[CANCER RESEARCH 34. 1952-1956, August 1974]
Mechanisms of Resistance to 6-Thiopurines in Human Leukemia1
Martin Rosman,
Men Hui Lee,3 William A. Creasey, and Alan C. Sartorelli
Departments of Medicine [M. K., W. A. C.] and Pharmacology [M. H. L., W. A. C., A. C. S.], Yale University School of Medicine
New Haven, Connecticut 06510
SUMMARY
The activities of enzymes involved in the formation and
catabolism of 6-thiopurine nucleotide were assayed in
leukocytes of leukemic patients considered either suscepti
ble or resistant to the 6-thiopurines. Altered ratios of the
activities of hypoxanthine-guanine phosphoribosyltransferase (H-GPRT) and adenine phosphoribosyltransferase
were noted in leukocytes from 3 of 11 resistant acute
nonlymphocytic leukemic (ANLL) patients, whereas nor
mal ratios were found in all 7 resistant acute lymphocytic
leukemic patients assayed. Leukocytes from one of the re
sistant ANLL patients had a marked decrease in H-GPRT
activity, while the remainder were relatively deficient, as
defined by an adenine phosphoribosyltransferase/H-GPRT
(A/H) ratio greater than the average ratio from leuko
cytes of untreated patients plus 3 standard deviation units.
Particulate-bound alkaline phosphatase was significantly
elevated in leukocytes from 2 of 11 resistant ANLL and 6
of 7 resistant acute lymphocytic leukemia patients. Little
correlation between acid phosphatase levels of leukemic
leukocytes and clinical status was noted. Increased alkaline
phosphatase activity tended to occur in resistant patients
with normal adenine phosphoribosyltransferase/H-GPRT
ratios. The findings suggested that either a decrease in
H-GPRT or an increase in alkaline phosphatase activities
was responsible, at least in part, for insensitivity to 6thiopurines in 5 of 11 ANLL and 6 of 7 acute lymphocytic
leukemia patients.
INTRODUCTION
Although the phenomenon of resistance to purine analogs
has been under intensive investigation for over a decade, the
mechanisms by which human tumors become unresponsive
to the 6-thiopurines remain unclear (4). In nonhuman
systems, the most frequent cause of acquired insensitivity to
purine analogs has been the loss or marked decrease of the
activity of the anabolic enzyme H-GPRT4 which is required
•¿This
work was supported by Grants CA-02817. CA-08341, and
CA-12317 from the National Cancer Institute, USPHS, and ACS DI IOH
from the American Cancer Society. Presented in part at the Annual
Meeting of the American Society of Hematology, Chicago, 1973.
2Special fellow of the Leukemia Society of America. To whom reprint
requests should be addressed at the Yale University School of Medicine,
Department of Pharmacology, New Haven, Conn. 06510.
3Special fellow of the Leukemia Society of America.
' The abbreviations used are: H-GPRT, hypoxanthine-guanine phosReceived February 6, 1974; accepted April 19, 1974.
1952
to convert these antimetabolites to the nucleotide form, a
process necessary for tumor inhibition. However, enzymatic
studies designed to test this possibility in man have indicated
that this mechanism is not operative in most human
leukemias (5, 14). In contrast, Kessel and Hall (8) reported
that leukemic leukocytes isolated from patients sensitive to
6-MP retained a significantly greater amount of radioactiv
ity from 6-MP to which they were exposed than did
leukocytes from resistant patients. While these findings
might be interpreted as being due to greater anabolic
enzyme activity and intracellular retention of 6-thiopurine
nucleotides by sensitive cells, an increase in the activity of
analog nucleotide degrading enzymes in the resistant cells
might also have accounted for the decreased retention.
Resistance to several purine analogs has been ascribed to
increased degradation of the active inhibitory analog nu
cleotide form (2, 3, 7, 19). Furthermore, Wolpert et al. (19)
provided evidence that an increase in catabolic activity
associated with resistance to 6-TG by a transplanted murine
neoplasm was due to a particulate-bound alkaline phospha
tase. This investigation represents an effort to expand upon
previous work on the anabolic enzyme profile of leukemic
leukocytes, and also includes a study of the phosphatase
activities within these cells, the goal being a characterization
of the role of anabolic and catabolic systems in determining
the sensitivity of leukemic leukocytes of humans to 6-thi
opurines.
