December 28, 2000 [605694]

1910
·
December 28, 2000
The New England Journal of Medicine
GENOMIC ABERRATIONS AND SURVIVAL IN CHRONIC LYMPHOCYTIC
LEUKEMIA
H
ARTMUT
D
ÖHNER
, M.D., S
TEPHAN
S
TILGENBAUER
, M.D., A
XEL
B
ENNER
, M.S
C
., E
LKE
L
EUPOLT
, M.D.,
A
LEXANDER
K
RÖBER
, M.D., L
ARS
B
ULLINGER
, M.D., K
ONSTANZE
D
ÖHNER
, M.D., M
ARTIN
B
ENTZ
, M.D.,
AND
P
ETER
L
ICHTER
, P
H
.D.
A
BSTRACT
Background
Fluorescence in situ hybridization has
improved the detection of genomic aberrations inchronic lymphocytic leukemia. We used this method toidentify chromosomal abnormalities in patients withchronic lymphocytic leukemia and assessed their prog-nostic implications.
Methods
Mononuclear cells from the blood of 325
patients with chronic lymphocytic leukemia were an-alyzed by fluorescence in situ hybridization for dele-tions in chromosome bands 6q21, 11q22–23, 13q14,and 17p13; trisomy of bands 3q26, 8q24, and 12q13;and translocations involving band 14q32. Molecularcytogenetic data were correlated with clinical findings.
Results
Chromosomal aberrations were detected in
268 of 325 cases (82 percent). The most frequentchanges were a deletion in 13q (55 percent), a deletionin 11q (18 percent), trisomy of 12q (16 percent), a dele-tion in 17p (7 percent), and a deletion in 6q (6 percent).Five categories were defined with a statistical model:17p deletion, 11q deletion, 12q trisomy, normal karyo-type, and 13q deletion as the sole abnormality; the me-dian survival times for patients in these groups were32, 79, 114, 111, and 133 m onths, respectively. Patients
in the 17p- and 11q-deletion groups had more ad-vanced disease than those in the other three groups.Patients with 17p deletions had the shortest mediantreatment-free interval (9 months), and those with 13qdeletions had the longest (92 months). In multivariateanalysis, the presence or absence of a 17p deletion,the presence or absence of an 11q deletion, age, Bi-net stage, the serum lactate dehydrogenase level, andthe white-cell count gave significant prognostic in-formation.
Conclusions
Genomic aberrations in chronic lym-
phocytic leukemia are important independent predic-tors of disease progression and survival. These find-ings have implications for the design of risk-adaptedtreatment strategies. (N Engl J Med 2000;343:1910-6.)
©2000, Massachusetts Medical Society.
From the Department of Internal Medicine III University of Ulm, Ulm
(H.D., S.S., E.L., A.K., L.B., K.D., M.B.); and the Deutsches Krebsfor-schungszentrum, Heidelberg (A.B., P .L.) — both in Germany. Address re-print requests to Dr. Hartmut Döhner at the Department of Internal Med-icine III, University of Ulm, Robert-Koch-Str. 8, 89081 Ulm, Germany,or at [anonimizat].
-CELL chronic lymphocytic leukemia is the
most common leukemia in adults. It has ahighly variable clinical course; some patientsdie from the disease within a few months of
the diagnosis, whereas others live for 20 years ormore.
1
The clinical staging systems devised by Rai et
al.
2
and Binet et al.
3
are the most useful methods for
predicting survival in chronic lymphocytic leukemia.However, these staging systems cannot be used to pre-dict the individual risk of disease progression and sur-vival in the early stages of chronic lymphocytic leuke-mia (Binet stage A or Rai stage 0 to 2 disease), whenthe disease is first diagnosed in most patients. The sub-stantial heterogeneity within clinical stages has prompt-ed searches for additional prognostic factors, but mostof them have not proved useful.
4
There is considerable interest in identifying chro-
mosomal aberrations that could pinpoint subgroupsof patients with chronic lymphocytic leukemia whohave different prognoses.
5
Conventional cytogenetic
analysis has been hampered by the low mitotic activityof the leukemic cells in vitro. With the usual method,clonal chromosomal aberrations are detected in only40 to 50 percent of cases, the most common beingtrisomy 12 and abnormalities of chromosome bands13q14 and 14q32.
6
Fluorescence in situ hybridization
allows the detection of chromosomal aberrations notonly in dividing cells but also in interphase nuclei, anapproach referred to as interphase cytogenetics. Ini-tial studies of chronic lymphocytic leukemia with thismethod demonstrated that the frequency and spec-trum of chromosomal aberrations it detected differedconsiderably from the results obtained by convention-al chromosome banding.
7
However, in these studies
only single aberrations were evaluated for their prog-nostic importance, and this was done mostly in smallseries of patients.
We designed a comprehensive set of DNA probes
for evaluating genomic changes in chronic lymphocyt-ic leukemia by interphase cytogenetics. Our objectivewas to assess the frequency and clinical relevance ofgenomic aberrations in a large group of patients with
the disease.B
The New England Journal of Medicine
Downloaded from nejm.org at N C STATE UNIVERSITY on January 19, 2013. For personal use only. No other uses without permission.
Copyright © 2000 Massachusetts Medical Society. All rights reserved.

