and acute myeloid leukemia (AML) - Nature

Leukemia (1997) 11, 964–970  1997 Stockton Press All rights reserved 0887-6924/97 $12.00

Detection of 16 p deletions by FISH in patients with inv(16) or t(16;16) and acute myeloid leukemia (AML) D Martinet1, D Mu¨hlematter1, M Leeman1, V Parlier1, U Hess2, J Gmu¨r3 and M Jotterand1 1

Division Autonome de Geńe´tique Me´dicale, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne; 2 on behalf of the SIAK/SAKK Leukemia Group, Klinik C fu¨r Innere Medizin, Kantonsspital, St-Gallen; and 3chairman of the SAKK/Hovon AML protocols, Abteilung Ha¨ matologie, Departement Innere Medizin Universita¨tsspital, Zu¨rich, Switzerland

Deletions of sequences centromeric to the p-arm breakpoint have been described in a subset of patients with inv(16) and acute myeloid leukemia (AML) and reported to be associated with a relatively good prognosis. We have investigated 16 p deletions in a cohort of 15 patients with AML and inv(16) or t(16;16) and compared non-deletion and deletion patients in terms of clinical course. Patients were studied by fluorescence in situ hybridization (FISH) using cosmid zit14 as a probe to detect the presence of 16 p deletions in metaphase chromosomes of leukemic cells. While seven patients (47%) revealed no evidence of a deletion, five patients (33%) presented 16 p deletions, thus bringing further support to the relatively frequent occurrence of this event in inv(16) patients. Remarkably, two patients with inv(16) and one patient with t(16;16) showed a mosaicism of deletion and non-deletion metaphases suggesting the presence of two distinct leukemic cell populations. Results let us assume that 16 p deletions are not restricted to inv(16) and may occur subsequently to inv(16) or t(16;16). The presence of a 16 p deletion in a case of inv(16) associated with CBFB–MYH11 transcript type E indicates that deletions are not limited to CBFB–MYH11 transcript type A rearrangements. Survival of deletion patients was compared with that of nondeletion and mosaic ones. No significant differences were observed. The advantage of FISH for enumerative and quantitative assessment of submicroscopic rearrangements of clinical significance is further emphasized. Keywords: inv(16)(p13q22); t(16;16)(p13;q22); 16 p deletions; FISH; acute myeloid leukemia; prognosis

Introduction In 1983, Arthur and Bloomfield1,2 reported five patients with acute myeloid leukemia (AML) of myelomonocytic type (M4), increased marrow eosinophils and a structurally abnormal chromosome 16. Soon after, a pericentric inversion of chromosome 16, inv(16)(p13q22), was described in 18 patients who had M4 leukemia and morphologically abnormal eosinophils with irregular, basophilic-staining granules and distinct cytochemical properties.3,4 A related chromosome abnormality, t(16;16)(p13;q22), was reported in one patient with similar characteristics.5 Further studies have suggested that these two abnormalities were uniquely associated with M4 and abnormal eosinophils and the correlation between the karyotype and morphology has led to a subcategory in the FAB classification, namely M4Eo.6 However, both rearrangements have been reported in other types of myeloid disorders.7,8 In most studies, the presence of inv(16)/t(16;16) has been associated with a high cure rate with standard chemotherapy and a relatively favorable prognosis.9–13 Inv(16) and t(16;16) have been shown to involve, on 16p13, a smooth muscle myosin heavy chain gene (MYH11) and, on

Correspondence: M Jotterand, Division autonome de geńe´tique me´ dicale, Centre Hospitalier Universitaire Vaudois, 1011 Lausanne, Switzerland Received 15 January 1997; accepted 3 March 1997

