Exhibit 99.6

Abstract 3243

An in Vitro Platform to Dissect Drug Responsiveness in Refractory Anemia with Ringed Sideroblasts (RARS)

Daniela Grazynska(1)*, Sheherzad Preisler(1)*, Michael Churchill, BS(1)*, Abdullah Mahmood Ali, PhD(1), Azra Raza, MD(1) and Siddhartha Mukherjee, MD, PhD(1),(2)*

(1)Herbert Irving Comprehensive Cancer Center, Columbia University Medical Center, New York, NY;

(2)Department of Medicine, Division of Oncology, Columbia University Medical Center, New York, NY

Background: Despite intensive investigation, the pathophysiology of low-risk myelodysplastic syndrome (LR-MDS) remains unknown and therapeutic options are limited. In a phase-II trial, treatment with oral rigosertib, a small molecule inhibitor of PI3K and PLK pathways, was shown to have a 39% response rate (i.e., transfusion independence), in patients with LR-MDS. Notably, rigosertib caused an increase in hemoglobin, a decrease in transfusion requirements, and, in some patients who were previously refractory, “re-sensitization” to erythropoietin (EPO) therapy. However, thus far, the mechanism of this responsiveness remains unknown. Whole exome sequencing revealed a broad spectrum of mutations in these patients, including mutations in SF3B1, SRSF2 and TET2. However, there was no correlation between mutational spectrum and responsiveness.

Methods: Here, we have devised an in vitro platform to predict responsiveness, and dissect the rigosertib mechanism of action. This study is approved by the Institutional Review Board of Columbia University and informed consent was obtained from all the individuals participated in the study. CD34+ cells were isolated from bone marrow (BM) mono nuclear cells derived from BM aspirates. Bone marrow stromal cells (BMSC) were isolated from bone biopsies of patients with MDS. Normal CD34+ cells and BMSC were obtained from commercial sources. CD34+ cells were co-cultured with stromal cells plated a day before and stimulated with various concentration of EPO and rigosertib. Differentiation was assessed by staining cells with various markers of erythroid differentiation including GPA, band 3, Integrin alpha 4 antibodies conjugated with fluorophores and analyzed by FACS. We show that co-culture of CD34+ stem cells from the bone marrow of patients with their own BMSC recapitulates key features of the MDS phenotype and rigosertib responsiveness.

Results: Normal (non-MDS) CD34+ cells co-cultured with normal BMSC showed striking responsiveness to EPO stimulation, as evidenced by the increased production of CD45- low/GPA+/ band 3+/Integrin-Alpha4+ erythroid cells (2.47 fold increase compared to no EPO). In contrast, RARS CD34+ cells co-cultured with MDS stromal cells showed no erythroid differentiation with EPO stimulation (1.09 fold change compared to EPO). Notably, LR-MDS co-cultures showed increased erythroid differentiation with the addition of 20 nm and 100 nm rigosertib in the presence of EPO (1.28 and 1.4 fold increase, respectively, compared to EPO alone). Moreover, co-cultures obtained from 3/3 patients that showed responsiveness in vivo were responsive to combined EPO/rigosertib stimulation in vitro, while 2/2 non-responsive patients were not responsive in vitro.

Conclusions: We have created an in vitro platform to dissect the mechanism of rigosertib responsiveness in RARS patients. This platform is a novel co-culture-based system that recapitulates key features of the RARS phenotype. In the absence of reliable animal models for this disease, this platform may offer a viable method to characterize drug responsiveness, dissect mechanisms, and offer patient-specific drug responsiveness information, since individual CD34+ cells are co-cultured with a patient’s own BMSC.  In future, we wish to use this in vitro co- culture test to prospectively predict responsiveness to experimental drugs in patients, thereby focusing the drug on only selected patients.