Annals of Oncology Advance Access originally published online on September 13, 2006
Annals of Oncology 2006 17(12):1826-1829; doi:10.1093/annonc/mdl309
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2006 European Society for Medical Oncology
melanoma |
A randomized phase II trial of gemcitabine plus treosulfan versus treosulfan alone in patients with metastatic uveal melanoma
1 Departments of Internal Medicine III (Hematology, Oncology and Transfusion Medicine), Campus Benjamin Franklin, Berlin, Germany
2 Departments of Biostatistics and Clinical Epidemiology, Campus Benjamin Franklin, Berlin, Germany
3 Departments of Ophthalmology, Charité, Campus Benjamin Franklin, Berlin, Germany
* Correspondence to: Dr A. Schmittel, Department of Internal Medicine III (Hematology, Oncology and Transfusion Medicine), Charité, Campus Benjamin Franklin, Hindenburgdamm 30, 12200 Berlin, Germany. Tel: +49-30-8445-3906; Fax: +49-30-8445-4468; E-mail: alexander.schmittel{at}charite.de
 |
Abstract |
|---|
 Top Abstract introductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Background: Several trials demonstrated efficacy of the gemcitabine/treosulfan (GeT) combination in metastatic uveal melamoma. This randomized phase II trial compared the GeT combination versus treosulfan alone (T) in this rare disease.
Patients and methods: Chemotherapy-naive patients with proven metastatic uveal melanoma were randomly assigned to receive 1000 mg/m2 of gemcitabine plus 3500 mg/m2 of treosulfan (GeT) or 3500 mg/m2 of T. Chemotherapy was administered on days 1 and 8 in both arms, cycles were repeated on day 29. Primary end point was rate of responses and disease stabilizations.
Results: Forty-eight patients were randomized. Seven confirmed stable diseases (SDs) and one partial remission (PR) were observed in 24 patients treated with the GeT regimen, whereas no PR and only three SDs were observed in the T arm (P = 0.08). Median progression-free survival (PFS) was 3 months (95% CI 1.1–4.9) and 2 months (95% CI 1.7–2.3) in the GeT and T arm (P = 0.008, log-rank). Six and 12 months PFS was 34.8% and 17.9% and 16.7% and 0% always favoring the GeT arm.
Conclusions: This first randomized trial in metastatic uveal melanoma showed a superior PFS and a trend for a higher response/stabilization rate of the GeT combination over T.
Key words: gemcitabine, randomized phase II, treosulfan, uveal melanoma
|
introduction |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Melanoma of the uvea has an annual incidence of six cases in one million [1]. Thus, it is the most common primary intraocular malignant tumor of the adult. Approximately 35% of patients will develop metastatic disease at various time points after primary therapy [2], with 16% metastases after a 5-year follow-up period [3].
Median survival in the metastatic stage is between 2 and 15 months after detection of metastases [4–8]. A variety of cytotoxic agents have been investigated, among them dacarbazine, temozolomide, 9-nitrocamptothecin, fotemustin, cisplatin and combination therapies like BOLD (bleomycin, vincristine, lomustine, dacarbazine). The response rates ranged from 0% to 20% in larger trials with a median survival of 4–12 months [4, 9–12]. A higher response rate of 35.5% and a median survival of 14.8 months were observed in a phase II trial with intra-arterial administration of fotemustine into the hepatic artery [13]. Intra-arterial chemoembolization using carmustine (BCNU) dissolved in ethiodized oil and gelatine sponge particles for transient arterial occlusion resulted in an objective response rate of 37% and a median overall survival (OS) of only 5.2 months [14].
On the basis of in vitro experiments suggesting a synergistic activity of gemcitabine and treosulfan [15–17], two phase I trials evaluating the gemcitabine/treosulfan (GeT) combination were carried out. One trial included patients with metastatic cutanous or uveal melanoma and recommended a dose of 2500 mg/m2 of gemcitabine plus 5000 mg/m2 of treosulfan both administered on day 1 every 3 weeks [18]. The other phase I trial defined a maximum tolerated dose of 1000 mg/m2 of gemcitabine plus 3000–3500 mg/m2 of treosulfan both administered on days 1 and 8 every 4 weeks [19]. This phase I trial was followed by a phase II trial, which comparatively analyzed two cohorts of patients treated with 1000 mg/m2 of gemcitabine plus 3000 mg/m2 or 3500 mg/m2 of treosulfan and suggested a favorable median OS of 9 months for patients treated with the higher dose of treosulfan and a benefit rate [partial remission (PR)+stable disease (SD)] of 58% [7]. On the basis of these results, we initiated this randomized phase II trial to investigate, whether the in vitro synergy of gemcitabine and treosulfan is clinically relevant.