16q22, a gene encoding for the b subunit (CBFB) of the core binding transcription factor (CBF) whose DNA target sequences have been identified in the promoter–enhancer regions of hematopoietic specific genes.8,14 Both rearrangements result in the formation of two abnormal fusion genes, one consisting of the 59 end of CBFB and the 39 end of MYH11 (CBFB–MYH11), the other of the 59 end of MYH11 and the 39 end of CBFB (MYH11–CBFB). A deletion of sequences centromeric to the p-arm breakpoint has been observed in addition to inv(16) in 14 out of 77 patients studied.15–18 Deletions have been shown to involve a region of 160 kb to 350 kb and to affect the 59 part of the MYH11 gene.18 In the deletion cases studied, no MYH11– CBFB transcript was identified by RT-PCR.17 As these patients exhibit the same phenotype as non-deletion ones, it can be concluded that it is the CBFB–MYH11 fusion product that is critical for the inv(16)/t(16;16)-associated leukemogenesis.18 Deletions were reported to be associated with a significantly longer time from diagnosis until death or relapse.16,17 As deletions were found to include at least the 59 part of the multidrug resistance-associated protein gene (MRP), the authors hypothesized that the partial loss of one MRP allele accounts for the relatively favorable outcome in this group. However, in another study, no difference in prognosis was observed between deletion and non-deletion patients.18 To assess further the incidence of the 16 p deletions and to improve the understanding of their prognostic significance, we studied 15 patients with AML and inv(16) or t(16;16) by fluorescence in situ hybridization (FISH) using the zit14 cosmid as a probe. Clinical, cytogenetic and FISH data are reported here. Patients, materials and methods

Patients Fifteen patients with either inv(16) or t(16;16) were diagnosed at the divisions of hematology of the University Hospitals of Zu¨rich (four patients), Basel (three), Bern (three), Lausanne (two) and of the Canton Hospitals of Sankt-Gallen (two) and Aarau (one) between 1992 and 1995 (Table 1). Each patient was assigned a unique patient number for this and previous publications.19–21 Additionally, 10 patients with AML (nine) or acute lymphoblastic leukemia (ALL) (one) and normal karyotypes or aberrations other than inv(16) or t(16;16) were investigated by FISH and served as controls. Informed consent was obtained from each patient. The diagnosis and classification of patients were based on morphologic and cytochemical examination of peripheral blood (PB) films, bone marrow (BM) aspirate and biopsy specimens obtained before therapy according to criteria proposed by the FAB and the MIC cooperative study groups.3,4,6,22–24 Patients were followed until death or to May 1996.

Detection of 16 p deletions in inv(16)/t(16;16) patients D Martinet et al

Table 1

Clinical data Case

965

Clinical outcome, cytogenetic findings and FISH data in 15 patients with inv(16)/t(16;16) and AML

Patient No.

Age/Sex Diagnosis FAB

Conventional cytogenetics Survival months

Karyotype

% of abnormal metaphases [metaphases analyzed]

FISH using zit 14 probe % of metaphases with a single signal c [metaphases scored]

Patient status

1 2 3 4 5

2353/92 1952/93 52/95 1473/95 1920/95

36/M 25/F 30/M 55/M 29/F

M4Eo M2Eo M4Eo M4Eo M4Eo

40+ 8 16+ 10+ 7+

inv(16) inv(16)b inv(16) inv(16)b inv(16)

100 97 100 100 96

[25] [31] [30] [25] [23]

90 88 94 96 89

[10] [17] [65] [26] [27]

del del del del del

6 7 8 9 10 11 12

1882/92 568/93 1147/93 1552/93 270/94 528/94 987/95

55/M 25/M 54/M 44/M 31/M 14/M 67/F

M4Eo M5 M4Eo M4 M4Eo AMLa M4Eo

13 38+ 2 5 days 2 27+ 12+

t(16;16) inv(16) inv(16) inv(16) inv(16)b inv(16) inv(16)b

100 92 100 100 100 75 96

[19] [13] [12] [10] [25] [28] [26]

4 0 0 9 4 9 10

[26] [3] [25] [11] [54] [11] [21]

non-del non-del non-del non-del non-del non-del non-del

13 14 15

184/93 1427/93 434/94

48/F 40/M 23/F

M4Eo M4 M4Eo

39+ 32+ 26+

inv(16) t(16;16) inv(16)b

92 [24] 100 [20] 100 [28]

25 [16] 35 [17] 29 [14]

mosaic mosaic mosaic

a

Secondary to T and B cell ALL. Presence of additional chromosome abnormalities. c A signal is defined as either one or two spot(s) on the same chromosome. del, deletion; survival +, still alive after x months survival. b

Conventional cytogenetics Chromosomes were prepared from PB or BM mononuclear cells and stained in G bands as described.21,25,26 The number of metaphases analyzed fully varied between 10 and 31 according to material available.

