1865

lntraperitoneal Mitoxantrone as
Consolidation Treatment for Patients
with Ovarian Carcinoma in Pathologic
Complete Remission
Patrick Dufour, M.D.,* Jean-Pierre Bergerat, M.D.,* Jean-Claude Barats, M.D.,t
Cathy Giron, M.D.,* Brigitte Duclos, M.D.,* Pierre Dellenbach, M.D.,$
Jean Ritter, M.D.,lI Robert Renaud, M.D., Bruno Audhuy, M.D.,t
and Francis Oberling, M.D.*

Background. Stages 11-IV ovarian cancer in pathologic complete remission (pCR) at second-look surgery
have a high relapse rate (50%) during the first 2 years.
Considering relapse sites (abdomen and/or pelvis), intraperitoneal (IP) therapy is a logical approach. Mitoxantrone is an effective drug against ovarian cancer cells in
vitro and is an attractive agent for IP therapy because of
its very low peritoneal clearance. The value of IP mitoxantrone was studied as consolidation treatment of ovarian cancer in pCR at second-look surgery.
Methods. Fifty patients with Stages 11-IV ovarian
cancer (8, Stage 11; 37, Stage 111; 5, Stage IV) were included
in this Phase I1 study, which began in June 1988. All patients had undergone initial cytoreductive surgery followed by 6 cyclophosphamide, doxorubicin, and cisplatin cycles. All patients were in pCR, as confirmed by second-look surgery. Consolidation treatment consists of 20
mg (total dose per cycle) IP mitoxantrone every 3 weeks
for six cycles.
Results. Toxicity was limited to mild abdominal pain
not requiring dose reduction (90% pain grade I 2). With a
median follow-up of 2 years, the 5-year predicted survival is 59.8% (95% confidence interval [CI], 48.3-71.3),
and the disease-free survival (DFS)rate is 47.3% (95% CI,
36.7-57.9). Patients with no or microscopic residual disease after initial surgery had a better 5-year DFS rate
(75.8%) than those with macroscopic residual disease
(31.2%)(P= 0.01).
From the *DCpartement d Onco-HCmatologie and the BService
GynCcologie, HBpitaux Universitaires de Strasbourg, Strasbourg; the
tService Onco-HCmatologie, HBpital Louis Pasteur, Colmar; and the
$Service GynCcologie, Centre MCdico Chirurgical et ObstCtrical,
Schiltigheim, France.
Address for reprints: Patrick Dufour, M.D., DCpartement d'
Onco-HCmatologie, HBpitaux Universitaires de Strasbourg, 67098
Strasbourg, France.
Accepted for publication November 12, 1993.

Conclusion. IP mitoxantrone (20 mg/cycle) is feasible with an acceptable abdominal toxicity. The results in
terms of DFS are encouraging, but a randomized study
versus no treatment is necessary to prove the value of
this consolidation treatment. Cancer 1994;73:1865-9.
Key words: mitoxantrone, intraperitoneal, ovarian
cancer, consolidation treatment.

Cisplatin or carboplatin-based combination chemotherapy allows a high clinical response rate in ovarian carcinoma (70-go%), but only 30-40% of patients are in
pathologic complete remission (pCR), as confirmed by
second look laparotomy.',* Depending on risk factors,
30-60% of pCR patients experience relapse during the
first 2 year^.^,^ The most important prognostic factors
are initial stage, extent of residual disease after primary
cytoreductive surgery, and performance
Initial
disease and sites of relapse are in most cases confined to
the abdomen or the pelvis or both, so that an intraperitoneal (IP) treatment is a logical approach. IP chemotherapy allows a very high dose effect that might overcome
resistance to systemic standard dose chemotherapy.
There have been many reports on IP chemotherapy in
patients with ovarian cancer, using cisplatin, the cisplatin-cytarabine combination, or other antineoplastic
agents. Among these drugs, mitoxantrone is especially
interesting, because it exerts a very potent cytotoxic effect in vitro against several ovarian carcinoma cell
lines.' Moreover, it has a low peritoneal clearance and a
high peritoneoplasmatic concentration ratio (>800),
leading to a very high exposure dose.' We previously
conducted a Phase I1 study of IP mitoxantrone in ovarian carcinoma, which showed that a dosage of 20 mg
per injection was feasible and that the best indications

