RCUS Q4 2025 Earnings Presentation

Page 1

Full Year 2025 Investor Presentation Oncology and Immunology Portfolio Summary February 25, 2026

Page 2

© Arcus Biosciences 2026 Forward-Looking Statements Safe Harbor: This presentation contains forward-looking statements about Arcus Biosciences, Inc. (“we,” “Arcus” or the “Company”) made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. All statements in this presentation that are not historical facts are forward-looking statements, including, without limitation, statements regarding: our anticipated cash runway (including our expectation of funding until at least the second half of 2028); our strategies, goals, opportunities, and potential advantages for our programs, including casdatifan and our inflammation and immunology programs; estimates of market potential or peak sales for our investigational products; and the timing, initiation, enrollment, advancement, conduct, and results of our clinical trials and other development activities (including, without limitation, expected timing for completion of enrollment for PEAK-1, initiation of a potential Phase 3 study in first- or early-line RCC, advancement of our inflammation and immunology programs into the clinic, and timing and results from ongoing clinical studies). Forward-looking statements are often identified by words such as “anticipate,” “believe,” “expect,” “intend,” “may,” “should,” “plan,” “project,” “target,” “will,” and similar expressions, although not all forward-looking statements contain these identifying words. These forward-looking statements are subject to numerous risks, uncertainties and assumptions that may cause actual results to differ materially from those expressed or implied by any forward-looking statements, including, but not limited to: the unexpected emergence of adverse events or other undesirable side effects with Arcus’s investigational products, including casdatifan; risks associated with interim or preliminary clinical data not being replicated in other studies evaluating the same investigational product, including PEAK-1 for casdatifan; difficulties or delays in conducting or completing our clinical trials due to regulatory review, site activation, patient identification or enrollment, or manufacturing and supply constraints of investigational or standard-of-care products for such clinical trials, all of which may be exacerbated by unfavorable global economic, political, public health and trade conditions; changes to Arcus’s cash runway due to changes in Arcus’s operating plans and expected resource allocation, including for domvanalimab; changes in the competitive landscape; our ability to obtain and maintain intellectual property protection for our product candidates; and the inherent uncertainty associated with pharmaceutical product development and clinical trials. It is not possible for our management to predict all risks, nor can we assess the impact of all factors on our business or the extent to which any factor, or combination of factors, may cause actual results to differ materially from those anticipated or implied in the forward-looking statements. Further information on these and other factors that could affect the forward-looking statements made herein is described in our most recent periodic reports filed with the U.S. Securities and Exchange Commission, including our most recent Annual Report on Form 10-K, and in other filings and reports we make with the SEC from time to time. You should not rely upon forward-looking statements as predictions of future events. Except as required by law, neither we nor any other person assumes responsibility for the accuracy or completeness of the forward-looking statements. We undertake no obligation to update any forward-looking statements for any reason after the date of this presentation to conform these statements to actual results or to changes in our expectations, except as required by law. Important Information Regarding Data Comparisons: This presentation includes comparisons between data from our Phase 1/1b ARC-20 trial and published data from separate trials that are not head-to-head studies. Cross-trial comparisons should be interpreted with caution due to differences in study populations, sample sizes, inclusion and exclusion criteria, trial design, dosing, endpoints, follow-up, and other factors that may limit direct comparability. Third-Party Sources Disclaimer: Additionally, this presentation contains certain information related to or based on studies, publications, surveys and other data obtained from third-party sources, and our own internal estimates and research, including without limitation relating to market size and potential sales or revenue opportunity. This information is based on a number of assumptions, projections and estimates, including with respect to our future performance and the future performance of markets in which we operate, and is necessarily subject to a high degree of uncertainty and risk, and you are cautioned not to place undue reliance onto to such estimates. No Regulatory Approval: All of Arcus’s molecules are investigational and Arcus (and Gilead for all of the molecules in each optioned program) has not received approval from any regulatory authority for any use globally, nor established the safety and efficacy of these investigational molecules. Notice of Trademarks: The Arcus name and logo are the property of Arcus. All other trademarks used herein are the property of their respective owners and are used for reference purposes only. Such use should not be construed as an endorsement of Arcus. 2 Forward-looking Statements/Safe Harbor

