Download Intraperitoneal Recombinant «

•
••:•'•
W äf.:v
[CANCER
RESEARCH
45, 4447-4453,
September 1985]
Intraperitoneal Recombinant «-Interferon for "Salvage" Immunotherapy ¡n
Stage III Epithelial Ovarian Cancer: A Gynecologic Oncology Group Study1
Jonathan S. Berek,2 Neville F. Hacker, Alan Lichtenstein, Tyler Jung, Gelsa Spina, Reba M. Knox, Jeanine Brady,
Teri Greene, Lupe M. Ettinger, Leo D. Lagasse, Eric M. Bonnern, Robert J. Spiegel,
and Jacob Zighelboim
Division of Gynecologic Oncology, the Departments of Obstetrics and Gynecology [J. S. B., N. F. H., T. J., R. M. K., T. G., L. M. E., L D. L], Medicine, and Microbiology
and Immunology [A. L., C. S., J. Z.], the UCLA School of Medicine, and Jonsson Comprehensive Cancer Center [J. S. B., N. F. H., L. D. L], Los Angeles, California 90024;
and the Schering Corporation ¡J.B., £.M. B., R. J. S.], Kenilworth, New Jersey 07033; and the Gynecologic Oncology Group [J. S. B., N. F. H., L. M. £.,L. D. L.],
Philadelphia, Pennsylvania 19107
ABSTRACT
Fourteen patients with persistent epithelial ovarian cancer
documented at second look laparotomy after combination chem
otherapy were treated with 146 cycles of a-recombinant interferon (r!FN-a2) administered i.p. The initial dose was 5 x 106
units which was escalated weekly to 50 x 106 units over 4 weeks
and then continued weekly for a total of 16 weeks. Eleven
patients underwent surgical réévaluation
after therapy which
confirmed four pathological complete responses (36%), one par
tial response (9%), and disease progression in six patients (55%).
Five of seven patients (71%) with residual tumor <5 mm had a
surgically documented response, whereas there was no re
sponse in the four patients whose tumors were >5 mm. Three
patients were évaluablefor clinical response only: one patient
who refused surgery had a complete clinical response with total
resolution of ascites; one had stable disease; and one had
disease progression. Fever >38°C was seen in 58%, fever
>39.0°C was seen in 18%, vomiting in 37%, abdominal pain
was reported in 22%, and one patient had infectious peritonitis.
Peripheral white blood cell counts and i.p. washings were ob
tained pretreatment and on days 1, 3, and 7 after treatment.
While there was no consistent alteration in peripheral white blood
cell counts, the numbers of i.p. monocytes and lymphocytes
showed a significant boost on day 1 after each dose of rlFN-a2.
Natural killer lymphocyte cytotoxicity was elevated in the i.p.
cavity fluid obtained from most patients on day 1 after treatment,
while blood natural killer lymphocyte cytotoxicity values showed
considerable variability. Pharmacokinetic studies show that i.p.
levels of rIFN-az were 30-1000 times blood levels. rlFN-«2i.p.
may act by increasing concentrations of drug and augmenting
regional host cells in patients with minimal residual ovarian
cancer.
INTRODUCTION
IFNs3 are a diverse group of locally active, naturally occurring,
regulatory glycoproteins which are currently undergoing exten
sive clinical testing as antineoplastic agents (1-9). Their biological
1Supported by grants from the Schering Corporation, the Concern Foundation,
and The National Cancer Institute (Grant CA 12800, NIH); a USPHS CRC Grant
RR865; and Gynecologic Oncology Group Grant CA 13630-13.
2 To whom requests for reprints should be addressed, at Division of Gynecologic
Oncology, Department of Obstetrics and Gynecology, UCLA School of Medicine,
Los Angeles, CA 90024.
'The abbreviations used are: IFNs, interferons; rlFN-a2, a-recombinant inter
feroni NK, natural killer lymphocyte cytotoxicity; MRD, minimal residual disease;
BRM, biological response modifier.
Received 3/18/85; revised 6/5/85; accepted 6/7/85.
CANCER
RESEARCH
activities are diverse, and in addition to their ability to modify
immunoregulatory systems (1, 2) they possess direct antineo
plastic activity. These agents alter tumor cell function and struc
ture and induce cellular inhibition of replication of many viruses
(1)Therapeutic responses have been observed in a variety of
tumor systems (1-5, 7, 8). Human a-interferon, a specific antigenie type produced mainly by leukocytes (2), is now produced
in highly purified, large quantities because genetic recombinant
DNA technology has permitted the production of specifically
cloned products (8).
In our experience patients treated with combination chemo
therapy for epithelial ovarian cancer and in clinical remission have
a 60-70% rate of persistent disease at second look laparotomy
(10). Therefore there is a need for improved primary or secondary
therapies in order to improve survival and cure rates.
One approach which has been suggested is to treat the
residual disease with the i.p. administration of antitumor agents
(11-13). Complete and partial responses have been surgically
documented in response to the i.p. administration of Corynebacterium parvum as second line therapy in epithelial ovarian cancer
patients (11,12). These responses have been documented only
in patients with MRD (;.e., <5 mm maximum diameter of the
residual tumor) at second look operations. However, since this
BRM is relatively toxic when administered i.p., BRMs which
induce a less intense peritonitis might be more practical and
effective (13,14). The i.p. instillation of BRMs for ovarian cancer
has clinical appeal because of the potential for increasing the
drug concentration at the tumor site and decreasing systemic
toxicity(11-15).
This report is a trial of the i.p. administration of recombinant
(rlFN-«2)in patients with persistent epithelial ovarian cancer
documented at second look laparotomy. Evaluation of efficacy,
toxicity, pharmacokinetics, and host cell parameters are pre
sented.
