Phase I and II Study of Fludarabine Phosphate in Leukemia:

Therapeutic Efficacy With Delayed Central Nervous System Toxicity

By Raymond P. Warrell, Jr, and Ellin Berman
Fludarabine phosphate (9-B-D-arabinofuranosyl-2fluoroadenine), a novel purine nucleoside, has demonstrated excellent preclinical antitumor activity and
little toxicity in phase I clinical trials. We evaluated
the clinical use of fludarabine given as a continuous
intravenous (IV) infusion for remission induction in
patients with relapsed or refractory leukemia. Thirty
infusions were administered to 25 patients. At doses
- 125 mg/m 2/d for five days, only three of 17 patients cleared their bone marrow of leukemic cells,
and none achieved complete remission (CR). Nine patients received doses of 150 mg/m 2/d for five days or
125 mg/m 2 /d for seven days. Four of these patients
achieved CR (three patients with acute nonlymphoblastic leukemia (ANLL), one patient with acute lymphoblastic leukemia (ALL)). However, severe CNS toxicity was encountered in five patients at the two highest dose levels. Initial symptoms of neurotoxicity

were delayed from 21 to 43 days after starting treatment and consisted of optic neuritis, cortical blindness, altered mental status, and generalized seizure.
Only one patient regained visual and neurologic
function; four other patients experienced progressive
neurologic deterioration and died. Clinicopathologic
evaluation suggested widespread, severe demyelination as the etiology of these reactions. We conclude
that fludarabine is an effective drug for remission
induction in acute leukemia. However, doses required
to achieve CR are associated with unacceptable CNS
toxicity. In view of its potent antileukemic activity,
further evaluation of fludarabine at lower doses
(: 75 mg/m 2 /d for five days) may be warranted in
combination with other chemotherapeutic agents for
the treatment of patients with acute leukemia.
J Clin Oncol 4:74-79. 1986 by American Society of
Clinical Oncology.

dephosphorylation to 2-F-ARA-A. 6' 7 The compound is then incorporated intracellularly and

converted to the triphosphate, which is a highly
competitive inhibitor of DNA synthesis.8,9
Preclinical studies showed activity against
P388 and L 1210 leukemia as well as other animal
and human tumor cells. 9'-91Several phase I studies in cancer patients have been completed. Myelosuppression has been the dose-limiting toxic
effect.'2.13 Although Hutton et al observed transient somnolence in several patients,' 2 the drug
has otherwise been extremely well tolerated. Because of the potent antileukemic activity of other
nucleosides, we initiated a clinical study in patients with relapsed or refractory acute leukemia.

LUDARABINE PHOSPHATE (9-/3-D-arabinofuranosyl-2-fluoroadenine 5'-monophosphate; 2-F-ARA-AMP) is a new purine nucleoside that has recently entered clinical trials.
The parent compound, 9-P3-D-arabinofuranosyl
adenine (ARA-A), is rapidly deaminated in
vivo,'-2 a problem that has limited both its antitumor and antiviral use. The 2-fluoro derivative of
ARA-A is resistant to adenosine deaminase34 but
is poorly soluble in water. Solubility was greatly
increased with the addition of a monophosphate
group at the 5' position. 5 .6 The resulting drug,
fludarabine phosphate (Fig 1), undergoes rapid
From the Developmental Chemotherapy and Hematology/
Lymphoma Services, MemorialSloan-Kettering Cancer Center
and the Cornell University Medical College, New York.
Submitted April 22, 1985; accepted Aug 15, 1985.
Supported in part by Contract NOI-CM-8426 from the National CancerInstitute, NationalInstitutes of Health, by a Junior Faculty Fellowship Awardfrom the American CancerSociety (R.P.W.), and by a grantfrom E.I. DuPont de Nemours.
Presented in part at the annual meeting of the American
Associationfor Cancer Research, May 1985.
Address reprint requests to Raymond P. Warrell, Jr, MD,
Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New
York, NY 10021.
C 1986 by American Society of Clinical Oncology.
0732- 183X186/0401-0001/$3.00/0

74

PATIENTS AND METHODS

Patient Selection
Patients with acute nonlymphoblastic leukemia (ANLL) or
acute lymphoblastic leukemia (ALL) who had failed to respond
to conventional chemotherapy or had relapsed after achieving
remission were eligible for entry. A medical history, physical
examination, chest roentgenogram, and electrocardiogram were
performed. Serum creatinine and bilirubin concentrations < 2.5
mg/dL were required. Pretreatment evaluation also included a
bone marrow aspirate, complete blood cell count (CBC), and
serum biochemical screening profile. There was no restriction
on the amount of prior therapy. Signed, informed consent was

Journal of Clinical Oncology, Vol 4, No 1 (January), 1986: pp 74-79

Downloaded from ascopubs.org by 212.39.125.182 on November 24, 2016 from 212.039.125.182

Copyright 2016 American Society of Clinical Oncology. All rights reserved.