1866

CANCER April 1, 1994, Volume 73, No. 7

Table 1. Patient Characteristics

~~

No. of patients
Total
Age
Median 58
Range2441
Stage
IIA
IIB
IIC
IIIA

IIIB
IV
Residual disease after first look
None
Microscopic
Macroscopic (< 2 cm)
Macroscopic (> 2 cm)
Unknown
Histologic type
Serous
Mucinous
Endometrial
Unclassified
Histologic grade
Well differentiated
Moderately differentiated
Undifferentiated
Not determined

50

3
1
4
7
30
5

16
4
11
17
2

40
2
3
5

4
6
17
23

were microscopic residual disease or pCR. We therefore

began in June 1988 a Phase I1 study of IP mitoxantrone
as consolidation treatment in pCR ovarian carcinoma.

las), peritoneal washing, and biopsy of retroperitoneal

lymph nodes. At this point, an IP catheter was positioned. Two weeks later, IP fluid distribution was
checked by abdominal ultrasound or computed tomography scan after infusion of 1000 ml Ringers solution.
Modality of IP mitoxantrone administration was
infusion of 500 ml Ringers solution over 30 minutes,
then 20 mg mitoxantrone (total dose per cycle) diluted
in 500 ml Ringers solution and infused over 30 minutes, then 1000 ml Ringers solution over 60 minutes.
There was no draining, and cycles were repeated every
3 weeks (six cycles in all).
Where there was Grade 3 abdominal pain, the dose
of mitoxantrone was reduced by 25%. Treatment was
definitively stopped when Grade 4 abdominal pain occurred (see Table 2 for grading definitions).
After six IP cycles, treatment was discontinued.
Clinical examination, tumor marker measurements,
and abdominal and pelvic ultrasound were performed
every 6 months for 5 years.
This protocol was accepted by the local ethical
committee, and informed consent has been obtained
from patients.
Statistical Methods

Overall survival and disease-free survival (DFS) were

calculated according to the Kaplan-Meier method, and
comparison was made with the log-rank test. Overall
survival was calculated from date of diagnosis to death.
DFS was calculated from date of pCR documentation
until relapse. The end point was February 1, 1993.
Results

Material and Methods

Survival and DFS

Fifty patients were included in this study from June
1988 to November 1992. Eligibility criteria were (1)
Stages 11-IV ovarian carcinoma, (2) pCR affirmed by
second-look surgery, and (3) IP catheter positioned at
time of second-look surgery.
The patients characteristics are summarized in Table 1. Histologic grading was not obtained for 23 patients who underwent their first-look surgery in other
centers. All had the same therapeutic scheme: initial
surgery with maximal debulking, including hysterectomy and bilateral salpingo-oophorectomy and omentectomy, was followed by six monthly chemotherapy
cycles of cyclophosphamide 500 mg/m2 intravenously
on day 1, doxorubicin 50 mg/m2 intravenously on day
1, and cisplatin 20 mg/m2 intravenously on days 1-5.
Responding patients underwent second-look surgery
with multiple peritoneal biopsies (includingparietocolic
gutters, diaphragms and liver surfaces, pouch of Doug-

With a median follow-up of 2 years, the 5-year predicted survival and DFS were 59.8% (95% confidence
interval [CI], 48.3-71.3) and 47.3% (95% CI, 36.757.9), respectively (Fig. 1). The 2-year predicted relapse

Table 2. Abdominal Pain: Maximal Grade Reported

bv Each Patient During Intraperitoneal Treatment
Grade
0
1
2

3
4

Definition
No pain
Transient abdominal pain
Abdominal pain requiring analgesic
treatment
Abdominal pain requiring analgesic
treatment and dose modification
Abdominal pain requiring permanent
discontinuation of treatment

No. of patients
6
21
18

4
1

IP Mitoxantrone for Ovarian Cancer/Dufour et al.

1867

rate was 21% (95% CI, 13-28). The median time to

relapse was 22 months. We have observed 13 relapses,
3 of them during IP therapy. All relapses were abdominal.
We did not observe any relapse in Stages IC and 11,
and there were 7 relapses among 37 Stage I11 and 3
relapses among 5 Stage IV patients.
According to the size of residual disease after firstlook surgery, we divided the patients into two subgroups: Group A, low-risk group (n = 20) with no or
microscopic residual disease, and Group B, high-risk
group (n = 28) with macroscopic residual disease after
primary cytoreductive surgery. The 5-year DFS was
75.8% (95% CI, 60.2-91.4) in Group A and 31.2%
(95% CI, 17.3-45.1) in Group B (P = 0.05) (Fig. 2).

U
R

60

__

40

._

L.

