Slides
Page 1
Burning Rock Biotech Limited 2Q 2026 results 10 September 2026
Page 2
2 Disclaimer This presentation has been prepared by Burning Rock Biotech Limited (the “Company”) solely for information purpose and has not been independently verified. No representations, warranties or undertakings, express or implied, are made by the Company or any of its affiliates, advisers, or representatives as to, and no reliance should be placed upon, the accuracy, fairness, completeness or correctness of the information or opinions presented or contained in this presentation. None of the Company or any of its affiliates, advisers or representatives accept any responsibility whatsoever (in negligence or otherwise) for any loss howsoever arising from any information presented or contained in this presentation or otherwise arising in connection with the presentation. The information presented or contained in this presentation is subject to change without notice and its accuracy is not guaranteed. Certain statements in this presentation, and other statements that the Company may make, are forward-looking statements within the meaning of Section 21E of the Securities Exchange Act of 1934, as amended. These statements reflect the Company’s intent, beliefs or current expectations about the future. These statements can be recognized by the use of words such as “expects,” “plans,” “will,” “estimates,” “projects,” “intends,” “anticipates,” “believes,” “confident” or words of similar meaning. These forward-looking statements are not guarantees of future performance and are based on a number of assumptions about the Company’s operations and other factors, many of which are beyond the Company’s control, and accordingly, actual results may differ materially from these forward-looking statements. The Company or any of its affiliates, advisers or representatives has no obligation and does not undertake to revise forward-looking statements to reflect future events or circumstances. This presentation does not constitute an offer to sell or issue or an invitation to purchase or subscribe for any securities of the Company for sale in the United States or anywhere else. No part of this presentation shall form the basis of or be relied upon in connection with any contract or commitment whatsoever. THE INFORMATION CONTAINED IN THIS DOCUMENT IS BEING GIVEN SOLELY FOR YOUR INFORMATION AND ONLY FOR YOUR USE IN CONNECTION WITH THIS PRESENTATION. THE INFORMATION CONTAINED HEREIN MAY NOT BE COPIED, REPRODUCED, REDISTRIBUTED, OR OTHERWISE DISCLOSED, IN WHOLE OR IN PART, TO ANY OTHER PERSON IN ANY MANNER. ANY FORWARDING, DISTRIBUTION OR REPRODUCTION OF THIS PRESENTATION IN WHOLE OR IN PART IS UNAUTHORIZED. By viewing, accessing or participating in this presentation, participants hereby acknowledge and agree to keep the contents of this presentation and these materials confidential. Participants agree not to remove these materials, or any materials provided in connection herewith, from the conference room where such documents are provided. Participants agree further not to photograph, copy or otherwise reproduce this presentation in any form or pass on this presentation to any other person for any purpose, during the presentation or while in the conference room. Participants must return this presentation and all other materials provided in connection herewith to the Company upon completion of the presentation. By viewing, accessing or participating in this presentation, participants agree to be bound by the foregoing limitations. Any failure to comply with these restrictions may constitute a violation of applicable securities laws.
