Exhibit 99.3

Safety and preliminary efficacy of vimseltinib in tenosynovial giant cell tumor (TGCT) Gelderblom H1, Abdul Razak AR2, Sánchez-Gastaldo A3, Rutkowski P4, Wilky BA5, Wagner AJ6, van de Sande M7, Michenzie M8, Vallee M8, Sharma M8, Sherman ML8, Ruiz-Soto R8, Tap WD9 1Medical Oncology, Leiden University Medical Center, Leiden, Netherlands; 2Toronto Sarcoma Program, Princess Margaret Cancer Center, Toronto, ON, Canada; 3Phase I Unit, Medical Oncology, Hospital Universitario Virgen del Rocío, Seville, Spain; 4Maria Sklodowska-Curie National Research Institute of Oncology, Abstract: Warsaw, Poland; 5Medicine, University of Colorado Cancer Center, Aurora, CO, United States; 6Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, United States; 7Department of Orthopedics, Leiden University Medical Center, Leiden, Netherlands; 8Deciphera Pharmaceuticals, LLC, Waltham, MA, United States; 1821P 9Medical Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, United States ePoster INTRODUCTION • Tenosynovial giant cell tumor (TGCT) is a rare, locally aggressive neoplasm, where overexpression of colony-stimulating factor 1 (CSF1) drives recruitment of macrophages leading to local inflammation and joint destruction1,2 • Patients with TGCT experience debilitating symptoms and significant disease burden. There remains an unmet need for treatment options for patients with TGCT not amenable to surgery • Vimseltinib (DCC-3014) is an investigational, oral, highly selective, switch-control kinase inhibitor of CSF1 receptor (CSF1R)3 • We report the safety and preliminary efficacy of patients with TGCT not amenable to surgery receiving vimseltinib in the Phase 1/2 study (NCT03069469) METHODS • NCT03069469, an ongoing, multicenter, open-label study of vimseltinib in patients with advanced solid tumors and TGCT consists of 2 phases: - Phase 1 (dose escalation) study, a pharmacologically guided 3 + 3 design, to determine the recommended phase 2 dose (RP2D) and the maximum tolerated dose (MTD) - Phase 2 (expansion) study to evaluate the safety, tolerability, and preliminary efficacy in 2 TGCT expansion cohorts • Cohort A: TGCT patients with no prior anti-CSF1/CSF1R therapy (previous therapy with imatinib or nilotinib is allowed) • Cohort B: TGCT patients with prior anti-CSF1/CSF1R therapy (previous therapy with imatinib or nilotinib alone would not be eligible) RESULTS Figure 1. TGCT enrollment and disposition in Phase 1/2 Study RP2D Phase 2 TGCT Phase 1 TGCT enrollment (30 mg twice weekly) enrollment (n = 32) (n = 45) • Cohort 5, n = 8 • Cohort A, n = 37 • Cohort 8, n = 12 • Cohort 9, n = 12 • Cohort B, n = 8 Pending entry of dosing data in EDC (n = 1) Phase 1 TGCT Cohort A n = 32 n = 36 (Safety population) (Safety population) Patient withdrew prior to assessment (n = 1) Phase 1 TGCT Pending first imaging assessment n = 32 (n = 16) (Efficacy-evaluable populationa) Cohort A n = 19 Loading dose Dose (Efficacy-evaluable populationa) Cohort 5 30 mg QD x 5 days 30 mg twice weekly Cohort 8 30 mg QD x 3 days 10 mg QD Data cutoff Cohort 9 20 mg QD x 3 days 6 mg QD 7 June 2021 Phase 2 NA 30 mg twice weekly ClinicalTrials.gov: NCT03069469 a At least one post-baseline efficacy assessment. EDC, electronic data capture; QD, once daily; RP2D, recommended phase 2 dose; TGCT, tenosynovial giant cell tumor. • Enrollment in phase 1 with dose escalation is complete (n = 32) • The target enrollment of 40 patients in phase 2 cohort A has been reached as of 13 July 2021 • The results of patients with TGCT in phase 1 (n = 32) and phase 2 Cohort A (n = 36) receiving vimseltinib as of 7 June 2021 are presented Table 1. Baseline characteristics of patients with TGCT receiving vimseltinib Phase 1 TGCT patients Phase 2 Cohort A patients n = 32 n = 36 Age, median (range), years 51 (23–73) 44 (21–71) Sex Female 17 (53) 26 (72) Male 15 (47) 10 (28) Race White 31 (97) 28 (78) Asian 1 (3) 2 (6) Not Reported or Missing 0 6 (17) Disease location Knee 20 (63) 20 (56) Ankle 5 (16) 5 (14) Hip 4 (13) 2 (6) Foot 1 (3) 6 (17) Othera 2 (6) 3 (8) Patients with at least one prior surgery 12 (38) 32 (89) Patients with at least one prior systemic therapy 5 (16) 2 (6) Imatinib or nilotinib 4 (13) 2 (6) Lacnotuzumab (MCS-110) 1 (3) 0 Data are presented as n (%) unless otherwise noted. Percentages might not add up to 100% due to rounding. a Other locations include wrist, shoulder, and jaw. TGCT, tenosynovial giant cell tumor. Table 2. TEAEs in ³15% of patients with TGCT receiving vimseltinib Phase 1 Phase 2 Cohort 5 All Patientsa Cohort Aa Preferred term, No. (%) (n = 8) (n = 32) (n = 36) All grades Grade 3/4 All grades Grade 3/4 All grades Grade 3/4 Blood CPK increased 7 (88) 4 (50) 20 (63) 10 (31) 19 (53) 9 (25) Periorbital oedema 3 (38) 0 17 (53) 0 8 (22) 0 Fatigue 3 (38) 0 15 (47) 0 6 (17) 0 AST increased 5 (63) 1 (13) 14 (44) 4 (13) 12 (33) 0 ALT increased 2 (25) 0 10 (31) 1 (3) 4 (11) 0 Myalgia 0 0 9 (28) 1 (3) 5 (14) 0 Arthralgia 2 (25) 0 8 (25) 1 (3) 2 (6) 0 Face oedema 0 0 8 (25) 0 0 0 Headache 3 (38) 0 8 (25) 0 10 (28) 0 Lipase increased 1 (13) 0 8 (25) 3 (9) 4 (11) 0 Oedema peripheral 1 (13) 0 8 (25) 0 5 (14) 0 Pruritus 1 (13) 0 8 (25) 0 3 (8) 0 Amylase increased 1 (13) 1 (13) 7 (22) 2 (6) 5 (14) 0 Diarrhoea 1 (13) 1 (13) 6 (19) 1 (3) 2 (6) 0 Generalised oedema 2 (25) 0 6 (19) 0 0 0 Hypertension 0 0 6 (19) 2 (6) 1 (3) 0 Nausea 2 (25) 0 6 (19) 0 8 (22) 0 Constipation 1 (13) 0 5 (16) 0 2 (6) 0 Parasthesia 0 0 5 (16) 0 1 (3) 0 Rash macular 0 0 5 (16) 0 0 0 Rash maculopapular 0 0 5 (16) 0 5 (14) 0 Asthenia 1 (13) 0 3 (9) 0 6 (17) 0 a TEAEs cutoff of >15% based on all grades for total phase 1 and phase 2 cohort A. ALT, alanine aminotransferase; AST, aspartate aminotransferase; CPK, creatine phosphokinase; TEAE, treatment-emergent adverse event. Table 3. Dose modifications due to any TEAEs Phase 1 Phase 2 Cohort 5 All patients Cohort A (n = 8) (n = 32) (n = 36) Any TEAEs leading to dose 5 (63) 19 (59) 10 (28) modification, n (%) Dose interruption 5 (63) 18 (56) 9 (25) Dose reduction 4 (50) 13 (41)a 3 (8)b Treatment discontinuation 1 (13) 2 (6)c 1 (3)d a Cohort 5: Gr3 urticaria (n = 1), Gr3 diarrheoa (n = 1), Gr2 AST increase (n = 1), Gr3 amylase, Gr3 CPK, and Gr3 LDH increased (n = 1); Cohort 8: Gr2 fatigue, Gr2 edema peripheral, and Gr2 rash maculopapular (n = 1), Gr2 rash macular (n = 1), Gr2 joint swelling and Gr1 pyrexia (n = 1), Gr3 CPK increased (n = 1), Gr3 CPK increased and Gr2 myalgia (n = 1), Gr3 pruritic rash (n = 1); Cohort 9: Gr1 generalized oedema and Gr1 periorbital oedema (n = 1), Gr1 rash maculopapular (n = 1), Gr3 AST increased (n = 1). b Cohort A: Gr1 periorbital oedema and Gr1 rash maculopapular (n = 1); Gr1 headache (n = 1), Gr2 headache, Gr2 nausea, and Gr2 vomiting (n = 1). c Cohort 5: Gr3 metabolic encephalopathy (n = 1); Cohort 8: Gr3 AST increase (DLT, n = 1); d Gr1 periorbital oedema and Gr1 rash maculopapular (n = 1). AST, aspartate aminotransferase; CPK, creatine phosphokinase; DLT, dose-limiting toxicity; Gr, Grade; LDH, lactate dehydrogenase; TEAE, treatment-emergent adverse event. Table 4. Best overall response in patients with TGCT Cohort 5 All patients Cohort A (n = 8b) (n = 32b) (n = 19) Best overall response, n (%) Complete response 1 (13) 1 (3) 0 Partial response 3 (38) 15 (47) 8 (42) Stable disease 4 (50) 16 (50) 11 (58) ORR, % 4 (50) 16 (50) 8 (42) a Of the 51 efficacy-evaluable patients in phase 1 across all dose cohorts and phase 2 cohort A, 24 patients had a response resulting in an ORR of 47%. b 1 patient had a local assessment for efficacy, but