Page 3

© Arcus Biosciences 2026 3 Arcus Has Created a Robust Portfolio of Potential Best-in- Class Medicines * Cash, cash equivalents and marketable securities as of December 31, 2025 ** Runway estimate based on cash, cash equivalents, marketable securities, and current planned operations FOUNDED 2015 R&D ENGINE World class medicinal chemistry Goal of 1-2 new molecules advancing to the clinic each year 1st immunology molecules should enter the clinic in 2026 Addressing most common diseases PancreaticKidney Urticaria RAIBD AD Psoriasis~$1B IN CASH* CAPITAL EFFICIENCY & RISK SHARING HAVE BEEN GROWTH ENABLING LATE-STAGE ONCOLOGY Casdatifan Quemliclustat MRGPRX2 TNF CCR6 & IMMUNOLOGY CD89 CD40L Funded until at least 2H:28** Cancers with Large Patient Populations Inflammation & Autoimmune Diseases

Page 4

© Arcus Biosciences 2026 4 Leveraging our Small Molecule Capabilities to Create Best-in- Class Oral Medicines *Arcus has world-wide rights including the US, Europe and China; Taiho has rights in Japan and other Asian territories Wholly-owned* HIF-2α inhibitor with "best in class" data; target validated clinically and commercially in RCC CASDATIFAN 2 PHASE 3 PROGRAMS IN ONCOLOGY EXPANDING ORAL SMALL MOLECULE I&I PORTFOLIO Expected in clinic in 2026 MRGPRX2 antagonist Enormous Opportunity to Displace Injectable, Blockbuster Biologic Drugs NEW Phase 3 Trial in 1L Targeted for YE:26 TNF inhibitor Expected in clinic late ‘26 / early ‘27Only small molecule CD73 inhibitor in clinical development; In Ph 3 for 1L pancreatic cancer QUEMLI Urticaria Atopic Dermatitis RAIBD Psoriasis Phase 3 enrolling

Page 5

© Arcus Biosciences 2026 Key Updates: New Casdatifan Data and Emerging I&I Portfolio IMPRESSIVE CASDATIFAN DATA IN LATE-LINE ccRCC TO BE PRESENTED AT ASCO GU With longer follow up, data have improved even further (Jan. 30, 2026 DCO1) •cORR increased to 45% with a mPFS of 15.1 months for the 100mg QD tablet cohort (going forward dose and formulation) •cORR increased to 35% and PFS was stable at 12.2 months in a pooled analysis of all four monotherapy cohorts (n=121) •New biomarker data demonstrated that greater EPO suppression with cas is associated with clinical benefit, including ORR and PFS Data support our ongoing, first phase 3 study, PEAK-1, and the initiation of our first Phase 3 in the 1L setting by year-end • Arcus has had a 2+ year discovery effort focused on I&I which has now yielded 5 active programs • The most advanced programs are small molecules against MRGPRX2 and TNF; initiation of first-in-human studies for MRGPRX2 and TNF expected in 2026 and late 2026 / early 2027, respectively DISCLOSURE OF EMERGING I&I PROGRAMS • Winding down Phase 3 STAR-221 and Phase 2 EDGE-Gastric studies • A futility analysis of Phase 3 STAR-121 will be conducted in the next couple of months to determine next steps for the program. DOMVANALIMAB PROGRAM

Page 6

© Arcus Biosciences 2026 6 Arcus Milestones Will Generate Near- and Long-Term Value for Shareholders  Cas monotherapy data at ASCO GU Cas + cabo PFS data Data for cas combinations in early-line RCC POC data for MRGPRX2 antagonistInitiate first Phase 3 study for cas in 1L RCC MRGPRX2 antagonist in clinic Phase 3 data TNF inhibitor in clinic Phase 3 data NEXT 3 MONTHS NEXT 6-9 MONTHS NEXT 9-12 MONTHS NEXT 12-24 MONTHS NEXT 24 MONTHS