Page 3
3 Our value-building blocks Extending leadership in NGS-based precision oncology from late-stage to earlier stage patients Early Detection Asymptomatic population Therapy Selection and MRD1 Assisting throughout the full process of oncology diagnosis, treatment and monitoring Biopharma Global biomarker services and CDx2 partners New Businesses Large market potential at early commercial phase Developed Business Commercial phase Common Infrastructure Accelerating growth of new businesses • Strong brand to support new product launches & attract talent • Broad industry network and synergy across different business units • Large volumes supporting lower cost & faster innovation Notes: 1 Molecular residual disease of solid tumors 2 Companion diagnostics Oversea Business Reagent kit for breast cancer
Page 4
4 Therapy Selection and MRD • OncoScreen® BCMatch Tissue Kit has officially entered the Priority Review Channel of the Center for Medical Device Evaluation(CMDE) under the NMPA. • CanCatch Custom Kit and CanCatch Custom Kit (cPAS) have officially entered the Special Review Procedure for Innovative Medical Devices of CMDE under the NMPA. • In August 2026, the Priority Review Channel application for OncoCompass Target Cancer Mutation Profiling Liquid Kit was resubmitted and subsequently accepted for review. Recent regulatory and academic progress
Page 5
5 Recent operating progress Continued efficiency gains, driving towards profitability Increasing revenues from in-hospital channel • In-hospital revenue keep increasing: +7.4% YoY; +27.3% QoQ, reflecting active collaboration with hospitals and continued transition towards in-hospital business Improving gross margin • Gross profit margin of in hospital business: 77.1% in 26Q2, +2.7 ppts YoY , reflecting optimized product mix and cost control Reducing operating expenses • Total opex: -3.2% YoY in 26Q2, most of these seeing disciplined cuts and efficiency improved • Improving inventory turnover • Inventory turnover: +20% YoY in 26Q2, freeing up working capital and improving overall efficiency 1 2 3 4
Page 6
6 Commercial operation cash flow and cash balance Notes: 1 Net cash used in operating activities 2 Purchase and prepayment of property and equipment and intangible assets, issuance of convertible loan, out of investing cashflow 3 Cash balance included cash and cash equivalents and restricted cash 1 RMB millions 26Q2 26H1 25Q2 25H1 FY25 FY24 FY23 Operating cash outflow1 22 52 44 68 28 92 256 Capex2 2 3 0 1 5 6 9 Sum 24 55 44 69 33 98 265 Cash balance3 at period-end 419 419 455 455 481 522 615
Page 7
7 Revenues by channel2 Revenues 26Q2 25Q2 YoY Revenues from central lab 36.8 40.9 -10% Revenues from in-hospital 67.1 62.5 7% Revenues from pharma R&D 30.8 45.1 -32% Total revenues 134.8 148.5 -9% Total gross profit margin 73% 73% 0% Increasing proportion of in-hospital revenues from 43% in 25Q2 to 53% in 26Q2, reflecting the continued transition towards in-hospital channel 25Q2 Central lab In-hospital Pharma R&D % of revenue 26Q2 % of revenue 43% 53%
Page 8
8 Reducing operating expenses Expect cost savings going forward 3 10% 60% 110% 22Q1 22Q2 22Q3 22Q4 23Q1 23Q2 23Q3 23Q4 24Q1 24Q2 24Q3 24Q4 25Q1 25Q2 25Q3 25Q4 26Q1 26Q2 R&D S&M G&A RMB millions 26Q2 25Q2 YoY R&D expense 31.2 49.8 -37% S&M expense 47.1 38.4 23% G&A expense 37.5 31.4 19% Total expense 115.8 119.6 -3% Driven by business cost control to improve operating efficiency, operating expenses have decreased in 26Q2. R&D expense, S&M expense and G&A expense % of revenues as follows:
Page 9
9 Improving inventory turnover Improved inventory management has steadily increased inventory turnover1 as follows: Notes: 1 Inventory turnover is calculated as cost of revenues divided by average inventory and contract costs. Average inventory and contract costs are the average of the beginning and ending carrying amounts of inventory and contract costs for the period. 2 2026H1 inventory turnover calculated using annualized cost of revenues. 4 1.31 1.72 2.5 2.47 2.77 FY22 FY23 FY24 FY25 26H1 0 0.5 1 1.5 2 2.5 3 Inventory turnover (x per year) 2
Page 10
10 Financials Notes: 1 Within in-hospital segment, over 95% revenues are kit revenues, which are recurring in nature; the remaining are instrument revenues. In-hospital primarily through direct-sales model 2 Excluding share based compensation (SBC) and depreciation and amortization (D&A) 3 Defined as net income (loss) adjusted to exclude interest income (expense), income tax expense, depreciation and amortization, impairment of accounts receivables and contract assets, impairment of long-lived assets and share-based compensation RMB millions 2024 2025 YoY 1Q25 2Q25 3Q25 4Q25 1Q26 2Q26 2Q26 YoY 2Q26 QoQ Revenues 515.8 539.6 5% 133.1 148.6 131.6 126.3 108.0 134.7 -9% 25% Central lab 175.6 160.0 -9% 38.3 40.9 36.8 44.0 32.4 36.8 -10% 14% In-hospital1 224.5 224.0 0% 57.7 62.5 52.8 51.0 52.8 67.1 7% 27% Pharma 115.7 155.6 35% 37.1 45.2 42.0 31.3 22.8 30.8 -32% 35% Gross profit 362.4 402.9 11% 97.4 108.1 98.8 98.6 78.0 97.8 -10% 25% Total opex 720.1 457.8 -36% 112.6 119.6 115.0 110.6 96.7 115.8 -3% 20% R&D2 191.1 161.4 -16% 38.0 48.9 40.3 34.2 27.1 30.2 -38% 11% S&M2 176.4 157.8 -11% 38.7 36.8 40.0 42.3 40.1 46.1 25% 15% G&A2 137.3 115.0 -16% 29.1 28.6 28.2 29.1 25.8 35.9 26% 39% SBC 153.7 12.6 4.2 2.1 3.5 2.8 1.7 1.9 D&A 26.5 11.0 2.6 3.2 3.0 2.2 2.0 1.7 Impairment 35.1 - - - - - - - Adjusted EBITDA3 (101.3) (8.8) (3.1) (0.6) (5.4) 0.4 (13.0) (8.0) Operating profit (357.5) (54.8) (15.2) (11.5) (16.1) (12.0) (18.8) (15.2) Net operating cash flows (92.2) (28.4) (23.5) (44.3) 16.4 23.0 (29.3) (22.4) Non-GAAP GP margin2 74.9% 76.6% 75.7% 74.4% 76.7% 80.3% 74.6% 74.3%
Page 11
Molecular Residual Disease (MRD)
Page 12
12 Burning Rock’s MRD clinical publications Covering adjuvant and relapse settings in lung, colorectal, gastric and other cancers Resectable Tumor Surgical Resection Neo- adjuvant Adjuvant Relapse The baseline ctDNA level Treatment effectiveness assessment Landmark MRD Longitudinal monitoring Treatment effectiveness assessment Non-small-cell lung cancer Colorectal cancer Pancreatic cancer Baseline, landmark and longitudinal monitoring timepoints completed AACR 2022 Abstract 5916, AACR 2023 Abstract 1039,MEDAL study publication Baseline, landmark timepoints and longitudinal monitoring AACR 2022 Abstract 5917, ASCO GI 2023 Abstract 213, PRECARE study publication Baseline, landmark and longitudinal monitoring timepoints completed ASCO GI 2023 Abstract 744 Biliary tract cancer Gastric cancer Baseline and landmark timepoints AACR 2023 Abstract 6682 Baseline and landmark timepoints AACR 2023 Abstract 6682 Esophageal cancer Baseline, post-neoadjuvant therapy, landmark and longitudinal monitoring timepoints completed preSINO study publication
Page 13
13 Overview of CanCatch® Custom (Previously named brPROPHET) An ultrasensitive and quantitative MRD assay Source: Chen et al., Individualized tumor-informed circulating tumor DNA analysis for postoperative monitoring of non-small cell lung cancer, Cancer Cell, Sep 2023 CanCatch® Custom Workflow