|
patients and methods |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
patient selection
Patients with histologically confirmed metastatic uveal melanoma were eligible for treatment within this trial. Eligibility criteria included no prior chemotherapy regimen, age
18 years, life expectancy of >3 months, Karnofsky performance status (KPS) 60%, measurable lesions, adequate hematologic function, hepatic, renal or cardiac function. One prior cytokine therapy was allowed, the trial was approved by the local ethics committee.
trial design
Randomization was stratified by KPS (90% versus <90%). The primary end point was the stabilization plus response rate. The number of patients to be included in the study had been calculated according to Simon's optimal two-stage minimax design [20]. The criterion for a further evaluation of one of the regimens in future phase III trials was a stabilization/response rate exceeding 25%. In case of a response/stabilization rate <5%, the study was planned to be terminated early. To minimize the sample size in case of low efficacy according to the minimax design, the response rate was analyzed after a first step of nine assessable patients in each arm. In case of at least one stabilization or objective response out of nine patients in step 1, 15 additional patients were included into step 2. If >2 out of 24 patients had a stabilization or an objective response after this second step, it can be concluded, with an alpha error of 10%, that the efficacy of the regimen warrants further investigation within future trials. In addition, toxic effect, progression-free survival (PFS) and OS were analyzed as secondary end points.
chemotherapy
In arm A, 1000 mg/m2 of gemcitabine plus 3500 mg/m2 of treosulfan were administered on days 1 and 8 and in arm B, 3500 mg/m2 of treosulfan was given on days 1 and 8. Cycles were repeated on day 29 in both the arms. All patients received antiemetic therapy consisting of 5-HT3 antagonists, before starting the infusion of chemotherapy.
response evaluation
Computed tomography (CT)-based response evaluation according to the RECIST criteria [21] was carried out before initiation of chemotherapy and after two, four and six cycles of chemotherapy. In the follow-up period, CT scans were carried out every 8 weeks.
statistical analysis
All analyses were carried out on an intent-to-treat basis considering all patients randomized. PFS was estimated by using the Kaplan–Meier method. The log-rank test was used to determine statistical significance. To analyze differences in toxicity and response rate, the 
2 test was used. The statistical analysis was carried out using the SPSS software (release 9.0).
|
results |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
patient characteristics
A total of 48 patients have been randomized with 24 patients into each arm. Median age was 63 (range 30–80) years in the GeT arm and 58 (range 30–79) years in the T arm. Randomization was stratified by KPS. KPS was well balanced with 19 patients with KPS
90% in the GeT arm and 18 in the T arm. Nine and seven patients in the GeT and T arm had prior cytokine therapy consisting of interleukin 2 ± histamine dihydrochloride. All patients had documented disease progression during cytokine therapy before entering the trial. Patients, who had no prior therapy had been randomized within 4 weeks after first diagnosis of metastatic disease. Lactate dehydrogenase level and the number of organs with metastatic involvement were evenly distributed (Table 1). A total of 15 patients in each arm received second line therapy, of which the majority had fotemustine chemotherapy (Table 1).
|
response analysis
According to the statistical design of the trial after a first step of nine patients in each arm, the numbers of objective remissions (ORs) and SDs were analyzed in both arms. Since in each arm at least one OR or SD was observed, accrual continued until 24 patients were included into each treatment arm. In the GeT arm, one patient had a confirmed PR and seven a disease stabilization (PR + SD = 33%), whereas no OR and only three SDs (13%) were observed in the T arm (P = 0.073; Table 2).
|
survival analysis
Median PFS of patients treated in GeT arm was 3 months (95% CI 1.1–4.9) and 2 months (95% CI 1.7–2.3) in the treosulfan arm (P = 0.008, log-rank). Six and 12 months PFS was 35% and 18% in the GeT arm and 17% and 0% in the T arm (Figure 1), respectively.
