Exhibit 99.2

Avapritinib is Highly Active and Well-tolerated in Patients With Advanced GIST Driven by a Diverse Variety of Oncogenic Mutations in KIT and PDGFRA Michael Heinrich, Margaret von Mehren, Robin L. Jones, Sebastian Bauer, Yoon-Koo Kang, Patrick Schöffski, Ferry Eskens, César Serrano, Philippe A. Cassier, Olivier Mir, William D. Tap, Piotr Rutkowski, Jonathan Trent, Shreyaskumar Patel, Sant P. Chawla, Teresa Zhou, Tamieka Lauz, Oleg Schmidt-Kittler, Khalid K. Mamlouk, Beni B. Wolf, Suzanne George Connective Tissue Oncology Society 2018 Annual Meeting Rome, Italy • November 15, 2018 Abstract no: 3027631

Disclosures • Avapritinib is an investigational agent discovered and currently in development by Blueprint Medicines Corporation (Blueprint Medicines) • Data are preliminary and based on a cutoff date of October 15, 2018 • Dr. Michael Heinrich is an investigator for Blueprint Medicines’ ongoing Phase 1 study in unresectable gastrointestinal stromal tumors • Dr. Michael Heinrich has the following disclosures: – Consultant: Blueprint Medicines, Novartis, Molecular MD, Deciphera – Research funding: Blueprint Medicines, Deciphera – Stock or stock options: Molecular MD – Patents: 4 patents on diagnosis and treatment of PDGFR-mutant GIST, 1 patent on imatinib treatment of GIST 2

Avapritinib kinome selectivity Avapritinib: a highly selective and potent KIT/PDGFRA inhibitor for GIST 3 GIST mutation(s) Medical need by mutation Avapritinib biochemical IC50 1 KIT Exon 11 deletion JM domain 1L imatinib is effective 2L sunitinib/3L regorafenib have low ORR/short PFS 0.6 nM KIT Exon 11 V560G 1 nM KIT Exon 11/13 ATP binding site Approved 2L/3L agents have low ORR/short PFS 11 nM KIT Exon 11/14 28 nM KIT Exon 11/17 Activation loop 0.1 nM PDGFRα D842V No highly effective therapy in any line 0.24 nM Phase 3 trial of avapritinib vs. regorafenib in 3L and 4L GISTPhase 1 advanced GIST Ongoing clinical trials KIT, KIT proto-oncogene receptor tyrosine kinase; PDGFRA, platelet-derived growth factor alpha; IC50, concentration causing 50% inhibition; L, line; JM, juxtamembrane; ORR, objective response rate; PFS, progression-free survival. Kinome illustrations reproduced courtesy of Cell Signaling Technology, Inc. (CSTI) (www.cellsignal.com). Blueprint Medicines is not responsible for the content of the CSTI site. 1Evans E, et al. Sci Transl Med. 2017;9(414). pii: eaao1690.

NAVIGATOR Phase 1 study design 4 Advanced GIST (N = 46) Part 1 dose escalation Part 2 dose expansion RP2D* KEY OBJECTIVES • Determine MTD/RP2D, safety, PK and clinical activity by line of therapy and mutational status • ORR/DOR per central radiology assessment (mRECIST 1.1) for planned NDA and MAA regulatory filings avapritinib PO QD RP2D, recommended Phase 2 dose; PO, orally; QD, once daily; MTD, maximum tolerated dose; PK, pharmacokinetics; DOR, duration of response; mRECIST, modified Response Evaluation Criteria in Solid Tumors; NDA, New Drug Application; MAA, Marketing Authorization Application. *MTD 400 mg; RP2D 300 mg. PDGFR D842V (n = 33) registration enabling – fully enrolled ≥3L (n = 116) ≥4L registration enabling – fully enrolled 2L Ongoing (n ~50)

Demography and baseline characteristics 5 Parameter All patients (N = 231) Age (years), median (range) 62 (25, 90) GIST mutational subtype, %(n) KIT PDGFR D842V PDGFR non-D842V 72% (167) 24% (56) 4% (8) Metastatic disease, % (n) 89% (205) Largest target lesion size, %(n) ≤5 cm >5–≤10 cm >10 cm Pending 34% (79) 40% (93) 20% (47) 5% (12) No. prior kinase inhibitors, %(n) Median (range) 0 1 2 3 4 ≥5 PDGFR 1 (0-6) 17% (11) 37% (24) 19% (12) 11% (7) 8% (5) 8% (5) KIT 4 (1-11) 0 19% (31) 8% (14) 20% (34) 23% (38) 30% (50) Efficacy populations *Similar to Phase 3 trial population (VOYAGER). Data are preliminary and based on a cutoff date of October 15, 2018. PDGFR D842V 3L/4L regorafenib-naïve* 2L ≥4L

