A Phase I Open-Label, Dose-Escalation, Multi-Institutional Trial of Injection with an E1B-Attenuated Adenovirus, ONYX-015, into the Peritumoral Region of Recurrent Malignant Gliomas, in the Adjuvant Setting

Abstract

ONYX-015 is an oncolytic virus untested as a treatment for malignant glioma. The NABTT CNS Consortium conducted a dose-escalation trial of intracerebral injections of ONYX-015. Cohorts of six patients at each dose level received doses of vector from 10(7) plaque-forming units (pfu) to 10(10) pfu into a total of 10 sites within the resected glioma cavity. Adverse events were identified on physical exams and testing of hematologic, renal, and liver functions. Efficacy data were obtained from serial MRI scans. None of the 24 patients experienced serious adverse events related to ONYX-015. The maximum tolerated dose was not reached at 10(10) pfu. The median time to progression after treatment with ONYX-015 was 46 days (range 13 to 452 + days). The median survival time was 6.2 months (range 1.3 to 28.0 + months). One patient has not progressed and 1 patient showed regression of interval-increased enhancement. With more than 19 months of follow-up, 1/6 recipients at a dose of 10(9) and 2/6 at a dose of 10(10) pfu remain alive. In 2 patients who underwent a second resection 3 months after ONYX-015 injection, a lymphocytic and plasmacytoid cell infiltrate was observed. Injection of ONYX-015 into glioma cavities is well tolerated at doses up to 10(10) pfu.

Full text

A Phase I Open-Label, Dose-Escalation, Multi-Institutional
Trial of Injection with an E1B-Attenuated Adenovirus,
ONYX-015, into the Peritumoral Region of Recurrent
Malignant Gliomas, in the Adjuvant Setting
E. Antonio Chiocca,
1,
*
Khalid M. Abbed,
1
Stephen Tatter,
1
David N. Louis,
1
Fred H. Hochberg,
1
Fred Barker,
1
Jean Kracher,
1
Stuart A. Grossman,
1
Joy D. Fisher,
1
Kathryn Carson,
1
Mark Rosenblum,
1
Tom Mikkelsen,
1
Jeff Olson,
1
James Markert,
1
Steven Rosenfeld,
1
L. Burt Nabors,
1
Steven Brem,
1
Surasak Phuphanich,
1
Scott Freeman,
2
Rick Kaplan,
3
and James Zwiebel
3
1
The NABTT CNS Consortium, Baltimore, MD 21231, USA
2
Onyx Pharmaceuticals, Redmond, CA, USA
3
Cancer Therapy Evaluation Program, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA
*To whom correspondence and reprint requests should be addressed c/o The NABTT CNS Consortium, 1650 Orleans Street, Room G93,
Baltimore, MD 21231. Fax: +1 410 614 9335. E-mail: jfisher@jhmi.edu.
Available online 24 August 2004
ONYX-015 is an oncolytic virus untested as a treatment for malignant glioma. The NABTT CNS
Consortium conducted a dose-escalation trial of intracerebral injections of ONYX-015. Cohorts of six
patients at each dose level received doses of vector from 10
7
plaque-forming units (pfu) to 10
10
pfu
into a total of 10 sites within the resected glioma cavity. Adverse events were identified on physical
exams and testing of hematologic, renal, and liver functions. Efficacy data were obtained from serial
MRI scans. None of the 24 patients experienced serious adverse events related to ONYX-015. The
maximum tolerated dose was not reached at 10
10
pfu. The median time to progression after
treatment with ONYX-015 was 46 days (range 13 to 452 + days). The median survival time was 6.2
months (range 1.3 to 28.0 + months). One patient has not progressed and 1 patient showed
regression of interval-increased enhancement. With more than 19 months of follow-up, 1/6
recipients at a dose of 10
9
and 2/6 at a dose of 10
10
pfu remain alive. In 2 patients who underwent a
second resection 3 months after ONYX-015 injection, a lymphocytic and plasmacytoid cell infiltrate
was observed. Injection of ONYX-015 into glioma cavities is well tolerated at doses up to 10
10
pfu.
Key Words: glioma, brain tumor, clinical trial, ONYX-015, gene therapy, oncolytic virus,
experimental therapy, virotherapy, adenovirus, conditionally replicating adenoviruses
INTRODUCTION
Current therapies such as surgery, radiotherapy, and
chemotherapy have had only limited success in treating
patients with malignant gliomas [1], and thus their
prognosis remains grim [2]. Novel treatment strategies
are needed [3–5]. Oncolytic viral therapy is one such
strategy [6,7]. This consists of the use of replication-
conditional viruses, genetically altered to render their
replication selective for tumor cells. An example of a
replication-conditional virus is ONYX-015, an adenovirus
mutant that is thought to replicate more efficiently in
cells with disruptions in the p53 tumor suppressor
pathway (such as tumor cells) [8], although this mecha-
nism remains controversial since other mechanisms of
replicative selectivity may be operative [9]. ONYX-015
has been tested in clinical trials for a variety of cancers,
including head and neck, ovarian, prostate, and lung [10–
23].
The standard therapy for patients with a suspected
malignant brain mass is surgical excision, if possible.
Resected tissue is analyzed and a diagnosis of glioma is
made dependent on the presence of histologic features.
