Phase II Study on Gemtuzumab Ozogamicin in Combination With All-trans-Retinoic Acid, High-dose Cytarabine and Mitoxantrone in Patients With Primary Refractory Acute Myeloid Leukemia

Primary refractory AML is associated with an extremely poor prognosis [1,2]. In the AMLHD93 trial conducted by the AMLSG ULM, patients refractory to the first induction therapy with ICE (idarubicin, cytarabine, etoposide) had an overall survival of 12% after 5 years [1]. All patients alive in this cohort had received allogeneic transplantation. Therefore, we assigned allogeneic transplantation in our consecutive trial, AMLHD98A, to all primary refractory patients [3]. However, the main problem in this patient group remains achieving a partial (PR) or complete (CR) remission to a salvage therapy. Additionally, the pre-transplant disease status is an important prognostic factor in most studies of allogeneic transplantation, regardless dose intensified or dose reduced conditioning regimens are used [4,5,6]. Since 1993, in all studies of the German-Austrian-AMLSG response-adapted treatment strategies had been used. Within the AMLHD93 trial, refractory patients were assigned to an intensified second induction regimen with S-HAM (age<55 years) [7] or HAM (age 55 to 60 years) [1], and in the AMLHD98A trial, with A-HAM [3]. The incorporation of all-trans-retinoic acid was based on in vitro data [8-13] and by our randomised AMLHD98B study for elderly AML-patients showing a benefit in primary response and survival for patients assigned to standard induction therapy in combination with ATRA [14].

To compare the different salvage therapy strategies, we performed an as-treated analysis in primary refractory patients of the different cohorts. Although refractory to the first induction therapy with ICE, nine patients received a second cycle ICE. The results summarized in table 1 showed an improved response rate (CR and PR) for patients treated with the A-HAM protocol and thus leading to a higher proportion of patients receiving an allogeneic transplantation. Survival analysis showed so far no difference between the 4 different groups. Gemtuzumab ozogamicin (GO) is a humanized anti-CD33 conjugated to Calicheamicin. The efficacy and the toxicity profile has been evaluated in several studies, so far the substance is approved for the monotherapy in relapsed AML-patients in a dose of 9mg/m² q 14d [15]. However, used as a single agent the efficacy is limited and not durable. Therefore, several trials have evaluated GO in combination with conventional chemotherapy [16,17]. In the MRC study a dose of 6 mg/m² given once at day 1 was associated with an increased liver toxicity and therefore the study continues with a dose of 3 mg/m² once at day 1 of induction therapy [17]. In summary, the available data for combination therapy showed efficacy of GO in phase II trials. The dose limiting toxicity was defined in the MRC trial at 6 mg/m². Therefore we consider GO in combination with A-HAM for primary refractory adult AML patients. Because all primary refractory patients are candidates for an allogeneic transplantation special considerations have to be taken with respect to the development of VOD after allogeneic transplantation. One recent report suggests a substantial risk for VOD for patients receiving an allogeneic transplantation after a therapy with GO [18]. In this report the odds ratio for VOD after a therapy with GO within 3.5 months before allogeneic transplantation was 21.6 (95%-confidence interval 4.2-112.2%). However, this report is based on 62 patients and the dosage of GO used was 6mg/m² and 9mg/m². Therefore, holding in mind the risk of VOD after GO exposure and the extremely poor prognosis of primary refractory patients the treatment approach combining A-HAM with GO with a dose of 3mg/m² is justified.