Document e5Zg9ypNBELVOGRQNEJo5KmY9
Oxymetholone
Treatment for the Anemia
of Bone Marrow Failure
By Raymond Alexanian, Judith Nadell, and Clarence Alfrey
Oxymetholone
was given to 28 adults
with chronic anemia from bone mar-
row disease. Changes in hematocrit
and red cell mass were correlated with
serial assessments of erythropoietin
and erythropoiesis.
Erythropoietin
ex-
cretion was enhanced more than five-
fold over the level expected for the
hematocrit in 70#{176}o/of the patients. Only
23#{176}o/fo the patients with an evaluable
treatment trial increased their red cell
mass by at least 20#{176}/Ion. all respond-
ers, the T#{189o} f 59Fe disappearance
ranged from 86-136 mm and erythron
iron turnover exceeded 0.25 mg/100
ml blood/day. A decline in serum iron
concentration
to the 50-100 lLg/100 ml
range after 1 mo of oxymetholone
was
frequently associated with a subse-
quent response to therapy. Patients
with severe bone marrow failure, for
whom frequent red cell transfusions
were required, did not improve. The
failure of other patients to respond
was attributed to complicating
factors
that either impaired maximal erythro-
poietin production
or restricted
iron
supply to the bone marrow. Hepatic
toxicity was detected in less than
10#{176o}f/o treated patients. Results sup-
port the use of oxymetholone
in the
treatment of patients with moderate
degrees of bone marrow failure and
symptomatic anemia.
S TUDIES
IN ANIMALS
AND MAN
hormones
stimulate
erythropoiesis,'
increased
production
of erythropoiefinfr7
have documented an effect probably Even when basal
that androgenic mediated by an levels of erythro-
poietin were elevated,
as in most anemias
resulting
from bone marrow
disease,
further increments
usually followed
androgenic
hormone
therapy.89
Never-
theless, red cell mass improved
infrequently,
presumably
because
marrow was unable to respond.3'9
Oxymetholone
is an oral androgen
considered
to be effective
the bone in many
patients with chronic, refractory
anemia.112
The capacity of oxymetholone
to stimulate erythropoietin
production,
and the clinical usefulness
of this
hormone in adults with anemia due to chronic marrow disease, has not been
determined. These questions were clarified by serial erythropoietin
and erythro-
kinetic studies before and after giving oxymetholone
to 28 anemic adults with
bone marrow failure. Results indicated a high frequency and degree of erythro-
poietin stimulation,
but a low incidence of bone marrow improvement.
Those
patients most likely to benefit were those with a greater magnitude
of marrow
proliferation,
with a marked degree of erythropoietin
stimulation,
and with an
adequate iron supply.
From the Departments
of Medicine,
The University
of Texas M. D. Anderson
Hospital
and Tumor institute, and Baylor College of Medicine, Houston, Texas.
Submitted
October 6, 1971; first revision March 1, 1972; second revision April 18, 1972;
accepted April 27, 1972.
Supported
by USPHS Grants AM 09155 and HE 05435
Raymond Alexanian,
M.D. : Associate Professor of Medicine,
The University
of Texas
M. D. Anderson Hospital and Tumor Institute, Houston, Texas. Judith Nadell, M.D. : As-
sociate Medical Director, Syntex Research, Palo Alto, Calif. Clarence Alfrey, M.D.: Asso-
ciate Professor of Medicine, Baylor College of Medicine, Houston, Texas.
Blood, Vol. 40, No. 3 (September), 1972
353
354 ALEXANIAN, NADELL, AND ALFAEY
Table 1. Aesponse of 28 Patients to Oxymetholone
Idiopathic marrow failure Aplastic
Normal and hyperplastic Myelofibrosis Lymphoma My#{232}loma
No. Patients
7 7 3 4 7
No. Evaluable Trial
6 7 3 4 6
No. Responsive
0 2 0 1 3
Total
28 26
6
S Response was defined by an increase in red cell mass of > 20#{176w}/ioth elimination of transfusion requirements. In all responders, the hematocrit also increased by at least 5 vol/100 ml.
MATERIALS AND METHODS
Twenty-eight
patients with chronic refractory
anemia from bone marrow failure were
studied before and during treatment with oxymetholone.
