Corporate Presentation June 2026 Exhibit 99.2

Disclaimer This presentation, including any printed or electronic copy of these slides, the talks given by the presenters, the information communicated during any delivery of the presentation and any question and answer session and any document distributed at or in connection with the presentation (collectively, the “Presentation”) has been prepared by Cabaletta Bio, Inc. (“we,” “us,” “our,” “Cabaletta” or the “Company”) and may contain “forward-looking statements” within the meaning of the Private Securities Litigation Reform Act of 1995 relating to our business, operations, and financial condition, and include, but are not limited to, express or implied statements regarding our current beliefs, expectations and assumptions regarding: our business, future plans and strategies for our technology; our ability to grow our autoimmune-focused pipeline; the ability to capitalize on and potential benefits resulting from our research and translational insights, including those related to any similarly-designed constructs or dosing regimens; the Company's business plans and objectives; our expectations around the potential success and therapeutic and clinical benefits of rese-cel, as well as our ability to successfully complete research and further development and commercialization of our drug candidates in current or future indications, including the timing and results of our clinical trials and our ability to conduct and complete preclinical and clinical trials; expectation that clinical results will support rese-cel's safety and activity profile; our plan to leverage increasing clinical data and a unique development program for rese-cel; the timing, clinical significance and impact of clinical data read-outs, including the progress, results and clinical data from each of the patients dosed with rese-cel in the Phase 1/2 RESET-Myositis, RESET-SLE, RESET-SSc, RESET-MG and RESET-PV trials and our other planned activities with respect to rese-cel; our expectations regarding the potential applicability and relevance of our preliminary and interim clinical data to results in larger patient populations, registrational studies, and future clinical trials; our belief that rese-cel has the potential to provide drug-free, durable transformative clinical responses, through an immune reset; the Company's advancement of separate Phase 1/2 clinical trials of rese-cel and advancement of the RESET-PV and RESET-SLE trials, with and without preconditioning, as applicable, including updates related to status, safety data, efficiency of clinical trial design and timing of data read-outs or otherwise; our ability to enroll the requisite number of patients, dose each dosing cohort in the intended manner and timing thereof, and advance the trial as planned in our Phase 1/2 clinical trials of rese-cel; the timing of any planned regulatory filings for our development programs, including IND applications, initiation of registrational cohorts, potential BLA submission and interactions with regulatory authorities, including such authorities' review of safety information from our ongoing clinical trials and discussions with regulatory agencies on potential registrational pathway for rese-cel in various indications, and the timing of trial design related thereto; our plans and expectations regarding automated scalable manufacturing and no preconditioning and its potential to expand and accelerate access; our expectations that automation and elimination of preconditioning and apheresis will enhance patient experience; our expectation and timing for clinical manufacturing data with Cellares' automated manufacturing process and its ability to confirm GMP readiness, including supply chain logistics, as well as its potential to provide scalability for thousands of patients per year and to facilitate post-approval expansion and timing thereof; our expectation and timing for completion of dosing of most disease-specific cohorts; our expectations regarding opportunities based on market research; our ability to accelerate our pipeline to approval and launch and to develop transformative therapies for patients, including in collaboration with academic and industry partners and the ability to optimize such collaborations on, including timing thereof, our development programs; our ability to contract with third-party suppliers and manufacturers; our ability to execute our manufacturing strategy to enable expansion of clinical supply and efficiently scale commercial supply for rese-cel; our potential commercial opportunities, including value and addressable market, for our product candidates; our expectations regarding the potential commercial and economic benefits of preconditioning elimination and automated manufacturing, including its potential to reduce costs of goods, minimize capital investment requirements, and support efficient global expansion of rese-cel; our expectations regarding potential reimbursement, pricing and coverage decisions for rese-cel by governmental and private payers, site-of-care utilization and its potential impact on reimbursement treatment, market adoption, our launch strategy and commercial execution of rese-cel if approved. Words such as, but not limited to, “look forward to,” “believe,” “expect,” “anticipate,” “estimate,” “intend,” “plan,” “would,” “should” and “could,” and similar expressions or words, identify forward-looking statements. Various risks, uncertainties and assumptions could cause actual results to differ materially from those anticipated or implied in our forward-looking statements. Such risks and uncertainties include, but are not limited to, risks related to the success, cost, and timing of our development activities and clinical trials, risks related to our ability to demonstrate sufficient evidence of safety, efficacy and tolerability in our clinical trials, the risk that the results observed with the similarly-designed construct, including, but not limited to, dosing regimen, are not indicative of the results we seek to achieve with rese-cel, the risk that signs of biologic activity or persistence may not inform long-term results, risks related to clinical trial site activation or enrollment rates that are lower than expected, risks that modifications to trial design or approach may not have the intended benefits and that the trial design may need to be further modified; our ability to protect and maintain our intellectual property position, risks related to our relationships with third parties, uncertainties related to regulatory agencies’ evaluation of regulatory filings and other information related to our product candidates, our ability to retain and recognize the intended incentives conferred by any regulatory designations, risks related to regulatory filings and potential clearance, the risk that any one or more of our product candidates will not be successfully developed and commercialized, the risk that the results of preclinical studies or clinical studies will not be predictive of future results in connection with future studies, risks related to volatile market and economic conditions and our ability to fund operations and continue as a going concern. New risks and uncertainties may emerge from time to time, and it is not possible to predict all risks and uncertainties. Except as required by applicable law, we do not plan to publicly update or revise any forward-looking statements contained herein, whether as a result of any new information, future events, changed circumstances or otherwise. Although we believe the expectations reflected in such forward-looking statements are reasonable, we can give no assurance that such expectations will prove to be correct. Accordingly, you are cautioned not to place undue reliance on these forward-looking statements. No representations or warranties (expressed or implied) are made about the accuracy of any such forward-looking statements. For a discussion of these and other risks and uncertainties, and other important factors, any of which could cause our actual results to differ materially from those contained in the forward-looking statements, see the section entitled "Risk Factors" in our most recent annual report on Form 10-K and quarterly report on Form 10-Q, as well as discussions of potential risks, uncertainties, and other important factors in our other filings with the Securities and Exchange Commission. Certain information contained in this Presentation relates to or is based on studies, publications, surveys and other data obtained from third-party sources and the Company’s own internal estimates and research. While the Company believes these third-party sources to be reliable as of the date of this Presentation, it has not independently verified, and makes no representation as to the adequacy, fairness, accuracy or completeness of, any information obtained from third-party sources. The Company is the owner of various trademarks, trade names and service marks. Certain other trademarks, trade names and service marks appearing in this Presentation are the property of third parties. Solely for convenience, the trademarks and trade names in this Presentation are referred to without the ® and TM symbols, but such references should not be construed as any indicator that their respective owners will not assert, to the fullest extent under applicable law, their rights thereto.

Develop and launch the first curative targeted cellular therapies for patients with autoimmune diseases

2H27 BLA planned to include adult and pediatric DM with potential PRV; PC-free & automated mfg advancing Rese-cel: Delivering on the promise of CD19-CAR T in autoimmunity BLA – biologics license application; DM – dermatomyositis; f/u – follow-up; JDM – juvenile dermatomyositis; MG – myasthenia gravis; PC – preconditioning; PRV – priority review voucher; PV – pemphigus vulgaris; rese-cel – resecabtagene autoleucel (CABA-201); SLE – systemic lupus erythematosus; TIS – total improvement score. As of data cut-off on 16 Apr 2026. As of data cut-off on 16 Apr 2026 (6 Mar 2026 for MG data). Nunez, Daniel, et al. "CD19 CAR T Therapy Is Feasible in Patients with Pemphigus Vulgaris Treated Without Lymphodepletion in the RESET-PV Trial." Blood Journal (2026): blood-2025032093. Rese-cel data with preconditioning: 63 myositis, SSc, lupus and MG patients with up to 1.5 years of follow-up Data suggest a single rese-cel dose after discontinuation of all IMs can deliver immune reset with compelling outcomes 83% of phase 1/2 adult DM patients would have met the registrational endpoint; responses maintained through latest f/u1 First JDM patient with clinical and PK/PD data consistent with adult DM patients and response maintained at 32 weeks1 Myositis: 2H27 BLA submission plans to include adult DM and JDM data, facilitating potential PRV A single dose of rese-cel after discontinuation of all immunomodulators is reliably delivering immune reset with compelling safety & efficacy profile; initial BLA submission including adult and pediatric DM patients in 2H27 PC-free lowest dose data: Lupus (1/2 pts) & PV (2/4 pts) achieved deep B cell depletion → potential threshold dose Favorable safety profile in first 63 patients dosed with preconditioning (PC) supports outpatient administration2 94% - Either no CRS (68%) or Grade 1 CRS (26%); 97% - No ICANS Published low dose PV data in Blood3; next higher dose cohorts enrolling Automated manufacturing by Cellares offers potential scale to thousands of patients with minimal capital investment Consistent manufacturing & translational data; commercial supply agreement includes among lowest COGS in industry SSc: 2nd pivotal indication; ~25 patient, single-arm registrational study expected to start in 4Q26

