Phase II study of a protracted irinotecan schedule in children with refractory or recurrent soft tissue sarcoma




Irinotecan (CPT-11) is a novel antineoplastic agent that takes effect by inhibiting topoisomerase I. The Italian Soft Tissue Sarcoma (STS) Committee performed a multiinstitutional Phase II study to evaluate its effect on STS.


Over a 2-year period between 2002 and 2004, 32 heavily pretreated patients were administered 60-minute infusions of irinotecan at 20 mg/m2/day, for 5 days a week, for 2 consecutive weeks. The courses were repeated every 4 weeks for at least 2 courses, unless there were signs of toxicity or disease progression. Thirty patients, 13 with peripheral primitive neuroectodermal tumor (PNET), 12 with rhabdomyosarcoma (RMS), 3 with desmoplastic small round cell tumor (DSRCT), and 2 with other STS were evaluable for response.


A total of 79 cycles were delivered. The main regimen-related toxicity was diarrhea, occurring in 58% of cycles with 9 episodes graded as 3 or 4. Grade 3–4 neutropenia was recorded in 10% of cycles. The overall response rate was 23% (2 complete remissions + 5 partial remissions of 30 patients), 38% for PNET and 16% for RMS. In addition, 4 minor responses were noted.


As a single agent in the treatment of recurrent and refractory STS, irinotecan administered on a daily ×5 ×2 schedule revealed a noteworthy response rate in a population of heavily pretreated patients, especially in the subset of patients with PNET. Its hematologic toxicity profile warrants further investigation in association with other myelotoxic agents. Cancer 2006. © 2005 American Cancer Society.

Although the prognosis for children with cancer has generally improved dramatically in the last 3 decades, little or no progress has been made for some groups of patients, which include children with metastatic or recurrent soft tissue sarcoma (STS), who have a poor prognosis and need new therapies.

Campthotecin derivatives are new anticancer agents that interfere with DNA replication and cell division, inhibiting topoisomerase I activity.1 Irinotecan (CPT-11) is a semisynthetic, water-soluble campthotecin analog with an S-phase specific activity. CPT-11 and its metabolite, SN38, have produced promising results in preclinical studies on a broad spectrum of human tumor xenografts derived from pediatric tumors such as rhabdomyosarcoma (RMS), neuroblastoma, and peripheral primitive neuroectodermal tumor (PNET).2, 3 In particular, xenograft studies have shown that the activity of CPT-11 is schedule-dependent. A given total dose of drug produced dramatically better responses when administered on a protracted schedule (using a daily administration for 2 consecutive 5-day courses, repeated every 21 days).2 This effect is consistent with the S-phase specific cytotoxic action of CPT-11. The protracted administration schedule was consequently translated into a Phase I study in children that demonstrated a low myelotoxicity but more severe gastrointestinal effects than the shorter term schedules.4

Initial findings showed that CPT-11 deserved appropriate studies to analyze its activity against pediatric tumors, but few Phase II studies have been published so far. The aim of this study was to ascertain the activity of CPT-11 administered in a protracted schedule to a group of children with STS.


Between July 2002 and September 2004, 32 patients younger than 19 years old joined this Phase II trial. They were registered from 13 centers belonging to the AIEOP (Associazione Italiana di Ematologia e Oncologia Pediatrica) and were taking part in studies coordinated by the AIEOP STS Committee.

Eligible patients were required to have a histologically confirmed diagnosis of surgically unresectable malignant STS and to be refractory or relapsing after 1 or more standard treatment regimens. Other eligibility criteria were a life expectancy of at least 8 weeks, a modified Lansky score of ≥50, recovery from toxic effects of prior chemotherapy, a hemoglobin level greater than 9 g/dL, an absolute neutrophil count greater than 1500/mm3, a platelet count higher than 100,000/mm3, adequate liver function (bilirubin level ≤1.5 mg/100mL; alanine aminotransferase ≤ twice the normal value), adequate renal function (serum creatinine concentration ≤1.5 mg/dL or creatinine clearance >60 mL/min/1.73 m2), and normal metabolic parameters (serum electrolytes, glucose, calcium, phosphorus). Patients with an interval of less than 3 weeks since the administration of radiotherapy (RT) or chemotherapy were excluded.

At baseline, the tumor was reassessed, with computed tomography (CT) or magnetic resonance imaging scans of disease sites, and measurements of all disease parameters, chest X-ray, chest CT scan, and whole-body technetium bone scan. Bone marrow aspirates and biopsies were collected if bone marrow was likely to be involved. Patients were monitored with twice-weekly physical examination, complete blood counts (until there was evidence of a recovery from hematopoietic toxicity), electrolytes (including calcium, magnesium, and phosphate), renal (creatinine, blood urea nitrogen) and liver (transaminases, bilirubin) function.

The study was approved by the Ethics Committees of each center taking part and informed consent was obtained from patients or parents, as appropriate.

