Phase II trial of gemcitabine in patients with advanced sarcomas (E1797)

A trial of the Eastern Cooperative Oncology Group


  • This study was conducted by the Eastern Cooperative Oncology Group (Robert L. Comis, M.D., Chair).

  • The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute.



The current study was conducted to evaluate the antitumor activity and toxicity of gemcitabine in patients with advanced sarcoma.


Twenty-five patients with advanced sarcomas, who previously were untreated for metastatic disease, were treated on an Eastern Cooperative Oncology Group Phase II study. Patients ranged in age from 27 to 79 years, with a median age of 59 years. The most common histology was leiomyosarcoma (54%). The grades of the tumors were high in 40%, moderate in 24% and low in 12%. Gemcitabine was given at a dose of 1250 mg/m2 as a 30-minute infusion weekly for 3 weeks followed by 1 week of rest.


One of the 25 patients (4%) (90% confidence interval [90% CI], 0–18%) achieved a partial response lasting 8 months. The estimated overall median survival was 15 months. The 1-year estimated survival rate was 63% (90% CI, 47–84%). The estimated median progression-free survival (PFS) was 13 months with a 1-year PFS rate of 56% (90% CI, 41–76%). Grade 3–4 toxicities (by CTC criteria) were observed in all 25 patients. No lethal toxicity (Grade 5) related to treatment was found.


The results of the current study demonstrated that gemcitabine given at this schedule and dose in this population of patients with advanced sarcoma had limited activity. Cancer 2003;97:1969–73. © 2003 American Cancer Society.

DOI 10.1002/cncr.11290

Sarcomas are a diverse group of tumors with varying biologic activity and responses to chemotherapy. The current management of patients with advanced sarcomas is largely palliative. Doxorubicin and ifosfamide currently are the most active agents against sarcomas with a response rate of approximately 25%.1 To our knowledge, combination regimens with or without doxorubicin have failed to demonstrate an improved overall survival.2, 3 Clinical trials of newer agents are needed to identify active drugs.

Gemcitabine, a nucleoside antimetabolite, is an analog of deoxycytidine that has demonstrated clinical activity in a variety of solid tumors including nonsmall cell lung, ovarian, head and neck, and pancreatic carcinomas.4 In preclinical studies, gemcitabine has shown activity against xenografts of sarcoma cell lines.5 Recent reports suggest some activity in patients with advanced sarcomas.6

The Eastern Cooperative Oncology Group (ECOG) conducted a Phase II trial of gemcitabine to determine the antitumor activity and toxicity in patients with advanced sarcomas.



Patients age ≥ 18 years; with an ECOG performance status of 0, 1, or 2; with histologically confirmed sarcoma; with measurable disease; who were previously untreated with chemotherapy for metastatic disease; with adequate organ function (defined as leukocytes ≥ 3500/uL, platelets ≥ 150,00/uL, total bilirubin ≤ 1.5 × upper limit of normal [ULN], aspartate aminotransferase ≤ 3 × ULN, and serum creatinine ≤ 1.5 × ULN), who were nonpregnant or nonlactating females; with no uncontrolled infection; with no prior history of malignancy except previously treated basal cell carcinoma, squamous cell carcinoma of the skin, or Stage 0-IIA squamous cell carcinoma of the cervix; and who had no active central nervous system metastasis were eligible. Patients whose measurable disease was previously irradiated also were eligible if there was a > 25% increase in size after radiation. Signed, informed consent was obtained from all patients. The study was approved by each institution's institutional review board.

Treatment Plan

Gemcitabine was administered at a dose of 1250 mg/m2 as a 30-minute infusion weekly for 3 weeks followed by 1-week of rest. Tumor assessment was performed every other cycle. If a response occurred, a confirmatory tumor assessment was performed at the next cycle. Dose modification (decrease of 25%) for an absolute neutrophil count (ANC) of 500–999 × 109/L or a platelet count of 50,000–99,999/uL was required. Treatment was held for any nonhematologic toxicities numbering ≥ 3 or an ANC < 500 or a platelet count < 50,000/uL. Treatment was resumed with a 25% dose reduction if nonhematologic toxicities per Common Toxicity Criteria (CTC) improved to Grade < 2. No dose escalations were planned. Treatment was continued until maximal response, tolerance, or disease progression.

Response Criteria

Patients who achieved a complete disappearance of all clinically detectable malignant disease for ≥ 4 weeks were considered to have achieved a complete response (CR). Patients who did not achieve a CR but had a response of a > 50% reduction in tumor size for ≥ 4 weeks without an increase in the size of any area of known malignant disease of > 25% or the appearance of new areas of malignant disease were considered to have achieved a partial response (PR). Patients with no significant change in measurable or evaluable disease for ≥ 4 weeks were considered to have stable disease (SD). Progressive disease was defined as a > 25% increase in size of the target lesion or the development of new lesions.

Statistical Analysis

This study used a two-stage design by Simon.7 If the response rate was ≥ 30%, this agent would be considered to have antitumor activity. Accrual to the study was to be stopped if < 3 responders were observed among the first 22 patients, otherwise an additional 19 patients would be entered to the study. Exact confidence intervals (90% CI)7 were used for the reporting of response rates. Overall survival corresponded from the time of entry to the study until death or censoring at the time of last follow-up. Progression-free survival (PFS) corresponded to the time to death, disease progression, or censoring, whichever came first. The median survivals were calculated using the method of Kaplan and Meier.8


Patient Characteristics

Twenty-five patients from 14 institutions were registered onto the current study. Each institution provided 1 or 2 patients other than the Mayo Clinic, which had 5 patients (20%), and the Washington Medical Center, which had 4 patients (16%). At the time of randomization, patients ranged in age from 27–79 years with a median age of 59 years. The majority of patients was white (72%) and female (64%). Baseline characteristics are shown in Table 1. Prior therapy and site of disease are shown in Tables 2 and 3. The histology, grade, and site of the primary tumor are listed in Table 4, with leiomyosarcoma being the most commonly reported histology (54%).