Beyond myositis and scleroderma, advancing innovation with PC-free rese-cel & industrialized automated mfg Immune reset with a single infusion and a favorable safety profile Transformative Value Proposition Elimination of PC Automated Manufacturing Expands the market Catalyzes outpatient use Reduces resource utilization Improves patient and provider experience Differentiated product profile Typical ‘biologic’ margins Potential to address larger autoimmune indications including UC, T1D and RA Minimal capital investment Scalability to produce thousands of patient runs per year Substantially reduced COGS Facilitates global expansion COGS – Cost of goods sold; PC – Preconditioning; PV – Pemphigus vulgaris; SLE – Systemic lupus erythematosus; RA – Rheumatoid arthritis; T1D – Type 1 diabetes; UC – Ulcerative colitis. PC-free rese-cel can expand access while automated manufacturing can reduce COGS & increase scale

Myositis and SSc programs advancing to registrational trials with preconditioning Innovative clinical strategy to support accelerated regulatory path Program Trial Preclinical Phase 1/2 Registrational Rese-cel (CABA-201) 4-1BB CD19-CAR T RESET-Myositis® CARTA Chimeric Antigen Receptor T cells for Autoimmunity RESET-SSc™ RESET-SLE™ RESET-MG™ RESET-PV® Dermatomyositis / Antisynthetase syndrome Immune-mediated necrotizing myopathy Lupus Nephritis Non-Renal SLE Skin + Organ Cohort AChR-Ab neg. gMG AChR-Ab pos. gMG Skin Cohort Rheumatology1 Neurology Dermatology FTD Pemphigus vulgaris Contains PC-Free cohort(s) Juvenile dermatomyositis Pediatric Indication RESET™ – REstoring SElf-Tolerance; Ab – Antibody; AChR – Acetylcholine receptor; gMG – Generalized myasthenia gravis; JDM – Juvenile dermatomyositis; PRV – Priority review voucher; PV – Pemphigus vulgaris; SLE – Systemic lupus erythematosus; SSc – Systemic sclerosis Myositis patients can also be treated by neurologists or dermatologists; lupus nephritis patients can also be treated by nephrologists. FDA Fast Track Designation received in dermatomyositis, SLE and lupus nephritis, systemic sclerosis, generalized myasthenia gravis and multiple sclerosis. FDA Regenerative Medicine Advanced Therapy (RMAT) received in myositis, SLE, LN and systemic sclerosis. RMAT RMAT RMAT Rare Pediatric Disease Designation Plan to include JDM data in 2H27 myositis BLA submission, facilitating potential for PRV

Broad portfolio of trials designed to address high unmet need and realize the potential of rese-cel RESET™ program advancing trials in a broad portfolio of diseases SLE – Systemic lupus erythematosus; DM – Dermatomyositis; SSc – Systemic sclerosis; gMG – Generalized myasthenia gravis; PC – Preconditioning; ESRD – End-stage renal disease; PV – pemphigus vulgaris SSc gMG PV ~90k ~100k ~15k Middle age onset common Progressive skin & organ fibrosis with lung, cardiac, renal damage Average survival of 12y Bimodal age of onset Profound weakness that can be disabling Risk for myasthenic crises, with respiratory failure Rheum Neuro Derm PC-free only Pure autoantibody & B-cell mediated autoimmune disease Characterized by painful blisters & erosions SLE/LN ~210k / ~110k Affects young women & people of color ~30-40% with lupus nephritis, which carries ~25% risk of death or ESRD within 10y > > > Myositis ~80k Typical onset middle age Only FDA-approved therapy is IVIg in DM Three-fold increased mortality due to lung & cardiac involvement > U.S. Prevalence PC-free cohorts

Rese-cel: Construct Design, PK/PD and Safety Data

Rese-cel binder with similar in vitro & in vivo activity to construct used in academic studies in autoimmunity1,3 Rese-cel: CD19-CAR T specifically designed for autoimmunity Peng BJ, et al. Mol Ther Methods Clin Dev. 2024;32(2):101267. Dai, Zhenyu, et al. "Development and functional characterization of novel fully human anti‐CD19 chimeric antigen receptors for T‐cell therapy." Journal of Cellular Physiology 236.8 (2021): 5832-5847. Müller, Fabian, et al. "CD19 CAR T-Cell Therapy in Autoimmune Disease—A Case Series with Follow-up." New England Journal of Medicine 390.8 (2024): 687-700. Maschan, Michael, et al. “Multiple site place-of-care manufactured anti-CD19 CAR-T cells induce high remission rates in B-cell malignancy patients.” Nature Communications 12, 7200 (2021) Transmembrane domain in rese-cel is CD8α vs. TNFRSF19 (Troy) utilized in the academic construct. The two transmembrane domains have not been shown to have a significant difference in function or IFN-γ production in preclinical studies. The CD8α transmembrane domain is employed in tisagenlecleucel. Volkov, Jenell, et al. “Case study of CD19 CAR T therapy in a subject with immune-mediate necrotizing myopathy treated in the RESET-Myositis phase I/II trial.” Molecular Therapy 32.11 (2024): 3821-3828. Abstract 1733: Safety and Efficacy of CABA-201, a Fully Human, Autologous 4-1BB Anti-CD19 CAR T Cell Therapy in Patients with Immune-Mediated Necrotizing Myopathy and Systemic Lupus Erythematosus from the RESET-MyositisTM and RESET-SLETM Clinical Trials. ACR 2024. Rese-cel product design & clinical / translational data 4-1BB costimulatory domain with fully human binder Binder with similar affinity & biologic activity to academic FMC63 binder while binding to the same epitopes1,2 Same weight-based dose as in academic studies Potential to provide immune reset based on clinical and translational data5 Patients treated with rese-cel have shown compelling clinical responses with safety data that supports outpatient use for autoimmune patients6 Fully human anti-CD19 binder 4-1BB costimulatory domain CD3- signaling domain Rese-cel4

Many of the RESET trials share common elements of preconditioning, dose, and study design RESETTM clinical trials have consistent design principles1 †Follow up period encompasses at least 15 years in total, ed to regulatory guidance for CAR T cell therapies. AE, adverse event; CABA, Cabaletta Approach to B cell Ablation; FLU, fludarabine; CY, cyclophosphamide; PBMC, peripheral blood mononuclear cell; PD, pharmacodynamics; PK, pharmacokinetics; RESET, REstoring SElf-Tolerance; SLE, systemic lupus erythematosus; SSc, systemic sclerosis. Cabaletta Bio: Data on file; 1. Peng BJ, et al. Mol Ther Methods Clin Dev. 2024;32(2):101267. Leukapheresis and rese-cel production Preconditioning Single weight-based infusion of rese-cel Weight-based dosing 1×106 cells/kg Day 1 Primary endpoint: Incidence and severity of AEs T cells isolated from patient’s own PBMCs (autologous CAR T) Day 29 Study follow-up through Year 3† Screening Additional Endpoints Clinical efficacy measuring: Drug-free responses Validated endpoints PK/PD analysis: Rese-cel expansion B cell depletion Peak BAFF levels B cell repopulation Adverse events & safety Biomarker analysis, including autoantibody levels FLU 25 mg/m2 x 3 days CY 1000 mg/m2 x 1 day (if required) Day -5 to Day -3 Discontinuation of all immunomodulatory agents

