1105

Effects of Age on Prognosis with Imatinib Mesylate

Therapy for Patients with Philadelphia Chromosome-
Positive Chronic Myelogenous Leukemia

Jorge Cortes, M.D.1 BACKGROUND. Older age is a consistent poor prognostic factor in patients with
Moshe Talpaz, M.D.2 Philadelphia chromosome (Ph)-positive chronic myelogenous leukemia (CML).
Susan OBrien, M.D.1 Whether this is related to an intrinsic worse disease biology or to inadequate drug
Francis Giles, M.D.1 delivery or excessive treatment-associated toxicity is unknown. The availability of
Mary Beth Rios, R.N.1 imatinib mesylate, a selective, Bcr-Abl-targeted therapy that is administered orally
Jianquin Shan, Ph.D.1 with minimal side effects, may clarify whether older age would remain an adverse
Stefan Faderl, M.D.1 factor (thus, implying a different age-related CML biology).
1
Guillermo Garcia-Manero, M.D. METHODS. Seven hundred forty-seven patients in different phases of Ph-positive
Alessandra Ferrajoli, M.D.1 CML who were treated with imatinib from 1999 until the time of last follow-up
William Wierda, M.D., Ph.D.1 were evaluated. Among them, 187 patients had newly diagnosed, early chronic
Hagop Kantarjian, M.D.1 phase CML; 351 patients had chronic phase CML after interferon (IFN) failure;
133 patients had accelerated phase CML; and 76 patients had blastic phase CML.
1
Department of Leukemia, The University of Texas The imatinib daily dose varied from 400 mg to 800 mg orally, according to the
M. D. Anderson Cancer Center, Houston, Texas. protocol design. Patients were categorized into a group of older patients (age 60
2
Department of Bioimmunotherapy, The University years or older) or younger patients (age younger than 60 years). Their character-
of Texas M. D. Anderson Cancer Center, Houston, istics, responses to therapy, and survival were compared by univariate and multi-
Texas.
variate analyses.
RESULTS. One hundred eighty-seven patients had newly diagnosed CML, and 49
patients (26%) were in the older age group. Older patients had similar cytogenetic
response rates and survival compared with younger patients. Among 351 patients
with late chronic phase CML after IFN failure, 120 patients (34%) were in the older
age group. Although the older patients had a lower incidence of achievement of
complete cytogenetic response (Ph, 0%) by univariate analysis (56% vs. 44%; P
0.05), age was not found to be an independent poor prognostic factor in the
multivariate analysis. Similarly, older age was not an adverse poor prognostic
factor for survival. Forty-two of 133 patients (32%) with accelerated phase CML
were older. The incidence of any cytogenetic response was lower in older patients
(53% vs. 33%; P 0.04), but age was not signicant in the multivariate analysis.
Older patients also had a trend toward worse survival (P 0.09) that was not
signicant in the multivariate analysis. Twenty-eight of 76 patients (37%) evaluated
in blastic phase were older. Older age was not a signicant prognostic factor either
Jorge Cortes M.D. is a Clinical Research Scholar
for the Leukemia & Lymphoma Society. for achieving response or for survival.
CONCLUSIONS. With imatinib therapy, older age appears to have lost much of its
Address for reprints: Hagop Kantarjian, M.D., De- prognostic relevance. This suggests that the previous poor prognosis observed with
partment of Leukemia, The University of Texas older age was related to treatment-associated factors (e.g., toxicity with allogeneic
M. D. Anderson Cancer Center, 1515 Holcombe
Boulevard, Box 428, Houston, TX 77030; Fax: transplantation or with IFN therapy) rather than to an intrinsic, different disease
(713) 792-2031; E-mail: hkantarj@mdanderson. biology of CML in older patients. Cancer 2003;98:110513.
org 2003 American Cancer Society.

Received February 12, 2003; revision received

May 27, 2003; accepted June 13, 2003. KEYWORDS: old age, chronic myelogenous leukemia, imatinib, Gleevec, STI571.