Page 14
14 MEDAL study Personalized MRD using CanCatch® Custom on non-small cell lung cancer (NSCLC) Source: Chen et al., Individualized tumor-informed circulating tumor DNA analysis for postoperative monitoring of non-small cell lung cancer, Cancer Cell, Sep 2023
Page 15
15 Source: Chen et al., Individualized tumor-informed circulating tumor DNA analysis for postoperative monitoring of non-small cell lung cancer, Cancer Cell, Sep 2023 Study design ❑ Cohort: ▪ 181 patients enrolled Stage I (63%), II (19%), and III (18%) ❑ Sampling Time: ▪ Tumor and adjacent paired tissue collected at surgery ▪ Blood samples collected at Pre-operative, 3 days, and 30 days post-surgery ▪ Median Follow-up Time: 30 months
Page 16
16 CanCatch® Custom demonstrates superior sensitivity in ctDNA detection Clinical validation with pre-operative plasma samples b e f ca d b e f c a d (a) (b) Sensitivity of pre-operative plasma ctDNA fraction distribution of MRD (+) samples detected by different methods • For pre-operative plasma from patients with different clinical stages, CanCatch® Custom has a higher sensitivity than the other two methods • The median ctDNA fraction of the 30 patients detected by CanCatch® Custom alone was significantly lower than the 25 patients detected by all three MRD assays The patient-specific CanCatch® Custom has a higher sensitivity than the two fixed panel detection methods Source: Chen et al., Individualized tumor-informed circulating tumor DNA analysis for postoperative monitoring of non-small cell lung cancer, Cancer Cell, Sep 2023
Page 17
17 CanCatch® shows strong prognostic value in post-surgery NSCLC patients Source: Chen et al., Individualized tumor-informed circulating tumor DNA analysis for postoperative monitoring of non-small cell lung cancer, Cancer Cell, Sep 2023 Prognostic analysis at Landmark time points Longitudinal MRD analysis
Page 18
18 preSINO study Personalized MRD using CanCatch® Custom on esophageal cancer Source: Liu et al. ctDNA detects residual disease after neoadjuvant chemoradiotherapy and guides adjuvant therapy in esophageal squamous cell carcinom a. Cell Rep Med. 2025.
Page 19
19 Study design Source: Liu et al. ctDNA detects residual disease after neoadjuvant chemoradiotherapy and guides adjuvant therapy in esophageal squamous cell carcinom a. Cell Rep Med. 2025.
Page 20
20 CanCatch® Custom demonstrates superior sensitivity in ctDNA detection • For baseline (before neoadjuvant therapy) plasma from patients with different clinical stages, CanCatch® Custom has a higher sensitivity than the other two methods • The median ctDNA fraction of the 9 patients detected by CanCatch® Custom alone was significantly lower than the 76 patients detected by all three MRD assays The patient-specific CanCatch® Custom has a higher sensitivity than the two fixed panel detection methods Source: Liu et al. ctDNA detects residual disease after neoadjuvant chemoradiotherapy and guides adjuvant therapy in esophageal squamous cell carcinom a. Cell Rep Med. 2025.
Page 21
21 CanCatch® can predict residual disease after neoadjuvant therapy in esophageal cancer • CanCatch® Custom deceases the false negative rate of predicting residual disease by 11.3%-13.6% when combined with clinical response evaluation (biopsy or biopsy+fine needle aspiration) Source: Liu et al. ctDNA detects residual disease after neoadjuvant chemoradiotherapy and guides adjuvant therapy in esophageal squamous cell carcinom a. Cell Rep Med. 2025.
