Exhibit 99.2

Early Safety and Efficacy Results of P-BCMA-ALLO1, A Fully Allogeneic Chimeric Antigen Receptor T-cell (CAR-T), in Patients with Relapsed / Refractory Multiple Myeloma (RRMM) Bhagirathbhai Dholaria, MBBS1, Mehmet H. Kocoglu, MD2, Andrew Kin, MD3, Adam Asch, MD4, Aravind Ramakrishnan, MD5, Carlos Bachier, MD6, Tulio Rodriguez, MD7, Leyla Shune, MD8, Katherine McArthur9, Joanne McCaigue, MS9, Samuel DePrimo, PhD9, Christopher Martin, PhD9, Sabrina Haag, PhD9, Jeff Eskew, PhD9, Hamid Namini, PhD9, Ellen Christie9, Rajesh Belani, MD9, Stacey A. Cranert, PhD9, Julia Coronella, PhD9, Devon J. Shedlock, PhD9, and Caitlin Costello, MD10 1Vanderbilt University Medical Center, Nashville, TN; 2University of Maryland Marlene and Stewart Greenebaum Comprehensive Cancer Center, Baltimore, MD; 3Karmanos Cancer Institute, Wayne State University, Detroit, MI; 4University of Oklahoma Health Sciences Center, Oklahoma City, OK; 5Sarah Cannon Transplant and Cellular Therapy Program at St. David’s South Austin Medical Center, Austin, TX; 6Sarah Cannon Transplant and Cellular Therapy Program at Methodist Hospital, San Antonio, TX; 7City of Hope Chicago, Chicago, IL; 8Division of Hematological Malignancies and Cellular Therapeutics, University of Kansas Medical Center, Westwood, KS; 9Poseida Therapeutics, San Diego, CA; 10UC San Diego Health, Moores Cancer Center, La Jolla, CA Study P-BCMA-ALLO1-001: open-label, multicenter, phase 1 study Higher cyclophosphamide LD doses markedly enhanced T -rich P-BCMA-ALLO1 cells traffic to bone marrow and BACKGROUND SCM to assess the safety of P-BCMA-ALLO1 in patients with RRMM P-BCMA-ALLO1 expansion differentiate to T as observed on day 44 marrow aspirate EFF Despite therapeutic advances, such as autologous (auto) CAR-T, MM remains an incurable malignancy. A Auto CAR-T pose significant challenges as each patient must undergo apheresis and each CAR-T dose must P-BCMA-ALLO1 Treatment: S: 2x106 cells/kg P1: 2x106 cells/kg P2: 2x106 cells/kg A B Key Inclusion Criteria: infusion on day 0 Follow up, potential for redosing CD8 • LD (Cy mg/m2): 300 500 1,000 CD4 be manufactured separately. This causes long wait times for the treatments resulting in disease RRMM as defined by the IMWG CAR-T cells CAR-T cells Conditioning • Must have received PI, IMiDs & CD38 mAb or 106 106 106 progression, need for bridging therapy and sometimes even death. Auto CAR-T are also hampered by SCREENING FOLLOW UP 100 100 Chemotherapy T triple refractory EFF occasional manufacturing failures with some patients not receiving the therapy or receiving an inadequate Long-term 105 105 105 80 80 T 2 • ECOG 0 or 1 A A cells cells EM Cyclophosphamide (300, 500, or 1,000 mg/m ) and NA CD27 T follow-up T—CAR-T dose. Broad patient access to auto CAR-T remains challenging because of inability to manufacture at Fludarabine (30 mg/m2) on days -5, -4, -3 D DN DN CD45RA—T R 60 60 CM g 104 g 104 g 104 scale and high costs associated with production. Auto CAR-T are also derived from patient T cells that are P-BCMA-ALLO1 doses and LD arms evaluated, mmm CA CAR T SCM PRIMARY ENDPOINTS and no. of patients† infused at each cohort: i es/ es/ i es/ i 40 40 impaired due to exposure to immunosuppressive anti-myeloma therapy. In contrast, allogeneic (allo) CAR- 103 103 103 CD4 CD8 • Assess safety and MTD based on DLT 6 A. 0.25 x 10 cells/kg (n=1) Cop Cop Cop f o of 20 20 T that