2003 American Cancer Society

DOI 10.1002/cncr.11629
1106 CANCER September 15, 2003 / Volume 98 / Number 6
O lder age has been a consistent poor prognostic
factor in patients with Philadelphia (Ph) chromo-
some-positive chronic myelogenous leukemia (CML)
and has been incorporated into most multivariate-
derived prognostic models.15 The poor prognostic ef-
fect of old age has been persistent across different
therapies, including busulfan, hydroxyurea, interfer-
on- (IFN), and allogeneic stem cell transplantation
(SCT), although the reasons for its prognostic effect
may differ. With allogeneic SCT, for example, a poor
prognosis with older age may be due primarily to
transplantation-related mortality, which is signi-
cantly higher in older patients regardless of the un-
derlying disease.6 8 Older patients have signicantly
worse side effects with IFN therapy; thus, their poor
prognosis may be due at least in part to poor treat-
ment tolerance and inadequate treatment delivery.9 It
is noteworthy that the median age in the IFN studies
was 4550 years,10 12 signicantly younger than the
median age of 55 years in the imatinib mesylate stud-
ies (Gleevec; imatinib; STI571 [Novartis, East Hanover,
NJ]),1320 or the median age of 67 years in the Surveil-
lance, Epidemiology, and End Results Program data.21
Thus, the prognostic effect of older age in these stud-
ies may not reect a universal effect in the CML pop-
ulation. Older patients tolerate hydroxyurea and
busulfan as well as younger patients. Therefore, their
poorer prognosis may be attributed to several factors:
lack of adjustment of survival data for disease specic
mortality; a different biology of CML in older patients
(that may be reected in a higher incidence of other
poor prognostic features, e.g., more aggressive dis-
ease); a delayed diagnosis in older patients who may
then present with more morbidity and tumor burden;
comorbid, age-related conditions; or poorer compli-
ance or less frequent follow-up or medical attention
provided to older patients, which may compromise
their care delivery.
Imatinib, a selective Bcr-Abl tyrosine kinase in-
hibitor, has changed the treatment paradigm and
prognosis for patients with Ph-positive CML. The in-
troduction of highly effective therapy in patients with
other malignancies (e.g., velban, bleomycin, and cis-
platin in patients with testicular carcinoma; 2-chloro-
deoxyadenosine in patients with hairy cell leukemia)
often has nullied or reduced signicantly the prog-
nostic signicance of previously well established prog-
nostic factors. Imatinib is taken orally and has been
tolerated extremely well across age groups. Thus, the
availability of a highly active Bcr-Abl-directed oral
therapy with minimal toxicity should eliminate age as
a prognostic factor in patients with Ph-positive CML,
unless age is associated with as yet undened biologic
disease differences. This is the focus of the current
study, which analyzes the clinicolaboratory differ-
ences of older patients versus younger patients with
Ph-positive CML in early chronic, chronic post-IFN
failure, accelerated, and blastic phases who were
treated with imatinib, and investigates potential dif-
ferences in treatment delivery, response to therapy,
and overall prognosis.
MATERIALS AND METHODS
Study groups of patients with Ph-positive CML who
were treated in the Leukemia Department at The Uni-
versity of Texas M. D. Anderson Cancer Center with
imatinib for newly diagnosed (early chronic phase)
CML, chronic phase CML after IFN failure, accelerated
phase CML, and blastic phase CML were analyzed.
Study eligibility criteria, treatment schedules, pre-
treatment and follow-up studies, and dose-schedule
modications have been detailed previously in publi-
cations on patients with CML in early chronic
phase,22,23 chronic phase post-IFN failure,13,15 accel-
erated phase,14,16 and blastic phase.17 Informed con-
sent was obtained for all study patients, according to
institutional guidelines. Accelerated phase CML was
dened by the presence of any of the following crite-
ria: 15% blasts, 30% blasts and promyelocytes,
20% basophils, or thrombocytopenia 100 109/L
not related to therapy.24 Blastic phase CML was de-
ned by the presence of 30% blasts or extremedul-
lary blastic disease.25
Response Criteria
Complete hematologic response/remission (CHR) re-
quired normalization for at least 4 weeks of the bone
marrow ( 5% blasts) and peripheral blood with a
leukocyte count 10 109/L, without peripheral
blasts, promyelocytes, or myelocytes, in addition to
the disappearance of all signs and symptoms of CML.
This was categorized further by the cytogenetic re-
sponse (suppression of Ph-positive cells) as follows:
no cytogenetic response, 100% Ph-positive; minimal
cytogenetic response, 3590% Ph-positive; partial cy-
togenetic response, 134% Ph-positive; and complete
cytogenetic response, 0% Ph-positive. A major cytoge-
netic response included complete and partial cytoge-
netic responses ( 35% Ph-positive).
A partial hematologic response (PHR) was the
same as a CHR, except for the persistence of immature
cells (myelocytes and metamyelocytes), splenomegaly
50% of the pretreatment level, or thrombocytosis
450 109/L but 50% of the pretreatment level.
Response denitions in blastic phase were detailed
previously.17
 Old Age and Prognosis in CML/Cortes et al. 1107