Table 1. Baseline Characteristics
Patient characteristicsNo.%
  1. ECOG: Eastern Cooperative Oncology Group.

Age (yrs)  
ECOG Performance status  
Table 2. Previous Treatment
Patient characteristicsNo.%
 Yes, external beam728
Systemic adjuvant therapy  
Table 3. Known Site of Metastasisa
Patient characteristicsNo.%
  • a

    Multiple sites of metastasis recorded. Numbers do not add up to 100%.

Subcutaneous tissue624
Table 4. Histology, Grade, and Primary Site
Patient CharacteristicsNo.%
  1. NOS: not otherwise specified.

 Spindle cells, NOS520
 Osteogenic sarcoma14
 Cannot be assessed624
Primary tumor site  
 Upper limb28
 Lower limb520

Response Data

There was 1 PR that lasted 8 months in a patient with an epithelioid sarcoma. Thus, the overall objective response rate was 4% (1 of 25 patients) (90% CI, 0–18%). SD occurred in 8 patients (32%) and disease progression occurred in 14 patients (56%), with 2 patients (8%) reported to be unevaluable. Among the 25 patients, 1 patient was lost to follow-up, this patient was censored at the time of last follow-up. The estimated median follow-up time for the current study was 10 months. Eleven of the 25 patients had died at the time of last follow-up. The estimated median survival was 15 months. The 1-year survival rate was 63% (90% CI, 47–84%). The estimated median PFS was 13 months, with a 1-year PFS rate of 56% (90% CI, 41–76%) (Fig. 1). Figure 2 shows the Kaplan–Meier estimate of survival.

Figure 1.

Progressive-free survival of patients in the current study.

Figure 2.

Overall survival of patients in the current study.

Toxicity Data

Table 5 summarizes the acute toxicities associated with treatment. All 25 patients experienced either Grade 3 or 4 toxicity. No Grade 5 toxicity related to treatment was observed.

Table 5. Incidence of Toxicity in 25 Patients
  1. DIC: disseminated intravascular coagulopathy.

Weight loss1-
Infection with ≥ Grade 3 neutropenia2-
Infection without neutropenia1-
Muscle weakness1-
Neuropathy, cranial1-
Pain, other11
Pneumonitis/pulmonary infiltrates1-
Erectile impotence1-


The results of chemotherapy for advanced sarcomas has been disappointing. Attempts to improve on the single-agent activity of doxorubicin or ifosfamide with dose intensification or combination chemotherapy has not demonstrated a consistent improvement in overall survival.2, 9, 10 The purpose of the current study was to evaluate the use of gemcitabine in the treatment of patients with advanced sarcomas. With only 1 PR reported, the current study failed to demonstrate that gemcitabine is active in patients with sarcoma.

Gemcitabine has been studied in other trials against sarcomas.6, 11–15 Using the same schedule of gemcitabine as in the current study, Okuno et al. reported only 1 PR (patient with uterine leiomyosarcoma) among 29 patients.12 In another study, previously treated patients were treated with gemcitabine at a dose of 1000 mg/m2 and 1 PR was reported in a patient with leiomyosarcoma among 18 patients.6 Patel et al., using gemcitabine at a dose of 1000 mg/m2, reported no responses in 17 patients with gastrointestinal (GI) leiomyosarcoma, but 7 PRs were reported among 39 patients with other soft tissue sarcomas, for a response rate of 18% (95% CI, 7–29%).11 Responses were reported in four of the patients with non-GI leiomyosarcomas.

It appears that gemcitabine does have activity against certain histologies in some Phase II studies of sarcoma. One can postulate that the poor response rate in the current trial as well as that from the Mayo Clinic can be attributed to the inclusion of a significant percentage of chemotherapy-resistant GI leiomyosarcomas.16 Unfortunately, because of the small numbers of individual histologies in each of these trials, firm conclusions regarding the activity of gemcitabine against specific histologies are unknown. However, there is a suggestion that gemcitabine is active in patients with non-GI leiomyosarcomas, especially uterine leiomyosarcoma, either alone or in combination with docetaxel (Table 6).13, 14

Table 6. Studies of Gemcitabine in Sarcoma
StudyNo. of patientsDose (mg/m2)Infusion (min)RR %Comments
  • NA, not available; RR: response rate; ECOG: Eastern Cooperative Oncology Group; 9: every; LMS: leiomyosarcoma; GIST: gastrointestinal stromal tumors; NA: not available.

  • a

    After run in of 7 weeks out of 8 weeks.

ECOG (current study)251250 q wk × 3304Response in epithelial sarcoma
Okuno et al. (Mayo)12251250 q wk × 3304Response in uterine LMS
Patel et al. (M. D. Anderson)1117 (GIST)1000 q wk × 3a300
 39 (non-GIST)1000 l wk × 3a3018Response in 3 uterine LMS
Spath-Schwalbe et al.1418200 q wk36011Response in uterine LMS
Brambilla et al.15111200 q wk × 2aNA100All patients with classical Kaposi sarcoma
Merimsky et al.6181000 q wk × 3aNA5.5Response in uterine LMS

Future trials in patients with sarcomas should be designed to include adequate numbers to determine the activity of newer agents for the common histologies. All sarcoma histologies do not behave the same way. In addition, histology-specific trials such as the randomized Phase III trial of STI-571 for the treatment of c-kit-positive GI stromal tumors will provide stronger evidence of the activity of newer agents.17, 18