MATERIALS AND METHODS
Patients. A total of 14 patients with persistent epithelial ovarÃ-ancancer
were treated with 146 cycles of ¡.p.r!FN-a2. All patients were évaluable
for clinical response and toxicity. Eleven patients were surgically reevaluated after therapy. Serial r!FN-a2 levels, studies of host cell populations,
and lymphocyte cytotoxicity assays were performed on peritoneal wash
ings and peripheral blood.
Patients ranged in age from 42 to 66 years with a mean age of 53.4
years. All patients were classified as International Federation of Gyne
cology and Obstetrics stage III. No patient had documented extraperitoneal disease at the initiation of immunotherapy.
VOL. 45 SEPTEMBER
1985
4447
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1985 American Association for Cancer Research.
INTERFERON
¡.p.IN OVARIAN
Residual tumor prior to rlFN-a2 treatments ranged from microscopic
disease (positive cytology only) to a 4- x 4-cm palpable pelvic mass. Six
patients (43%) had only positive peritoneal cytology (classes IV and V),
two patients (14%) had measurable disease <5 mm in diameter, one
patient (7%) had disease 5 to 15 mm in diameter, and five patients (36%)
had disease >15 mm. Papillary serous cystadenocarcinorna was the
histological type in 11 of the 14 patients. Two patients had grade I
tumors, six had grade II lesions, and six had grade III lesions.
Cytology of peritoneal fluids were graded as follows: class I or II,
negative; class III, equivocal or suspicious; and class IV or V, positive for
frankly malignant cells.
Prior to i.p. rlFN-«2therapy each patient had undergone one to three
exploratory laparotomies to resect tumor, including a total abdominal
hysterectomy, bilateral salpingooophorectomy,
as well as 5 to 13 (me
dian, 9) cycles of combination chemotherapy with c/s-platinum, doxorubicin, and cyclophosphamide. Two patients previously had received ex
ternal pelvic radiation therapy. Immediately prior to entry onto this study,
a second look minilaparotomy was performed at which time two Tenckhoff catheters were implanted using standard techniques (11): the tip of
one catheter was directed toward the pelvis; and the tip of the second
catheter was directed cephalad up the right paracolic gutter. An operation
to evaluate response to rlFN-«2were performed 1 to 6 weeks following
the final course of r!FN-a2 as described previously (10,16). The protocol
was approved by the Human Subject Protection Committee of the
University of California, UCLA.
Dialysis fluid was instilled into the peritoneal cavity (Dyalasate with
1.5% dextrose) after being warmed to 37°Cand was supplemented with
sodium heparin (500 units/liter; Panheparin; Abbott Laboratories).
i.p. Therapy. After catheter placement several peritoneal lavages were
performed using 1 to 2 liters of dialysis fluid until peritoneal rinses were
free of blood. Three consecutive blood free peritoneal lavages were
required before the first rlFN-«2treatment. Catheters were flushed every
other day with 10 ml of heparinized saline except for treatment days in
which only saline solution was used.
Once rlFN-«2treatment began peritoneal lavages were performed on
days 2 and 7. Aliquots obtained prior to each lavage were examined
cytologically for malignant cells, and total WBC differential counts and
immunological assays of peritoneal host cells were performed.
Interferon Administration. Human rlFN-«2(SCh 30500, Schering
Corporation, Kenilworth, NJ) was administered i.p. in 250 ml normal
saline after 1750 ml of dialysis solution were instilled i.p. via Tenckhoff
catheters.
Ten patients in the first dose schedule of the study received an
escalating weekly dosage of r!FN-a2. The dosage of rlFN-a2 was initially
5 x 10" units, which was escalated to 10 x 10" and then 20 x 10" units
and then followed by five consecutive courses at 50 x 106 units. The
second dose schedule initially called for patients to be infused with rlFN«2three times a week (every other day) at 50 x 106 units. This regimen
was to be followed every other week for 16 weeks. However, because
of toxicity the second dosage schedule was modified to 50 x 10" units
a week for 8 more weeks.
Patients were premedicated with 25 mg diphenhydramine
hydrochlo-
ride and 650 mg acetaminophen p.o. The former was discontinued during
the second dose schedule because no major allergic reactions from rlFN«2were observed and the patients complained of somnolence. Acetami
nophen was given every 4 h for the duration of the hospital visit (usually
24 h) and was then taken by patients upon discharge as needed.
When 24 h had elapsed after rlFN-«2treatment, 2 liters of warmed
dialysis fluid were instilled into the peritoneal cavity and recovered for
cytological and immunological assays. The exposed ends of the Tenck
hoff catheters were resealed aseptically.
Response Criteria. Response was defined as follows: complete re
sponse, no evidence of disease at the completion of therapy, including
the absence of malignant cells in the peritoneal fluid cytological evaluation
in those patients whose cytologies had been positive; partial response,
the presence of a positive cytology only at the completion of therapy in
CANCER RESEARCH
CANCER
those patients who had had macroscopic disease, or the regression of
macroscopic disease by at least 50% in tumor dimension; stable disease,
no change in the positive cytology and/or the tumor dimension; progres
sive disease, evidence of tumor growth during or at the completion of
treatment.
Toxicity. Toxicity was monitored and graded according to the criteria
outlined by the Gynecologic Oncology Group (15) and modified to adapt
to the expected effects of rlFN-a2. rlFN-«2routinely causes fevers which
at times are above 39°Cand this was not considered a toxicity requiring
dose reduction. For rlFN-«2dose-limiting toxicity is usually a cumulative
one based on the patients' general condition and symptoms, e.g., fatigue
and malaise, rather than objective laboratory parameters (1, 2).