Drug Dosage and Administration

The CPT-11 dose was 20 mg/m2/day; the drug was diluted in 0.9% sodium chloride solution to a final total volume of 100 mL and administered by 1-hour intravenous infusion once daily for 5 consecutive days followed by a 2-day rest and then another 5 consecutive days of treatment ([qd x 5] × 2) (see Fig. 1).

Figure 1.

Irinotecan schedule.

Cycles were given every 28 days, with neutrophils >1.0 × 109/L and platelets to >100 × 109/L and after resolution of nonhematopoietic toxicity. No use of colony-stimulating factors was allowed as a preventive measure.

Toxicity was graded using the National Cancer Institute Common Toxicity Criteria version 2.0. In the event of Grade 3–4 reversible nonhematologic toxicity or a treatment delay >1 week due to toxicity, a 25% dose reduction was recommended in the subsequent cycles. The administration of loperamide (0.1 mg/kg/dose) every 6–8 hours was recommended at the first episode of loose stools.

Response Evaluation

After 2 courses, a formal assessment of the primary tumor and all sites of metastases had to be performed. Response criteria were as follows: complete response (CR) = resolution of all evidence of disease; partial response (PR) = a tumor size reduction of >50% in the sum of the products of the 2 maximum perpendicular diameters of all measurable lesions; minor response (MR) = a reduction of <50% but >25% in the sum of the products of the 2 maximum perpendicular diameters of all measurable lesions. Stable disease or a reduction in size of <25% was recorded as no response, whereas an increase in tumor size or the detection of new lesions was considered as progressive disease (PD). CT scans had to be sent to the STS Committee and reviewed by the same radiologist in the case of CR or PR. Responses had to last at least 4 weeks after the assessment of the response.

Because of the difficulty in judging tumor response on bone marrow aspirates, we decided not to consider bone marrow in assessment of tumor response unless there was clear evidence of PD or a new lesion.

After the initial two cycles, any further CPT-11 courses were only to be administered if there was some degree of response or in patients with stable disease.

Statistical Method

This was an open label, nonrandomized, Phase II trial using a Gehan 2-step design. The expected effectiveness (π) was 20% for the whole group. If at least 1 response was recorded in the first 14 eligible patients, recruitment was to continue to at least 25 patients so that the standard error of the observed response rate would be ≤0.10.


Clinical Features

A total of 32 patients joined the study, 2 of whom were evaluable only for toxicity because their neoplastic lesion was not clearly evident on imaging investigations.

Patient characteristics are shown in Table 1. The age range was 1–18.5 years (median, 10.6 yrs), and 21 patients were male. The most common histotypes were PNET in 14 children and RMS in 13 (7 alveolar, 6 embryonal). The other STS were 3 desmoplastic small round cell tumors (DSRCT), 1 clear cell sarcoma, and 1 rhabdoid tumor.

Table 1. Patient Characteristics
No. of patients 
  1. RMS: rhabdomyosarcoma; PNET: primitive peripheral neuroectodermal tumor; DSRCT: desmoplastic small round cell tumors; hdCT: high-dose chemotherapy; PBSC: peripheral blood stem cells.

Gender, M/F21/11
 >10 yrs14
 <10 yrs18
 Desmoplastic small round cell tumor3
 Extrarenal rhabdoid tumor1
 Clear cell sarcoma1
Primary tumor site 
 Head and neck nonparameningeal2
 Head and neck parameningeal3
At initial diagnosis 
 Localized disease16
 Metastatic disease16
At the beginning of CPT-11 treatment 
 Locoregional relapse9
 Metastatic relapse10
 Local and metastatic relapse2
 Persistent disease4
 Progressive disease7
No. of previous chemotherapy regimens 
Prior radiotherapy23
Prior hdCT with PBSC rescue11

Tumors were located mainly in the extremities (5 RMS and 4 PNET), chest wall (5 PNET), or abdomen (2 PNET, 2 DSRCT, 1 RMS, 1 rhabdoid tumor). Five patients had head and neck RMS. Sixteen patients were metastatic at initial diagnosis (4 RMS, 8 PNET, 3 DSRCT, 1 clear cell sarcoma).

Patients had been treated previously with a median of two chemotherapy regimens (range, 1–3) before joining our study. Prior therapy included high-dose chemotherapy with stem cell rescue in 10 patients. A further patient had had an allogeneic bone marrow transplant. Second-line treatment had included topotecan in 2 patients. Radiation therapy had been administered to 23 patients.

At baseline, 9 patients had a locoregional and 12 had a distant relapse (lung in 7 cases, liver in 2, soft tissue and bone in 3, with concomitant lymph node involvement in 5), 4 had persistent disease at the end of first- or second-line treatment, and 7 had PD after second-line chemotherapy.


Considering only the 30 patients evaluable for chemotherapy response, 2 CR and 5 PR were documented, for an overall response rate of 23% (Table 2). A minor response was recorded in 4 cases.