The histologic features of the more malignant forms of
glioma include nuclear atypia, endothelial proliferation,
necrosis, and mitoses. Based on how many of these
features are present, further classification of a malignant
glioma into an anaplastic astrocytoma (World Health
Organization grade III astrocytoma), an anaplastic oligo-
1525-0016/$30.00
TRIALS
doi:10.1016/j.ymthe.2004.07.021
MOLECULAR THERAPY Vol. 10, No. 5, November 2004
958
Copyright C The American Society of Gene Therapy

dendroglioma (World Health Organization grade III
oligodendroglioma), or a glioblastoma multiforme
(World Health Organization grade IV astrocytoma) is
made. Because these tumors are characterized by exten-
sive infiltration of single and multiple cells throughout
the brain, which obviously cannot be resected, further
treatment of patients with radiation is started 2 weeks
after surgical excision. A total of 5400–6000 cGy is
delivered to the tumor cavity and to the margins over a
period of 6 weeks. At the end of this treatment, addition
of chemotherapy with alkylating agents, such as BCNU,
may provide additional therapeutic benefit. Throughout
this time, patients are usually maintained on cortico-
steroids to reduce brain edema and anticonvulsants to
reduce the incidence of seizures. Almost all tumors will
recur during or after the above treatments, usually locally
at the margin or site of the previous resection or, less
often, at a distance from the main tumor mass. Generally,
phase I/II experimental therapies are reserved for this
group of patients.
The objectives of our study were to determine the
safety and evaluate the efficacy of multiple injections
of escalating doses of an E1B-attenuated adenovirus,
ONYX-015, into the margins of a recurrent malignant
glioma that has been resected in adult patients. We
found that injection was very well tolerated without
evidence of toxicities attributable to ONYX-015, up to
a dose of 10
10
plaque-forming units (pfu). Although
previous trials of direct, stereotactic injection of repli-
cating, tumor-selective herpes simplex viruses into the
malignant gliomas of patients were reported [24–26],
this study provides the first demonstration that injec-
tions of a replicating, tumor-selective virus into brain
tissue, adjacent to a freshly excised glioma, are well
tolerated. This provides justification for additional
studies of such modalities in patients with malignant
gliomas.
RESULTS
Twenty-four patients were enrolled in the study bet-
ween January 2000 and May 2002. All twenty-four
patients were treated and all included in the intent-to-
treat population. Patient characteristics are shown in
Table 1.
A summary of the adverse events that occurred
during the treatment period is shown in Table 2. Ten
of the 24 patients experienced one or more adverse
events. None of the adverse events were judged as
possibly related to ONYX-015 treatment, since other
etiologies provided more likely explanations. This judg-
ment was made by the clinicians involved in the care of
these patients and was reviewed by the Data Safety
Monitoring Board after each enrollment into each
cohort was completed. This judgment was based on
the likelihood that additional etiologies for the adverse
events were more likely than the injection of ONYX-
015. For instance, dyslexia, dyscalculia, and dysgraphia
in the postoperative period after resection of a tumor
near eloquent cortex responsible for such function was
more likely due to the trauma of surgery than to ONYX-
015 injection.
One patient in the high-dose (10
10
) cohort, whose
on-study histology was anaplastic astrocytoma, has not
progressed (stable disease). The remaining 23 patients
have progressed. However, it should be noted that 1 of
the patients who had been declared to have progressed
because of increased enhancement on the MRI scan
was found to have decreased enhancement on a
subsequent scan (Fig. 1). This episode of increased
enhancement with subsequent decreased enhancement
occurred again a few months later, as detailed in the
legend to Fig. 1. Median time to progression of disease
after ONYX for all patients was 46 days with a range of
13 to 452 + days.
TABLE 1: Baseline demographic and clinical characteristics for all patients and stratified by on-study diagnosis
All patients (n = 24) GBM (n = 17) Other (AA, AO) (n =7)
Sex, male 17 (71) 11 (65) 6 (86)
Race, white 24 (100) 17 (100) 7 (100)
Age, years 52 (35–70) 55 (37–70) 38 (35–61)
Karnofsky performance status 90 (60–100) 90 (60–100) 90 (70–100)
Prior chemotherapy
a
22 (92) 16 (94) 6 (86)
Anticonvulsant therapy 18 (75) 11 (65) 7 (100)
Original diagnosis histology
b
Glioma 1 (4) 1 (6) 0 (0)
Grade 2 glioma 2 (8) 0 (0) 2 (29)
Anaplastic astrocytoma 5 (21) 0 (0) 5 (71)
Glioblastoma multiforme 16 (67) 16 (94) 0 (0)
Median (range) or N (%) is shown.
a
All patients had prior surgery and XRT.
b
Refers to the histology at initial presentation of the patient with disease. Patients then underwent conventional, standard treatment. At re-presentation with a recurrence, this histology
was either the same or had progressed to a more malignant grading (the histology at recurrence is textually described under Results).
ARTICLE
doi:10.1016/j.ymthe.2004.07.021
MOLECULAR THERAPY Vol. 10, No. 5, November 2004
959
Copyright C The American Society of Gene Therapy

Twenty patients have died due to tumor progression, 1
died of non-tumor-related events (ruptured intestine),
and 3 remain alive. Of the survivors, 1 belongs to the
10
9
pfu cohort, and 2 belong to the 10
10
pfu cohort.
