A phase II study of topotecan with vincristine and doxorubicin in children with recurrent/refractory neuroblastoma

Abstract

BACKGROUND

A Phase II trial in children with advanced neuroblastoma was carried out in five Italian institutions to evaluate the antitumor activity and tolerability of topotecan followed by vincristine and doxorubicin.

METHODS

Children older than age 1 year with Stage III or Stage IV neuroblastoma, all of whom had been treated previously with chemotherapy and were diagnosed with either refractory or recurrent disease, were treated with topotecan at an intravenous dose of 1.5 mg/m² (the dose was 0.75 mg/m² for patients who were treated within 1 year of previous megatherapy) per day for 5 days followed by 48-hour intravenous infusions of 2 mg/m² vincristine and 45 mg/m² doxorubicin. Cycles of therapy were repeated every 3 weeks.

RESULTS

Twenty-five patients (2 with Stage III disease and 23 with Stage IV disease; 19 with refractory disease and 6 with recurrent disease) were treated with a total of 115 cycles. Four patients had complete responses, 12 patients had partial responses, 4 patients had minor responses or stable disease, and 5 patients had tumor progression. The overall response rate (including complete and partial responses) was 64% (95% confidence interval, 43–82%). Fifteen patients were alive at the time of the current report and were progression free at 4–16 months (median, 9 months) after the first course of this treatment. Toxicity generally was limited to the hematopoietic system. Dose-limiting toxicity was observed in only 1 patient (Grade 4 liver toxicity). There were no deaths due to infectious or toxic causes.

CONCLUSIONS

The topotecan-vincristine-doxorubicin combination was active and well tolerated in previously treated patients with advanced neuroblastoma. Cancer 2003. © 2003 American Cancer Society.

Neuroblastoma, a neoplasm of the sympathetic nervous system, is the second most common malignant solid tumor in childhood. The prognosis for patients with this neoplasm has improved thanks to advances in medical care, but the outcome in children older than age 1 year with metastatic disease remains poor, and innovative therapies still are needed.1, 2

Camptothecin analogues have emerged as a very attractive new class of antitumor drugs acting as topoisomerase I inhibitors. They exhibit a novel mechanism of action, because none of the antitumor drugs currently used in the treatment of patients with pediatric malignancies are targeted against that enzyme.3 Topotecan (TOPO), a water-soluble, semisynthetic camptothecin derivative, has exhibited antitumor activity in both in vitro and in vivo tumor models.4–6 Phase I trials of TOPO in adults have identified neutropenia as the dose-limiting toxicity (DLT), establishing a recommended dose of 1.5 mg/m² as a 30-minute infusion daily for 5 days.7–9 At the half-dose of 0.75 mg/m² for 5 days combined with cyclophosphamide 250 mg/m², TOPO was tolerable even in patients who had previously received high-dose chemotherapy with autologous bone marrow rescue.10 The presence of broad antitumor activity also was suggested in those studies.7–10

A preclinical study explored the in vivo therapeutic effects of sequential combinations of topoisomerase I–targeting and topoisomerase II–targeting drugs in a murine tumor model system. After treatment with camptothecins, it was observed that levels of topoisomerase I mRNA and protein in the tumor decreased, whereas levels of topoisomerase IIα mRNA and protein rose. The conclusion was that topoisomerase I and II agents, if administered in this sequence, have synergic activity and show no increase in toxicity.11 In addition, initial treatment with TOPO was associated with increasing responsiveness of the xenografts to subsequent doses of doxorubicin (DOXO).12 Furthermore, in a study on pediatric solid tumor xenografts, the therapeutic effect of combining TOPO with vincristine (VCR) was greater than additive, and toxicity was moderate.13

Over the past few years, after favorable data were reported by Helson et al.,14 we used a 72-hour, continuous infusion of VCR plus DOXO as palliative therapy in children with Stage IV neuroblastoma who developed a recurrence after high-dose chemotherapy with hematopoietic stem cell rescue (HSCR). It was very well tolerated, with only mild hematologic toxicity, and it induced palliative effects by reducing symptoms of the neoplasia.

