PRECISE-AD PMN310 · PHASE 1b BLINDED 6-MONTH INTERIM ANALYSIS Safety & Biomarker Assessment Update Blinded Interim Analysis Topline results expected Q1 2027 1

LEGAL DISCLOSURES Forward-Looking Statements & Disclaimers Blinded Interim Analysis Topline results expected Q1 2027 This slide deck may contain certain forward-looking information and “forward-looking statements” that are made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. Such information involves known and unknown risks, uncertainties and other factors that may cause actual results, performance or achievements to be materially different from those implied by statements herein, and therefore these statements should not be read as guarantees of future performance or results. Such forward-looking statements include, among others, statements pertaining to the ProMIS Neurosciences Inc.'s (the “Company”) PRECISE-AD Phase 1b clinical trial, target engagement and biomarker findings, the results and nature of the blinded interim clinical data of PMN310 and anticipated topline clinical data of PMN310, its mechanism of action and potential benefits and the Company's development plans and anticipated milestone timing, among other factors. Forward-looking information is based on a number of opinions, assumptions and estimates that, while considered reasonable by the Company as of the date of this slide deck, are subject to known and unknown risks, uncertainties and assumptions and other factors that may cause the actual results, level of activity, performance or achievements to be materially different from those expressed or implied by such forward-looking information, including, but not limited to, the risk that the results of early clinical trials are not necessarily predictive of future results with PMN310 and the Company’s ability to fund its operations. Important factors that could cause actual results to differ materially from those indicated in the forward-looking information include, among others, the factors discussed throughout the “Risk Factors” section of the Company's most recently filed Annual Report on Form 10-K for the year ended December 31, 2025 and in its subsequent filings filed with the United States Securities and Exchange Commission. Except as required by applicable securities laws, the Company undertakes no obligation to publicly update any forward-looking information, whether written or oral, that may be made from time to time, whether as a result of new information, future developments or otherwise. Data presented as of July 22, 2026 2

Table of Contents 3 Slides 4-5: Company Overview, KOL Bios Slides 6-11: PRECISE-AD Phase 1b Trial Design and Patient Demographics Slides 24-27: Summary and Conclusions Slides 12-16: Blinded Safety Assessment Slides 17-23: Blinded Biomarker Assessment

Presenters & KOL Participants PRESENTERS Neil Warma President & CEO ProMIS Neurosciences Dr. Larry Altstiel Chief Medical Officer (M.D., Ph.D.) ProMIS Neurosciences KEY OPINION LEADER Dr. Will Mantyh Behavioral Neurologist University of Minnesota Focused on early detection and diagnosis of Alzheimer's and related neurodegenerative diseases. Develops blood and imaging biomarkers to bring new blood tests to real-world and underrepresented patient populations. A tenured Associate Professor at the University of Minnesota, he leads a $3.7M NIH R01 bringing Alzheimer's blood tests to Native American communities. His honors include the AAN's Robert W. Katzman Award and the Fesler-Lampert Chair. KEY OPINION LEADER Dr. Michael Weiner Professor Emeritus, UCSF Principal Investigator, ADNI* Principal Investigator of ADNI, the world's largest observational Alzheimer's study, and founder of the Brain Health Registry. A pioneer in MRI/MRS development who helped bring nuclear magnetic resonance imaging into clinical use, he has published over 1,030 peer-reviewed articles. His honors include the Alzheimer's Association's Nancy and Ronald Reagan Award, the AAN's Potamkin Prize, and the Henry Wisniewski Lifetime Achievement Award (2021). *Alzheimer's Disease Neuroimaging 4

COMPANY SNAPSHOT Clinical-stage biotechnology company with a pipeline designed to selectively target specific, disease-causing misfolded proteins Unique selectivity may create potential to address the unmet need for safer, more efficacious therapies. PMN : NASDAQ Cambridge, MA TOP-LINE DATA Early Q1 2027 PRECISE-AD Phase 1b readout PEAK SALES POTENTIAL >$10B PMN310 in early Alzheimer's FINANCIAL Up to $175M raised Cash through 2027 · A-list syndicate LEADERSHIP Global development team Deep neuroscience domain experience DIFFERENTIATION Oligomer-selective by design Aim to reduce ARIA and improve clinical efficacy 5