The median age was 67, ranging
from 20 to 78; 18 of our patients were male. Fourteen patients had idiopathic bone marrow
failure; three had myelofibrosis;
four had lymphoma
with no marrow infiltration;
and
seven had multiple myeloma in a remission status (Table 1). Bone marrow hemosiderin
was present in all, and no patient had marked marrow infiltration
by malignant
cells
(i.e., > 20%). Patients with a significant hemolytic component,
or a serum iron concen-
tration less than 60 g/100 ml before treatment were excluded. Six patients were included
with normal pretreatment serum iron values, even though postoxymetholone
values fell
later to less than 60 zg/100 ml. The creatinine
clearance
exceeded 50 mI/mm in all but
two patients whose clearances
were between
35-50 mI/mm ; these two patients
were
included only after urinary erythropoietln
excretion was found to be appropriate
for the
degree of anemia. Serum bilirubin, alkaline phosphatase,
and transaminase
were measured
before treatment and at monthly intervals thereafter. Red cell transfusions
(i.e., two on
more per mo) were given to 14 patients when the hematocnit was less than 20 vol/i00 ml.
No patient received chemotherapy
for malignant
disease on corticosteroids
during the
treatment
trial.
Erythropoiesis
was evaluated from measurements
of plasma iron turnover, erythnon iron
turnover, and red cell mass. Plasma iron turnover was measured using 59Fe after the intra-
venous injection of 5-10 iCi of citnated 59Fe previously incubated in normal plasma. Blood
was collected at measured time intervals for 3 hr, the T#{188}of plasma clearance was deter-
mined, and plasma iron turnover (mg/100 ml blood/day)
was calculated.13
Erythnocytic
iron turnover (EIT) was derived by subtracting
the calculated nonerythron
turnover (serum
iron X plasmatocrit
X 0.0035) from the measured
plasma iron turnover.'4
Fixed red
blood cell iron turnover (FRIT) was derived from the product of the plasma iron turnover and the per cent utilization of 59Fe in the measured red cell mass. No patient with signifi-
cant hemolytic
anemia, defined by a FRIT (mg/be
ml blood/day)
more than 2% of the
red cell iron content (mg) in 100 ml blood, was included in this study. Red cell iron con-
tent was derived by assuming that 1 ml of red cells contained 1 mg of iron. Using a FRIT/ red cell iron index for those anemic patients not requiring red cell transfusions, the median
rate of hemolysis was 1.5%/day.
Five patients were considered
to have significant
ineffec-
tive enythnopoiesis
when total enythropoiesis
derived from the enythnon iron turnover'4
(plasma iron turnover minus nonerythron
turnover) exceeded, by more than 0.60 mg/100
ml blood/day,
the rate of effective enythropoiesis
derived from the fixed red cell iron
turnover. This level of ineffective enythropoiesis
was 1#{188t}imes the total red cell production
in normal man.'5 A linear body scanner was used to define the distribution of 59Fe in the
body 1 day after the injection of 59Fe.18 The fraction of administered
isotope found over
the pelvis and upper abdomen was calculated for each patient. Erythropoietin
production
ANEMIA OF BONE MARROW FAILURE
355
was determined
from the urinary excretion by assay of urine on urine concentrates
In
polycythemic
mice and expressed in standard B units/day
as previously
described.7
Since
erythropoietin
excretion
reached
maximal erythropoietin
production
Oxymetholone
(Syntex Research)
maximum
levels in man 1 mo after oral andnogens,9
was evaluated after this duration of therapy.
was initiated in a daily dose of 1.0 mg/kg for women
and 5.0 mg/kg for men. The dose was increased in 50% increments
at monthly intervals
if side effects were not distressing
(muscle cramps, virilization,
nausea), but not to more
than 100 mg/day for women or 450 mg/day for men. An evaluable treatment trial required
at least 3 mo of therapy with a minimum total dose of 5.0 g of oxymetholone.
The red cell
mass was measured with 51Cr-tagged
red cells before and after at least 3 mo of treatment;
clinical response was defined as an increase of at least 20% with elimination
of ned cell
transfusion
requirements.
Table 2. Hematologic Status of 26 Evaluable Patients Treated With Oxymetholone
Diagnnsis
Hct (Vol/ 100 ml)
WBC Gran Platelets ( x 10' cu mm)
Bone Marrow
cellularityf
cytologyf
Responsive patients
D.H. MM
32
2.6 1.6 124 N; 20#{176}/oRBC; 10#{17P6C}/o
R.C.t
L 26
2.4 1.3
44 N;20#{176A}B/Co
A.W. IRA 295
4.8 1.0
87 N; 40#{176}A/oBC
Z.G.t MM 185
5.1 4.2 276 N; 30#{176}A/aBC; 20#{176}P/oC
ES. MM 34
6.0 3.1 177 N; 30#{17R6B}C/;a 20#{17P6C}/o
B.T. IRA 26
2.3 0.7
92 N;40#{176}A/oBC
Unresponsive patients
Decreased marrow
M.R.