On the basis of the observations made in preclinical and clinical studies, we designed a Phase II trial of sequentially administered TOPO at a dose of 1.5 mg/m² for 5 days followed by a combined, prolonged infusion of VCR with DOXO at two-thirds of the dose that had been employed previously as palliative treatment. This regimen was to be followed for children with advanced neuroblastoma. The results of this study are reported herein.

MATERIALS AND METHODS

Patient Selection

Starting in December 2000, 5 Italian institutions began enrolling patients with Stage III or IV neuroblastoma who were older than age 12 months at diagnosis, who had achieved only a minor response after first-line or second-line chemotherapy (in patients with refractory disease), or who had responded to first-line treatment and then developed a recurrence after a treatment-free interval ≥ 3 months (in patients with recurrent disease). Other eligibility criteria included neutrophil count > 500/μL; platelet count > 50,000/μL; and serum transaminase, bilirubin, and creatinine levels less than twice the normal values. Exclusion criteria included any severe organ dysfunction; life expectancy < 4 weeks; known human immunodeficiency virus, hepatitis C virus, or hepatitis B virus infection; or treatment with other experimental drugs during the 30 days prior to the study and a previous, cumulative dose of doxorubicin > 300 mg/m². The study was approved by the ethical committees of the five institutions that enrolled patients (Giannina Gaslini Children's Hospital [Genoa, Italy]; Istituto Nazionale Tumori [Milan, Italy]; Chairs of the Pediatrics Universities of Bologna [Bologna, Italy] and Padua [Padua, Italy]; and the Policlinico Gemelli [Rome, Italy]). The decision to enroll a patient in the study was made by the referring physician after discussion with the study coordinator. All study entry and response data were reviewed centrally. Written informed consent, according to the Helsinki Declaration, was obtained from the patients' parents before enrollment into the trial.

Treatment Plan

TOPO, DOXO, and VCR were prepared from commercially available supplies. Patients were hydrated intravenously for 7 days at a rate of 2 L/m² every 24 hours. TOPO (Hycamtin; GlaxoSmithKline Beecham, Philadelphia, PA) was administered intravenously at a dose of 1.5 mg/m² over 30 minutes daily on Days 1–5. DOXO (Adriblastina; Pharmacia, Gaithersburg, MD) 45 mg/m² and VCR (Vincristina; Teva, Mijdrecht, The Netherlands) 2 mg/m² then were administered as a 48-hour, continuous, intravenous infusion on Days 5 and 6 starting 1 hour after the last dose of TOPO. The daily dose of TOPO was reduced by 50% (to 0.75 mg/m²) in children who had previously received megatherapy (MGT) with HSCR within 1 year before study entry.10 All patients received filgrastim 5 μg/kg subcutaneously daily beginning on Day 8 and continuing until the neutrophil count was ≥ 2500/μL after the time of the nadir. Toxicity was graded according to the National Cancer Institute Common Toxicity Criteria.15

The next cycle of TOPO-VCR-DOXO (TVD) chemotherapy started 3 weeks after beginning the previous cycle, provided that no progressive disease (PD) had developed or when the neutrophil and platelet counts were > 1000/μL, and > 100,000/μL, respectively, and the nonhematologic toxicity was Grade 1 or lower. When the cumulative dose of 480 mg/m² of DOXO was reached, it was no longer included in the remaining courses.

If DLT occurred after any course, then the following course was administered at 66% of the dose. DLT was defined as Grade 4 hematologic toxicity (a reduction in neutrophils < 500/μL, platelet count < 20,000/μL) associated with life-threatening infections or bleeding that required transfusions. Any nonhematologic, Grade 4 toxicity also was considered DLT.

Assessment

Disease status was assessed before study entry and then every two cycles afterward by physical examination, computed tomography or ultrasonography scans, and urine catecholamine measurement. Bilateral bone marrow aspirates and biopsies were required before study entry and were repeated every two cycles only in patients with evidence of bone marrow infiltration at study entry. Metaiodobenzylguanidine (MIBG) scintigraphy and/or bone scans were performed at study entry, after four cycles, and after eight cycles.