no central assessment was performed. ORR, objective response rate; QD, once daily. SAFETY Majority of the common (³15%) TEAEs were £Grade 2 (Table 2) • Observed transaminase, pancreatic, and CPK enzyme elevations were mostly low grade and not associated with symptoms; are consistent with the mechanism of action of CSF1R inhibitors • No abnormalities in bilirubin levels reported • In phase 1, 2 patients had TEAEs leading to treatment discontinuation (Table 3) and 2 patients had treatment-related grade 3 serious AEs (SAE): metabolic encephalopathy (possibly related) and vaginal hemorrhage (probably related) • In phase 2 Cohort A, 1 patient had a TEAE leading to treatment discontinuation (Table 3) and no treatment-related SAEs were reported EFFICACY • Phase 1: ORR of 50%; responses observed across all dose cohorts (Table 4, Figure 2, Figure 3) • Phase 2 Cohort A: ORR of 42% (all partial responses; Table 4, Figure 2, Figure 3) Figure 2. Duration of treatment and response in patients with TGCT receiving vimseltinib Phase 1a Phase 2 – Cohort Ab 5 Cohort # 8 Cohort SD PR 9 NE SD Cohort CR PR Ongoing Ongoing # From Local Data 0 2 4 6 8 10 12 14 16 18 20 22 24 26 0 1 2 3 4 5 6 7 8 9 10 11 12 Months on Study Months on Study # 1 patient had a local assessment for efficacy, but no central assessment was performed. a Median duration of treatment of 10.1 months across all phase 1 dose cohorts. b Median duration of treatment of 1.9 months in phase 2 cohort A. CR, complete response; NE, not evaluable; PR, partial response; SD, stable disease; TGCT, tenosynovial giant cell tumor. Figure 3. Best percentage change in target lesions in patients with TGCT receiving vimseltinib Phase 1 Phase 2 – Cohort A 100 Cohort 5 100 Expansion A from Cohort 8 80 80 Cohort 9 Lesions 60 60 Target Baseline 40 40 - in Post 20 20 of SD 0 0 SD SD SD SD SD SD Diameters Nadir SD SD SD SD SD SD of SD SD the -20 SD SD SD SD -20 SD SD to Local SD SD Sum SD SD SD SD PR PR in -40 -40 PR PR PR PR PR PR PR PR PR PR Baseline PR PR Change -60 PR PR -60 PR PR PR PR PR -80 -80 Percent PR PR -100 -100 CR CR, complete response; NE, not evaluable; ORR, objective response rate; PR, partial response; SD, stable disease; TGCT, tenos ynovial giant cell tumor. CONCLUSIONS • In patients with TGCT not amenable to surgical resection, - Vimseltinib was well tolerated in both phase 1 and phase 2 Cohort A. The safety profile remains manageable with longer-term follow-up across all phase 1 dose cohorts - Vimseltinib demonstrated encouraging preliminary efficacy • Of the 32 patients in phase 1, ORR of 50% with durable responses observed across all dose cohorts, including 1 complete response in Cohort 5 • Of the 36 patients enrolled in phase 2 Cohort A, 19 patients were evaluable for efficacy and had an ORR of 42%. Of the 19 patients, 10 had >1 follow-up imaging assessment and 2 responses occurred at later scans. The study is ongoing and follow-up evaluation is continuing • These results support further evaluation of vimseltinib in the MOTION study, a randomized, placebo-controlled, phase 3 trial in patients with TGCT not amenable to surgical resection To be presented at the 2021 ESMO Congress, Acknowledgments Corresponding Author References The study was sponsored by Deciphera Pharmaceuticals, LLC, Waltham, MA, USA Dr. H. Gelderblom (a.j.gelderblom@lumc.nl) 1) Staals EL, et al. Eur J Cancer. 2016;63:34-40; 2) West RB, et al. Proc Natl Acad Sci USA. 2006;103:690-695; 3) Smith BD, et al. Mol Cancer 16-21 Sep 2021 (Virtual) No financial disclosures Medical writing was provided by Uma Chandrasekaran, PhD (Deciphera Pharmaceuticals, LLC); editorial support was provided by Ashfield Ther. 2021 Aug 25. doi:10.1158/1535-7163.MCT-21-0361. Health and was funded by Deciphera Pharmaceuticals, LLC