|
toxicity analysis
Grades III and IV leucopenia only occurred in the GeT arm in four of 24 patients (17%) (P = 0.001). Two of these patients (8%) experienced a febrile neutropenia (Table 3). Frequencies of anemia, nausea, vomiting and infections were not significantly different in both treatment arms (for details see Table 3).
|
|
discussion |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Until now, treatment of metastatic uveal melanoma has not been investigated systematically in randomized trials. In retrospective analyses, cisplatin- or dacarbazine-based chemotherapy regimens showed low efficacy [9]. The BOLD plus interferon regimen, which consists of bleomycin, vincristine, lomustine (CCNU) and dacarbazine (DTIC), demonstrated a promising response rate of 20% in an early phase II trial [22]. Therefore, this regimen was investigated more intensively in a multicenter European Organization for Research and Treatment of Cancer phase II study. This trial failed to show an acceptable efficacy and revealed a high toxicity, which lead to the conclusion that BOLD plus interferon will no longer be investigated in metastatic uveal melanoma [12]. Therefore, until today no standard chemotherapy regimen has been developed.
New aspects for the development of chemotherapy regimens in uveal melanoma came from a variety of ex vivo investigations. In vitro studies demonstrated that highly chemoresistant tumor cells, including human uveal melanoma cells, show considerable sensitivity to the alkylating agent treosulfan [15]. The capability of treosulfan to alkylate or cross-link DNA is preserved in multidrug-resistant cell lines [16]. Using an adenosine triphosphate (ATP)-based chemosensitivity assays (ATP-trichloroacetic acid), a considerable synergistic in vitro interaction between treosulfan and the nucleoside analogs cytarabine and gemcitabine was demonstrated in uveal melanoma cells [17]. The mechanism of synergy may be explained by the inhibition of DNA repair mechanisms by both nucleoside analogs [23]. The present trial was carried out to evaluate, whether the postulated mechanism of synergy is clinically relevant.
Objective responses reported in previous trials varied between 0.3% and 28.6% [19, 24] and the rate of disease stabilization between 45% and 57%. Although a prospective comparison of both arms of our trial had not been calculated in the statistical design of the trial, the post hoc analysis revealed a favorable rate of remissions and SDs of 33% in the GeT arm when compared to 12.5% in the treosulfan-alone arm. Median PFS of the GeT arm was 3 months, which was similar to our earlier trial [7], but still much shorter than reported by Pföhler et al. [24]. This difference might be due to selection of patients. However, in the comparison of GeT versus T, our randomized phase II trial revealed a PFS and a response and stabilization rate favoring the GeT arm in line with the in vitro synergy of both drugs. Since the toxicity of the GeT regimen was only slightly higher as compared to T, we recommend the GeT regimen as a reference for further randomized trials, although a direct comparison of both arms had not been prospectively calculated.
|
Acknowledgements |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
The authors thank Susanne Beier for excellent data management and patient care.
Received for publication May 30, 2006. Revision received July 19, 2006. Accepted for publication July 19, 2006.
|
References |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
1. Egan KM, Seddon JM, Glynn RJ, et al. (1988) Epidemiologic aspects of uveal melanoma. Surv Ophthalmol 32:239–251.[CrossRef][Web of Science][Medline]
2. Diener-West M, Hawkins BS, Markowitz JA, Schachat AP. (1992) A review of mortality from choroidal melanoma. II. A meta-analysis of 5-year mortality rates following enucleation, 1966 through 1988. Arch Ophthalmol 110:245–250.