Tumor response was assessed according to the International Response Criteria for Neuroblastoma,16, every 2 cycles. Response was classified into one of four categories: complete response (CR), partial response (PR), stable disease (SD), and PD. Very good PRs and minor responses were included in PR and in SD, respectively. All data at study entry and response evaluations were centrally reviewed.

Required monitoring included weekly physical examinations, complete blood cell counts, and complete blood chemistry measurements. Toxicity was assessed and reported after each course. The time to progression was calculated from study entry to progression, death, or last follow-up, whichever occurred first.

Statistical Methods

The current study was designed as a sequential two-stage trial, protecting patients from treatment with toxic or ineffective agents.17 The response rate was reported after two courses of treatment and was defined as the proportion of patients whose best response was CR or PR. Because the prognosis for children with recurrent/refractory neuroblastoma is very poor, we assumed that a response rate ≤ 5% was of no interest (null hypothesis; H₀), whereas a response rate ≥ 25% was acceptable (alternative hypothesis; H₁). We designed the study so that if H₀ was true, then the probability of concluding that the treatment was promising was < 10% (α); whereas, if H₁ was true, then the probability of rejecting the treatment was < 10% (β). Similarly, we assumed that DLT that occurred in ≥ 90% of patients was unacceptable (H₀), whereas a toxicity rate ≤ 60% was acceptable (H₁), with the same values for α and β. According to this design, after entering the first 10 patients into the study, accrual was to be suspended until response was evaluated. If no patients responded (CR or PR) or if three or more patients experienced DLT, accrual would be terminated and the conclusion would be that treatment either was not active or was toxic, respectively. After this step, 14 additional patients would be entered into the study, for a total of 24 patients. At the end of the study, if CR or PR had been achieved by more than two patients, then the treatment would be declared active, and if DLT had occurred in no more than six patients, then the treatment would be declared nontoxic.

RESULTS

A total of 25 patients were enrolled into the study, and all were fully assessable for response and toxicity. All patients previously had received combined chemotherapy that included at least 5 drugs (range, 5–9 drugs; median, 7 drugs). All but 3 patients had previously received DOXO, and 11 patients also had received VCR. No patients previously had received camptothecin analogues. Five patients previously had been treated with metabolic radiotherapy with I¹²³-MIBG, and four patients had received external radiotherapy to the primary tumor. The characteristics at study entry of the 25 evaluable patients are listed in Table 1.

Table 1. Patient and Tumor Characteristics at Study Entry

Characteristic	No. of patients
Chemo: chemotherapy; MGT + HSCR: megatherapy plus hematopoietic stem cell rescue.
Patients
Enrolled	25
Evaluable	25
Age (yrs)
Median	6.1
Range	1.9–12.6
Gender
Male	22
Female	3
Tumor stage
T3	2
T4	23
Tumor status
Refractory	19
Recurrent	6
Previous treatment
Chemo	11
Chemo + MGT + HSCR within 12 mos	5
Chemo + MGT + HSCR after 12 mos	9
Time from diagnosis (mos)
Median	14
Range	3–65

Nineteen patients were enrolled in the study after their disease was deemed refractory to first-line treatment (n = 7) or second-line treatment (n = 12) (Table 2), and 6 patients were treated for recurrences that occurred after elective discontinuation of therapy with a median interval between off therapy and recurrence of 11 months (range, 7–42 months) (Table 2). In particular, 5 of those patients developed recurrences within 1 year of receiving MGT.

Table 2. Characteristics of Patients at Diagnosis and at the Time of Topotecan, Vincristine, and Doxorubicin Chemotherapy, Along with Data on Response to Therapy

