From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
Clinical
By
Staging
Kanti
R. Rai,
of Chronic
Arthur
Sawitsky,
Robert
N.
Levy,
Lymphocytic
Eugene
and
HRONIC
while
have
others
LYMPHOCYTIC
a variable
live for
measurable,
him
diagnosis
an
accumulative
According
therefore,
the
by physical
than
clinical
From
the
Division
N. Y. 11040.
the
and
found
of
Long
of Environmental
Department
the
Center,
Research
with
CLL
inactive
population
Island
on
New
to
York
be
We
of
CLL
is
University
lymphocytes.
span,
during
more
abnormal
at
a system
concept
Medical
and,
the
et al.3 reveal
data
obtained
propose
Laboratory,
aid
that
a long
life
in a patient
Medical
National
might
time
of
Dameshek’s
Jewish-Hillside
Brookhaven
Medicine,
tend
to
at the
suggested
survivors.
is based
which
et al.2 and Zippin
is short,
all the
studies
longer
known
after
diagnosis,
accepted
and
either
Dameshek’
of Boggs
of diagnosis
which
of Hematology.
Medical
to a physician
illness.
laboratory
among
CLL
is a disease
CLL
lymphocytes
have
lymphocytes
accumulate
The observations
from the time
data
staging
of the
the
such
examination
diagnosis
available
of a patient
of a functionally
to Dameshek,
more
and more
course
of the disease.
that, when the survival
the
course
disease
(CLL)
some
patients
die within
1 yr
than
10 yr. There
are no generally
of prognosis
or during
D. Chanana,
stage III, 19; stage
IV, 19. The median
survival
for the entire
series was 71 me.
The prognostic
significance
of the stage remained
even after adjustment
was made
for age and sex. However,
both sex and
age were shown
to be poor predictors
of
survival
after adjustment
for stage.
The
method
of staging
proved to be a reliable
predictor
of survival
whether
used
at
diagnosis
or during
the course of the disease. The proposed
staging
system was an
equally
accurate
indicator
for survival
when
applied
to two
other
previously
published
studies of large series of patients.
parameters
assessment
Arjun
S. Pasternack
LEUKEMIA
course;
longer
standardized
in the
P. Cronkite,
Bernard
A method
of clinical
staging
of chronic
lymphocytic
leukemia
(CLL) has been proposed which is based on the concept
that
CLL is a disease
of progressive
accumulation of nonfunctioning
lymphocytes:
stage
0, bone marrow
and blood lymphocytosis
only; stage I, lymphocytosis
with enlarged
nodes;
stage
II, lymphocytosis
with
enlarged
spleen
or liver or both;
stage
Ill,
lymphocytosis
with anemia;
and stage IV:
lymphocytosis
with
thrombocytopenia.
Analysis
of 125 patients
in the present
series showed
the following
median
survival
times
(in months)
from
diagnosis:
stage 0, >150;
stage I, 101; stage II, 71;
C
Leukemia
for
and
have
Hyde
Park.
Center.
New
Upton.
N. Y.
11973.
and
Center,
New
York,
N. Y.
10016.
Submitted
January
Supported
by
Ca
11028-07
from
Environmental
7, 1975;
the
United
the
Health
accepted
States
February
Atomic
28.
1975.
Energy
National
Cancer
Institute
Sciences;
United
Leukemia
Commission;
and
No.
Fund,
PHS
ES-00260
inc.;
Research
from
and
the
Grants
National
National
Leukemia
noted
institutions.
No.
3R10
Institute
of
Association.
Inc.
The
work
human
described
patients
in
who
this
were
report
treated
was
in
performed
accordance
at
with
the
the
above
precepts
established
It
in
the
involved
Helsinki
Declaration.
Address
Hillside
©
Blood,
for
reprint
Medical
1975
by Grune
requests:
Center,
Dr.
New
Hyde
& Stratton,
Inc.
Vol. 46, No. 2 (Au9ust),
1975
Kanti
Park,
R.
Rai.
Division
of
Hematology.
Long
Island
Jewish-
N. Y. 11040.
219
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
220RAI
ET AL.
tested
its validity
follow-up
study
in predicting
survival
in a retrospective
of a large number
of patients.
MATERIALS
Criterionfor
The
criterion
lymphocytosis
cell
of diagnosis
tion.
The
staging
tered
the
study
A ccrual
The
The
diagnosis
oratory
practice
the
other
Long
Utah,
on this
series
patients
to
based
on
extensive
data
allow
data
on
at diagnosis
were
adequate
fever,
alive
data
without
then
other
at
excep-
or
who
en-
follow-up
the
and
to allow
between
the
or
by
literature2’4
one
data
very
time
at
of
diagnosis.
we were
able
us to “stage”
to find
them
1971.
Lab-
the
private
December
analyzed
our
31,
seen
available
152
the
on
recently
84
published
Hansen’s
patients
in
re-
staging
in Salt Lake
us with
1963.
with
proposed
patients
were
and
in
was
provided
Hansen4
1954
us
also
to
130
diagnosis
between
of
study
were
kindly
and
National
1, 1973.
according
comprised
1941
Brookhaven
was June
diagnosis
et al.2
disease
made
the
Center
R. Boggs
in Copenhagen
breath,
easy
bruising,
recurrent
considered
this
analysis
to
to be enlarged
separate
enlargement
Proposed
data
of
these
a report
report
this
raw
includes
study
whose
at diagnosis.
stage
marrow
in
enlarged
spleen
findings
at
the
made
nodes.
or
at
mm).
