Phase II study of XR5000 (DACA) administered as a 120-h infusion in patients with recurrent glioblastoma multiforme

C. Twelves1,+, M. Campone2, B. Coudert3, M. Van den Bent4, M. de Jonge4, C. Dittrich5, R. Rampling1, R. Sorio6, D. Lacombe7, C. de Balincourt7 and P. Fumoleau2

1Cancer Research UK, Department of Medical Oncology and Beatson Oncology Centre, Glasgow, UK; 2Centre Rene Gauducheau, Nantes, France; 3Centre G.F. Leclerc, Dijon, France; 4Rotterdam Cancer Institute (Daniel den Hoed Kliniek) and University Hospital Rotterdam, The Netherlands; 5Kaiser Franz Josef Spital, Vienna, Austria; 6Centre di Referimento Oncologico, Aviano, Italy; 7European Organisation for Research and Treatment of Cancer Data Centre, Brussels, Belgium

Received 10 September 2001; revised 30 October 2001; accepted 16 November 2001 .


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background

XR5000 is a tricyclic carboxamide that intercalates DNA and inhibits both topoisomerase I and II. The aim of this study was to evaluate the efficacy and tolerability of XR5000 in patients with recurrent glioblastoma multiforme previously untreated with chemotherapy at relapse.

Patients and methods

Patients received XR5000 at a dose of 3010 mg/m2 as a 120-h central venous infusion every 3 weeks. An independent panel assessed response every two cycles using McDonald’s criteria (tumour size, steroid intake and neurological status); toxicity was graded according to the National Cancer Institute-Common Toxicity Criteria, version 2.0.

Results

Sixteen patients were enrolled (one ineligible patient was excluded from efficacy evaluation). Performance status was zero (five patients), one (nine patients) or two (one patient). They received 30 cycles of XR5000 (median 2, range 1–5). Haematological toxicity was mild, with only one patient experiencing grade 3 neutropenia. Other related grade 3/4 adverse events included chest pain (one patient), axillary vein thrombosis (one patient) and rigors/fever in the absence of neutropenia (one patient). There were no objective responses, 14 patients progressing on XR5000 and one having stable disease.

Conclusions

Although XR5000 was generally well tolerated, these results do not support further evaluation in patients with glioblastoma multiforme using this dose and schedule.

Key words: chemotherapy, glioblastoma multiforme, phase II


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
XR5000 {N-[2-(dimethylamino)ethyl]-acridine-4-carboxamide}, formerly known as DACA, is one of a series of tricyclic carboxamide cytotoxic drugs that bind to DNA by intercalation, and acts as an inhibitor of both topoisomerase I and II [1]. This dual inhibition and its activity in preclinical models, including those exhibiting drug resistance mediated by P-glycoprotein or the multi-drug resistance protein, make XR5000 an attractive agent for clinical development [2].

When administered as a 3-h infusion on a single day [3] or over 3 successive days [4] pain in the infusion arm was the main toxicity. One patient, who received XR5000 over 3 h through a central venous catheter, also experienced severe chest pain. Myelosuppression was minimal but dose-related somnolence or agitation, peri-oral parasthesiae and lacrimation were also reported. The maximum tolerated dose (MTD) of XR5000 given as a short infusion was 800 mg/m2, with the arm pain precluding further dose escalation. Accordingly, a subsequent study evaluated XR5000 given as 120-h infusion through a central venous catheter (Propper DJ, de Bono J, Saleem A et al, unpublished observations). This enabled dose escalation to 4060 mg/m2, which was associated with grade 4 chest and abdominal pain. The 3010 mg/m2 dose level was well tolerated although vomiting and somnolence were reported with administration of XR5000 over 120 h.