75

FLUDARABINE IN LEUKEMIA

RESULTS
Thirty infusions of fludarabine were adminis-

NH2

NI

IK%
IT

tered to 25 patients. Characteristics of this patient population are presented in Table 1. Twelve
patients were treated at the time of first relapse.
More than 50% of the patients had an outpatient
performance status at the time of entry. However, one third of the treatment courses were initiated after antibiotic therapy had been started for
fever and presumed infection. One patient with

II

0
OH
Fig 1. Structure of fludarabine phosphate (9-B-Darabinofuranosyl-2-fluoroadenine; 2-F-ARA-AMP).
obtained. The study was approved in advance by this center's
Investigational Review Board.
During the induction phase of therapy, patients were monitored with daily blood counts. Serum electrolyte determinations
and a 12-channel automated screening profile were performed
twice weekly.

Assessment of Response
Response to fludarabine was evaluated twice during the
study. First, we evaluated the ability of the drug to produce bone
marrow hypoplasia without evidence of leukemic cells. This
assessment was made on the basis of a marrow aspirate and/or
biopsy performed ten to 14 days after initiation of therapy.
Second, patients were evaluated for remission status by serial
bone marrow aspirates performed every seven to 14 days thereafter. Complete remission (CR) was defined as a cellular bone
marrow containing < 5% blasts, a peripheral leukocyte count
> 3000/ILL, a hemoglobin concentration > 10 g/dL, and a
platelet count > 100,000/L. Patients achieving less than a CR
were counted as failures. At least four evaluable patients were
entered at each dose level. Escalation to subsequent dose levels
was performed if less than one half of the patients achieved
marrow hypoplasia at the preceding level.

Drug Administration
The starting dose approximated twice the dose recommended
for phase II studies in patients with solid tumors. 12, 13The prescribed daily dose of fludarabine phosphate was diluted in 1,000
mL of normal saline and was administered over 24 hours using
an infusion pump. Both peripheral and central venous catheters
were used. Patients received a single infusion of fludarabine. If
they failed to achieve CR, they were removed from the study and
received alternative treatment. Multiple, consecutive infusions
were not administered. However, patients who showed some
response to a lower dose of fludarabine (eg, marrow hypoplasia
without CR) were eligible for retreatment at a higher dose if they
had subsequently failed to respond to other chemotherapy administered during the interim period.

ALL had CNS leukemia and had received intrathecal methotrexate. No other patient had prior neurologic symptoms, and none had previously received high-dose cytosine arabinoside or
cranial irradiation.
Six dose levels were examined (Table 2).
Twenty-one infusions were administered to 17
patients at doses - 125 mg/m 2/d for five days.

Although a decrease in peripheral blast counts

was observed in all patients, only three patients
cleared their bone marrow of leukemic cells, and

none achieved remission. Nine patients were entered at the two highest dose levels, 150 mg/m 2/d

for five days and 125 mg/m 2/d for seven days.
Three of these patients died before day 10 of
non-drug-related causes; thus, their marrow was
not examined for hypocellularity. Marrow hypoplasia developed in four of the six remaining
patients, and each of these individuals achieved
CR. Three patients who achieved CR had ANLL
and were treated at the time of their first relapse.
The fourth patient had ALL and was treated after
Table 1.

Characteristics of Patient Population

Treated With Fludarabine

Characteristics
No. patients entered
No. treatment courses
Median age (range)
Median performance status,
Karnofsky (range)
Diagnoses (n)
ANLL
ALL
AML/ALL*
Prior therapy (n)
First relapse
Two (+) relapses
Primary resistant
Antibiotics at entry (n)
Abbreviation: n, number of patients.
*Biphenotypic leukemia.

Downloaded from ascopubs.org by 212.39.125.182 on November 24, 2016 from 212.039.125.182

Copyright 2016 American Society of Clinical Oncology. All rights reserved.

Value
25
30
39 (23-64)
60 (30-80)
21
3
1
12
5
8
10

76

WARRELL AND BERMAN

Table 2.

