Updated Phase II Study of Targeted Subcutaneous (SC) Bortezomib for Patients with Low- or Intermediate-1 (Int-1)-Risk Myelodysplastic Syndrome (MDS) with Evidence of NF-κb Activation

Introduction

MDS has been linked to constitutive activation of genes involved in the nuclear factor-kappaB (NF-κB) pathway [Wei et al. Leukemia 2013; Braun et al. Blood 2006]. Hence, NF-kB is an attractive therapeutic target in this disease. Bortezomib is a proteosome inhibitor with inhibitory activity against NF-κB. We designed a phase II trial of SC bortezomib in patients with low- to intermediate (low/int-1)-risk MDS and evidence of NF-κB activation to determine its therapeutic activity in these patients.

Methods

In this single-arm phase II study in low/int-1-risk MDS patients, bortezomib was administered at 1.3 mg/m² SC on days 1, 4, 8, and 11 in a 21-day cycle for a maximum of 2 years of therapy. Eligibility criteria included age ≥18 years, adequate performance status and organ function, and having received at least 1 prior therapy. Patients with grade 2 or greater peripheral neuropathy at baseline were excluded. Patients were prescreened prospectively for enrollment by assessing cellular levels of the phosphorylated NF-κB subunit p65 (pp65) in their marrow. This was performed by immunofluorescence with phospho-Ser276 in bone marrow aspirate smears in the Department of Hematopathology (CLIA regulations). Patients were eligible if at least 5% of all nucleated marrow cells were positive for pp65 staining. pp65 levels were assessed again on day 21 of cycles 1 and 2 and then as clinically indicated. Responses were assessed according to IWG [Cheson et al. Blood 2000; Cheson et al, Blood 2006]. The study could accrue a maximum of 40 patients if there were at least a 95% chance of at least a 15% ORR.

Results

Beginning 9/2013, we enrolled 15 patients with a median age of 71 years (range 56 - 87). Median marrow blast percentage was 1.9% (range 0 - 5%). Eleven patients (73%) had diploid cytogenetics, 2 had del(20q), 1 had Y-, and 1 had del(5q). All had lower-risk MDS by IPSS (low 33.3%, int-1 66.7%). All were transfusion dependent (6 both platelets and red cells (PRBCs), 1 only platelets, 8 only PRBCs). Hypomethylating agents had failed in 12 of the patients.

At a median follow-up interval of 22 weeks, the ORR was 20%; 3 patients had hematologic improvement with a mean duration of response of 14.3 weeks (range 4-21). Eight patients had stable disease, and 4 had progression. No correlation between clinical response and molecular or cytogenetic data was observed. Eventually, all patients were taken off study: 7 due to increasing transfusion requirements, 2 worsening cytopenia, 1 lack of response, 1 increased blasts, and 1 grade 2 neuropathy; 2 withdrew by choice, and 1 patient died from causes unrelated to the study. Four patients experienced ≥ 1 grade 3 toxicity. No grade 4 toxicity was observed. Seven patients experienced grade 1 (n=4) or grade 2 (n=3) neuropathy. Morphologic review (n=14) in the responders group (n=3) showed reduction in ring sideroblasts (RS) in 2 patients (60 to 47% and 43% to 30%). Two patients, including 1 of those with reduced RS, had improvement of dysplasia (from severe trilineage dysplasia to moderate bilineage dysplasia). No changes in cytogenetic studies were found in the responder group.

Among nonresponders (n=11), 3 had new acquired cytogenetic abnormalities, 3 had worsening dysplasia, and 2 had increase in blast count (1% to 3% and 1% to 4%). Two had no morphologic changes during treatment. One showed improvement in dysplasia and blast count (2% to 0%) and no changes in cytogenetics. Interestingly, 3 nonresponders showed RS reduction during treatment (18% to 1%, 85% to 45%, and 6% to 0%).

The median pp65 level at baseline was 30.87% (range 7 - 70%). The pp65 level decreased in 7 of the 15 patients (46.7%), in 6 of them by the end of cycle 1. Interestingly, the 3 responders were among the 7 patients who had a decrease in pp65 level. Eventual loss of response in these patients was accompanied by return to a higher pp65 level. In nonresponders, the pp65 level increased in 7, decreased in 2, and remained unchanged in 1; for 1, the sample was suboptimal.

Conclusions

In previously treated lower-risk MDS patients, SC Bortezomib was well tolerated and resulted in hematologic improvement and decrease in RS. NF-κB activation, measured by pp65 level, can be a useful biomarker to select patients with lower-risk MDS who could benefit from therapies that target this pathway. The NF-κB pattern of expression suggests an inverse relationship between treatment response and NF-κB level, with associated improvement in bone marrow morphology.