Median survival time for all patients was 6.2 months
(range 1.3–28.0 months). On-study histology for the
patients by cohort was 6/6 glioblastoma multiforme
(GBM) for the 10
7
pfu cohort, 4/6 GBM and 2/6
anaplastic astrocytoma (AA) for the 10
8
pfu cohort, 5/6
GBM and 1/6 AA for the 10
9
pfu cohort, and 2/6 GBM,
3/6 AA, and 1/6 anaplastic oligodendroglioma (AO) for
the 10
10
pfu cohort. For the 3 surviving patients, 2
patients had histologies of AA and the other had AO.
The Kaplan–Meier survival curve for GBM and non-GBM
(AA and AO) patients is shown in Fig. 2. Median survival
for the GBM patients was 4.9 months, and for AA and
AO patients it was 11.3 months.
We obtained serum from all of the patients prior to the
introduction of ONYX-015 and again on day 42 to assess
for antibodies to adenovirus. Of the 24 patients, only 2
patients tested positive for adenovirus antibodies before
inoculation. After delivery of ONYX-015, 2 patients
seroconverted from negative to positive for adenovirus
antibodies. One patient belonged to the 10
9
cohort and
the other patient belonged to the 10
10
cohort.
One patient in the 10
10
cohort (maximum dose) and
one patient in the 10
7
cohort had a recurrent mass 3
months after inoculation with ONYX-015. On MRI scans
these masses were manifested as increases in gadolinium
enhancement (Fig. 3). Because of their good performance
status, these patients thus underwent reoperation for
resection. On histologic examination of the recurrent
tumor, profound lymphocytic and plasmacytoid cellular
infiltrates in perivascular locations were noted within the
tumor but not in the surrounding brain parenchyma in
both cases (Fig. 4).
DISCUSSION
The primary objective of this study was to determine if
injection of an oncolytic virus (i.e., a virus that can still
replicate in a relatively selective fashion in tumor cells)
into human brain that surrounds a resected malignant
TABLE 2: Serious adverse events
Adverse event Grade
a
Number of patients
Neuropathy–motor 3/4 2
Dyslexia, dyscalculia, dysgraphia 31
Headache 31
Diarrhea 32
Confusion 32
Hypertension 31
Decreased LOC 31
Hyponatremia 31
Abnormal PT 31
Thrombosis/embolism 31
Febrile neutropenia 3/4 1
Fever 31
Nausea 31
Vomiting 31
Fatigue 41
Abnormal SGPT 41
Ataxia 41
Hydrocephalus 31
Ten of the 24 patients experienced one or more serious adverse events.
a
The relationship to treatment was coded as unlikely for all serious adverse events.
FIG. 1. Serial coronal MRI scans with gadolinium
enhancement in a patient treated with ONYX-015.
This patient was initially treated with resection of
recurrent glioma and injection of ONYX-015 in August
2001. (A) The patientTs immediate postoperative scan
showed minimal evidence of enhancement at the
margin. However, 2 months later (10/08/2001),
evidence for increased enhancement at this margin
(B, yellow arrow) was observed, reflecting increased
blood–brain barrier breakdown, possible postoperative
gliotic reaction, and/or possible recurrence. However,
on subsequent monthly scans, such enhancement
gradually decreased, eventually returning to baseline
levels on 2/18/2002 (C). Yet again, on 8/6/2002 (D)
and 8/8/2002 (E), enhancement in the same region
returned (yellow arrows). (F) Again, by 1/28/2003,
such enhancement returned to baseline levels.
TRIALS
doi:10.1016/j.ymthe.2004.07.021
MOLECULAR THERAPY Vol. 10, No. 5, November 2004
960
Copyright C The American Society of Gene Therapy

glioma would be tolerated. We determined that the
treatment was well tolerated by all patients even at the
highest dose of ONYX-015 that was available.
This study was conducted as a phase I dose-escalation
trial in six different institutions that belonged to the New
Approaches to Brain Tumor Therapy (NABTT) CNS Con-
sortium. The major novelty in this trial is that for the first
time a tumor-selective virus that maintains replication
was injected into brain that surrounds a surgically
resected glioma. Previous trials with another oncolytic
virus (based on the HSV-1 viral mutant designated G207
and another HSV-1 mutant designated 1716) have tested
the viral vector via stereotactic intratumoral administra-
tion [24,26]. Other trials have involved injections of a
replication-defective adenovirus vector, and not of a
replication-conditional virus such as ONYX-015, to
deliver p53 or the herpes simplex virus thymidine kinase
gene [27–30]. Although these trials did not see evidence of
toxicity until doses of 10
12
viral particles, a very appro-
priate concern relates to the possibility that an oncolytic
virus may still provoke clinically significant damage to
brain if its action was not confined to a tumor. In fact,
reports have shown evidence of cerebral inflammation in
animals whose brains were inoculated with replication-
defective adenoviral vectors [31–34]. It has been suggested
that this may represent an immune-mediated reaction to
adenoviral gene products/proteins and/or antigens
released from dying tumor cells. By injecting this onco-
lytic adenovirus directly into human brain tissue that
surrounded a resected glioma, we have demonstrated that
ONYX-015 is unlikely to cause clinically significant
disease in humans at doses up to 10
10
pfu. We did not
detect clinical or radiologic evidence of neurologic or
systemic injury and the maximum tolerated dose was not
reached at 10
10
pfu. However, in two patients whose
recurrent tumors were available for analyses we did find
evidence of lymphocytic and plasma cell infiltrates that
would be relatively unusual for this patient population
FIG. 2. Kaplan–Meier survival curve for patients treated with ONYX-015 (n =
24). Kaplan–Meier survival curves for GBM (n = 17) and other histology (n =7)
patients treated with ONYX-015.
