Exhibit 99.9

Abstract 2598

Treatment Patterns Among Patients with Myelodysplastic Syndromes: Observations of 1st-Line Therapy, Discontinuation and the Need of Additional Therapies

Sudipto Mukherjee, MD, MPH(1) , Christopher R Cogle, MD(2), Tanya GK Bentley, PhD(3), Michael S Broder, MD, MSHS(3), Eunice Chang, PhD(3), Moira E. Lawrence, PhD(4), Thomas J. McKearn, MD, PhD(4), Scott Megaffin(4), Rita Percy(4), Michael E. Petrone, MD, MPH(4) and Gordon H Sun, MD, MS(3)

(1)Cleveland Clinic Foundation, Cleveland, OH;

(2)University of Florida, Gainesville, FL;

(3)Partnership for Health Analytic Research, LLC, Beverly Hills, CA;

(4)Onconova Therapeutics, Inc., Newtown, PA

Background: The myelodysplastic syndromes (MDS) include a group of hematopoietic malignancies that typically affect older individuals. Healthcare claims data have not been previously used to describe treatment patterns of MDS patients who progressed while on, failed to respond to, or became intolerant to hypomethylating agents (HMA; azacitidine [AZA] and decitabine [DEC]) and were eligible for 2nd-line therapy.

Methods: We conducted a retrospective cohort study using a US commercial health insurance claims database. The database contained adjudicated pharmacy and medical claims on 13 million annual lives submitted by providers, healthcare facilities, and pharmacies and included data on physician visits, medical procedures, hospitalizations, dispensed drugs, and performed tests. We identified patients with MDS (ICD-9-CM diagnosis code 238.7x) who were using an HMA in the identification (ID) period (1/1/2009—12/31/2011). The index date was the date on which the patient had used the same HMA for > 7 months, discontinued HMA treatment for > 2 months, or switched to another HMA. Patients were defined as eligible for 2nd-line therapy after reaching the index date and were followed for >6 months after they became candidates for 2nd-line treatment until the end of enrollment or study end (12/31/2012).

We examined treatment use patterns from 1/1/2008 to the index date. MDS-specific chemotherapy included AZA, DEC, and lenalidomide (LEN). We defined AZA or DEC cycles as continuous treatment periods without a gap of >7 days, and the gap between observed cycles could not be > 60 days. Hematologic supportive care agents included erythropoiesis-stimulating agents (ESA), growth factors (ie, granulocyte and granulocyte-macrophage colony- stimulating factors), and blood transfusions.

Results: Of 38,702 patients diagnosed with MDS in the ID period, we identified 1,366 who used an HMA. Three hundred eighty-six (96.0%) patients had a cytopenia potentially associated with HMA use, including anemia (92.8%), neutropenia (53.0%), and thrombocytopenia (52.7%). Of this group, 402 patients were eligible for 2nd-line MDS therapy: 283 (70.4%) had used AZA, 123 (30.6%) DEC, and 16 (4.0%) LEN. Patients averaged 6.1 HMA treatment cycles (6.4 for AZA,

5.4 for DEC) prior to being considered potential candidates for 2nd-line therapy. One hundred ninety-three (48.0%) patients had “early” discontinuation, defined as <5 treatment cycles before stopping or switching among HMAs. The proportions of patients who discontinued either AZA or DEC early were similar (47.1% vs. 50.0%). Among all 402 patients eligible for 2nd-line therapy, 320 (79.6%) received at least 1 hematologic supportive care agent: 208 (51.7%) received transfusions, 206 (51.2%) ESA, and 170 (42.3%) growth factors. Furthermore, among those eligible for 2nd-line therapy, 367 specifically switched to another HMA or did not have HMA treatment for 2 months. In this group, the rate of early discontinuation was 52.6%.

Conclusions: About half of patients completed < 5 HMA treatment cycles before discontinuation. AZA was used more often than DEC as 1st-line MDS therapy. These data suggest that a significant number of patients treated with AZA and DEC discontinue and become candidates for 2nd-line MDS therapy. The clinical reasons for early HMA discontinuation warrant further investigation; effective 2nd-line therapies for these patients are needed.