Tosedostat Plus Low Dose Cytarabine Induces a High Rate of Responses That Can be Predicted By Genetic Profiling in Elderly AML

Background: Older (age ≥60 years) patients with acute myeloid leukemia (AML) have poor outcomes and intensive induction chemotherapy is frequently unsuitable. Thus, new safe and effective therapies are urgently needed. Tosedostat, a new, orally bioavailable inhibitor of members of the M1 and M17 classes of aminopeptidases, was proven to be effective in both de novo and relapsed AML. We hypothesize that the addition of tosedostat to cytarabine may improve the response rate and remission duration over what is expected with chemotherapy or tosedostat alone.

Methods: This was a phase II, prospective, multicenter study, designed according to Fleming’s method. Fixing the lowest acceptable rate as 10% and the successful rate as 25%, with a significance level alpha=0.05 and a power 1-beta =0.80, the sample size was estimated in 33 patients. Thirty-three patients (median age 75 years) received Tosedostat 120 milligrams orally once daily until disease progression, coupled with intermittent low-dose cytarabine given subcutaneously at 20 milligrams twice/day for 10 days. Courses of cytarabine were repeated every 4 weeks in the absence of disease progression or unacceptable toxicity, up to 8 cycles. Global gene expression profiling (GEP, Affymetrix Human Gene 2.0 Array) was performed on purified AML blasts of 29 patients from peripheral blood or bone marrow at diagnosis before treatment initiation. Unsupervised clustering was generated using a hierarchical algorithm based on the average-linkage method. Principal component analysis (PCA) and supervised gene expression analysis was performed by using GeneSpring GX 12.0 (Agilent, USA).

Results: The characteristic of enrolled patients are listed in table 1 Induction-period mortality was 12%, with 4 deaths occurring in aplasia. According to intent-to-treat, the CR rate was 48.5% (16/33 patients); 2 additional patients obtained a partial response, for an overall response rate of 54.6%. In addition, 4/33 patients remained in stable disease for a median time of 9 months (range: 4-14). Seven patients did not respond and died with progressive disease after having received a median of 2 cycles of cytarabine and 45 days of tosedostat. In responding patients, the median time to best response was 74 days (range 22-145). Responding patients (CR+PR) had a longer median overall survival than non-responders (P=0.018). Six out of 18 (33%) responding patients are still in CR after a median follow-up of 425.5 days (range 208-758);  5 additional patients are alive with stable disease. Twenty-two patients died [while in aplasia (4), in CR (1), due to resistant disease (9) or due to progressive disease (8) after a median CR duration of 192 days (87-535)]. 29 patients had GEP analysis, and a molecular signature associated with the clinical response (CR vs. no CR) was identified. By supervised analysis, 212 genes differentially expressed based on the clinical response (complete remission (CR) vs no CR) were identified (Mann-Whitney, p<0.05; fold change >2). The 212 genes differentially expressed were significantly associated with six relevant biological functions and pathways: β-catenin (βcat); TNFα signaling pathway via NFκB; ERB2; STK33/SKM (serine/threonine kinase 33 expression using the SKM cell line); inflammatory response; and epithelial-mesenchymal transition pathways. 

Conclusions: The tosedostat and low-dose cytarabine combo produced a CR rate superior to what expected (45.4% versus 25%), and thus met the primary endpoint of study. Further, potential biomarkers were identified by GEP. Specifically, the achievement of CR could be efficiently predicted by the gene expression patterns with an overall accuracy exceeding 90%. A validation analysis is currently being conducted on additional 14 patients in order to confirm the ability of GEP to identify potential responders to TST. The study was registered at European Union Drug Regulating Authorities Clinical Trials (EudraCT) n.2012-000334-19.

Acknowledgements: CTI is gratefully acknowledged for providing Tosedostat for the patients. The study was supported in part by AIL Pesaro Onlus.

Table 1.

Characteristic	N = 33
Median age, years (range)	75 (62-85)
Median WBC count, × 109/L (range)	3.05 (0.26–24.53)
Blasts,   %  (range)	60   (20-96)
Cytogenetic risk group*, n
Not evaluable	3 (9%)
Intermediate Karyotype	17 (52%)
Unfavourable Karyotype	13 (39%)
AML, n
De novo	16 (48%)
Secondary, n	17 (52%)