Page 7

© Arcus Biosciences 2026 SUPERIOR PHARMACO- DYNAMIC PROFILE4,5 7 Clinical Data from >120 Patients Support Casdatifan’s Potential as the Best-in-Class HIF-2α Inhibitor Patient and sales opportunity based on drug treatable population in major markets; from Decision Resources Group and Arcus analy sis. Data are not from head-to-head studies. Cross-trial data interpretation should be considered with caution as it is limited by di fferences in study population, sample size, inclusion and exclusion criteria and many other factors. Longer PFS than other HIF-2α inhibitors and TKIs in late-line kidney cancer Median PFS, months (95% CI) Median follow-up (range) 12.2 (9.4, 16.5) 20.8 (11.8, 31.5) 12.2 months mPFS compares to 5.6 mos PFS for belzutifan in LITESPARK- 005 2,3 Top Priority is to Establish Casdatifan as the SOC Across Multiple Settings of RCC ~$2B market potential Post-IO metastatic cas + cabo | Phase 3 1L TKI-free regimen ~$3B market potential 19K patients 21K patients 100mg QD of cas results in substantial and sustained EPO suppression 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 -100 -80 -60 -40 -20 0 20 Study Week Cas 100mg QD (N=31) Belz 120mg QD IMPROVED CLINICAL OUTCOMES1

Page 8

Page 9

© Arcus Biosciences 2026 Cohorts Evaluating Cas Mono in anti-PD-1/TKI Experienced ccRCC 9 2L+ ccRCC Casdatifan mono 50mg BID capsule 2L+ ccRCC Casdatifan mono 50mg QD capsule 2L+ ccRCC Casdatifan mono 150mg QD tablet 2L+ ccRCC Casdatifan mono 100mg QD tablet Favorable-risk 1L ccRCC Casdatifan mono 100mg QD 1L ccRCC Casdatifan 100mg QD + Zimberelimab 360mg Q3W Post-IO ccRCC Casdatifan mono 100mg QD Post-IO ccRCC Casdatifan 100mg QD + Cabozantinib 60mg QD Casdatifan monotherapy DOSE EXPANSION N = ~30 per cohort DOSE ESCALATION Patients with advanced solid tumors 150mg QD 50mg BID 50mg QD 20mg QD 200mg QD To date, over 240 patients have received casdatifan across all cohorts of ARC-20

Page 10

© Arcus Biosciences 202610 Confirmed ORR for the “Pooled” Cohort (n=121) is Now 35% Efficacy-Evaluable Populationa Dose Expansion: 2L+ ccRCC Belzutifan2 50mg BID (n = 31) 50mg QD (n = 28) 100mg QD (n=31) 150mg QD (n = 31) All Pooled (n = 121) 120mg QD (n = 374) Median Follow-Up, mos (range) 28.3 (19.9, 31.5) 25.2 (21.4, 26.7) 17.9 (11.8, 19.0) 19.8 (18.0, 21.0) 20.8 (11.8, 31.5) 18.4 (9.4–31.7) Median Time to Response, mos 2.7 4.1 2.6 2.7 2.8 3.8 Confirmed ORR (95% CI) 26% (12, 45) 36% (19, 56) 45% (27, 64) 32% (17, 51) 35% (26, 44) 22% (18, 27) Complete Response, % (n) 0% (0) 4% (1) 0% (0) 0% (0) 1% (1) 3% (10) Partial Response, % (n) 26% (8) 32% (9) 45% (14) 32% (10) 34% (41) 19% (72) Stable Disease, % (n) 55% (17) 50% (14) 39% (12) 42% (13) 46% (56) 39% (147) Progressive Disease, % (n) 19% (6) 14% (4) 16% (5)* 26% (8) 19% (23)* 34% (126) *Includes two patients in the 100mg QD Tablet cohort, and also included in the pooled analysis, who had clinical progression before the first scan and therefore did not meet criteria for progressive disease per RECIST. PD based on RECIST criteria was 10% (n=3) for the 100mg QD Tablet cohort and 17% (n=121) DCO date: Jan. 30, 2026 a. Efficacy-evaluable population for this expansion cohort is defined as all eligible participants who received any study treatm ent and have at least one post-baseline efficacy assessment, or who discontinued study treatment due to progressive disease or death, regardless of whether they had a scan.