TRIAL MILESTONES Execution on Track, Progressing Quickly Toward Topline Results DEC 2025 Enrollment n = 144 complete JUN 2026 6-Month Dosing complete EARLY Q3 2026 Interim Data this presentation DEC 2026 12-Month Dosing expected to complete for all patients EARLY Q1 2027 Topline Results final readout BLINDED 6-MONTH INTERIM · EARLY Q3 2026 Trial design & execution overview Safety observations & ARIA snapshot Directional trend on key biomarker UNBLINDED 12-MONTH TOPLINE RESULTS · EARLY 2027 Clinical endpoints: cognition change Active vs. placebo group comparisons Safety, biomarker & efficacy analysis Imaging results . Patient demographics 6

PRECISE-AD Study Design & Patient Demographics 7

PRECISE-AD: Phase 1b Trial Design KEY STUDY PARAMETERS & PATIENT DEMOGRAPHICS Study PRECISE-AD · Phase 1b Patient population Mild Cognitive Impairment (MCI) due to AD / early AD Total enrolled 144 patients Safety-evaluable 136 patients RANDOMIZATION · 3:1 DRUG TO PLACEBO Drug ~75% DOSE COHORTS · MULTIPLE ASCENDING DOSE 12 MONTHLY IV INFUSIONS Cohort 1 350 mg (5 mg/kg) Cohort 2 700 mg (10 mg/kg) Cohort 3 1400 mg (20 mg/kg) 75 136 patients Safety-Evaluable Drug (~75%) Placebo (~25%) Mean age 73.3 years Sex 58% F · 42% M Race White 69% · Hispanic 24% · Black 6% · Asian 1% APOE4 carriers 50% heterozygotes · 11% homozygotes Agent PMN310 - humanized IgG1 mAb TRIAL DEMOGRAPH ICS AND DESIGN 8

TRIAL EXECUTION & DEMOGRAPHICS Subjects Evaluated to Confirm Mild Cognitive Impairment (MCI) due to AD or Early AD CONFIRMATION CRITERIA Clinical Criteria NIA-AA criteria: MCI due to AD or mild AD dementia 100% Amyloid-Positive Confirmed on PET imaging Cognitive Staging MMSE* 20–30 & CDR Global 0.5–1.0 confirm early-stage AD Plasma Biomarker p-WDX$ȕ​UDWLR​SRVLWLYH​FRQVLVWHQW​ZLWK​$'​SDWKRORJ\ 9 *Mini-Mental State Examination, 30-point questionnaire

TRIAL DEMOGRAPHICS PRECISE-AD: Enrollment Consistent with Real World AD Population 11% APOE4/4 Homozygous Highest ARIA-risk subgroup with plaque-binding therapies 50% APOE4 Carrier (1 allele) E2/E4, E3/E4, E4/E3 and E4/E2 combined 61% Total APOE4 Carriers $Q\​SDUWLFLSDQW​ZLWK​•​$32(​DOOHOH​ WHY THIS MATTERS ض Representative population, including highest ARIA-risk patients Treatment options may open for ~15% of AD patients who are APOE4 homozygotes, the group current therapies largely exclude. APOE allele frequency: trial vs. reference AD population In Hardy-Weinberg Equilibrium Observed APOE genotype frequencies match those expected under Hardy-Weinberg equilibrium (p > 0.05), confirming a genetically representative sample with no enrollment selection bias. Source: AD population genotype percentage and allele frequencies; Yamazaki et al. Nat Rev Neurol 2019 PRECISE-AD PATIENT POPULATION ض PRECISE-AD 10