The rationale for studying XR5000 in patients with relapsed glioblastoma multiforme was three-fold. First, there remains an urgent need for effective new treatments in patients with recurrent glioblastoma multiforme for whom the prognosis is extremely poor. Secondly, XR5000 is lipophilic and crosses the blood–brain barrier in pre-clinical models [5]. Finally, some of the toxicities seen in the phase I trials, such as somnolence or agitation, peri-oral paraesthesiae and lacrimation, suggest that XR5000 may also penetrate the central nervous system in man.

The principle objective of this trial was to define the objective response rate of XR5000 in patients with recurrent glioblastoma multiforme who had not received chemotherapy for recurrent disease. Secondary objectives were to determine the duration of any responses and to characterise the toxicities of XR5000 in this setting. The evaluation of new agents in patients with glioblastoma multiforme is difficult and we modified our standard phase II solid tumour protocol to meet the specific demands for brain tumour studies.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
This was an open label, non-randomised phase II study using a Gehan two-step design. If at least one response was seen in the first 14 eligible patients, recruitment was to continue to a total of 25 patients. With this design, after evaluating 14 patients the chance of wrongly rejecting a drug with a true response rate of 20% is 0.044. The Ethics Committee of each hospital approved the study and all patients gave written informed consent.

Patients were eligible if they were at least 18 years of age with histologically or cytologically proven glioblastoma multiforme and recurrent disease which was measurable [defined as a contrast enhancing lesion at least 20 mm in diameter on magnetic resonance imaging (MRI)]. They were required to be on a stable or decreasing dose of corticosteroids for at least 2 weeks, have an Eastern Cooperative Oncology Group performance status of 0, 1 or 2 and a life expectancy of at least 3 months. Adequate haematological reserve (WBC >3.0 x 109/l and platelets >100 x 109/l), liver biochemistry [bilirubin <1.5-fold upper limit of normal (ULN), transaminases and alkaline phosphatase <=2-fold ULN] and renal function (serum creatinine <=150 µmol/l). Patients were ineligible if they had previously received high-dose radiotherapy, a stereotactic boost or brachytherapy. They were also excluded if they had undergone surgery, radiotherapy or received chemotherapy for recurrent disease.

XR5000 was administered at a dose of 3010 mg/m2, diluted in 250 ml of normal saline. It was delivered as a 120-h (5 day) continuous infusion through an indwelling central venous catheter using an ambulatory pump. Prophylactic anti-emetics were given according to local practice. Chemotherapy was repeated every 3 weeks, with each cycle comprising 5 days of XR5000 and 16 days off treatment. Treatment could be delayed by up to 2 weeks if toxicities had not resolved by day 21. Dose modifications for toxicity were not planned. Patients were to be treated until disease progression or unacceptable toxicity occurred, or until they or their physician considered further XR5000 inappropriate.

Response was assessed according to McDonald’s criteria [6] based on the bi-dimensional size of the enhancing tumour mass on MRI, steroid intake and neurological status according to the Edinburgh Functional Impairment Test (EFIT) score [7]. MRI evaluations were to be carried out at baseline and after every two cycles of treatment on the same machine and using the same scanning protocol. Neurological examination was undertaken prior to each treatment cycle. In brief, the EFIT score has four components: (i) the Nine Hole Peg Test is a timed assessment of manual dexterity; (ii) the Ten Metre Walk is a timed test of mobility; (iii) the Williams Delayed Recall Test is a short memory test and (iv) the Boston Aphasia Severity Rating Scale is a subjective test of dysphasia. The initial EFIT score of 0–4 is based on the number of abnormal subsets at baseline. After the tests have been repeated an index of change is calculated for each one, with a value of +1 indicating significant improvement, 0 no change and –1 significant deterioration. These four scores are summed to generate the EFIT score as follows: EFIT >0, neurological improvement; EFIT = 0, no change; EFIT <0, neurological deterioration. Treatment toxicity was graded according to CTC, version 2.0.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Sixteen patients were entered from October 1999 to May 2000 at six institutions. One patient was ineligible, not having had a baseline MRI scan, but received two cycles of XR5000 before going off study because of clinical progression and is included only in the analysis of safety.