MATERIALS
AND METHODS
Adenine-8-14C (4.05 mCi/mmole), hypoxanthine-8-14C
(5.32 mCi/mmole), and 6-MP-8-14C hydrate (2.48 mCi/
mmole) were purchased from New England Nuclear, Bos
ton, Mass. PRPP tetrasodium salt (Lot 42C 8631-9) and
unlabeled purines were purchased from Sigma Chemical
Company, St. Louis, Mo. Cellex D (anión exchange
cellulose) was obtained from Bio-Rad Laboratories, Rich
mond, Calif. Dextron lOOc(M.W. 135,000) was a product
of Sigma Chemical Company, and Aquasol was obtained
from New England Nuclear.
Patients at the Yale New Haven Medical Center with the
diagnosis of active acute leukemia were studied. No initial
selection was imposed other than the availability of either
peripheral blood with a significant blast count (all but 2 of
25 samples had a blast content >40%) or the availability of
phoribosyltransferase (EC 2.4.2.8); 6-MP, 6-mercaptopurine; 6-TG, 6-thioguanine; PRPP, 5-phosphoribosyl 1-pyrophosphate; ANLL, acute non
lymphocytic leukemia; APRT, adenine phosphoribosyltransferase (EC
2.4.2.7); A/H, ratio of the specific activity of APRT to H-GPRT.
CANCER RESEARCH
VOL. 34
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1974 American Association for Cancer Research.
Resistance to 6-Thiopurines in Human Leukemia
adequate relapse bone marrow from the leukopenic patient of the enzyme either from the supernatant fraction or from
(5 samples). Of the 30 samples from 29 patients (1 patient the suspension of the sonically disrupted pellet, as described
before and after thiopurine therapy), 11 were collected from above. With the sonically disrupted suspension as the
patients with acute lymphocytic leukemia (4 untreated with enzyme source, the reaction was allowed to continue for at
thiopurines) and 19 with ANLL (8 with ANLL before least 20 min to overcome interference from turbidity, and
thiopurine treatment). Patients were further divided into 2 the linear portion of the slope was taken for calculation of
groups on the basis of drug exposure. Twelve patients enzyme activity. One unit of activity is expressed as
classified as untreated had not received any purine analog hydrolysis of 1 nmole of substrate per min under the assay
but may have been given other chemotherapy. Eighteen conditions, assuming a molar absorbancy of p-nitrophenol
patients, classified as treated, had a history of receiving in the buffers used of 1.7 x 104/mole/liter/cm. Specific ac
either 6-TG or 6-MP alone or in combination with other tivity refers to nmoles PÌ
per min per mg of protein.
agents and had relapsed during or after the regimen.
Leukocytes were prepared free of contaminating erythrocytes as described previously ( 14). The supernatant obtained RESULTS
after centrifugation at 105,000 x g was used for assay of
Anabolic Enzymes. A study of the nucleotide-forming
both H-GPRT and APRT activities. Both the supernatant
fraction and the remaining pellet were used to measure acid capacity of leukemic cells, enlarging upon previously pub
and alkaline phosphatase activities. The pellet was sus lished series (5, 14), was undertaken to determine the
pended in 0.01 MTris-HCl (pH 7.6) and homogenized in an relative contribution of decreased anabolic activity in
leukocytes resistant to the 6-thiopurines. The findings
Omnimixer before phosphatase assays. Protein was deter
mined by the method of Lowry et al. (11). The standard obtained with leukemic leukocytes from purine analogphosphoribosyltransferase assay contained 50 mMTris-HCl treated and untreated groups of patients are illustrated in
(pH 8.0), 5 mM MgCl2, 1 HIMPRPP tetrasodium, and a Table 1 and Chart 1. A wide scatter of values occurred in
radioactive purine base in a total volume of 100n\. The final both untreated and treated categories, and no statistically
concentrations of purine substrates were 0.47 mM hypoxan- significant difference existed between the mean values of the
thine, 0.62 mM adenine, and 0.51 mM 6-MP. The reactions 2 groups; however, it should be noted that there is only 1
patient with a phosphoribosyltransferase activity (6-MP
were initiated by the addition of enzyme (usually 40 to 80 ¿¿g
protein). Control tubes in which PRPP was omitted were used as the substrate) less than 50 nmoles per mg protein per
included routinely. Incubation lasted 30 min at 37°with hr in the group that did not receive a purine analog, while 6
adenine and hypoxanthine and lasted 1 hr with 6-MP. The samples in the treated population had lower enzyme activ
reaction was terminated by the addition of 50 /il of 0.2 M ity. It has been found useful to compare, in cell-free ex