SAFETY Many Patients are Currently Well Beyond 6 Months of Treatment TREATMENT EXPOSURE TO DATE 0 (initiation) 6 mo 9 mo 12 mo • All patients have been treated for at least 6 months. Many (~49%) have completed all 12 doses • 144 subjects were enrolled, 136 have been included in this safety analysis • Safety data is reported as of July 22, 2026 for all patients • Biomarker data presented through the 6-month time point MRI is performed at baseline, 2, 4, 6, 9 and 12 months per protocol; ARIA incidence reflects all scans completed as of the interim cutoff. Follow-up distribution shown; illustrative of assessment maturity. n=136 (100%) of patients are past 6 months of dosing n=107 (78%) of patients are past 9 months of dosing n=67 (49%) of patients have completed the full 12-month trial 11

12 Blinded Safety Assessment

INTERIM SAFETY · 6-MONTH BLINDED Favorable Safety Data, Observed Across Key Measures 6-month blinded interim · N = 144 dosed, 136 safety-evaluable · a representative population 0.0% ARIA-E The most severe treatment driven form of ARIA No treatment-related serious AEs Across all genotypes No treatment-related discontinuations Low overall dropout rate to date 4.4% total ARIA In-line with placebo-range ARIA-H, all mild, asymptomatic, non-serious Minimal infusion reactions A recognized liability of approved anti-amyloids and brain-shuttle candidates. PRECISE-AD: one non-serious event, non-systemic, dosing continued. PMN310 <1% 1 event, non-serious Donanemab ~9% Lecanemab ~26% APOE4 carriers included 61% FDUU\​•​$32(​DOOHOH 11% İİ​KRPR]\JRWHV​— the highest ARIA-risk group ض No ARIA-E in any genotype ARIA detection using Susceptibility-Weighted Imaging (SWI) for increased sensitivity. 6-month blinded. Infusion-reaction comparators: lecanemab 26.4% (CLARITY AD, van Dyck NEJM 2023); donanemab 8.7% (TRAILBLAZER-ALZ 2, Sims JAMA 2023). Descriptive comparison only. No head-to-head studies have been conducted to compare PMN310 with approved products or other product candidates in development. 13

TIME-MATCHED SAFETY Cumulative ARIA-E Over Time: Zero Edema Observed First 6 Months ARIA-E (the most serious ARIA type) has been observed to occur almost entirely in the first six months of dosing. Comparator ARIA-E plateaus after the first few months; rates are time-matched to published data where available, with ~ indicating values estimated between published time points. AT 6 MONTHS Time-matched to estimated or published 6-month rates Donanemab 23.7% Lecanemab ~12% Historic, third-party placebo data (avg across both trials) ~0.6%* PRECISE-AD 0% Illustrative cross-trial comparison — not head-to-head; ~ denotes values estimated between published time points. * Placebo: 6-mo ARIA-E (~0.6%) is ProMIS-calculated by linear accrual — (6/18) × ~1.9%, the average of the two trials’ 18-mo placebo rates (1.7% and 2.1%). Lecanemab: 12.6% treated / 1.7% placebo ARIA-E at 18 mo; 6-mo (~12%) approximated from this front-loaded cumulative (van Dyck et al., NEJM 2023;388:9– 21). Donanemab, TRAILBLAZER-ALZ 2: 24.0% treated / 2.1% placebo over 76 wk (Sims et al., JAMA 2023;330:512–527). Donanemab standard regimen, TRAILBLAZER-ALZ 6: 23.7% ARIA-E at wk 24 (6 mo), 24.2% at 76 wk (Wang et al., Alzheimer’s & Dementia 2025;21:e70062); pooled standard-dosing 18-mo ~24.4%. PRECISE-AD: blinded pooled cohort (active + placebo, 3:1); 0/136 ARIA-E at 6-mo interim — observed, not projected. Donanemab, 23.7% 24.4% Lecanemab, ~12% 12.6% Historic Placebo ~0.6% ~1.9% PRECISE-AD, 0% 0% 10% 20% 30% 0 6 12 18 Cumulative ARIA-E incidence (%) Months on study 6-months 14