Page 11

© Arcus Biosciences 2026 100mg QD (n=32) 5 mos med. Follow-up cORR: 33% mPFS: Not reached PD: 15% 12 mos med. Follow-up cORR: 35% mPFS: Not reached PD: 16%** 17.9 mos med. Follow-up cORR: 45% mPFS: 15.1mos PD: 16%** Pooled Analysis (n=121)* -- 15 mos med. Follow-up cORR: 31% mPFS: 12.2 months PD: 19%** 20.8 mos med. Follow-up cORR: 35% mPFS: 12.2 mos PD: 19%** 11 * Includes all four late-line monotherapy cohorts: 50mg BID, 50mg QD, 100mg QD, 150mg QD from the ARC-20 study ** Includes two patients in the 100mg QD Tablet cohort, and also included in the pooled analysis, who had clinical progression before the first scan and therefore did not meet criteria for progressive disease per RECIST. PD based on RECIST criteria was 10% (n=3) for the 100mg QD Tablet cohort and 17% (n=121). Oct.2025 Investor Event Data Have Been Consistently Robust and Have Improved Over Time, Exceeding Benchmarks ASCO GU 2025 DCO: January 3, 2025 Oct. 6 Arcus IR Event DCO: August 15, 2025 ASCO GU 2026 DCO: Jan. 30, 2026

Page 12

© Arcus Biosciences 2026 12 15.1 Months mPFS for Patients Treated with Casdatifan Phase 3 Dose and Formulation DCO: Jan. 30, 2026 Median Follow Up: 17.9 months In the unlikely scenario that all censored patients (still on therapy) progress at the next scan, the mPFS would only decline to 14.4 months

Page 13

© Arcus Biosciences 2026 13 12.2 Months mPFS for Patients Treated with Casdatifan Across All Four Monotherapy Cohorts DCO: Jan. 30, 2026 a.The efficacy evaluable population (N = 121) was defined as all eligible patients who received any study treatment and had ≥ 1 pos t-baseline efficacy assessment or discontinued study treatment due to progressive disease or death. Median Follow Up: 20.8 months Substantial number of patients remain on treatment beyond 14 months and 24 months

Page 14

© Arcus Biosciences 2026 14 Systemic Suppression of HIF-2α–Regulated sEPO Is Highly Correlated with Clinical Benefit with Casdatifan (ORR) Normal EPO range **** n = 58 N = 1292 4 8 16 32 64 128 256 512 Healthy volunteer ARC-20 (ccRCC) Baseline sEPO (mlU/mL) Patients with ccRCC had significantly higher baseline levels of sEPO than healthy volunteers (P < .0001) 1 0 −20 −40 −60 −80 −100 0 25 50 75 100 125 150 175 200 T otal daily casdatifan dose (mg) Pharmacodynamic response (% EPO change from baseline) 71% (92/129) of patients reached maximal sEPO reduction during the first cycle of treatment with casdatifan 1 Across all doses, most patients (88/129 [68%]) experienced > 80% maximal sEPO reduction; Deeper reductions in sEPO were highly correlated with clinical responses1 a. PD was defined according to RECIST v1.1 as a ≥ 20% increase in the sum of diameters of target lesions relative to the smal lest sum on study (including baseline, if smallest), with an absolute increase of ≥ 5 mm. The appearance of 1 or more new lesions was also considered progression.

Page 15

© Arcus Biosciences 2026 15 Systemic Suppression of HIF-2α–Regulated sEPO Is Highly Correlated with Clinical Benefit with Casdatifan (PFS) The degree of maximal sEPO reduction was significantly greater in patients who achieved CR, PR, or SD compared with those with PD (P < .05) 1 Maximal sEPO reduction was associated with longer PFS1 a. PD was defined according to RECIST v1.1 as a ≥ 20% increase in the sum of diameters of target lesions relative to the smal lest sum on study (including baseline, if smallest), with an absolute increase of ≥ 5 mm. The appearance of 1 or more new lesions was also considered progression.