Therapeutic Effects of Fludarabine in

Patients With Acute Leukemia

Dose
mg/m2
X days
50
75
100
125
150
125

x
x
x
x
x
x

5
5
5
5
5
7

No.
Patients
Entered

No.
Patients
Hypoplastic/
Evaluable

No.
CRs

4
8
4
5
5
4

1/4
1/8
0/4
1/5
2/3
2/3

0
0
0
0
2
2

CR
Duration
(mo)

1, 2.5
1, 1.5+

two relapses, having failed an induction attempt

with other therapy immediately before entering
this study. Because of neurotoxicity encountered
at these levels, consolidation and maintenance
treatment with fludarabine, initially planned for
this protocol, was not administered. Patients
achieved CR by all criteria 34, 45, 49, and 123
days after beginning treatment, the latter patient
having experienced prolonged thrombocytopenia. None of these patients received further
chemotherapy with fludarabine. CRs in patients
with ANLL persisted from 1 to 2.5 months. The
single patient with ALL died in remission from
CNS toxicity.
TOXICITY

As predicted by the phase I studies, acute toxic

reactions to fludarabine were mild. Only two
patients experienced nausea or vomiting; one patient complained of excessive fatigue during the
infusion. We did not observe any evidence of
somnolence. All treatment courses produced
moderate to severe myelosuppression that required antibiotics and platelet transfusions.
Severe CNS toxicity was observed at the two
highest dose levels. The first reaction occurred in
a patient in whom an altered mental status developed 29 days after beginning treatment in the
setting of high fever, intravascular coagulation,
and renal insufficiency. Subsequently, cortical
blindness, incontinence, and further deterioration in mental status developed, and the diagnosis of progressive multifocal leukoencephalopathy was considered. Lumbar punctures
revealed acellular spinal fluid with normal protein and glucose concentrations. Two computed
tomographic brain scans were normal. Requests

for brain biopsy were declined. The patient

achieved a CR but experienced a generalized tonic-clonic seizure. Postictally, he was unresponsive with decerebrate posturing.
Because the onset of the initial reaction was
quite delayed, the possibility c - a fludarabinerelated toxic reaction was not considered until
eight more patients were treated at the two highest dose levels. Three heavily pretreated patients
with low performance status died of septicemia
within ten days of treatment. Another patient
died 3 weeks later, having failed to achieve either
marrow hypoplasia or remission. Five patients
experienced CNS reactions, four of which were
considered definitely drug related. Generalized
seizures developed in a fifth patient as a terminal
event, and the patient died from disseminated
candidiasis. The neurotoxic symptoms are described in Table 3. The initial manifestations of
neurotoxicity were delayed from 21 to 43 days
after starting treatment. Four of these patients
died, three of whom had achieved CR. One patient in whom total blindness and quadriparesis
developed gradually regained completely normal
vision and strength. Her only neurologic sequela
was an asymptomatic delay in visually evoked
response.
Clinical manifestations of neurotoxicity varied
as noted in Table 3. Three patients were found to
have optic neuritis. Computerized tomography
of the brain in four patients failed to reveal any
abnormalities. However, NMR scans of two patients showed an extensive, diffuse loss of white
matter. Visually evoked potentials were absent
in the two patients tested. Electroencephalography revealed nonfocal abnormalities with diffuse
Table 3.

Characteristics of Fludarabine-Related
Neurotoxicity in Five Patients

Initial symptoms*
Altered mental status
Blurred vision
Amaurosis
Generalized seizure
Progressive symptoms*
Blindness
Coma
Spastic paralysis
Flaccid paralysis
Quadriparesis

2
1
1
1
5
4
3
1
1

NOTE. Days of onset = 21, 29, 34, 41, and 43.

*Number of patients affected.

Downloaded from ascopubs.org by 212.39.125.182 on November 24, 2016 from 212.039.125.182

Copyright 2016 American Society of Clinical Oncology. All rights reserved.