B Cell Kinetics Peak rese-cel expansion and transient peripheral B cell depletion occurred by ~2 weeks post infusion Rese-cel expansion & B cell kinetics across indications* RESET-Myositis† RESET-SLE RESET-SSc Rese-cel Pharmacokinetics Rese-cel Pharmacokinetics Rese-cel Pharmacokinetics§ B Cell Kinetics‡ B Cell Kinetics Peripheral B cells begin repopulating ~2 to 3 months after rese-cel in patients with sufficient follow-up* RESET-MG Rese-cel Pharmacokinetics B Cell Kinetics ‡ CAR T cells/µL Blood All data is as of 16 Apr 2026, except MG which is as of 6 Mar 2026. *Data shown as median and IQR for myositis, SLE and SSc. †Note baseline (pre-preconditioning) B cell count for the JDM patient was not available. ‡B cell count data excluded from any patient after receiving B cell-depleting rescue therapy. §Median time includes subjects with depletion and repopulation; one Organ subject did not exhibit peripheral B cell depletion at the time points sampled. ¶AChR-pos-1: Azathioprine, a prohibited medication, was continued until the day of infusion (Day 1) that may have inhibited CAR T activity. AChR, acetylcholine receptor; ASyS, antisynthetase syndrome; CAR, chimeric antigen receptor; DM, dermatomyositis; IMNM, immune-mediated necrotizing myopathy; IQR, interquartile range; JDM, juvenile dermatomyositis; JIIM, juvenile idiopathic inflammatory myopathy; LN, lupus nephritis; MG, myasthenia gravis; PC, preconditioning; rese-cel, resecabtagene autoleucel; RESET, REstoring SElf-Tolerance; SLE, systemic lupus erythematosus, SSc, systemic sclerosis. Cabaletta Bio: Data on file. ¶

Across 4 RESET™ studies, 94% of patients have either no CRS or Grade 1 (fever) and 97% have no ICANS1 Demographics & CRS/ICANS in first 63 rese-cel patients by indication RESET-Myositis RESET-SLE RESET-SSc RESET-MG Number of patients 17 18 15 13 Age, years, mean (SD) 57.0 (15.1) 29.9 (7.4) 50.4 (14.6) 53.7 (10.0) Sex, % female 52.9 88.9 73.3 69.2 Duration of disease, years, mean (SD) 4.9 (3.7) 10.2 (5.0) 1.9 (1.2) 7 (6.8) Baseline characteristics of autoimmune disease patients treated with rese-cel Incidence, severity and onset of CRS and ICANS in the 1st 28 days in patients treated with rese-cel RESET-Myositis RESET-SLE RESET-SSc RESET-MG Total CRS‡, n (%) 5 (29.4) 6 (33.3) 7 (46.7) 2 (15.4) 20 (31.7% CRS) CRS Grade 1, n (%) 5 (29.4) 5 (27.7) 5 (33.3) 1 (7.7) 16 (25.4% G1 CRS) CRS Grade 2, n (%) – 1 (5.6) 2 (13.3) 1 (7.7) 4 (6.3% G2 CRS) Time to CRS onset, days* (mean) 7.4 6.7 9.0 5.5 7.6 days CRS duration†, days (mean) 4.6 3.0 2.9 5.0 3.6 days ICANS‡ n (%) (Grade) – 1 (5.6) (G4) 1 (6.7) (G3) - 2 (3.2% ICANS) Time to ICANS onset, days (mean) – 9.0 8.0 - 8.5 days ICANS duration, days (mean) – 3.0 3.0 - 3.0 days *Days relative to rese-cel infusion. †Events occurring within 7 days of each other are considered as 1 episode. IMNM-3 CRS duration includes preceding event of fever which was consistent with CRS definition. ‡Graded per ASTCT Consensus Grading Criteria. 1. Presented at EULAR 2026 with data cut-off as of 16 Apr 2026 (Myositis, SLE and SSc), Presented at AAN 2026 with data cut-off as of 6 Mar 2026 (MG).

Myositis: Unmet Need, Clinical Data and Commercial Opportunity

Highly concentrated treatment network in the US; dermatomyositis represents ~75% of this market Myositis: High rates of disability & increased risk of mortality High disease burden: disability & mortality Typical patient is a middle-aged female who experiences muscle weakness, fatigue, pain, shortness of breath and difficulty swallowing Moderate to severe disability (40% to 65%)1 Assisted walking devices (18% to 38%)1 The risk of mortality is ~3 times higher than the general population, primarily due to cancer and lung & cardiac complications2 ~20% mortality < 5 years with standard immunosuppressive treatment3 “I find it very difficult to get up from a regular chair, I need boosters or assistance from somebody else. Walking, my gait has really suffered. My stability walking has suffered as well, and I can't lift anything more than five or eight pounds. So doing stuff is difficult. Bending down is very difficult. I can't get up from the floor if I fall.” “John” 61-year-old male with ASyS4 ~10 yrs since diagnosis "It just affected every aspect of my life. Just work, family, social life, own wellbeing. It just pours into everything else with that." “Erica” 44-year-old female with DM4 ~2.5 yrs since diagnosis Opinc AH, Brzezinska OE, Makowska JS. Disability in idiopathic inflammatory myopathies: questionnaire-based study. Rheumatol Int. 2019;39(7):1213-1220. Marie I. Morbidity and mortality in adult polymyositis and dermatomyositis. Curr Rheumatol Rep. 2012;14(3):275-285. Schiopu E, Phillips K, MacDonald PM, Crofford LJ, Somers EC. Predictors of survival in a cohort of patients with polymyositis and dermatomyositis: effect of corticosteroids, methotrexate and azathioprine. Arthritis Res Ther. 2012;14(1):R22. Primary market research conducted via third-party, blinded interviews with myositis patients, conducted in 2024. Khoo 2023 6. Kronzer 2023 7. Coffey 2021 8. Dahal 2022 9. Papadopoulou 2022 ~2,000 pts9 Juvenile-onset disease (JDM) ~15,000 pts7,8 Anti-synthetase syndrome (ASyS) ~60,000 pts5,6 Dermatomyositis (DM) Subtype prevalence in the U.S.

IVIg is the only approved therapy (only for patients with the adult DM subtype) Dermatomyositis (DM): Limited treatment options for ~60k U.S. patients Lundberg, Ingrid E., et al. "Idiopathic inflammatory myopathies." Nature Reviews Disease Primers 7.1 (2021): 86. Analysis from quantitative survey of U.S. myositis-treating physicians, conducted 2Q25. N = ~240. Limited treatment options1 Common therapies: steroids plus an immunomodulator (i.e. methotrexate, azathioprine, mycophenolate, rituximab) IVIg (intravenous immunoglobulin), the only FDA-approved therapy, is approved in adult dermatomyositis Therapies can carry potential long-term side effects such as serious infections and organ damage Despite existing therapies, disease is often refractory Two therapies in Phase 3 development, Brepocitinib and Vyvgart®, demonstrated improvement with chronic administration added onto existing immunomodulatory medications Adult DM patients with moderate to severe, refractory disease potentially eligible for rese-cel (per analysis of quantitative research with ~240 myositis-treating physicians) Eligible U.S. adult DM patients2 ~12k to 15k U.S. patients potentially eligible for rese-cel U.S. Prevalence: ~60k adult DM patients

Single-arm cohort including DM/ASyS patients with a primary endpoint at 16 weeks Myositis registrational cohort – Key design elements Registrational Cohort2 ASyS DM & ASyS (n=17) Phase 1/2 Cohorts DM (combined n ~6) JDM, juvenile dermatomyositis; TIS, total improvement score. Pediatric JDM submission based on data available at the time of adult submission from ongoing Ph 1/2 study (no new study) to support pediatric label claim Size of myositis registrational cohort based on key statistical parameters and estimated background remission rate in myositis. Low dose steroids is defined as 50% reduction from baseline or ≤7.5 mg/day. Inclusion of JDM in 2H27 myositis BLA submission facilitates potential for priority review voucher, pending supportive data, based on rare pediatric disease designation RESET-Myositis trial now enrolling registrational cohort in DM (14 patients) / ASyS (3 patients) Plan to include JDM data in myositis BLA submission1 Primary Endpoint: Moderate or Major TIS response @ Week 16 off all immunomodulators and off or on low-dose3 steroids Weight-based single dose of rese-cel at 1 million cells/kg with safety profile appropriate for outpatient dosing Safety database ~100 autoimmune patients at ≥1-month of follow-up (with at least 35 myositis patients) P P P P JDM1 (n ≥ 6)