V
A

2o
0

t
I

12

18

24

30

36

42

48

54

60

MONTHS

Toxicity
As expected, the major side effect was abdominal pain
(Table 2) with a predominance of grades less than or
equal to Grade 2. Of the patients, 45 of 50 experienced
only mild abdominal pain not requiring dose modification. However, we observed four patients with Grade 3
disease requiring a 25% dose reduction and one patient
with Grade 4 disease requiring treatment discontinuation. We did not observe any delayed abdominal complication.
We did not observe any hematologic or cardiac toxicity or any infectious episodes.
Fourteen patients underwent explorative laparotomy when the IP catheter was removed. We observed
in all cases a chemical peritonitis with a blue aspect and
thickening of the peritoneum. In nine cases, peritoneal
adhesions were present. In the other cases (n = 36), no
third-look laparotomy was performed and the IP cath-

100

V
I
A
L

t
. O t

I
0

12

18

24

30

36

42

48

MONTHS

Figure 1. (Solid line) DFS and (dotted line) overall survival.

54

60

Figure 2. DFS according to the size of residual disease after initial

cytoreductive surgery. (Solid line) Group A: no or microscopic
residue. (Dotted line) Group B: macroscopic residue.

eter was withdrawn without abdominal or pelvic exploration.

Discussion
There is no consensus about the treatment of patients
with ovarian cancer in pCR after second-look surgery.
Some authors advocate a wait-and-see attitude only.
Others recommend a maintenance therapy using three
to six more cycles of the same induction chemotherapy.
Due to the high observed relapse rate, however, several
modalities of consolidation treatment have also been
proposed. The three main options are (1)whole abdominal radiation therapy, (2) high-dose chemotherapy followed by autologous bone marrow transplant, and (3)
IP therapy.
None of these modalities has proved a clear and
definitive advantage in terms of DFS or survival, and
there are no randomized comparative studies in this
field.
The wait-and-see attitude confirms the finding that
approximately half of the pCR patients relapse during
the first 2 years. Rubin et al.9 reported results of 91
patients in pCR and observed a 44% recurrence rate at 5
years with a median time to recurrence of 24 months.
However, 75% of patients in this series had a good
prognosis with a minimal residual disease after firstlook surgery. Neijt,' in a series of 41 patients in pCR
after cisplatin-induction chemotherapy, observed a 5year survival rate of only 42%. Similar results have
been observed by Luesley et a1.I'
Watring et al." studied the value of increasing the
number of induction chemotherapy cycles and com-

1868

CANCER April 2, 2994, Volume 73, No. 7

pared in a nonrandomized study six versus nine cycles

of cyclophosphamide, doxorubicin, and cisplatin. There
was no statistical difference in terms of DFS or survival.
Hakes et al. reported similar results in a randomized
trial comparing 5 to 10 cycles of cisplatin-based combination chemotherapy in 78 patients. Twelve patients
receiving five cycles experienced a partial response;
these partial responders received five more cycles, and
only one of them demonstrated a complete response.
Whole abdominal radiation therapy is probably the
most widely used therapeutic modality, but the reported results are conflicting. Fuks et al.,13 in a series of
25 Stage I11 pCR patients, reported that whole abdominal radiation therapy did not enhance the cure rate,
with a 37% 5-year survival. Menczer et al.,14 in a study
of 18 Stages II-IV patients who received abdominopelvic irradiation as consolidation treatment, found no advantage in terms of survival compared with a historical
group without radiation therapy.
High-dose chemotherapy with autologous bone
marrow transplant in pCR patients is still investigational. We have reported 6 cases of pCR patients treated
by high-dose melphalan (140 mg/m2) and autologous
bone marrow transplant. We observed three relapses at
13, 14, and 20 months, and only three patients experienced a long DFS (>3 years).15Dauplat et a1.16 found
similar results among nine patients.
IP treatment has been extensively studied in ovarian carcinoma. It has been shown that such an approach is ineffective in the face of bulky disease and
that tumor responses may be observed in cases of minimal residual disease (<1-2 cm diameter). Cisplatin
alone or combined with cytarabine is the drug most
commonly used. Howell et al. reported the results observed with IP cisplatin in 25 patients with minimal
residual disease and showed an improvement of DFS
and overall survival. Markman et a1.18treated a series of
58 advanced ovarian cancers with minimal residual disease (10.5 cm) at second-look surgery with a combination of IP cytarabine and cisplatin. They observed that
patients with microscopic residual disease experienced
a longer DFS (32 versus 15 months) than patients with
minimal macroscopic residual disease (<0.5 cm). These
results show that patients with microscopic residual
disease can experience a very long DFS, some of them
perhaps being cured by IP therapy. There are few studies concerning IP treatment for pCR patients. Menczer
et al.19reported a nonrandomized study with a group of
25 patients receiving three IP cisplatin cycles after negative second-look surgery, compared with 12 patients
undergoing no further treatment. There was a trend
toward better survival times in the IP group. Spencer et
al. reported a nonrandomized study in 31 pCR patients
at second-look surgery; 14 patients received IP radioac-