The prior treatments and clinical characteristics including EFIT score at study entry for the eligible patients are shown in Table 1. Of these 15 patients, 13 were receiving a stable dose of steroids on entry to the study; seven were on dexamethasone (median dose 9 mg/day, range 4–16), three on methylprednisolone (median dose 20 mg/day, range 20–32) and three on prednisolone or prednisone (median dose 80 mg/day, range 20–80). The remaining two patients were not on steroids at study entry.


View this table:
[in this window]
[in a new window]
 
Table 1. Characteristics of eligible patients
 
The 15 eligible patients received a total of 30 cycles (median 2, range 1–5). One patient received a single cycle of XR5000 over 3 days so the median delivered dose intensity was 99% (range 77–110%). Fourteen patients came off study because of progressive disease; the remaining patient had a cardiac arrhythmia (grade 1 supraventicular tachycardia judged related to study drug) but soon after was confirmed as having disease progression.

An independent assessment panel assessed the best response for all patients based on MRI, EFIT score and steroid requirement. There were no responses and the study closed at the first stage. Fourteen patients had progressive disease; all had radiological progression, 10 with a deterioration in EFIT score and three with a stable EFIT (there was one patient in whom a repeat EFIT was not done). Two of the patients with progression on MRI and a stable EFIT had their dose of steroids increased and the other had them reduced over this period. One patient had stable disease, with no change on MRI and an improved EFIT score whilst on a stable dose of steroids. The median time to progression was 37 days (95% confidence intervals: 28 to 42 days). Twelve patients died during the follow-up period.

Treatment was generally well tolerated (Table 2). In particular, haematological toxicity was mild. During the first cycle of treatment one patient had uncomplicated grade 3 leucopenia and grade 1 neutropenia together with grade 1 thrombocytopenia; one had grade 1 leucopenia alone. A further patient had grade 2 thrombocytopenia and two had grade 1 thrombocytopenia. One patient with no past history of ischaemic heart disease developed chest pain lasting 5–10 min shortly after starting their first infusion of XR5000. The pain was described as ‘tight’ and similar in character to angina pectoris but resolved spontaneously and did not recur during the second cycle. A second patient experienced constrictive chest pain that was transient and mild (grade 1). A further patient developed right bundle branch block and supraventricular tachycardia that again resolved and were not severe (grade 1). Central nervous system signs and symptoms were reported but were judged not attributable to XR5000. Alopecia was mild (grade 1) but most patients received only two cycles of treatment.


View this table:
[in this window]
[in a new window]
 
Table 2. Grade 3 or 4 toxicities (related to XR5000)
 

    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The main finding of this study is that although XR5000 was well tolerated, further evaluation in patients with glioblastoma multiforme using this dose and schedule of administration is not justified. However, this study illustrates many points of importance for future trials of novel agents in patients with glioblastoma multiforme.

Assessing response in patients with brain tumours is difficult. Although many trials focus on imaging alone, there is often a disparity between radiological changes and neurological function [8]. Imaging studies may show apparent improvement while the patient is deteriorating clinically [9] and subjective reports by neurosurgeons of altered neurological function can be unreliable. We, therefore, incorporated EFIT, a robust and validated measure of impairment in patients with malignant gliomas [7] in the current trial. From a practical point of view the EFIT assessments and radiology were largely in agreement. All patients had disease progression on MRI; of the three patients with stable EFIT scores, two were on increased doses of steroids. Hence, although it is unclear why XR5000 was not active, we can be confident that we did not overlook any clinically relevant benefits from XR5000 in patients with glioblastoma multiforme.