GENOMIC ABERRATIONS AND SURVIVAL IN CHRONIC LYMPHOCYTIC LEUKEMIA
Volume 343 Number 26
·
1911
METHODS
Patients
Between October 1990 and August 1998, 325 consecutive pa-
tients with chronic lymphocytic leukemia from a single institutionwere enrolled in the study and followed with regard to survival.There were 199 men and 126 women; their ages at the time of en-rollment ranged from 30 to 87 years (median, 62). The diagnosisof chronic lymphocytic leukemia required persistent lymphocyto-sis (>5000 lymphocytes per cubic millimeter).
8
Immunopheno-
typic data, available for 314 of the 325 patients, showed that allthe cases of leukemia were CD19+, 298 of 308 tested were CD5+,and 300 of 308 tested were CD23+. All these cases were there-fore of the B-cell type. At the time of enrollment, 63 patients wereat Rai stage 0, 48 at stage 1, 146 at stage 2, 33 at stage 3, and 34at stage 4.
2
According to the Binet system, 170 patients had stage
A, 102 stage B, and 52 stage C disease.
3
In one patient, clinical data
were incomplete. Two hundred forty-eight patients had received noprevious treatment, 39 patients had received one chemotherapeuticregimen, and 38 patients had received two or more chemothera-peutic regimens before interphase cytogenetic analysis. The mediantime from the date of diagnosis to the date of interphase cytoge-netic study was 15 months (interquartile range, 1 to 43 months).
Interphase Cytogenetic Analysis
DNA Probes
A set of DNA probes was developed to diagnose genomic ab-
errations by interphase cytogenetics. Chromosomal regions wereselected on the basis of data from conventional chromosome-banding studies and comparative genomic hybridization.
6,9
The
DNA probes allowed us to screen for the following partial dele-
tions, partial trisomies, and translocations (the clone designation
and the gene or locus detected are shown in brackets): +(3q26)[yeast artificial chromosome 866_e_7],
10
del(6q21) [963_d_6],
11
+(8q24) [935_a_12],
10
del(11q22–q23) [755_b_11],
12
+12q13
[754_a_1],
10
and del(13q14) [
l
-phage clones, which recognize
RB1
(kindly provided by Dr. Thaddeus Dryja, Boston); cosmid
c1325, which identifies
D13S25
],
13
t(14q32) [cosmid cos-C
a
1/2
,
which recognizes the c
a
1
and c
a
2
gene segments proximal to the
J
H
region; yeast artificial chromosome Y6, which identifies V
H
segments telomeric to the J
H
break points in the immunoglob-
ulin heavy-chain gene (
IgH
)],
12
and del(17p13) [cosmids
ICRFc105BO195–75, ICRFc105CO275–77 , ICRFc105EO675–78, and ICRFc105AO144–79 for p53].
14
In cases showing splitting of one fluorescence signal with the
IgH
probes, the leukemia cells were analyzed for two reciprocal
translocations: t(11;14) and t(14;18). For the diagnosis of t(11;14),the
IgH
probes were combined with the differently labeled 540-kb
yeast artificial chromosome 55_g_7 , which recognizes DNA se-quences spanning the region between the major translocation clus-ter and the
CCND1
gene in the
BCL1
locus at 11q13
12
; for the
detection of t(14;18), the
IgH
probes were combined with yeast
artificial chromosome yA153_A_6, which spans the
BCL2
proto-
oncogene (kindly provided by G. Silverman, Boston).
Detection of Genomic Aberrations by Fluorescence in Situ
Hybridization
DNA probe sequences from yeast artificial chromosome clones
were generated by an inter-Alu polymerase-chain-reaction (PCR)protocol.
15
Cosmid DNA was prepared according to the plasmid
Midi Kit protocol (Qiagen, Hilden, Germany). The probes werelabeled by nick translation with biotin–16-deoxyuridine triphos-phate or digoxigenin–11-deoxyuridine triphosphate (Roche, Mann-heim, Germany). Fluorescence in situ hybridization was performedas described previously.
12,14
Statistical Analysis
The primary end point was survival from the time of diagnosis.
Survival times and censored waiting times measured from the dateof diagnosis were plotted with the use of Kaplan–Meier esti-
mates. The median duration of follow-up was calculated accordingto the method of Korn.
16
The proportional-hazards regression mod-
el of Cox was used to identify differences in survival due to prog-nostic factors.
17
As possible prognostic factors, age, sex, Binet and
Rai stages, hemoglobin level, white-cell count, platelet count, serumlactate dehydrogenase and alkaline phosphatase levels, presence orabsence of splenomegaly and lymphadenopathy, extent of periph-eral lymphadenopathy, greatest lymph-node diameter measured,and presence or absence of genomic aberrations (deletion in 17p,deletion in 11q, trisomy of 12q, deletion in 13q, and deletion in6q) were included in the regression model. We estimated missingdata using a multiple-imputation technique with 10 random draws.A limited backward-selection procedure was used to exclude re-dundant or unnecessary variables.
18
Groupwise comparisons of the distributions of clinical and lab-
oratory variables at the time of the genetic study were performedwith the Kruskal–Wallis test (for quantitative variables) and Fish-er’s exact test (for categorical variables). All tests were two-sided.An effect was considered statistically significant if the P value was0.05 or less. To provide quantitative information on the relevanceof statistically significant results, 95 percent confidence intervalsfor hazard ratios were computed. The statistical analyses were per-formed with the following software packages: StatXact (Cytel Soft-ware, Cambridge, Mass.), S-Plus (MathSoft, Seattle), and the De-sign software library.
18
RESULTS
Interphase Cytogenetic Analysis
All 325 cases could be evaluated by interphase cy-
togenetics. Of these cases, 268 (82 percent) exhibitedabnormalities. Table 1 lists these aberrations, in orderof decreasing frequency. In 175 patients there was oneaberration, 67 patients had two aberrations, and 26patients had more than two aberrations. Among the178 patients with 13q deletion, the deletion was thesole abnormality in 117 (66 percent). In the remain-ing 61 patients (34 percent), 13q deletion was ac-
companied by 11q deletion (28 patients), 12q trisomy(13 patients), 11q deletion and 12q trisomy (1 pa-
*One hundred seventy-five patients had one aber-
ration, 67 had two aberrations, and 26 had morethan two aberrations.
T
ABLE
1.
I
NCIDENCE