Page 22
22 CanCatch® shows strong prognostic value in post-surgery esophageal cancer • The ctDNA (+) detected by CanCatch® Custom at 3-4 weeks post-surgery was associated with poor disease-free survival. • In patients with ctDNA (+) post-surgery, non-pCR patients treated with adjuvant immunotherapy had better disease- free survival than those without adjuvant immunotherapy, which was not observed in patients with ctDNA (-), suggesting that the ctDNA status post-surgery can guide adjuvant immunotherapy in patients with esophageal cancer. ctDNA (+) ctDNA (-)
Page 23
Early detection
Page 24
24 Competitive technology Methylation + machine learning to overcome challenges of low ctDNA abundance Multi-cancer validation data Burning Rock’s multi-cancer early detection technology AACR 2022 ESMO 2022 Regulatory breakthrough breakthrough device designation granted China NMPA breakthrough designation granted
Page 25
25 Product development roadmap Proof-of-concept 2016 – 2019 3-cancer 2017 – 2020 6-cancer 2018 – 2020 22-cancer 2019 – Ongoing • Proof of concept on our methylation based, machine learning aided technology platform • Results published on Nature Biomedical Engineering, “Ultrasensitive detection of circulating tumour DNA via deep methylation sequencing aided by machine learning” • Lung, Colorectal Cancer (CRC), Hepatocellular Carcinoma (HCC) • Lung, CRC, HCC, Ovarian, Pancreatic, Esophageal • THUNDER study (N=2,385) completed, 98.9% specificity and 69.1% sensitivity, top-2 predicted origin accuracy of 91.7% (independent validation cohort) • PREVENT study (prospective and interventional, IU population) ongoing • Covering 88% of China’s cancer incidence • PROMISE study (N=2,035) completed, improved performance vs. 6-cancer test • PREDICT and PRESCIENT studies ongoing
Page 26
26 Running the largest clinical programs in China supported by top physicians PREDICT PRESCIENT • Leading site: Shanghai Zhongshan Hospital − One of China's largest comprehensive academic hospitals − Performs c.104,000 operations and serves c.169,000 inpatients and over 4,236,000 outpatients on an annual basis1 − Ranked top 5 in the 2019 China’s general hospital rankings2 • Fellow of the Chinese Academy of Sciences • President of Shanghai Zhongshan Hospital • Leading site: Cancer Hospital of the Chinese Academy of Medical Sciences3 − The first and top cancer-specialist hospital in China − The National Clinical Center for Cancer Research, the National Center for Quality Control on Standardized Cancer Treatment and Diagnosis, the National Clinical Center for Drug Research • Fellow of the Chinese Academy of Sciences • President of CHCAMS Principal Investigator: Prof. Jia Fan Principal Investigators Prof. Jie He Prof. Jie Wang Notes: 1 Based on 2018 statistics 2 http://rank.cn-healthcare.com/rank/general-best 3 CHCAMS PREVENT • Head of the Dept. of Medicine, CHCAMS Principal Investigator: Prof. Weiming Li • President of West China Hospital • Leading site: West China Hospital − One of the largest hospitals in China, performed 196,000 surgeries and 7.8 million out-patient services in 2021 − Ranked #2 in the Fudan Best Hospital in China Rankings (2009-2020)
Page 27
Appendix 1 Early detection
Page 28
28 Product Development Roadmap Proof-of-concept 2016 – 2019 3-cancer 2017 – 2020 6-cancer 2018 – 2020 CE Mark, FDA BDD 22-cancer 2019 – Ongoing • Proof of concept on our methylation based, machine learning aided technology platform • Results published on Nature Biomedical Engineering, “Ultrasensitive detection of circulating tumour DNA via deep methylation sequencing aided by machine learning” • Lung, Colorectal Cancer (CRC), Hepatocellular Carcinoma (HCC) • Lung, CRC, HCC, Ovarian, Pancreatic, Esophageal • THUNDER study (N=2,385) completed, 98.9% specificity and 69.1% sensitivity, top-2 predicted origin accuracy of 91.7% (independent validation cohort) • PREVENT study (prospective and interventional, IU population) ongoing • Covering 88% of China’s cancer incidence • PROMISE study (N=2,035) completed, improved performance vs. 6-cancer test • PREDICT/PRESCIENT studies ongoing
Page 29
29 6-cancer test marker discovery and model training The THUNDER study, 2395 participants Source: Gao et al., Unintrusive multi-cancer detection by circulating cell-free DNA methylation sequencing (THUNDER): development and independent validation studies, ASCO 2022 A customized panel of 161,984 CpG sites was constructed and validated by public and in-house (cancer: n = 249; non-cancer: n = 288) methylome data, respectively. The cfDNA samples from 1,693 participants (cancer: n = 735; non-cancer: n = 958) were retrospectively collected and divided into training and validation sets to establish and test two multi-cancer detection blood test (MCDBT-1/2) models. Both models was blindly validated on a prospectively enrolled, independent validation cohort of age-matched 1,010 participants (cancer: n = 505; non-cancer: n = 505).