In preclinical testing XR5000 was a promising anti-cancer agent, so why has it so far proved inactive, not only in glioblastoma multiforme but also in ovarian [10], colorectal [11] and non-small-cell lung cancer [C. Dittrich, personal communication]. The first possibility is that dual inhibition of topoisomerase I and II may not be desirable in practice. In some cell lines resistance to inhibitors of one topoisomerase is associated with increased sensitivity to the other [12]. Combining inhibitors of topoisomerases I and II is beneficial in some human cell lines [13] but not in others [14]. In the clinic the combination of irinotecan and etoposide was active but associated with an unacceptable level of myelosuppression [15]. In contrast to the use of established inhibitors of topoisomerases I and II, XR5000 has the ability to overcome a number of mechanisms of drug resistance so there were additional reasons for anticipating clinical efficacy.

Another possible explanation for the disappointing results to date with XR5000 may be that the wrong schedule has been evaluated. In the earlier phase I trials, administration over 3 h was limited by infusion-related toxicities so prolonged administration appears the only feasible approach. Indeed, an initial phase II programme evaluating the 3-h infusion over 3 successive days was terminated when a patient had a myocardial infarct judged to be definitely related to XR5000 (Bevan P, Xenova, Slough, UK; personal communication). Infusion over 120 h was an attractive alternative as it allowed approximately a four-fold escalation in dose and achieved an area under the concentration time curve about seven-fold greater than seen with administration over 3 h on a single day (Propper DJ, de Bono J, Saleem A et al, unpublished observations) [16].

Alternatively, the dose of XR5000 tested here may have been too low. The low incidence of myelotoxicity in a study of an antiproliferative agent does raise this possibility. How-ever, in the phase I trial one of eight patients treated at the 3010 mg/m2 dose level did develop grade 4 neutropenia. Moreover, XR5000 pharmacokinetics showed approximately four-fold variability and appeared non-linear at these doses (Propper DJ, de Bono J, Saleem A et al, unpublished observations). Hence, the occurrence of severe chest and abdominal pain with XR5000 at a dose of 4060 mg/m2 was a cause for concern and precluded use of XR5000 at a dose above 3010 mg/m2 in the current trial.

Although the results of this trial are disappointing, there remains an urgent need for new treatments in patients with glioblastoma. Radiotherapy and surgery, either alone or in combination, remain the mainstay of treatment for these patients. Chemotherapy, comprising the nitrosoureas (either BCNU alone or CCNU in combination with procarbazine and vincristine) or temozolamide as a single agent are both used either as adjuvant therapy or at relapse. However, the benefits of such chemotherapy are modest, either because many cytotoxics do not easily cross the blood–brain barrier or because these tumours are inherently resistant to chemotherapy. There is increasing interest not only in new cytotoxics but also in novel therapies building on advances in the understanding of the molecular biology of gliomas. These approaches include targeting platelet-derived growth factor (PDGF), epithelial growth factors (EGFR) and their receptors, and gene therapy. It is unclear what impact these treatments will have but there is no doubt that there will be an increasing need to assess novel therapies in patients with gliomas.

In conclusion, XR5000 given by this route and schedule does not have clinically useful activity in patients with recurrent glioblastoma multiforme. Given the toxicity associated with shorter infusions, it would be difficult to justify further evaluation of XR5000 unless an alternative method of delivery, perhaps using liposomal or polymer systems, can be developed. This trial does, however, illustrate how the design of early clinical trials can be adapted in patients with glioblastoma multiforme.


    Acknowledgements
 
We are grateful to Xenova, Slough, UK as sponsors of the trial, also to P. Bevan (Xenova) and J. Steiner (Oxford Therapeutics Consulting, Oxford, UK) for their assistance in the conduct of the trial.