Hematologic
hematologic
counting
Clinical
at
liver
At
At
were
33%).
regard
diagnosis
Anemia
on
Nodes,
or
type
were
and
spleen,
of
anemia,
lymphocytosis
organomegaly
the
or
with
or
I, the
II,
may
bleeding.
more
in
may
at
or
may
e.g.,
hemolytic
with
thrombocytopenia
or may
not
or
and
made
had
in
barely
At
as well
more
lymphocytosis
lymphocytosis
clinical
than
be enlarged.
otherwise.
as in
40#{176}/aor
were
were
less
not
who
with
diagnosis
(hemoglobin
was
blood
blood,
be enlarged.
severe:
of Diagnosis
in
at diagnosis
not
anemia
may
Time
loss,
(3)
spleen,
attempt
those
only,
findings
findings
may
liver
at the
or
nodes,
No
from
weight
liver;
Lymph
lymphocytosis
mm
moderate:
or
examination.
of CLL
were
stage
stage
spleen,
organomegaly
Staging
Nodes
and
physical
massive
diagnosis
clinical
nodes,
infections,
lymphocytosis
to the
large
15,000/cu
clinical
or both.
from
with
lymphocytes,
diagnosis
with
100,000/cu
for
findings
marrow).
less than
findings
organs.
0, the
lymph
enlarged
the
of these
(absolute
lymphocytes
was
patients
(2)
as (1) mild; only weakness;
if palpable
the
Criteria
clinical
hematocrit
recorded
symptoms
of
palpable
were
pressure
were
All
in
at
and
liver
platelet
study
all
patient entry into the
of
Boggs
sweats,
shortness
bone
other
lymphosarcoma
hematologic
were
was
of
information
Dane
patients,
related to CLL
night
At
all
any
Observations
Symptoms
The
of
Dr.
their
seen
have
this
basis;
Center
Medical
time
data
“stage”
of CLL
who
125 patients,
reported
study
into
absolute
to
records.
Research
follow-up
the
1964.
patients
and
Period
for
adequate
clinical
Clinical
with
The
us to
all
to
If adequate
entered
a prospective
a sustained
be attributed
known
study.
were
on
comprising
from
was
not
Patients
and
cut-off date for new
1945 and
189 cases
type.
of medical
Medical
patients
study
could
in the
Jewish-Hillside
of
of patients;
1968,
this
which
patients
in
and
date
in
included
observed
series,
survival
below).
between
series
not
available,
were
Island
The
of
(described
City,
were
at the
closing
large
duration
criteria
cell
review
in our
the common
to
mature
devised
4 yr
seen either
of hematology.
Two
small
Sources,
ofCLL
at
and
spect
METHODS
CLL
marrow
a retrospective
were
or
1971,
were
of
bone
were
next
of Patients,
patients
and
leukemia
criteria
from
diagnosis
of the
of CLL
in the
obtained
the
blood
were
or lymphosarcoma
were
for
in peripheral
Lymphocytes
the time
as prospective
Diagnosis
essential
cause.
AND
as well
At
(platelet
stage
II
with
III,
g/l00
No
the
ml
or
distinction
clincal
stage
count
less
IV,
than
be present.
Methods
tests
varied
were
performed
between
the
by
different
the
prevailing
institutions.
standard
The
methods.5
indirect
method
The
of
method
calculating
of
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
STAGING
OF
CLL
221
the platelet
count
cytes
peripheral
in the
stitutions
from
gradually
Cases
Included
four
smear
in our
series
at the
were
on follow-up
examination
any
had
and
patients
in
Asymptomatic
when
treatment
ascribed
for
large
masses
Duration
In
patients
in
were
available
Statistical
The
the p value
t test
curves
squared
for
series
Boggs
et
often
the
or
therapy
to enlarged
of
local
and
of
15,000
less
than
patients)
40%.
and
All
or
these
more
those
al.2
and
reports;
were
the
in
and
the
change
the survival
of
duration
stage
from
was
present
nodes
was
Some
of
diagnosis
and
or
for
employed
patients
also
re-
of treatment.
to date
dates
series,
the
for
were
lost
to
of
the
disease,
into
the
new
stage
to
data
follow-up
change
in disease
adequate
follow-up
course
of entry
or
lost
follow-up
patients
from
When
of death
of whom
closing
during
choice,
modalities
alive-none
these
the time
therapy;
of
Stage
still
of
or
methods.
used.
Hansen,4
both
drug
other
date
patients
of
first
symptoms.
of Disease
from
for
antileukemia
spleen
or without
rarely
Change
accepted
any
thrombocytopenia
pressure
with
was calculated
1, 1973
receive
anemia
(ECIB)7
and
respective
was
death
rates
group,
i.e.,
for
prognosis
two
with
(the
latter
Differences
either
in
at
the
differences
the
logrank
it yields
stages
count
with
(lower
data
to
on
such
pa-
was
noted.
respect
in blood,
and
statistically
survival
to the
bone
mar-
significant
subgroups
the
average
aforementioned
was divided
survival
said
0.001,
having
into
to
would
a p value
in
this
four
groups:
significant
or
0.05
these
if
greater
than
0.05
were
have
rate
a relative
study
to
be
made
be
p
understood
levels
are said
(p
involved.9
than
groups
chithe
chia sig-
of
rate
1.00.
A
greater
than
death
rate
associated
sex,
>70
chi-
patients
between
the groups
both
methods,
relative
included
6 1-70,
value
death
as
of
the
summary
among)
The
death
of
methods9’1#{176} provided
(or
50, 51-60,
should
method
use
curves
in distribution
survival.
In
described
were
probability
by
two
precise
compared.
was
adjustments
be
0.01,
of
a relative
considered
actuarial
entirety
than
more
between
being
have
than
more
Both
combined,
would
their
one-degree-of-freedom
into
account
differences
to be associated
with
patients
factors
product-limit
in
a somewhat
in
average
the
Mantel’0
differences
the
by
test when
the
test.
all
better
the
different
considered
compared
logrank
for
The
which
were
obtained
the
taking
thought
for
a prognosis
diagnosis
on
while
factors
obtained
for
in the
lymphocyte
means
were
curves,
since
based
Appendix).
and
were
curves
employed,
than
patients
mean
between
series
survival
average
worse
Statistical
three
survival
evaluating
were
the
comparing
of symptoms,
differences
the
approximation
test
for
to the conditional
statistic
nificance
used
The
of
other
in-
0.05.
stratified
by one factor
with respect
to other
All
20
28%
conventionally
not
most
cyclophosphamide
incidence
approximation
nificant
our
was
Meier.8
comparison
squared
CLL
presence
and
percentage;
and
group
all
of the disease.
to
did
was
of blood
of
was less than
Survival
Kaplan
square
the
to detect
age at diagnosis,
lymphocyte
For
between
was
in
lymphocytosis
according
Radiation
of June
and
erythro-
l960s,
Methods
two-tailed
mean
in
their
restaged,
1000
the
count
11-14,000
ranged
alive
until
the date of last contact.