TABLE 1 TABLE 2
Patient Accrual on Different Chronic Myelogenous Leukemia Trials Characteristics of Younger Patients versus Older Patients Treated
(n 747 patients) with Imatinib for Early Chronic Phase Chronic Myelogenous
Leukemia (N 187 patients)
CML phase/protocol No. of patients
No. of patients (%)
Early chronic phase
DM01-015 (imatinib 400 mg) 50 Age < 60 Age > 60
DM01-151 (imatinib 800 mg) 100 Parameter yrs yrs P value
DM01-163 (imatinib 400 mg) (Novartis 106) 23
Expanded access (Novartis 113)a 14 No. treated 138 49
Chronic phase post-IFN failure Splenomegaly (yes) 46 (33) 7 (14) 0.01
FDA pivotal (Novartis 109) 149 Hemoglobin 12 g/dL 59 (43) 17 (35) 0.40
Expanded access (Novartis 110) 111 Leukocyte count 50 109/L 43 (31) 15 (31) 1.00
Expanded access (Novartis 113)a 91 Platelets 450 109/L 49 (36) 19 (39) 0.73
Accelerated phase Peripheral blasts (yes) 53 (38) 12 (24) 0.08
FDA pivotal (Novartis 109) 59 % Bone marrow blasts ( 5) 11 (8) 4 (8) 1.00
Expanded access (Novartis 113) 74 % Peripheral basophils ( 7) 30 (22) 9 (18) 0.69
Blastic phase % Bone marrow basophils ( 4) 42 (30) 15 (31) 1.00
FDA pivotal (Novartis 102) 13 Additional chromosomal abnormalities (yes) 6 (4) 4 (8) 0.29
Expanded access (Novartis 115) 33
Phase I study (Novartis 001) 29 TABLE 3
Expanded access (Novartis 113)b 1 Response to Imatinib by Age in Patients with Early Chronic Phase
Chronic Myelogenous Leukemia
CML: chronic myelogenous leukemia; IFN: interferon FDA: Food and Drug Administration.
a
These patients had newly diagnosed chronic myelogenous leukemia (n 14 patients) or chronic No. of patients (%)
phase post-IFN failure (n 91 patients) but were categorized in accelerated phase based on the looser
denition of accelerated phase in the study (Novartis 113), which included spelenomegaly 10 cm Age < 60 yrs Age > 60 yrs
below the costal margin, pretreatment additional chromosomal abnormalities only other than Phila- Response (n 128 patients)a (n 46 patients)a P value
delphia chromosome (Ph) abnormalities (but excluding Ph variants, complex Ph translocations, and
loss of chromosome Y), or the presence of 10% blasts (see Kantarjian et al.16). Cytogenetic response 123 (96) 45 (98) 1.00
b
One patient who was treated on the accelerated phase study was reclassied in blastic phase. Complete 101 (79) 40 (87) 0.28
Partial 14 (11) 5 (11)
Minor 8 (6) 0 (0)
Major (complete partial) 115 (90) 45 (98) 0.12
Statistical Analysis
Evaluation of cytogenetic response was based on rou- a
Of 138 patients age 60 years, 1 patients was inevaluable, and 9 patients were treated for 3 months;
tine cytogenetic studies. Cytogenetic studies were per- thus, response was analyzed in 128 evaluable patients. Of 49 patients Age 60 years, 3 patients were
treated for 3 months; thus, response was analyzed in 46 evaluable patients.
formed every 3 months in the rst year and every 3 6
months thereafter. Pretreatment additional chromo-
somal abnormalities referred to the presence of cyto-
genetic abnormalities other than Ph (excluding Ph phase, and 76 patients in blastic phase (Table 1). We
variants, complex Ph translocations, and loss of chro- chose the cut-off age of 60 years in subsequent anal-
mosome Y). Response rates were computed on an yses for several reasons: 1) this was the cut-off age that
intention-to-treated basis. Survival was analyzed from was identied as prognostically relevant in previous
the date therapy started until death from any cause. multivariate analysis studies of CML prognosis at our
Differences in survival curves were compared using institution; 2) these patients mostly are ineligible for
the log-rank test. Statistical methods and multivariate allogeneic SCT and are offered non-SCT approaches
analysis studies were conducted as described previ- to improve their prognosis; and 3) they have histori-
ously.13,15 Multivariate analyses used the logistic re- cally tolerated IFN therapy poorly9 and/or were not
gression model for response and the Cox proportional offered IFN therapy, hence, it is not known whether
hazards model for survival. Variables that demon- their poor prognosis is related to a different intrinsic
strated signicant differences in the univariate analy- disease biology or to ineffective, poorly tolerated ther-
sis (P 0.05) were included in the multivariate anal- apy.
yses.
Early Chronic Phase CML
RESULTS One hundred eighty-seven patients were treated for
A total of 747 patients with CML were analyzed: 187 early chronic phase CML: Seventy-three patients re-
patients in early chronic phase, 351 patients in chronic ceived imatinib at a dose of 400 mg daily, 14 patients
phase after IFN failure, 133 patients in accelerated received 600 mg daily, and 100 patients received 800
1108 CANCER September 15, 2003 / Volume 98 / Number 6