Patient	Stage (INSS)	MYCN amplification	Previous megatherapy	Time from diagnosis to TVD therapy (mos)	Tumor lesions at TVD therapy				No. of courses of TVD	Maximal response/timea to progression (mos)
					Primary	Metastases	I^234 positivity	Urinary
TVD: topotecan, vincristine, and doxorubicin chemotherapy; INSS: International Neuroblastoma Staging System; BM: bone marrow; B: bone; PD: progressive disease; PR: partial response; ND: not determined; PT: primary tumor; LN: lymph node; CR: complete response; L: lungs; SD: stable disease. a Time to progression is indicated as of entry into the study.
Patients with refractory disease at study entry
1	3	Yes	No	8	Yes	BM, B	BM, B	Elevated	2	PD/2
3	4	Yes	Yes	21	No	BM	BM	Normal	5	PR/9
6	4	ND	No	3	Yes	B, BM	PT, B	Normal	2	PR/12+
7	4	ND	Yes	48	Yes	B, BM	PT, B	Elevated	1	PD/1
8	4	ND	No	4	Yes	B, BM	B	Normal	2	PR/11+
9	4	No	No	4	Yes	—	PT, BM	Elevated	9	PR/10+
10	4	No	Yes	32	No	BM, B, LN	BM, LN	Elevated	7	PR/10+
11	4	ND	Yes	10	Yes	BM, B	PT, B	Elevated	4	PR/12+
13	4	ND	Yes	42	No	BM, B	B	Normal	9	CR/9
14	4	Yes	No	5	Yes	L	PT, L	Elevated	2	PD/2
16	4	No	No	6	Yes	BM	B, PT	Elevated	4	PR/6+
17	4	No	No	3	Yes	BM	BM	Elevated	5	PR/5+
18	4	No	Yes	11	No	BM	B, BM	Normal	1	PD/2
20	4	ND	No	3	Yes	BM	PT, BM	ND	4	PR/5+
21	4	Yes	Yes	10	Yes	—	PT	Normal	2	SD/4+
22	3	No	Yes	17	Yes	BM	ND	Normal	5	SD/7
23	4	ND	No	5	Yes	BM	ND	Elevated	4	PR/8+
24	4	No	No	4	Yes	B, BM	PT, B, BM	Elevated	4	SD/8+
25	4	No	Yes	17	No	B, BM	B, BM	Normal	4	PR/8+
Patients with recurrent disease at study entry
2	4	No	Yes	19	Yes	BM, B	BM	Elevated	8	CR/14
4	4	ND	Yes	32	Yes	BM, LN, B	BM	Elevated	6	PR/13+
5	4	ND	Yes	21	Yes	BM, B	PT, B	ND	2	SD/7
12	4	ND	Yes	65	Yes	BM	NE	Elevated	8	CR/9+
15	4	No	Yes	37	No	BM	NE	Elevated	7	CR/7+
19	4	Yes	Yes	22	No	B, BM	B, BM	Normal	2	PD/3

The number of TVD treatment cycles, the best response achieved, and the time to progression for each patient are described in Table 2. A total of 115 treatment courses (median, 4 courses; range, 1–9 courses) were administered, and none of the patients received the total DOXO dose of 480 mg/m².

Toxicity

Of 115 treatment courses, 106 courses were fully evaluable for toxicity (Table 3). Only 1 course was followed by an episode of DLT, corresponding to a toxicity rate of 4% (95% confidence interval [95% CI], 0–20%). The episode was characterized by both Grade 4 liver toxicity during aplasia and fever when the patient suffered a concomitant Epstein–Barr virus infection. The DLT episode occurred after the first course of TVD, and the patient received 4 further courses of TVD at 66% of the dosage with no severe toxicity.

Table 3. Treatment Toxicity Data (106 Evaluable Cycles)

Toxicity	No. of patients (%)
Dose-limiting toxicity	1 (1)
Leukopenia
Grade 3	37 (35)
Grade 4	63 (59)
Neutropenia
Grade 3	8 (7.4)
Grade 4	95 (90)
Thrombocytopenia
Grade 3	11 (10)
Grade 4	91 (86)
Anemia
Grade 3	52 (49)
Grade 4	4 (4)
Hepatotoxicity
Grade 3	1 (1)
Grade 4	1 (1)
Emesis
Grade 3	1 (1)
Grade 4	0 (0)
Mucositis
Grade 3	2 (2)
Grade 4	0 (0)
Hematuria
Grade 3	1 (1)
Grade 4	0 (0)

Overall, the toxicity of TVD therapy was limited mostly to the hematopoietic system. All 25 patients experienced at least 1 episode of Grade 4 neutropenia and 1 episode of Grade 4 thrombocytopenia, whereas 24 patients experienced at least 1 episode of leukopenia. Grade 4 anemia was observed in 2 patients, and Grade 3 anemia was observed in 18 patients.