EDTA, and the mixture was kept at 4°until it was subjected tracts from leukocytes of individual patients, the activity of
to chromatography. Separation of the nucleotide products H-GPRT, which anabolizes 6-TG and 6-MP to their active
from the substrates was accomplished with small DEAE nucleotide forms, with that of APRT, a supernatant enzyme
columns as previously described (14). The formation of that converts adenine to adenosine 5'-phosphate, in order to
product was linear over the time interval measured and was assess changes in the relative potential for formation of
linearly proportional to enzyme concentration. Phosphatase purine nucleotides by the cells under study (6). The means of
activities were determined at 25°with a Gilford thermo- the A/H ratio of untreated and treated groups were
stated recording spectrophotometer by measuring the initial equivalent; in particular, the values of the untreated popula
rate of hydrolysis of p-nitrophenyl phosphate, using the tion formed a tight cluster about the mean. The mean for
change in absorbancy at 410 nm. Alkaline phosphatase was the untreated group, 2.71 units, plus 3 S.D. units (arbitrarily
assayed in 0.05 M sodium bicarbonate buffer, pH 9.2, and chosen) was equal to 4.60 units. Three treated ANLL
acid phosphatase activity was determined in 0.05 M sodium patients exceeded this value. The patient with the lowest
acetate buffer, pH 5.8. The reaction mixtures contained 2.8 H-GPRT activity, with either hypoxanthine or 6-MP as the
ml of the appropriate buffer and 10 3 M p-nitrophenyl
substrate, had an A/H value of 101; this individual is a clear
phosphate; the reaction was initiated by addition of 0.2 ml example of H-GPRT deficiency in purine antimetaboliteTable 1
Anabolic and calabolic leukocyte enzyme activities in treated and untreated patients with acule leukemia
Phosphatase activity (nmoles P,/mg protein/min)
Phosphoribosyltransferase activity
No. of
patients12
thine372
purine176
ratio2.71
±513°
±0.27"
untreated
±196
±114
±0.632.71
±4.26
1.36
±0.61
±1.14'Acid3.85
18treatedAdenine1007644 ±392Hypoxan
234 ±141Mercapto1.76 ±2.13Alkaline0.53
0.71 ±0.94Particulate-boundAcidl.84±
91 ±65SupernatantA/H
1.48 ±2.01Alkaline0.15
1.41 ±1.72
" Mean ±S.D.
" Excluding I patient with a particulate-bound alkaline phosphatase level of 12.8 nmoles P, per mg protein per min.
' Excluding I patient with an A/H ratio of 100.9.
AUGUST
1974
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1974 American Association for Cancer Research.
1953
M. Rosman et al.
treated leukemia. As described previously (14), enzymic
activities of leukocytes from all patients with A/H ratios
greater than 4.60 units were evaluated kinetically to deter
mine whether mutant enzymes with altered affinity for the
substrate PRPP could account for the relatively low HGPRT activity of these patients. No enzymatic defects
similar to those reported previously were observed (1, 12,
14).
Catabolic Enzymes. The activities of phosphohydrolase
enzymes that might be expected to affect the cellular
concentration of analog nucleotide are listed in Table 1 and
Chart 2. Phosphatase activity of the supernatant fraction
after centrifugation of cell sonic extracts at 105,000 x g for
1 hr, and of the residual particulate fraction, was measured
at both acid and alkaline pH values. Particulate-bound
alkaline phosphatase was the only measured enzyme that
showed a clear elevation in leukocytes from some of the
treated patients. A breakdown by cell type demonstrated
that all 4 patients with acute lymphocytic leukemia that
were never treated with purine antimetabolites had particu
late alkaline phosphatase activity less than 1 nmole per mg
protein per min, while 6 of 7 patients previously treated with
these agents had higher levels of activity. Among patients
with ANLL, the distinction was less clear. Of the 8
untreated patients, 1 had the highest value for particulatebound leukocyte alkaline phosphatase recorded in our
laboratory to date. This patient was treated for 21 days with
a full course
of cytosine
arabinoside
(1-/3-D-
OOk
*700e*•2
Chart I. Specific activities of anabolic
enzymes. Values are expressed for the
enzyme H-GPRT with both hypoxanthine
and 6-MP as substrates and for the enzyme
APRT with adenine as the substrate. The
ratio of specific activities of APRT to
H-GPRT (hypoxanthine) is shown in the
A/H ratio column. O, Patients with
ANLL; G, patients with acute lymphocytic
leukemia; •¿,
Patient I;
, mean for
each group; and vertical bars, 3 S.D. units.