TABLE 4
Characteristics of Younger Patients and Older Patients Treated with Imatinib for Chronic Phase Chronic Myelogenous
Leukemia after Failure on Interferon Therapy

No. of patients (%)

Age < 60 yrs Age > 60 yrs

Parameter (n 231 patients) (n 120 patients) P value

Last response to IFN- immediately before entry on imatinib study

(n 223 patients age 60 yrs; n 115 patients age 60 yrs)
Hematologic resistance/recurrence 34 (15) 18 (16) 0.14
Cytogenetic resistance/recurrence 111 (50) 44 (38)
IFN intolerance 78 (35) 53 (46)
Spelenomegaly (yes) 40/229 (17) 23 (19) 0.81
Hemoglobin 10 g/dL 206 (89) 110 (92) 0.57
Leukocyte count 10 109/L 106 (46) 71 (59) 0.04
Platelets 450 109/L 45 (20) 30 (25) 0.49
% Peripheral blasts (any) 51 (22) 35 (29) 0.18
% Bone marrow blasts ( 5) 30 (13) 19 (16) 0.57
% Peripheral basophils ( 7) 24 (10) 19 (16) 0.19
% Bone marrow basophils ( 4) 46 (20) 36 (30) 0.05
% Ph positive at start of therapy ( 90) 160/227 (70) 96/119 (81) 0.05
Additional chromosomal abnormalities (yes) 54 (24) 14 (12) 0.01
CML duration
12 mos 28 (12) 11 (9)
1235 mos 91 (39) 45 (38) 0.59
36 mos 112 (48) 64 (53)
Imatinib dose (mg/day)
400 170 (74) 90 (75) 0.88
600 61 (26) 30 (25)
No. of prior treatments
1 190 (82) 87 (72)
23 41 (18) 33 (28) 0.05

IFN- interferon Ph: Philadelphia chromosome; CML: chronic myelogenous leukemia.

mg daily. Forty-nine patients (26%) were age 60 years
or older. The characteristics of patients in the two age
groups are detailed in Table 2. No signicant differ-
ences in important clinical or laboratory features were
noted, except for splenomegaly, which was more fre-
quent in younger patients (P 0.05). Older patients
represented 18 of 73 patients (25%) who received ima-
tinib at a dose of 400 mg daily and 31 of 114 patients
(27%) who received imatinib at a dose of 600 800 mg
daily. Response to imatinib therapy was similar in
younger and older patients (Table 3). With a median
follow-up of 16 months (range, 230 months), 7 pa-
tients have transformed to accelerated or blastic
phases (2 of 49 patients age 60 years or older [1 patient
in accelerated phase and 1 patient in blastic] vs. 5 of
138 patients younger than 60 years [3 patients in ac-
celerated phase and 2 patients in blastic phase]), and
only 3 of 187 patients have died (1 patient died after
developing complications of graft-versus-host disease,
1 patient died a natural death from old age, and 1
patient died from blastic phase complications).
Chronic Phase CML Post-IFN Failure
Three hundred fty-one patients were treated with
imatinib for chronic phase CML after IFN failure; 120
patients (34%) were age 60 years or older. The char-
acteristics of younger and older patients are compared
in Table 4. Older patients had a lower incidence of
pretreatment additional chromosomal abnormalities
(P 0.01), which constitute a well known adverse
prognostic factor. Older patients also had more fre-
quent leukocytosis (P 0.04), bone marrow baso-
philia (P 0.05), Ph positivity 90% at the start of
therapy (P 0.05), and more therapies prior to ima-
tinib (P 0.05). Older patients had a signicantly
lower incidence of complete cytogenetic response
compared with younger patients (Table 5). With a
median follow-up of 24 months (range, 136 months),
35 patients have died, 21 of 231 younger patients (9%)
versus 14 of 120 older patients (12%). Their causes of
death are shown in Table 6. Survival of patients by age
was not different between patients in the two age
groups (Fig. 1).
 Old Age and Prognosis in CML/Cortes et al. 1109

TABLE 5
Response to Imatinib by Age in Patients with Chronic Phase Chronic
Myelogenous Leukemia after Failure on Interferon Therapy

No. of patients (%)

Age < 60 yrs Age > 60 yrs

Response (n 231 patients) (n 120 patients) P value

Complete hematologic remission 217 (94) 113 (94)

Cytogenetic response 169 (73) 76 (63) 0.08
Complete 129 (56) 53 (44) 0.05
Partial 24 (10) 15 (12)
Minor 16 (7) 8 (7)
Major (complete partial) 153 (66) 68 (57) 0.10