AA 195
2.4 0.7
70 0; <5#{176}/o ABC
C.D. MF 215
4.6 1.7 554
I; 90#{176}/ofibrosis
B.M. AA 275
1.8 0.7
45 0; 100/0 ABC
L.C.
AA 315
3.3 1.4 105 N; <5#{176}/oABC
W.P. MF 255
1.1 0.9 308
I; 90#{176}/ofibrosis
J.M.
AA 185
2.6 0.6
24 0; <5#{176}/o ABC
T.S.
AA 305
5.7 3.2 434
N; <5#{176}/o RBC
L.L.
AA 295
2.2 1.0
34 D; <5#{176}/o ABC
Normal and Increased marrow
M.B. MM 19
3.6 2.0
33 N : 20#{17A6B}C/;a 20#{17P6C}/o
F.G. IRA 225 0.9 0.5
62 N; 20#{17A6B}C/a
E.T.
L 33
3.3 1.6 171 N; 20#{17R6B}C/a
E.P. IRA 38
4.7 0.9 265 N; 40#{176A}B/aC
H.C.
L 39
28.8 3.5
86 N; 20#{17A6B}C/a
N.G. MM
31
3.3 2.6 183 N: 20#{176A}/BaC; 20#{176P}/Ca
J.G. MM 35
5.0 3.9 211 N; 20#{17R6B}C/;a 20#{17P6C}/o
R.M.
L 36
3.1 1.9 162
I:60#{17A6B}C/a
J.c. MF 305 3.4 1.2 40 I;50#{176}/aABC
FL.
IRA 315
2.6 1.1 245
I: 50#{176}/aRBC
S.C. IRA 32
8.2 5.0 364 I; 30#{176}/aABC
W.O. IRA 34
2.1 1.3 211
I: 40#{176}/aABC
a AA, aplastic anemia; IRA, idiopathic
refractory
anemia; MF, myelofibrosis;
multiple myeloma; L, lymphoma.
t Marrow cellularity
on clot section. N, normal; I, increased;
0, decreased;
erythroid cells; PC, plasma cells; WBC, white blood cells: Gran, granulocytes.
t Creatinine clearance 35-50 mI/mm.
S Required red cell transfusions.
MM, ABC,
356
ALEXANIAN, NADELL, AND ALFREY
2 % .el
-2 #{149}<
01
-cx: IC-'). 0C') cc LCIJ)
CsJ
>< ><
Cl . .?
SI
E"
-> ->
ol
0 tO
Q.
0.
I -2 I
Cd?)
0 csJcf)
0 N.Lt)
CsJ OJ
C'J OJ OJ
I 0.
(C0O `-
-`L- () N`` - `J' CtO') NLO- `-
0` )
CJ CJ . `- `- `- I-
i:l `
cJ L() L() U)
N.0)i-C')0)'-O)
"
0 C
0
.I
: 0
EI eI 0 l>
:
;
>01100
V El
4) i I-l'
,! F
U,
-0 V
.)
-V
U. d
tq d?
,,
0 C
2 -. ,2.
a
,_ ,.
Co csJ C') `: `
t..C')
#{228}ciodod#{228}
I- tO 0 . N. 0 0) CDCD')U) oocjdQoo
. ,I! I
CJ
0. i e . `E #{176}
I
.
(I)
.!
V
I
IC,)
.
0
1I
C 2l'
: -I
2 EI
CD
aa II
LI. lc'i I
I
IF-
CD CigO Cit)D 0CO I-.. CD
:c`, e II
I-
U)
-C')
C'J 0 0 0
Lt) CO F0 `- 0
dooddddd
N. `
CO C')
dddoddodci
CO CO CD
0 CD 0 0 0 0) 0 CD C')
- ycJ (%J (%4 %J C,J C'J CSJ
0C0%1J 0L- U C`- 'J `C- D C(0') 0U`- ) CN#{D.182}- LOt)J 0LC))
C')0)LOOLOC4CD0)C')
-CtosJ `i CD CC'%J J 0) C`- D I1-- "CtJ 0(`J
oo
C'J CD 0) C')CSJC'4
CD CD
c'c-J
.U 0) `
N. -
11) CD 0 0) 0 C')C'JC'J
V
.