Numbers in parentheses, actual values.
0Q)
600(L0x
Oo-—
500T)|o
200.0
1600o8
rfl_"3"-a
°
ß>400U.^Õ
120°Obo
O8o•s-.
150a90
Z>
3
Qi
D
O
so°"ö
100-8-Qi
0
eooWa400o
a>
*!
-ooo-
Z200sI
Da]
°£OTreated
rb
50a8CD
oUntreated20000
100cn0-Oo"
fi)r*
*
f.-
0400290-
TreatedHYPOXANTHINE Untreated
Treated
Untreated
ADENINE
Untrtattd Treated
A/H RATIO
MERCAPTOPURINE
(13.9)O
o (*-')
0(7.61O-0o
(11.ZI
C"Ee
5o£d
^E•o«
o8a8o?
O
O
3£(Ce>oEC1oO
E
Chart 2. Specific activities of catabolic
enzymes. Phosphatase activities were mea
sured with the use of either supernatant or
particulate fractions of leukemic cells as
sayed at both acid and alkaline pH values.
Symbols are as in Chart I.
o
D
°-
8oCD943Z\n
o
o
o
—¿
CCOlCD— 0
Untraattd
Treated
ACID
Untreated
ALKALINE
SUPERNATANT
1954
Treated
—¿Mini——Q
Untreated
Treated
ACID
Untreated
Treated
ALKALINE
PARTICULATE-BOUND
CANCER
RESEARCH
VOL. 34
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1974 American Association for Cancer Research.
Resistance to 6-Thiopurines in Human Leukemia
arabinofuranosylcytosine) and 6-TG (100 mg/sq mm of
each drug every 12 hr) without remission. The remainder of
the untreated patients with ANLL had little or no particulate-bound leukocyte alkaline phosphatase activity. Of the
11 treated patients with ANLL, 2 had high, 2 intermediate,
and the remainder no particulate-bound alkaline phos
phatase activity. No correlation between patient status and
the activities of either soluble alkaline phosphatase or
soluble or particulate acid phosphatase was apparent.
DISCUSSION
The levels of the enzyme H-GPRT in leukemic leukocytes
of patients deemed susceptible or resistant to the antileukemic activity of 6-MP and 6-TG are in general
agreement with previously published results (5, 14). To date,
severe deficiency of this enzyme has been reported in human
leukemia by Davidson and Winter (5) in 1 of 14 patients, by
Rosman and Williams (14) in 1 of 8 patients, and in this
investigation in 1 of 18 treated patients. However, use of the
A/H ratio discriminator has allowed the detection of the
possible presence of a moderate anabolic enzyme deficiency
in an additional 6 resistant cases [i.e., in 3 additional
patients in the study by Davidson and Winter (5), in 1 other
case in the series of Rosman and Williams (14), and in 2
additional individuals in this study]. Decreases in H-GPRT
activity associated with insensitivity to purine analogs have
occurred only in patients with ANLL. Thus, in this disease,
severe H-GPRT deficiency may account for 12% of resist
ant cases, while moderate enzyme deficiency may be
responsible for an additional 23%. Loss of analog nucleotide-forming potential was not implicated in the mechanism
of resistance to 6-thiopurines in acute lymphocytic leukemia
to a significant extent.
