared to those with favorable ones at either time point. Discussion A genomic instability may contribute to the development of refractoriness of AML to anti-leukemic therapy. The results of the current study support this hypothesis and provide evidence for correlations between the currently used prognostic grouping of patients with AML according to the karyotype aberrations of their disease, the stability of these karyotypes and the duration of the interval from diagnosis to relapse. Hence, an evolution of the karyotype, which is considered to be related to genomic instability, occurred significantly more

Figure 4 relapse.

Overall survival according to karyotye aberrations at

Figure 5 at relapse.

Event-free survival according to karyotype aberrations

often in patients with unfavorable aberrations at diagnosis as compared to patients with favorable or prognostically intermediate aberrations. Furthermore, in these cases the duration of the interval from diagnosis to relapse was significantly shorter as compared to all other cases. In addition, the data on the prognostic impact of karyotype aberrations indicate that aberrations detected at relapse influence the prognosis more strongly than aberrations detected at diagnosis. This suggests that a change in karyotype between diagnosis and relapse may be of major relevance in this regard. The types of changes in karyotype appear to be less significant as compared to the occurrence per se of an evolution of initially aberrant karyotypes. Along this line, as would have been anticipated, the acquisition at relapse of aberrations associated with an unfavorable prognosis occurred in patients with complex aberrations at diagnosis and not in patients with other karyotypes. In contrast, the other cases acquired abnormalities which are usually observed very infrequently at diagnosis or which at diagnosis are considered secondary abnormalities (eg trisomies). This is in accordance with the concept of karyotype evolution not only in patients with initially aberrant karyotypes but also in patients with initially normal karyotypes. This concept is further supported by the data on patients with alterations of the FLT3 or of the MLL genes which were consistently detected both at diagnosis and at relapse irrespective of the acquisition of additional cytogenetic aberrations at relapse. Five previous reports have been published addressing this topic. In two preliminary ones, eight and six patients, respectively, have been analyzed cytogenetically at diagnosis and at Leukemia


2090

relapse.25,26 Changes occurred in six of eight and in one of six, respectively, mostly being gains of trisomies. In the other three reports, 168, 103 and 60 patients, respectively, were analyzed at both time points.27–29 Changes were reported in 52%, 61% and 28%, respectively, which is in the range of the 38% observed in the current series. Changes occurred in patients with initially normal karyotypes in 30%, 51% and 25% (current series: 36%) and in patients with initially aberrant karyotypes in 67%, 68% and 31% (current series: 39%). In the current series an evolution of the karyotype occurred more frequently in patients with unfavorable aberrations at diagnosis, which was also observed in the only other report providing detailed data (seven of 10 cases with evolution had unfavorable karyotypes at diagnosis).29 In the previous reports, the acquisition of aberrations typical for secondary AML, ie aberrations of chromosomes 5 and 7 and aberrations of 11q23, or of t(8;21) in patients with normal karyotypes at diagnosis were reported in 25%, 7% and 0% (current series: 0%) which is also in accordance with the concept of karyotype evolution rather than occurrence of secondary leukemia being artificially diagnosed as relapsed disease. This is further supported by the types of aberrations frequently acquired being considered secondary ones when observed at diagnosis, ie trisomy 8, trisomy 13. Also in two of the reports, complex aberrant karyotypes appeared to be particularly instable.28,29 Contrasting patients with an evolution of karyotype, cases with aberrations independent on the initially diagnosed karyotype seem to have secondary or therapy-related AML. Only one of the reports provided data on a comparison of the remission duration for cases with different types of changes; however, no differences were detected.27 In contrast, the current series demonstrated that the evolution of aberrant karyotypes was associated with a significantly shorter interval from diagnosis to relapse as compared to other cases. This particular analysis, however, may not have been performed in the previous study.27 The importance of changes in karyotype for the prognosis is further supported by aberrations detected at relapse influencing the patients’ outcome more strongly as compared to aberrations detected at diagnosis. Recent analyses have proved the prognostic value of cytogenetics at diagnosis in patients undergoing salvage therapy;8,30 however, no conclusive data on the value of cytogenetics at relapse is available yet. Thus, the results of the current study strongly suggest the need to gain further data on karyotype changes between diagnosis and relapse and to analyze their prognostic significance. The data discussed above may in total suggest that the differences of the prognosis in patients with normal karyotypes as compared to those with unfavorable aberrations are due to differences in the stability of the karyotype. Hence, in cases with more unstable karyotypes the acquisition of aberrations conferring resistance to anti-leukemic therapy may occur more frequently and may thus contribute to the manifestation of refractory disease. Additional cytogenetic and molecular analyses, in particular of genetic changes between diagnosis and relapse, are needed to better characterize the phenotype of refractory disease and to potentially derive novel treatment approaches from the insights gained.