FIG. 3. Serial MRIs with representative findings of trial. In
(A), a left temporal lobe recurrence of a malignant glioma
was observed and confirmed intraoperatively. (B and C)
Immediate postoperative scans (following resection and
margin injection of 10
10
pfu of ONYX-015) reveal gross total
resection of tumor. Approximately 3 months later, probable
recurrence of glioma was visualized in resection cavity
infiltrating the brain. Histological findings from the re-
resection of this tumor are presented in Figs. 4A and 4B.
ARTICLE
doi:10.1016/j.ymthe.2004.07.021
MOLECULAR THERAPY Vol. 10, No. 5, November 2004
961
Copyright C The American Society of Gene Therapy

and that would be consistent with the aforementioned
animal studies.
No definite anti-tumor efficacy could be demonstrated
in this trial. All but one patient experienced progression of
disease, as determined by N25% increase in gadolinium
enhancement. One patient showed evidence for increased
enhancement approximately 1 year after treatment,
which decreased on a subsequent scan. This episode of
increased enhancement, which then decreased, had also
occurred postoperatively. The postoperative changes in
enhancement could be attributed to postoperative gliotic
reactions, but the changes that occurred 1 year later
would be more difficult to explain on this basis. Although
no histologic data are available, one can speculate that it
may represent an episode of transient breakdown of the
blood–brain barrier due to a waxing and waning inflam-
matory reaction to recurrent tumor and/or residual
replicative virus, in agreement with the histologic find-
ings of the recurrent tumors described above.
The finding that such an inflammatory event may
occur within a glioma that is recurring several weeks after
ONYX-015 injection calls into question the current
criterion of using an increase in gadolinium enhance-
ment to decide whether there is progression of a glioma
treated with such biologic agents. Although it still
remains likely that observed N25% increases in gadoli-
nium enhancement were due to tumor progression, we
cannot exclude that a very localized inflammatory
reaction within the injected tumor bed could have been
partially responsible for such radiologic images.
Only two patients displayed evidence of seroconver-
sion. Such a low number was likely due to the relatively
immunocompromised state of this patient population,
which is on steroid medication and has been treated
with radiation and chemotherapy, and to the injection
within the brain, an organ that is relatively immuno-
privileged.
Although survival and time to progression evidence
were relatively encouraging for four patients, three of
these had diagnoses other than glioblastoma multiforme.
It is known that patients with anaplastic astrocytomas
and oligodendroglioma display more favorable average
survival rates than patients with glioblastoma multiforme
and thus the result of this trial should be judged in this
context. It was encouraging, though, that one of the
patients with glioblastoma multiforme remained alive for
over a year after treatment.
In conclusion, this trial has shown the relative safety
of injection of ONYX-015 into brain surrounding a
resected malignant glioma. Further studies to determine
a maximum tolerated dose and potential for efficacy are
warranted.
PATIENTS AND METHODS
Approval of the trial was accomplished at each
institutionTs Institutional Review Board, in accord with
an assurance filed with and approved by the Department
of Health and Human Services. Informed consent was
obtained from each subject. The following NABTT
institutions participated: Massachusetts General Hospital
(Boston, MA, USA), Wake Forest University (Wake Forest,
NC, USA), Henry Ford Hospital (Detroit, MI, USA), Emory
University (Atlanta, GA, USA), Moffitt Cancer Center
(Tampa, FL, USA), and University of Alabama (Birming-
ham, AL, USA).
FIG. 4. Histological findings of gliomas resected months
after ONYX-015 injection. In (A), the histological findings
of the recurrent glioma from the patient presented in Fig.
3 are shown before ONYX-015 injection, with character-
istics consistent of malignant glioma. Approximately 3
months later, a new recurrence was resected. (B) The
extensive lymphocytic and plasmacytoid cell infiltrate in
this recurrence are shown. Similarly, (C) represents the
histological picture of a recurrent malignant glioma
before injection of 10
7
pfu of ONYX-015. Approximately
3 months later, this tumor was judged by gadolinium
enhancement to have recurred and reexcision was
performed. (D, white arrow) The presence of numerous
perivascular lymphocytic and plasma cell infiltrates is
indicated.
TRIALS
doi:10.1016/j.ymthe.2004.07.021
MOLECULAR THERAPY Vol. 10, No. 5, November 2004
962
Copyright C The American Society of Gene Therapy

Inclusion and Exclusion Criteria
Patients were 18 years or older and had to have a
histologically documented supratentorial malignant
glioma (glioblastoma multiforme, anaplastic astrocy-
toma, or anaplastic oligodendroglioma) that had pro-
gressed after initial external beam radiation therapy.
Radiation therapy had to have been between 5400 and
6700 cGy delivered in 180-to 200-cGy fractions.