The relatively low frequency of attainment of H-GPRT
mutants in 6-thiopurine-resistant leukemic leukocytes of
humans compared with murine neoplasms and bacterial
systems (4), may be a consequence of little or no de novo
biosynthesis of purine nucleotides by the leukemic leuko
cytes (15, 17), coupled with major dependence by these cells
on the utilization of preformed purines for the formation of
coenzymes and nucleic acids (9, 13). Should such mech
anisms be operative, leukemic leukocytes from patients
achieving insensitivity to 6-TG and 6-MP by lowering
H-GPRT activity would be expected to be highly dependent
upon preformed adenine utilization for survival. This popu
lation of neoplastic cells should, therefore, be susceptible to
cytotoxic analogs of adenine and, for example, to 5(4)amino-4(5)-thioimidazole carboxamide which, following ex
pected activation by APRT to the nucleotide level and ring
closure, would generate 6-thioinosinic acid by an alternate
route. This latter possibility is under consideration by our
laboratory.
A change in catabolic potential for nucleotide degrada
tion by leukemic cells of humans (i.e., increased phosphohydrolase activity), as described for a transplanted murine
6-thiopurine-resistant neoplasm (3, 19), theoretically ap
peared to be a less difficult alteration for these cells to
accomplish resistance to either 6-MP or 6-TG. In support of
AUGUST
this concept, particulate-bound alkaline phosphatase activ
ity was elevated in leukocytes from 6 of 7 patients with acute
lymphocytic leukemia deemed resistant to purine antimetabolites, while most untreated patients had negligible
activity (Chart 2). Of the patients with ANLL, 2 had high
values for particulate alkaline phosptratase activity. Leuko
cytes from 5 of 11 patients with ANLL had either elevated
particulate alkaline phosphatase activity or decreased HGPRT activity associated with clinical resistance. No
correlation between the clinical status of patients and the
activities of particulate-bound acid phosphatase or of solu
ble acid or alkaline phosphohydrolases was evident. Of the
untreated patients, 1 showed an extremely high level of
particulate alkaline phosphatase activity and a 2nd had an
intermediate level of activity (0.91 nmole PÌ
per mg per
min). However, the predictive value of elevated particulate
alkaline phosphatase levels in untreated patients as to their
subsequent response to chemotherapy with purine analogs is
unclear from this study. Although the patient with ANLL
with a particulate alkaline phosphatase activity of 12.8
nmoles P¡per mg protein per min proved resistant to a
rather extraordinary course of chemotherapy with cytosine
arabinoside and 6-TG, as described above, and the patient
with acute lymphocytic leukemia with an activity of 0.91
nmole P, per mg protein per min relapsed after 2 months on
the St. Jude regimen (16), there was an insufficient number
of responders in our untreated group and an insufficient
number of total untreated patients with elevated phospha
tase levels to make the response data meaningful. Unfortu
nately, in this period of multidrug therapy, when the
6-thiopurines are no longer used to induce remission in
acute lymphocytic leukemia, a prospective study would be
extremely difficult to carry out. Nevertheless, it appears
that in patients with acute lymphocytic leukemia, an
increase in the capability of leukemic leukocytes to degrade
analog nucleotide may be, at least in part, responsible for
the attainment of insensitivity to the 6-thiopurines. The
development of an effective inhibitor of alkaline phospha
tase to be used in combination with 6-MP or 6-TG may,
therefore, be important in restoring the sensitivity of such
resistant cells to these agents. It appears that promising
leads in this area are represented by the findings of our
laboratory that derivatives of «-(A')-heterocycliccarboxyaldehyde thiosemicarbazones (10) and tetramisole (18) are
potent inhibitors of alkaline phosphatase isolated from the
murine neoplastic model system, Sarcoma 180/TG.
ACKNOWLEDGMENTS
The authors wish to thank the clinical staffs of Hunter 5 and the
pediatrie hematology service for their cooperation; and Charles Shansky
for his excellent technical assistance.
REFERENCES
1. Benke, P. J., Herrick, N., and Herbert, A. Hypoxanthine-Guanine
Phosphoribosyltransferase Variant Associated with Accelerated Pu
rine Synthesis. J. Clin. Invest., 52: 2234-2240, 1973.
2. Bennett, L. L., Jr., Allan, P. W., Smithers, D., and Vail, M. H.
1974
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1974 American Association for Cancer Research.
1955
M. Rosman et al.
Resistance to 4-Aminopyrazolo(3,4-<y)pyrimidine.
macol., 18: 725-740, 1969.
Biochem.