Page 16

© Arcus Biosciences 2026 SETTING TRIAL REGIMEN cORR mPFS 3L+ ARC-20 (Phase 1b/2) Cas (100mg QD) Cas (Pooled) 45% 35% 15.1m 12.2m LITESPARK-005 (Phase 3) 3 Belz 22% 5.6m TIVO-3 (Phase 3)6 Tivo 18% 5.6m 2L+ METEOR (Phase 3)7 Cabo 17% 7.4m Lenva + ev (Phase 2) 8 Lenva 27% 7.4m IO- Experienced (1L/2L) AXIS (Phase 3) 9 Axi 19% 6.7m CONTACT-03 (Phase 3)10 Cabo 41% 10.8m CANTATA (Phase 3) 11 Cabo 28% 9.3m 16 Casdatifan Data Exceed All PFS Benchmarks for Both Monotherapy HIF-2α Inhibition (Belz) and TKIs DCO date for casdatifan: Jan. 30, 2026 Data are not from head-to-head studies. Cross-trial data interpretation should be considered with caution as it is limited by differences in study population, sample size, inclusion and exclusion criteria and many other factors. EARLIER-LINE PATIENT POPULATIONS

Page 17

© Arcus Biosciences 2026 17 Casdatifan Was Well Tolerated in All Cohorts, With a Comparable Safety Profile to That of Belzutifan DCO date for casdatifan: August 15, 2025; safety profile remained consistent at the Jan. 30, 2026 DCO Data are not from head-to-head studies. Cross-trial data interpretation should be considered with caution as it is limited by differences in study population, sample size, inclusion and exclusion criteria and many other factors. Safety-Evaluable Population 100mg QD (n=32) Pooled (n=127) Belzutifan (LITESPARK- 005)2 Anemia, n (%) Anemia: All grades 29 (91) 117 (92) All grade: 83% Grade ≥3 related to casdatifan 8 (25) 52 (41) Grade ≥3: 33% Related to casdatifan leading to interruptions 9 (28) 45 (35) Leading to dose reductions 3 (9) 18 (14) Leading to discontin. 0 0 Hypoxia, n (%) Hypoxia: All grades 5 (16) 23 (18) All Grade: 15% Grade ≥3 related to casdatifan 3 (9) 14 (11) Grade ≥3: 11% Related to casdatifan leading to interruptions 3 (9) 18 (14) Leading to dose reductions 1 (3) 9 (7) Leading to discontin. 1 (3) 3 (2)

Page 18

Page 19

© Arcus Biosciences 2026 First Phase 3 Study Evaluating a Differentiated TKI Combination in Post-IO ccRCC is Enrolling 19 PRIMARY ENDPOINT • PFS KEY SECONDARY ENDPOINTS • OS • ORR, DOR, DCR PATIENT POPULATION: • Unresectable, locally advanced or metastatic ccRCC • Have had prior anti-PD- 1/ PD-L1 (either in adjuvant or 1L metastatic setting) • Have not received cabozantinib 100MG CASDATIFAN + CABOZANTINIB PLACEBO + CABOZANTINIB R 2:1 N~700 Goal Is to Fully Enroll by Year-End Only Phase 3 study evaluating a HIF-2α inhibitor with cabozantinib, the most widely used TKI in ccRCC

Page 20

© Arcus Biosciences 2026 20 1L Strategy for Cas Will Focus on TKI-Free Regimens, Enabled by its Low Rate of Primary Progression Primary PD rate: The percent of patients whose disease progressed at or before their first post -baseline scan. *All data from ongoing ARC-20 cohorts with DCO dates as follows: a. Jan. 30, 2026; b. March 14, 2025; c. Spotfire data as of January 7, 2026, includes patients who had received at least one scan at time of DCO. Efficacy-Evaluable Population* Primary PD Rate % (n) cas 100mg QD (n=31)a 16% (5)1 Monotherapy pooled (n=121) a 19% (23) cas + cabo (n=24)b 4% (1) 1L cas + zim (n=23)c 9% (2) 1L cas mono, favorable risk (n=14) c 0% (0) cas mono, post-IO, TKI naive (n=27)c 7% (2) 2L+ ccRCC 1L+ ccRCC 1L ccRCC cas + volrustomig (anti-PD-1/CTLA-4 bispecific) 1L ccRCC cas + zim (anti -PD-1) 1L ccRCC cas + anti-PD-1 + ipilimumab ARC-20 Data To Date Show a Consistently Low Rate of Primary PD, Avoiding the Need for a TKI 1L ccRCC cas-based TKI-free combination Four TKI-Free Cohorts in the 1L Setting Will Inform 1L Strategy for Cas with the Goal of Initiating Phase 3 Study by YE:26 Ongoing Plan to open Enrollment completed Enrolling Initiation planned 2026