77

FLUDARABINE IN LEUKEMIA

slowing. Only one of the five patients had an

increase in spinal fluid protein. The myelin basic
protein concentration was increased to 8.7
ng/mL (normal, - 4.0 ng/mL) without immunoglobin G (IgG) elevation or oligoclonal banding
in the one patient tested. An autopsy of one of the
four patients who died with neurotoxic symptoms showed severe demyelination with reactive
gliosis. Another patient whose brain could not be
examined showed extensive hemorrhagic necrosis of the white matter of the spinal cord. Similar
findings have been observed in patients from other centers in whom neurotoxicity developed from
fludarabine administered by serial bolus injections rather than continuous intravenous (IV) infusion.14
DISCUSSION

The synthesis of fludarabine represented a major achievement in purine pharmaceutical chemistry with respect to solubility and resistance to
deamination. Laboratory studies suggested that
this drug would have important anticancer activity. Preliminary clinical trials showed that it was
well tolerated. Unfortunately, the CNS toxicity
observed in this trial was not predicted by either
animal toxicologic studies nor preceding phase I
clinical trials.
Since the onset of neurotoxic symptoms occurred 3 to 6 weeks after drug administration, we
reviewed the subsequent course of all patients
who received fludarabine at doses - 125
mg/m 2/d for five days. Follow-up was complete
on all patients. Clinical signs of CNS toxicity
developed in no patient in this group; the median
survival after administration of fludarabine was
11 weeks. Thus, it seems unlikely that significant toxicity was overlooked at the lower dose
levels in this study. However, we have since
learned of one patient who experienced severe
neurotoxicity after receiving a dose of 100
mg/m 2/d for five days (Dr H. Chun, personal
communication). No cases of CNS toxicity have
occurred with single courses of fludarabine at
doses 5 75 mg/m2/d, although neurotoxicity has
been observed with repeated dosing at this level.
Using higher doses of fludarabine, it is unclear
what interval between infusions can be considered safe. One patient in this study received infusions of 50, 75, and 100 mg/m 2/d for five days
over a 4-month period without developing toxic-

ity. We (R.P.W., unpublished observations) and

othersl' 16 have not observed neurotoxicity in ongoing clinical trials using doses of 25 to 30
mg/m 2/d for 5 days administered by bolus injection or continuous infusion. Thus, it appears that
large cumulative doses of fludarabine can be
safely administered over prolonged periods-an
important factor for patients with diseases other
than acute leukemia.
CNS reactions have been associated with both
purine and pyrimidine derivatives. Despite different biochemical pathways of metabolism, pyrimidine-related compounds appear to principally cause cerebellar toxicity. Cerebellar reactions
are commonly observed after the administration
of high doses of cytosine arabinoside (ARAC),16-21 which can produce loss of Purkinje cells
and reactive gliosis. 1 Patients who have received
prior cranial irradiation or intrathecal chemotherapy may be at greater risk for CNS toxicity
from high-dose ARA-C.20 Transient cerebral

toxicity, including somnolence, confusion, and

dysarthria, are also common reactions."9 Breuer
et al have noted spinal cord demyelination after
intrathecal injections of ARA-C. 22 However, lethal CNS toxicity from ARA-C is rare." Similarly, the pyrimidine nucleotide 5-fluorouracil
(5-FU) can also cause persistent cerebellar
dysfunction. 23 The experimental antiviral compound 1-(2'-deoxy-2' fluoro 1-P/-D-arabinofuranosyl)-5-methyluracil (FMAU) has also produced severe, irreversible extrapyramidal
reactions.24 Thymidine has been tolerated at extremely high doses, 25 and its fluorinated derivative, trifluorothymidine, has 27not caused neuro26
toxicity in limited studies. ,
Conversely, purine-related derivatives are
principally associated with signs of cerebral dysfunction. Mercaptopurine (6-MP) competes with
guanine for hypoxanthine-guanine phosphoribosyltransferase. Myelosuppression occurs as the
dose-limiting reaction for 6-MP, and neurotoxicity has not been observed. Deoxycoformycin, a
drug that inhibits adenosine deaminase, has
caused severe ocular toxicity and lethal encephalopathy. 28 Adenine arabinoside (ARA-A; vidarabine), the compound most closely related to fludarabine, has produced fatal CNS reactions
preceded by tremors, myoclonus, dysarthria, and
coma. 29, 30 In vivo, ARA-A is rapidly deaminated
to hypoxanthine arabinoside.31 However, fludar-

Downloaded from ascopubs.org by 212.39.125.182 on November 24, 2016 from 212.039.125.182

Copyright 2016 American Society of Clinical Oncology. All rights reserved.