Toxicities which may be encountered by the patients receiving rlFN<*2include: (a) fatigue and malaise. If a subject experienced fatigue to the
extent that she was unable to ambulate, the rlFN-«2was held. One to 2
weeks after the patient was again ambulatory, the rlFN-a2 was resumed
on the regular schedule at the last maximum tolerated dosage, or
reduced by 25%, whichever dose was higher; (ù) weight loss. The
guidelines listed below were followed for weight loss believed to be
solely induced by the rlFN-«2(for >15 pounds reduce dosage by 50%;
for >20 pounds, hold treatment); (c) abdominal pain if the pain was
intermittent over a time period of <24 h and required no analgesics, it
was considered mild (grade I), if it lasted >24 h and necessitated p.o.
analgesics which relieved the pain it was grade II, and severe (grade III)
pain included that which was continuous weekly and not responsive to
p.o. analgesics; (d) chills: grade I, those that lasted <10 min at a time
and did not restrict patient activity in any way; grade II, those that lasted
>10 but <30 min at a time and did not confine the patient to bed; and
grade III, rigorous shivering that required the patient to stay in bed and
that lasted continuously for a period of >30 min; (e) headache: grade I,
those lasting <24 h relieved by p.o. analgesics; grade II, those that
lasted >24 h and <48 h and are temporarily relieved by p.o. analgesics;
and grade III, those lasting >48 h and temporarily relieved by p.o.
analgesics; (f) temporary anorexia: grade I, <24 h; grade II, >24 h and
<72 h; and grade III, >72 h; (g) pruritis: grade I, <24 h; grade II, >24 h
and <48 h; and grade III, >48 h.
Blood Lymphocyte Analysis. Ten ml of blood were drawn before
treatment for base-line analysis and one to three times/week during
treatment. Samples were assayed for differential WBC, lymphocyte
populations, and cytotoxic activity.
Peritoneal Cell Analysis. Peritoneal lavage fluid specimens were
obtained two to three times/week during rlFN-«2treatment. An aliquot
of 50 ml of fluid was examined cytologically for malignant cells. The
remaining dialysis fluid was centrifugea at approximately 700 rpm for 5
min and resuspended with minimal essential media. Total WBC/ml were
counted, Cytoprep slides were made, and differential cell counts/ml were
performed.
Separation of peritoneal cells was performed with a FicolhHypaque
density gradient as described previously (18). Specimens which had high
concentrations of macrophages were treated with GAF iron fillings and
removed with a magnet (19). Differential WBC were performed. Lympho
cyte populations were isolated and a specimen was assayed by the
UCLA Clinical Immunological Research Laboratory for lymphocyte cytctoxicity.
Natural Killer Cell Activity. A constant number of "Cr-labeled target
cells (1 x 104) was mixed with varying numbers of lymphocyte effector
cells at an effectortarget ratio of 25:1. Cultures were incubated in V-well
mterotiter plates for 3 h at 27°C. Supernatant was collected and 51Cr
release was assayed by -y-émissionaccording to the method of Jondal
ef al. (20). The percentage of lysis was calculated as
[cpm of target + effector]
- [cpm of target alone]
% of lysis
Addition
[cpm of target + detergent (Triton X)]
- [cpm of target alone]
of detergent
VOL. 45 SEPTEMBER
to targets evaluated
x 100
maximal release (always
1985
4448
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1985 American Association for Cancer Research.
INTERFERON
Lp. IN OVARIAN
greater than 85% of incorporated cpm) while culture of targets alone
served as background release (always less than 20% of incorporated
cpm). The myeloid line K562 served as the target cell of the NK assay.
Neutralizing Antibodies. Frozen serum samples from each patient
were obtained before and after treatment and were sent to the Schering
Corporation Clinical Laboratories to assay for generation of antibodies
specific for the rIFN-az. The procedure for detection of anti-IFN-aj
antibody in human sera is a radioimmunological technique to screen for
antibodies as described previously (21).
Pharmacokinetics.
From each patient 10 ml of blood were drawn
prior to rIFN-az (50 x 106) and 24 h afterward. Additionally five patients
were studied after the administration of a rlFN-«2treatment at time 0
(pretreatment = base line), 4, 8, 12, 20, and 24 h from initiation of the
i.p. infusion. Blood samples were centrifugea and subsequent frozen
serum specimens were sent to the Schering Corporation Clinical Labo
ratories to determine rIFN-az levels. The method for measuring rIFN-az
levels utilized a human foreskin diploid cell line, the murine EMC virus,
and a crystal violet stain to visually detect cytopathic effects as described
(22).
Approximately 50 ml of dialysis fluid were obtained from peritoneal
lavages at pretreatment (time = 0) and at 4 and 24 h following 50 million
units of rIFN-az infused at time 0. Additional peritoneal fluid samples (48
and 72 h) were obtained by peritoneal lavages consisting of approxi
mately 1 liter of dialysis fluid. Specimens were frozen immediately and
sent to the Schering Corporation Clinical Laboratories to determine rlFN<i2levels.
RESULTS
CANCER
and progressed, one developed a 2- x 2-cm retroperitoneal tumor
nodule which was removed. After whole abdominal radiation
treatment she remained clinically free of disease for 12 months
before abdominal recurrence. The other patient with microscopic
disease was found to be free of disease by laparoscopy following
her first eight cycles of treatment, but progressive abdominal
disease was documented at the completion of the protocol by
laparotomy where 1-2 mm studding on the small bowel mesen
tery was documented.