All patients had active, refractory disease despite multiple medications, including IVIg and B cell-targeting therapies Baseline characteristics: First 17 patients in RESET-Myositis DM N=6 ASyS N=4 IMNM N=6 JDM N=1 Age, years, median (min, max) 57 (45, 72) 44 (26, 57) 59 (33, 64) 14 Female, n (%) 5 (83) 2 (50) 1 (17) 1 (100) Disease duration, years, median (min, max) 3.5 (2.0, 10.3) 2.7 (0.9, 14.8) 4.7 (1.4, 8.8) 8.5 Myositis-specific autoantibody 50% TIF1-γ 17%: NXP, SAE, MDA-5 100% Jo-1 67% HMGCR 33% SRP NXP-2 Baseline disease activity, median* MMT-8 CK CDASI-A 123.0 40.0 23.5 129.5 257.5 N/A 127.5 2214.5 N/A 134.0 176.0 5 Prior RTX† (%) Prior IVIg† (%) 50% 67% 100% 75% 83% 83% 100% 100% Therapies at Screening Systemic GCs ≤2 IMs ≥3 IMs 67% 67% 33% 75% 75% 25% 67% 100% 0 0 0 100% As of 16 Apr 2026.*Baseline disease activity = activity before preconditioning. †Reflects any exposure to RTX and IVIg prior or at time of study entry. RTX is not allowed within approximately 6 months of screening.ASyS, antisynthetase syndrome; CDASI-A, Cutaneous Dermatomyositis Disease Area and Severity Index – Activity; CK, creatine kinase; DM, dermatomyositis; GC, glucocorticoid; HMGCR, 3-hydroxy-3-methylglutaryl-coenzyme A reductase; IM, immunomodulatory medication; IMNM, immune-mediated necrotizing myopathy; IVIg, intravenous immunoglobulin; JDM, juvenile dermatomyositis; MDA-5, melanoma differentiation-associated gene 5; MMT-8, manual muscle testing 8; NXP, nuclear matrix protein; N/A, not applicable; RESET, REstoring SElf-Tolerance; RTX, rituximab; SAE, small ubiquitin-like modifier activating enzyme; SRP, signal recognition particle; TIF1, transcription intermediary factor 1; U/L, units per liter. Cabaletta Bio – Data on File.

6 of 7 patients achieved moderate or major IM-free TIS response at Week 16 which was maintained through latest follow-up Efficacy data in DM and JDM patients following rese-cel infusion As of 16 Apr 2026. *Low-dose steroids is defined as 50% reduction from baseline or ≤7.5 mg/day. †Missing data were imputed using last observation carried forward. ‡TIS threshold for a moderate response is ≥45 in patients with JDM; TIS scale on the Y-axis reflects adult thresholds. §Median and n numbers are based on DM patients (excluding JDM patient) not receiving rescue immunomodulatory medications. BL, baseline; DM, dermatomyositis; GC, glucocorticoids; IM, immunomodulatory medication; JDM, juvenile dermatomyositis; mg, milligrams; rese-cel, resecabtagene autoleucel; TIS, total improvement score. Cabaletta Bio: Data on File. 5 of 6 adult Phase 1/2 DM patients and the JDM patient would have achieved the 16-week primary endpoint for the pivotal study and all of them maintained IM-free TIS response through latest follow-up, as long as 1.5 years Assessment at Week 16 DM and JDM patients (N=7) Complete B cell depletion (%) 100% IM-free & low-dose* or no GC (%) 100% Moderate or major TIS response (%) 86% Meets moderate or major TIS off IM therapy & on low-dose or no GCs* (Pivotal primary endpoint) 86% TIS Mod. Major Min. None TIS TIS (N=7)† Discontinued all immunomodulators n§ 6 6 6 6 6 3 2 1 Visit Week

Rese-cel safety profile facilitates outpatient administration which could allow for favorable reimbursement CAR T may eliminate active disease & use of expensive medications Cancer patients experience early and frequent CRS/ICANS following CAR T therapy, which increases inpatient admissions and shifts Medicare reimbursement to the DRG system. Majority of oncology patients treated with CAR T therapy experience CRS within first 5 days post-infusion1 Cancer CAR T: Safety profile often requires inpatient infusion, affecting reimbursement Rese-cel: Safety profile facilitates outpatient infusion, which could favorably impact reimbursement Many cancer patients are insured under Medicare, which has inpatient DRG-018 reimbursement Myositis & SSc patients often commercially insured (60%-75%)2,3 CRS less frequent & severe, delayed onset  potential outpatient administration Outpatient CAR T infrastructure exists at many centers RESET clinical site footprint can be leveraged to generate early adopters Outpatient administration supports viable Part B Medicare payments Ferreri, Christopher J., and Manisha Bhutani. "Mechanisms and management of CAR T toxicity." Frontiers in Oncology 14 (2024): 1396490. Smoyer-Tomic KE, et al. BMC Musculoskelet. Disord. 2012 Jun 15;13:103. doi: 10.1186/1471-2474-13-103. Gale, Sara L., et al. "Characterizing disease manifestations and treatment patterns among adults with systemic sclerosis: a retrospective analysis of a US healthcare claims population." Rheumatology and therapy 7.1 (2020): 89-99. Commercial

Outpatient administration reduces administrative burden and improves patient and provider accessibility RESET™ program designed for outpatient administration at launch INPATIENT MODEL OUTPATIENT MODEL Limited patient beds and resource infrastructure Increases inpatient resource pressure: ↑ total cost of care, human resource and bed space demands Reduces eligible patients treated More favorable safety profile reduces need for inpatient admission Reduces use of hospital resources; Increases throughput Reduces conflicts with cancer patient use of in-patient beds

Health status of patient population and slower disease progression improve manufacturing cost efficiency Rese-cel commercial model – manufacturing and COGM Late-stage oncology patients have high drop-off rate due to rapid disease progression and compromised T cell fitness, leading to higher manufacturing OOS rates1,2,3 Increased OOS rates; ↑ COGM + ↓ revenue since out of spec products not reimbursed In oncology, disease progress & out of specification (OOS) rates increase costs and reduce margins In autoimmunity, less pretreated patients & automated rese-cel mfg should support lower COGM Disease progression reduces revenue capture because unused product is not reimbursed Manufacturing capacity constraints  delayed commercial ramp-up Autoimmune patients are not heavily pretreated with chemotherapy  more fit immune cells4 that support reliable manufacture, reducing COGM Building manufacturing capacity at CDMOs to support successful launch; Cellares automation has the potential to facilitate post-approval expansion Autoimmune patients rarely progress as rapidly as cancer patients  more reliable revenue realization for manufactured product Reduced eligible patients, resulting in economies of scale not being achieved COGM – Cost of goods manufactured U.S. Food and Drug Administration. Kymriah (tisagenlecleucel) Prescribing Information. Revised 2025, U.S. Food and Drug Administration, https://www.fda.gov/media/107296/download U.S. Food and Drug Administration. Breyanzi (lisocabtagene maraleucel) Prescribing Information. Revised 2025, U.S. Food and Drug Administration, https://www.fda.gov/media/145711/download U.S. Food and Drug Administration. Yescarta (axicabtagene ciloleucel) Prescribing Information. Revised 2025, U.S. Food and Drug Administration, https://www.fda.gov/media/108377/download Sharma, Sagar, et al. Consistency in resecabtagene autoleucel product quality across RESET phase 1/2 clinical trials and manufacturing platforms American Society of Gene & Cell Therapy (ASGCT) 2026 Annual Meeting, 12 May 2026, Boston, MA. Oral presentation.