OF
C
HROMOSOMAL

A
BNORMALITIES

IN
325 P
ATIENTS

WITH

C
HRONIC
L
YMPHOCYTIC
L
EUKEMIA
.
A
BERRATION
N
O
.
OF
P
ATIENTS
(%)*
13q deletion 178 (55)
11q deletion 58 (18)12q trisomy 53 (16)17p deletion 23 (7)6q deletion 21 (6)8q trisomy 16 (5)t(14q32) 12 (4)3q trisomy 9 (3)Clonal abnormalities 268 (82)Normal karyotype 57 (18)
The New England Journal of Medicine
Downloaded from nejm.org at N C STATE UNIVERSITY on January 19, 2013. For personal use only. No other uses without permission.
Copyright © 2000 Massachusetts Medical Society. All rights reserved.

1912
·
December 28, 2000
The New England Journal of Medicine
tient), 17p deletion (8 patients), or other abnormal-
ities (11 patients). An 11q deletion occurred as thesole aberration in 19 of 58 patients (33 percent),12q trisomy in 22 of 53 patients (42 percent), 17pdeletion in 4 of 23 patients (17 percent), and 6q de-letion in 6 of 21 patients (29 percent). All deletionswere monoallelic except for the 13q14 region: in 43of the 178 patients with 13q deletions (24 percent),there were biallelic or concomitant monoallelic andbiallelic deletions. In all cases, biallelic deletion affect-ed the
D13S25
locus, and in 2 of the 43 patients there
was also biallelic
RB1
deletion. Of the 12 patients with
the translocation t(14q32), 7 had t(14;18), and therest had t(14q32) with an unidentified partner. Weincluded patients with t(14;18) in the analysis, sincethey had the typical morphologic features and im-munophenotype of chronic lymphocytic leukemia.No patient had t(11;14).
Correlation with Clinical and Laboratory Data
The proportional-hazards regression model with
backward selection identified six significant prognosticfactors: 17p deletion (P<0.001), 11q deletion (P=0.004), age (P<0.001), Binet stage (B as comparedwith A, P=0.36; C as compared with A, P=0.002),serum lactate dehydrogenase level (P=0.002), andwhite-cell count (P=0.02). There was a statisticallysignificant interaction effect between age and the pres-ence or absence of an 11q deletion (P=0.02): the neg-ative prognostic effect of an 11q deletion was seenprimarily in younger patients. The hazard ratios to-gether with their 95 percent confidence limits areshown in Table 2.
On the basis of the regression analysis, we con-
structed a hierarchical model of genetic subgroups inwhich each case was allocated to one category only.Table 3 lists the five major categories to which 300of the 325 cases could be assigned with this model.
After a median follow-up of 70 months, 112 of
the 325 patients had died. The median survival timeof the entire group was 108 months (95 percent con-fidence interval, 94 to 119). The estimated mediansurvival times from the date of diagnosis for the fivegenetic categories listed in Table 3 were as follows:17p deletion, 32 months; 11q deletion, 79 months;12q trisomy, 114 months; normal karyotype, 111months; and 13q deletion as the sole abnormality, 133months (Fig. 1). The remaining 25 patients were com-bined into the group with various abnormalities. Thisheterogeneous group included patients with 3q tri-somy, 6q deletion, 8q trisomy, or t(14q32). Patientsin this category had a high probability of survival (themedian survival time was not reached).
Table 4 shows the clinical and laboratory data for
the patients in the five major categories at the time ofenrollment. Patients with 17p or 11q deletions hadmore advanced disease than those in the other threegroups (P<0.001), whereas patients with 13q dele-tions had the highest proportion at Binet stage A (72
percent). The groups with 17p and 11q deletions weremore likely to have splenomegaly, mediastinal lym-phadenopathy, and abdominal lymphadenopathy andhad more extensive peripheral lymphadenopathy. Theextent of lymph-node involvement was particularly
*Hazard ratios and confidence intervals (CIs) are
computed for a 10-year increment in age, dependenton 11q deletion; for 11q deletion at the age of 55and 65 years; for an increment of 50 IU per liter inlactate dehydrogenase; and for an increment of20,000 per cubic millimeter in the white-cell count.
T
ABLE
2.
R
ESULTS

OF
C
OX
R
EGRESSION

A
NALYSIS

OF
S
URVIVAL
T
IME

FROM
D
IAGNOSIS

(F
INAL
M
ODEL
).*
V
ARIABLE
H
AZARD
R
ATIO

FOR
D
EATH
(95% CI)
17p deletion 8.08 (4.24–15.40)
Binet stage
B vs. AC vs. A1.27 (0.76–2.13)
3.77 (1.64–8.66)
Age (10-yr increment)
No 11q deletion11q deletion2.04 (1.56–2.67)
1.12 (0.74–1.69)
11q deletion
Age 55 yrAge 65 yr2.89 (1.73–4.84)
1.58 (0.91–2.76)
Lactate dehydrogenase (incre-
ment of 50 IU/liter)1.30 (1.10–1.53)
White-cell count (increment
of 20,000/mm
3
)1.08 (1.01–1.15)
*The model was constructed on the basis of the
regression analysis.
†The five major categories are defined as follows:
patients with a 17p deletion; patients with an 11qdeletion but not a 17p deletion; patients with 12qtrisomy but not a 17p or 11q deletion; patients witha normal karyotype; and patients with a 13q deletionas the sole aberration. Twenty-five of the 325 patientswith various chromosomal abnormalities could notbe assigned to one of these five major categories.
T
ABLE
3.
H
IERARCHICAL
M
ODEL

OF
C
HROMOSOMAL
A
BNORMALITIES

IN
C
HRONIC
L
YMPHOCYTIC
L
EUKEMIA
.*
K
ARYOTYPE
†N
O
.
OF
P
ATIENTS
(%)
17p deletion 23 (7)
11q deletion 56 (17)12q trisomy 47 (14)Normal karyotype 57 (18)13q deletion as sole abnor-
mality117 (36)
Various abnormalities 25 (8)
The New England Journal of Medicine
Downloaded from nejm.org at N C STATE UNIVERSITY on January 19, 2013. For personal use only. No other uses without permission.
Copyright © 2000 Massachusetts Medical Society. All rights reserved.