TABLE 6
Causes of Death by Age Group with Imatinib Therapy in Patients
with Chronic Phase Chronic Myelogenous Leukemia after Failure on FIGURE 1. Survival by age group with imatinib therapy for patients with
Interferon Therapy chronic phase chronic myelogenous leukemia after failure on interferon
therapy.
No. of patients

Age < 60 yrs Age > 60 yrs

Cause of death (n 21 patients) (n 14 patients)
Blastic/accelerated phase 10/0 3/2
Infections/organ failure 3 3
Cardiac event 1 2
Intercurrent illness 1 1
Bleeding 1 0
Unknown (death outside the institution) 5 3
A multivariate analysis of prognostic factors asso-
ciated with lower complete cytogenetic response iden-
tied thrombocytosis (P 0.03), Ph positivity 90%
at start of therapy (P 0.01), bone marrow blasts
5% (P 0.02), and more than 1 prior therapy (P
0.04) as independent poor prognostic factors for
achieving a complete cytogenetic response. Older age
was not an independent poor prognostic factor.
A multivariate analysis of survival identied the
following factors as indicative of a poor prognostic
trend: pretreatment additional chromosomal abnor-
malities (P 0.10) and higher numbers of prior ther-
apies (P 0.05). Older age was not associated inde-
pendently with poor survival.
Accelerated Phase CML
Forty-two of 133 patients (32%) who were analyzed in
accelerated phase were age 60 years or older. The
characteristics of older and younger patients who re-
ceived treatment with imatinib for accelerated phase
CML are shown in Table 7. Older patients had a sig-
nicantly higher incidence of bone marrow blasts
15% (P 0.04) and trends for higher percentages of
peripheral blasts (P 0.10), higher hemoglobin levels
(P 0.10), and leukocytosis (P 0.07) (Table 7).
Although there was a trend toward higher complete
response rates (P 0.07) and major cytogenetic re-
sponse rates (P 0.08) with imatinib in younger pa-
tients, it was not signicant; however, the incidence of
any cytogenetic response was 53% versus 33%, respec-
tively (P 0.04) (Table 8). With the current median
follow-up of 22 months (range, 137 months), 47 of
133 patients have died from reasons shown in Table 9.
There was a trend toward better survival with imatinib
in younger patients (30-month estimated survival
rates: 66% vs. 45%; P 0.09) (Fig. 2).
In the multivariate analysis, signicant poor inde-
pendent prognostic factors for achieving complete cy-
togenetic response were splenomegaly 10 cm bcm
(P 0.02), any peripheral blasts (P 0.03), and Ph
positivity 90% at the initiation of therapy (P 0.08).
For survival, the multivariate-derived, independent
poor prognostic factors were pretreatment additional
chromosomal abnormalities (P 0.02), bone marrow
basophilia (P 0.04), and lower imatinib dose (P
0.03). Age, however, was not found to be an inde-
pendent poor prognostic factor for achieving com-
plete cytogenetic response and had a marginal inde-
pendent effect on survival (P 0.08).
Blastic Phase CML
Seventy-six patients received imatinib for CML in
blastic phase, including 28 patients (37%) age 60 years
or older. The characteristics of younger and older pa-
tients were similar, except for a longer duration of
chronic phase in older patients (P 0.01), and more
patients in the older group received imatinib as their
1110 CANCER September 15, 2003 / Volume 98 / Number 6

TABLE 7 TABLE 9
Characteristics of Younger Patients and Older Patients Treated with Causes of Death by Age in Patients with Accelerated Phase Chronic
Imatinib for Accelerated Phase Chronic Myelogenous Leukemia Myelogenous Leukemia Treated with Imatinib
(N 133 patients)
No. of patients
No. of patients (%)
Age < 60 yrs Age > 60 yrs
Age < 60 yrs Age > 60 yrs Cause of death (n 28 patients) (n 19 patients)
Parameter (n 91 patients) (n 42 patients) P value
Blastic phase 18 7
Spelenomegaly (yes) 43 (47) 21 (50) 0.44 Infections/organ failure 1 4
Hemoglobin 10 g/dL 44 (48) 27 (64) 0.10 Cardiac events 0 1
Leukocyte count 50 Postallogenic SCT complications 3 0
109/L 15 (16) 13 (31) 0.07 Cause unknown (death outside the
Platelets ( 109/L) institution) 6 7
100 39 (43) 14 (33) 0.48
450 26 (28) 16 (38) SCT: stem cell transplantation,.
% Peripheral blasts ( 10) 16 (18) 14 (33) 0.10
% Peripheral blasts and
promyelocytes ( 20) 8 (9) 8 (19) 0.15
% Peripheral basophils (
20) 21 (23) 6 (14) 0.26
% Bone marrow blasts (
15) 24 (26) 21 (50) 0.04
% Bone marrow blasts and
promyelocytes ( 20) 22 (24) 14 (33) 0.30
Additional chromosomal
abnormalities (yes) 34 (37) 17 (40) 0.88
Imatinib dose (mg/day)
400 19 (21) 7 (17)
600 72 (79) 35 (83) 0.64
Chronic phase duration
12 mos 29 (32) 7 (17)
1235 mos 26 (28) 11 (26) 0.11
36 mos 36 (40) 24 (57)
% Ph positive at initiation
of therapy ( 90) 79/89 (89) 38/41 (93) 0.75
FIGURE 2. Survival by age group with imatinib therapy for patients with
Ph: Philadelphia chromosome.
accelerated phase chronic myelogenous leukemia.