Page 30
30 6-cancer test, detection-of-cancer performance in case-control cohorts The THUNDER study Source: Gao et al., Unintrusive multi-cancer detection by circulating cell-free DNA methylation sequencing (THUNDER): development and independent validation studies, ASCO 2022 Data set Specificity (%) Sensitivity (%) Accuracy of top predicted origin (%) Accuracy of top two predicted origins (%) Training set 99.7 (98.9-100.0) 75.2 (70.6-79.4) 89.7 (85.7-92.9) 94.7 (91.5-96.9) Validation set 100.0 (97.0-100.0) 69.4 (63.9-74.6) 82.8 (77.0-87.6) 89.4 (84.5-93.3) Independent validation set 98.9 (97.6-99.7) 69.1 (64.8-73.3) 83.2 (78.7-87.1) 91.7 (88.2-94.5)
Page 31
31 6-cancer test, top-predicted-origin performance in case-control cohorts The THUNDER study Source: Gao et al., Unintrusive multi-cancer detection by circulating cell-free DNA methylation sequencing (THUNDER): development and independent validation studies, ASCO 2022 Data set Specificity (%) Sensitivity (%) Accuracy of top predicted origin (%) Accuracy of top two predicted origins (%) Training set 99.7 (98.9-100.0) 75.2 (70.6-79.4) 89.7 (85.7-92.9) 94.7 (91.5-96.9) Validation set 100.0 (97.0-100.0) 69.4 (63.9-74.6) 82.8 (77.0-87.6) 89.4 (84.5-93.3) Independent validation set 98.9 (97.6-99.7) 69.1 (64.8-73.3) 83.2 (78.7-87.1) 91.7 (88.2-94.5)
Page 32
32 Product Development Roadmap Proof-of-concept 2016 – 2019 3-cancer 2017 – 2020 6-cancer 2018 – 2020 CE Mark, FDA BDD 22-cancer 2019 – 2022 • Proof of concept on our methylation based, machine learning aided technology platform • Results published on Nature Biomedical Engineering, “Ultrasensitive detection of circulating tumour DNA via deep methylation sequencing aided by machine learning” • Lung, Colorectal Cancer (CRC), Hepatocellular Carcinoma (HCC) • Lung, CRC, HCC, Ovarian, Pancreatic, Esophageal • THUNDER study (N=2,385) completed, 98.9% specificity and 69.1% sensitivity, top-2 predicted origin accuracy of 91.7% (independent validation cohort) • PREVENT study (prospective and interventional, IU population) ongoing • Covering 88% of China’s cancer incidence • PROMISE study (N=2,035) completed, improved performance vs. 6-cancer test • PREDICT/PRESCIENT studies ongoing
Page 33
33 Source: Duan et al. Exploration of multi-omics liquid biopsy approaches for multi-cancer early detection: The PROMISE study a multi-cancer early detection model based on liquid biopsy of multi-omics. The Innovation, 2025. The PROMISE study Exploration of multi-omics classifier for 9 cancer
Page 34
34 Study design The PROMISE study, 1,706 participants Source: Duan et al. Exploration of multi-omics liquid biopsy approaches for multi-cancer early detection: The PROMISE study a multi-cancer early detection model based on liquid biopsy of multi-omics. The Innovation, 2025.
Page 35
35 Multimodal classifier for 9-cancer, detection-of-cancer performance The PROMISE study Source: Duan et al. Exploration of multi-omics liquid biopsy approaches for multi-cancer early detection: The PROMISE study a multi-cancer early detection model based on liquid biopsy of multi-omics. The Innovation, 2025. Performance of the multimodal classifier based on both methylation and protein markers (A) Sensitivity, specificity, and top predicted origin accuracy. (B) Sensitivity reported by cancer stage and cancer type. Error bars represent 95% CI.