Adverse events ≥20% 6 Safety population; all doses (N = 231) AE, % (n) Any Grade Grade 1 Grade 2 Grade 3 Grade 4 Nausea 61% (142) 46% (106) 13% (30) 3% (6) 0 Fatigue 55% (127) 21% (48) 28% (64) 6% (15) 0 Anemia 46% (107) 5% (11) 15% (35) 25% (58) 1% (3) Periorbital edema 40% (93) 34% (79) 6% (13) <1% (1) 0 Diarrhea 39% (90) 22% (50) 13% (30) 4% (10) 0 Vomiting 38% (88) 30% (69) 6% (14) 2% (5) 0 Decreased appetite 35% (82) 23% (54) 9% (20) 3% (8) 0 Peripheral edema 33% (77) 23% (53) 10% (22) <1% (2) 0 Increased lacrimation 31% (72) 28% (64) 3% (8) 0 0 Memory impairment* 26% (60) 19% (45) 6% (15) 0 0 Constipation 23% (53) 14% (32) 8% (18) <1% (2) <1% (1) Face edema 23% (53) 19% (43) 4% (9) <1% (1) 0 Hair color changes 21% (49) 20% (46) <1% (2) <1% (1) 0 Dizziness 20% (47) 16% (38) 3% (8) <1% (1) 0 • Most AEs are grade 1 or 2 • No treatment-related grade 5 AEs • 8.7% (20) of patients discontinued due to related AEs • Grade 3-4 treatment-related AEs ≥2%: anemia, fatigue, hypophosphatemia, increased bilirubin, decreased white blood count/neutropenia, and diarrhea AE, adverse event. *The most commonly reported cognitive AE

Best response by central radiology in PDGFR D842V GIST 7 98% of patients with tumor reduction PD SD PR CR-100 -80 -60 -40 -20 0 20 Maximum reduction – sum of diameter change from baseline, % n = 56 patients across all dose levels PD, progressive disease; SD, stable disease; PR, partial response; CR, complete response.

ORR and DOR by central radiology in PDGFR D842V GIST 8 Best response* n = 56 mRECIST 1.1 %(n) [95% CI] ORR 84% (47) [71.7-92.47] CR/PR* 9% (5)/75% (42) SD 16% (9) CBR† 96% (54) [87.7-99.6] DOR 12-month DOR 76.3%; 12-month PFS 81.3% 0 3 6 9 12 15 18 21 24 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 All doses 47 36 25 19 14 8 5 3 1 Duration of response probability (%) Months from first dose Censored CI, confidence interval; CBR, clinical benefit rate. *4 PR pending confirmation. Patients who have had ≥1 post-baseline radiographic assessment. Response evaluable includes all doses. † PR + SD lasting ≥4 months.

Best response by central radiology in ≥4L GIST 9 PD SD PR CR KIT D842V Other PDGFRA 60% of patients with tumor reduction n = 109 patients 300/400 mg -100 -80 -60 -40 -20 0 20 40 60 80 100 Maximum reduction – sum of diameter change from baseline, %

ORR and DOR by central radiology ≥4L GIST 10 mDOR, median duration of response; NE, not estimatable *1 PR pending confirmation. Patients who have had ≥1 post-baseline radiographic assessment. Response evaluable includes 300 mg and 400 mg. †PR + SD lasting ≥4 months Best response* n = 109 mRECIST 1.1 %(n) [95% CI] ORR 20% (22) [13.1-29.0] CR/PR* 1% (1)/19% (21) SD 46% (50) CBR† 40% (44) [31.1-50.2] DOR 0 3 6 9 12 15 18 300-400 mg 22 18 6 3 2 1 0 Duration of response probability (%) Months from first dose Censored mDOR 7.3 months (95% CI: 7.2-NE) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Best response by central radiology in 3L/4L regorafenib-naïve GIST* 11 PD SD PR CR-100 -50 0 50 Maximum reduction – sum of diameter change from baseline, % n = 23 patients 300/400 mg 78% of patients with tumor reduction KIT Other PDGFRA *Similar to Phase 3 trial population (VOYAGER), except that PDGFR D842V patients (ORR 80%) are not included here. PDGFRα D842V patients, n=10, and ORR 80% are not included here

