A Phase I Study of Irinotecan in Pediatric Patients: A Pediatric Oncology Group Study¹

Irinotecan (CPT-11, or 7-ethyl-10-(4-[1-piperidino]-1piperidino)-carbonyloxy-camptothecin)is a semisynthetic, water soluble analogue of camptothecin that differs from other camptothecin analogues, such as topotecan and 9-aminocamptothecin, in that it is a prodrug that undergoes de-esterification to a much more potent topoisomerase-I inhibitor,SN-38. Like other camptothecin analogues, both irinotecan and its active metabolite, SN-38, undergo pH-dependent reversible hydrolysis from an active lactone species to a relatively inactive hydroxy acid(carboxylate) form. Irinotecan itself possesses a marginal antiproliferative effect, whereas its metabolite SN-38 is 100-fold more active in vitro (1).

In preclinical studies, irinotecan was highly active in vitro and in vivo against a broad spectrum of human and murine tumor cell lines. Substantial antitumor activity was observed in xenografts derived from pediatric tumors such as neuroblastoma,rhabdomyosarcoma, peripheral primitive neuroectodermal tumors, and CNS³tumors, as well as in rhabdomyosarcoma xenografts selected in vivo for resistance to vincristine, melphalan, and topotecan (2, 3, 4, 5, 6, 7). In addition, marked activity was observed against a variety of xenografts derived from adult tumors, including colon adenocarcinoma Co-4; mammary carcinoma MX-1; gastric adenocarcinomas ST-15 and SC-6; squamous cell carcinoma QG-56; and lung tumor xenografts, Mqnul, Msnul, and LX1 (8, 9).

In recently completed adult Phase II trials, the clinical antitumor activity of irinotecan has been confirmed in a wide variety of tumor types. There was a 27% objective response rate (PRs) in patients with metastatic colorectal cancer (10). The response rate for irinotecan as a single agent in previously untreated non-small cell lung cancer was 32%, and in combination with cisplatin the response rate was 54% (11, 12). In patients with refractory or relapsed small cell lung cancer, the response rate was 47%(13). Responses have also been observed in adults with refractory leukemias and lymphomas (17–34%; Refs. 14, 15) or refractory gliomas (16). Objective responses have also been observed in a variety of other malignancies in patients treated on Phase I/II trials (10, 17, 18, 19, 20). The high level of antitumor activity resulted in recent Food and Drug Administration approval of irinotecan for commercial use in patients with refractory colorectal cancer.

In this report we present the results of a Phase I trial and pharmacokinetic study of irinotecan given daily for 5 days every 3 weeks in pediatric patients with refractory cancer. The objectives of this study were to identify the optimal irinotecan dose for phase II pediatric trials, to determine the incidence and severity of toxicities associated with irinotecan administration, and to determine the pharmacokinetics of irinotecan and its metabolites in children. Results of the pharmacokinetic study will be reported separately.

The eligibility criteria initially limited enrollment to children ≥6 year and ≤21 years of age with a histologically confirmed solid tumor refractory to standard therapy. After preliminary experience in older children treated at the first two dose levels demonstrated that irinotecan-induced diarrhea was readily managed with supportive care,the patient eligibility criteria were expanded to include patients ≥1 and <6 years of age. Patients in this younger age group were enrolled in a separate stratum (stratum 3) at one dose level below the dose level for older children until the DLT and MTD were defined. Other eligibility criteria included the following: (a) an ECOG performance status ≤2; (b) a life expectancy>8 weeks; (c) adequate bone marrow function (an absolute neutrophil count >1500/mm³, a hemoglobin >9.0 g/dl, and a platelet count >100,000/mm³);(d) adequate liver function (serum bilirubin <1.5 mg/100 ml; alanine aminotransferase <2× normal); (e)adequate renal function (serum creatinine <1.5 mg/100 ml or creatinine clearance >60 ml/min/1.73m²); (f)recovery from the toxicity of prior therapy; (g) no other chemotherapy within 2 weeks (6 weeks for prior nitrosourea therapy) of entering onto this protocol; and (h) no prior extensive radiotherapy (e.g., pelvic or craniospinal) or bone marrow transplantation with total body irradiation. Initially, there was no limitation on the number of chemotherapy regimens that the patient could have received before entry onto this trial.