TABLE 8
Response by Age Group in Patients with Accelerated Phase Chronic vival of older and younger patients was similar (Fig. 3).
Myelogenous Leukemia Multivariate analysis for achievement of an objective
No. of patients (%) response (CHR, PHR, hematologic improvement, and
second chronic phase) identied thrombocytopenia
Age < 60 yrs Age > 60 yrs 50 109/L (P 0.04) and high peripheral blasts
Response (n 91 patients) (n 42 patients) P value 50% (P 0.07) as poor independent prognostic
factors, whereas older age was not. Similarly, the mul-
Complete hematologic remission 70 (77) 34 (81) 0.66
Cytogenetic response 48 (53) 14 (33) 0.04 tivariate analysis for survival identied only spleno-
Complete 33 (36) 8 (19) 0.07 megaly (P 0.01) as a factor that had an adverse
Partial 6 (7) 3 (7) effect.
Minor 9 (10) 3 (7)
Major (complete partial) 39 (43) 11 (26) 0.08
Prognostic Signicance of Increasing Age in Patients
with Chronic Phase CML
The cut-off age of 60 years was chosen for the reasons
rst blastic phase salvage therapy (P 0.01) (Table stated above (established prognostic cut-off age, the
10). Response to imatinib salvage therapy was similar role of allogeneic SCT in younger patients, and intol-
in younger and older patients (Table 11). With a me- erance to IFN in older patients). However, a different
dian follow-up of 20 months (range, 6 37 months), 66 cut-off age or increasing age still may be relevant
patients died from blastic phase complications. Sur- prognostically and would be important clinically (e.g.,
 Old Age and Prognosis in CML/Cortes et al. 1111

TABLE 10
Characteristics of Younger Patients and Older Patients Treated with Imatinib for Blastic Chronic Myelogenous Leukemia

No. of patients (%)

Age < 60 yrs Age > 60 yrs

Parameter (n 48 patients) (n 28 patients) P value

Spelenomegaly ( 5 cm bcm) 18/44 (41) 10/25 (40) 1.00

Hemoglobin 10 g/dL 26 (54) 12 (43) 0.48
Leukocyte count 50 109/L 14 (29) 8 (29) 1.00
Platelets ( 109/L)
50 12 (25) 12 (39) 0.37
50100 12 (25) 7 (25)
Additional chromosomal abnormalities (yes) 26/45 (58) 17/27 (63) 0.80
% Peripheral blasts ( 50) 18 (38) 8 (29) 0.46
% Bone marrow blasts ( 50) 23 (48) 17 (61) 0.34
Blastic morphology (lymphoid) 7 (15) 3 (11) 0.74
Chronic phase duration
12 mos 14 (29) 3 (11)
1235 mos 16 (33) 4 (14) 0.01
36 mos 18 (38) 21 (75)
Blastic phase salvage ( second) 21 (44) 4 (14) 0.01

TABLE 11
Response by Age Group in Patients with Blastic Phase Chronic Myelogenous Leukemia

No. of patients (%)

Age < 60 yrs Age > 60 yrs

Response (n 48 patients) (n 28 patients) P value

Complete hematologic remission hematologic improvement 95 96

Cytogenetic response
Complete 2 3
Partial 2 2
Minor 2 1
Partial hematologic response 1 2
Second chronic phase 5 0
Early death 0 2
Response
Objective response 20 (42) 17 (61) 0.15
Any cytogenetic response 6 (13) 6 (21) 0.34