Patients had to have recovered from toxicity of prior
therapy. An interval of at least 3 months had to have
elapsed since the completion of the most recent course
of radiation, while at least 3 weeks had to have elapsed
since the completion of a non-nitrosourea-containing
chemotherapy regimen and at least 6 weeks since the
completion of a nitrosourea-containing chemotherapy.
They may not have received more than two prior
chemotherapy regimens. They had to have been eligible
for resection of a portion of the recurrent tumor that
was at least 1 cm in greatest dimension. There must
have been no anticipated physical connection between
the postresection tumor cavity and the cerebral ven-
tricle. A Karnofsky performance status (KPS) of at least
60, a life expectancy of at least 3 months, and the
ability to provide informed consent were required.
Criteria for baseline organ function determined within
2 weeks of the start of treatment included the follow-
ing: an absolute neutrophil count z1500 mm
3
, platelet
count z100,000 mm
3
,creatinineV1.7 mg/dl, total
bilirubin V1.5 mg/dl, transaminases V4 times above
the limits of the institutional norm, PT and PTT V
upper limit of normal, and CD4 lymphocyte count
N200/Al. Finally, at the time of tumor resection, a
frozen biopsy confirmation of malignant glioma was
required.
Patients were excluded if they required immediate
excision because of impending neurological decline or
if a postsurgical connection between the resection
cavity and the ventricular system was anticipated.
Patients who had prior treatment of the tumor with
gene therapy, brachytherapy, radiosurgery, or implants
of polymers containing chemotherapeutic agents were
excluded. Any patient with the presence of an immu-
nosuppressive disorder (e.g., HIV infection) or iatro-
genic immunosuppression (with the exception of
corticosteroid use) was excluded. Patients with any
active infection (defined as a clinically diagnosed viral,
bacterial, or fungal infection that required active treat-
ment and caused oral temperature N38.58Cand/or
clinically significant leukocytosis) were excluded. Like-
wise, any viral syndrome clinically diagnosed within 2
weeks prior to treatment on this protocol led to
exclusion. To be included, patients had to have no
concurrent malignancy except curatively treated basal
or squamous cell carcinoma of the skin or carcinoma in
situ of the cervix and breast. Patients with prior
malignancies had to be disease-free for z5years.
Known diagnosis of Li-Fraumeni syndrome or known
germ-line defect in the p53 gene was grounds for
exclusion. Pregnant or lactating females were not
included. Women of child-bearing potential were
required to practice birth control for the duration of
the treatment. Men were advised to use barrier protec-
tion for the duration of treatment. This exclusion was
based on the potential risks of adenoviral encephalitis
to the fetus and newborn. Finally, patients with
gliomatosis cerebri were excluded.
Description of ONYX-015
ONYX-015 was manufactured under contract to Onyx
Pharmaceuticals, Inc., by MAGENTA Corp. (currently
BioReliance, Inc., Bethesda, MD, USA) in compliance
with Good Manufacturing Practice Regulations. It was
distributed to the individual trial sites by the Cancer
Therapy Evaluation Program/National Cancer Institute.
The purified virus was stored frozen below 608Cin
aliquots. Prior to release for clinical use, each lot was
tested to ensure that it met the following criteria: (1)
sterility, (2) bacterial endotoxin b10 EU/ml, (3) general
safety, (4) adenovirus titer in HEK293 cells for low-dose
vials N2

10
8
pfu/ml, (5) adenovirus titer in HEK293 cells
for high-dose vials N2

10
10
pfu/ml, (6) E1B deletion
confirmed by PCR assay, and (7) selective replication in
p53-deficient cells confirmed.
The product was formulated as a sterile viral solution
in Tris buffer (10 mM Tris, pH 7.41; 1 mM MgCl
2
; 150
mM NaCl; 10% glycerol). The product was supplied
frozen in a single-use, plastic screw-cap vial. Prior to
use, vials were thawed at room temperature. The appro-
priate dilution was admixed using aseptic procedures and
transported on ice to the operating room in a sealed
package. Injections were performed within a 4-h time
frame after the virus was thawed. The injection protocol
is described below (dose-escalation scheme). All injec-
tions were performed with a clinical lot, before the date
of expiration. Although a standard for replication-defec-
tive vectors is to provide the viral particle to plaque-
forming unit ratio and to provide doses as viral particles
per milliliter, this was not a required release criterion for
ONYX-015 and regulatory agencies did not require this
release criterion.
Treatment
Pretreatment testing. All pretreatment testing/evalua-
tions were conducted within 2 weeks of treatment and
consisted of: (1) a complete history (disease history and
prior oncologic therapies); (2) a physical examination
including vital signs (heart rate, respiratory rate, blood
pressure, temperature, and pulse), height and weight,
KPS, and neurologic exam; (3) laboratory exams includ-
ing CBC with differential and platelet count, serum
electrolytes (Na
+
,K
+
,Cl, bicarbonate), BUN, creatinine,
glucose, total protein, calcium, phosphorous, magne-
ARTICLE
doi:10.1016/j.ymthe.2004.07.021
MOLECULAR THERAPY Vol. 10, No. 5, November 2004
963
Copyright C The American Society of Gene Therapy

sium, AST, ALT, total bilirubin, alkaline phosphatase,
PT/PTT, CD3, CD4, CD8, total lymphocyte counts,
serum antibody to group C adenovirus (total and
neutralizing), serum pregnancy test within 72 h of
treatment (for all premenopausal women with child-
bearing potential), and urinalysis with microscopy; (4)
an EKG (12 lead); (5) a chest X-ray (PA and Lat) (within
6 weeks); (6) an MRI of the brain with and without
gadolinium.