References 1 Rowley JD. Molecular genetics in acute leukemia. Leukemia 2000; 14: 513–517. Leukemia

2 Melnick A, Licht JD. Deconstructing a disease: RARalpha, its fusion partners, and their roles in the pathogenesis of acute promyelocytic leukemia. Blood 1999; 93: 3167–3215. 3 Downing JR. The AML1-ETO chimaeric transcription factor in acute myeloid leukaemia: biology and clinical significance. Br J Haematol 1999; 106: 296–308. 4 Goger M, Gupta V, Kim WY, Shigesada K, Ito Y, Werner MH. Molecular insights into PEBP2/CBF beta-SMMHC associated acute leukemia revealed from the structure of PEBP2/CBF beta. Nat Struct Biol 1999; 6: 620–623. 5 Schoch C, Haferlach T, Haase D, Fonatsch C, Loffler H, Schlegelberger B, Staib P, Sauerland MC, Heinecke A, Buchner T, Hiddemann W. Patients with de novo acute myeloid leukaemia and complex karyotype aberrations show a poor prognosis despite intensive treatment: a study of 90 patients. Br J Haematol 2001; 112: 118-126. 6 Buchner T, Hiddemann W, Wormann B, Loffler H, Gassmann W, Haferlach T, Fonatsch C, Haase D, Schoch C, Hossfeld D, Lengfelder E, Aul C, Heyll A, Maschmeyer G, Ludwig WD, Sauerland MC, Heinecke A. Double induction strategy for acute myeloid leukemia: the effect of high-dose cytarabine with mitoxantrone instead of standard-dose cytarabine with daunorubicin and 6-thioguanine: a randomized trial by the German AML Cooperative Group. Blood 1999; 93: 4116–4124. 7 Grimwade D, Walker H, Oliver F, Wheatley K, Harrison C, Harrison G, Rees J, Hann I, Stevens R, Burnett A, Goldstone A. The importance of diagnostic cytogenetics on outcome in AML: analysis of 1,612 patients entered into the MRC AML 10 trial. The Medical Research Council Adult and Children’s Leukaemia Working Parties. Blood 1998; 92: 2322–2333. 8 Kern W, Schoch C, Haferlach T, Braess J, Unterhalt M, Wormann B, Buchner T, Hiddemann W. Multivariate analysis of prognostic factors in patients with refractory and relapsed acute myeloid leukemia undergoing sequential high-dose cytosine arabinoside and mitoxantrone (S-HAM) salvage therapy: relevance of cytogenetic abnormalities. Leukemia 2000; 14: 226–231. 9 Gale RP, Horowitz MM, Weiner RS, Ash RC, Atkinson K, Babu R, Dicke KA, Klein JP, Lowenberg B, Reiffers J. Impact of cytogenetic abnormalities on outcome of bone marrow transplants in acute myelogenous leukemia in first remission. Bone Marrow Transplant 1995; 16: 203–208. 10 Schneider E, Cowan KH, Bader H, Toomey S, Schwartz GN, Karp JE, Burke PJ, Kaufmann SH. Increased expression of the multidrug resistance-associated protein gene in relapsed acute leukemia. Blood 1995; 85: 186–193. 11 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, Sultan C. Proposal for the recognition of minimally differentiated acute myeloid leukaemia (AML-MO) (see comments). Br J Haematol 1991; 78: 325–329. 12 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, Sultan C. 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. 13 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, Sultan C. A variant form of hypergranular promyelocytic leukemia (M3) (letter). Ann Intern Med 1980; 92: 261–261. 14 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, Sultan C. Proposals for the classification of the acute leukaemias. French– American–British (FAB) co-operative group. Br J Haematol 1976; 33: 451–458. 15 Kern W, Schoch C, Haferlach T, Braess J, Wormann B, Buchner T, Ludwig WD, Hiddemann W. Prognostic significance of cytogenetics at relapse in patients receiving sequential high-dose cytosine arabinoside and idarubicin (S-HAI) + fludarabine therapy for refractory and relapsed acute myeloid leukemia. Blood 2000; 96: 504a. 16 Kern W, Aul C, Maschmeyer G, Scho¨nrock-Nabulsi R, Ludwig WD, Bartholomaus A, Bettelheim P, Wormann B, Buchner T, Hiddemann W, for the German AML Cooperative Group. Superiority of high-dose over intermediate-dose cytosine arabinoside in the treatment of patients with high-risk acute myeloid leukemia: results of a age-adjusted prospective randomized comparison. Leukemia 1998; 12: 1049–1055. 17 Schoch C, Bursch S, Kern W, Schnittger S, Hiddemann W, Haferlach T. Gain of an isochromosome 5p: a new recurrent chromo-