Rese-cel safety data facilitates plan to target academic centers at launch with rapid community expansion Three-phase approach to rese-cel launch in myositis and beyond1 Rese-cel is an investigational product candidate and has not been approved by the FDA. L+2yr+ Phase 3 Academic, Community Hospitals and Outpatient Clinics Objective: Increase patient reach Patient Population: Moderate/Severe Site of Care: Outpatient clinics and supporting hospitals through network agreements Phase 1 Academic Hospitals & Clinics Patient Population: Moderate/Severe Objective: Migrate from hospital to outpatient Site of Care: Academic hospital In & outpatient product administration Launch +1yr L+1yr-2yr Phase 2 Academic & Community Hospitals Patient Population: Moderate/Severe Objective: Increase patient reach and improve physician referral dynamic Site of Care: Academic and Community hospitals through Hub and Spoke model

Systemic Sclerosis: Unmet Need & Clinical Data

Associated with progressive morbidity and high mortality1,2 Systemic sclerosis: Profound unmet need & limited options AHSCT, autologous hematopoietic stem cell transplantation; ILD, interstitial lung disease; SSc, systemic sclerosis. 1. Allanore Y, et al. Nat Rev Dis Primers. 2015;1:15002. 2. Denton CP, et al. Lancet. 2017;390(10103):1685–1699. 3. Thoreau B, et al. Front Immunol. 2022;13:933468. 4. Truchetet ME, et al. Clin Rev Allergy Immunol. 2023;64(3):262–283. 5. Pope JE, et al. Nat Rev Rheumatol. 2023;19(4):212–226. 6. Results from quantitative survey of U.S. SSc-treating physicians (rheumatologists), conducted 3Q25. N = ~100. Rare, potentially life-threatening autoimmune disease1 Typically, middle age onset and more common in females1 Highest mortality of all rheumatological diseases & significant burden from persistent skin & organ manifestations4,5 Mean survival is ~12 years from diagnosis Need for disease-modifying therapies across all SSc subsets5 FDA-approved agents for SSc-ILD slow but do not stabilize or improve lung progression Approved based on 1-year primary endpoints No existing treatments capable of halting SSc pathology other than AHSCT, which carries high risk Patients experience a progressive & often fatal course SSc patients with early, active disease potentially eligible for rese-cel (per quantitative research with ~100 SSc-treating physicians) Eligible U.S. SSc patients6 ~12,000-15,000 U.S. SSc patients potentially eligible for rese-cel ~90k SSc patients (~40% with clinically significant ILD) Characterized by progressive skin & internal organ fibrosis1 Deep, tissue-level B cell-driven autoimmunity, with activated B cells & autoantibodies, promotes inflammation & organ damage3

All patients had active, refractory disease despite multiple SSc therapies, and 12/15 had ILD Baseline characteristics: First 15 Patients in RESET-SSc As of 16 Apr 2026. *Time of diagnosis to screening. †Per HRCT. ‡Baseline disease activity = activity before preconditioning. §The severe skin cohort includes patients with severe skin involvement and the organ cohort includes patients with significant organ involvement (e.g., pulmonary, renal, or cardiac involvement), irrespective of skin disease severity. ¶Missing data were imputed using last observation carried forward; FVC and DLCO data for 1 patient were imputed using values from screening visit. **Missing baseline data for SSc-Organ-9.DLCO, % predicted diffusing capacity for carbon monoxide; FVC, forced vital capacity; HAQ-DI, Health Assessment Questionnaire Disability Index; HCQ, hydroxychloroquine; HRCT, high-resolution computed tomography; ILD, interstitial lung disease; IVIg, intravenous immunoglobulin; MMF, mycophenolate mofetil; MPA, mycophenoloic acid; mRSS, modified Rodnan skin score; PGA, Physician Global Assessment; RESETTM, REstoring SElf-Tolerance; RNA pol, ribonucleic acid polymerase; Scl-70, anti–topoisomerase I antibody; SSc, systemic sclerosis. Cabaletta Bio: Data on File. Cohort Severe skin cohort§ (N=6) Organ cohort§ (N=9) All (N=15) ILD (N=12) Age, years, median (min, max) 57 (42, 66) 43 (19, 70) 54 (19, 70) 54.5 (19, 70) Female, n (%) 4 (67) 7 (78) 11 (73) 9 (75) Disease duration,* years, median (min, max) 1.6 (0.5, 2.2) 2.1 (0.4, 5.0) 1.8 (0.4, 5.0) 1.7 (0.4, 5.0) ILD at screening,† n (%) 4 (67) 8 (89) 12 (80) 12 (100) Cardiac involvement at screening, n (%) 0 (0) 3 (33) 3 (20) 2 (17) Autoantibodies, % Scl-70 Anti-RNA pol III 17 83 78 11 53 40 58 25 Baseline disease activity, median‡ mRSS HAQ-DI FVC (% predicted) DLCO (% predicted) PGA 40 2.19 92¶ 75.5¶ 6 19 1.63 77 66 5.5** 24 1.88 79 72 6** 23 1.81 78.5 68 6** Therapies at screening, n (%) MMF/MPA Tocilizumab Prednisone Methotrexate HCQ IVIg 5 (83) 1 (17) 1 (17) 1 (17) 0 (0) 0 (0) 7 (78) 5 (56) 2 (22) 1 (11) 3 (33) 2 (22) 12 (80) 6 (40) 3 (20) 2 (13) 3 (20) 2 (13) 11 (92) 6 (50) 3 (25) 2 (17) 2 (17) 2 (17) Number of prior SSc therapies, median (min, max) 3 (2, 5) 4 (3, 6) 4 (2, 6) 3 (2, 6)

After discontinuing immunomodulators, FVC improvement or stabilization was observed in most patients with ILD* Efficacy data in SSc following rese-cel infusion As of 16 Apr 2026.*Based on HRCT at screening. †Missing data were imputed using last observation carried forward. ‡Median and n numbers are based on SSc patients not receiving rescue immunomodulatory medications. §Absolute change. ¶ n numbers are based on evaluable SSc patients. BL, baseline; FVC, forced vital capacity; ILD, interstitial lung disease; HRCT, high-resolution computed tomography; rCRISS, revised Composite Response Index in Systemic Sclerosis; rese-cel, resecabtagene autoleucel; SSc, systemic sclerosis. Cabaletta Bio: Data on File. Based on these data and FDA discussions, Cabaletta plans to advance a single-arm registrational study of ~25 patients with SSc-associated ILD using an FVC-based primary endpoint at 52 weeks with initiation anticipated in 4Q26 Change from baseline in FVC % predicted§ Discontinued all immunomodulators n‡ 12 12 8 4 2 1 Visit week FVC % predicted (Patients with ILD*; N=12)† Proportion of patients achieving rCRISS response n¶ 15 10 6 2 rCRISS response (All cohorts, N=15) Visit week

Rese-cel Manufacturing Strategy & Innovation

Rese-cel Manufacturing Process – Commercial Readiness T cell collection Rese-cel manufacture (9 day) Rese-cel administration apheresis cell selection and activation cell expansion cell harvest preconditioning infusion Product release QC testing and QA release Partial automation anticipated at launch with full automation planned for post-launch supply Process B – Commercial-ready manufacturing process Substantially closed and partially automated to improve process capacity and consistency over original process Implemented prior to IIM registrational study, with FDA alignment on comparability1 Planned for initial launch through proven commercial CDMO partners Process C – Industrialized, automated manufacturing process with Cellares Comparability data, including multiple engineering runs, supported treatment of Phase 1/2 patients2 Planning for commercial implementation ~1-year post-launch3 1 Current comparability data allows use of safety data from patients treated with original process (“Process A”) and Process B; additional data submission planned in 2026 to allow inclusion of efficacy data. 2 Current comparability data allows use of safety data from patients treated with Process A, full analytical comparability between Process A, B and C pending process optimization. 3 Following completion of process optimization, process validation work, and based on full analytical comparability data set.