FISH Slides were prepared from methanol/acetic acid fixed cell pellets, fresh or stored at −20°C for up to 3 years. To detect deletions proximal of the p-arm breakpoint, one-color FISH was performed on metaphase chromosomes with the pHuR 195 probe specific for the heterochromatic region of chromosome 1627 and the cosmid probe zit14.15 To confirm results obtained in mosaic patients, the zit18 cosmid probe was used in addition to the pHuR 195 and zit14 probes in dual-color FISH. Both zit14 and zit18 probes, each spanning 40 kb, overlap and are specific for sequences located in the deletion region, 50 kb centromeric to the 16 p-arm breakpoint. Zit14 hybridizes to sequences slightly proximal to zit18 target sequences. Probes were labeled by nick-translation with biotin-16-dUTP (pHuR 195 and zit14 probes) or digoxygenin (DIG)-11-dUTP (zit18 probe). Slides were denatured at 80°C in formamide 70%/2 × SSC for 5 min, dehydrated in ice-cold ethanol and air-dried. pHuR 195 (4 ng/ml) and cosmid probes (20 ng/ml) including 50 × human Cot 1 DNA were dissolved in 50% formamide/2 × SSC/10% dextran sulfate, denatured at 80°C for 5 min. DNA was allowed to reanneal at 37°C for 10 min and 15 ml hybridization mixture was applied under a 22 × 26 mm coverslip. After overnight hybridization at 37°C in a moist chamber, slides were washed in 50% formamide/2 × SSC at 42°C, then in 1 × SSC at 60°C three times 5 min each and in the washing buffer (WB)

4 × SSC/0.05% Tween 20 for 3 min at room temperature (RT). After a preincubation in WB/5% non-fat dry milk (NFDM) at RT for 10 min, slides were washed twice 5 min each in WB. Biotin-labeled probes were detected after an incubation with 6 mg/ml fluorescein isothiocyanate (FITC)-avidin (DCS, Vector, Burlingame, CA, USA) and an amplification with 5.5 mg/ml biotinylated goat anti-avidin (Vector) followed by an incubation with FITC-avidin.28 All reagents were diluted in 4 × SSC/5% NFDM. Each incubation was performed in a moist chamber at 37°C for 30 min and followed by three washes of 5 min in WB. For dual-color experiments, signal revelation was performed as described by Dauwerse et al,29 except that FITC-avidin was used for detecting biotin-labeled probes and 10 mg/ml sheep anti-mouse Cy3 (Sigma, Buchs, Switzerland) for detecting mouse anti-DIG. DNA was counterstained with DAPI or propidium iodide and slides were mounted with an anti-fade solution (Vectashield; Vector). Hybridization signals were defined as follows. Either one or two spots on the same chromosome were considered as one signal. As a consequence, metaphases with two spots on both chromosomes 16 (2/2), two spots on one chromosome 16 and one spot on the other (2/1), or one spot on both chromosomes 16 (1/1) were considered as harboring two signals. Metaphases with two or one spot(s) on one chromosome 16 only (2/0 and 1/0 respectively) were considered as presenting one signal. Hybridization signals were scored by at least two observers on a Zeiss Axioplan fluorescence microscope (Zeiss, Zu¨rich, Switzerland) equipped with FITC and rhodamine filters, as well as with double (FITC/rhodamine) and triple (FITC/rhodamine/DAPI) band pass filters. Photographs were taken with a Kodak Ektachrome 320T film (Kodak, Lausanne, Switzerland).


966

Statistical analysis

Conventional cytogenetics

Continuous variables were compared using the non-parametric Wilcoxon rank-sum test. Comparisons of counts were performed using the Fisher’s exact test. A P value ,0.05 was considered statistically significant. Survival was calculated from the date of diagnosis to the date of last follow-up or death. Survival curves were established by the Kaplan–Meier method and differences in the survival of subgroups assessed by log-rank test. All statistical computations were performed using STATA 4.0 (Stata Corporation, College Station, TX, USA) for Apple Macintosh.

Thirteen patients presented with inv(16)(p13q22) and two with t(16;16)(p13;q22) (Table 1). Ten patients had inv(16) or t(16;16) as a single defect. The remaining patients had additional chromosome abnormalities. A clone with a normal karyotype was observed in six patients. Cytogenetic analysis of BM obtained in patient 13 during complete remission (39 months after diagnosis) showed disappearance of the abnormal clone. Cytogenetic findings will be described in detail elsewhere (Jotterand et al, manuscript in preparation).