Page 21

Page 22

© Arcus Biosciences 2026 22 RCC Is a Growing Multi-Billion Dollar Market, Driven by the Introduction of HIF-2α Inhibition and Long DoT's Source: DRG | 7 Major Markets = US, EU5 (France, Germany, Italy, Spain, and the United Kingdom), Japan $- $2,000 $4,000 $6,000 $8,000 $10,000 $12,000 $14,000 2024 2025 2026 2027 2028 2029 2030 Annual RCC Sales – US ($M) Other TKI IO $- $2,000 $4,000 $6,000 $8,000 $10,000 $12,000 $14,000 2024 2025 2026 2027 2028 2029 2030 Annual RCC Sales – 7 Major Markets ($M) Other TKI IO • RCC is currently a ~$9B market in Major Markets − This is despite multiple generic TKIs (e.g., axitinib, pazopanib, everolimus, sunitinib) − The market is almost entirely dominated by 2 classes of therapy: IO (5 approved agents) and TKI (7+ approved agents) − Each class has approx. $4B in sales • The RCC market is expected to grow to $13B by 2030 driven by: − Increasing patient population − Introduction of HIF-2α MOA − Increasing DoT and “IO-like” durations of response for HIF-2α-containing regimens • There are only 2 agents currently in Phase 3 development in the HIF-2α class

Page 23

© Arcus Biosciences 2026 $88012 ~ $2,500 ~ $2,800 $- $500 $1,000 $1,500 $2,000 $2,500 $3,000 Annualized Belz Run Rate (3L, 5.6mo DoT ) Projected IO-Experienced Opportunity at mPFS Projected 1L Opportunity at mPFS Potential HIF-2α Peak Sales ($M) 3 Belzutifan Today is Only "Scratching the Surface" on the RCC Opportunity (Major Markets) Arcus is pursuing early lines of treatment, and cas has the potential to achieve long durations of therapy $5B+ for 1L and "IO- Experienced" 15k pts 21k pts 24k pts 23 Annualized run-rate belzutifan sales based on Q4 2025 results. 5.6mo DoT assumed based on LS -005 PFS. Projected potential opportunities based on DRG (Clarivate) epi and Arcus analysis across US, EU5 & Japan in 2036. Assumes HIF -2α containing regimens capture 55-75% share of patient population. Major Markets: France, Germany, Italy, Japan, Spain, United Kingdom, United States

Page 24

© Arcus Biosciences 2026 24 Oncologists clearly preferred the PD1/CTLA4 + cas regimen over the PD1/TKI + cas regimen when presented with both options Source: Arcus Primary Qualitative Research, n=20, Fielded December 3 2025 - January 7 2026 # of Oncs Preferred anti-PD-1 + anti-CTLA-4 + cas 15 out of 20 Neutral 1 out of 20 anti-PD1 + TKI + cas 4 out of 20  Preferred for patients with rapidly progressing disease • Uses 3 MoAs in 1L, limiting sequencing options • Concerns related to AEs from TKIs  Familiarity with PD1+CTLA4 combination across other tumors  Allows for sequencing of different MoAs in 2L+  PD1+CTLA4 currently most used regimen in 1L  May not be suitable for patients w/ rapid disease progression anti-PD-1 + TKI + casanti-PD-1 + anti-CTLA-4 + cas Oncologist Regimen Preferences Pros Pros Cons Cons

Page 25

Page 26

© Arcus Biosciences 2026 26 Quemli Could be the First Transformative Therapy for All- Comer 1L Pancreatic Cancer in 30+ Years GnP data constructed into a synthetic control arm, on a post-hoc basis, from Phase 2 and 3 clinical studies in 1L metastatic pancreatic cancer setting ~$4B Market Potential PHASE 1 STUDY SHOWED 5.9 MONTH mOS IMPROVEMENT VS G/nP13 Readout in 1H 2027 Kaplan-Meir Estimate of Overall Survival (%) Overall Survival (months) 1.00 0.75 0.50 0.25 0.00 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 122 108 89 72 59 47 34 23 8 3 1 0 122 104 85 61 49 32 26 21 12 7 4 3 1 0 All pooled Q100 Q(±Z) + G/nP (n=122) SCA (n=122) Median OS, m 15.7 9.8 HR [95% CI] 0.634 (0.471-0.854) Log-rank nominal p-value 0.0030 Events, % (n) 58.2% (71) 86.1% (105) All pooled Q100 Q(±Z) + G/nP Synthetic Control Arm Censored Median follow-up, All pooled Q100 Q(±Z) + G/nP NUMBER OF PATIENTS AT RISK Phase 3 trial in 1L PDAC Enrollment completed

