Phase II Trial of the Pegylated Liposomal Doxorubicin in

Previously Treated Metastatic Endometrial Cancer: A

Gynecologic Oncology Group Study

By Franco M. Muggia, John A. Blessing, Joel Sorosky, and Gary C. Reid

Purpose: To determine whether pegylated liposo- 2.1, 3.3, and 5.4 months; the overall response rate was
mal doxorubicin (PLD) has antitumor activity in pre- 9.5% (95% confidence interval, 2.7% to 22.6%). Three
treated patients with persistent or recurrent endome- of these responses (in liver and in lymph node) occurred
trial carcinoma and to define the nature and degree of in patients who had progressed after doxorubicin with
toxicity of PLD. either paclitaxel or cisplatin. The median number of
Patients and Methods: Women with histologically courses was 2.5 (range, one to 14). Toxicity was gen-
documented recurrent or persistent measurable endo- erally mild: only 25 patients experienced leukopenia,
metrial carcinoma and with failure of one prior treat- with a median WBC count of 2,900 (range, 800 to
ment regardless of prior anthracycline therapy were 3,900) at nadir. The only grade 4 toxicities were one
enrolled. PLD was administered intravenously over a episode each of esophagitis, hematuria, and vomiting.
1-hour period at a dose of 50 mg/m2 every 4 weeks; the The median overall survival was 8.2 months.
dosage was modified in accordance with observed toxicity. Conclusion: PLD has only limited activity in pre-
Results: Of 46 patients entered, 42 were assessable treated advanced, recurrent endometrial cancer, but
for response, as three were declared ineligible on cen- further trials in anthracycline-naive patients and in
tral pathology review and one was not assessable for previously untreated patients are ongoing. Its toxicity
response. Forty had received prior chemotherapy, 11 profile should permit its use in combination with myelo-
hormonal therapy, and 29 radiation therapy. Doxoru- suppressive drugs.
bicin had been given to 32 patients, carboplatin with J Clin Oncol 20:2360-2364. © 2002 by American
paclitaxel to six, carboplatin to one, and fluorouracil to Society of Clinical Oncology.
one. Four patients had partial responses lasting 1.1,

OST PATIENTS with endometrial cancer present stage I endometrial cancer2 but not alter their survival. Sys-
M with localized stages amenable to treatment with
surgery. A small percentage of these women have unfavor-
temic chemotherapy has been explored in an effort to improve
the outcome of those patients with poorly differentiated tumors
able features that place them at increased risk for local or that have recurred or are destined to metastasize. However, the
distant recurrences. An even smaller percentage are found to role of cytotoxic agents remains uncertain because of both
have advanced disease with distant metastases when ini- efficacy and toxicity considerations.
tially diagnosed. Several decades ago, progestational agents Doxorubicin has been the most frequently studied cyto-
were noted to be effective, particularly against pulmonary toxic drug in endometrial cancer, after demonstration of its
metastases.1 Unfortunately, patients responding to such activity when given alone and in combination.3,4 The
treatments are those with well-differentiated tumors that are Gynecologic Oncology Group (GOG) subsequently estab-
least likely to metastasize. Postoperative radiation therapy has lished that response rates and progression-free survival after
been found to decrease recurrences in a subset of patients with doxorubicin plus cisplatin were superior to those achieved
by doxorubicin as a single agent.5 More recently, paclitaxel
was established as another active drug (Lincoln et al,
manuscript submitted for publication),6 and the results of a
From the Department of Medicine, Kaplan Cancer Center, New York
University Medical Center, New York, and Gynecologic Oncology study by the GOG comparing doxorubicin plus cisplatin
Group, Roswell Park Cancer Institute, Buffalo, NY; Division of versus doxorubicin plus paclitaxel were recently reported.7
Gynecologic Oncology, University of Iowa Hospitals and Clinics, Iowa Unfortunately, the therapeutic achievements of these com-
City, IA; and Riverside Methodist Hospital, Columbus, OH.
binations remain modest with regard to improving survival,
Submitted August 31, 2001; accepted January 14, 2002.
Supported by National Cancer Institute grant no. CA 27469 to the and the toxicities of the regimens often pose an obstacle for
Gynecologic Oncology Group Administrative Office and grant no. CA initial or continued treatment with these cytotoxic drugs.
37517 to the Gynecologic Oncology Group Statistical Office. Pegylated liposomal doxorubicin (PLD; doxorubicin HCl
Address reprint requests to Gynecologic Oncology Group Adminis- liposome injection [Doxil, Alza Pharmaceuticals, Palo Alto,
trative Office, 4 Penn Center, Suite 1020, 1600 John F. Kennedy Blvd,
Philadelphia, PA 19103.
CA]) has been documented to have a markedly different
© 2002 by American Society of Clinical Oncology. pharmacology and toxicity spectrum in relation to the free
0732-183X/02/2009-2360/$20.00 drug.8 It also has been shown to have antitumor activity