FISH Results

Patients’ characteristics and clinical features Study patients were five women and 10 men (Table 1). Their age ranged from 14 to 67 years. Median age was 36 years. At diagnosis, 10 patients were categorized as M4Eo. Two additional patients had the features of M4. Other FAB categories observed included one patient with M5 and one patient with M2Eo. One patient, a 14-year-old male, had an AML of probable M1 type secondary to an early B CD10+ ALL which occurred 2 years after a T cell ALL diagnosed in 1988; at the time of ALL, this patient did not reveal any evidence of inv(16). Patients will be described in detail elsewhere (Jotterand et al, manuscript in preparation). Fourteen patients received standard chemotherapy according to the Hovon 4A – SAKK 30/92 (12 patients, seven of them included in the Dutch–Swiss study) and Hovon 29 – SAKK 30/95 (two patients, both of them included in the Dutch–Swiss study) protocols. The first cycle of remission induction treatment consisted of cytarabin for 7 days (200 mg/m2) and of either daunomycin (40–60 mg/m2) for 3 days (11 patients, 4 days in one elderly patient) or idarubicin for 3 days (12 mg/m2, two patients). The second cycle consisted of cytarabin for 6 days (1000 mg/m2) and amsacrine for 3 days (120 mg/m2). Patient 14 underwent allogeneic BM transplantation 5 months after diagnosis. The young patient with secondary AML (case 11) received an induction therapy comprising etoposide/cytarabin and etoposide/endoxan, alternatively. Median overall survival of the whole group has not been reached. Survival curves according to the presence/absence of the deletions were not significantly different (P = 0.17) (Figure 1). Median survival was 13.4 months in the seven nondeletion patients (four deceased patients); it was not reached in the five deletion patients (one deceased patient) and in the three mosaic cases (no deceased patient).

Figure 1 patients.

Kaplan–Meier survival curves in 15 inv(16)/t(16:16)

Based on zit14 results, three distinct groups were observed among the inv(16)/t(16;16) patients (Figures 2 and 3; Tables 1 and 2). In the first group (five patients), the proportion of metaphases with one signal ranged from 88.2 to 96.2% (mean 92.4%) and patients were considered as deletion patients; the 3.4% metaphases with signals on both chromosomes 16 are probably due to the presence of some normal metaphases as demonstrated by conventional cytogenetics. In the second group (seven patients), the proportion of metaphases with signals on both chromosome 16 ranged from 90.5 to 100% (mean 94.7%); as this proportion is similar to that observed in the control group, patients were considered as non-deletion patients (Table 2). Case 7 was included in this group as two signals were observed in all three metaphases available. In the third group (three patients), the proportion of metaphases with a single signal ranged from 25.0 to 35.3% (mean 29.8%) and

Figure 2 Schematic representation of p-arm and q-arm breakpoints, probes localization (zit14 and pHuR 195) and deletion region in inv(16), inv(16)/deletion and t(16;16), t(16;16)/deletion situations (drawings are not to scale).


967

Figure 3 FISH analysis in control, inv(16), inv(16)/deletion and t(16;16)/deletion cells using the zit14 and pHuR 195 probes. (a) normal chromosomes 16; (b) metaphase cell from inv(16) patient showing signal on both chromosomes 16; (c–e) metaphase cells from inv(16)/deletion patients demonstrating a single signal from the normal chromosome 16; (f) metaphase cell from t(16;16)/deletion mosaic patient showing loss of signal from one rearranged chromosome 16 (16 p+).

patients were considered as mosaic patients. In all inv(16) metaphases with a single signal, the signal was localized on the normal chromosome 16, as assessed by pHuR195 localization with regard to arm ratio. One mosaic patient (case 13) was studied in complete remission; a signal was present on both chromosomes 16 in all metaphases analyzed. Spot loss was evaluated in each group of patients and the proportion of metaphases with 2/1, 1/1, 2/0, 1/0 and 0/0 were compared to those of metaphases with optimal hybridization (either 2/2 in control and non-deletion or 2/0 in deletion patients) (Table 2). Results indicate that, in control, non-

deletion and deletion patients, more than 50% metaphases do not present any spot loss (mean values are equal to 58.3, 61.6 and 71.0%, respectively). In these patients, the proportion of metaphases with one spot loss (2/1 or 1/0) amount to 35.4, 25.2 and 21.4%, respectively, those of metaphases with two spots loss represent 6.3, 12.5 and 4.1%. In control and nondeletion patients, metaphases with 2/0 (1.6% and 4.6%) are less frequent than metaphases with 1/1 (4.7% and 7.9%). The situation is reversed in the mosaic group, where metaphases with 2/0 (21.3%) occur much more frequently than those with 1/1 (10.6%).