GENOMIC ABERRATIONS AND SURVIVAL IN CHRONIC LYMPHOCYTIC LEUKEMIA
Volume 343 Number 26 ·1913striking in the 11q-deletion group. Moreover, patients
with 11q and 17p deletions were more likely thanthe others to have fever, night sweats, or weight loss(B symptoms) and had lower hemog lobin values and
lower platelet counts; patients with 17p deletionshad higher serum lactate dehydrogenase and alkalinephosphatase levels and lower serum albumin levels.
There were statistically significant differences in
disease progression among the five genetic categories,as indicated by the treatment-free interval (Fig. 2). Pa-tients in the groups with 17p and 11q deletions hadmore rapid disease progression: the median time fromthe date of diagnosis to the date of first treatment inthese two groups was only 9 and 13 months, respec-tively, and eventually all these patients required ther-apy. The median treatment-free interval was longerin the 12q-trisomy group (33 months) and the nor-
mal-karyotype group (49 months), and it was thelongest by far in the 13q-deletion group (92 months).In the last group, nearly one third of the patients didnot require therapy.
DISCUSSION
We found that molecular cytogenetic methods can
detect genomic aberrations in over 80 percent of pa-tients with chronic lymphocytic leukemia, or abouttwice as frequently as chromosome banding.
5 The
most frequent abnormality we found was a deletioninvolving chromosome band 13q14, which occurredin 55 percent of cases. This result is consistent withstudies using microsatellite and quantitative Southernblot analysis.
13,19-21 The second-most-frequent changeFigure 1. Probability of Survival from the Date of Diagnosis among the Patients in the Five Genetic Categories.
The median survival times for the groups with 17p deletion, 11q deletion, 12q trisomy, normal karyotype, and 13q de-
letion as the sole abnormality were 32, 79, 114, 111, and 133 m onths, respectively. Twenty-five patients with various
other chromosomal abnormalities are not included in the analysis.
0100
0 16820406080
12 24 36 48 60 72 84 96108 120 132 144 156 180
Months
NO. AT RISK
17p deletionk0R
11q deletionk0R12q trisomyk0RNormalk0R13q deletion as solek0R
abnormality23k0R
56k0R47k0R57k0R
11718k0R
53k0R44k0R51k0R
11713k0R
47k0R41k0R45k0R
1068k0R
43k0R29k0R37k0R915k0R
33k0R24k0R30k0R804k0R
27k0R17k0R27k0R631k0R
20k0R14k0R20k0R450k0R
15k0R13k0R17k0R360k0R
10k0R12k0R12k0R240k0R
4k0R
11k0R11k0R160k0R
2k0R4k0R6k0R
120k0R
2k0R3k0R5k0R
110k0R
1k0R2k0R2k0R30k0R
0k0R1k0R2k0R10k0R
0k0R1k0R1k0R10k0R
0k0R0k0R1k0R0Patients Surviving (%)17p deletionk0R
11q deletionk0R12q trisomyk0RNormalk0R13q deletion as solek0R
abnormality
The New England Journal of Medicine
Downloaded from nejm.org at N C STATE UNIVERSITY on January 19, 2013. For personal use only. No other uses without permission.
Copyright © 2000 Massachusetts Medical Society. All rights reserved.