Page 36
36 Multimodal classifier for 9-cancer, top-predicted-origin performance The PROMISE study Source: Duan et al. Exploration of multi-omics liquid biopsy approaches for multi-cancer early detection: The PROMISE study a multi-cancer early detection model based on liquid biopsy of multi-omics. The Innovation, 2025. Performance of the multimodal classifier based on both methylation and protein markers (C) Top predicted origin accuracy reported by cancer type.
Page 37
37 Burning Rock’s 22-cancer test covers 88% of China’s cancer incidence China Cancer Incidence1 per 100,000 population, across all ages Notes: 1 Incidence data per “2018 China cancer registry annual report ”, J He et al., ISBN 978-7-117-28585-8 2 Final number of cancer types subject to development progress 28 28 0 50 100 150 200 250 300 350 Uterus Prostate Thyroid Cervix Esophagus Colorectal Liver Gastric Breast Lung BR-6 BR-22 All cancer types
Page 38
38 Leadership in multi-cancer early detection First-in-class, high entry-barrier, multi-year efforts Low amount of cancer signal in the circulating bloodstream, much more challenging compared to tissueTechnology Clinical Regulatory Commercial Challenges BNR position Proprietary chemistry and algorithm • On par with global leader, competitive sensitivity in earlier stages for certain cancers • Multi-year lead vs. China peers (most showing liver-cancer and colon-cancer data only) Large, multi-year studies required from case-control to intend-to-use population, from observational to interventional (e.g. CCGA study: 15,254 participants, 8,584 with cancer, 6,670 without cancer) Sponsorship from top physicians • Catching up with global leader, to improve specificity and tissue-of-origin performance through large clinical studies • Multi-year lead in China as the only company that has launched studies with over 10,000+ subjects 1 2 3 4 First-in-class with no established regulatory pathway Leading regulatory capability in China • Exploring possible pathway, leveraging experience through the country’s first NMPA-approved NGS kit Unprecedented product Multi-pronged approach • Initially working with hospitals’ health check-up departments, leveraging synergy from in-hospital therapy selection business
Page 39
Appendix 2 MRD
Page 40
40 How do MRD studies advance utility Example 1: IMvigor010, enrich the high-risk group and "tune-up" adjuvant treatment Atezo, MIUC Adjuvant Therapy “All comers” demonstrated NO efficacy TMB/PD-L1 showed NO prediction Landmark MRD+ pts (39%) had worse prognosis Maybe only those patients can benefit? Indeed, only baseline MRD+ pts showed benefit Nature. 2021 Jun 16. doi: 10.1038/s41586-021-03642-9.
Page 41
41 How do MRD studies advance utility Example 2: Dynamic, identify low-risk patients and “tune-down” adjuvant treatment ctDNA-guided vs standard The ctDNA-guided MRD- group had fewer patients with adjuvant chemotherapy than the standard group (15% vs. 29%) with non-inferiority of 3-year RFS (92.4% vs 91.7%) 2022 ASCO Annual Meeting
Page 42
42 CanCatch® – Burning Rock’s MRD solution Whole Exome Sequencing Tumor Profiling with add-on region for SV detection SNV/SV/CNV/MSI/HLA/Therapy Selection Personalized Panel Intelligent selection of 50 tracking sites CanCatch® Custom Assay 100,000x Raw Depth/UMI error correction/Tumor Fraction Estimation PROPHET Techology – CanCatch® Custom Assay Patient-specific pROgnostic and Potential tHErapeutic marker Tracking
Page 43
43 Colorectal cancer cohort publication at AACR 2025 Source: Li et al. Circulating tumor DNA-based molecular residual disease guides treatment in metastatic colorectal cancer patients with no evidence of disease: An open-label, prospective, phase II study #4567 AACR 2025 Flow chart Patients with negative ctDNA results detected by CanCatch® after curative treatment demonstrated superior recurrence-free survival Longitudinal ctDNA monitoring with CanCatch® demonstrates potential prognostic value