Aliquots of peritoneal fluid were obtained weekly during the
course of therapy and examined cytologically for malignant cells
(Chart 1). The six patients who were clinically free of disease
and evaluated surgically at the completion of therapy retained
evidence of malignant cells throughout treatment (class IV or V
cytology), whereas those four patients who had surgically doc
umented complete responses developed negative cytology by
week 8 of treatment. This corresponds to the week after the
patients had received their fourth treatment cycle at the maxi
mum dose of 50 x 106 units/week.
Of the three patients who were only clinically évaluable,one
had a complete resolution of ascites during treatment and re
mains clinically free of disease 16 months posttreatment. One
patient had stable disease during treatment and remains stable
after 4 months with 2 cm disease palpable in the rectovaginal
septum. The third had a 3-4-cm mass palpable in the rectovaginal septum progress to 5-6-cm disease after eight treat
ments.
Response
Toxicity
Four of 11 surgically evaluated patients demonstrated com
plete responses to rlFN-<v2treatment. Tumor disappearance was
documented by a negative reassessment laparotomy and at
least six consecutive negative (class I or II) peritoneal cytological
washings at the completion of rlFN-«2therapy. Three of the
complete responders entered the program with microscopic
disease as documented by positive cytology only, and the other
had 5-10 residual tumor nodules, approximately 1-2 mm in
diameter in the pelvic peritoneum. All four patients remain alive
without recurrence from 5 to 14 months posttreatment.
An additional patient had a partial response. This patient began
the study with 6-10 peritoneal tumor nodules of 2-3-mm tumors
in the pelvic cul-de-sac. Upon reassessment laparotomy the
patient had no macroscopic evidence of disease, but a specimen
for cytological evaluation obtained from the upper abdomen was
positive for persistent malignant cells. Surgically documented
responses by residual disease are listed in Table 1.
Six surgically evaluated patients had disease progression dur
ing treatment. Four of these patients entered treatment with
intraabdominal tumor nodules >1.5 cm in diameter, and two
patients had only microscopic disease following optimal debulking. Of the two patients who entered with microscopic disease
Objective and subjective systemic reactions were monitored
throughout the study in all patients (Table 2). Of the 146 courses
monitored for subjective side effects, fatigue was noted in 65%,
chills in 53%, temporary anorexia in 39%, headaches in 38%,
nausea and/or vomiting in 37%, diarrhea in 27%, abdominal pain
in 22%, and total body pruritis in 12%. Other common complaints
included "abnormal" taste, "dry" mouth, back pain, and muscle
aches, all of which were mild to moderate.
Vital signs were monitored during the first 24 h after treatment
in 132 courses. Fever (>38°C) was documented in 77 courses
Non-responders
I
ffiII
. n-
Responders
Itti
i-
IFN Dose*5
t
LAP
IO 20 SOxIcf uiits •
Table 1
Surgically documented response by residual disease
Maximum
diameterCytology
positive
00
of
patients5a
2»4
1
04Partial0
11ResponseComplete3
CANCER
RESEARCH
2
3
4
5
6 7
Time
(«neks)
8
9
IO II
I2
I3 I4 I5
I6
Chart 1. Plot of mean of the cytologicalclasses of peritonealwashings by grade
for responders and nonresponders in a group of 11 surgically évaluable
patients
treated with i.p. rlFN-.»2.
Cytological grades I and II are negative, grade III is
equivocal, and grades IV and V are frankly positive for malignant cells. Following
two baseline(time = 0) cytological assays, specimens were obtained weekly while
on treatment. The maximum dose of 50 x 10* units r!FN-a2was reached by week
4 and repeated weekly thereafter. LAP, laparotomy.
1
100
01%60
0
45
* Five of seven (71%) of patients with <5 mm disease responded to therapy.
<5 mm
>5mm
TotalNo.
I
VOL. 45 SEPTEMBER
1985
4449
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1985 American Association for Cancer Research.
INTERFERON
i.p. IN OVARIAN
CANCER
Table 2
Toxicity of patients receiving i.p.
% of patients with side effects
Leukopenia0Day
Grade of
Toxicity1.
Paina38°C:
Fever8
40
Mild
>39°C:17
II. Moderate
>40°C: 111
III. SevereAbdominal
Total
58
and
vomiting21
Chills19
anorexia31
131
1347
Day
794
16
10
33
1Malaise30 2Headache3
36
16
16Diarrhea19 8Pruritis6 6Temporary7
22
24
1Nausea
1Anemia0 5Day 4
26
9
1Day
22
53
36
65
38
37
12
27
39
58
62
13
* Fever assessed in 132 of 146 cycles.
" Anemia and leukopeniaassessed in 140 of 146 cycles.
(58%). The degree of fever is listed in Table 2. Response could
not be correlated with the fever because both nonresponding
and responding patients had systemic temperature evaluations.
A slight increase in pulse without tachycardia was measured in
some patients during this first 24-h period; however, in all cases
pulse rates returned to normal values within 24 h. Respiration
and blood pressure remained normal during treatment.
Hematological analysis in 140 courses was performed pre
treatment and 24 h and 7 days after treatment. Base-line (pre
treatment = time 0) blood hemoglobin levels showed only 5%
anemia, but by day 1, 31 % had grade I anemia, 22% had grade
II anemia, and 5% had grade III anemia. By day 7, 34% had
grade I toxicity, 24% had grade II toxicity, and 4% had grade III
toxicity. With base-line (pretreatment = time 0) values of leuko
penia in only 1% of patients, leukopenia was noted in 36% of
cases on day 1 and 13% of cases on day 7. In 90% of the
courses the total WBC on day 1 was less than pretreatment
value, but only 36% were <3000/ml (10% grade III), and those
values returned to normal by day 3 in most cases. Additionally
in 88% of these courses the WBC on day 7 was greater than
the WBC on day 1, with only 13% having WBC <3000/ml on
day 7. The mean base-line WBC for the 14 treated patients was
5350/ml. This level decreased slightly during treatment with
mean values of 4400 on day 1 and 4900/ml on day 7. WBC for
responders and nonresponders were not significantly different.