1914 ·December 28, 2000The New England Journal of Medicine
was a deletion in 11q (found in 18 percent of patients).
Previous evidence from banding studies of chromo-somal loss from 11q in chronic lymphocytic leukemiais inconsistent.
5,22 Sixteen percent of our patients had
12q trisomy, which was long considered the most fre-quent chromosomal abnormality in chronic lympho-cytic leukemia; in our study it was the third most fre-quent aberration.
Little is known about the molecular correlates of
these chromosomal abnormalities. The tumor suppres-sor gene p53 is affected by 17p deletions.
14,23 Recent
studies suggest that the gene encoding the ataxia–tel-angiectasia mutated protein is altered in some cases ofchronic lymphocytic leukemia with 11q deletion.
24-26Band 13q14 probably contains a tumor-suppressorgene with a role in chronic lymphocytic leuke-mia.
13,19-21 No disease-related genes have yet been as-
sociated with the other aberrations.
These aberrations are among the most important
factors in predicting survival. Patients with 17p de-letions had by far the worst prognosis, followed bypatients with 11q deletions, those with 12q trisomy,and those with normal karyotypes, whereas patientswith 13q deletions as the sole abnormality had thelongest estimated survival times (Fig. 1). These obser-vations parallel the more frequent finding of advanceddisease at enrollment in patients with 17p or 11q de-letions. In a smaller series of patients, extensive lym-*Median values are given for quantitative variables. Because of rounding, percentages do not always
total 100. Twenty-five patients with various other chromosomal abnormalities are not included in theanalysis.
†The P value is for the overall comparison among the subgroups and was calculated by the
Kruskal–Wallis test.
‡The P value is for the overall comparison among the subgroups and was calculated by Fisher’s
exact test.
§The values are the medians of the products of the diameters of the largest cervical, axillary, and
inguinal lymph nodes in centimeters.
¶B symptoms consist of fever, night sweats, or weight loss.¿The P value is for the overall comparison among the subgroups and was calculated by the log–
rank test.TABLE 4. COMPARISON OF CLINICAL AND LABORATORY DATA AMONG THE MAJOR
CYTOGENETIC SUBGROUPS .*
VARIABLE17p
DELETION11q
DELETION12q
TRISOMY NORMAL13q
DELETION P V ALUE
No. of patients 23 56 47 57 117
Median age (yr) 63 58 63 61 63 0.11†Male sex (%) 74 68 55 49 60 0.17‡
Disease stage at enrollment (%)
Binet
ABC
Rai
0123423
4136
0
5
45232725
5025
5
2043161651
3415
13
115913
453
3017
28
1142
5
1472
21
8
31
1544
55<0.001‡
<0.001‡
White-cell count (¬10
¡3/mm3) 48.3 43.9 32.2 30.9 27.7 0.27†
Hemoglobin (g/dl) 12.5 12.5 13.3 13.5 13.6 0.004†Platelet count (¬10
¡3/mm3) 148 159 192 174 191 0.004†
Lactate dehydrogenase (IU/liter) 247 188 204 169 155 <0.001†Alkaline phosphatase (IU/liter) 152 130 124 124 115 0.004†Albumin (g/liter) 43 45 44 45 46 0.024†Splenomegaly (%) 86 71 66 61 53 0.004‡Mediastinal lymphadenopathy (%) 23 28 5 4 1 <0.001‡Abdominal lymphadenopathy (%) 64 83 53 49 23 <0.001‡Peripheral lymphadenopathy (cm
2)§ 12 17 3 3 0 <0.001†
Largest lymph-node diameter (cm) 3 5 2 2 0 <0.001†B symptoms (%)¶ 33 32 19 16 6 <0.001‡Time from diagnosis to first treat-
