A Phase I/II Study of a 72-h Continuous Infusion of Etoposide in Advanced Soft Tissue Sarcoma

Purpose. The study was performed to assess the antitumour activity and toxicity of a 72-h continuous infusion of single-agent etoposide as second-line treatment for patients with locally advanced or metastatic soft tissue sarcoma (STS), following reports of substantial activity using this schedule of etoposide administration as first-line treatment in combination with ifosfamide. Patients/method. This was an open phase I/II trial performed at a single institution in patients with metastatic or locally advanced STS who had failed first-line treatment with doxorubicin + ifosfamide combination chemotherapy or, less commonly, single-agent treatment with doxorubicin or ifosfamide. Etoposide was given as a continuous intravenous infusion over 72 h. The starting dose level was 200 mg m-2 day-1 × 3 escalating in 10% steps in cohorts of three patients until dose-limiting toxicity was encountered. Results. Seventeen patients were treated, median age 47 years (range 26–71 years). No responses were seen in 16 assessable patients despite etoposide levels in the cotoxic range. The steady-state plasma concentration exceeded 8 μg ml−1 in all patients and in patients treated at ≥ 600 mg m −2 the mean steady-state level was 14.4 μg ml −1. The median event-free survival was 6 weeks (95% confidence interval (CI) 3.31–8.69) and the overall survival 16 weeks (95% CI 9.28–22.72). The maximum tolerated dose in this pretreated patient group was 200 mg mm-2 day-1 × 3. The dose-limiting toxicity was myelosuppression. Discussion. Etoposide given by 72-h infusion is inactive as second-line chemotherapy in STS. It is associated with significant toxicity when given in these doses, in this patient group.


Introduction
Etoposide (VP-16-213) is a semi-synthetic derivative of podophyllotoxin. It first underwent clinical trials 20 years ago. It has significant activity in lymphoma, leukaemia and small cell lung cancer and is included in most standard regimens for the treatment of germ cell tumours. It acts primarily by inhibition of topoisomerase II resulting in doubleand single-strand DNA breaks.
The initial trials of etoposide as a single agent in the treatment of soft tissue sarcoma (STS), almost all carried out in pretreated patients, have been disappointing. Welt et al. reported a dose escalation study in pretreated patients with a 3-day alternating schedule. 2 The starting dose was 120 mg m-2 given intravenously every other day for three doses, repeated every 3 weeks, increasing to a maximum of 240 mg mper dose. While toxicity was low, there was only one minor response, which included complete resolution of lung metastases, and 4/26 patients had stable disease. Similarly, a phase II EORTC study using a treatment schedule of 130 mg mpo daily for 5 days every 3 weeks, in mostly heavily pretreated patients, reported only one partial response (PR), which lasted 19 months, in 29 patients. 3 However, responses have been reported using more prolonged exposure. Hainsworth et al. 4  50 mg m-2 21 days every 28 days, failed to show significant activity. [6][7][8] Etoposide has a schedule-dependent mechanism of action, as elegantly demonstrated by Slevin et al. in small cell lung cancer. 9 This study compared the activity of 5 daily 2-h infusions with the same total dose given as a continuous intravenous infusion over 24 h. The 5-day course was significantly superior with a response rate of 89% vs 10% despite the area under the concentration time curve (AUC) being identical for both schedules. However, when treatment duration was extended to 8 days, no additional benefit was seen compared with 5 days, 1 and a 15-day infusion study had to be stopped early because of a worse response in the 15-day arm. 11 Thompson et al. performed a study using a protracted intravenous infusion of etoposide. This study, carried out in patients with potentially etoposide-sensitive malignancies, was performed at doses of 18-25 mg mday -1 for 21-153 days. 12 The median duration of treatment was 17 weeks (range 3-80 weeks) and the overall response rate was 47%. The mean serum etoposide concentration was 0.7 _ _ 0.42 #g m-1 and antitumour activity was observed with levels from 0.5 to 1.0 #g m-1. This was lower than the effective level suggested by Clark et al. 1 in relation to small cell lung cancer, who reported that serum etoposide levels > 1 #g mwere associated with antitumour activity and levels > 3 #g m-1 with haematological toxicity. This is consistent with the model of a therapeutic window for etoposide with both the cytotoxicity and haematological toxicity thresholds falling as the duration of exposure increases. 13 The Scandinavian Sarcoma Group reported a study of first-line chemotherapy in STS using etoposide by 72-h continuous infusion (200 mg mday-1) plus ifosfamide 1.5 g m-2 day-by 2-h infusion daily 3.14 They reported a PR rate of 40% in 33 patients with a median time to progression of 8 months. Confidence intervals (CIs) were not given but were presumably quite wide. Nevertheless, this level of activity suggests that etoposide may be active when given by 72-h infusion, given that response rates for single-agent ifosfamide at a standard dose of 5 g m-2 have been in the region of 250/0.15 Evaluation of this schedule of etoposide as a single agent as second-line treatment in STS seemed warranted.