Grade 3 or 4 neutropenia, thrombocytopenia, and leukopenia occurred after 103 courses (97%), 102 courses (96%), and 100 courses (94%) of the 106 evaluable courses, respectively; whereas Grade 3 or 4 anemia was documented after 56 courses (53%) (Table 4). Transfusional support with packed red cells (PRCs) was required after 88 TVD courses (82%), whereas platelet transfusions were required after 99 of 106 courses (93%) that were evaluable for toxicity. The median number of transfusions per patient per course was 1, both for PRCs and platelets (range, 0–5 transfusions and 0–4 transfusions, respectively).

Table 4. Treatment Efficacy Data (25 Patients, 106 Cycles)

Variable	Response after two cycles	Best response
Evaluable patients	25	25
Complete response	—	4
Partial response	14	12
Stable disease	6	4
Progressive disease	5	5

Other nonhematologic Grade 3 toxicities included nausea and emesis (one patient), oral mucositis (two patients), increased transaminase levels (one patient), and hematuria (one patient). Despite the intensely myelosuppressive nature of this therapy, infectious complications were infrequent. In fact, only 32 episodes of fever were reported, and bacteremia was only documented in 3 patients.

Hospitalization for intravenous antibiotic therapy or other supportive therapy during the interval between cycles was necessary after 35 of 106 evaluable courses (33%) and was required most often after the first course of TVD (16 events after the first 25 courses). There was no difference in toxicity rates between patients who had previously undergone MGT and those who had not. Furthermore, toxicity was manageable in the 5 patients who received TOPO at a dose of 0.75 mg/m² per day because of more recent MGT. Finally, no deaths due to infectious or toxic causes were observed.

Antitumor Activity

All 25 patients were evaluable for response (Table 4). Two patients progressed after the first course of TVD. Of the remaining 23 children, after 2 courses of therapy, 14 patients achieved a PR, 6 patients had SD, and 3 patients showed PD, for a response rate of 56% (95% CI, 35–76%). Among patients who achieved PR, we also included four patients with bone marrow and bone disease who exhibited clearing of bone marrow disease and normalization of biochemical markers. After the first 2 cycles, 2 patients in PR and 2 patients with SD were withdrawn from the study on their parents' decision; therefore, 16 patients received 2 further courses of TVD, after which we observed 1 patient who achieved CR, 12 who achieved PR, 2 with persistent SD, and 1 with PD.

After the fourth cycle of TVD, 1 patient in PR and 1 with SD were withdrawn from the study on the basis of the investigator's decision; at the time of the current report, 3 patients already had completed evaluation after the fourth cycle and 10 patients received further courses of the TVD combination, resulting in 4 CRs, 5 PRs, and 1 patient with SD. Overall, as shown in Table 3, the best responses we observed were 4 CRs, 12 PRs, 4 patients with SD, and 5 patients with PD, which was equivalent to a response rate (CRs plus PRs) of 64% (95% CI, 43–82%).

Better responses were observed among the 6 children who were treated for recurrent disease (3 CRs and 1 PR) compared with the 1 CR and 11 PRs obtained among 19 patients who were treated for refractory disease. Of 5 patients who were treated with TVD within 1 year after MGT (daily dose of TOPO, 0.75 mg), 3 patients responded by achieving a PR, and 2 patients developed PD. Of the 9 patients who were treated with TVD > 1 year after MGT (thus, receiving the full dose of TOPO), 6 patients responded by achieving CR (4 patients) or PR (2 patients), 1 had SD, and 2 patients had PD.

The time to disease progression ranged from 1 to 3 months (median, 2 months) for the 5 patients whose best response was PD. Three responsive patients later developed PD, at 9 months, 9 months, and 14 months, respectively, and the 2 patients whose best response was SD both experienced disease progression after 7 months. At the end of the study, 15 patients were alive and progression free at 4–16 months (median, 9 months) after the first course of TVD.