ment (mo)9 13 33 49 92 <0.001¿
The New England Journal of Medicine
Downloaded from nejm.org at N C STATE UNIVERSITY on January 19, 2013. For personal use only. No other uses without permission.
Copyright © 2000 Massachusetts Medical Society. All rights reserved.

GENOMIC ABERRATIONS AND SURVIVAL IN CHRONIC LYMPHOCYTIC LEUKEMIA
Volume 343 Number 26 ·1915phadenopathy was particularly striking in patients with
an 11q deletion.12 In the multivariate analysis, both
17p deletion and 11q deletion provided statisticallysignificant prognostic information, with 17p deletionbeing the strongest predictor of poor survival.
Most previous studies of chromosomal aberrations
in chronic lymphocytic leukemia did not identifychromosomal abnormalities that provided independ-ent prognostic information.
6 The poor prognosis of
patients with 17p deletion or p53 mutation has beenreported in only a few studies.
14,23,27 El Rouby et al.
found that mutation of p53 was the strongest inde-pendent prognostic factor.
23 In a prospective study
using chromosome banding, abnormality of chromo-some 17 was associated with poor survival, and itwas the only cytogenetic finding with independent
prognostic value.
27 Neilson et al. found that 11q de-
letions were associated with rapid disease progressionand shorter survival times.
22 The prognostic effect of
12q trisomy has been controversial5,6,28,29 ; our data
indicate that patients with 12q trisomy have shortersurvival than those who have a 13q deletion as thesole aberration. The finding of a favorable outcomefor patients with 13q deletions supports other data.
6
Two recent studies further illuminate the biologic
basis of the clinical variability of chronic lymphocyticleukemia.
30,31 They indicate that chronic lymphocytic
leukemia can arise at different stages of B-cell matu-ration, as indicated by the presence or absence of mu-tations of immunoglobulin variable genes: the latterFigure 2. Probability of Disease Progression, as Indicated by the Treatment-free Interval in the Patients in the Five Ge-
netic Categories.
The median treatment-free intervals for the groups with 17p deletion, 11q deletion, 12q trisomy, normal karyotype, and
13q deletion as the sole abnormality were 9, 13, 33, 49, and 92 months, respectively. The differences between the curveswere significant (P<0.001). Twenty-five patients with various other chromosomal abnormalities are not included in theanalysis.
0100
0 16820406080
12 24 36 48 60 72 84 96108 120 132 144 156 180
Months
NO. UNTREATED
17p deletionk0R
11q deletionk0R12q trisomyk0RNormalk0R13q deletion as solek0R
abnormality23k0R
56k0R47k0R57k0R
1177k0R
29k0R32k0R42k0R
1084k0R
20k0R26k0R33k0R941k0R
13k0R18k0R24k0R720k0R
7k0R
12k0R17k0R580k0R
5k0R9k0R
13k0R450k0R
2k0R7k0R8k0R
280k0R
0k0R6k0R5k0R
210k0R
0k0R6k0R3k0R
130k0R
0k0R4k0R2k0R70k0R
0k0R1k0R1k0R60k0R
0k0R0k0R1k0R60k0R
0k0R0k0R0k0R10k0R
0k0R0k0R0k0R10k0R
0k0R0k0R0k0R10k0R
0k0R0k0R0k0R0Patients Treated (%)17p deletionk0R
11q deletionk0R12q trisomyk0RNormalk0R13q deletion as solek0R
abnormality
The New England Journal of Medicine
Downloaded from nejm.org at N C STATE UNIVERSITY on January 19, 2013. For personal use only. No other uses without permission.
Copyright © 2000 Massachusetts Medical Society. All rights reserved.