Page 27

© Arcus Biosciences 2026 Phase 3 Study of Quemli + Chemo in 1L Metastatic PDAC Arm A: Quemli 100mg Q2W + G/nP Arm B: Placebo Q2W + G/nP 2:1 N = 610 Stratification • ECOG 0/1 • Liver metastases • *Region 1L mPDAC • Confirmed PDAC untreated in metastatic setting • Metastatic disease diagnosed within 6 weeks prior to screening • Measurable metastatic disease per RECIST 1.1 • ECOG PS 0 or 1 PRIMARY ENDPOINT: • OS KEY SECONDARY ENDPOINTS: • PFS ENROLLMENT COMPLETED

Page 28

Page 29

© Arcus Biosciences 2026 29 I&I Small-Molecule Strategy is Targeting Validated Blockbuster Biologics IN-HOUSE EXPERTISE IN IMMUNOLOGY has been a core aspect of our discovery group since Arcus’s founding MINIMIZE BIOLOGICAL RISK by leveraging validated mechanisms with applications to common diseases with large addressable populations 2-PRONG I&I STRATEGY: • Small-molecule improvements of cytokine-targeted therapeutics • Target immune cell types that play key roles in human disease and have been historically “under-studied” TARGET MODALITY DISEASE AREA STATUS MRGPRX2 SM CSU, AD FIH Expected in 2026 TNF (TNFR1) SM RA, Psoriasis, IBD FIH Expected Late 2026 / Early 2027 CCR6 SM Psoriasis Advanced Discovery CD89 mAb RA Advanced Discovery CD40L SM SLE; MS Discovery I&I DRUG DISCOVERY PORTFOLIO

Page 30

© Arcus Biosciences 2026 -6 -5 -4 -3 -2 -1 0 1 2 0 20 40 60 80 100 120 Log [AB102] (µM) Percent of Activity Validated biology with multi- billion dollar potential OPPORTUNITY FOR IMPROVEMENT PROGRAM STATUS: Expect FIH in 2026 30 Potential Best-in-Class MRGPRX2 Antagonist to Treat Atopic Skin Diseases Improved potency/PK relative to early small-molecule entrants into the clinic Based on the predicted human PK and potency of our leading molecule, required clinical exposures could be >90% lower than those associated with the leading small-molecule competitor in the clinic Approved biologics are highly successful and effective in treating AD and CSU, etc. and generate >$15B in LTM sales14 Anti-IgE (e.g., omalizumab) and anti-IL-4R (e.g., dupilumab) are not sufficient to address clinical need in CSU and/or AD IC50: 8 nM IC90: 46 nM Mast cell (LAD2) degranulation (CD107a) in 100% human serum Modeled steady-state human PK 0 4 8 12 16 20 24 10 100 Time (hr) AB102 Concentration (ng/mL)

Page 31

© Arcus Biosciences 2026 Anti-TNF antibodies block TNF signaling at 2 receptors: • TNFR1 (pro-inflammatory) • TNFR2 (pro-Treg – blocking can drive inflammation) As a result, TNF antibodies can drive a fraction of patients to develop “paradoxical inflammation” (e.g., psoriasis) Small-molecule disruptors of TNF selectively block only the TNFR1 biology, with potential for efficacy & better safety Opportunity for molecules with better potency / human PK, relative to early small-molecule entrant into the clinic VALIDATED BIOLOGY WITH MULTI-BILLION $ POTENTIAL 31 Small-molecule Inhibitors of TNF with Potentially Improved Efficacy/Safety Relative to anti-TNF Biologics Anti-TNF antibodies are among the most successful biologic drugs ever developed Humira® was the world’s top-selling drug for nearly a decade, with peak sales over $20B15,16 OPPORTUNITY FOR IMPROVEMENT PROGRAM STATUS: Expect FIH in late 2026 / early 2027