Blood chemistry tests performed every week of treatment
revealed no elevation of liver enzymes, renal function tests, or
other abnormalities.
The dose limiting toxicity occurred in the second dose sched
ule phase of the study during which patients received 50 x 106
units rlFN-a2 three times every other week, but after two cycles
persistent moderate toxicity (fatigue, abdominal pain, and head
aches) required that the treatment schedule be modified so that
the rlFN-a2 was given weekly thereafter.
In an additonal two patients Tenckhoff catheters were inserted
with the intention of treating the patients with i.p. r!FN-o2. One
patient developed a clotted catheter and decided not to continue
the study, and the catheter was removed prior to any r!FN-a2
treatment. Another patient had a catheter placed at the time of
laparotomy but developed a postoperation thrombophlebitis and
was thus excluded from the protocol.
Immunological
Results
Peripheral Blood. Differential counts revealed that lympho
cytes decreased each cycle for a mean base-line value of 1530/
ml to 550/ml on day 1 and then rapidly returned to near baseCANCER
100r
•Total
A Mono
o Lymph
•Poly
IE 50
r
0
1
"""
;>
"3
4
5
6
7
Time (days)
Chart 2. Fluid WBC of patients treated with i.p. r!FN-a2 are presented. After
three baseline values (pretreatment = time 0), specimens were obtained on days
1, 3, and 7 after most weekly treatments. Day 7 also represents the time immedi
ately prior to the subsequenttreatment. Valuesrepresent the mean integrated sum
of all treatment cycles (oars, SD) showing a rapid rise in total WBC, lymphocytes,
and mononuclear cells by day 1 after treatment, and a gradual return to below
baselineby day 7. Total WBC are plotted, as well as polymorphonuclearleukocytes
(Poly),lymphocytes, and monocytes.
line levels of 1300/ml by day 7. Blood monocytes and polymor
phonuclear leukocytes remained relatively constant throughout
treatment.
Peritoneal Fluid. The mean base-line peritoneal WBC for
treated patients was 48,100/ml. A consistent increase was seen
in the WBC by day 1 in all patients (83,100/ml), a mean value
which returned to levels slightly below base line by day 7, i.e.,
22,000/ml (Chart 2). The absolute numbers of peritoneal mono
cytes and lymphocytes increased on day 1 of each cycle after
treatment and returned to base-line values by day 7. Monocyte
numbers increased from a mean base-line value of 16,300/ml to
52,600/ml on day 1, while lymphocytes increased from a mean
base-line value of 12,300/ml to 26,500/ml on day 1 and returned
to 6,600/ml by day 7. By contrast polymorphonuclear leukocyte
levels dropped from a base line of 19,500/ml to 4,000/ml on day
1 and rose to 7,700/ml on day 7. There were no significant
differences in WBC counts and differential populations between
responders and nonresponders. Some increase in total cell
counts were observed by the seventh and eighth weeks of
therapy.
NK
Evaluation of NK (effectortarget
ratio, 25:1) in blood and
peritoneal fluid is presented (mean ±SE) in Chart 3. The NK
RESEARCH VOL. 45 SEPTEMBER
1985
4450
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1985 American Association for Cancer Research.
INTERFERON
i.p. IN OVARIAN
CANCER
io" i
40
id1 •
30
k
•H20
io1 •
10
iplFN
io'O
2
3
4
Time (days)
Chart 3. NK. Points, the mean percentage of specific lysis of each patient; bars,
SE. The peripheral blood lymphocytes (
) and the peritoneal cavity lymphocytes
(
) are presented. Time zero = baseline levels prior to initial treatment, and day
1 is the mean of NK 24 h after i.p. administration of rlFN-a2- Specimens taken 7
days after treatment are also those taken immediately prior to subsequent treat
ment.
base line in the blood was 27 ±3, which was 24 ±3 by day 1,
and rose slightly to 32 ±2 by day 7. In contrast the i.p. NK
values were markedly boosted from a base line of 6 ±2 to 31
±4 on day 1 and fell to 10 ±2 by day 7. Over the successive
treatment weeks alterations in NK were relatively homogeneous
with no major variation over time; I.e., there was no cumulative
boost or depression in the NK over 8 to 16 weeks. Boosting of
NK did not uniformly relate to response; i.e., while all patients
who had a documented response had significant elevation of i.p.
NK on day 1, five of the eight patients who did not have a
response also demonstrated some NK elevation.
5
IO 15 20 25 30 35 40 45 50 55 60 65 70 75
Tim«
(hours)
Chart 4. Mean values of blood (
) and ¡.p.fluid (
) r!FN-a2 levels following
the i.p. administration of 50 x 10' r!FN-«2;oars, SE.
DISCUSSION
This is the first report demonstrating efficacy for the i.p.
administration of rlFN-a2 in patients with persistent MRD epithe
lial ovarian cancer detected at second look laparotomy. The
documentation of complete responses in 4 of the 11 surgically
evaluated patients suggests that this biological response modi
fier may prove useful in the treatment of MRD ovarian cancer.