1916 ·December 28, 2000The New England Journal of Medicine
represents naive B cells before they enter the germi-
nal center, and the former memory B cells that havepassed through germinal centers. Patients with chron-ic lymphocytic leukemia originating from naive B cellshad significantly shorter survival than patients withchronic lymphocytic leukemia arising from memoryB cells. It will be necessary to assess the relative prog-nostic value of the currently used clinical, biochemical,and genetic markers in large, prospective trials. Ourresults with molecular cytogenetic techniques mayalready have implications for the risk-adapted clinicalmanagement of chronic lymphocytic leukemia, par-ticularly in younger patients.
Supported by grants from the Deutsche Krebshilfe (70-2434-Dö I and
10-1289-St I), the European Community (QLGZ-1999-CT 00786), andthe Tumorzentrum Heidelberg–Mannheim (I/I.1 and I/I.2).
We are indebted to Kathrin Wildenberger, Edeltraud Weilguni,
and Petra Schramm for technical assistance and to Dr. Lutz Edlerfor statistical advice.
REFERENCES
1.Rozman C, Montserrat E. Chronic lymphocytic leukemia. N Engl J
Med 1995;333:1052-7 . [Erratum, N Engl J Med 1995;333:1515.]2.Rai KR, Sawitsky A, Cronkite EP , Chanana AD, Levy RN, Pasternack
BS. Clinical staging of chronic lymphocytic leukemia. Blood 1975;46:219-34.3.Binet JL, Auquier A, Dighiero G, et al. A new prognostic classification
of chronic lymphocytic leukemia derived from a multivariate survival analy-sis. Cancer 1981;48:198-206.4.Zwiebel JA, Cheson BD. Chronic lymphocytic leukemia: staging and
prognostic factors. Semin Oncol 1998;25:42-59.5.Juliusson G, Merup M. Cytogenetics in chronic lymphocytic leukemia.
Semin Oncol 1998;25:19-26.6.Juliusson G, Oscier DG, Fitchett M, et al. Prognostic subgroups in
B-cell chronic lymphocytic leukemia defined by specific chromosomal ab-normalities. N Engl J Med 1990;323:720-4.7.Döhner H, Stilgenbauer S, Döhner K, Bentz M, Lichter P . Chromo-
some aberrations in B-cell chronic lymphocytic leukemia: reassessment based on molecular cytogenetic analysis. J Mol Med 1999;77:266-81.8.Cheson BD, Bennett JM, Grever M, et al. National Cancer Institute-
sponsored Working Group guidelines for chronic lymphocytic leukemia: revised guidelines for diagnosis and treatment. Blood 1996;87:4990-7 .9.Bentz M, Huck K, du Manoir S. et al. Comparative genomic hybridi-
zation in chronic B-cell leukemias reveals a high incidence of chromosomal gains and losses. Blood 1995;85:3610-8.10.Bentz M, Plesch A, Bullinger L, et al. t(11;14)-Positive mantle cell
lymphomas exhibit complex karyotypes and share similarities with B-cell chronic lymphocytic leukemias. Genes Chromosomes Cancer 2000;27:285-94.11.Stilgenbauer S, Bullinger L, Benner A, et al. Incidence and prognostic significance of 6q deletions in B cell chronic lymphocytic leukemia. Leu-
kemia 1999;13:1331-4.12.Döhner H, Stilgenbauer S, James MR, et al. 11q Deletions identify a
new subset of B-cell chronic lymphocytic leukemia characterized by exten-sive nodal involvement and inferior prognosis. Blood 1997;89:2516-22.13.Stilgenbauer S, Nickolenko J, Wilhelm J, et al. Expressed sequences as
candidates for a novel tumor suppressor gene at band 13q14 in B-cell chronic lymphocytic leukemia and mantle cell lymphoma. Oncogene 1998;16:1891-7 .14.Döhner H, Fischer K, Bentz M, et al. p53 Gene deletion predicts for