tive phosphorus 32 (15 mCi) and 17 patients received

no treatment. With a mean follow-up of 4 years, they
observed no relapse in the group treated with phosphorus 32 and four relapses in the group receiving no further therapy (P = 0.076).20
Some studies of IP mitoxantrone therapy showed
promising results in patients with minimal residual disease, but abdominal pains leading in most cases to permanent treatment discontinuation was a major limitation for this treatment. Mitoxantrone dosage was higher
(20-35 mg/m2), however, than in our
Markman et a1. recently reported a better tolerance with
low-dose (10 mg/m2) mitoxantrone in a series of 28
patients. IP therapy could be repeated weekly or biweekly, and in 13 patients whose residual tumor was
less than 1 cm they observed a 31% surgically demonstrated response rate. In our preliminary study, we did
not observe any response in patients with macroscopic
disease, but IP mitoxantrone was administered only
once every 3 weeks.23
In the current study, tolerance was acceptable, because 90% of patients experienced only mild abdominal
pains, allowing the administration of the six cycles at
full doses in 75% of cases. We observed a satisfactory
5-year DFS (47.3%), and only 21% of patients have
experienced relapse during the first 2 years. However,
we wish to emphasize that these results must be interpreted with great caution, because this study is not comparative and confidence intervals for DFS and survival
are broad. The heterogeneity of our group of patients in
terms of prognostic factors may cause a bias in the interpretation of results. We therefore studied the DFS
according to the main prognostic factor (size of residual
tumor after first-look surgery) and observed a 5-year
DFS of 75.8% for good risk and 31.2% for bad risk.
These DFS figures are higher than those commonly
published in the literature for similar groups without
consolidation treatment. Chemical peritonitis leading to
adhesions is a major problem in IP mitoxantrone therapy. Markman et al. observed in patients who showed
minimal residual disease at initiation of IP therapy that
biopsy specimens were positive in areas not exposed to
mitoxantrone and negative in areas exposed (blue
aspect). Thus, adhesions impairing a good distribution
could be a cause of IP therapy failure. We think that a
maximum dose of mitoxantrone of 20 mg per cycle
could limit the occurrence of such adhesions.
In conclusion, IP mitoxantrone (20 mg per cycle) as
consolidation treatment in patients with pCR ovarian
cancer is feasible, with acceptable toxicity. This modality may improve DFS, but a broader and randomized
study versus no treatment is necessary to assess definitively the value of this approach. Such a consolidation
treatment might be offered to patients in pCR at sec-

IP Mitoxantrone for Ovarian Cancer/Dufour et al.

ond-look surgery but with high risk factors of relapse,

particularly in patients with microscopic or minimal residual disease at the conclusion of first-look surgery.

1869

13.

References
1. Advanced Ovarian Cancer Trialists Group. Chemotherapy in

2.

3.

4.

5.

7.

8.

9.

10

11.

12.