Responses appeared to occur after four to six cycles of 50 x
106 units of rlFN-a2, as indicated by the results obtained with
serial peritoneal fluid cytologies performed on these patients.
The toxicity of the i.p. rlFN-a2 was moderate; the principal
problem related to the use of the i.p. catheters.
In this study the rlFN-a2 pharmacokinetic data suggest that
very high i.p. concentrations of the agent are present for at least
24 h, at which time dialysis solution is withdrawn for study.
Neutralizing Antibodies
However, a low level of rlFN-a2 is detectable 48 and 72 h after
treatment. Peripheral blood levels remained elevated for 5 to 6
days. These data might help to explain the biological efficacy of
Multiple serum samples obtained prior to treatment and on
rlFN-02. Also the serum levels of rlFN-«2are quite high in
days 1 and 7 after treatment were tested for neutralizing anti
bodies specific to the human rlFN-<*2. No antibodies were de
comparison to the usual levels achieved with standard i.m. or
s.c. administration, which usually ranges approximately 200 to
tected at any time.
300 units/ml (1-8). Thus the peritoneal cavity may be acting as
a reservoir, leading to both constant surface exposure and to a
IFN Pharmacokinetics
persistent serum level. This feature has implications for cell cycle
studies as indicated by in vitro data which demonstrate that
Serum. Serum r!FN-a2 levels rose from pretreatment base line constant exposure to r!FN-«2produces more marked cytotoxic
of <38 units/ml to levels of 75-300 units/ml 8 h after treatment
and cytostatic effect than does pulse exposure (3, 4). This
(Chart 4). By 24 h serum rlFN-«2levels consistently ranged from
antitumor activity might result from exposure of the ovarian
cancer cells to high concentrations of rlFN-a2.
150 to 600 units/ml. Serum samples taken from day 2 through
day 7 were scarce due to the logistics of patient travel to and
It is of interest that only patients with MRD (<5 mm maximum
from the hospital on "off" days; however, from the small number
tumor diameter) achieved a surgically documented response and
of samples that we were able to obtain, it appears that rlFN-a2
that patients with larger disease did not respond. Indeed five of
levels remained high or above base-line values up to day 6.
seven surgically evaluated patients (71%) with MRD had a
Serum levels returned to base line by day 7.
response. As such, future trials of this agent should probably be
Peritoneal Fluid. Peritoneal fluid rlFN-o2 levels rose from a confined to patients with MRD. These data are similar to our
pretreatment level of <38 units/ml up to approximately 38,400
previous experience using i.p. immunotherapy (11,12), chemo
units/ml during the first 24 h of treatment at Interferon dosage
therapy (13), and whole abdominal radiation therapy (23) and
of 50 x 106 units of rlFN-a2 (Chart 4). Thereafter coinciding with
with data obtained using a murine ovarian cancer model (14,24removal of dialysis fluid at the 24-h mark, levels of rlFN-a2 in the
26). Therefore it would appear that the tumor burden prior to the
peritoneal cavity returned to base line. Measurements of rlFN-«2 initiation of therapy is critical in these patients; thus surgical
levels in lavages at day 2 (48 h), day 3 (72 h), and day 7 (168 h) resection of tumor might be important if larger tumors are present
were equal to base-line levels of <38 units/ml.
at the time of exploratory surgery (27).
CANCER
RESEARCH
VOL. 45 SEPTEMBER
1985
4451
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1985 American Association for Cancer Research.
INTERFERON
Lp. IN OVARIAN
The only previous studies of the use of systemically adminis
tered IFNs for the treatment of ovarian cancer have been carried
out in Scandinavia (2, 9) and the United States (28). The first
group of authors reported partial clinical responses in 2 of 18
previously treated patients, 3 patients with "minimal" responses,
and 3 patients who had stable disease for up to 1 year. The
second group of authors (27) reported no responses, while 4 of
their 13 patients had stable disease (median, 6 months).
In our experience ovarian cancer often remains confined to
the peritoneal cavity, even after clinical recurrence (10). Intracav-
patients after primary cytoreductive surgery in combination with
chemotherapy or in an adjuvant setting after a negative second
look laparotomy.
REFERENCES
itary therapies are appealing because of their potential for in
creasing the concentration of the agent and bringing the agent
into direct contact with the residual tumor (6,11 -15). Historically
BRMs have been most effective in patients in whom the agent
has been administered intralesionally or intracavitarilly (6,11-15,
24, 25). Also the presence of tumor cytotoxic chemotherapy
appears to be immunosuppressive in patients with ovarian can
cer (19).
BRMS have generally been unsuccessful in the treatment of
ovarian cancer (13). However, some studies have suggested
that combinations of systemically administered Bacillus Calmette-Guer/n (18) and C. parvum (30) administered concomitantly
with cytotoxic chemotherapy induce a higher proportion of clini
cally documented complete responses than the use of combi
nation chemotherapy alone. The i.p. administration of C. parvum
in patients with MRD documented at second look has produced
surgically documented responses (11,12).
NK and antibody dependent cell mediated cytotoxicity may be
partially responsible for tumor control and/or proliferation in the
peritoneal cavity (11,12,19,31).
These effector cells have been
shown to be deficient in the peritoneal cavities of women with
ovarian cancer previously treated with combination chemother
apy (18). The i.p. administration of C. parvum has been reported
to augment peritoneal cavity cytotoxic effector lymphocytes,
both NK (12,31) and antibody-dependent cell mediated cytotox
icity (11, 12), in responding patients. Since rlFN-<*2 has been
shown to be a potent stimulator of NK (6), its i.p. administration
to patients with persistent epithelial disease thus seems rational.