ONYX-015 administration, schedule, and dose
escalation. Four cohorts of six patients each were
treated. The number of patients per cohort was selected
based on the estimate that between one-third and one-
half of patients may have a mutation in the p53 tumor
suppressor gene or a defect within the p53 tumor
suppressor pathway [1]. Each patient received 10
injections (volume of 100 Al per injection) into the
cavity wall after resection of the recurrent tumor. A
period of 4 weeks (28 days) was allowed to elapse after
treatment of the last patient at each dose level before
escalation to the next higher dose level. At the end of
each cohort study, the Data Safety Monitoring Com-
mittee (DSMC) evaluated administration methods, dos-
ages, and overall treatment plan. The DSMC was
composed of Drs. Chiocca and Barker (Massachusetts
General Hospital), Dr. Stuart Grossman (Johns Hopkins
University), Drs. James Zwiebel and Rick Kaplan (NCI/
CTEP), Dr. Philip Gutin (Memorial Sloan-Kettering),
and Dr. Michael Walker (NCI/NIH). The starting dose
of ONYX-015 consisted of 10
7
pfu inoculated into the
resected tumor cavity (as 10 single doses into 10
separate locations). Dose levels were successively esca-
lated by a factor of 10 in subsequent cohorts. The
selection of peritumoral injection sites was left to the
discretion of the operating neurosurgeon. Injections
were carried out using 25-gauge needles on tuberculin
syringes. The maximum depth thus did not exceed
1 cm.
Treatment schedule. On the first day, the patient was
admitted to the hospital. A craniotomy with resection of
the recurrent tumor was then performed. After tumor
resection was complete and an intraoperative diagnosis of
malignant glioma was rendered, free-hand injections of
100 Al of ONYX-015 virus were performed by the neuro-
surgeon into each of 10 sites in the wall of the resection
cavity. Separate tuberculin syringes were used for each
injection (10 total injections). The choice of injection site
was left to the judgment of the operating neurosurgeon,
but the sites had to be separated by a least 1 cm. Injections
were performed slowly, usually over a period of a few
minutes, to avoid spillage. The depth of injection did not
exceed 1 cm (length of the needle of the tuberculin
syringe). The sites were selected by the neurosurgeon to
avoid injections into adjacent motor or speech cortex or
the cerebral ventricle or spillage into the subarachnoid
space. After the injections, the wound was closed as per
routine. Postoperatively, the patients were admitted to the
Intensive Care Unit. Postoperative care followed standard
neurosurgical practice and it included a brain MRI F
gadolinium at day 3, followed by one at day 14, one at day
42, and then every 6 weeks after. A postoperative serum
sample for adenovirus antibodies was obtained on day 42.
Toxicity Assessment
A dose-limiting toxicity was defined as any one of the
following: (1) NIH Common Toxicity Criteria (CTC;
version 2.0) grade 4 toxicity for flu-like symptoms (fever,
fatigue, myalgia) attributed to ONYX-015 or (2) CTC
grade 3 toxicity for neurologic symptoms or for symptoms
in other organ systems lasting longer than 5 days and
attributed to ONYX-015. Because in the CTC (version 2.0)
the grading for neurosurgical oncologic complications,
such as cerebral edema, hydrocephalus, seizures, and
hemorrhages, was not available or was unclear, we devised
a supplemental table of neurosurgical toxicity criteria that
was followed by the six participating institutions in
addition to the CTC (Table 3).
TABLE 3: Supplemental table of toxicities for neurological/neurosurgical disease employed in trial
Grade of toxicity
CNS toxicity Grade I Grade II Grade III Grade IV
Hydrocephalus Asymptomatic
ventricular dilation
Ventriculomegaly
with headache
Ventriculomegaly with
severe headache, nausea,
and vomiting
Ventriculomegaly
requiring permanent
CSF drainage
Meningitis Asymptomatic
diffuse meningeal
enhancement on MRI
Mild signs of
meningeal irritation
(headache, photophobia)
Moderate signs of
meningeal irritation
(severe headache,
photophobia, vomiting,
nuchal rigidity)
As in Grade III with
altered mental status
(stupor or coma)
Edema Asymptomatic
edema on MRI
Focal edema on MRI with
corresponding new focal
neurological deficit
Diffuse edema on MRI
with corresponding new
neurological deficit
As in Grade III, but with
altered mental status
Seizures — 3 seizures or fewer per day N3 seizures in 1 day Status epilepticus
TRIALS
doi:10.1016/j.ymthe.2004.07.021
MOLECULAR THERAPY Vol. 10, No. 5, November 2004
964
Copyright C The American Society of Gene Therapy

Assessment of Response
Neurosurgical resections did not have to be gross total
resections for ONYX-015 injection to occur. The scan done
postoperatively was to assess the tumor configuration after
injection and evaluate the extent of tumor resection. It
was not used to determine response. However, it was this
postoperative scan, not the scan obtained prior to surgery,
that was used as the bbaseline scan.Q The scan obtained at
day 42 was compared with the postoperative scan to
determine response. Responses to ONYX-015 were deter-
mined at this 42-day time and every 6 weeks thereafter.