TIME-MATCHED SAFETY Cumulative Total ARIA Over Time: Blinded PRECISE-AD Data Tracks the Placebo Range Reported in Other Trials Approved anti-amyloids accrue ARIA steeply in the first months of dosing, with 90%+ ARIA-E occurring in the first 6 months. Comparator rates are time-matched to published data where available, with ~ indicating values estimated between published time points. Donanemab-trial placebo Lecanemab-trial placebo PLACEBO RANGE 14.2% 9.3% AT 6 MONTHS * Calculated based on published data Donanemab *~31% Lecanemab *~16% Historic, third-party placebo data (both trials) *~3-5% PRECISE-AD 4.4% MODELED SHAPE, ACTUAL ENDPOINTS — comparator and placebo curves are reconstructed from published trial time-course; trial-end values are as reported. Illustrative, not a head-to-head comparison. Any ARIA (ARIA-E and/or ARIA-H): Lecanemab 21.5% / 9.3% placebo at 18 mo (van Dyck, NEJM 2023); Donanemab 37.0% / 14.2% placebo, TRAILBLAZER-ALZ 2 ~76 wk (Sims, JAMA 2023; Zimmer, JAMA Neurol 2025). 6-mo points calculated from the published time-course (>90% of ARIA-E within 6 mo): Donanemab ~31%, Lecanemab ~16%, placebo ~3–5%; placebo accrues linearly to reported endpoints. PRECISE-AD: blinded pooled cohort (active + placebo, 3:1); 4.4% total ARIA (6/136) at 6-mo interim — observed, not projected. 15 0% 10% 20% 30% 40% 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Months on study · cumulative total ARIA incidence (%) Donanemab 37.0% Lecanemab 21.5% PRECISE-AD 4.4% observed interim — no projection

ARIA risk rises sharply with APOE4 carrier status. Approved drugs carry boxed warnings and the worst risk/benefit profile, especially for homozygotes, who represent ~15% of patients. ARIA-E rates at 6 months by APOE4 genotype Genotype Lecanemab* PMN310 Noncarrier ~5% 0% APOE4 Heterozygote ~10% 0% APOE4 Homozygote ~30% 0% ~61% of the PRECISE-AD population were APOE4 carriers. No ARIA-E events were reported in any study participants to date. A low ARIA-E profile across genotypes could help address a significant barrier to treating higher risk patients. *Lecanemab data (van Dyck CH, Sperling R, Johnson K, et al. Long-term safety and efficacy of lecanemab in early Alzheimer's disease: Results from the clarity AD open-label extension study. Alzheimer's Dement. 2025; 21:e70905); 16 APOE4: Higher-risk Patient Population for ARIA-E APOE4 GENOTYPE · CLASS DIFFERENTIATION

12 Blinded Biomarker Assessment

6-MONTH BLINDED INTERIM ANALYSIS Two complementary biomarkers: plasma pTau217 and CSF MTBR-tau243 Presenting 6-month blinded interim data on two biomarkers chosen to bracket the disease cascade. UPSTREAM · PLASMA · AMYLOID-DRIVEN Plasma pTau217 The earliest-moving, best-validated plasma marker of AD pathology. It rises before clinical change, predicts 12-month clinical outcome, and is a key plasma biomarker in current diagnostic criteria, potentially the marker most likely to show an early drug effect by 6 months. DOWNSTREAM · CSF · TANGLE-SPECIFIC CSF MTBR-tau243 A CSF marker specific for insoluble tau tangles, the pathology most tightly linked to tau-PET and cognitive decline. It captures the downstream, disease-driving process that amyloid-oriented markers do not. Rationale Bracketing the disease cascade triggered by amyloid-beta oligomers: pTau217 reports upstream, amyloid-beta-driven tau phosphorylation; MTBR-tau243 reports the ensuing downstream tangle accumulation. A favorable move in both is far stronger evidence of a potential drug effect than either alone. 17