Our measurement of both blood and i.p. fluid NK in relationship
to the rlFN-02 treatments shows that i.p. NK is significantly
boosted after 1 day, while blood values showed considerable
variability. No cumulative boosting or suppression is demon
strated. Also the values do not correlate with the objective
response. The temporal dynamics of NK effectors relative to the
rlFN-(>2 pharmacokinetics are the subject of a separate report.4
Although mononuclear cells were not tested for cytotoxicity in
these patients, future studies will focus on these potentially
important cells.
IFNaS have been shown to have an antiproliferative effect
against ovarian cancer cells grown in human tumor stem cell
assays (32). Also there is evidence that IFNs may act synergistically with cytotoxic chemotherapeutic agents in vitro (33-36).
These data suggest that the exposure of cancer cells to rlFN-a2
prior to or concomitant with cytotoxic agent(s) can increase cell
kill. Therefore i.p. rlFN-<*2may be potentially useful in addition to
cytotoxic chemotherapy in patients with ovarian cancer. Indeed
there is the potential for the use of rlFN-a2 in the treatment of
4 C. Spina, A. Uchtenstein, J. S. Berek. and J. Zkjhelboim. Regional modulation
of natural immunity by Interferon therapy to patients with ovarÃ-ancancer, submitted
for publication.
CANCER RESEARCH
CANCER
1. Borden, E. C. Progress toward therapeutic application of interferons. Cancer
(Phila.), 54: 2770-2776, 1984.
2. Kirkwood, J. M., and Ernstoff, M. S. Interferons in the treatment of human
cancer. J. Clin. Oncol., 2: 336-352, 1984.
3. Grossberg, S. E. The interferons and their inducers: molecular and therapeutic
considerations. Part I. N. Engl. J. Med., 287:13-19,1972.
4. Grossberg, S. E. The interferons and their inducers: molecular and therapeutic
considerations. Part II. N. Engl. J. Med., 287: 79-85,1972.
5. Grossberg, S. E. The interferons and their inducers: molecular and therapeutic
considerations. Part III. N. Engl. J. Med., 287:122-128,1972.
6. Oldham. R. K. Biologicals and biological response modifiers: fourth modality
of cancer treatment. Cancer Treat. Rep., 68:221-232,1984.
7. Quesada, J. R., Reuben, J., Manning, J. T., Hersh, E. M., and Gutterman, J.
U. Alpha Interferon for induction of remission in hairy-cell leukemia. N. Engl. J.
Med., 370:15-18,1984.
8. Foon, F. A., Sherwin, S. A., Abrams, P. G., Longo, D. L., Fer, M. F., Stevenson,
H. C., Ochs, J. J., Bottino, G. C., Schoenberger, C. S., Zeffren, J., Jaffe, E.
S., and Oldham, R. K. Treatment of advanced non-Hodgkin's lymphoma with
recombinant leukocyte a Interferon. N. Engl. J. Med., 311:1148-1152,1984.
9. Einhorn, N., Cantell, K., Einhorn, S., and Strander, H. S. Human leukocyte
Interferon therapy for advanced ovarÃ-ancancer. Am. J. Clin. Oncol., 5: 167172,1982.
10. Berek, J. S., Hacker N. F., Lagasse, L. D., Poth, T., Resnick, B., and Nieberg,
R. K. Second-look laparotomy in stage III epithelial ovarÃ-ancáncer: clinical
variables associated with disease status. Obstet. Gynecoi., 64: 207-212,
1984.
11. Bast, R. C., Berek, J. S., Obrist, R., Griffiths, C. T., Berkowitz, R. S., Hacker,
N. F., Parker, L., Lagasse, L. D., and Knapp, R. C. IntraperÃ-toneal immunotherapy of human ovarÃ-ancarcinoma with Corynebacterium parvum. Cancer
Res., 43:1395-1401,1983.
12. Berek, J. S., Knapp, R. C., Hacker, N. F., Uchtenstein, A., Jung, T., Spina, C.,
Obrist, R., Griffiths, C. T., Berkowitz, R. S., Parker, L, Zighelboim, J., and
Bast, R. C. IntraperÃ-tonealimmunotherapy of epithelial ovarÃ-ancarcinoma with
Corynebacterium parvum. Am. J. Obstet. Gynecoi., in press, 1985.
13. Berek, J. S., and Hacker, N. F. Immunotherapy of gynecologic cancer. In: G.
Deppe (ed.), Chemotherapy of Gynecologic Cancer, pp. 363-374. New York:
Alan R. Uss, Inc., 1984.
14. Berek, J. S., Cantre«,J. L., Uchtenstein, A. K., Hacker, N. F., Knox, R. M.,
Nieberg, R. K., Poth, T., Elashoff. R. M., Lagasse, L. D., and Zighelboim, J.
Immunotherapy with biochemically dissociated fractions of Propionibacterium
acnes in a murine ovarian cancer model. Cancer Res., 44:1871-1875,1984.
15. Pretorius, R. G., Hacker, N. F., Berek, J. S., Ford, L. C., Hoeschete, J. D.,
Butler, T. A., and Lagasse, L. D. al. The pharmacokinetics of intraperitoneal
cisplatinum in refractory ovarian cancer. Cancer Treat. Rep., 67: 1085-1092,
1983.
16. Berek, J. S., Griffiths, C. T., and Leventhal, J. L. Second-look evaluation in
ovarian cancer. Obstet. Gynecoi., 58:192-198,1981.
17. Blessing, J. A. Design, analysis, and interpretation of chemotherapy trials in
gynecologic cancer. In: G. Deppe (ed.), Chemotherapy of Gynecologic Cancer,
pp. 49-83. New York: Alan R. Uss, Inc., 1984.