in the choice of allogeneic SCT in younger patients
ages 40 60 years). For this purpose, a Martingale re-
sidual plot was examined for survival with increasing
age for all patients with CML in chronic phase in this
study (Fig. 4A). Increasing age was not associated with
worse survival, as expected. When a similar plot was
examined for patients who were treated with IFN for
early chronic phase CML from 1982 to 1995, increas-
ing age appeared to be associated with worse survival
with IFN, with an optimal cut-off age older than 50 55
years (Fig. 4B).
DISCUSSION
Older age has been a consistently poor prognostic
factor for outcome in patients with Ph-positive CML.
This may be related to biologic differences of the dis-
ease in older patients, poor follow-up or compliance,
inadequate delivery of effective therapy (e.g., with al-
logeneic SCT or IFN), or non-CML-related deaths.
Several prognostic models previously reported that
older age was an independent poor prognostic factor
for outcome in patients with CML.1 4
This analysis focused on the prognostic signi-
cance of older age in the era of imatinib therapy. We
hypothesized that, if the poor prognosis for older pa-
tients with CML was not related to intrinsic age-asso-
ciated biologic features, then the prognostic effect of
older age would be reduced signicantly with an oral,
well tolerated, highly effective therapy. With the cur-
rent follow-up, the complete cytogenetic response
1112 CANCER September 15, 2003 / Volume 98 / Number 6
FIGURE 3. Survival by age group with imatinib therapy for patients with
blastic phase chronic myelogenous leukemia.
FIGURE 4. (A) A Martingale residual plot of survival with increasing age
demonstrating no effect of increasing age on survival in patients with chronic
phase chronic myelogenous leukemia (CML) in the era of imatinib therapy. (B)
Increasing age was an adverse prognostic factor for survival in patients with
early chronic phase CML who received interferon therapy from 1982 until
1995.
FIGURE 5. Survival by age group with interferon therapy for patients with
early chronic phase chronic myelogenous leukemia (19821995).
rate and the survival rate were similar by age group in
patients who received imatinib in early chronic phase.
This is in contrast to the previously consistent poor
prognosis associated with older age (e.g., with IFN
therapy) (Fig. 5). Among patients with chronic phase
CML post-IFN failure, older patients had a lower inci-
dence of complete cytogenetic response, but survival
was similar. In our previous experience in patients
with late chronic phase CML with IFN therapy, age
remained an independent poor prognostic factor,26,27
which was not the case with imatinib therapy.
In both the accelerated and blastic phases of CML,
older age did not appear to be of prognostic relevance.
There are no denite data to suggest the prognostic
impact of older age in the literature in these patients
with CML in advanced phases. However, in patients
with blastic phase CML, older age often has not been
of denite prognostic relevance, perhaps because the
prognosis of those patients already is extremely poor
and is determined by more important poor prognostic
features.
The poor prognostic relevance of older age in
patients with CML already appears to be minimized
signicantly with the availability of imatinib, an effec-
tive Bcr-Abl-targeted therapy that is delivered orally
with minimal side effects. This nding suggests that
the previously described prognostic effect of age may
not have been related to a major extent to biologic
differences in CML among older patients versus
younger patients.
REFERENCES
1. Kantarjian H, Smith TL, McCredie K, et al. Chronic myelog-
enous leukemia: a multivariate analysis of the associations
of patient characteristics and therapy with survival. Blood.
1985;66:1326 1335.