968

Figure 3

Table 2

(Continued).

zit14 spot distribution in inv(16)/t(16;16) and control patients

Signal numbera

2

Spot numberb Deletion patients 1 2 3 4 5 Total

2/2

[10] c [17] [65] [26] [27] [145]

0.0d 0.0 1.5 [1]c 0.0 3.7 [1] 1.4 [2]

2/1

10.0 [1] 0.0 0.0 0.0 7.4 [2] 2.1 [3]

1

1/1

0.0 0.0 0.0 0.0 0.0 0.0

10.0 0.0 1.5 0.0 11.1 3.4

Non-deletion patients 6 [26] 7 [3] 8 [25] 9 [11] 10 [54] 11 [11] 12 [21] Total [151]

53.8 66.7 80.0 54.5 64.8 36.4 57.1 61.6

[14] [2] [20] [6] [35] [4] [12] [93]

34.6 33.3 20.0 27.3 24.1 18.2 23.8 25.2

[9] [1] [5] [3] [13] [2] [5] [38]

3.8 0.0 0.0 9.1 7.4 36.4 9.5 7.9

[1]

Mosaic patients 13 14 15 Total

50.0 11.8 35.7 31.9

[8] [2] [5] [15]

18.8 35.3 28.6 27.7

[3] [6] [4] [13]

6.3 17.6 7.1 10.6

[16] [17] [14] [47]

Total control patients [254]

58.3 [148]

35.4 [90]

Total

[1] [1] [3] [5]

2/0

70.0 47.1 76.9 69.2 74.1 71.0

[7] [8] [50] [18] [20] [103]

1/0

20.0 41.2 16.9 26.9 14.8 21.4

[24] [3] [25] [10] [52] [10] [19] [143]

3.8 0.0 0.0 9.1 3.7 9.1 9.5 4.6

[1]

[1] [4] [4] [2] [12]

92.3 100.0 100.0 90.9 96.3 90.9 90.5 94.7

[1] [2] [1] [2] [7]

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

[1] [3] [1] [5]

75.0 64.7 71.4 70.2

[12] [11] [10] [33]

18.8 29.4 14.3 21.3

[3] [5] [2] [10]

6.3 5.9 14.3 8.5

4.7 [12]

98.4 [250]

1.6 [4]

0

0.0

[2] [7] [11] [7] [4] [31]

[1] [1] [2] [4]

Total

90.0 88.2 93.8 96.2 88.9 92.4

[9] [15] [61] [25] [24] [134]

Total 0/0

0.0 11.8 4.6 3.8 0.0 4.1

[2] [3] [1] [6]

3.8 0.0 0.0 9.1 3.7 9.1 9.5 4.6

[1] [1] [2] [1] [2] [7]

3.8 [1] 0.0 0.0 0.0 0.0 0.0 0.0 0.7 [1]

25.0 35.3 28.6 29.8

[4] [6] [4] [14]

0.0 0.0 0.0 0.0

1.6 [4]

0.0

The number of metaphases scored in each case ranged between 15 and 32 in the control patients and between 10 and 65 in inv(16)/t(16;16) patients, except in patient 7, where only three metaphases could be analyzed. a A signal is defined as either one or two spot(s) on the same chromosome. b Spot distribution on both chromosomes 16. c Metaphase number. d Proportion of metaphases with 2/2, 2/1, 1/1, 2/0, 1/0, and 0/0 spot(s).


