Salvage Therapy with the Combination of ARA-C and Gemtuzumab Ozogamicin in Post-Allo-SCT High Risk AML Patients Is Feasible and Results in Modest Responses.

Abstract 2629

Acute myeloid leukemia (AML) relapse is often associated with a poor outcome, in particular in the post allogneic stem cell transplantation ( Allo-SCT) setting. Therapeutic options in patients relapsing early after SCT are further limited partially due to the risk of increased toxicity. In patients who did not develop graft versus host disease ( GVHD), reduction or cessation of immunosuppression therapy may result in the achievement of a graft versus leukemia (GvL) effect. However, this is usually only beneficial in patients with a low disease burden. Gemtuzumab Ozogamicin (GO) was introduced into therapy of CD33 positive AML patients as a targeted therapy with a relatively safe extra-medullary toxicity profile. Although withdrawn from the USA market by the FDA, recent studies have shown efficacy of combining GO with chemotherapy in newly diagnosed AML patients. We hypothesized that in patients with AML relapsing early after an Allo-SCT, combination of GO with single agent chemotherapy will result in disease control for a sufficient amount of time to allow a trial of GvL induction. Therapy consisted of high dose ARA-C (1 gr/m²) for 4 days followed by one dose of GO, 9 mg/m² on day 5. To decrease hematopoietic toxicity, therapy was supported by donor stem cells, when available. Responding patients which did not develop GVHD, where eligible for cell mediasted immunotherapy with a donor lymphocyte infusion(DLI) or a second Allo-SCT. 16 patients (F=10, M=6), median age 52 years (range 26–63) are included in this analysis. The majority (14/16) of patients were previously transplanted for a high risk AML (secondary AML-6, unfavorable cytogenetics-5, FLT3-ITD-3); disease status at SCT were CR1 (n=8), CR2 (n=4), untreated 1^st relapse (n=1) and at diagnosis (n=1, MDS with excess of blasts). AML relapse occurred at a median of 104 days (range 15–403) after SCT. ARA-C/GO therapy was supported with donor stem cells in 8/16 patients and was relatively well tolerated. One patient died in aplasia in day +2 due to sepsis resulting in overall 6% induction mortality. Five patients developed veno occlusive disease of the liver ( VOD )(36% of evaluable patients); all were mild and resolved with symptomatic care. Responses were achieved in 8/15 (53%) evaluable patients, all CR/CRp. Four patients in CR/CRp received further therapy with DLI (n=3) or Allo SCT from a different donor (n=1), and 2 non-responding patients received a second Allo-SCT after second line salvage chemotherapy. At long follow up, 1 patient (6%) is alive in and continuous remission (1040 days after therapy). In patients surviving induction, death occurred due to relapse (n=10) or severe GVHD (n=5). Median EFS and OS were 55 and 133 days, respectively in the entire cohort. Median EFS was longer in responding patients (156 vs. 16 days, p<0.0001), while median OS was not significantly different in these 2 patients cohorts (204 vs. 33 days, p=0.2).

ARA-C/GO combination is a valid option for salvage therapy in AML patients relapsing after an Allo-SCT and is well tolerated. Although responses are usually of a short duration, they may allow immunogenic manipulation in the absence of an already established significant GVHD, and may result in a long term disease control in a minority of patients.