(1)0 00. 0)
C)
` ao<rL1 `-U)
00
C!)
vi)
>0
IC 0
(I) 0
ci :2 d ui:2#{176}-
> C
i
ANEMIA OF BONE MARROW FAILURE
CD N. 0 CD cJcJc')cJ
N. 0 N. CO 0) 0) CJ-cjcsj
C') 0) c')c'
0) C') C') 0 0 0 0 0 1 J C') C') CJ C%J C%J OJ
0 ` N. tO N. C%J C') C') OJ (`J C'J
CJ CJ OJ CJ
0)0)N.CDN.c')N. ,. I- CD
CO CD N. CD
N.CDi-CJ0Oj C')
CsJ C') CSJ
CDCD - CD
Co'a)CoJaLqaaao
LCOD-C0%C'JLCtJ)NC'.)C')N.
aaaooo
-CJcCDJ
oo
0 L() C') CD It) 0 Cl).j.Lq od00000o
tr)
LO () `-C')
C') N.
aa
N.CD0c')cj.. tnC0CD0)CDN.CD
0)CDLflO'U) CLOLOLO
gCD
-;: 0 (C)J 0CD C0)) CtoD -N. c0o
csj 00 1IC) `` C` ') `
C') 1C%J-
1 (j--OJ0cOCD0oJ 0) CD 0
`- cvi
`-
N. CO 0)
--C'4C'),-
-0) CD C'J N. CD
cJ I-
CJ
`.L()
,jC') CD
C
.2?
?j
.`i;;-
o0
0.
0
2
0
E
d-i(icocD
z
- g,,C) C') (C0') .4-
oU2)
`-
0 > CV.?
.!t
CCO')
U0 C')
U
C- ')
CCJ') C')
C-C4-'S-) J
Co
.i; .2?
0
20.
.C a,
.
a, >.C
CO
.2?
#(c1{s1)j49} J0)
I:2
.Ez:
w
0.
Va)
20(1)
.-
C
(1)_
0 =C'J
0oV
`-I VU)
O'U)
357
358 ALEXANIAN, NADELL, AND ALFREY
0
0
`I, C
z 0
ILU C-, >< Li
z
ILi
0 I00
0. 0
N.
>a- :
Li
>a- :
.cz
z a:
:D
0
0 0
0
\ \ \
.`
0
.
\
\0
\ \
.@
S
0
0
0
0
.
\
10 20 30 40 HEMATOCRIT (Vol.%)
Fig. 1. Relationship
of
urinary
erythropoietin
excretion to hematocrit.
Studies
before andro-
gens (solid circles) are
compared
with studies
after 1 mo of treatment
(open circles) in 23 patients with evaluable ery-
thropoietin
assays. The
solid line was calculated
from basal measure-
ments and the dashed
line from studies after 1
50 mo of treatment. Values
for responsive
patients
are circled.
RESULTS
Erythropoietin
and erythropoiesis
were assessed before and during therapy
in the 26 patients with an evaluable treatment trial of at least 3 mo. Only six
patients (23%) increased
their red cell mass by more than 20% , and five were
women. In these responders,
the median increment in hematocrit
was 9 vol/100
ml (range 5 to 20), and in red cell mass 260 ml/sq m (range, 130-380).
This
elevation was recognized
within 3 mo of treatment,
and was equivalent
to
about 25% of the red cell mass in normal man.15 Only two of the 13 evaluable
patients
who had been dependent
on red cell transfusions
eliminated
this
requirement
after oxymetholone
(Table 2). Within
2 mo after the cessation
of
androgen treatment,
the hematocrit
declined by at least 5 vol/100 ml in all
responding
patients.
Erythropoietin
Oxymetholone
produced an increment in urinary erythropoietin
excretion
of at least five times the value expected for the hematocrit
in 16 of the 23
patients (70%) with evaluable studies before and after treatment.
Seven pa-
tients showed less than a fivefold enhancement
and all were unresponsive
to
ANEMIA OF BONE MARROW FAILUAE
359
oxymetholone.
Figure 1 compares urinary erythropoietin
values after 1 mo of
treatment with control measurements
in relation to the hematocrit.
Erythrokinetics
The clinical response to oxymetholone
was correlated with the degree of
bone marrow activity in each patient (Tables 2 and 3). In responding
patients,
bone marrow erythroid cellularity was normal, the T#{189}for 59Fe disappearance
ranged between 86-136 mm, and the erythron iron turnover exceeded 0.25
mg/100 ml blood/day
(normal range 0.30-0.50).