Survival
was also separately determined
for our series.
were
tients
for
studies
as described
were considered
higher number)
lymphocyte
patients,
blood
In
mm
the course
activity
when
Duration
date
the
treated
therapy
mustard,
closing
showed
per
1950s.
method.
absolute
in four
during
disease
activity.
or
of survival
follow-up.
were
to
of Survival
the common
whose
chlorambucil
irratiation
nitrogen
Recording
row
added
extracorporeal
Vincristine,
were
stable
to the disease
discrete
ceived
late
Modalities
series
indicated,
were
of platelets
the
15,000/cu
lymphocytes
as confirmed
with
was
corticosteroids
fever
our
patients
patients
marrow
number
until
phase-contrast
Than
eventually
CLL
the
often
(8-10,000
24) whose
without
Therapy-Criteria
24
of diagnosis
these
question
from
most
ofLess
also
time
(among
and
used
to the Brecher-Cronkite6
patients
All
count
was
Lymphocytosis
mm
patients
erythrocyte
switched
ofinitial
15,000/cu
the
blood
for
the
subgroup
whole
with
a
I,
while
a sub-
less
than
I. (see
with
stage,
survival
in
and
at
age
yr).
to
values)
mean
which
to be nonsignificant
statistically
are
(NS).
indicated.
sig-
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
222
RAI
ET AL.
RESULTS
Clinical
the
Data
(Present
Series)
One hundred
twenty-five
patients
closing
date,
and 79 had died.
and
1949
on four
patients,
between
and 1969 on 84 patients,
records
on 65 patients
clinically
observed
are
The
included
diagnosis
1950
during
the
(1) the patients
in stages
(2) the stage
0 patients
in Table
ready
dead
1, the
(except
differences
were
blood
of stage
stage
I mean
stage,
and
stages
III and
was less than
past
patients
Among
70%
tient
by each
differences
were
the
at the 0.05
significantly
count
II mean
alive
I in which
not significant
0 patients
was
patients
was
not
lymphocyte
basis.
generally
reverse
was
nodes.
and about
II at
18%
Among
50%
had
was
diagnosis,
III patients
Although
younger
than
true).
However,
different
lower
patients
in stage
had enlarged
spleen
had enlarged
nodes,
and about
each
in stages
III and IV had
had
of both
an
and
not
those
althese
level in
stage
IV;
from
significantly
92%
enlargement
of diagnosis
significant:
level. The mean
lymphocyte
lower
than
that
of only
significantly
level
1960
The medical
patients
were
stage
at the time
were statistically
any
than
other
those
of
count
at diagnosis
in stage
0, and two
and IV. The bone
marrow
lymphocyte
lower
than that of stage IV.
in stage
liver,
between
11 patients.
and
60
IV. Among
the 24 patients
whose
lymphocyte
15,000/cu
mm,
14 were in stage
II, four were
an enlarged
enlarged
nodes,
26 patients,
6 yr on a prospective
still
in stage
each were in stages
I, III,
0 and II were significantly
had
on
0, I, and II were younger
than the stage
were younger
than
the stage
IV patients.
lymphocyte
stage
1959
and during
1970 and 1971 on
were
reviewed
retrospectively,
The relevant
clinical
data of patients
are shown
in Table
1. The following
age
shown
and
in our series;
46 were alive on
was first made
between
1941
ratio
enlarged
organs,
and
III at diagnosis,
or liver;
among
of stages
spleen,
64%
90% had
stage
IV
50% had enlarged
spleen
neither
nodes,
spleen,
nor
64%
also
had
enlarged
patients,
or liver. One
liver palpable
paat
diagnosis.
Table
1.
Clinical
Data
(Chroni
of Patients
c Lymphocy
According
to Stage
at Diagnosis
tic Leukemia)
Per Cent
Blood
With
Stage
at
Diagnosis
0
Mean
No.
of
Per Cent
at Diagnosis
Patients
Male
(Yr
±
SD)
22
28
54
±
12
Per Cent
Still
Alive
Bone
Lymphocyte
Symptoms
Age
Count
at Diagnosis
x
10
lymphocytes
(Mean)
(Range)
(Severity)
(Mean)
5
25
(15-130)
52
(40-85)
42
(15-150)
63
(50-88)
38
(15-175)
61
(40-83)
78
(40-600)
79
(40-93)
80
(15-190)
92
(60-93)
40
(15-600)
63
(40-93)
82
(Range)
Morrow
Per Cent
(mild)
I
29
59
57
±
12
28
62
(mild)
II
39
82
57 ±
12
49
36
(mild to
moderate)
III
21
57
66
±
11
5
71
(moderate
to
severe)
IV
14
64
62
± 8
0
100
(moderate
to
severe)
0 to IV
125
62
58
±
12
37
50
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
STAGING
OF
223
CLL
Table 2. Data
on Therapy Administered
According to Stage at Diagnosis
(Chronic Lymphocytic Leukemia)
Mean
Duration
Between
Disease
No.
Stage
of
at Diagnosis
Per Cent
Diagnosis
and
Initiation
Therapeutic’
of First Rx
Patients
of
Patients
Agent
(yr)
Never
Used
the Course
During
Receiving
of Disease
0
22
5.3t
C(8),
I
29
2.8t
C(14),P(12),RT(4),E(4)
II
39
1.6k
Therapy
P(3), RT(2), E(2)*
59
21
V(1), Cy(1)
C(29),P(12),RT(14),E(5)
V(1),
*C
21
0.4
P(14),C(13),RT(5),E(3)
IV
14
0.3
P(9), C(9),
chlorambucil;
P, prednisone
irradiation
in parenthesis
or
of
denotes
other
blood;
number
V,
corticosteroids;
vincristine;
of patients
RT,
Cy,
23
HN2(1)
Ill
E, extracorporeol
Number
Cy(1),
5
RT(7),
local
E(3)
0
radiation
to
cyclophosphamide;
spleen
HN2,
or
nitrogen
nodes;
mustard.
so treated.
tWhen
therapy
was begun nearly all patients
were in stage II, III, or IV.