are mass produced from healthy donor cells and are available “off-the-shelf’ may be able to B. 0.75 x 106 (n=7) 102 102 102 % % SECONDARY OBJECTIVES Arm S (Cy 300); N = 21 overcome the auto CAR-T challenges described above. C. 2 x 106 (n=9‡) 101 101 101 0 0 • Evaluate the anti-myeloma effect of P-BCMA-ALLO1 6 0 7 14 21 28 35 42 49 56 63 0 7 14 21 28 35 42 49 56 63 0 7 14 21 28 35 42 49 56 63 P 6 P 6 • Study effect of cell dose & LD regimen selection to guide D. 6 x 10 (n=4) F W K F W K P-BCMA-ALLO1, a BCMA-targeting allo CAR-T for RRMM, is manufactured from healthy donor T cells with a Study Day dose selection for pivotal studies E. 2 x 106 Arm P1 (Cy 500); N = 5 Study Day Study Day proprietary non-viral transposon-based integration (piggyBac® DNA Delivery System) system to express a F. 2 x 106 Arm P2 (Cy 1,000); N = 6 Figure 4 – A) CAR staining and CAR-T phenotype of bone human anti-BCMA V -based CAR, Cas-CLOVER™ gene editing and a transiently expressed Booster B marrow mononuclear cells (BMMC) for patient #23 (Fig. 2A, H †minimum of 4 weeks follow-up; ‡Includes one patient non-evaluable for C T AUCD1-WK3 max max Figure 1 – A) CK in Arm S, P1 and P2 at 2x106/kg cell Molecule, to produce a T stem cell memory (T )-rich product. Gene editing in resting T cells eliminates response assessment purple) at Week 6 (Day 44) post-infusion. Top row CD4+ SCM P = 0.0096 7 P = 0.0159 106 25 10 dose and increasing Cy dose levels. Missing data CAR-T cells, bottom row CD8+ CAR-T cells. B) Final drug endogenous T cell receptor expression to prevent graft-vs-host disease (GvHD), along with beta-2 106 points are due to insufficient DNA yield. B) C/Tmax product (FP) phenotype for patient #23 in comparison to microglobulin to decrease host-vs-graft response against MHC class I while maintaining a desirable T Rapid and convenient CAR-T administration for entire intent to 105 20 A SCM and AUC by cohort shown in A. Only n=7 (S) n=4 y 5 CAR-T phenotype by % TSCM, TCM, TEM and TEFF of CD4 and phenotype. T -based CAR-T cells can be considered a prodrug, as they must expand and differentiate in DNA DN 10 SCM 15 treat (ITT) population without need for apheresis g 104 Da g (P1) shown for Tmax, since no signal >LOD for one CD8 CAR-T cells (as shown in quadrant gates in A) in bone vivo to yield effector CAR-T, or drug. While T cells correlate with clinical benefit in auto CAR-T m m 104 SCM s/ es/ subject per group. Kruskal-Wallis test with Dunn’s marrow specimen at Week 6. Gating strategy of BCMA- 10 (NCT03288493), this has yet to be demonstrated for allo CAR-T. Questions remain about optimal • 100% of ITT population underwent LD and received P-BCMA-ALLO1 103 Study 103 51 patients screened (1 patient had not begun LD by data cut) Copie Copi multiple comparisons test; Median with range is CAR CAR-T cells and CAR-T sub-populations described in Fig. 3. lymphodepletion (LD) needed for allo CAR-T expansion, engraftment, differentiation, and persistence. 