DISCUSSION

The current Phase II trial demonstrated that the TVD combination (TOPO, VCR, and DOXO) was active and well tolerated in pediatric patients with neuroblastoma. All the preclinical data on camptothecin derivatives in pediatric tumor models, as well as the preliminary results of the single-agent TOPO clinical evaluation in children, suggest that topoisomerase I inhibitors represent an interesting new class of antitumor drugs for the treatment of childhood malignancies.4

Clinical studies using topoisomerase I inhibitors in combination with alkylating or platinating agents have demonstrated a greater degree of hematopoietic toxicity than what would be expected from either agent delivered alone.18–22 In contrast, the toxicity that results from combining TOPO with VCR and DOXO, which was supported by preclinical data, does not appear to be additive.11–13 However, the timing of the sequence seems to be a determinant, at least for toxicity.21, 22 In fact, high hematologic toxicity precluding TOPO dose escalation was observed when carboplatin20 or cyclophosphamide23 was administered before TOPO. In a previous Phase I study on adults who received DOXO before TOPO,24 the maximum tolerated dose was 25 mg/m² for DOXO and 5.25 mg/m² for TOPO, representing approximately two-thirds of the doses that we administered.

Neutropenia was the main toxicity that was encountered in the current study. We observed Grade 4 neutropenia after almost all courses of TVD, but it was associated with fever or hospitalization in only one-third of the courses, most often with the first course. Furthermore, the severity of TOPO-induced thrombocytopenia was greater in the first cycle, as has been observed previously in patients with ovarian carcinoma,25 and no cumulative effects were evident. Nonhematopoietic toxicity was rare; in fact, only 1 episode of severe emesis, which may have been attributable to DOXO, was reported, suggesting that the 5 days of TOPO could be administered safely on an outpatient basis. Furthermore, no episodes of severe constipation or neuropathy occurred, suggesting that there is no additive toxicity or pharmacokinetic interference with VCR. Finally, the single episode of Grade 4 liver toxicity probably was due to a concurrent Epstein–Barr virus infection.

In our experience, the quality of response was better in the group of patients who were treated at recurrence compared with patients who had refractory disease during first-line or second-line therapy or patients who developed recurrent disease shortly after MGT. However, some long-term responses were observed in both groups of patients. Most patients already had been treated with DOXO and VCR, and they still responded to TVD, suggesting that the clinical activity of DOXO and VCR may be increased by TOPO, as shown in vitro.

We believe that the 56% overall response rate we observed after 2 courses of TVD in this heavily pretreated group of children with neuroblastoma warrants attention. Previous studies in similar populations who received combinations of either platinum derivates and etoposide or cyclophosphamide and etoposide reported overall response rate ranging between 39% and 45%,26–29 not far from what has been reported in front-line studies.30, 31 Furthermore, the results we obtained using TOPO with VCR and DOXO also compared favorably with the results obtained using TOPO alone or combined with and cyclophosphamide. In fact, in a window study with TOPO alone conducted by the Pediatric Oncology Group (POG) on 32 children with untreated, Stage IV neuroblastoma, the highest response rate (CRs plus PRs) was 37% and reached 66% when minor responses were included.32 A greater degree of antitumor activity was observed when TOPO was combined with cyclophosphamide. This combination resulted in 6 major responses among 13 patients with recurrent/refractory disease in another POG study.19 Furthermore, in a similar study with higher doses of TOPO and cyclophosphamide, Kushner et al. reported 5 minor responses out of 11 evaluable neuroblastoma patients.18

In conclusion, in the current study, the TVD regimen with TOPO followed by prolonged infusion of VCR and DOXO demonstrated antitumor activity with acceptable toxicity in children with advanced, previously treated neuroblastoma. The clinical utility of sequencing TOPO with VCR and DOXO should be investigated in a first-line therapy setting as induction treatment or, considering the absence of DOXO in the regimen that is employed currently in the European Cooperative protocol,33 as intensification treatment before MGT with HSCR.

Acknowledgements

The authors thank Barbara Galleni and Paola Padovani for their assistance in the preparation of this article.