In mosaic patients, dual-color FISH with zit18 and zit14 cosmids was performed on the slides used for zit14 hybridization assessment. In all metaphases available, the presence or absence of zit18 signals correlated with that of zit14 signals (data not shown). Discussion We have investigated 16 p deletions in 13 patients with inv(16) and two patients with t(16;16). Seven patients (47%) showed no evidence of a deletion, whereas five patients (33%) presented 16 p deletions. The presence of the deletion was confirmed by Southern blot analysis with a MYH11 6.9 kb BamHI probe in three patients.20 Our rate of deletion patients (33%) is close or superior to those in other reports.16–18 This finding provides further evidence of the relatively frequent occurrence of this event in inv(16) patients and to the presence of some degree of genomic instability in this region predisposing to such molecular rearrangements. 18 Remarkably, two inv(16) and one t(16;16) patients showed a mosaic pattern suggesting the existence of two distinct leukemic cell populations, one with the deletion and one without. It could be argued that, in these cases, the presence of one and two signals in respectively 30 and 70% of metaphases (mean values), may be due to an admixture of normal and abnormal cells. This seems unlikely according to the results of conventional cytogenetics (100% abnormal metaphases in two patients and 92% in the third one whatever the culture condition) which suggest that the metaphases with two signals do not all represent normal cells. A second argument views the presence of 30% metaphases with a single signal as reflecting defective probe hybridization in these cases. There are several reasons for not supporting this assumption, although it cannot be excluded. In control patients, more than 96% metaphases present a double signal. Experiment variability in the hybridization yield can be ruled out as each analysis relied on several slides, all hybridized in separate assays. Defective hybridization is expected to generate a decrease in the proportion of metaphases with 2/2 (control and non-deletion patients) or 2/0 (deletion patients) and an increase in metaphases with spot loss. To characterize better mosaic patients and to test if defective hybridization may account for the presence of more than 25% metaphases with a single signal in these cases, the distribution of fluorescent spots was analyzed in detail. Whereas the proportion of metaphases with three spots (2/1) is similar to those with 1 spot loss (2/1 or 1/0) in all other groups, the proportion of metaphases with two spots loss (2/0 or 1/1) is much higher and the reduction of the number of 2/2 metaphases correlates with a specific increase in 2/0 metaphases without any significant increase in the proportion of 1/1 metaphases. Moreover, in all inv(16) metaphases with a single signal, the signal was recognized in the normal chromosome 16. The use of the zit18 probe in addition to zit14 (dual-color FISH) in a number of metaphases from mosaic patients with two or one zit14 signal(s) did not reveal any discrepancy between zit14 and zit18 signal distribution. Thus, if the mosaicism interpretation is correct, these results mean that, in a subset of patients, 16 p deletions may arise subsequently to inv(16)/t(16;16). Case 14 would represent the first reported t(16;16) patient with a deletion detected by FISH. Unfortunately, no more material was available in the mosaic patients for further study and additional patients have to be tested so as to confirm definitively the mosaicism hypothesis.

To our knowledge, one case of inv(16) with a 16 p deletion mosaicism has been reported.16 Subsequently however, this case was considered as non-deleted based on additional molecular data.17 In the other studies, the number of metaphases observed in each case and the distribution of FISH signals were not reported in detail; therefore, possible mosaic cases may have been considered as non-deleted when only a few metaphases were available. Fourteen of the 15 patients reported here have been studied by RT-PCR.19,20 CBFB–MYH11 transcript type A (nomenclature after Liu et al8) was found in 12 cases (86%), transcript type E in two (14%) (cases 5 and 7). The presence of the deletion in a patient with transcript type E indicates that the p-arm deletion is not restricted to CBFB–MYH11 transcript type A and suggests that it can be associated with other CBFB– MYH11 fusion types as well. Deletions in 16 p have been reported to be associated with improved survival.16,17 As the deleted region was shown to include the promoter and the 59 part of the MRP gene located 150 kb proximal to the p-arm breakpoint, Kuss et al suggested that MRP hemizygosity may be relevant in determining the outcome of inv(16) AML patients. Although MRP status was not assessed specifically, part of the MRP gene was supposed to be codeleted in our deletion patients and these patients were compared with non-deleted and mosaic ones in terms of survival. No significant difference could be demonstrated; the result was unaltered when mosaic patients were grouped with deletion or with non-deletion ones. This finding does not corroborate the formally reported positive correlation between deletion and improved prognosis. However, although in accordance with other observations,18 our data have to be considered with caution as larger numbers of patients have to be longitudinally studied to draw any definitive conclusion regarding prognosis. Moreover, multidrug resistance is a complex and multifactorial phenomenon that involves different mechanisms and several distinct genes such as MRP and MDR1.30 If the MRP gene is expected to play a role in the clinical response to chemotherapy, follow-up studies should be performed not only at the DNA level but also at the mRNA and protein levels before any definitive statements concerning its role in leukemic drug resistance can be made. Our study reinforces the usefulness of FISH for assessing submicroscopic rearrangements of possible diagnostic and prognostic significance that escape conventional cytogenetics due to their small size. While molecular techniques such as Southern blotting or microsatellite marker assessment may allow detection of the presence of the deletion in inv(16)/t(16;16) patients, metaphase FISH with appropriate probes has the advantage of being enumerative and quantitative. Due to the existence of mosaic cases, the FISH approach requires the observation of possibly a large number of metaphases for quantitative values to be significant. Acknowledgements We are indebted to Dr A Tichelli, Prof Dr A Tobler, Prof Dr M Fey, Dr V Spataro, Dr PM Schmidt, Prof Dr A Hirt, Dr M Wernli and the members of the Leukemia Group (Chairman, Dr U Hess) of the SIAK (Chairman Prof Dr H Wagner)/SAKK (Chairman Prof Dr A Goldhirsch) for providing patient samples and clinical data. We thank Dr C Castagne´, O Bruzzese, M Wicht and the other technicians who assisted in the chromosome analysis. We are grateful to B van der Reijden for helpful comments and interest in the project. We thank Dr