5 shown. LOD of qPCR is 100 cp/µg. CK = cellular 102 102 • No patient required bridging therapy 101 101 kinetics; PBMC = peripheral blood mononuclear Here, we present early results from a phase 1 study (NCT04960579) of P-BCMA-ALLO1 demonstrating 0 • Median time from enrollment to: S P1 P2 S P1 P2 S P1 P2 cells; cp/mg = transposon copies/mg of DNA. compelling safety and auto CAR-T-like response rates in heavily pretreated RRMM patients. P-BCMA- Deep responses and a high response rate in BCMA naïve and 39 patients enrolled (ITT) – Start of LD was 1 (range 0-10) day ALLO1 can be rapidly administered to patients without the need for apheresis or long manufacturing times and shows activity in patients who have received prior auto CAR-T. As hypothesized, T -rich P-BCMA- – P-BCMA-ALLO1 infusion was 7 (range 5-15) days prior BCMA therapy exposed RRMM patients SCM P-BCMA-ALLO1 showed durable persistence in peripheral ALLO1 cells traffic to the bone marrow, expand, differentiate to effector phenotypes and demonstrate 38 patients lymphodepleted † Total includes: one patient enrolled in cyclic admin arm (cohort 2); one patient retreated at cohort 2 dose; and dosed P-BCMA-ALLO1 † one non-evaluable patient and 4 patients with 4 weeks follow-up. Interim safety analysis on patients (n = 34) durable persistence. Our data highlight the impact of optimized cyclophosphamide dosing on in vivo given an infusion of P-BCMA-ALLO1 (including retreated) and with a minimum of 4 weeks follow-up. Efficacy blood and bone marrow cellular kinetics and clinical activity of P-BCMA-ALLO1 in RRMM. analysis performed on dose-level 2 × 106 cells/kg on response evaluable patients (n = 19). ITT = intent-to-treat defined at enrollment. Data cutoff for safety and efficacy analysis was Oct. 23rd, 2023. 3 A B sCR Proprietary, non-viral approach to produce T -rich, fully Patient #23 & #30 Patient #30** (all PBMC) Patient #23** 3 SCM CR Study enrolled a heavily pretreated patient population D10 WK2 WK3 WK4 WK6 (BMA) WK6 (PBMC) 600000 Patient #23 GPR allogeneic P-BCMA-ALLO1 CAR-T from healthy donors 350000 qPCR (cp/µg) 345,562 114,115 37,549 19,566 ~69k* 932 Patient #30 † 100000 PR Characteristic Number of Patients (total n = 33 ) DNA 100000 % CAR-T of all g 80000 Lymphocytes 31.1 n/a 5.94 4.6 13.8 n/a Median age (min, max) years 68 (33, 85)m BMA #23* piggyBac® Gene Insertion 5 60000 *qPCR was performed on DNA derived from BMA. The standard curve efficiency was Female n (%) / Male n (%) 21 (64) / 12 (36) Copies/ 40000 86%, which is below our acceptance criteria ( 90%); consequently, the result was P-BCMA-ALLO1 Transposon Design expressed as an approximation. ‘n/a’: not assayed due to insufficient PBMC counts for 20000 Median time (min, max) since diagnosis, years 6.27 (1.48, 18.95) FACS and qPCR assays. 3 Insulator Insulator • Preferentially 0 **MM response achieved: sCR (MRD-) for patient #23 and VGPR (patient #30) ITR Promoter Poly(A) ITR transposes naïve T IgG, 20 (61), IgA, 9 (27), IgD, 1 (3) 0 7 14 21 28 35 42 49 56 63 TTAA TTAA cells resulting in T PiggyBac transposon SCM Myeloma Diagnosis Subtype, n (%) Kappa FLC, 21 (64), Lambda FLC, 12 (36) Study Day rich product system inserts Safety Switch CAR Molecule Selection Gene Other: Abnormal CBC, 1 (3) Figure 2 – A) CK in patient #23 and #30 in peripheral blood (PBMC) and bone marrow aspirate (BMA) for subject #23. B) transgene encoding • Large cargo capacity piggyBac piggyBac delivers CAR, inducible High-Risk cytogenetics , n (%) 10 (30) qPCR signal compared to frequency of CAR-T cells identified by Flow (expressed as % of total CD45+ live single cells). BCMA-directed CAR DNA Transposon RNA Transposase ITR CARGO ITR safety switch, and Flow performed on PBMC and BMMC for bone marrow sample for patient #23. Missing data points due to insufficient 3 5 selectable marker in Median