The serum iron concentration
was between 100-170 g/100 ml before treatment in all but one; within 1 mo
after institution of oxymetholone,
the serum iron declined by at least 40 g/100
ml, but not to a concentration
less than 50 jg/100
ml (Table 3). Figure 2 con-
trasts the relationship
between erythropoietin
and erythropoiesis
in the six
responding
patients with identical studies in the 20 unresponsive
patients.
IOO(
U
U
U
0
0 UU U
=,
0U
FLi
U
a: U
C-)
><
Li
zU
FLi
0 0 0 a: :r F-
>a-:
Li
>-
a:
a:
:D
U
IJ U
U
U
0.5
0.4 0.8 .2 .6
ERYTHRON IRON TURNOVER (mg/IOOmI Blood/Day)
Fig. 2. Relationship
between
erythropoietmn
excretion
and
turnover in 26 patients with an evaluable oxymetholone
treatment
cates values for responsive patients (all of whom fell within the
and U for patients unresponsive
to oxymetholone.
The stippled
the range for normal man and the solid line was calculated
studies before and/or after bleeding 13 normal volunteers.'5
erythron
iron
trial. A mdi-
dashed lines),
area includes
from previous
360 ALEXANIAN, NADELL, AND ALFREY
Patients
failing to improve
red cell production
with oxymetholone
were
analyzed in three different groups (Table 3). In one group (8 patients),
bone
marrow
function
was depressed
and erythron
iron turnover
0.25 mg/100
ml blood/day
(median 0.16). All were markedly
was less than anemic and de-
pendent on frequent red cell transfusions
(two or more units/mo).
Erythroid
marrow cellularity
was depressed
and some patients also had severe granulo-
cytopenia
and thrombocytopenia
(Table 2). Serum iron levels were high
(median 203 tg/100
ml) and usually remained
elevated
after oxymetholone.
The T1/2 of 59Fe disappearance
was markedly
prolonged
in all (median
mm) (Table 3). Although basal erythropoietin
excretion was appropriately
223 high
for the degree of anemia oxymetholone
produced
further marked elevations
to more than 500 units/day
in all.
The second group of unresponsive
patients
(7 patients)
had a calculated
erythron iron turnover within the normal range. Anemia was less severe, mar-
row erythroid
cellularity
was normal, and there was no evidence of significant
ineffective
erythropoiesis.
The third group of unresponsive
patients
(5 pa-
tients) showed an erythron iron turnover greater than twice normal, increased
marrow erythroid
cellularity,
and significant
ineffective
erythropoiesis
(Table
3). Two of the four patients
in this group who had bone marrow
culture
studies showed major cytogenetic
abnormalities
(courtesy
of Dr. J. Trujillo).
While the maximum
erythropoietin
excretion
after 1 mo of oxymetholone
was stimulated
to more than 5 times control to more than 40 units/day
(i.e.,
> 20 times normal) in all "responsive"
patients with evaluable
studies, this
occurred
in only 6 of 13 "unresponsive"
patients
with normal or increased
marrow function (Fig. 1). Four of the seven patients with "impaired"
erythro-
poietin stimulation
also had a marked fall in serum iron concentration
to less
than 30 zg/100 ml (Table 3). Serum iron values remained
low, in association
with an iron binding capacity less than 300 ,g/100
ml, despite oral iron sup-
plements
for at least 2 mo. Only two of these four patients had overt clinical
evidence of infection (pneumonia
and urinary tract infection).
When the analy-
sis was limited to the eleven patients with an erythron
iron turnover
greater
than 0.25 mg/100
ml blood/day
without ineffective
erythropoiesis,
in whom
the serum iron exceeded 50 g/100
ml after 1 mo of treatment,
six showed a
significant
elevation in red cell volume after oxymetholone
(Table 3).
Organ Scanning
Linear body scanning
of 59Fe distribution
demonstrated
other differences
between
patients
responsive
and unresponsive
to oxymetholone.
Figure 3 cor-
relates 59Fe uptake in the pelvis and upper abdomen with different levels of
red cell production.
Only patients with serum iron levels above 60 tg/100 ml
and without irradiation
therapy to the pelvis were included in this analysis.
Subjects with normal marrow function had a radioiron
uptake in the pelvis
of at least 20% of the injected dose (median 28%). The per cent uptake re-
mained
constant
with increased
red cell production,
but declined with sub-
normal
levels of erythropoiesis
(i.e., EIT < 0.30 mg/100
ml blood/day).