BIOMARKER RATIONALE pTau217: early blood-based readout of AD biology What it measures Disease progression: a downstream marker of amyloid-positive AD pathology and tau pathway activation. Why it matters In untreated (placebo) patients, pTau217 rises ~6% over 18 months. A decline in pTau217 would be an encouraging sign, indicating biological / pharmacodynamic activity and is potentially supportive of disease modification. 6-month pTau217 is a sensitive early signal, not a standalone efficacy claim. Potential to inform dose selection and bridge to later clinical outcomes. Refs: Pontecorvo et al., JAMA Neurology 2022 (TRAILBLAZER-ALZ Ph2 placebo-arm natural history); Leveraging recent advances in plasma biomarkers to optimize early proof of concept trials in Alzheimer's disease." Alzheimer's & Dementia: Translational Research & Clinical Interventions (TRC), 2025. DOI 10.1002/trc2.70183. Predictive ability Recent Pentara / ProMIS analysis found 6-month plasma pTau treatment effects correlated with later CDR-SB effects and showed ~2.6× larger effect size than CDR-SB. 18 6% 0% 10% 20% 30% Baseline 6 mo 12 mo 18 mo pTau217 (% change from baseline) Months on treatment Representative Natural History AD population plasma pTau217 rises from baseline Placebo arm — donanemab Phase 2 (TRAILBLAZER-ALZ), ~76 weeks; Pontecorvo et al., JAMA Neurology 2022. Interim time points illustrative.

6-MONTH BLINDED ANALYSIS · PLASMA PRECISE-AD RESULTS: Plasma pTau217 declined steadily through Day 169 PRECISE-AD Mean % change from baseline over time -25 -20 -15 -10 -5 0 5 0 29 85 141 169 % change from baseline Study day Linear trend PRECISE-AD patients with a day-169 decline 68.5% 31.5% 0 20 40 60 80 ĞĐůŝŶĞĚ​;чϬͿ Increased (>0) (Improved) Why this is compelling at 6 months Untreated pTau217 rises as the disease progresses. The 15% decline shown in the PRECISE-AD data (including both PMN and placebo treated patients) indicates a possible early treatment-associated reversal. Although responder data remains blinded, the ~68% responder fraction closely aligns with the 75% active allocation in the 3:1 design based on a prior ProMIS analysis. A 6-month pT217 change could be an early leading indicator of benefit and a predictor of 12-month clinical outcome (Pentara/ProMIS). Randomization 3:1 Aggregated drug + placebo (3:1) 19 (Worsened) Ref: “Leveraging recent advances in plasma biomarkers to optimize early proof of concept trials in Alzheimer’s disease.” Alzheimer’s & Dementia: Translational Research & Clinical Interventions (TRCI), 2025. DOI 10.1002/trc2.70183. -15%

BIOMARKER RATIONALE CSF MTBR-tau243: tangle-specific marker of AD pathology What it measures A CSF marker of tauopathy, the pathology most closely tied to symptoms. Why it’s strong Correlates with tau-tangle burden and cognition comparable to tau-PET, and more strongly than other CSF markers. Tau pathology tends to become more evident as patients become symptomatic. Why it tracks change Tau tangles correlate with cognitive decline. MTRB-tau243 rises as tangles accumulate and moves in step with disease progression over time. This biomarker tends to respond more slowly than pTau217. 1 Horie et al., Nature Medicine 2023; BioFINDER-2 n = 448; Knight ADRC n = 219. 20 Natural history progression What the published evidence supports Time matched longitudinal data for CSF MTBR-tau243 is limited but has been shown to steeply increase as the disease progresses: • As a tangle-specific marker, it is elevated at more advanced disease stages.1 • Closely related tau markers, notably p-tau217, increase measurably over time across the AD continuum. In untreated patients, MTBR-tau243 is therefore expected to trend upward. A downward trend with treatment would indicate a possible drug effect.