2. Kantarjian H, Keating M, Smith TL, et al. Proposal for a
simple synthesis prognostic staging system in chronic my-
elogenous leukemia. Am J Med. 1990;88:1 8.
3. Sokal J, Cox EB, Baccarani M, Tura S, et al. Prognostic
discrimination in good-risk chronic granulocytic leuke-
mia. Blood. 1984;63:789 799.
4. Hasford J, Prrmann M, Hehlmann R, et al. A new prognos-
tic score for survival of patients with chronic myeloid leu-
kemia treated with interferon alfa. J Natl Cancer Ins. 1998;
90:850 858.
5. Kantarjian H, Keating M, McCredie K, et al. Old age is a sign
of poor prognosis in patients with chronic myelogenous
leukemia. Southern Med J. 1988;80:12331235.
6. Goldman JM, Szydlo R, Horowitz MM, et al. Choice of pre-
transplant treatment and timing of transplants for chronic
myelogenous leukemia in chronic phase. Blood. 1993;82:
22352238.
7. Horowitz MM, Rowlings PA, Passweg JR. Allogeneic bone
marrow transplantation for CML: a report from the Interna-
tional Bone Marrow Registry. Bone Marrow Transplant.
1996;17:S5S6.
8. Gratwohl A, Hermans J, Niederwieser D, et al. Bone marrow
transplantation for chronic myeloid leukemia: long-term
results. Chronic Leukemia Working Party of the European
Group for Bone Marrow Transplantation. Bone Marrow
Transplant. 1993;12:509 516.
9. Cortes J, Kantarjian H, OBrien S, Robertson LE, Pierce S,
Talpaz M. Result of interferon-alpha therapy in patients
with chronic myelogenous leukemia 60 years of age and
older. Am J Med. 1996;100:452 455.
10. Kantarjian H, Smith TL, OBrien S, Beran M, Pierce S, Talpaz
M. Prolonged survival in chronic myelogenous leukemia
after cytogenetic response to interferon-alpha therapy. The
Leukemia Service. Ann Intern Med. 1995;122:254 261.
11. Italian Cooperative Group. Interferon alfa-2a compared
with conventional chemotherapy for the treatment of
chronic myeloid leukemia. The Italian Cooperative Study
Group on Chronic Myeloid Leukemia. N Engl J Med. 1994;
330:820 825.
12. Chronic Myeloid Leukemia Trialists Collaborative Group.
Interferon alfa versus chemotherapy for chronic myeloid
leukemia: a meta-analysis of seven randomized trials. J Natl
Cancer Inst. 1997;89:1616 1620.
13. Kantarjian H, Sawyers C, Hochhaus A, et al. Hematologic
and cytogenetic responses to imatinib mesylate in chronic
myelogenous leukemia. N Engl J Med. 2002;346:645 652.
14. Talpaz M, Silver R, Druker B, Goldman JM, et al. Imatinib
induces durable hematologic and cytogenetic responses in
patients with accelerated phase chronic myeloid leukemia:
results of a Phase 2 study. Blood. 2002;99:1928 1937.
15. Kantarjian H, Talpaz M, OBrien S, et al. Imatinib mesylate
Old Age and Prognosis in CML/Cortes et al. 1113
for Philadelphia chromosome-positive, chronic phase my-
eloid leukemia after failure of interferon-alpha: follow-up
results. Clin Cancer Res. 2002;8:21772187.
16. Kantarjian H, OBrien S, Cortes JE, et al. Treatment of Phil-
adelphia chromosome-positive, accelerated-phase chronic
myelogenous leukemia with imatinib mesylate. Clin Cancer
Res. 2002;8:21672176.
17. Kantarjian H, Cortes J, OBrien S, et al. Imatinib mesylate
(STI571, Gleevec) therapy for Philadelphia chromosome-
positive chronic myelogenous leukemia in blast phase.
Blood. 2002;99:35473553.
18. Druker BJ, Talpaz M, Resta DJ, et al. Efcacy and safety of a
specic inhibitor of the BCR-ABL tyrosine kinase in chronic
myeloid leukemia. N Engl J Med. 2001;344:10311037.
19. Druker BJ, Sawyers CL, Kantarjian, H. Activity of a specic
inhibitor of the BCR-ABL tyrosine kinase in the blast crisis of
chronic myeloid leukemia and acute lymphoblastic leuke-
mia with the Philadelphia chromosome. N Engl J Med. 2001;
344:1038 1042.
20. Larson R. Imatinib (STI571, Gleevec) as initial therapy for
patients with newly diagnosed Ph chronic myeloid leuke-
mia (CML): results of a randomized Phase III study vs inter-
feron-alfa cytarabine (IFN AraC). Blood. 2002;100:4a.
21. Surveillance, Epidemiology, and End Results (SEER) Pro-
gram. Public use CD-ROM (19731994). Bethesda: National
Cancer Institute, DCPC, Surveillance Program, Cancer Sta-
tistics Branch, 1997.
22. Kantarjian H, Cortes J, OBrien S, et al. Imatinib mesylate
therapy in newly diagnosed patients with Philadelphia chro-
mosome-positive chronic myelogenous leukemia: high in-
cidence of early complete and major cytogenetic responses.
Blood. 2003;101:97100.
23. Kantarjian H, Cortes J, OBrien, et al. High rates of early
major and complete cytogenetic responses with imatinib
mesylate therapy given at 400 mg or 800 mg orally daily in
patients with newly diagnosed Philadelphia chromosome-
positive chronic myeloid leukemia in chronic phase (Ph
CML-CP). Proc Am Soc Clin Oncol. 2002;21:261a.
24. Kantarjian HM, Dixon D, Keating MJ, et al. Characteristics of
accelerated disease in chronic myelogenous leukemia. Can-
cer. 1988;61:14411446.
25. Kantarjian HM, Keating MJ, Talpaz M, et al. Chronic my-
elogenous leukemia in blast crisis. Analysis of 242 patients.
Am J Med. 1987;83:445 454.
26. Rodriguez J, Cortes J, Smith T, et al. Determinants of prog-
nosis in late chronic-phase chronic myelogenous leukemia.
J Clin Oncol. 1998;16:37823787.
27. Sacchi S, Kantarjian H, OBrien S, et al. Long-term fol-
low-up results of alpha-interferon-based regimens in pa-
tients with late chronic phase chronic myelogenous leuke-
mia. Leukemia. 1997;11:1610 1616.