18. Boyum, A. Separation of lymphocytes from blood and bone marrow. Scand.
J. Clin. Lab. Invest., 27 (Suppl. 97V 77, 1968.
19. Berek, J. S., Bast, R. C., Uchtenstein, A., Hacker, N. F., Spina, C., Lagasse,
L. D., Knapp, R. C., and Zighelboim, J. Lymphocyte cytotoxicity in the
peritoneal cavity and blood of patients with ovarÃ-ancancer. Obstet. Gynecoi.,
64:708-714,1984.
20. Jondal, M., Spina, C., and Targan, S. Human spontaneous killer cells: selective
cytotoxicity against tumor-derived target cells. Nature (Lond.), 272: 62-64,
1978.
21. Protzman, W. P., Jacobs, S. L., Minnicozzi, M., Oden, E. M., and Kelsey, D.
K. A radioimmunologic technique to screen for antibodies to alpha-2 Interferon.
J. Immunol. Methods, in press, 1985.
22. Rubinstein, S., Samiltetti, P., and Pestka, S. A convenient assay for interferon.
J. Viral., 37:755-758,1981.
23. Hacker, N. F., Berek, J. S., Bumison, C. M., Heintz, A. P. M., Eisen, K. P. S.,
Juillard, G. J. F., Fu, Y. S., and Lagasse, L. D. Whole abdominal radiation as
salvage therapy for epithelial ovarian cancer. Obstet. Gynecoi., 65: 60-66,
1985.
24. Knapp, R. C., and Berkowitz, R. S. Corynebacterium parvum as an immunotherapeutic agent in an ovarian cancer model. Am. J. Obstet. Gynecoi., 728:
782-786, 1977.
25. Bast, R. C., Knapp, R. C., Mitchell, A. K., Thorston, J. F., Tucker, R. W., and
Schtossman, S. F. Immunotherapy of a murine ovarian carcinoma with Cory
nebacterium parvum and specific heteroantiserum. J. Immunol., 723: 1945-
VOL. 45 SEPTEMBER
1985
4452
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1985 American Association for Cancer Research.
INTERFERON
i.p.
IN OVARIAN
1951,1979.
26. Vanhaelen, C. P. J., and Fisher, R. I. Requirements for successful immunetherapy and chemo-immunotherapy
of a murine model of ovarian cancer.
Cancer Res., 47: 980-983, 1981.
27. Berek, J. S., Hacker, N. F., Lagasse, L. D., Nieberg, R. K., and Elashoff, R. M.
Survival of patients following secondary cytoreductive surgery in ovarian
cancer. Obstet. Gynecol., 61:189-193,1983.
28. Freedman, R., Wharton, J. T., and Rutledge, F. Human leukocyte interferon
(IFNa) in patients with epithelial ovarian cancer (a preliminary report). Proc.
Am. Assoc. Cancer Res., 22: 372, 1981.
29. Alberts, D. S., Salmon, E. S., and Moon, T. E. Chemoimmunotherapy for
advanced ovarian carcinoma with Adriamycin-cyclophosphamide ±BCG: early
report of a Southwest oncology group study. Recent Results Cancer Res.,
68:160-165,1978.
30. Creasman, W. T., Gall, S. A., and Blessing, J. A. Chemoimmunotherapy in the
management of primary stage III ovarian cancer: a Gynecologic Oncology
Group study. Cancer Treat. Rep., 63: 319-323,1979.
31. LJchtenstein, A., Berek, J. S., Bast, R. C., Spina, C., Hacker, N. F., Knapp, R.
CANCER
32.
33.
34.
35.
36.
CANCER
C., and Zighelboim, J. Activation of peritoneal lymphocyte cytotoxicity in
patients with ovarian cancer by intraperitoneal treatment with Cornyebacterium
parvum. J. Biol. Resp. Modifiers, 3.-1-8,1984.
Willson, J. K. V.. Bittner, G., and Border, E. C. Antiproliferative activity of
interferons against ovarian cancer cells grown in human tumor stem cell assay.
J. Interferon Res., 4: 441-447,1984.
Welander, C., Gaines, J., Homesley, H., and Rudnick, S. In vitro synergistic
effects of recombinant human interferon alpha2 (r!FN-«2)and doxorubicin on
human tumor cell lines. Proc. Am. Soc. Clin. Oncol., 2: 42,1983.
Inoue, M., and Tan, Y. H. Enhancement of actinomycin D- and c/s-diamminedichloroplatinum(ll) induced killing of human fibroblasts by human /3-interferon.
Cancer Res., 43: 5484-5488,1983.
Tozawa, M., Kidowaki, T., Terufusa, T., Schmidt, H. J., Abu-Ghazaleh, S.,
Whisnant, J. K., and Di Saia, P. J. Combination therapy of interferon and
cyclophosphamide on murine neuroblastoma. J. Jpn. Soc. Cancer Ther., 17:
613,1982.
Aapro, M. S., Salmon, S. E., and Alberts, D. S. Schedule dependent synergism
of vinWastine and cloned leukocyte interferon«. Stern Cells, 7:303-304,1981.
RESEARCH VOL. 45 SEPTEMBER
1985
4453
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1985 American Association for Cancer Research.
Intraperitoneal Recombinant α-Interferon for ''Salvage''
Immunotherapy in Stage III Epithelial Ovarian Cancer: A
Gynecologic Oncology Group Study
Jonathan S. Berek, Neville F. Hacker, Alan Lichtenstein, et al.
Cancer Res 1985;45:4447-4453.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/45/9/4447
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. © 1985 American Association for Cancer Research.