With
the exception
of two patients
who were in Stage II, all others
were
in Stage
Ill or
IV when
E
was used.
§With
one exception:
The
initiation
mean
of
agents
detailed
used and
in Table
serves
special
published
received
the percentage
2. Therapy
comment.
other
or
disease.
The
Only
in an
two
other
agents
the first
is shown
who
which
method
this
treatment
and
had
either
terminal
were
of
diagnosis
in Table
recently
time
III
and
in stage
were
still
0 or
II,
or
been
already
to the
IV
of ECIB
diagnosed,
of the patients
these
patients
had
nonresponsive
stage
also
de-
have
patients
become
of
therapy
are
in 17 patients,
treatment
15
II at the
in either
stage
closing
date.
of CLL
and
the
2; the therapeutic
of this
phase
in stage
previously,
diagnosis.
never
received
was applied
and
were
Nearly
two-thirds
2); a majority
17 yr after
of
or
who
therapy
of
had
were
their
not
re-
in good
0 never
received
alive
and in our
while
two
patients
had
of Survival
Overall
survival
ies (125 patients),
and survival
according
to stage
at diagnosis.
the data of Boggs
et al. (84 patients),
and
(152 patients)
which
shows
are represented
overall
survival
at diagnosis.
The
end
time
advanced
care
and remained
to stage IV at the
Duration
between
therapy
rationale
the
patients
clinical
condition.
any therapy
(Table
follow-up
progressed
At
therapy
of patients
with ECIB,
conventional
were
any
first received
interval
in years
first anti-leukemia
elsewhere.7
agents
ceived
one patient
at 144 mo
curve
because,
in the panels
A,
as well as survival
for
stage
with
the
0 of the
B, and C,
according
present
product-limit
series
actuarial
the
The present
serdata of Hansen
respectively,
to the stage
(Fig.
IA)
method,
the
of Fig. 1
of disease
is shown
to
cumulative
per cent surviving
is computed
and plotted
each time a death
occurs;
thus
the
last point
on the curve
represents
the last observed
death.
Seven
patients
were
still alive in this group
at 144 mo. The duration
of survival
of every
patient
of
the present
series,
dead
or alive on the closing
date,
is shown
in Fig. 2. Figure
lB shows
24 mo,
only
one
lost
three
patients
to follow-up
in stage
at 37 mo,
0 in the
and
one
series
alive
of Boggs
at
116
et al.,
mo.
one
Therefore
alive
stage
at
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
224
RAI
ET AL.
5
z
5
>
z
>
>
>
z
z
w
U
U
a-
a-
z
>
>
U,
z
U
a.
Fig. 1.
ent series.
series.
0 is represented
median
survival
in
Survival
according
survival
curves
for
3A)
survival
et al. (Fig.
the younger
Fig.
times
the
appears
3B) the age
age group,
lB
for the
to age
three
by
a dashed
three
studies
straight
are
with
advancing
group
of 51-60
yr appears
as is also the case in the
(B)
line
shown
at diagnosis
and according
series
by age at diagnosis.
to be poorer
Survival
curves
Boggs’
at
100%
in Table
by stage.
series.
(C)
survival.
presHansen’s
(A)
The
3.
to sex.
Figure
3 shows
In the present
series
(Fig.
age;
in the
series
of Boggs
to have a better
survival
series of Hansen
(Fig. 3C).
than
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
STAGING
OF
CLL
225
CHRONIC
0
32
LYMPHOCYTIC
I
I
I
LEUKEMIA
I
II
I
I
PATIENTS
PATIENTS
#{176}
0
STILL
ALREADY
ALIVE
DEAD-I
-0
0
o
e
e
6
0
(I,
#{149}o
e
Ui
>-
0
z
II
0
Ui
12
#{149}oo
0
0
I-
0
-j
0
o#{149}.e
10
0
00
>
I
e
0
0I
b.c
ID
e
6
#{149}
o
o
cf
0
00
ee
4
I..
e
0
o?.Cb
I
00
eel
00
I
2-
I.e
I
0
0
I
DISEASE
Fig. 2.
Survival
times
Table 3. Median
(A)
Stage
at
by stage
In
U
STAGE
for dead
AT
and
DIAGNOSIS
alive
patients
in the present series.
Duration of Survival According to Stage at Diagnosis
(Chronic Lymphocytic Leukemia)
(B) Bog
Prese nt Series
gs
et
al.
(C)
Hansen
No.
Median
No.
Median
No.
Median
of
Survival
of
Survival
of
Survival
Patients
(mo)
Patients
0
22
>150
3
-
I
29
101
7
130
52
60
II
39
71
41
108
23
47
Ill
21
19
13
9
32
26
IV
14
19
20
42
39
20
125
71
83
77
152
40
Diagnosis
Overall
(no)
Patients
6
(no)
180
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
226
RAI
I
I
I
I
I
TOTAL
90
80
I
ET AL
I
DEAD
AGE
GROUPS
25
45
14
25
A
5O
51-60
36
27
+
61-70
19
13
x
>70
120
144
o
V.,
z
>
\
6O
>
a.
U,
\4\
\
50
I-
z
w
U
40
L1
a.
30
0
24
I
72
48
96
168
A
MONTHS
FROM
DIAGNOSIS
C
MONTHS
FROM
DIAGNOSIS
192
216
V.,
z
>
>
a.
U,
I-
z
w
U
a.
U
a.
Fig. 3. Survival curves by age.
ries. (B) Boggs’ series. (C) Hansen’s
Survival
curves
for the three
Females
appear
to have a better
the
series
between
of Hansen
males
Statistical
stages
the
the
after
series
females
analysis
of each
differences
survival
(Fig.
and
series
4C),
series
according
survival
in the
but
in the
of survival
(Fig.
of Boggs
for
et al.
stage
were
in
4A)
appears
to be no
difference
in survival
of Boggs
et al. (Fig.