Process C: Consistent manufacturing and translational rese-cel data in 1st 2 patients vs. traditional CDMOs Automated manufacturing initial manufacturing & translational data Cellares uses Celleca for viability measurement to enable automated testing. Viability data for Process C shown in this figure were measured using NC200 to enable direct comparison with Process A and Process B, for which NC200 was also used Data points with asterisks had low DNA input per assay recommendation. Median and 50th percentile intervals of values from subjects manufactured with conventional manufacturing (conv) shown in gray. 1st 2 rese-cel clinical manufacturing runs met release specifications including product quality attributes within historical ranges; initial data suggest rese-cel manufactured by Cellares exhibited comparable in vivo behavior to rese‑cel products manufactured using conventional processes Viability 1 (accesses T cell health) %CD45+ CD3+ (Total T cells produced) %CD3+ CAR+ (Total CAR-T cells produced) Vector copy number (Amount of vector/transduced cell) Potency (Cytotoxicity) CAR T Cell Expansion2,3 B Cell Depletion3 * *

Automation plus elimination of preconditioning and apheresis could impact patient preference & experience Advancing breakthrough innovations to improve scalability and costs Stratton et al, ESGCT 2024. Poster available at https://www.cabalettabio.com/technology/posters-publications (https://d1io3yog0oux5.cloudfront.net/_cdcc45a1b07d9c1e0fc529e815f21ec3/cabalettabio/db/947/8240/pdf/Whole+Blood+Mfg+Poster+ESGCT+2024.pdf) Automation run feasibility completed under TAP program. Video on Cellares technology can be viewed here: https://vimeo.com/947203843/cd59569f16. Under evaluation in an ongoing study in Pemphigus Vulgarus (NCT004422912); presented at ESGCT Conference 2025, presentation is available at https://www.cabalettabio.com/technology/posters-publications. T cell collection Rese-cel manufacture and harvest (9 day) Rese-cel administration apheresis cell selection and activation cell expansion cell harvest preconditioning infusion Product release post-production testing and QA review CURRENT INNOVATIONS One outpatient infusion without preconditioning4 Blood draw to replace apheresis1,2 Cellares Cell Shuttle3 Fully closed, end-to-end automation Rapid & global scalability Shorter turn-around-time

Rese-cel – PC-Free Data at Lowest Dose in PV and Lupus

In the RESET-PV trial, 4 refractory patients received rese-cel at the lowest dose without preconditioning and had follow-up between 24 and 36 weeks as of the data cut-off 2 of 4 patients demonstrated compelling clinical activity through 6 months follow-up Complete peripheral B cell elimination was observed in 3 of 4 patients CRS was observed in 1 patient (Grade 1); ICANS – none Based on the safety profile observed at the lowest dose, multiple additional patients have been enrolled at a higher dose cohort in the RESET-PV trial and longer-term data at the higher dose is anticipated in 2H26 In the RESET-SLE trial evaluating PC-free rese-cel in patients with lupus, 1 of 2 patients experienced deep B cell depletion as measured by flow cytometry and peak BAFF levels After discontinuing all immunomodulators (IM), unanticipated activity observed without PC at lowest dose Summary of rese-cel without preconditioning (PC), lowest dose cohort1 Data cut off as of April 2, 2026. Cabaletta Bio: Data on file.

Similar CAR T expansion to PC-treated rese-cel pts with complete B cell depletion in 3 of 4 PC-free patients PK / PD for PC-free rese-cel in patients with pemphigus vulgaris1 LD, lymphodepletion Note: Gray vertical dotted line indicates day of rese-cel infusion (study visit Day 1). Median and 50th percentile intervals of values from LD subjects shown in gray. The LD group consisted of 38 rese-cel treated subjects that were evaluated across the RESET program in myositis, lupus, systemic sclerosis and myasthenia gravis. Cabaletta Bio: Data on file. As of 2 April 2026. 1. Nunez, Daniel, et al. "CD19 CAR T Therapy Is Feasible in Patients with Pemphigus Vulgaris Treated Without Lymphodepletion in the RESET-PV Trial." Blood Journal (2026): blood-2025032093. CAR T cell expansion Peak BAFF levels PV-1M-2, PV-1M-3 & PV-1M-4 had complete depletion of peripheral B cells at lowest rese-cel dose BAFF levels in these three patients were at the low end of the range observed with rese-cel with PC Days post infusion Cells/μL pg/mL LD Non-LD B cell depletion PC-free data in pemphigus vulgaris

Near complete resolution of clinical symptoms off all medicines in 2 of 4 patients through 6 months follow-up1 Unanticipated efficacy at lowest PC-free dose supports dose escalation *As of 2 April 2026. Cabaletta Bio: Data on file. Disease severity intervals as defined Krain RL, et al. Br J Dermatol. 2021;184(5): 975–977. Gray vertical dotted line indicates day of rese-cel infusion (study visit Day 1). 1. Nunez, Daniel, et al. "CD19 CAR T Therapy Is Feasible in Patients with Pemphigus Vulgaris Treated Without Lymphodepletion in the RESET-PV Trial." Blood Journal (2026): blood-2025032093. PDAI improvements were the greatest in the three patients who experienced complete peripheral B cell depletion; Next higher dose cohort enrolling based on efficacy, translational data and favorable safety profile to date PC-free data in pemphigus vulgaris PDAI Total Activity Score Anti-DSG3 and Anti-DSG1 Titers Anti-DSG3 Anti-DSG1 U/mL U/mL Anti-DSG3 and anti-DSG1 antibodies declined with initial improvement in PDAI total activity scores

Similar CAR T expansion as PC-treated patients with deep B cell depletion in 1 of 2 PC-free patients Lupus: PC-free rese-cel PK and PD1 LD, lymphodepletion 1. Shading indicates interquartile range, IQR. Individual dashed lines represent PC-free SLE and LN subjects. Rese-cel infusion shown as vertical gray dotted line and PC window represented as vertical gray shading. Cabaletta Bio: Data on file. As of 16 April 2026 for PC SLE/LN, as of 16 May 2026 for PC-free patients. CAR T cell expansion1 Peak BAFF levels PK profiles with the lowest dose of PC-free rese-cel were similar to rese-cel with PC Deep B cell depletion observed in 1 of 2 patients based on flow cytometry and peak BAFF levels Gr 1 CRS observed in 1 of 2 patients; no ICANS observed B cell depletion1 PC-free data in lupus Days post infusion CAR T cells/μL Days post infusion B cells/μL Max BAFF (pg/mL) SLE/LN PC-free All Trials with LD

Lupus: Unmet Need & Clinical Data

Increased mortality risk & negative impact on quality of life for patients with SLE & LN SLE & LN: Represent a high unmet clinical need ESRD, end-stage renal disease; LN, lupus nephritis; SLE, systemic lupus erythematosus. 1. Zen M, et al. Eur J Intern Med. 2023;112:45–51. 2. Rahman P, et al. Lupus. 2001;10(2):93-96. 3. Singh, R, et al. Lupus 27.10 (2018): 1577-1581.4. Murimi-Worstell, I., et al. BMJ 10.5 (2020): e031850. 5. Lichtnekert, J. Nature reviews rheumatology 20.11 (2024): 699-711. 6. Tektonidou, M. Arthritis & rheumatology 68.6 (2016): 1432-1441. 7. Results from quantitative survey of U.S. lupus-treating physicians (rheumatologists & nephrologists), conducted 2Q25. N = ~150. SLE patients with moderate to severe, refractory disease & LN patients with refractory disease potentially eligible for rese-cel (per analysis of quantitative research with ~150 lupus-treating physicians) Eligible U.S. Non-Renal SLE & LN patients7 ~40,000 U.S. non-renal SLE patients potentially eligible for rese-cel ~210k non-renal SLE patients ~50,000 U.S. LN patients potentially eligible for rese-cel ~110k LN patients SLE is a chronic autoimmune condition that can affect nearly every organ system1 Most common in women, with disease onset generally between ages of 20-40 years Common symptoms include severe fatigue, joint pain and swelling, skin rashes, ulcers & Raynaud’s phenomenon >50% of patients develop permanent widespread organ damage, caused by disease & current treatments2 Standardized mortality ratio from 2.4-4.5 for SLE patients3,4 ~30-40% of SLE patients develop LN, with inflammation & damage within the kidneys LN may present silently or with symptoms such as proteinuria, hematuria, swelling & elevated blood pressure 10-30% of patients with LN will progress to ESRD, requiring dialysis or transplantation within the first decade of their disease5,6 Market research indicates opportunity to achieve superior penetration and potentially further expand the market through introducing a PC-free CAR T alternative for patients