6-MONTH BLINDED ANALYSIS · CSF PRECISE-AD RESULTS: CSF MTBR-tau243 declined steadily through Day 169 PRECISE-$'​0HDQ​​FKDQJH​VFUHHQ​ĺ​'D\​ -25 -20 -15 -10 -5 0 5 Screen Day 169 % change from baseline Timepoint PRECISE-AD: Patients with a Day-169 decline 62.5% 37.5% 0 20 40 60 80 Declined Increased Why this matters at 6 months MTBR-tau243 is tangle-specific and normally rises as aggregates accumulate, so a downward move is a favorable, disease-relevant signal. Across the trial participants, it declined on average (-13.3%) and in the majority of patients. The effect is early, not yet powered for significance, but its direction, against a rising natural history, is the meaningful result and is directionally consistent with the upstream plasma pT217 decline and tracks the 3:1 randomization pattern. Aggregated drug + placebo (3:1) 21 (Improved) (Worsened) -13.3% Randomization 3:1

BIOMARKER SUMMARY Two biomarkers, one coherent 6-month signal UPSTREAM · PLASMA pT217 declined through Day 169 A steady overall decline of 15% vs baseline with ~68% of all patients showing a reduction, indicating potential improvement. This tracks the 75% (3:1) randomization pattern, in a blinded analysis. A meaningful reversal versus a rising natural history. DOWNSTREAM · CSF MTBR-tau243 declined through Day 169 This tangle-specific marker declined by 13.3% in the blinded and aggregated analysis with ~62% of all patients showing a reduction, indicating potential improvement. This closely aligns with the 3:1 randomization. This is an early signal, but opposite the usual biomarker increase expected in AD progression. The takeaway Upstream (amyloid-beta-driven pT217) and downstream (tangle-specific MTBR-tau243) markers both moved in a favorable direction at 6 months, against a natural history of rising levels. Although data remain blinded, this suggests an early, biologically coherent signal suggesting disease modification. 22

23 Summary & Conclusions

CONCLUSION A Differentiated, Precision Approach to Alzheimer’s — With a Clear Path Forward PRECISE-AD’s blinded interim analysis shows favorable safety data and early, mechanism-consistent biomarker signals, supporting a differentiated approach for patients underserved by approved anti-amyloid therapies. 01 DIFFERENTIATED MECHANISM Oligomer-selective by design. PMN310 is engineered to bind toxic amyloid-ȕ​ROLJRPHUV​ZKLOH​avoiding plaque, decoupling efficacy from the ARIA liability that defines the approved class. 02 FAVORABLE SAFETY PROFILE Zero ARIA-E; 4.4% total ARIA — all mild, asymptomatic ARIA-H. Zero treatment-related SAEs and zero drug-related discontinuations, minimal (n=1) drug-related infusion reactions as of data cutoff date. 03 TARGET ENGAGEMENT WITH RELEVANT BIOMARKERS Coherent, biologically aligned signal. Upstream plasma pTau217 declined significantly (~69% of patients, tracking 3:1 randomization) and downstream CSF MTBR-tau243 moved favorably (~63% of patients declined), both opposite a rising natural history. 04 OPENING ACCESS TO APOE4 CARRIERS No ARIA-E in any genotype. With 61% APOE4 carriers and 11% homozygotes enrolled, a favorable safety profile could remove the class’s single biggest barrier, especially for the patients at highest risk. 05 POTENTIAL IN PRECLINICAL AD A safety profile suited to earlier intervention. A placebo-level ARIA profile makes PMN310 a strong candidate to move upstream into preclinical AD, where prevention has the greatest impact, but tolerability is paramount. 24

Topline results early Q1 2027 Favorable safety profile No ARIA-E, zero SAEs — across all genotypes. Early biomarker signals Directional trends consistent with potential target engagement. Oligomer-selective MOA Designed to decouple efficacy from ARIA liability with precision selectivity. PMN : NASDAQ · Cambridge, MA 25

NASDAQ: PMN For further information, contact: info@promisneurosciences.com 26