    Footnotes
 
+ Correspondence to: Dr C. Twelves, Cancer Research Campaign Department of Medical Oncology, CRC Beatson Laboratories, Alexander Stone Building, Switchback Road, Bearsden, Glasgow G61 1BD, UK. Tel: +44-141-330-3890; Fax: +44-141-330-4063; E-mail: c.twelves@beatson.gla.ac.uk Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Atwell GJ, Rewcastle GW, Baguley BC, Denny WA. Potential anti-tumour activity agents. In vivo solid tumour activity of derivatives of N-[2-(dimethylamino)ethyl]acridine-4-carboxamide. J Med Chem 1987; 30: 664–669.[ISI][Medline]

2. Baguley BC, Holdaway KM, Fray LM. Design of DNA intercalators to overcome topoisomerase II-mediated multi-drug resistance. J Natl Cancer Inst 1990; 82: 398–402.[Abstract]

3. McCrystal MR, Evans BD, Harvey VJ et al. Phase I study of the cytotoxic agent N-[2-(dimethylamino)ethyl] acidine-4-carboxamide. Cancer Chemother Pharmacol 1999; 44: 39–44.[ISI][Medline]

4. Twelves CJ, Gardner C, Flavin A et al. Phase I and pharmacokinetic study of DACA (XR5000): a novel inhibitor of topoisomerase I and II. Br J Cancer 1999; 80: 1786–1791.[ISI][Medline]

5. Cornford EM, Young D, Paxton JW. Comparison of the blood–brain barrier and liver penetration of acridine antitumor drugs. Cancer Chemother Pharmacol 1992; 29: 439–444.[ISI][Medline]

6. MacDonald DR, Cascino TL, Schold SC, Cairncross JG. Response criteria for phase II studies of supratentorial malignant glioma. J Clin Oncol 1990; 8: 1277–1280.[Abstract]

7. Grant R, Slattery J, Gregor A, Whittle IR. Recording neurological impairment in clinical trials of glioma. J Neuro Onol 1994; 19: 37–49.

8. Gregor A, Rampling R, Aapro M et al. Phase II study of tauromustine in malignant glioma. Rain Tumor Co-operative Group Trial 8001. J Neurosurg 189; 71: 1–9.

9. Imperato JP, Paleologos NA, Vick NA. Effects of treatment on long-term survivors with malignant astrocytomas. Ann Neurol 1990; 28: 818–822.[ISI][Medline]

10. Dieras V, Kerbrat P, Punt C et al. Phase II study of XR5000 administered as a 120-hour intravenous infusion in advanced ovarian cancer. Proc Am Assoc Cancer Res 2001; 42: 48 (Abstr 253).

11. Caponigro F, Dittrich C, Sorensen JB et al. Phase II study of XR5000 administered as a 120-hour intravenous infusion in advanced colorectal cancer. Proc Am Assoc Cancer Res 2001; 42: 620 (Abstr 3335).

12. Vasey PA, Kaye SB. Combined inhibition of topoisomerase I and II—is this a worthwhile/feasible strategy? Br J Cancer 1997; 76: 1494–1499.[ISI][Medline]

13. Kano Y, Suzuki K, Akatsu M et al. Effects of CPT-11 in combination with other anti-cancer agents in culture. Int J Cancer 1992; 50: 604–610.[ISI][Medline]

14. Kaufman SH. Antagonism between camptothecan and topoisomerase II-directed chemotherapeutic agents in a human leukaemia cell line. Cancer Res 1991; 51: 1129–1136.[Abstract]

15. Karato A, Saski Y, Shiraishi J et al. Phase I study of CPT-11 and etoposide in patients with refractory solid tumours. J Clin Oncol 1993; 11: 2030–2035.[Abstract]

16. Kestell P, Dunlop IC, McCrystal MR et al. Plasma pharmacokinetics of N-[2-(dimethylamino)ethyl]acridine-4-carboxamide in a phase I trial. Cancer Chemother Pharmacol 1999; 44: 45–50.[ISI][Medline]





This Article
Abstract
Full Text (PDF)
E-letters: Submit a response
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (5)
Disclaimer
Request Permissions
Google Scholar
Articles by Twelves, C.
Articles by Fumoleau, P.
PubMed
PubMed Citation
Articles by Twelves, C.
Articles by Fumoleau, P.