2360 Journal of Clinical Oncology, Vol 20, No 9 (May 1), 2002: pp 2360-2364
DOI: 10.1200/JCO.2002.08.171
Information downloaded from [Link] and provided by at Virginia Tech on March 13, 2015 from [Link]
Copyright © 2002 American Society of Clinical Oncology. All rights reserved.
PLD IN METASTATIC ENDOMETRIAL CANCER 2361

against platinum- and paclitaxel-resistant ovarian cancer9 Table 1. Patient Characteristics (n ⴝ 42)
and against previously untreated breast cancer.10 Activity in Characteristic No. of Patients
patients who had previously received doxorubicin suggests Age, years
that one might overcome resistance with this liposomal Median 62.5
formulation.11 This could occur if the amount delivered to Range 40-79
the tumor were to be considerably greater than that provided Ethnicity
White 34
by the free drug or if continued exposure were to result in Black 6
some advantage in overcoming MDR1 P-glycoprotein– Other 2
mediated drug efflux as a cause of resistance. Moreover, GOG performance
PLD has the potential advantage of allowing repeated status
0 19
administration with a lesser likelihood of cumulative car-
1 17
diotoxicity12,13 and, therefore, a greater acceptability by 2 6
older patients with cardiac risk factors. Accordingly, the Grade
GOG initiated a study of PLD in patients with persistent or 1 7
recurrent endometrial cancer for whom one treatment pro- 2 17
tocol of higher priority had failed. 3 18
Prior radiotherapy 29
Prior chemotherapy 40
PATIENTS AND METHODS Prior hormonal therapy 11
Eligible patients had to have histologically confirmed persistent or Courses, n
recurrent endometrial carcinoma and the presence of measurable Median 2.5
disease. All epithelial cell types were eligible. The GOG performance Range 1-14
status was to be 0, 1, or 2. At least 3 weeks had to elapse from prior
therapy, and recovery had to take place from any recent treatment.
Patients were required to have adequate organ function as evidenced by count was greater than 1,500/␮L, the platelet count was greater than
WBC count ⱖ 3,000/␮L, granulocyte count ⱖ 1,500/␮L, platelet count 100,000/␮L, and the patient had recovered fully from any nonhema-
ⱖ 100,000/␮L, creatinine level ⱕ 2.0 mg/100 mL, bilirubin level ⱕ 1.5 tologic toxicity. Dose modification by 25% was based on nadir grade 4
times the institutional normal, and AST and alkaline phosphatase ⱕ platelets or neutrophils, or any grade 3 or greater nonhematologic
three times the institutional normal. A normal left ventricular ejection toxicity (expected to be primarily skin or mucosal toxicities).
fraction by radionuclide multiple-gated acquisition (MUGA) scan was Patients were assessable for response after receiving PLD and being
required. Patients had to have received one and only one prior observed for at least 4 weeks. Objective responses were defined by
chemotherapy regimen or to have qualified for one prior study for standard GOG criteria.14
recurrent endometrial cancer. Prior treatment with doxorubicin was
RESULTS
allowed, as long as the cumulative dose level was ⱕ 350 mg/m2.
Therefore, patients who entered and completed GOG Protocol 163 After at least three objective responses were recorded
(doxorubicin/cisplatin versus doxorubicin/paclitaxel) were eligible to
during the entry of the first 24 patients, the second-stage
enter this study.
No patient could have had a previous or concomitant malignancy accrual was opened and completed with 46 entries. Three
with the exception of nonmelanoma skin cancer. Written informed patients were ineligible: one patient (who actually exhibited
consent was obtained from all patients before entry onto the study, a partial response) was disqualified because of a synchro-
fulfilling all institutional, state, and federal regulations. nous ovarian primary, and one each with wrong pathology
Pretherapy evaluation included history and physical examination,
and wrong primary site. One patient was assessable only for
performance status, tumor measurement(s), complete blood cell count
with differential and platelet count, serum creatinine, bilirubin, AST toxicity. Thus 42 patients were assessable for response.
and alkaline phosphatase, chest roentgenogram, ECG MUGA radionu-
clide heart scan, urinalysis, and serum CA-125 level. Complete blood Patient Characteristics
cell count, differential, and platelet count were repeated weekly, and Table 1 summarizes patient characteristics. The median age
hematologic toxicity grading was based on nadir counts. The MUGA
was 62.5 years (range, 40 to 79 years). Nearly half had
scan and ECG were repeated when a cumulative dose of 300 mg/m2
was reached, and they were repeated every other cycle of therapy. performance status 0. Histologic features included a predomi-
Blood chemistry analyses were repeated every 4 weeks except for the chest nance of grades 2 and 3, and several patients had unfavorable
roentgenogram and the CA-125 test, which were repeated if results were histologic characteristics: three had adenosquamous features,
initially abnormal. Other tests were repeated as clinically indicated. five had uterine papillary serous cancers, and two had predom-
PLD was administered as a 1-hour infusion at the initial dose of 50
inant clear-cell components. Two patients had ureteral stents at
mg/m2 every 4 weeks. Premedication including dexamethasone 20 mg
intravenously, together with diphenhydramine 50 mg and cimetidine the outset, and one had hypercalcemia. The initial CA-125 was
300 mg, was recommended during the first and subsequent courses. greater than 30 units in 27 patients and less than 30 in 13
Subsequent courses of PLD were not administered until the granulocyte patients; in two it was not obtained. More than half (29