Radioiron
uptake in the upper abdomen was less than 40% of the injected
ANEMIA OF BONE MARROW FAILURE
361
dose with normal and increased erythropoiesis,
but increased with declining
red cell production
(Fig. 3). None of the eight patients with markedly de-
pressed radioiron
uptake in the pelvis (< 18% of injected dose) or increased
uptake in the upper abdomen
(> 40% of injected dose) responded
to oxy-
metholone
(Table 3). In contrast,
four of nine patients
without
ineffective
40
0
(1)
>
-J
LaU-
30
LU
Fa--
D
a, 20
a) U-
I0
U
:
U UU U
R R #{149}o
. U#{149}
0
CU
0 0
0
LU
60
cr
LU
a-
a-
40
LU Fa-
20
a,
`C,
a) U-
U 0
U U
0
U
0 o
0 0.4 0.8 .2 6 2.0
ERYTHRON IRON TURNOVER (mg/IOOmI Blood/Day)
Fig. 3. Relationship
between pelvis
and erythron iron turnover. R indicates
20 unresponsive
patients with evaluable
cate other anemic patients not treated
values for normal man before and after
secondary
polycythemia.
Boxes include
and upper abdominal
uptake of 59Fe
four responsive patients, and U shows
scanning studies. Solid circles mdi-
with androgens.
Open circles show
phlebotomy,
and for four patients with
ranges for normal man.
362 ALEXANIAN, NADELL, AND ALFREY
erythropoiesis (> 19%) oxymetholone
and with an approximately
normal 59Fe distribution
and upper abdomen
(< 35%) elevated
their red
(Fig. 3).
in the pelvis cell mass after
Clinical Effects
The doses of oxymetholone
used in this study were well tolerated
in most
patients. The minimum
3 mo treatment
trial was not completed
by two patients
because of death from progressive
bone marrow disease (myeloma
and aplastic
anemia). Side effects from oxymetholone
were frequent, but mild and re-
versible. Most patients complained
of intermittent
cramping
muscle pain, but
this never interfered
with normal activities.
Four women showed slight viriliz-
ing changes. Mild fluid retention developed
in ten patients, four of whom had
preexisting
heart disease,
but symptoms
abated when the daily dose was
reduced or an oral diuretic was added. No patient received less than 50 mg/day
of oxymetholone.
Two patients developed
hepatic insufficiency
as their treat-
ment trial was completed
(total doses 9 and 30 g). In one, the bilirubin
in-
creased
(normal
to 17 mg/100 ml, the alkaline phosphatase <3 units), and the serum glutamic-oxaloacetic
to 15 Bessey-Lowrv transaminase
units (SCOT)
to 110 mU/mi.
In the second patient, the bilirubin
increased
to 2.7 mg/100
ml, the SCOT to 155 mU/ml, while the alkaline phosphatase
remained normal.
Values returned
to normal in both within 2 mo after the cessation
of oxy-
metholone.
DISCUSSION
This study was designed
to evaluate
the utility of oxymetholone
in the
treatment
of adults with chronic refractory
anemia resulting
from bone mar-
row failure. Emphasis was placed on a systematic
evaluation
before and during
therapy of those factors of major importance
in the maintenance
of normal
erythropoiesis.
These included erythropoietin
production,
iron supply, and the
level of bone marrow activity. In order to provide a clear appraisal of the
effect of androgens
on patients with bone marrow disease, patients with renal
disease, hemolytic
anemia, or extensive
marrow infiltration
by malignant
cells
were excluded from this study. Erythropoietin
production
was evaluated
from
measurements
of the daily urinary excretion.
Previous
studies had ruled out
the likelihood
of an indirect stimulation
of endogenous
erythropoiesis
in assay
animals by androgens.7
Iron supply to the bone marrow was assessed from
the serum iron concentration,
and was considered
depressed
when the level
fell to less than 60 ,g/100 ml.'3 Because assessments
of erythropoiesis
from
measurements
of iron turnover
are probably
invalid with iron deficiency,'7
patients with serum iron values less than 60 ig/100 ml before treatment
were
considered
ineligible
for this study. The body distribution
of radioiron
was
evaluated
with a linear body scanner, and primary emphasis
was placed on
the per cent uptake over the pelvis and upper abdomen.