prior lines of therapy (min, max), n 7 (2, 17) 5 DNA yield. BM = bone marrow. BMA = bone marrow aspirate. BMMC = bone marrow mononuclear cells. single step Prior anti-BCMA CAR-T and anti-BCMA therapy, n (%) 13 (39) “Cut and Paste” Genomic DNA †Includes one patient non-evaluable for response assessment ITR CARGO ITR Peripheral blood samples of patient demonstrate P-BCMA-ALLO1 Figure 5—Data cutoff: October 23rd, 2023. ORR Objective Response Rate, attaining sCR, CR, VGPR or PR, including T cells differentiation to T phenotype confirmed and unconfirmed responses. Evaluable patients: Obtained first response assessment by IMWG m-protein P-BCMA-ALLO1 is well tolerated in RRMM patients SCM EFF criteria or PD/death. Fisher exact test used to calculate P-values for arm S compared to arms P1 and P2. Cas-CLOVER™ Gene Editing A †Two P2 patients who had received prior BCMA auto CAR-T (idecabtagene vicleucel and P-BCMA-101), both achieved VGPR. A small subset (n=5) of patients were tested for MRD status, with two patients being MRD—(10-5). One arm S LESS MORE patient who received prior BCMA auto CAR-T achieved SD. The two nonresponding patients in arm P1 and P2 had DIFFERENTIATED DIFFERENTIATED Left gRNA received teclistamab as previous therapy; excluding these two patients the ORR for these arms is 100%. One arm P1 Clo051 • High editing efficiency in patient with t(4:14) achieved MRD – VGPR. High fidelity gene editing resting T cells results in SCM CM EM EFF system used to address 5’ 3’ high % of T Stem cell Central memory T cell Effector memory T cell Effector T cell SCM ORR for study cohorts not shown in bar graph: arm S cohort -1 (0%; 0/1 pt); arm S cohort 1 (43%; 3/7 pts); arm S cohort graft vs. host and host 3’ 5’ memory T cell • Low to no off target • Non self renewing 3 (0%; 0/4 pts); arm C cohort 2 (0%; 0/1 pt); retreated patient cohort 2 (100%; 1/1 pt) vs. graft alloreactivity • Self renewing cutting • Short lived • Long lived • Specialized—Cytotoxic Right gRNA • Edits include TCR and • Multipotent CONCLUSIONS B2M (MHC I) knockouts dCas9 dCas9 C • P-BCMA-ALLO1 is a promising “off-the-shelf” TSCM-rich allogeneic CAR-T therapy manufactured using non-viral gene Madison et al., Molecular Therapy – Nucleic Acids, 2022. (https://www.sciencedirect.com/science/article/pii/S216225312200155X) B Day 10 Week 3 Week 4 editing technologies that these Phase 1 results show is: – Delivered to patients from bulk inventory to enable rapid treatment of 100% of the ITT population with drug product meeting all quality specifications CD27 CD4 CAR-T cells CD8 CAR-T cells CD45RA 100 100 – Well tolerated, with no GvHD and low rates of CRS, neurotoxicity Gr 2 High-yield Clinical Manufacturing TEFF cells 80 cells 80 T • P-BCMA-ALLO1 demonstrates activity comparable to autologous BCMA CAR-T in heavily pretreated RRMM pts EM T T—60—TCM – 82% overall ORR including both BCMA treatment-naïve and BCMA treatment-experienced patients in pooled CAR CAR 60 TSCM P1/P2 cohorts (total n=11 pts) receiving adequate lymphodepletion Manufacturing 40 40 CD4 CD8 – 100% ORR in all P1/P2 patients receiving any treatment, including auto BCMA-directed CAR-T, other than T Cell Isolation Safety Summary: S=Cy 300 of of • Production process 20 20 BCMA-directed bispecific T cell-engaging therapy teclistamab (n=9) Allogeneic Non-viral Gene Editing P1=Cy 500 % % preserves T • Dose-levels through 6 × 106 cells/kg