4B).
data.
1) were
When
compared
remained
highly
significant,
curves
of the four age groups
adjustment
to sex are shown
present
series
(Fig.
there
series
and
sex,
no
longer
(A) Present
series.
the
after
survival
curves
adjustment
for
as shown
in Table
of each series
(Fig.
differences
significant;
in the
present
in the
series
Fig.
and
of
age
4.
in
the
five
and
sex,
4. However,
when
3) were compared
series
of
and
Hansen,
in the
sig-
se-
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
STAGING
OF
227
CLL
00
90
80
70
V.,
z
z
? 60
>
a.
U,
>
>
a.
D
U,
50
I-
z
I-
z
U
U
40
a.
a.
U
U
a.
30
20
lO
0
O
24
48
A
72
MONTHS
FROM
DIAGNOSIS
V.,
z
>
>
a.
U,
I-.
z
U
U
a.
U
a.
Fig. 4.
Survival
curves
by sex.
ries. (B) Boggs’ series. (C) Hansen’s
nific#{224}nce was
curves
in any of the
in Table
6.
Change
In the
their
barely
according
to
three
During
present
series,
disease,
stage
(p
4) were
not
adjusted
for
so it was
II at
C
obtained
(Fig.
when
the
that
Course
we had
possible
one stage to another.
For
sis first showed
evidence
sidered
se.
sex
series
of Stage
(A) Present
series.
data
for
0.05),
<
and
shown
found
FROM
in Table
to
be
and
Survival
during
when
the
a patient
of the
duration
who
spleen
of
survival
as shown
of the Disease
all patients
The
at diagnosis,
age
us to ascertain
5.
different
and
on
DIAGNOSIS
significantly
stage
example,
when
of enlargement
time,
as
MONTHS
(Present
the
Series)
clinical
a patient
course
moved
of
from
was in stage I at diagnoor liver,
he was con-
survival
from
that
time
to
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
228
RAI
Table 4.
Survival
by Stage Adj usted for Age at Diagnosi sand
Fro m Diagnosis
at
N’
0
E
(p value)
O/E
series
0
22
4
17.51
0.23
I
29
20
21.57
0.93
II
39
21
27.24
0.77
III
21
20
8.22
2.43
IV
14
14
4.46
3.14
All
125
79
79.00
1.00
of Boggs
49.63
(p <0.001)
et al.
0
3
0
1.06
0.00
20.90
I
7(l)t
1
3.64
0.27
(p <0.001)
II
41(2)
14
21.30
0.66
Ill
13
13
4.69
2.77
IV
20(1)
13
10.31
1.26
All
84(4)
41
41.00
1.00
(C) Data
*N,
Diagnosis
of present
(B) Data
Sex
2(4df)
Stage
(A) Data
ET AL.
of Hansen
0
6(2)
0
8.38
0.00
I
52(8)
36
54.30
0.66
II
23
21
32
30
20.68
18.48
1.02
III
IV
39(2)
36
21.16
1.70
All
152(12)
123
123.00
number
of patients
based
on survival
mated
by chi-square
tNumber
in group;
for all patients;
O/E,
denotes
number
of deaths
x2
relative
death
rate.
under
the null hypothesis.
deaths
included
not
1.62
1.00
in group;
of freedom
with 4 degrees
in parenthesis
duration
0, observed
(4 df),
32.27
(p <0.001)
E, expected
logrank
in 0 due to inability
number
test statistic
which
to cross-match
of
deaths
is approxi-
for sex, age,
and
of follow-up.
death
or to the
stage)
was
closing
entered
date
into
the
(if the
patient
survival
was
data
for
still
alive,
patients
irrespective
changed
of current
to stage
II.
When
that patient
became
anemic,
the duration
of survival
from
that
point
was entered
into
survival
data
for patients
changed
to stage
III, and
so on. The
median
duration
of survival
of patients
originally
in a stage
at diagnosis,
as
well as the median
survival
of patients
entering
that
stage
from
other
stages
can be seen in Fig. 5. Differences
in survival
time in any stage,
whether
a group
of patients
was
diagnosed
initially
during
the course
of the disease,
Among
the patients
in stages
times
of
mained
stage
70
mo
in stage
II patients
follow-up
or
longer),
0, 27%
is rather
50%
of stage
have
of
for the
stage
patients
I patients
remained
short
in that
or the
were not significant.
0, I, and II (the
stages
in stage
have
group
remained
entered
with
0 at
the
median
survival
diagnosis
in stage
date.
have
I, and
in stage
II at the
closing
The
patients
in stages
I or II at diagnosis
stage
re-
50%
duration
who
of
of
have
so far not changed
their
disease
stage
(and therefore
will not b&detailed
here);
for the 11 patients
in stage 0 at diagnosis
who have remained
in stage 0, the survival data
are as follows:
three
patients,
3.5-6.0
yr; five patients,
7.0-12.5
yr;
and
one
patient
each
included
in the 59%
antileukemia
therapy.
at 18, 23, and
of stage
32 yr after
0 patients
(Table
diagnosis.
2) who
These
have
not
patients
are
also
been
given
any
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
STAGING
OF
CLL
229
Table 5. Survival
Age
by Age at Dia gnosis Adjust ed for Stage and Sex
Fro m Diagnosis
x2
at Diagnosis
(yr)
(A) Data
N’
of Present
0
(3 df)
E
O/E
(p Value)
Series
50
51-60
25(4)t
10
15.75
0.63
4.04
45(4)
21
22.65
0.93
(NS)
61-70
36(1)
26
20.56
1.26
>70
19(1)
12
10.04
1.20
All
125(10)
69
69.00
1.00
(B) Data of Boggs et al.