All patients had active, refractory disease and had failed multiple B cell-targeted therapies Baseline characteristics: First 18 patients in RESET-SLE As of 16 Apr 2026. *Time from diagnosis to screening. †Baseline disease activity = activity before preconditioning.‡SLE medications may include biologics, antimalarials, and immunosuppressants. §Four patients in the non-renal SLE cohort had renal involvement that did not meet criteria for the LN cohort. dsDNA, double-stranded DNA; GC, glucocorticoid; LN, lupus nephritis; RESET, REstoring SElf-Tolerance; SLE, systemic lupus erythematosus; SLEDAI-2K, SLE Disease Activity Index 2000; Sm, Smith; UPCR, urine protein-to-creatinine ratio. Cabaletta Bio: Data on File. Cohort Non-renal SLE (N=12) LN (N=6) Age, years, median (min, max) 31 (21, 44) 26 (18, 35) Female, n (%) 11 (92) 5 (83) Disease duration,* years, median (min, max) 12 (5, 17) 6 (2, 16) Autoantibodies (%) dsDNA Sm 75 67 67 83 Baseline disease activity† SLEDAI-2K (mean) 12.75 16 UPCR (mg/mg) (mean) 1.7§ 3.1 Therapies at screening (%): Systemic GCs ≤2 SLE immunomodulators‡ ≥3 SLE immunomodulators‡ 92 50 50 50 33 67 GC dose at screening, mg/day, mean (min, max) 13.25 (0, 30) 5.83 (0, 20)

Improvements in SLEDAI-2K and PGA over time after discontinuing immunomodulators Efficacy data in SLE following rese-cel infusion As of 16 Apr 2026. *Missing data were imputed using last observation carried forward.†Dashed single patient trend lines represent patients receiving rescue immunomodulatory medications. ‡Mean/median and n numbers are based on SLE/LN patients not receiving prohibited rescue immunomodulators.§PGA measured on a 0–3 scale. CAR, chimeric antigen receptor; DORIS, definition of remission in SLE; GC, glucocorticoid; LN, lupus nephritis; PD, pharmacodynamic; PGA, Physician Global Assessment; PK, pharmacokinetic; rese-cel, resecabtagene autoleucel; SLE, systemic lupus erythematosus; SLEDAI-2K, Systemic Lupus Erythematosus Disease Activity Index 2000. Cabaletta Bio: Data on File. SLEDAI-2K, non-renal SLE (N=12)*,† SLEDAI-2K, LN (N=6)* PGA (N=18)*,†,§ These data suggest the potential for rese-cel to reset the immune system and achieve meaningful clinical responses off immunomodulators with no or low-dose GCs in non-renal SLE and LN n‡ 12 12 7 6 5 5 3 3 2 Visit week Visit week n‡ 6 5 5 4 4 4 4 4 1 Visit week n‡ 18 17 12 10 9 9 7 7 3 DORIS threshold <0.5 Discontinued all SLE immunomodulators‡ Discontinued all SLE immunomodulators Discontinued all SLE immunomodulators‡

Myasthenia Gravis: Unmet Need & Clinical Data

High impact of disease due to patient symptoms & cost burden, particularly for refractory patients Myasthenia gravis: Significant disease & treatment burden Gilhus NE, et al. Eur J Neurol. 2024 2. Dresser L, et al. J Clin Med. May 2021. 3. Results from quantitative survey of U.S. MG-treating physicians (neurologists), conducted 3Q25. N = ~100. Generalized MG patients with refractory disease potentially eligible for rese-cel (per quantitative research with ~100 MG-treating physicians) Eligible U.S. MG patients3 ~35,000 U.S. MG patients potentially eligible for rese-cel ~100k MG patients Serious, chronic autoimmune neuromuscular disorder1 Available therapeutic options focus on specific symptoms and can be associated with serious long-term side effects Mainstays include steroids, immunosuppressants (e.g., mycophenolate), FcRn antagonists, complement inhibitors and rituximab MG represents a significant healthcare cost burden in the US, particularly for patients whose disease is inadequately controlled Treatments have transient effect & involve long-term broad immunosuppression1 Characterized by defective transmission at the neuromuscular junction, resulting in weakness of the skeletal muscles Typically associated with autoantibodies (e.g. AChR, MuSK, LRP4) Symptoms range from ocular involvement, including double vision and ptosis, to severe weakness of the limb, bulbar, trunk, and respiratory muscles, which is worsened with exertion Mortality rate estimated to be 5-9%, primarily driven by myasthenic crises, or respiratory crises requiring ventilation2

All patients had active, refractory disease despite multiple immunomodulatory agents Baseline Characteristics: 13 RESET-MG Patients* AChR Positive (n=7) AChR Negative (n=6) Age, years, mean (min, max) 54.0 (41, 65) 53.3 (37, 70) Female, n (%) 3 (42.9%) 6 (100.0%) Time from diagnosis to screening, years, mean (min, max) 7.10 (1.4, 19.1) 6.83 (0.6, 16.2) Autoantibodies (%) AChR: 100% Seronegative: 50% MuSK: 33.3% LRP4: 16.7% Baseline disease activity† MG-ADL (mean) 12.3 12.8 QMG (mean) 14.1 16.8 Prior MG therapies (excluding GCs), mean (min, max) 4.6 (0, 8) 3.5 (1, 6) Therapies at screening: Systemic GCs ≤2 MG therapies‡ ≥3 MG therapies‡ 57% 71% 29% 50% 83% 17% GC dose at screening§, mg/day, mean (min, max) 10 (0, 25) 10.8 (0, 30) *As of 6 March, 2026. †Baseline disease activity = activity before preconditioning. ‡MG therapies include acetylcholinesterase inhibitors, FcRn inhibitors, biologics, IVIg, and immunosuppressants. §GC dose = glucocorticoid dose expressed in equivalent dose of prednisone (mg/day). AChR, acetylcholine receptor; FcRn, neonatal Fc receptor; GC, glucocorticoid; IVIg, intravenous immunoglobulin; LRP4, low-density lipoprotein receptor-related protein 4; MG, myasthenia gravis; MG-ADL, MG – Activities of Daily Living; MuSK, muscle-specific tyrosine kinase; QMG, Quantitative Myasthenia Gravis Score; RESET, REstoring SElf-Tolerance; rese-cel, resecabtagene autoleucel.Cabaletta Bio – Data on File.

No CRS was observed in 11 of 13 patients; CRS was mild and resolved with no sequelae; no ICANS observed Incidence of Relevant and Related Serious Adverse Events* Cohort AChR Positive AChR Negative Patient AChR-pos-1 AChR-pos-2 AChR-pos-3 AChR-pos-4 AChR-pos-5 AChR-pos-6 AChR-pos-7 AChR-neg-1 AChR-neg-2 AChR-neg-3 AChR-neg-4 AChR-neg-5 AChR-neg-6 CRS† None Grade 2‡ None None None None Grade 1‡ None None None None None None ICANS† None None None None None None None None None None None None None Serious infections§ None None None None None None None None None None None None None Related SAEs¶ (Grade) (Excluding CRS/ICANS) None Physical deconditioning, anorexia (3) None None None None None None None None None Neutropenic fever (3) None *As of 6 March, 2026; (N=13 dosed); primary endpoint is incidence and severity of adverse events through Day 29. †Graded per ASTCT Consensus Grading Criteria. ‡The median time to onset of observed CRS was 5 days (range 2–8 days) relative to the rese-cel infusion (events occurring within 7 days of each other were considered a single event).§Coded in System Organ Class of Infections and Infestations and meets seriousness criteria. ¶As assessed per US Food and Drug Administration guidelines. AChR, acetylcholine receptor; AE, adverse event; ASTCT, American Society for Transplantation and Cellular Therapy; CRS, cytokine release syndrome; ICANS, immune effector cell-associated neurotoxicity syndrome;SAE, serious adverse event. Cabaletta Bio – Data on File.