Information downloaded from [Link] and provided by at Virginia Tech on March 13, 2015 from [Link]
Copyright © 2002 American Society of Clinical Oncology. All rights reserved.
2362 MUGGIA ET AL

Table 2. Characteristics of Four Partial Responders

Patient No. Age (years) Features at Initial Diagnosis Months to Recurrence Treatment Before PLD (time) Comments on Response

1 71 Stage IV (pulmonary metastases), 5 AP ⫻ 3, RT to pelvis, PR in lung, refused

well-differentiated, some areas progression in lung lesion additional doses,
of squamous differentiation lasted 1.1 month
2 43 Stage III, well-differentiated 63 Surgery, RT to supraclavicular PR in lymph nodes,
lymph nodes, AT ⫻ 7, then lasted 2.1 months
5 months, progression
3 69 Stage III (ovarian metastases 19 Surgery ⫹ RT, CPX ⫻ 4, CAP PR in liver, lasted 3.3
moderately well-differentiated) ⫻ 2, progression months
4 73 Stage I bulky adenosquamous, 84 Surgery, hormones (⫾); FU/ PR in lungs, lasted
25% myometrial invasion LV for lung metastases 5.4 months

Abbreviations: FU, fluorouracil; LV, leucovorin; PR, partial response; RT, radiation therapy; CPX, cyclophosphamide; CAP, cyclophosphamide, doxorubicin, and
cisplatin; AT, doxorubicin and paclitaxel; AP, doxorubicin and platinum (cisplatin).

patients) had received prior radiation. Prior chemotherapy had therapy, with stable disease as best response in one patient
been given to 40 patients: 32 had received prior doxorubicin receiving paclitaxel.
alone or in combination, six had received carboplatin and
paclitaxel in combination, one had received carboplatin alone, Adverse Effects
and one had received fluorouracil with leucovorin. Eleven Forty-three patients were assessable for toxicity (Table
patients were also pretreated with hormones (eight with meges- 3). There were no treatment-related deaths. Fourteen pa-
trol acetate and three with goserelin); one of the 11 had also tients had subsequent MUGA scans, and seven of these had
received anastrazole. three or more additional determinations. One patient, age
64, experienced a 25% decrease in ejection fraction (from
Treatment Outcome and Responses 68% to 51%). She had received five courses of doxorubicin
in GOG 163 (combined with paclitaxel 150 mg/m2 per
The median number of courses was 2.5 (range, one to 14).
course) prior to 200 mg/m2 of PLD, and two other patients
Four patients achieved a partial response, out of 42 assess-
also had a decrease in their left ventricular ejection fraction
able patients (9.5%). Response duration was 1.1, 2.1, 3.3,
(22% and 24%, respectively). These two patients had
and 5.4 months. Three of the responses occurred in the 32
patients who had previously received doxorubicin, one of
whom had completed seven cycles in GOG 163. The 95%
Table 3. Adverse Effects (n ⴝ 43)
confidence interval for the true response rate in these
Grade
pretreated patients is 2.7% to 22.6%. One partial response
Adverse Effect 1 2 3 4
was observed among the eight patients pretreated with
nondoxorubicin combination chemotherapy. Table 2 shows Neutropenia 3 6 5 2
Thrombocytopenia 9 0 0 0
the characteristics of the four responders. The number of
Anemia 6 11 4 1
courses received by responding patients was two, three, 10, Mucositis* 6 7 0 1
and 14. The short response was observed in a patient who Anorexia 2 3 0 0
preferred to discontinue therapy after only two courses. Nausea and/or vomiting 4 8 1 0
Twenty-nine percent experienced stable disease for at least Other gastrointestinal† 6 3 3 0
Dermatologic 9 4 4 0
two cycles, and the median overall survival was 8.2 months.
Neurotoxicity 4 0 2 0
It is noteworthy that four patients developed brain metasta- Infusion reaction 1 1 0 0
ses (two including the cerebellum) during the course of Alopecia 2 4 0 0
therapy, and four additional patients developed vertebral Fever 2 2 1 0
metastases with neurologic complications in two. One Pulmonary 1 1 1 0
Cardiovascular‡ 0 3 1 0
patient had resection of lung metastases, and another had a
complete response to subsequent megestrol acetate in a *Includes stomatitis, esophagitis, bronchitis, vaginitis, nasal soreness, and
rectal bleeding.
vaginal recurrence arising in a previously radiated field.
†Includes two events beyond grade 1 for diarrhea and weight loss, and one
These two patients were still alive and had no evidence of each of constipation, dehydration, and indigestion.
disease at 32 and 43 months since study entry, respectively. ‡Refers to three patients who had decreased left ventricular ejection fraction
Five other patients went on to receive additional chemo- (see text).