3. Schmittel A, Bechrakis NE, Martus P, et al. (2004) Independent prognostic factors for distant metastases and survival in patients with primary uveal melanoma. Eur J Cancer 40:2389–2395.[CrossRef][Web of Science][Medline]
4. Bedikian AY, Legha SS, Mavligit G, et al. (1995) Treatment of uveal melanoma metastatic to the liver: a review of the M. D. Anderson Cancer Center experience and prognostic factors. Cancer 76:1665–1670.[CrossRef][Web of Science][Medline]
5. Gragoudas ES, Egan KM, Seddon JM, et al. (1991) Survival of patients with metastases from uveal melanoma. Ophthalmology 98:383–389.[Web of Science][Medline]
6. Kath R, Hayungs J, Bornfeld N, et al. (1993) Prognosis and treatment of disseminated uveal melanoma. Cancer 72:2219–2223.[CrossRef][Web of Science][Medline]
7. Schmittel A, Schuster R, Bechrakis NE, et al. (2005) A two-cohort phase II trial of gemcitabine plus trosulfan in patients with metastatic uveal melanoma. Melanoma Res 15:441–451.[CrossRef][Web of Science][Medline]
8. Kodjikian L, Grange JD, Baldo S, et al. (2005) Prognostic factors of liver metastases from uveal melanoma. Graefes Arch Clin Exp Ophthalmol 243:985–993.[CrossRef][Web of Science][Medline]
9. Flaherty LE, Unger JM, Liu PY, et al. (1998) Metastatic melanoma from intraocular primary tumors: the Southwest Oncology Group experience in phase II advanced melanoma clinical trials. Am J Clin Oncol 21:568–572.[CrossRef][Web of Science][Medline]
10. Pyrhonen S, Hahka-Kemppinen M, Muhonen T. (1992) A promising interferon plus four-drug chemotherapy regimen for metastatic melanoma. J Clin Oncol 10:1919–1926.[Abstract]
11. Becker JC, Terheyden P, Kampgen E, et al. (2002) Treatment of disseminated ocular melanoma with sequential fotemustine, interferon alpha, and interleukin 2. Br J Cancer 87:840–845.[CrossRef][Web of Science][Medline]
12. Kivelä T, Suciu S, Hansson J, et al. (2003) Bleomycin, vincristine, lomustine and dacarbazine (BOLD) in combination with recombinant interferon alpha-2b for metastatic uveal melanoma. Eur J Cancer 39:1115–1120.[CrossRef][Web of Science][Medline]
13. Leyvraz S, Bosshard W, Salmon R, et al. (2002) Prolonged survival of patients with liver metastases from ocular melanoma: multicentric experience with fotemustin hepatic arterial infusion. J Clin Oncol 20:1360.
14. Patel K, Sullivan K, Berd D, et al. (2005) Chemoembolization of the hepatic artery with BCNU for metastatic uveal melanoma: results of a phase II study. Melanoma Res 15:297–304.[CrossRef][Web of Science][Medline]
15. Hartley JA, O'Hare CC, Baumgart J. (1999) DNA alkylation and interstrand crosslinking by treosulfan. Br J Cancer 79:264–266.[Web of Science][Medline]
16. Fritz G, Hengstler JG, Kaina B. (1997) High-dose selection with mafosfamide results in sensitivity to DNA cross-linking agents: characterization of hypersensitive cell lines. Cancer Res 57:454–460.
17. Neale MH, Myatt N, Cree IA, et al. (1999) Combination chemotherapy for choroidal melanoma: ex vivo sensitivity to treosulfan with gemcitabine or cytosine arabinoside. Br J Cancer 79:1487–1493.[CrossRef][Web of Science][Medline]
18. Corrie PG, Shaw J, Spanswick VJ, et al. (2005) Phase I trial combining gemcitabine and treosulfan in advanced cutaneous and uveal melanoma patients. Br J Cancer 92:1997–2003.[CrossRef][Web of Science][Medline]
19. Keilholz U, Schuster R, Schmittel A, et al. (2004) A clinical phase I trial of gemcitabine and treosulfan in uveal melanoma and other solid tumors. Eur J Cancer 40:2047–2052.[CrossRef][Web of Science][Medline]
20. Simon R. (1989) Optimal two-stage designs for phase II clinical trials. Control Clin Trials 10:1–10.[Medline]
21. Therasse P, Arbuck SG, Eisenhauer EA, et al. (2000) New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 92:205–216.
22. Nathan FE, Berd D, Sato T, et al. (1997) BOLD+interferon in the treatment of metastatic uveal melanoma: first report of active systemic therapy. J Exp Clin Cancer Res 16:201–208.[Web of Science][Medline]
23. Plunkett W and Gandhi V. (2001) Purine and pyrimidine nucleoside analogs. Cancer Chemother Biol Response Modif 19:21–45.[Medline]
24. Pföhler C, Cree IA, Ugurel S, et al. (2003) Treosulfan and gemcitabine in metastatic uveal melanoma patients: results of a multicenter feasibility study. Anticancer Drugs 14:337–340.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D.K. Wilkins and P.D. Nathan Review: Therapeutic opportunities in noncutaneous melanoma Therapeutic Advances in Medical Oncology, July 1, 2009; 1(1): 29 - 36. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||