Pelvis uptake of 59Fe
provided
an index of red cell production
in a major bone marrow segment;
upper abdominal
uptake reflected
the incorporation
of radioiron
in liver
parenchymal
cells.1#{176}
ANEMIA OF BONE MARROW FAILURE
363
Oxymetholone
was used because of the convenience
of oral administration,
the high frequency
of clinical benefit reported by others,'#{176}'2 and our desire
to evaluate the capacity of this drug to stimulate
erythropoietin
production
in man. Maximal doses were given in order to achieve maximal erythropoietin
stimulation
and a more adequate
opportunity
for enhanced
bone marrow ac-
tivity. Thus, only treatment
trials of more than 3 mo duration
with a total
dose of at least 5.0 g were considered
evaluable.
Despite the high doses used,
side effects from oxymetholone
were usually mild and reversible.
Marked ele-
vations in urinary erythropoietin
to more than five times the value expected
for the hematocrit
were produced
in 70% of the patients with evaluable
stud-
ies. This degree of urinary erythropoietin
stimulation
after oxymetholone
was
comparable
to previous studies using fluoxymesterone.9
When erythropoietin
values failed to increase markedly,
red cell production
did not improve despite
an "adequate
level" of bone marrow activity. These observations
support the
conclusion
that marked elevations
in erythropoietin
constitute
the principal
mechanism
for the occasional
efficacy of androgens
in patients with anemia
due to bone marrow disease. Whether
oxymetholone
also affects red cell pro-
duction by other mechanisms,
such as by a direct marrow stimulation
inde-
pendent of erythropoietin,
was not evaluated
in this study.
Significant
elevation
in red cell mass was observed
in about one-fourth
of our patients with an evaluable
patients
had levels of erythropoiesis
treatment
trial. Benefit occurred only when
close to or within the normal range. Even
in these patients, the increment
in red cell production
(i.e., change in red cell
mass) was only about 10% of the elevation (i.e., change in erythron iron turn-
over) induced in normal man following a comparable
degree of erythropoietin
stimulation
by phlebotomy.'5
Improvements
in erythropoiesis
occurred only in
patients with a normal or slightly prolonged
T1/2 of plasma 59Fe disappear-
ance, an almost normal distribution
of radioiron
to the pelvis and liver, and
moderate declines in serum iron concentration
after 1 mo of treatment.
Patients
with more severe depression
of bone marrow function,
who re-
quired frequent red cell transfusions,
did not benefit from oxymetholone.
These
patients
usually
had elevated
serum iron values unaffected
by treatment,
markedly
prolonged
disappearance
half-times
of radioiron,
and diminished
red
cell precursors
in the bone marrow. Decreased
pelvis and increased
abdominal
uptake of radioiron
provided
further evidence
for the severe bone marrow
damage in these patients.
Even when marrow
erythropoiesis
appeared
to be
well preserved,
oxymetholone
failed to increase the red cell mass of many
anemic patients.
Some of these patients
did not show a marked degree of
erythropoietin
stimulation,
while others demonstrated
substantial
rates of
ineffective
erythropoiesis.
In some unresponsive
patients, the serum iron con-
centration
fell with oxymetholone
to very low levels (< 30 ig/100
ml) despite
adequate marrow iron stores and oral iron supplements.
These patients were
considered
to have additional
complications,
such as the "anemia of chronic
disease,"8
that restricted
iron supply to the bone marrow and prevented
an
opportunity
for further enhancement
of erythropoiesis.
The ineffectiveness
of
androgens
in this clinical setting differs from the experience
of Haurani who
364 ALEXANIAN, NADELL, AND ALFAEY
reported
improved
iron reutilization
and increased
red cell mass with testo-
sterone in patients with hypoferremia.'9
Other investigations
have defined changes in the "erythropoietin-sensitive
stem cell population"
from the relationship
between the rates of erythropoiesis
in irradiated
and unirradiated
polycythemic
mice following a standardized
dose
of erythropoietin.20'2'
Using similar assumptions,
the erythropoietin-sensitive
bone marrow capacity of each of our anemic patients was calculated
as a per
cent of normal. This was derived from the ratio between
the erythron
iron
turnover
of each patient and the measured
turnover
of normal man for the
urinary
erythropoietin
excretion
of each anemic patient.
In all patients
re-
sponding
to oxymetholone,
the rate of erythropoiesis
exceeded
20% of the
level found in normal human beings after phlebotomy.
No patient with a more
depressed
erythropoietin-sensitive
marrow
capacity,
in whom frequent
red
cell transfusions
were required,
improved
with oxymetholone.