poor survival and non-response to therapy with purine analogs in chronic
B-cell leukemias. Blood 1995;85:1580-9.
15.Lengauer C, Green ED, Cremer T. Fluorescence in situ hybridization
of YAC clones after Alu-PCR amplification. Genomics 1992;13:826-8.16.Korn EL. Censoring distributions as a measure of follow-up in survival
analysis. Stat Med 1986;5:255-60.17.Cox DR. Regression models and life-tables. J R Stat Soc [B] 1972;34:
187-220.18.Harrell FE. Predicting outcome: applied survival analysis and logistic
regression. Charlottesville: University of Virginia, 1998.19.Kalachikov S, Migliazza A, Cayanis E, et al. Cloning and gene map-
ping of the chromosome 13q14 region deleted in chronic lymphocytic leu-kemia. Genomics 1997;42:369-77 .20.Liu Y, Corcoran M, Rasool O, et al. Cloning of two candidate tumor
suppressor genes within a 10 kb region on chromosome 13q14, frequently deleted in chronic lymphocytic leukemia. Oncogene 1997;15:2463-73.21.Bullrich F , Veronese ML, Kitada S, et al. Minimal region of loss at
13q14 in B-cell chronic lymphocytic leukemia. Blood 1996;88:3109-15.22.Neilson JR, Auer R, White D, et al. Deletions at 11q identify a subset
of patients with typical CLL who show consistent disease progression and reduced survival. Leukemia 1997;11:1929-32.23.el Rouby S, Thomas A, Costin D, et al. p53 Gene mutation in B-cell
chronic lymphocytic leukemia is associated with drug resistance and is in-dependent of MDR1/MDR3 gene expression. Blood 1993;82:3452-9.24.Stankovic T, Weber P , Stewart G, et al. Inactivation of ataxia telangi-
ectasia mutated gene in B-cell chronic lymphocytic leukaemia. Lancet 1999;353:26-9.25.Bullrich F , Rasio D, Kitada S, et al. ATM mutations in B-cell chronic
lymphocytic leukemia. Cancer Res 1999;59:24-7 .26.Schaffner C, Stilgenbauer S, Rappold G, Döhner H, Lichter P . Somat-
ic ATM mutations indicate a pathogenic role of ATM in B-cell chronic lym-
phocytic leukemia. Blood 1999;94:748-53.27.Geisler CH, Philip P , Christensen BE, et al. In B-cell chronic lympho-
cytic leukaemia chromosome 17 abnormalities and not trisomy 12 are the single most important cytogenetic abnormalities for the prognosis: a cyto-genetic and immunophenotypic study of 480 unselected newly diagnosed patients. Leuk Res 1997;21:1011-23.28.Oscier DG, Stevens J, Hamblin TJ, Pickering RM, Lambert R, Fitch-
ett M. Correlation of chromosome abnormalities with laboratory features and clinical course in B-cell chronic lymphocytic leukaemia. Br J Haematol 1990;76:352-8.29.Escudier SM, Pereira-Leahy JM, Drach JW, et al. Fluorescent in situ
hybridization and cytogenetic studies of trisomy 12 in chronic lymphocytic leukemia. Blood 1993;81:2702-7 .30.Damle RN, Wasil T, Fais F , et al. Ig V gene mutation status and CD38
expression as novel prognostic indicators in chronic lymphocytic leukemia. Blood 1999;94:1840-7 .31.Hamblin TJ, Davis Z, Gardiner A, Oscier DG, Stevenson FK. Unmu-
tated Ig V
H genes are associated with a more aggressive form of chronic
lymphocytic leukemia. Blood 1999;94:1848-54.
The New England Journal of Medicine
Downloaded from nejm.org at N C STATE UNIVERSITY on January 19, 2013. For personal use only. No other uses without permission.
Copyright © 2000 Massachusetts Medical Society. All rights reserved.