After determination of the MTD in this heavily pretreated patient population (stratum 1), the eligibility criteria were revised to study a less-heavily pretreated patient population (stratum 2). The revised eligibility criteria excluded patients who had received >2 prior chemotherapy treatment regimens and patients who had received any prior central axis radiation (skull, spine, pelvis, or ribs) or a bone marrow transplant. The definition of the less-heavily pretreated group was designed to closely mirror pediatric patient populations in classic Phase II trials.

Informed consent was obtained from the patient or his/her legal guardian before entry on this study in accordance with federal and individual institutional policies.

Irinotecan, supplied by the Division of Cancer Treatment, National Cancer Institute, Bethesda, MD, was administered as a 60-min i.v. infusion immediately after premedication with dexamethasone, 4.0 mg/m², and ondansetron, 0.15 mg/kg daily for 5 days. The appropriate dose of drug was diluted with normal saline to a final total volume of 30 ml and administered over a period of 60 min.

The starting dose of irinotecan was 30 mg/m². Subsequent escalations were to 39 mg/m² and 50 mg/m². Because the DLT at the 50 mg/m² dose level in the heavily pretreated patients was myelosuppression, additional dose escalation was attempted in the less-heavily pretreated patient population starting at the 50 mg/m² dose level and escalating to 65 mg/m².

A minimum of three patients evaluable for both hematological and nonhematological toxicity were treated at each dose level. If one of the first three patients entered at any dose level experienced a dose-limiting toxicity during the first course of therapy, up to three additional patients were entered at that dose level. Toxicities were graded according to the National Cancer Institute Common Toxicity Criteria (Version 1; Ref. 21). Dose-limiting nonhematological toxicity was defined as any grade-3 or grade-4 nonhematological toxicity, with the specific exclusion of grade-3 nausea and vomiting, grade-3 fever, and grade-3 hepatic toxicity that returned to grade 1 before the scheduled time for the next treatment course. Dose-limiting hematological toxicity was defined as grade-4 neutropenia (<500/mm³), anemia (Hg <6.5 g/dl), or thrombocytopenia (<25,000/mm³) of >7 days duration. Each course was evaluated for both hematological and nonhematological toxicity.

Patient histories, physical examinations, and laboratory studies were obtained before treatment and then weekly throughout the course of the study. Laboratory evaluation included electrolytes, blood urea nitrogen, creatinine, and liver function tests. Complete blood counts were obtained at least twice weekly throughout the course of the study. Patients with measurable disease had appropriate radiographic or bone marrow evaluations at baseline, after the second cycle of irinotecan, and then every other cycle of irinotecan, to assess tumor response.

The MTD of irinotecan was defined as the dose level immediately below the level at which two or more patients of a cohort of up to six patients experienced dose-limiting toxicity. Courses were repeated every 21 days in the absence of DLT. In the absence of PD, patients with reversible DLT could receive additional cycles of irinotecan at one dose level below the dose that resulted in DLT in that patient.

Patients with measurable disease at the time of study enrollment were considered evaluable for response. A complete response was defined as the complete resolution of all measurable tumors and no progression of bony disease for a duration of ≥3 weeks. PR was defined as a ≥50%reduction in the sum of the products of the two longest perpendicular diameters of all measurable tumors for a duration of ≥3 weeks, and a minimal response was a ≥25% but <50% reduction in the sum of the products of the two longest perpendicular diameters of all tumors. SD or NR were defined as a <25% decrease in the sum of the products of the maximum perpendicular diameters of all measurable lesions, no evidence of progression of any lesion, and no evidence of new lesions. PD was defined as the appearance of new tumors or a ≥25% increase in the product of the two longest perpendicular diameters in any measurable lesion (excluding bone). Patients with PD after one or more courses of irinotecan were removed from study.

A total of 35 patients were entered on the study, of which one was ineligible (because of late institutional registration). Four additional patients were not evaluable for toxicity or response: two patients did not receive irinotecan, one patient did not complete the first cycle of irinotecan secondary to bacterial sepsis, and one patient died of PD 7 days after receiving the drug. Of the remaining 30 patients, 27 were evaluable for nonhematological toxicity, 24 were evaluable for hematological toxicity, and 29 were evaluable for response. Patient characteristics are listed in Tables 1 and 2.