Thus, those
patients with the greatest need for more red cell production
seldom derived
any benefit. These observations
are in conformance
with those of others who
found that enhanced
red cell production
from androgens
occurred more fre-
quentlv in anemic patients with some evidence of marrow erythropoiesis,
as
in children with idiopathic
refractory
anemia4 and adults with myelofibrosis.22
Therapeutic
trials with oxymetholone
are justified primarily
in those patients
with symptomatic
levels of chronic anemia who have moderate
levels of
residual erythropoiesis
and who sustain an adequate iron supply to the bone
marrow.
REFERENCES
1. Steinglass,
P., Gordon,
A. S., and
Charipper,
H. A. : Effect of castration
and
sex hormones on the blood of rats. Proc.
Soc. Exp. Biol. Med. 48 :169, 1941.
2. Kennedy, B. J., and Gilbertson,
A. S.:
Increased erythropoiesis
genic-hormone
therapy.
induced by andro-
New Eng. J. Med.
256:719,
1957.
3. Gardner,
F. H., and Pningle, J. C.:
Androgens
and enythropoiesis.
Arch. Intern.
Med. (Chicago) 107:112, 1961.
4. Shahidi, N. T., and Diamond,
L. K.:
Testosterone-induced
remission
in aplastic
anemia
of both acquired
and congenital
types. New Eng. J. Med. 264 :953, 1961.
5. Mirand,
E. A., Gordon, A. S., and
Wenig, J. : Mechanism
of testosterone
ac-
tion in erythnopoiesis.
Nature (London) 206:
270, 1965.
6. Fried, W., and Gurney, C. W. : Erythro-
poietin effect of plasma from mice receiving
testosterone.
Nature
(London)
206:1170,
1965.
7. Alexanian,
R., Vaughn,
W. K., and
Ruchelman,
M. W. : Erythropoietin
excretion
in man following androgens.
J. Lab. Clin.
Med. 70:777, 1967.
8. Rishpon-Meyerstein,
N., Kilbridge, T.,
Simone, J., and Fried, W. : The effect of
testosterone
on erythropoietin
levels in
anemic patients.
Blood 31 :453, 1968.
9. Alexanian,
R. : Erythropoietin
and ery-
thropoiesis
in anemic man following andro-
gens. Blood 33:564, 1969.
10. Sanchez-Medal,
L., Gomez-Leal,
A.,
Duante, L., and Rico, M. C. : Anabolic
androgenic
steroids in the treatment of ac-
quired aplastic anemia. Blood 34 :283, 1969.
11. AlIen, D. M., Fine, M. H., Necheles,
T. F., and Dameshek,
W. : Oxymetholone
therapy in aplastic anemia. Blood 32:83,
1968. 12. Silink,
S. J., and Firkin,
B. C. : An
analysis of hypoplastic
anemia with special
reference to the use of oxymetholone
in its
therapy. Aust. Ann. Med. 17:224, 1968.
13. Finch, C. A., et al. : Ferrokinetics
in
man. Medicine 49 :17, 1970.
14. Cook, J. D., Marsaglia,
C., Eschbach,
J. w., Funk, D. D., and Finch, C. A. : Ferno-
ANEMIA OF BONE MARROW FAILURE
365
kinetics : A biologic model for plasma iron
exchange in man. J. Clin. Invest. 49:197,
1970.
15. Alexanian,
R., and Alfrey, C. : Ery-
thropoiesis in the anemia of bone marrow
failure. J. Clin. Invest. 49 :1986, 1970.
16. Alfrey, C. P., Jr., Lynch, E. C., and
Hettig, R. A. : Studies of Iron kinetics using
a linear scanner. J. Lab. Cliii. Med. 73:405,
1969.
17. Hillman, R. S., and Henderson,
P. A.:
Control
of marrow
production
by the level
of iron supply. J. Clin. Invest. 48 :454, 1969.
18. Cartwnight,
C. E. : The anemia of
chronic disorders. Seminars Hemat. 3:351,
1966.
19. Haurani, F. I., and Green, D. : Primary
defective iron reutilization,
Amer. J. Med.
42:151, 1967.
20. Gurney,
C.W., Lajtha, L.G., and
Oliver, R. : A method for investigation
of
stem-cell kinetics. Bnit. J. Haemat. 8:461,
1962.
21. Alexanian,
R., Ponteous, D. D., and
Lajtha, L. C. : Stem cell kinetics after ir-
radiation. Tnt. J. Radiat. Biol. 7:87, 1963. 22. Gardner, F. H., and Nathan, D. C.:
Androgens
and erythropoiesis.
J. Med. 274:420, 1966.
New Eng.