The area of contrast enhancement was determined on
each slice and the total was then multiplied by slice
thickness to obtain a total volume. Response criteria were:
(1) Complete Response, complete disappearance of all
tumor on MR/CT scan and not taking glucocorticoids,
with a stable or improving neurologic exam for at least 6
weeks; (2) Partial Response, greater than or equal to 50%
reduction in tumor size on volumetric MR/CT scan, on a
stable or decreasing dose of glucocorticoids, with a stable
or improving neurologic exam for at least 6 weeks; (3)
Progressive Disease, progressive neurologic abnormalities
not explained by causes unrelated to tumor progression
(e.g., anticonvulsant or corticosteroid toxicity, electrolyte
abnormalities, hyperglycemia) or a greater than 25%
increase in the volume of the tumor by MR/CT scan; if
neurologic status on a stable or increasing dose of steroids
deteriorated or if new lesions appeared on serial MR/CT,
the patient was removed from the study and became
eligible for other therapies; (4) Stable Disease, a patient
whose clinical status and MR/CT scan volumetrics did not
meet the criteria for Partial Response or Progressive
Disease.
Endpoints
The main endpoint of the study was safety. Patients were
evaluated for toxicity if they received at least one dose of
ONYX-015. An adverse event was defined as any unfav-
orable or unintended sign (including an abnormal labo-
ratory finding), symptom, or disease temporally
associated with the use or procedure regardless of whether
it was considered related to the medical procedure. The
investigator documented his/her opinion of the relation-
ship of the event to the study treatment (unrelated,
unlikely, possible, probable, or definite). Serious adverse
events were defined as an experience that was fatal or life-
threatening, was disabling, or required inpatient care.
Efficacy of treatment was assessed as survival time and
time to progression of disease, both measured from the
first day of ONYX-015 treatment. As previously noted,
progressive disease was defined as progressive neurologic
abnormalities not explained by causes unrelated to tumor
progression (e.g., anticonvulsant or corticosteroid toxic-
ity, electrolyte abnormalities, hyperglycemia) or a greater
than 25% increase in the volume of the tumor by MRI/CT
scan. Responses were determined at the 42-day time and
every 6 weeks thereafter until documented tumor pro-
gression or another treatment was started.
Statistical Considerations
Survival distributions were estimated using the product
limit method. The analysis was intention-to-treat and
included all eligible patients. SAS software version 9 (SAS
Institute, Cary, NC, USA) was used to perform analyses.
A
CKNOWLEDGMENTS
This trial was supported through NCI/CTEP and NABTT grants.
RECEIVED FOR PUBLICATION MAY 4, 2004; ACCEPTED JULY 19, 2004.
REFERENCES
1. Kleihues, P., et al. (2002). The WHO classification of tumors of the nervous system.
J. Neuropathol. Exp. Neurol. 61: 215 – 225; discussion 226–229.
2. Prados, M. D., and Levin, V. (2000). Biology and treatment of malignant glioma.
Semin. Oncol. 27: 1 – 10.
3. Prados, M. D. (2000). Future directions in the treatment of malignant gliomas with
temozolomide. Semin. Oncol. 27: 41 – 46.
4. Yung, W. K. (2000). Temozolomide in malignant gliomas. Semin. Oncol. 27: 27 – 34.
5. Castro, M. G., et al. (2003). Current and future strategies for the treatment of
malignant brain tumors. Pharmacol. Ther. 98: 71 – 108.
6. Chiocca, E. A. (2002). Oncolytic viruses. Nat. Rev. Cancer 2: 938 – 950.
7. Kirn, D., Martuza, R. L., and Zwiebel, J. (2001). Replication-selective virotherapy for
cancer: biological principles, risk management and future directions. Nat. Med. 7:
781 – 787.
8. Bischoff, J. R., et al. (1996). An adenovirus mutant that replicates selectively in p53-
deficient human tumor cells. Science 274: 373 – 376.
9. Edwards, S. J., et al. (2002). Evidence that replication of the antitumor adenovirus
ONYX-015 is not controlled by the p53 and p14(ARF) tumor suppressor genes. J. Virol.
76: 12483 – 12490.
10. Nemunaitis, J., et al. (2003). Pilot trial of intravenous infusion of a replication-selective
adenovirus (ONYX-015) in combination with chemotherapy or IL-2 treatment in
refractory cancer patients. Cancer Gene Ther. 10: 341 – 352.
11. Hamid, O., et al. (2003). Phase II trial of intravenous CI-1042 in patients with metastatic
colorectal cancer. J. Clin. Oncol. 21: 1498 – 1504.
12. Makower, D., et al. (2003). Phase II clinical trial of intralesional administration of the
oncolytic adenovirus ONYX-015 in patients with hepatobiliary tumors with correlative
p53 studies. Clin. Cancer Res. 9: 693 – 702.
13. Hecht, J. R., et al. (2003). A phase I/II trial of intratumoral endoscopic ultrasound
injection of ONYX-015 with intravenous gemcitabine in unresectable pancreatic
carcinoma. Clin. Cancer Res. 9: 555 – 561.
14. Warren, R. S., and Kirn, D. H. (2002). Liver-directed viral therapy for cancer p53-
targeted adenoviruses and beyond. Surg. Oncol. Clin. North Am. 11: 571 – 588, vi.