Information downloaded from [Link] and provided by at Virginia Tech on March 13, 2015 from [Link]
Copyright © 2002 American Society of Clinical Oncology. All rights reserved.
PLD IN METASTATIC ENDOMETRIAL CANCER 2363

received two and five doses of doxorubicin-containing survival varying between 7 and 11 months, have been
therapy and then received two and eight courses of PLD, observed. Alternating sequences of both of these drugs
respectively. The pulmonary toxicity represented dyspnea, are under study.
and the two grade 3 neurologic toxicities were fatigue and Against this background, the results achieved in the
weakness. Neutropenia was the most common hematologic current study with PLD are of interest in two respects: (1)
toxicity, thrombocytopenia being confined to only nine some antitumor activity was observed despite prior cyto-
patients with grade 1 changes. Among nonhematologic toxic therapy (including prior doxorubicin in most patients),
toxicities, dermatologic problems were most common, and and (2) the toxicity profile should allow testing of PLD-
severe gastrointestinal toxicities were rare. containing combinations even with myelosuppressive
Dose reductions were required in six patients. These oc- drugs. Four instances of severe (grade 3) skin toxicity may
curred at the second doses in three patients and at the third, reflect the slightly higher dose intensity (12.5 mg/m2/wk)
fifth, and sixth doses in the others. Seven patients discontinued than is commonly used (10 mg/m2/wk).
drug treatment because of side effects, and two others (one of Additional clinical findings in these patients receiving
whom was a responder) refused further treatment for no clear
second-line therapy for endometrial cancer deserve com-
reason. The given reasons for discontinuation of therapy
ment. Prolonged treatment may predispose to observing new
include stomatitis in two instances and loss of appetite,
complications of the disease, such as the four patients with
weakness, fatigue, abdominal pain with dyspnea, and a de-
brain metastases and four others with involvement of the
crease in ejection fraction in one instance each. Thus, except
vertebral column. Encountering these complications may be
for mucosal toxicity, other reasons were quite variable.
the result of either lengthening survival with metastatic disease,
DISCUSSION or the greater medical surveillance provided to patients receiv-
When patients with endometrial cancer manifest recur- ing treatment, or both. Furthermore, the occurrence of cerebel-
rent or metastatic disease, endocrine therapy and/or chemo- lar metastases in two patients may indicate an unusual tropism
therapy is usually administered. Toxicity considerations of endometrial cancer for this site: three instances of cerebellar
often preclude using chemotherapy such as doxorubicin metastases were recorded among 10 patients with endometrial
alone or with cisplatin or paclitaxel. Therefore, efforts continue cancer metastatic to the CNS.15 By contrast, in a literature
in identifying active drugs or drug combinations that may have review of ovarian cancer metastatic to the brain, only seven
advantages over current chemotherapy regimens. Recent stud- (8.7%) of 80 had cerebellar metastasis.16
ies have defined better the indications and limitations of In summary, PLD had only limited antitumor activity in this
hormonal therapy and have identified poor-risk entities such as extensively pretreated population with endometrial cancer, but
papillary serous and clear-cell histologies. In GOG studies, minimal myelosuppression, even in patients who had received
response rates of 25% or less for medroxyprogesterone acetate, radiation. This should facilitate the study of combinations with
and of 10% for tamoxifen, with median durations of other active agents, such as taxanes or platinum compounds.