50
23(3)
9
9.60
0.94
1.64
51-60
24
7
9.78
0.72
(NS)
61-70
23(3)
14
12.61
1.11
>70
13(1)
8
6.01
1.33
All
83t(7)
38
38.00
1.00
(C) Data
of Hansen
50
51-60
18(3)
13
18.72
0.69
33(4)
24
33.57
0.71
61-70
63(4)
50
43.07
1.16
37
28.65
1.29
124
124.00
38
>70
152(11)
All
*N
number
of patients
in group;
0,
based
on survival for all patients; O/E,
mated
by chi-square
tNumber
and
duration
Does
with
3 degrees
in parenthesis
denotes
observed
number
relative
death
of freedom
deaths
of deaths
rate.
under
x2
<
logrank
E, expected
test statistic
number
of deaths
is approxi-
which
the null hypothesis.
not included
in 0 due to inability
to cross-match
for
sex,
include
not
6.
one
patient whose
Survival
age
Fro m Diagnosis
at diagnosis
was
not available.
by Sex Adjus ted for Stage and
Age at Diagnosis
x(l
N’
(A) Data
of Present
0
E
O/E
df)
(p value)
Study
Males
77(l8)t
35
35.13
1.00
0.02
Females
48(4)
22
21.87
1.00
(NS)
125(22)
57
57.00
1.00
63(17)
16
16.04
1.00
0.05
9
8.96
1.00
(NS)
1.00
All
(B) Data
Males
of Boggs
et al.
Females
21(3)
All
84(20)
25
25.00
101(24)
69
61.58
1.12
3.08
51(11)
31
38.42
0.81
(NS)
152(35)
100
100.00
1.00
(C) Data
of Hansen
Males
Females
All
number
of patients
in group;
0,
observed
based on survival
for all patients;
O/E, relative
with 1 degree of freedom
under the null hypothesis.
tNumber
and duration
stage,
of follow-up.
Table
*N,
0.05)
1.00
in group;
(3df),
8.03
(p
in parenthesis
of follow-up.
denotes
deaths
number
death
not included
of deaths
rate;
x2
in group;
(1 df),
in 0 due to inability
E, expected
summary
number
chi-square
to cross-match
of deaths
test
for
statistic
stage,
age,
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
230
RAI
H
0
C
,
,i
U,
c
N
V’s
4
a
V’s
‘-4
0
o
o
‘fl
N
a
,-4
H
o
a
I
x
Om
O
‘4J
on,
C
H
-.
‘S
4
0
-
C
‘
so
e>
4J4-4
U,
a
0
V’s
H
‘-4
z
‘0
C
so
no>
‘0
‘0
C
C
so
,o>
4J
4-4
U
no>
4JH
,n
,.
04J#.4
.
U
‘oowso
so
no>
OCO
n
‘0
C>
>
0
S
so
S
0
C
,
U
U
U,CU
U,
m
a
-a
C
0
0
U,
0
0
N
H
‘
H
N
.C
H
,.4
O
-4
N
0
H
4
V’s
N
‘“V
‘0
a
2
a
2
C.-.
x
C
Os
.
OO
C
I
N
-
‘0
‘0
C
so
‘0
C
so
no
CH
U,
0
“4
z
so
ao
-
C1J4
,
U
C
-
so
no
5’.p-4
‘4
-
‘0
,
U
U
S
C#’4
OCsoH
so
so
C4.H
,n
a
H
H
0
X
0
C
U,5Q4J
C
iJ5
-‘-4
0
V
V
‘“V
0
0
O
sO
-e
a
0
sO
0
U,
H
a
3
S
os
r
.4J
n,
as
H
,-4
z
0
r
0
C’s
C
0
‘0
C
C
s-
ao
si’.n
so
JH
U,
‘
so
H
VeQI’
so
o
o,-j
on H
U
U,
4
CO
3
-a
so
.
U
H
Q
s4J
4J5
U,
C
a
-a
I
E
0
0
a’
H
C
H
0
H
2
.C
a
E
Os
.
V’s
05
on
N
z
C
so
C
‘U
C
0
U
a
E
sI
so
so
no
CH
4J4
4.4
0
.
3
H
s
-CO
‘“V
U,n
0
U,
H
A
V3
‘44
‘a
20
Os
z
N
US
C
In.
C
so
so
.!
‘V
U,
C
‘
#{163}
V
ET AL.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
STAGING
OF
CLL
231
DISCUSSION
In addition
recently
have
which
may
presented
influence
prognosis
in-depth
confirmed
the
evidences
short
recent
to the earlier
report
provided
excellent
in CLL,
analyses
of their
observation
of disease
of Boggs
et al.,2 Zippin
reviews
of the literature
of
and
Boggs
at diagnosis
each
respective
et
tended
al.2
younger
people
than
reveal,
no
we believe
when
the
The
marized
for
men
relationship
survival
and
these
proposed
and
older
are
scheme
all
only
was
shorter
necessarily
establish
We
record
that
the
respectively,
the
was
recognize
that,
when
in the group
reviewed
liver
was
enlarged,
with
patients.
women
The
and
they
Boggs
corwill
disappear
present
series
1 .6 : 1 and the
was
the
case
not correlate
of organomegaly
we have
count
higher
for
are summean
age
,
the liver
was
retrospectively,
but
shorter
than
laboratory
patients
although
because
of symptoms
did
a higher
incidence
total
lymphocyte
tended
to be
groups
and
among
also
studies3’4
and sex and only a slight
As the following
discussion
apparent
tion such an observation.
The
marrow
lymphocyte
percentage
surviving
reports
is used.
58 yr. The incidence
surviving
group
had
the long-term
survivors.
palpable
in a patient’s
Hansen4
factors
These
longer
surviving
prognosis
for
clinical
characteristics
of patients
comprising
in Tables
1 and 2. The male : female
ratio
at diagnosis
vival. The
three
physical
severe
people,
between
survival
age at diagnosis.
discrepancies
staging
that
the
better
of these
of patients.
to be more
survival
time than was the case among
studies3’4
indicated
a significantly
et al.2 observed
relation
between
one
series
et al.3 and
concerning
no
and
long-term
surthan
noted
to be
it did not
reason
in blood
at diagnosis
the
with
to questhe
among
bone
the
survivors.