In the heavily pretreated patient group (stratum 1), the median number of prior chemotherapy regimens was three and the range was one to eight. In the less-heavily pretreated group, one patient had not received any prior chemotherapy, four had received one prior chemotherapy regimen, and five had received two prior chemotherapy regimens.

Irinotecan was well tolerated by these pediatric patients with refractory cancer. Dose-related myelosuppression was the primary dose-limiting toxicity of irinotecan in the heavily pretreated patient population treated at the 50 mg/m²/day dose level. At this dose level, one patient with neuroblastoma experienced dose-limiting grade 4 neutropenia (absolute neutrophil count<500/μl for >7 days) and a second with pinealoblastoma and a prior history of craniospinal irradiation had prolonged neutropenia (grade 3–4 for ∼16 days) accompanied by thrombocytopenia (grade 2 for almost 1 month). As a result, this patient had a long treatment delay between cycles 1 and 2 of therapy, and thus this toxicity was considered dose-limiting. The MTD for heavily pretreated patients was 39 mg/m²/dose. Subsequent protocol accrual was limited to less-heavily pretreated patients. Dose-limiting diarrhea occurred in two of three patients treated at the 65 mg/m² dose level. The maximum tolerated dose for less-heavily treated children was therefore 50 mg/m²/dose.

Although neutropenia was dose-limiting in two heavily pretreated patients, myelosuppression was not a significant toxicity for the majority of patients enrolled in this study. Only 20% of patients experienced grade 4 neutropenia and <15% had grade 4 thrombocytopenia. Even grade 3 myelotoxicity was infrequent with fewer than 15% of patients experiencing grade 3 neutropenia and fewer than 10% either grade 3 thrombocytopenia or anemia.

Three patients experienced diaphoresis and flushing which in some was also accompanied by diarrhea (“early diarrhea”) during or immediately after completion of the irinotecan infusion. When this symptom complex was observed, it resolved spontaneously or after the administration of a single dose of atropine. Early diarrhea did not always occur with subsequent doses of irinotecan. Late diarrhea,primarily grade 1 or 2, occurred in approximately two-thirds of patients. Only one patient with diarrhea required hospitalization for fluid administration. The diarrhea in this particular instance was probably not drug-related because the patient’s entire family had gastroenteritis.

Other toxicities associated with irinotecan administration were primarily grade 1 or 2 (Table 3). In addition to these toxicities, one patient with a brainstem glioma developed severe erythema multiforme. This patient was hospitalized approximately 2 weeks after irinotecan administration with a history of diarrhea, lethargy, and respiratory distress. The next day, she had fever, mucosal bleeding, and skin lesions consistent with varicella and erythema multiforme. She also developed grade 4 neutropenia and thrombocytopenia during this time. It is not known whether the erythema multiforme was irinotecan-related or secondary to the patient’s underlying infection. The direct fluorescence antigen for varicella was negative. The patient died of PD ∼2 weeks after the onset of this event.

Although this study initially excluded patients <6 years of age, seven patients who were <6 years were subsequently enrolled on study (four at 39 mg/m² and three at 50 mg/m²). One of the six did not complete the first cycle of drug because of the onset of bacterial sepsis (not drug related), and the second is the patient with erythema multiforme described above. The five other patients who were enrolled in stratum 3 had a toxicity profile similar to that in children ≥6 years of age. Dose-limiting diarrhea did not occur in any of the patients <6 years of age.

The responses to irinotecan are summarized in Table 4. A PR was documented in one patient with recurrent hepatic neuroblastoma who was treated at the 30 mg/m²dose level. Almost complete radiographic resolution of the hepatic disease occurred gradually over many cycles of therapy. The patient developed leptomeningeal dissemination of his tumor after his fifteenth cycle of irinotecan. A PR was also observed in a patient with hepatocellular carcinoma who was treated at the 65 mg/m² dose level. This patient also had a gradual reduction in his tumor mass that was documented on serial imaging studies and additionally by a reduction in the serum α-fetoprotein level from 253.5 μg/liter to 26.9 μg/liter. In addition to the objective responses observed in these two patients, ∼25% of the patients who were evaluable for response had documented stable disease(Table 4). The patients with prolonged stable disease included one with hepatoblastoma who had a marked reduction in his serum α-fetoprotein level from an initial 122,030 μg/liter to 534 μg/liter during his thirteenth cycle of irinotecan. At that time the patient was removed from the study by his treating physician to receive a liver transplant.