cleared with no DLTs – 100% ORR in pooled P1/P2 BCMA treatment-naïve patients (n=7), with sCR rate 40% (2/5 pts) in highest LD SCM 0 0 phenotype P2=Cy 1000 CAR (P2) cohort P 0 3 4 P 0 3 4 • No GvHD observed at any dose F 1 K K F 1 K K CD3 CD45RA D W W D W W • Nearly all CAR-carrying • – Deep clinical response was observed in a patient with high-risk disease Grade 3 TEAEs were associated mainly with LD and myeloma cells Figure 3 – A) TSCM differentiation pathway. B) Phenotype of CAR-T cells (% CD3-BCMA+ of live, single cells, CD45+CD2+, • Consistent with the hypothesized mechanism of a T -rich product, prodrug-like P-BCMA-ALLO1 cells demonstrate ON • • Low CRS incidence (21%) , Grade 2 in severity SCM DOSES “On demand” Delivery CD4+ or CD8+; CAR-T identified with BCMA protein) in peripheral blood by Flow over time (Day 10, Week 3 and Week 4) engraftment, expansion, differentiation to T phenotype and persistence, as observed in a day 44 bone marrow Arm S 14% (n=3), arm P1 20% (n=1), arm P2 50% (n=3) EFF in patient #30 based on CD45RA and CD27 expression (Fig. 2A, green). Representative CAR staining shown for D10; aspirate specimen • Neurotoxicity (Grade 2) observed in 2 patients endogenous T cells are defined as BCMA-CD3+. Top row CD4 CAR-T cells, bottom row CD8 CAR-T cells. C) Final drug • Optimized LD with cyclophosphamide is essential to promote robust cell expansion (Cmax) and unlock the activity One patient in arm S and one patient in arm P2 developed neurotoxicity product (FP) phenotype for patient #30 in comparison to CAR-T phenotype by % TSCM, TCM, TEM and TEFF of CD4 and CD8 potential of the allogeneic CAR-T • Serious Infections were uncommon even in higher LD arms CAR-T cells (as shown in quadrant gates in B) in PBMC samples at indicated timepoints. • Further clinical development of P-BCMA-ALLO1 is ongoing vu ACKNOWLEDGMENTS: The authors and Poseida Therapeutics, Inc. thank the patients, caregivers, investigators, and study site staff for their involvement in this Auto = Autologous; AUC = Area under the curve; BCMA = B-cell maturation antigen; B2M = 2 microglobulin; BMA = Bone marrow aspirate; CK = Cellular kinetics; CRS = Cytokine release syndrome; Cy = Cyclophosphamide; DLT = Dose limiting toxicity; ECOG = Eastern Cooperative Oncology Group; FLC = Free light chains; Flu = Fludarabine; GvHD = Graft-versus-host disease; IMiD = study. This study was sponsored and funded by Poseida Therapeutics (San Diego, CA). MWG = International Myeloma Working Group; ITR = inverted terminal repeats; ITT = Intention-to-treat; LD = Lymphodepletion; LOD = Limit of detection; MHC = Major histocompatibility complex; MM = Multiple myeloma; MTD = Maximum tolerated dose; PBMC = Peripheral blood mononuclear cells; PR = Partial response; RRMM = Relapsed and Refractory Multiple Myeloma; sCR = = Central memory T cells; TEAE = Treatment Emergent Adverse Event; Teff = Effector T cells; TEM = Effector memory T cells; TSCM = Stem cell memory T cells; VGPR = Very Good Partial Response PLEASE SEE OUR OTHER POSEIDA PRESENTATIONS—A Tumor-Bearing Murine Xenograft Model as a Bioassay for Assessing CAR-T Product Potency Shows Positive Predictive Value for Clinical Performance; Publication Number: 2293—Effective Gene Therapy for Hemophilia A: Novel Re-Dosable Non-Viral Formulation That Provides Stable, and Durable FVIII Expression with Improved Tolerability; Publication Number: 1232