The
long-surviving
groups
had a greater
proportion
of patients
who never
received
any antileukemia
therapy,
and they also had a longer
interval
between
diagnosis
and start of such therapy,
but the choice
of agents
used among
the longor the
short-survivor
groups
was not
the poor prognosis
for patients
amount
of therapy
given
to
different.
in stages
them
or
disease.
The
increase
and tested
laboratory
clinical
staging
proposed
in the physical
and
lymphocytosis
of spleen
or
marrow
function
less
than
tive
and
as revealed
As
within
100,000/cu
alive,
II,
and
analysis
represents
of the disease,
in stage
mm).
All
in stage
I, enlargement
compromise
of bone
and
thrombocytopenia
were also anemic,
It should
be noted
as set forth
(hemoglobin
criteria
III
in
less
proposed
this
than
(although
a stepwise
from
mere
in
although
that the
staging
system
11 g/100
ml,
arbitrary)
none
criteria
include
platelets
are
objec-
measurable.
can be seen
3.5 yr from
In stages
still
by anemia
of stage IV patients
thrombocytopenic.
and thrombocytopenia
relatively
severe
forms
readily
in this
evidence
of stage 0 to enlargement
of lymph
nodes
liver
in stage
II, and thereafter
progressive
stage IV. A large majority
of stage III patients
were
of anemia
only the
From
our study
we cannot
tell whether
III and IV was because
of the greater
because
of the natural
history
of the
from
Fig.
diagnosis,
I, and
overall
0, there
survival
2 (the present
series),
all stage
and only one patient
in stage
III
was
a progressively
was
clearly
longer.
larger
The
IV
has
proportion
longest
patients
died
lived to 5 yr.
of
survivors
patients
of
the
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
232
RAI
entire
series
were
the
after diagnosis.
The association
three
patients
ofstage
in stage
at diagnosis
0 who
with
were
survival
alive
ET AL
at 23, 24, and
in the
present
32 yr
series
is evi-
dent from Figs.
IA and 2. Age and sex also appear
to be significant
prognostic
factors,
as indicated
in Figs.
3A and 4A. When
the five stages
are arranged
in
ascending
numerical
order
from 0 to IV, there
is a continuously
decreasing
survival as one proceeds
from
stage
0 to IV even after adjusting
for age and sex.
Table
4(A)
patients
shows
compared
parent
relative
not
the
to 3.14
for
by age
in Fig.
death
length
at diagnosis
Similarly,
differences
age
a role
sex
and
age
ported
in the
viz.,
by stage
0.00 1) after
observation
analyzed
the
were
data
while
not
of Hansen
are
and
Boggs
series
by Fig.
two series
for
stage
significant
0.05).
than
in the
other
two
Comparison
series.
This
series
fered
would
stages
CLL
series.
sex
difference
may
It will
be observed
from
Fig.
1 and
be maintained
being
an
apart
accumulative
has been
proposed,
hepatosplenomegaly
for
the
disease,
as an indicator
two other
series
of
re-
the
survival
comparable
to
in sur(p <
sex (Table
4). This
these
series
were
5), the
and
differences
just
and
partly
in
significant
expected
for
deaths
in either
of the
of Hansen
was
be ascribed
to the
of patients
in stages
III
frequency
and extent
of
different
from the other
4 that
in each
of the
three
rather
than
five, groups
which
I and II, and stages
III and IV.
follow-up
and treatment
the
following
reasons:
concept
enlargement
of nodes
with or without
nodes
accumulation
of lymphocytes-and
different
therapeutic
schedules.
cytopenia
and thrombocytopenia
grees
of compromise
of marrow
Table
of
of
not
unadjusted
(Table
data
of observed
examined
there were essentially
three,
with respect
to survival:
stage 0, stages
nonetheless
suggest
that
for future
after
which
were
and
When
fact that Hansen’s
series contained
a higher
percentage
and IV than did the other
two series.
Also,
perhaps
the
therapy
of patients
in Hansen’s
series
might
have been
two
entirely
the analysis
are predictive
do
by sex after adjustment
for stage and age revealed
no difference
two series (Table
6), again
confirming
our own findings.
Figure
1 and Table
3 show
that overall
survival
in the series
poorer
ap-
1. The observed
differences
remained
highly
significant
Boggs’
on
0
is
believed.
et al.2 The
and
for
stage
survival
differences
factors
of patients
made
for age at diagnosis
with
our own results.
controlling
for
increase
in the
are statistically
disappear
of staging
criteria
to
two
statistically
<
sex,
as heretofore
method
staging
Hansen4
as evidenced
for these
(p
and
by sex
at diagnosis
for these
adjustments
were
was in agreement
by age
survival
for stage
prognosis
curves
of our own series,
according
to stage
0.23
when
5(A) shows
a slight
but the differences
in survival
our
proposed
we applied
our
at diagnosis
from
Conversely,
at diagnosis
(Table
6(A)).
Thus,
that our proposed
staging
criteria
in determining
literature,
increasing
0.001).
<
adjustment
In order
to validate
prognosis
for survival,
those
vival
(O/E)
(p
after
whereas
as important
rate
disappear.
Table
age at diagnosis,
for stage
and
strongly
suggests
of survival,
play
death
IV patients
3A almost
rate with
significant.
adjustment
this series
relative
stage
(I)
which
From
the
the
present
without
hepatosplenomegaly
ought
to reflect
different
perhaps
in the future
(2)
Similarly,
anemia
with
or without
anemia
function
and
possibly
difWe
five
concept
of
staging
degrees
and
of
may be treated
by
without
thromboreflect
different
deof body
burden
of
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
STAGING
OF
CLI
lymphocytes
future,
233
and
of
different
the
future
this
time
by doing
the
staging
so, and
progression
tially
stage
biology
of
the same
from
an
whether
earlier
and
disease.
Perhaps,
be indicated.
(3)
into
three
nothing
important
of disease,
the
may
groups;
prospective
stages
I and II and stages
III and
(Fig.
5) our initial
observations
believe
like
overall
oftreatment
to redesign
combining
We
the
lines
the
outlook
he is found
stage
during
for
data
that
survival
large
has
followed
for
more
than
7 yr, and
none
is essen-
has
that in stage
time.