969


970

M Breuning, J Dauwerse and R Giles for providing probes and for critical reading of the manuscript. Dr G van Melle is acknowledged for advice in statistical analysis. This work was supported by grants from the Ligue Suisse contre le Cancer (FOR 209), the Ligue Argovienne contre le Cancer, the Recherche Suisse contre le Cancer, the Cramer Foundation, the Ligues Vaudoise and Zougoise contre le Cancer, the Leukemia group of the Swiss Institute for Applied Cancer Research/Swiss Group for Clinical Cancer Research (SIAK/SAKK), the Swiss National Science Foundation (3132319.91), the Stiftung zur Krebsbeka¨mpfung, the Sandoz, the SICPA Foundations and the Division autonome de geńe´tique me´dicale (Prof G Pescia) to Martine Jotterand.

References 1 Arthur DC, Bloomfield CD. Partial deletion of the long arm of chromosome 16 and bone marrow eosinophilia in acute nonlymphocytic leukemia: a new association. Blood 1983; 61: 994–998. 2 Arthur DC, Bloomfield CD. Correspondence. Blood 1983; 62: 931. 3 Le Beau MM, Larson RA, Bitter MA, Vardiman JW, Golomb HM, Rowley JD. Association of an inversion of chromosome 16 with abnormal marrow eosinophils in acute myelomonocytic leukemia. New Engl J Med 1983; 309: 630–636. 4 Bitter MA, Le Beau MM, Larson RA, Rosner MC, Golomb HM, Rowley JD, Vardiman JW. A morphologic and cytochemical study of acute myelomonocytic leukemia with abnormal marrow eosinophils associated with inv(16)(p13q22). Am J Clin Pathol 1984; 81: 733–741. 5 Hogge DE, Misawa S, Parsa NZ, Pollak A, Testa JR. Abnormalities of chromosome 16 in association with acute myelomonocytic leukemia and dysplastic bone marrow eosinophils. J Clin Oncol 1984; 2: 550–557. 6 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. 7 Mitelman F. Catalog of Chromosome Aberrations in Cancer. Wiley-Liss: New York, 1994. 8 Liu PP, Hajra A, Wijmenga C, Collins FS. Molecular pathogenesis of the chromosome 16 inversion in the M4Eo subtype of acute myeloid leukemia. Blood 1995; 85: 2289–2302. 9 Larson RA, Williams SF, Le Beau MM, Bitter MA, Vardiman JW, Rowley JD. Acute myelomonocytic leukemia with abnormal eosinophils and inv(16) or t(16;16) has a favorable prognosis. Blood 1986; 68: 1242–1249. 10 Schiffer CA, Lee EJ, Tomiyasu T, Wiernik PH, Testa JR. Prognostic impact of cytogenetic abnormalities in patients with de novo acute nonlymphocytic leukemia. Blood 1989; 73: 263–270. 11 Plantier I, Lai JL, Wattel E, Bauters F, Fenaux P. Inv(16) may be one of the only ‘favorable’ factors in acute myeloid leukemia: a report on 19 cases with prolonged follow-up. Leukemia Res 1994; 18: 885–888. 12 Burnett AK. Karyotypically defined risk groups in acute myeloid leukaemia. Leukemia Res 1994; 18: 889–890. 13 Marlton P, Keating M, Kantarjian H, Pierce S, O’Brien S, Freireich EJ, Estey E. Cytogenetic and clinical correlates in AML patients with abnormalities of chromosome 16. Leukemia 1995; 9: 965– 971. 14 Liu P, Tarle´ SA, Hajra A, Claxton DF, Marlton P, Freedman M, Siciliano MJ, Collins FS. Fusion between transcription factor CBFb/PEBP2b and a myosin heavy chain in acute myeloid leukemia. Science 1993; 261: 1041–1044.