Phar-
3. Bieber, A. L.. and Sartorelli. A. C. The Metabolism of Thioguanine in
Furine Analog-resistant Cells. Cancer Res., 24: 1210-1215, 1964.
4. Brockman, R. W. Mechanisms of Resistance to Anticancer Agents.
Advan. Cancer Res., 7: 129 234, 1963.
5. Davidson, J. D., and Winter, T. S. Furine Nucleotide Pyrophosphorylases in 6-Mercaptopurine-sensitive
and -resistant Human
Protein Measurement with the Folin Phenol Reagent. J. Biol. Chem.,
193: 265 275. 1951.
12. McDonald, J. A., and Kelley, W. N. Lesch-Nyhan Syndrome: Altered
Kinetic Properties of Mutant Enzyme. Science, ///.- 689 691, 1971.
13. Pritchard, J. B., Chavez-Peon. F.. and Berlin, R. D. Purines: Supply
by Liver to Tissues. Am. J. Physiol., 219: 1263 1267, 1970.
14. Rosman, M., and Williams. H. E. Leukocyte Purine Phosphoribosyltransferases in Human Leukemias Sensitive and Resistant to 6-Thiopurines. Cancer Res., 33: 1202 1209, 1973.
Leukemias. Cancer Res., 24: 261 267, 1964.
15. Scott, J. L. Human Leukocyte Metabolism In Vitro. I. Incorporation
6. Gartler, S. M., Scott, R. C., Goldstein, J. L., Campbell, B., and
of Adenine-8-C14 and Formate-C"
into the Nucleic Acids of LeuSparkes, R. Lesch-Nyhan Syndrome: Rapid Detection of Hétérozy
gotes by Use of Hair Follicles. Science, 172: 572-574, 1971.
kemic Leukocytes. J. Clin. Invest., 41: 67-79, 1962.
16. Simone. J., Aur, R. J. A., Hustu, H. O., and Pinkel, D. "Total
7. Henderson, J. F., Caldwell, I. C., and Paterson, A. R. P. Decreased
Therapy" Studies of Acute Lymphocytic Leukemia of Children.
Feedback Inhibition in a 6-(Methylmercapto)purine
Ribonucleoside-
resistant tumor. Cancer Res., 27: 1773 1778. 1967.
8. Kessel, D., and Hall. T. C. Retention of 6-Mercaptopurine
Cancer, 30: 1488-1494, 1972.
Derivatives
by Intact Cells as an Index of Drug Response in Human and Murine
Leukemias. Cancer Res., 29: 21162119, 1969.
9. Lajtha, L. G., and Vane, J. R. Dependence of Bone Marrow Cells on
the Liver for Purine Supply. Nature, 182: 191 192, 1958.
10. Lee, M. H., Sznycer-Bochner, E., Agrawal. K. C., Wolpert, M. K.,
and Sartorelli. A. C. Development of inhibitors of Alkaline Phosphatase. An Enzyme Involved in Resistance of Sarcoma 180/TG to
Thiopurines. Biochem. Pharmacol., 22: 1477 1485, 1973.
11. Lowry, O. H.. Rosebrough, N. J., Farr, A. L., and Randall. R. J.
1956
17. Smelile, R. M. S.. Thomson. R. Y., and Davidson, J. N. The Nucleic
acid Metabolism of Animal Cells In Vitro. I. The Incorporation of
"C-Formate. Biochim. Biophys. Acta, 29: 59-74, 1958.
18. Van Belle, H. Kinetics and Inhibition of Alkaline Phosphatases from
Canine Tissues. Biochim. Biophys. Acta, 289: 158 168, 1972.
19. Wolpert, M. K.. Damle, S. P., Brown, J. E., Sznycer, E., Agrawal, K.
C., and Sartorelli, A. C. The Role of Phosphohydrolases in the
Mechanism of Resistance of Neoplastic Cells to 6-Thiopurines.
Cancer Res.. 31: 1620 1626, 1971.
CANCER RESEARCH
VOL. 34
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1974 American Association for Cancer Research.
Mechanisms of Resistance to 6-Thiopurines in Human Leukemia
Martin Rosman, Men Hui Lee, William A. Creasey, et al.
Cancer Res 1974;34:1952-1956.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/34/8/1952
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1974 American Association for Cancer Research.