Topoisomerase I inhibitors are a novel class of antitumor agents that have marked preclinical and clinical activity in a wide spectrum of adult and pediatric malignancies. Topotecan was the first topoisomerase I inhibitor to be evaluated in children with refractory cancer. The contribution of topoisomerase I inhibitors in the front-line management of pediatric cancer is not yet known. However the promising antitumor activity of topoisomerase I inhibitors that has been observed in Phase II studies in recurrent disease (22) and in therapeutic windows in tumors such as neuroblastoma and alveolar rhabdomyosarcoma (23) has resulted in the development of additional topoisomerase I inhibitors such as irinotecan. Preclinical studies in vitro and in xenografts derived from childhood neuroblastomas have demonstrated that the active metabolite of irinotecan, SN-38, is much more potent than topotecan (24).

Although the cytotoxic activity of topoisomerase I inhibitors is schedule-dependent, the optimal dosing schedule for topoisomerase I inhibitors in humans is not known. Houghton et al.(24) have demonstrated in a variety of xenograft models that prolonged exposure to topoisomerase I inhibitors such as topotecan or irinotecan is associated with increased efficacy. On the basis of these preclinical studies, they advocate a protracted schedule of drug administration, e.g., daily for 5 days for 2 consecutive weeks with cycles repeated every 3 weeks. In the clinical setting,objective antitumor activity has been observed in a wide spectrum of malignancies despite a multitude of topoisomerase I treatment schedules. Randomized studies in humans to evaluate whether a protracted administration schedule affects efficacy have not been performed. However, a comparison of the clinical pharmacodynamics of four different schedules of oral topotecan (daily ×5, every 21 days;(daily ×5 × 2, every 21 days; b.i.d. ×10, every 21 days; and b.i.d. ×21, every 28 days) has been published (25). In this report, the total drug exposure [area under the concentration versus time cuve (AUC)] per course was relatively consistent regardless of the schedule of drug administration. In contrast, toxicity seemed to be related to schedule rather than to AUC, with myelosuppression dose-limiting on shorter schedules and intestinal side effects dose-limiting on protracted schedules (25). There was no apparent difference in antitumor activity among these four schedules; and based on patient convenience, the daily ×5 schedule was favored. For similar reasons, this pediatric study evaluated a daily ×5, every-21-days schedule.

In this trial, we evaluated the clinical toxicities of the topoisomerase I inhibitor, irinotecan, administered as a 60-min i.v. infusion following premedication with dexamethasone and ondansetron. Myelosuppression was dose-limiting in heavily pretreated patients at 50 mg/m²/dose, and diarrhea was dose-limiting in less-heavily pretreated children at 65 mg/m²/dose. Thus, the recommended Phase II doses of irinotecan are 39 mg/m²/dose for heavily pretreated patients and 50 mg/m²/dose for less-heavily pretreated patients.

The toxicity profile of irinotecan in children treated in this study was similar to that reported in adults and to children treated with a more protracted dosing schedule, (daily [times 5 × 2,every-21-days schedule (26) with diarrhea and myelosuppression, primarily neutropenia, being the predominant toxicities.

The antitumor activity observed in this study is very encouraging. Nine of 29 (30%) of the evaluable patients had disease stabilization or response, and of these 9 patients, 7 received ≥10 cycles of drug.

In summary, irinotecan is well-tolerated in children with refractory cancer. As with docetaxel, the recommended dosage for and toxicity profile of irinotecan in heavily versus less-heavily treated children differs (27). Such studies underscore the continued need to evaluate less-heavily pretreated pediatric patients in a separate stratum in an attempt to recommend the optimal dose for Phase II studies. The objective responses and prolonged stable disease observed in this study occurred in a wide spectrum of pediatric tumors. Phase II studies of irinotecan in children with refractory solid tumors and refractory CNS tumors at a dose of 50 mg/m²administered daily ×5 every 21 days are in progress.