ORR and DOR by central radiology in 3L/4L regorafenib-naïve GIST 12 mDOR 10.2 months (95% CI: 4.2-NE) *All responses are confirmed. Patients who have had ≥1 post-baseline radiographic assessment. Response evaluable includes 300 mg and 400 mg. †PR + SD lasting ≥4 months Non-D842V patients best response* n = 23 mRECIST 1.1 %(n) [95% CI] ORR 26% (6) [10.0-48.4] CR/PR 0% (0)/26% (6) SD 57% (13) CBR† 70% (16) [47.1-86.8] DOR 0 3 6 9 12 300-400 mg 6 5 2 2 0 Duration of response probability (%) Months from first dose Censored 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Best response by mutational profile in ≥4L GIST 13 -100 -50 0 50 100 Best response (RECIST) PD SD CR PR V654A or T670I positive V654A and T670I negative n = 109 patients 300/400 mg; baseline genotype per ct-DNA and tumor sequencing Best response* n = 109 V654A or T670I POSITIVE, % (n) (n = 25) V654A and T670I NEGATIVE, % (n) (n = 84) ORR 0 26% (22)* CR/PR 0/0 1% (1)/ 25% (21)† SD 28% (7) 51% (43) CBR‡ 8% (2) 49% (41) PD 72% (18) 23% (19) ct-DNA, circulating tumor DNA. *Patients who have had ≥1 post-baseline radiographic assessment. Response evaluable includes 300 mg and 400 mg. † Includes 1 unconfirmed PR. ‡PR + SD lasting ≥4 months

Avapritinib has important clinical activity in advanced GIST 14 PDGFRα D842V n = 56 ≥4L all patients n = 109 3L/4L regorafenib- naïve non-D842V n = 23 2L non-D842V n = 20 ORR (central radiology), % (n) [95% CI] 84% (47) [72-92] 20% (22) [13.1-29.0] 26% (6) [10.2-48.4] 25% (5) [9-49] mDOR (central radiology), months [95% CI] NE [NE, NE] 7.3 [7.2-NE] 10.2 [4.2-NE] NR CBR (central radiology), % (n) [95% CI] 96% (54) [88-100] 40% (44) [31.1-50.2] 70% (16) [47.1-86.8] NR mPFS (central radiology), months [95% CI] NE [NE, NE] 3.7 [3.5-5.6] 8.6 [5.6-14.7] NR mPFS (investigator), months [95% CI] 22.8 [20.8-28.4] 5.5 [3.8-6.8] 10.2 [5.7-NE] NR Benchmarks PDGFRα D842V Approved agents: ORR ~0% mPFS ~3 mo mOS ~15 mo 4L imatinib re-treatment: ORR ~0% PFS 1.8 mo 3L regorafenib: ORR ~5% PFS 4.8 mo 2L sunitinib: ORR ~7% PFS 6 mo NR, not reported; mPFS, median progression-free survival; mOS, median overall survival. ORR is not an endpoint for 2L but is early signal readout.

• Phase 1 NAVIGATOR study demonstrates favorable tolerability and encouraging clinical activity across lines of therapy – Most AEs were grade 1 or 2, with manageable on-target toxicity – Important efficacy in PDGFRα D842V GIST and refractory, ≥4L GIST supports regulatory filing – Encouraging activity in 3L/4L regorafenib-naïve GIST indicates the potential for a favorable outcome in the ongoing randomized Phase 3 VOYAGER study – Mutational profiling analyses and promising 2L data provide strong rationale for genotype-selected 2L study 15 Avapritinib has the potential to change GIST treatment paradigms

Acknowledgments 16 We would like to thank the participating patients, their families, all study co-investigators, and research coordinators at the following institutions: • OHSU Knight Cancer Institute • Fox Chase Cancer Center • Royal Marsden Hospital/Institute of Cancer Research • University of Duisburg-Essen • Asan Medical Centre • University Hospitals Leuven • Erasmus MC Cancer Institute • Vall d’ Hebron Institute of Oncology • Centre Leon Berard • Institut Gustave Roussy • Memorial Sloan Kettering Cancer Center • Maria Sklodowska-Curie Institute – Oncology Center • University of Miami Sylvester Comprehensive Cancer Center • MD Anderson Cancer Center • Sarcoma Oncology Centre • Dana-Farber Cancer Institute Editorial and medical writing support were provided by Lauren Fink, PhD, of Cello Health Communications, and were funded by Blueprint Medicines.