15. Reid, T., et al. (2002). Hepatic arterial infusion of a replication-selective oncolytic
adenovirus (dl1520): phase II viral, immunologic, and clinical endpoints. Cancer Res.
62: 6070 – 6079.
16. Vasey, P. A., et al. (2002). Phase I trial of intraperitoneal injection of the E1B-55-kd-
gene-deleted adenovirus ONYX-015 (dl1520) given on days 1 through 5 every 3
weeks in patients with recurrent/refractory epithelial ovarian cancer. J. Clin. Oncol. 20:
1562 – 1569.
17. Reid, T., et al. (2001). Intra-arterial administration of a replication-selective adenovirus
(dl1520) in patients with colorectal carcinoma metastatic to the liver: a phase I trial.
Gene Ther. 8: 1618 – 1626.
18. Nemunaitis, J., et al. (2001). Intravenous infusion of a replication-selective adenovirus
(ONYX-015) in cancer patients: safety, feasibility and biological activity. Gene Ther. 8:
746 – 759.
19. Mulvihill, S., et al. (2001). Safety and feasibility of injection with an E1B-55 kDa gene-
deleted, replication-selective adenovirus (ONYX-015) into primary carcinomas of the
pancreas: a phase I trial. Gene Ther. 8: 308 – 315.
20. Nemunaitis, J., et al. (2001). Phase II trial of intratumoral administration of ONYX-015,
a replication-selective adenovirus, in patients with refractory head and neck cancer.
J. Clin. Oncol. 19: 289 – 298.
21. Lamont, J. P., Nemunaitis, J., Kuhn, J. A., Landers, S. A., and McCarty, T. M. (2000). A
prospective phase II trial of ONYX-015 adenovirus and chemotherapy in recurrent
squamous cell carcinoma of the head and neck (the Baylor experience). Ann. Surg.
Oncol. 7: 588 – 592.
ARTICLE
doi:10.1016/j.ymthe.2004.07.021
MOLECULAR THERAPY Vol. 10, No. 5, November 2004
965
Copyright C The American Society of Gene Therapy

22. Khuri, F. R., et al. (2000). A controlled trial of intratumoral ONYX-015, a selectively-
replicating adenovirus, in combination with cisplatin and 5-fluorouracil in patients with
recurrent head and neck cancer. Nat. Med. 6: 879 – 885.
23. Ganly, I., et al. (2000). A phase I study of Onyx-015, an E1B attenuated adenovirus,
administered intratumorally to patients with recurrent head and neck cancer. Clin.
Cancer Res. 6: 798 – 806.
24. Markert, J. M., et al. (2000). Conditionally replicating herpes simplex virus mutant,
G207 for the treatment of malignant glioma: results of a phase I trial. Gene Ther. 7:
867 – 874.
25. Papanastassiou, V., et al. (2002). The potential for efficacy of the modified (ICP
34.5()) herpes simplex virus HSV1716 following intratumoural injection into human
malignant glioma: a proof of principle study. Gene Ther. 9: 398 – 406.
26. Rampling, R., et al. (2000). Toxicity evaluation of replication-competent herpes simplex
virus (ICP 34.5 null mutant 1716) in patients with recurrent malignant glioma. Gene
Ther. 7: 859 – 866.
27. Trask, T. W., et al. (2000). Phase I study of adenoviral delivery of the HSV-tk gene and
ganciclovir administration in patients with current malignant brain tumors. Mol. Ther.
1: 195 – 203.
28. Smitt, P. S., Driesse, M., Wolbers, J., Kros, M., and Avezaat, C. (2003). Treatment of
relapsed malignant glioma with an adenoviral vector containing the herpes simplex
thymidine kinase gene followed by ganciclovir. Mol. Ther. 7: 851 – 858.
29. Germano, I. M., Fable, J., Gultekin, S. H., and Silvers, A. (2003). Adenovirus/herpes
simplex-thymidine kinase/ganciclovir complex: preliminary results of a phase I trial in
patients with recurrent malignant gliomas. J. Neurooncol. 65: 279 – 289.
30. Lang, F. F., et al. (2003). Phase I trial of adenovirus-mediated p53 gene therapy for
recurrent glioma: biological and clinical results. J. Clin. Oncol. 21: 2508 – 2518.
31. Dewey, R. A., et al. (1999). Chronic brain inflammation and persistent herpes
simplex virus 1 thymidine kinase expression in survivors of syngeneic glioma treated
by adenovirus-mediated gene therapy: implications for clinical trials. Nat. Med. 5:
1256 – 1263.
32. Boviatsis, E. J., et al. (1994). Gene transfer into experimental brain tumors mediated by
adenovirus, herpes simplex virus, and retrovirus vectors. Hum. Gene Ther. 5: 183 – 191.
33. Byrnes, A. P., Rusby, J. E., Wood, M. J., and Charlton, H. M. (1995). Adenovirus gene
transfer causes inflammation in the brain. Neuroscience 66: 1015 – 1024.
34. Bhat, N. R., and Fan, F. (2002). Adenovirus infection induces microglial activation:
involvement of mitogen-activated protein kinase pathways. Brain Res. 948: 93 – 101.
TRIALS
doi:10.1016/j.ymthe.2004.07.021
MOLECULAR THERAPY Vol. 10, No. 5, November 2004
966
Copyright C The American Society of Gene Therapy