18

19

20

21

22

23

some abnormality in acute monoblastic leukemia. Cancer Genet Cytogenet 2001; 127: 85–88. Schnittger S, Schoch C, Kern W, Staib P, Wuchter C, Sauerland C, Serve H, Buchner T, Haferlach T, Hiddemann W. FLT3 length mutations in AML: correlation to cytogenetics, FAB-subtype, and prognosis in 652 patients. Blood 2000; 96: 826a. Schnittger S, Wormann B, Hiddemann W, Griesinger F. Partial tandem duplications of the MLL gene are detectable in peripheral blood and bone marrow of nearly all healthy donors. Blood 1998; 92: 1728–1734. Buchner T, Hiddemann W, Wormann B et al. Threapeutic outcome in AML is mainly determined by cytogenetics, LDH in serum, early response and, in a poor risk subgroup, by intensified induction treatment. Blood 1997; 90 (Suppl. 1): 504a. Schoch C, Haase D, Haferlach T, Gudat H, Buchner T, Freund M, Link H, Lengfelder E, Wandt H, Sauerland MC, Loffler H, Fonatsch C. Fifty-one patients with acute myeloid leukemia and translocation t(8;21)(q22;q22): an additional deletion in 9q is an adverse prognostic factor. Leukemia 1996; 10: 1288–1295. Schoch C, Haase D, Haferlach T, Freund M, Link H, Lengfelder E, Loffler H, Buchner T, Fonatsch C. Incidence and implication of additional chromosome aberrations in acute promyelocytic leukaemia with translocation t(15;17)(q22; q21): a report on 50 patients. Br J Haematol 1996; 94: 493–500. Schoch C, Haase D, Fonatsch C, Haferlach T, Loffler H, Schlegelberger B, Hossfeld DK, Becher R, Sauerland MC, Heinecke A, Wormann B, Buchner T, Hiddemann W. The significance of trisomy 8 in de novo acute myeloid leukaemia: the accompanying

24

25 26 27 28

29 30

chromosome aberrations determine the prognosis. German AML Cooperative Study Group. Br J Haematol 1997; 99: 605–611. Yates J, Glidewell O, Wiernik P, Cooper MR, Steinberg D, Dosik H, Levy R, Hoagland C, Henry P, Gottlieb A, Cornell C, Berenberg J, Hutchison JL, Raich P, Nissen N, Ellison RR, Frelick R, James GW, Falkson G, Silver RT, Haurani F, Green M, Henderson E, Leone L, Holland JF. Cytosine arabinoside with daunorubicin or adriamycin for therapy of acute myelocytic leukemia: a CALGB study. Blood 1982; 60: 454–462. Rowley JD, Potter D. Chromosomal banding patterns in acute nonlymphocytic leukemia. Blood 1976; 47: 705–721. Li YS, Khalid G, Hayhoe FG. Correlation between chromosomal pattern, cytological subtypes, response to therapy, and survival in acute myeloid leukaemia. Scand J Haematol 1983; 30: 265–277. Estey E, Keating MJ, Pierce S, Stass S. Change in karyotype between diagnosis and first relapse in acute myelogenous leukemia. Leukemia 1995; 9: 972–976. Garson OM, Hagemeijer A, Sakurai M, Reeves BR, Swansbury GJ, Williams GJ, Alimena G, Arthur DC, Berger R, de la Chapelle A. Cytogenetic studies of 103 patients with acute myelogenous leukemia in relapse. Cancer Genet Cytogenet 1989; 40: 187–202. Testa JR, Mintz U, Rowley JD, Vardiman JW, Golomb HM. Evolution of karyotypes in acute nonlymphocytic leukemia. Cancer Res 1979; 39: 3619–3627. Kantarjian HM, Keating MJ, Walters RS, McCredie KB, Freireich EJ. The characteristics and outcome of patients with late relapse acute myelogenous leukemia. J Clin Oncol 1988; 6: 232–238.

2091

Leukemia