Table 2. Response to CPT-11 according to Histology
 Embryonal RMSAlveolar RMSPNETDSRCTOtherTotal
  1. RMS: rhabdomyosarcoma; PNET: primitive peripheral neuroectodermal tumor; DSRCT: desmoplastic small round cells tumors; Other STS: clear cell sarcoma (minor response), rhabdoid tumor (no response).

No. of evaluable patients48133230
Complete response2  2
Partial response113  5
Minor response3 14
No response14  5
Progressive disease643114

In addition, 2 patients achieved a PR after the first CPT-11 cycle, followed by disease progression after the second cycle. Two partial responses were reported in 12 patients with RMS (16% response rate). Five responses (2 CR and 3 PR) were documented among the PNET patients, giving a 38% response rate.

When previous treatment is taken into account, responses were seen in two-thirds of patients already treated with 3 chemotherapy regimens. The response rates were 18% in the group of patients who had received high-dose chemotherapy (2 PR) and 26% in the group who had not (2 CR and 3 PR), whereas no response was recorded in the 2 patients previously treated with topotecan.

When any type of tumor size reduction (complete, partial, minor, and temporary) was considered, a 43% response rate was calculated.

After further treatment, including surgery, chemotherapy, and RT, 4 patients are currently alive in second CR and one in third CR 4.9–24.2 months since starting CPT-11. Time to progression ranged from 0.5 to 22 months (median, 2.8 mos).


The 32 patients received a total of 79 cycles (median, 2; range, 1–6) of irinotecan, which was well tolerated in almost all patients. Hematologic toxicity was minimal, with Grade 3 anemia and Grade 4 thrombocytopenia occurring after 5 cycles in 3 patients (including the one who had had an allogeneic transplant). Grade 3–4 neutropenia was evident after 8 cycles. One case of gastrointestinal candidiasis and 5 episodes of fever and neutropenia were reported.

Diarrhea was reported very frequently, occurring in 58% of cycles with 9 episodes graded as Grade 3 or 4, typically beginning in the second to third week of treatment and requiring hospitalization. No other major toxicities were reported.

Because of its gastrointestinal toxicity, the dose of CPT-11 was reduced by 25% in the second cycle in 3 patients, and 1 had the second cycle delayed because of an infection. One adolescent patient refused further CPT-11 treatment because of severe diarrhea, despite evidence of tumor reduction after the first 2 cycles.


Experiments in human tumor xenografts have concluded that CPT-11 is a promising drug and have shown a schedule-dependent activity, with protracted schedules producing better results.4

Initial experiences with different schedules indicated some response in children with solid tumors.4, 5, 6, 7 The effect on childhood STS has not been thoroughly evaluated to date.

On the basis of preclinical data and Phase I trial results, we performed a Phase II trial in pediatric patients with STS. Two consecutive 5-day courses were administered, repeated every 28 days.

We decided to enroll patients with different histologies, such as RMS, PNET, and DSRCT, for two main reasons, because 1) RMS and extraosseous PNET are currently treated according to the same protocol in Italy; and 2) they show similar rates of response to the same drugs.

In a group of heavily pretreated patients, we observed an overall response rate of 23%, plus some minor and a few temporary responses. We chose to adopt a 4-week interval between cycles because of these patients' previous treatments (most of them had already had high-dose chemotherapy with stem cell infusion). A short-lived response was seen in 2 patients. These findings lead us to suggest that a 28-day interval may be too long in only relatively chemosensitive tumors, and a 21-day interval (as used in other experiences) might have produced better results.

Irinotecan proved active against previously treated RMS. Our response rate (2 PR, 16%) is lower than we expected, but results obtained in other limited series5 and preclinical studies using a similar prolonged schedule4 seemed more promising. Cosetti et al. reported 6 responses in 7 patients with RMS,5 and 3 responses were reported in 4 RMS patients enrolled in another Phase I study.4 Prior treatment and the longer interval we adopted between cycles may explain these differences.

In our experience, the most remarkable result was obtained in the group of PNET patients, where a 38% response rate was observed. Topotecan, administered as a 72-hour continuous infusion, is reportedly active against the Ewing family of tumors,8 but, as far as we know, no positive data have been published concerning irinotecan apart from the study published by Wagner et al., where 3 patients showed a tumor reduction after the administration of CPT-11 and temozolamide.9 The data emerging from our study seem promising and are in line with results obtained in xenografts.3

Successful responses to irinotecan treatment have been reported in DSRCT patients,10 but were not confirmed by results obtained in our small series of these aggressive tumors.

Finally, our experience confirms that a protracted CPT schedule is well tolerated in children. Despite the occurrence of diarrhea in 58% patients, only 11% had symptoms severe enough (Grade 3–4) to necessitate a suspension of treatment in 1 patient and a dose reduction in 2. Myelosuppression was a minor problem in our series.

In conclusion, used as single agent in the treatment of recurrent or refractory STS, irinotecan, administered on a protracted schedule, showed a noteworthy response rate in a population of heavily pretreated patients, especially in the subset of patients with PNET. Its toxicity profile warrants further investigations in association with myelotoxic agents.