advanced ovarian cancer: an overview of randomised clinical

trials. BM/ 1991; 303:884-93.
Neijt J P. Treatment of advanced ovarian cancer: 10 years of
experience. Ann Oncol 1992; 3:17-27.
Podczaski E, Manetta A, Kaminski P, Ricelli A, Larson J, De
Geest K, et al. Survival of patients with ovarian epithelial carcinomas after second look Iaparotomy. Gynecol Oncol 1990;
36:43-7.
Neijt JP, ten Bokkel Huinink WW, Van der Burg MEL, Van Oosterom AT. Complete remission at laparotomy: still a gold standard in ovarian cancer? Lancet 1986; 1:1028.
Maloisel F, Dufour P, Bergerat JP, Bellocq JP, Duclos B, Boilletot
A, et al. Resultats d u n e polychimiothkrapie B base de cisplatine
et analyse des facteurs pronostiques dans les cancers de lovaire:
a propos de 106 cas. Bull Cancer 1989; 76:1083-93.
Lund B, Williamson P. Prognostic factors for overall survival in
patients with advanced ovarian carcinoma. Ann Oncol 1991;
2:281-7.
Alberts DS, Young L, Mason N, Salmon SE. In-vitro evaluation
of anticancer drugs against ovarian cancer at concentrations
achievable by intraperitoneal administration. Semin Oncol 1985,
12:38-42.
Nagel JD, Varossieau FJ, Dubbelman R, ten Bokkel Huinink
WW, Mc Vie JG. Clinical pharmacokinetics of mitoxantrone
after intraperitoneal administration. Cancer Chenzother PharnzaC O ~1992, 29:480-4.
Rubin SC, Hoskins WJ, Saigo PE, Chapman D, Hakes TB, Markman M, et al. Prognostic factors for recurrence following negative second look laparotomy in ovarian cancer patients treated
with platinum based chemotherapy. Gynecol Oncol 1991;
42: 137-4 1,
Luesley D, Lawton F, Blackledge G, Hilton C, Kelly K, Rollason
T, et al. Failure of second-look laparotomy to influence survival
in epithelial ovarian cancer. Lancet 1988, 1:599-603.
Watring W, Semrad N, Alaverdian V, Latino F, Pretorius G.
Second-look procedures in ovarian cancer patients receiving six
versus nine courses of platinum, adriamycin, cytoxan (PAC) chemotherapy: the SCPMG experience, 1982-1985. Gynecol Oncol
1989; 32:245-7.
Hakes T, Chalas E, Hoskins WJ, Jones WB, Markman M, Rubin
SC, et al. Randomized prospective trial of 5 versus 10 cycles of

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

cyclophosphamide, doxorubicin and cisplatin in advanced ovarian carcinoma. Gyiiecol Oncol 1992; 45284-9.
Fuks 2, Rizel S, Biran S. Chemotherapeutic and surgical induction of pathological complete remission and whole abdominal
irradiation for consolidation does not enhance the cure of stage
111 ovarian carcinoma. ] Clin Oncol 1988; 6:509-16.
Menczer J, Modan M, Brenner J, Ben-Baruch G. Abdominopelvic irradiation for stage 11-IV ovarian carcinoma patients with
limited or no residual disease at second look laparotomy after
completion of cisplatinum based combination chemotherapy.
Gynecol Oncol 1986; 24:149-54.
Dufour P, Bergerat JP, Liu KL, Bohbot A, Maloisel F, Duclos B, et
al. High dose melphalan and ABMT with or without radiotherapy as consolidation treatment for ovarian carcinoma in complete remission or with microscopic residual disease. Eur ] Gynecol Oncol 1991; 12457-61.
Dauplat J, Legros M, Condat P, Ferriere JP, Ben Ahmed S,
Plagne R. High dose melphalan and autologous bone marrow
support for treatment of ovarian carcinoma with positive second
look operation. Gynecol Oncol 1989; 34:294-8.
Howell SB, Zimm S, Markman M, Abramson IS, Clearly S,
Lucas WE, et al. Long-term survival of advanced refractory
ovarian carcinoma patients with small volume disease treated
with intraperitoneal chemotherapy. Clin Oncol 1987; 5:160712.
Markman M, Hakes T, Reichman 8, Hoskins W, Rubin S, Jones
W, et al. Intraperitoneal cisplatin and cytarabine in the treatment of refractory or recurrent ovarian carcinoma. ] Clin Oncol
1991; 9:204-10.
Menczer J, Ben-Baruch G, Rizel S, Brenner H. Intraperitoneal
chemotherapy versus no treatment in patients with ovarian carcinoma who are in complete remission. Cancer 1992; 70:1956-9.
Spencer TR, Marks RD, Fenn JO, Jenrette JM, Lutz MH. Intraperitoneal P32 after negative second look laparotomy in ovarian
carcinoma. Cancer 1989; 63:2434-7.
Markman M, George M, Hakes T, Reichman B, Hoskins W, Rubin S, et al. Phase I1 trial of intraperitoneal mitoxantrone in the
management of refractory ovarian cancer. ] Clin Oncol 1990;
8:146-50.
Markman M, Hakes T, Reichman B, Lewis JL, Rubin S, Jones W,
et al. Phase I1 trial of weekly or biweekly intraperitoneal mitoxantrone in epithelial ovarian cancer. ] Clin Oncol 1991; 9:97882.
Dufour P, Maloisel F, Bergerat JP, Barats JC, Dellenbach P, Renaud R, et al. Intraperitoneal mitoxantrone as consolidation treatment for stage 111 ovarian carcinoma: a pilot study. Bull Cancer
1991; 78~273-80.