APPENDIX

The appendix listing the Gynecologic Oncology Group institutions that participated in this study is available online at
[Link].

REFERENCES
1. Thigpen JT, Brady M, Alvarez RD, et al: Oral medroxyproges- 4. Muggia FM, Perloff M, Chia GA, et al: Adriamycin (NSC
terone acetate in the treatment of advanced or recurrent endometrial 123127) in combination with cyclophosphamide (NSC 262710): A
cancer: A dose-response study by the Gynecologic Oncology Group. phase I/II evaluation. Cancer Chemother Rep 58:919-926, 1974
J Clin Oncol 17:1736-1744, 1999 5. Thigpen JT, Blessing JA, Homesley H, et al: Phase III trial of
2. Look KY: Who benefits from radiotherapy treatment of endo- doxorubicin ⫹/⫺ cisplatin in advanced or recurrent endometrial
metrial cancer and at what price? Lancet 355:1381-1382, 2000 carcinoma: A Gynecologic Oncology Group (GOG) study. Proc Am
(editorial) Soc Clin Oncol 12:261, 1993 (abstr 830)
3. Thigpen JT, Blessing JA, DiSaia PJ, et al: A randomized 6. Ball HG, Blessing JA, Lentz SS, et al: A phase II trial of Taxol in
comparison of doxorubicin alone versus doxorubicin plus cyclophos- advanced and recurrent adenocarcinoma of the endometrium: A Gyneco-
phamide in the management of advanced or recurrent endometrial logic Oncology Group study. Gynecol Oncol 56:120, 1995 (abstr 42)
carcinoma: A Gynecologic Oncology Group study. J Clin Oncol 7. Fleming GF, Brunetto VL, Bentley R, et al: Randomized trial of
12:1408-1414, 1994 doxorubicin (DOX) plus cisplatin (CIS) versus DOX plus paclitaxel

Information downloaded from [Link] and provided by at Virginia Tech on March 13, 2015 from [Link]
Copyright © 2002 American Society of Clinical Oncology. All rights reserved.
2364 MUGGIA ET AL

(TAX) plus granulocyte colony-stimulating factor (G-CSF) in patients 12. Berry G, Billingham M, Alderman H, et al: The use of cardiac
with advanced or recurrent endometrial cancer: A report on Gyneco- biopsy to demonstrate reduced cardiotoxicity in AIDS Kaposi’s sar-
logic Oncology Group (GOG) Protocol 163. Proc Am Soc Clin Oncol coma treated with pegylated liposomal doxorubicin. Ann Oncol 9:711-
119:379a, 2000 (abstr 1498) 716, 1998
8. Uziely B, Jeffers S, Isaacson R, et al: Liposomal doxorubicin: 13. Safra T, Muggia F, Jeffers S, et al: Pegylated liposomal
Antitumor activity and unique toxicities during two complementary doxorubicin (Doxil): Reduced clinical cardiomyopathy in patients
phase I studies. J Clin Oncol 13:1777-1785, 1995 reaching or exceeding cumulative doses of 500 mg/m2. Ann Oncol
9. Muggia FM, Hainsworth J, Jeffers S, et al: Phase II study of Doxil 11:1029-1033, 2000
in refractory ovarian cancer: Antitumor activity and toxicity modifica- 14. Blessing JA: Design, analysis and interpretation of chemother-
tion by liposomal encapsulation. J Clin Oncol 15:987-983, 1997 apy trials in gynecologic cancer, in Deppe G (ed): Chemotherapy of
10. Ranson MR, Carmichael J, O’Byrne K, et al: Treatment of Gynecologic Cancer (ed 2). New York, NY, Alan R Liss Inc, Scientific
advanced breast cancer with sterically stabilized liposomal doxorubicin: and Medical Publications, 1990, pp 63-97
Results of a multicenter phase II trial. J Clin Oncol 15:3185-3191, 1997 15. Cormio G, Lissoni A, Losa G, et al: Brain metastases from
11. Gabizon A, Uziely B, Lotem M, et al: Doxil in patients with endometrial carcinoma. Gynecol Oncol 61:40-43, 1996
pretreated metastatic breast cancer (MBC): A dose-schedule findings 16. Sood A, Kumar L, Sood R, et al: Epithelial ovarian carcinoma
study with pharmacokinetics. Proc Am Soc Clin Oncol 16:147, 1997 metastatic to the central nervous system: A report on two cases with
(abstr 516) review of literature. Gynecol Oncol 62:113-118, 1996

Information downloaded from [Link] and provided by at Virginia Tech on March 13, 2015 from [Link]
Copyright © 2002 American Society of Clinical Oncology. All rights reserved.