It might
sults
sider
of new therapeutic
all patients
with
expected
duration
measures
separately
CLL
as a homogeneous
for
rather
with
anti-
useful
in
agents,
nonprogressive
to record
each stage
population
a
received
considering
very long
uniquely
invesitgators
that
basis
is needed
among
the surin stage
0, a
therapy.
Clinical
staging
of CLL
should
therefore
prove
the patients
in stage 0 who need not be exposed
to cytotoxic
remains
for future
by
or enters
However,
leukemia
identifying
0 the disease
be advisable
at
be lost
is a pipeline-
to be in a stage
at diagnosis
the course
of the
disease.
of a larger
number
of patients
on a prospective
observation.
It is noteworthy
that,
in our series,
in stage 0 at diagnosis
who have so far remained
been
may
in
gained
a patient
close follow-up
to confirm
this
viving
patients
majority
be
there
for
in the
possible
will
follow-up
IV now.
indicate
again
It is always
for a
the re-
than
respect
to conto the
of survival.
ACKNOWLEDGMENT
We
wish
to
acknowledge
with
thanks
Fisichelli, Dr. Carrie
Hunter,
Mr.
Clyde
Morton
Urivetzky.
Sipe,
Brecher,
Dr.
Sylvan
Silver
and
B.
Dr.
David
P. Byar,
Green,
for
Dr.
reviewing
Dr.
Robert
this
work
the
Ruby
Bayard
are
Dr.
Dr.
As an illustration,
time
for
the
sex and
from
diagnosis
since there
are
1-yr period
of
combination
can
steps
age
of
four
used
using
the
the
patients
age
followup;
groups
the
for
Dane
R.
Coleman,
Dr.
Gee,
Dr.
Richard
T.
stages
ad-
suggestions.
periods,
sex)
the
in
survival
be described
(2)
for
construct
series
of
among
as follows:
each
the
age-sex
the
(eight
table
representing
for
a specific
Dying
No.
Alive
Total
Ath
Bth
A11
B11
II
A21
B21
N21
III
A31
B31
N31
N11
4L
All
the
[E(A31)
tables
combinations-denote
4L
M,1
stage III
contingency
expected
=
number
N31M11/T,],
(each
series
by
0.
each
age-sex
I
calculate
survival
combination
a 5 x 2 contingency
tables
five
(1) divide
0
for
all
George
as
No.
(3)
Mr.
Dr.
Timothy
and
Vincent
Muniz,
Boggs,
Dr.
Schneiderman,
differences
can
1-yr
each
table
be characterized
Stage
into
i-th
A.
Dr.
Francisco
APPENDIX
test
within
B. Cowen,
Mr.
Dr.
Morton
constructive
to evaluate
logrank
Richard
Leichter,
to
Marvin
STATISTICAL
justing
Mr.
Carl
grateful
Clarkson,
for their
of
Mr.
We
Greenwald,
and
assistance
Huttner,
the
M21
of
deaths
for
each
stage
(4) sum the observed
within
each age-sex
summed
observed
subgroup
for
the
and expected
numbers
combination
and then
deaths,
and
by
E,
the
i-th
of
over
summed
table,
e.g.,
deaths
over
all age-sex
expected
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
234
RAI
deaths),
and
(5) compute
the
logrank
test statistic
-
where
with
the
summation
four
degrees
the number
is over
of subgroups
difference
all
of freedom
in survival
stages;
(the
being
among
ET AL.
this
statistic
of
freedom
for
each
other)
degrees
compared
E)2/E
with
is approximately
the
distributed
chi-square
under
the
as chi-square
is always
null
one
less
than
hypothesis-there
is no
stages.
REFERENCES
I.
Dameshek
kemia-an
ically
W:
Chronic
accumulative
incompetent
lymphocytic
disease
leu-
of immunolog-
lymphocytes.
Blood
29:566,
1967
2.
Boggs
of
phocytic
DR.
Sofferman
GE:
Factors
survival
Survival
lowed
for
(Suppl)
J Med
MM,
the
chronic
40:243,
lymphocytic
lym-
1966
WJ, Lum
D:
leukemia.
microscopy,
in Tocantins
Chronic
studies
a
long
lymphocytic
based
time.
on
189
Methods
(ed
14).
the blood.
Semin
8.
Kaplan
Scand
J
fol-
Haematol
Assoc
9.
Henry
Diagnosis
Philadelphia,
JB
(eds):
by
Saunders,
1969,
and
Mantel
two
consideration.
1966
new
1964
Cronkite
EP,
H,
Extracorporeal
irradiation
of
Hematol
1966
3:154,
Meier
P:
Nonparametric
es-
observations.
J Am
1958
Pike
data
29:579,
(eds):
Throm-
Schnappauf
MC:
(0
survival
10.
Todd-
Laboratory
R,
LA
and
DD,
53:457,
Peto
Kazal
AD,
incomplete
approximation
for
Joel
EL,
from
metrics
I,
Clinical
PA:
timation
18, 1973
Davidsohn
Stryckmans
phase
& Stratton,
Chanana
and
by
LM,
Grune
LM,
ML,
Morphology
platelets
Hemorhage
York,
Greenberg
leucases
New
7. Schiffer
EP:
blood
Coagulation,
bosis.
Stat
MM:
G, Cronkite
of human
dura-
1973
Sanford
p 158
with
chronic
Hansen
Clinical
patients
Am
Wintrobe
Si, Reeves
42:367,
kemia.
SA,
influencing
C, Cutler
in
Blood,
5.
of
leukemia.
3. Zippin
4.
Brecher
Blood
Cartwright
tion
6.
enumeration
-
Conservatism
E)2/E
or tumor
of
the
in the
logrank
test
incidence
data.
Bio-
1973
N:
rank
Evaluation
order
Cancer
of
statistics
Chemother
survival
data
arising
in
Rep
50:163,
its
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1975 46: 219-234
Clinical staging of chronic lymphocytic leukemia
KR Rai, A Sawitsky, EP Cronkite, AD Chanana, RN Levy and BS Pasternack
Updated information and services can be found at:
http://www.bloodjournal.org/content/46/2/219.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of
Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.