After discontinuation of all immunomodulators Efficacy data in AChR-positive patients following rese-cel infusion1* *As of 6 March, 2026.†Baseline disease activity = activity before preconditioning. ‡Mean and n numbers are based on dosed patients not receiving rescue medication for MG. §AChR-pos-1: AZA, a prohibited medication, was continued until the day of infusion (Day 1). IVIg was stopped prior to rese-cel infusion and restarted 4 weeks after infusion for continued MG symptoms; patient discontinued study at Week 20 due to visit refusal. AChR-pos-1, Day 29 visit data unavailable. ¶AChR-pos-4 missed Week 16 visit. **AChR-pos-7 received rescue IVIg due to MG exacerbation 3 days post rese-cel infusion; IVIg rescue therapy ongoing. AChEI, acetylcholinesterase inhibitors (i.e PYR); AChR, acetylcholine receptor; AZA, azathioprine; EFG, efgartigimod; GC, glucocorticoid; IM, immunomodulatory medication; IVIg, intravenous immunoglobulin; MG, myasthenia gravis; MG-ADL, MG – Activities of Daily Living; MMF, mycophenolate mofetil; PYR, pyridostigmine; QMG, Quantitative Myasthenia Gravis Score; rese-cel, resecabtagene autoleucel.1. Cabaletta Bio – Data on File. 2. Muppidi S, et al. Muscle Nerve. 2022;65(6):630–639. 3. EMA. Available at: www.ema.europa.eu/en/documents/overview/soliris-epar-medicine-overview_en.pdf (accessed April 2026). 4. Barnett C, et al. Neurol Clin. 2018;36(2):339–353. After discontinuation of all immunodulators, 5 of 7 AChR-positive patients showed clinically meaningful improvements on the MG-ADL scale AChR-pos-1§ AChR-pos-2 AChR-pos-3 AChR-pos-4 AChR-pos-5 AChR-pos-6 AChR-pos-7** MG medications (screening visit) AZA, IVIg, GC, PYR EFG, PYR EFG, MMF, GC EFG, MMF, GC, PYR PYR GC, PYR IVIg GC-free (latest follow-up) Taper  Taper   Taper  AChEI-free (latest follow-up) Ongoing    Taper   MG-ADL response Received IVIg      Received IVIg MG-ADL point change from baseline† Clinically meaningful improvement threshold2,3 MG-ADL QMG Clinically meaningful improvement threshold4 Discontinued all immunomodulators ≥2 QMG point change from baseline† n‡ 5 5 5 5 3¶ 3 2 1 n‡ 5 5 5 5 3¶ 3 2 1 ≥3 Discontinued all immunomodulators AChR-pos-1 not evaluable due to continued use of a prohibited medication, azathioprine, until the day of infusion. AChR-pos-1 not evaluable due to continued use of a prohibited medication, azathioprine, until the day of infusion.

After discontinuation of all immunomodulators Efficacy data in AChR-negative patients following rese-cel infusion1* *As of 6 March, 2026. †Baseline disease activity = activity before preconditioning. ‡Mean and n numbers are based on dosed patients not receiving rescue medication for MG. §AChR-neg-1, no Week 44 QMG performed (unrelated AE prevented assessment being completed); AChR-neg-2, no Week 28 visit data available (missed visit). ¶AChR-neg-1 receiving low dose IVIg for ongoing hypogammaglobinemia every 2 months from Week 8; AChR-neg-2 received rescue EFG from Week 13 through Week 16 and IVIg every 3 weeks from Week 24 visit due to MG symptoms; AChR-neg-3 receiving chronic GC for adrenal insufficiency. AChEI, acetylcholinesterase inhibitors (i.e PYR); AChR, acetylcholine receptor; AZA, azathioprine; EFG, efgartigimod; GC, glucocorticoid; IM, immunomodulatory medication; IVIg, intravenous immunoglobulin; LRP4, low-density lipoprotein receptor-related protein 4; MG, myasthenia gravis; MG-ADL, MG – Activities of Daily Living; MMF, mycophenolate mofetil; MuSK, muscle-specific tyrosine kinase; PLA, plasmapheresis; PYR, pyridostigmine; QMG, Quantitative Myasthenia Gravis Score; rese-cel, resecabtagene autoleucel; ROZ, rozanolixizumab, 1. Cabaletta Bio – Data on File. 2. Muppidi S, et al. Muscle Nerve. 2022;65(6):630–639. 3. EMA. Available at: www.ema.europa.eu/en/documents/overview/soliris-epar-medicine-overview_en.pdf (accessed April 2026). 4. Barnett C, et al. Neurol Clin. 2018;36(2):339–353. After discontinuation of all immunomodulators, 5 of 6 AChR-negative patients showed clinically meaningful improvements on the MG-ADL scale AChR-neg-1¶ (seronegative) AChR-neg-2¶ (seronegative) AChR-neg-3 (MuSK) AChR-neg-4 (MuSK) AChR-neg-5 (LRP4) AChR-neg-6 (seronegative) MG medications (screening visit) PLA, GC, PYR MMF, ROZ, PYR PLA,GC AZA EFG EFG, GC, PYR GC-free (latest follow-up)   No¶   Taper AChEI-free (latest follow-up)      Taper MG-ADL response  Received EFG and IVIg     MG-ADL point change from baseline† Clinically meaningful improvement threshold2,3 MG-ADL Discontinued all immunomodulators QMG Clinically meaningful improvement threshold4 QMG point change from baseline† Discontinued all immunomodulators ≥2 n‡ 6 6 6 6 3 3 2 1 1 1 1 1 Seronegative MuSK LRP4 Seronegative MuSK LRP4 ≥3 n‡ 6 6 6 6 3 3 2 1 1 1 1

Corporate Summary

Track record of operational success evaluating & developing novel cell therapy candidates in autoimmunity Cabaletta Bio leadership LEADERSHIP TEAM Anup Marda Chief Financial Officer Nicolette Sherman Chief HR Officer Michael Gerard General Counsel Steven Nichtberger, M.D. President, CEO & Chairman Heather Harte-Hall Chief Compliance Officer Gwendolyn Binder, Ph.D. President, Science & Technology SCIENTIFIC ADVISORY BOARD Aimee Payne, M.D., Ph.D. Co-Founder and Co-Chair Michael C. Milone, M.D., Ph.D. Co-Founder and Co-Chair Carl June, M.D. Jay Siegel, M.D. Iain McInnes, Ph.D., FRCP, FRSE, FMedSci Georg Schett, M.D. David J. Chang, M.D., M.P.H., FACR Chief Medical Officer Samik Basu, M.D. Chief Scientific Officer Arun Das, M.D. Chief Business Officer Sarah Yuan Chief Technology Officer Steve Gavel Chief Commercial Officer Francisco Ramírez-Valle, M.D., Ph.D.

Appendix

TIS is a composite tool measuring a patient’s relative improvement from their baseline Myositis outcomes captured through validated composite endpoint Major Total Improvement Score 100 80 0 Minimal Moderate 60 40 20 Week 24 Baseline Week 12 None Extramuscular Activity 20 max achievable score Manual Muscle Testing (MMT8) 32.5 max achievable score Patient Global Activity 10 max achievable score Physician Global Activity 20 max achievable score Health Assessment Questionnaire (HAQ) 10 max achievable score Muscle Enzymes 7.5 max achievable score Total improvement score (TIS) components TIS developed via conjoint analysis based continuous model using absolute percentage change in 6 core set measures (CSM): MMT8, Extramuscular Activity, Physician Global Activity, Health Assessment Questionnaire, Patient Global Activity, and Muscle Enzymes TIS is the sum of improvement scores in the 6 CSMs, with ceiling of potential effect likely higher in DM and ASyS than in IMNM given minimal extramuscular involvement Illustrative patient-specific TIS ceiling for milder disease Illustrative patient-specific TIS ceiling for more severe disease ASyS – antisynthetase syndrome; CSM – core set measure; DM – dermatomyositis; IMNM – immune-mediated necrotizing myopathy; IVIg – intravenous immunoglobulin. Aggarwal R et al. NEJM. 2022;387(14):1264-1278.

Corporate Presentation June 2026