78

WARRELL AND BERMAN

abine is resistant to deamination. Since several

analogues of adenosine depress cortical activity,32 accumulation of phosphorylated deoxyadenosine metabolites was proposed as the
mechanism for the transient somnolence observed by Hutton et al.12 However, it is not
known which, if any, of these metabolites are
responsible for the demyelination observed in
this study.
Fludarabine is clearly an effective drug for
remission induction in both ANLL and ALL.
However, there appears to be little or no therapeutic index since doses required to achieve remission are associated with unacceptable CNS
toxicity. Clinicopathologic evaluation suggests
that fludarabine, in high doses, causes progressive demyelination that (with one exception) has
been lethal. Lower doses (ie, 5 75 mg/m 2/d for
five days) appear to be safe, and the drug is

otherwise exceptionally well tolerated. Therefore, given its potent antileukemic activity, further clinical evaluation at a low dose in combination with other chemotherapeutic agents may be
warranted in patients with acute leukemia.
ADDENDUM
The NCI has since been informed of a case of
progressive CNS dysfunction which occurred
after three courses of fludarabine treatment at
relatively low dosage (18 to 25 mg/m 2/d for five
days). The reaction is considered "probably"
drug related (H. Chun, personal communication).
ACKNOWLEDGMENT
We thank the chemotherapy research nursing staff for monitoring the drug infusions and Catherine Cassidy for data base
management. Fludarabine phosphate was supplied by the Division of Cancer Treatment, the National Cancer Institute, Bethesda, Md.

REFERENCES
1. Brink JJ, LePage GA: Metabolism and distribution of 9-13D-arabinofuranosyladenine in mouse tissues. Cancer Res 24:
1042-1049, 1964
2. LePage GA, Khaliq A, Gottlieb JA: Studies of 9-3-Darabinofuranosyladenine in man. Drug Metab Dispos 1:756759, 1973
3. Brockman RW, Schabel FM Jr, Montgomery JA: Biologic
activity of 9-B-o-arabinofuranosyl-2-fluoroadenine, a metabolically stable analog of 9-B-D-arabinofuranosyladenine. Biochem
Pharmacol 26:2193-2196, 1977
4. Montgomery JA, Hewson K: Nucleosides of 2-fluoroadenosine. J Med Chem 12:498-504, 1969
5. Brockman RW, Cheng Y-C, Schabel FM Jr, et al: Metabolism and chemotherapeutic activity of 9-I3-D-arabinofuranosyl2-fluoroadenine against murine leukemia L1210 and evidence
for its phosphorylation by deoxycytidine kinase. Cancer Res
40:3610-3615, 1980
6. LePage GA, Naik SR, Katakkan SB, et al: 9-13-o-arabinofuranosyladenine 5'-phosphate metabolism and excretion in humans. Cancer Res 35:3036-3040, 1975
7. Noker PE, Duncan GF, El Dareer SM, et al: Disposition of
9-B-o-arabinofuranosyl-2-fluoroadenine 5'-phosphate in mice
and dogs. Cancer Treat Rep 67:445-456, 1983
8. White EL, Shaddix SC, Brockman RW, et al: Comparison
of the actions of 9-1-o-arabinofuranosyl-2-fluoroadenine and 9/3-o-arabinofuranosyladenine on target enzymes from mouse tumor cells. Cancer Res 42:2260-2264, 1982
9. Plunkett W, Chubb S, Alexander L, et al: Comparison of
the toxicity and metabolism of 9-P/-D-arabinofuranosyl-2-fluoroadenine and 9-P3-o-arabinofuranosyladenine in human lymphoblastoid cells. Cancer Res 40:2349-2355, 1980
10. Avramis VI, Plunkett W: Metabolism and therapeutic
efficacy of 9-3-D-arabinofuranosyl-2-fluoroadenine against
murine leukemia P388. Cancer Res 42:2587-2591, 1982
11. Dow LW, Bell DE, Poulakos L, et al: Differences in

metabolism and cytotoxicity between 9-3-o-arabinofuranosyladenine and 9-0-o-arabinofuranosyl-2-fluoroadenine in human