Page 32

© Arcus Biosciences 2026 INDICATION MARKET POTENTIAL (Major markets*) CAS Small molecule HIF-2α inhibitor Post-IO ccRCC ~$2B IO-naive ccRCC ~$3B QUEMLI Small molecule CD73 inhibitor 1L PDAC >$4B MRGPRX2 Small molecule antagonist CSU & Atopic dermatitis Multi-Billion $ TNF Small molecule inhibitor RA, Psoriasis & IBD Multi-Billion $ 32 Multiple Programs Targeting Substantial Market Opportunities and Unmet Medical Need *Major Markets (US, EU5, JP) - total projected 2034 quemli opportunity & cas opportunity

Page 33

© Arcus Biosciences 2026 33 Multiple Catalysts are Expected over the Next 12 Months Cas + cabo PFS data Data for cas combinations in early-line RCC POC data for MRGPRX2 antagonistInitiate first Phase 3 study for cas in 1L RCC MRGPRX2 antagonist in clinic Phase 3 data TNF inhibitor in clinic Phase 3 data NEXT 3 MONTHS NEXT 6-9 MONTHS NEXT 9-12 MONTHS NEXT 12-24 MONTHS NEXT 24+ MONTHS  Cas monotherapy data at ASCO GU

Page 34

Page 35

© Arcus Biosciences 2026 References and Acronym Key 1. Choueiri T. et al 2026 (ARC-20) 9. Rini B. et al. 2011 (AXIS) 2. Albiges L. et al. 2023 (LITESPARK-005) 10. Pal S. et al. 2023 (CONTACT-03) 3. Choueiri T. et al. 2024 (LITESPARK-005) 11. Tannir N. et al. 2022 (CANTATA) 4. Choueiri T. et al. 2025 (ARC-20) 12. Merck 4QFY 2025 earnings release. Accessed February 18, 2026. 5. Marathe D. et al. 2024 (belzutifan) 13. Wainberg Z. et al. 2024 (ARC-8) 6. Rini B. et al 2020 (TIVO-3) 14. Sanofi earnings releases. Accessed February 18, 2026. 7. Choueiri T. et al. 2016 (METEOR) 15. 2022 Annual Report. Abbvie. Accessed February 18, 2026. 8. Motzer R. et al. 2015 (lenvatinib +everolimus) 16. Sagonowsky E. The top 20 drugs by worldwide sales in 2020. FiercePharma. Published May 3 2021. Accessed January 8, 2026. 1H first half Cabo cabozantinib ECOG Eastern Cooperative Oncology Group mAb monoclonal antibody OS overall survival Q2/3W every 2/3 weeks SLE systemic lupus erythematosus 1L first-line Cas casdatifan EPO erythropoietin mg milligrams PD primary progression QD once daily SM small molecule 2H second-half ccRCC clear cell renal cell carcinoma FIH first-in-human MOA mechanism of action PDAC pancreatic ductal adenocarcinoma quemli quemliclustat SOC standard of care 2L second-line CI confidence interval G/nP gemcitabine/nab- paclitaxel Mono monotherapy PD-L1 programmed death- ligand 1 R randomized TKI tyrosine kinase inhibitor 3L third-line cORR confirmed overall response rate HR hazard ratio mos months PFS progression-free survival R&D research & development Tivo tivozanib AD atopic dermatitis CR complete response I&I inflammation & immunology mOS median overall survival Ph Phase RA rheumatoid arthritis YE year-end AE adverse event CSU chronic spontaneous urticaria IBD inflammatory bowel disease mPFS median progression-free survival PK pharmacokinetics RCC renal cell carcinoma Z zimberelimab Axi axitinib DCO data cutoff IO immunotherapy MRGPRX2 mas-related G protein- coupled receptor member X2 PR partial response RECIST Response Evaluation Criteria in Solid Tumors zim zimberelimab B billion DCR disease control rate K thousand MS multiple sclerosis POC proof of concept SCA synthetic control arm belz belzutifan DOR duration of response Lenva lenvatinib NE not evaluable Q quemliclustat sEPO serum erythropoietin BID twice daily DoT duration of treatment m months ORR objective response rate Q3 third quarter SD stable disease

Slides