Eligibility criteria
The following entry criteria were required: (1) histological evidence of STS; the following tumour types were excluded: Ewing's sarcoma, embryonal or alveolar rhabdomyosarcoma, osteosarcoma and malig-nant mesothelioma; (2) at least one bidimensionally measurable lesion with evidence of progression within 6 weeks prior to treatment, (3)  Standards were prepared freshly for each HPLC run. Samples were extracted using dichloromethane, the organic phase was evaporated to dryness and reconstituted in methanol-water (51 "49) and centrifuged at 300 g before loading on to the autosampler.

Results
A total of 17 patients were treated with a total of 44 cycles of chemotherapy (median 2, range 1-5).
Patient characteristics are detailed in Table 1. All patients had received one line of prior chemotherapy, most commonly a combination of ifosfamide and doxorubicin as part of an EORTC study in which ifosfamide was given at 5 g m -z, and doxorubicin at either 50 or 75 mg m -2, the latter dose supported by granulocyte-macrophage colonystimulating factor (BM-CSF). Single-agent doxorubicin was given to three patients at 75 mg m-2 and a combination of doxorubicin with ifosfamide at 5-9 g m-2 to two patients. Three patients were initially treated elsewhere, two received doxorubicin and one epirubicin. Two patients had received two prior chemotherapy regimens. All 17 patients were assessable for toxicity, 16 were assessable for response and one patient died due to neutropeniarelated infection prior to any response assessment.
One patient withdrew from the study after two courses with stable disease and subsequently received four more courses of etoposide off study. He currently remains alive with stable disease. The initial dose level was 200 mg m-2 day-1 for 3 days; four patients were treated at this dose before escalat-Etoposide in advanced soft tissue sarcoma 151 ing to a dose level of 220 mg m-2 day-1 for 3 days.
Five patients were treated at this dose but toxicity proved unacceptable and the remaining eight patients were treated at the lower dose level. Despite this, five patients required dose reductions (four at 200 mg mday-and one at 220 mg mday-1).

Response
No responses were seen. Eight patients had stable disease and eight patients progressed through chemotherapy. The median EFS was only 6 weeks (95% CI 3.31-8.69) and the median OS was 16 weeks (95% CI 9.28-22.72) (Fig. 1). In the eight patients with stable disease the median progressionfree survival was 3.5 months. Of the other two surviving patients, one subsequently received abdominal radiotherapy resulting in a PR. She has subsequently progressed with hepatic metastases, while the other has had no further therapy.

Plasma etoposide concentrations
Data were available on 12 patients and 70 courses.
Where steady-state levels at two time points during   was not used. Three patients died on treatment, one from Escherichia coli sepsis when neutropenic, one at home from an acute abdominal catastrophe during her third course, her disease having been stable after the second course. She had a normal blood count at the time of the last chemotherapy 6 days before her death. The third patient died due to progressive pulmonary disease but was neutropenic at the time of his death. Early progression was seen in two additional patients who both received only one course of treatment. Seven patients were admitted for neutropenic sepsis. Other admissions (total three) were for problems related to disease progression. Non-haematological toxicity was less of a problem. Two patients (11%) experienced grade III emesis and one patient developed pulmonary oedema due to the fluid load. Alopecia was almost universal but mostly predated etoposide The remaining toxicities were relatively mild and included mucositis in seven patients (grade 1-2), diarrhoea in three patients (grade 2), fatigue and asymptomatic elevation of liver enzymes.

Discussion
In this study, 17