15 Dauwerse JG, Wessels JW, Giles RH, Wiegant J, van der Reijden BA, Fugazza G, Jumelet EA, Smit E, Baas F, Raap AK, Hagemeijer A, Beverstock GC, Van Ommen GJ, Breuning MH. Cloning the breakpoint cluster region of the inv(16) in acute nonlymphocytic leukemia M4Eo. Hum Mol Genet 1993; 2: 1527–1534. 16 Kuss BJ, Deeley RG, Cole SP, Willman CL, Kopecky KJ, Wolman SR, Eyre HJ, Lane SA, Nancarrow JK, Whitmore SA, Callen DF. Deletion of gene for multidrug resistance in acute myeloid leukaemia with inversion in chromosome 16: prognostic implications. Lancet 1994; 343: 1531–1534. 17 Kuss BJ, Deeley RG, Cole SP, Willman CL, Kopecky KJ, Wolman SR, Eyre HJ, Callen DF. The biological significance of the multidrug resistance gene MRP in inversion 16 leukemias. Leuk Lymphoma 1996; 20: 357–364. 18 Marlton P, Claxton DF, Liu P, Estey EH, Beran M, Le Beau M, Testa JR, Collins FS, Rowley JD, Siciliano MJ. Molecular characterization of 16 p deletions associated with inversion 16 defines the critical fusion for leukemogenesis. Blood 1995; 85: 772–779. 19 Van der Reijden BA, Lombardo M, Dauwerse HG, Giles RH, Mu¨hlematter D, Jotterand Bellomo M, Wessels HW, Beverstock GC, Van Ommen GJ, Hagemeijer A, Breuning MH. RT-PCR diagnosis of patients with acute nonlymphocytic leukemia and inv(16)(p13q22) and identification of new alternative splicing in CBFB–MYH11 transcripts. Blood 1995; 86: 277–282. 20 Van der Reijden BA, Martinet D, Dauwerse JG, Giles RH, Wessels JW, Beverstock GC, Smit B, Mu¨hlematter D, Jotterand Bellomo M, Gabert J, Lafage-Pochitaloff M, Reiffers J, Bilhou-Nabera C, Van Ommen GJ, Hagemeijer A, Breuning MH. Simple method for detection of MYH11 DNA rearrangements in patients with inv(16)(p13q22) and acute myeloid leukemia. Leukemia 1996; 10: 1459–1462. 21 Castagne´ C, Mu¨hlematter D, van Melle G, Gachoud V, Jotterand Bellomo M. Effect of conditioned media, nutritive elements and mitotic synchronization on the accuracy of the cytogenetic analysis in acute nonlymphocytic leukemia patients presenting inv(16)/t(16;16) or t(15;17). Cancer Genet Cytogenet (in press). 22 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. 23 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, Sultan C. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol 1982; 51: 189–199. 24 Second MIC Cooperative Study Group. Morphologic, immunologic and cytogenetic (MIC) working classification of the acute myeloid leukaemias. Br J Haematol 1988; 68: 487–494. 25 Jotterand Bellomo M, Parlier V, Schmidt PM, Beris P. Cytogenetic analysis of 54 cases of myelodysplastic syndrome. Cancer Genet Cytogenet 1990; 46: 157–172. 26 Jotterand Bellomo M, Parlier V, Mu¨hlematter D, Grob JP, Beris P. Three new cases of chromosome 3 rearrangement in bands q21 and q26 with abnormal thrombopoiesis bring further evidence to the existence of a 3q21q26 syndrome. Cancer Genet Cytogenet 1992; 59: 138–160. 27 Moyzis RK, Albright KL, Bartholdi MF, Cram LS, Deaven LL, Hildebrand CE, Joste NE, Longmire JL, Meyne J, Schwarzacher-Robinson T. Human chromosome-specific repetitive DNA sequences: novel markers for genetic analysis. Chromosoma 1987; 95: 375–386. 28 Pinkel D, Straume T, Gray JW. Cytogenetic analysis using quantitative, high-sensitivity, fluorescence hybridization. Proc Natl Acad Sci USA 1986; 83: 2934–2938. 29 Dauwerse JG, Wiegant J, Raap AK, Breuning MH, van Ommen GJ. Multiple colors by fluoresence in situ hybridization using radiolabelled DNA probes create a molecular karyotype. Hum Mol Genet 1992; 1: 593–598. 30 Zo¨chbauer S, Haas OA, Schwarzinger I, Lechner K, Pirker R. Multidrug resistance in acute myeloid leukaemia with inversion in chromosome 16 of FAB M4Eo subtype. Lancet 1994; 344: 894.