leukemic lymphoblasts. Cancer Res 40:1405-1410, 1980
12. Hutton JJ, Von Hoff DD, Kuhn J, et al: Phase I clinical
investigation of 9-13-o-arabinofuranosyl-2-fluoroadenine 5'monophosphate (NSC 312887), a new purine antimetabolite.
Cancer Res 44:4183-4186, 1984
13. Casper ES, Mittelman A: Phase I trial of fludarabine (2FAMP) in patients with solid tumors. Proc Am Assoc Cancer
Res 25:203, 1984 (abstr)
14. Stillman MJ, Navia B, Mast J, et al: Delayed neurotoxicity associated with fludarabine. Neurology (NY) 35:291, 1985
(suppl)
15. Mittelman A, Ashikari R, Ahmed T, et al: Phase II trial
of 2 fluoro-Ara-AMP (fludarabine phosphate) in patients (pts)
with advanced breast cancer. Proc Am Assoc Cancer Res
26:170, 1985 (abstr)
16. Leiby JM, Grever MR, Staubus AE, et al: A phase I
clinical investigation of fludarabine phosphate by loading dose/
continuous infusion. Proc Am Assoc Cancer Res 26:169, 1985
(abstr)
17. Salinsky MC, Levine RL, Aubuchon JP, et al: Acute
cerebellar dysfunction with high-dose ARA-C therapy. Cancer
51:426-429, 1983
18. Breithaupt H, Pralle H, Eckhardt T, et al: Clinical results
and pharmacokinetics of high-dose cytosine arabinoside (HD
ARA-C). Cancer 50:1248-1257, 1982
19. Lazarus HM, Herzig RH, Herzig GP, et al: Central nervous system toxicity of high-dose systemic cytosine arabinoside. Cancer 48:2577-2582, 1981
20. Early AP, Preisler HD, Slocum H, et al: A pilot study of
high-dose I-,-D-arabinofuranosylcytosine for acute leukemia
and refractory lymphoma: Clinical response and pharmacology.
Cancer Res 42:1587-1594, 1982
21. Wolff SN, Marion J, Stein RS, et al: High-dose cytosine

Downloaded from ascopubs.org by 212.39.125.182 on November 24, 2016 from 212.039.125.182

Copyright 2016 American Society of Clinical Oncology. All rights reserved.

FLUDARABINE IN LEUKEMIA
arabinoside and daunorubicin as consolidation therapy for acute
nonlymphocytic leukemia in first remission: A pilot study.
Blood 65:1407-1411, 1985
22. Breuer AC, Pitman SW, Dawson DM, et al: Paraparesis
following intrathecal cytosine arabinoside. Cancer 40:28172822, 1977
23. Weiss HD, Walker MD, Wiemik PH: Neurotoxicity of
commonly used antineoplastic agents. N Engl J Med 291:75-81,
127-133, 1974
24. Fanucchi MP, Leyland-Jones B, Young CW, et al: Phase
I trial of 1-(2'-deoxy-2'fluoro 1-/3-D arabinofuranosyl)-5-methyluracil (FMAU). Cancer Treat Rep 69:55-59, 1985
25. Blumenreich MS, Woodcock TM, Andreef M, et al:
Effects of very high dose thymidine infusion on leukemia and
lymphoma patients. Cancer Res 44:2203-2207, 1984
26. Helson L, Yagoda A, McCarthy M, et al: Clinical trials
with trifluoromethyl 2' deoxyuridine (F 3TdR). Proc Am Assoc
Cancer Res 11:35, 1970 (abstr)
27. Warrell RP Jr, Currie V, Kempin S, et al: Phase II trial of
pyrazofurin, alone and in combination with trifluorothymidine,

79
in non-Hodgkin's lymphoma. Cancer Treat Rep 63:1423-1425,
1979
28. Grever MR, Siaw MFE, Jacob WF, et al: The biochemical and clinical consequences of 2'-deoxycoformycin in refractory cancer. Proc Am Assoc Cancer Res 20:47, 1979 (abstr)
29. Van Etta L, Brown J, Mastri A, et al: Fatal vidarabine
toxicity in a patient with normal renal function. JAMA 246:
1703-1705, 1981
30. Marker SC, Howard RJ, Groth KE, et al: A trial of
vidarabine for cytomegalovirus infection in renal transplant patients. Arch Intern Med 140:1441-1444, 1980
31. Kinkel AW, Buchanan RA: Human pharmacology, in
Pagan-Livingston D, Buchanan RA, Alford CA Jr (eds): Adenine Arabinoside: An Antiviral Agent. New York, Raven, 1975,
pp 197-204
32. Phillis JW, Kostopolous GK, Edstrom JP, et al: Role of
adenosine and adenine nucleotides in central nervous function,
in Baer HP, Drummond GI (eds): Physiological and Regulatory
Functions of Adenosine and Adenine Nucleotides. New York,
Raven, 1979, 343-359

Downloaded from ascopubs.org by 212.39.125.182 on November 24, 2016 from 212.039.125.182

Copyright 2016 American Society of Clinical Oncology. All rights reserved.