Page 44
44 Gastrointestinal stromal tumor publication at ASCO 2025 The positive post-operative ctDNA detected by CanCatch® informed tumor recurrence prior to radiological evidence with a lead time of 3 months in a 71-year-old patient. Study Flow Patients with positive MRD detected by CanCatch® at landmark showed marginally worse DFS compared with those with negative MRD (p = 0.08, left). The longitudinal MRD positivity was associated with inferior DFS (p = 0.012, right) Source: Zhidong Gao et al. Personalized tumour-informed circulating tumor DNA analysis in monitoring recurrence following resection of high - risk locally advanced stage gastrointestinal stromal tumor #11522 ASCO 2025
Page 45
45 Colorectal cancer cohort publication at ASCO 2024 CanCatch® assay showed a positive rate of 96% in preoperative samples. Day 7 Day 30 ctDNA positivity detected by CanCatch® at postoperative day 7 and 30 predicted worse RFS Among patients with results from all three approaches (n=168), the CanCatch® assay outperformed TI and TN fixed-panel assays in predicting RFS at postoperative day 7. Note: TI: fixed panel tumor-informed approach; TN: tumor naï ve (TN) approach. Source: Di Cao et al. Individualized Tumor-Informed Circulating Tumor DNA Analysis for Molecular Residual Disease Detection in Predicting Recurrence and Efficacy of Adjuvant Chemotherapy in Colorectal Cancer #3615 ASCO 2024
Page 46
Appendix 3 Therapy selection
Page 47
47 First NMPA-approved kit Second NMPA-approved kit EGFR, ALK, BRAF, KRAS Approved in Jul 2018 First approved NGS kit in China EGFR, KRAS, MET, ERBB2, BRAF, PIK3CA, ALK, ROS1, RET Approved in Mar 2022 Novogene 诺禾 EGFR, KRAS, BRAF, PIK3CA, ALK, ROS1 Approved in Aug 2018 Geneseeq 世和 EGFR, ALK, ROS1, BRAF, KRAS, ERBB2 Approved in Sep 2018 BGI 华大 EGFR, KRAS, ALK Approved in Aug 2019 Gene+ 吉因加 EGFR, KRAS, ALK Approved in Dec 2019 Genetron 泛生子 EGFR, KRAS, BRAF, ERBB2, PIK3CA, ALK, ROS1, MET Approved in Feb 2020 Genecast 臻和 KRAS, NRAS, BRAF, PIK3CA Approved in Mar 2021 3DMed 思路迪 NMPA approved NGS panels Notes: 1 The list is not exhaustive. A total of 13 kits have been approved by the NMPA as of the date of this presentation 2 Copy number variation NMPA approved testing kits by major NGS- focused companies1 • Only 30ng DNA input required, applicable to small tissue samples • First NMPA approved NGS kit with CNV2 mutation type, with MET exon14 skipping Highlights on our second NMPA- approved kit
Page 48
48 NGS testing Diagnostics companies focus on steps 1 and 3 Throughput High-throughput Cost Low-cost per biomarker Efficiency Parallel testing Biomarker Profiling Comprehensive profiling & superior accuracy Fit for Liquid Sample Liquid availability NGS Library Preparation EGFR ALK MET Targeted genes enrichment 1 Sequencing 22 Bioinformatics Analysis and Mutation Annotation Final report: including mutation type, targeted therapy, drugs in R&D, clinical trials, etc. 3
Page 49
49 Leading liquid-biopsy product in China, with globally competitive performance Demonstrated in high-impact analytical validation study Source: Slides from “Establishing the analytical validity of circulating tumor DNA sequencing for precision oncology“, 5th Annual Liquid Biopsy for Precision Oncology Summit, Feb 2021 Further information in Appendix 2 SEQC2 Study Overview Liquid Biopsy
Page 50
50 Participating assays and study design Source: “Evaluating the analytical validity of circulating tumor DNA sequencing assays for precision oncology“, Nature Biotechnology, Apr 2021
Page 51
51 Overall analytical accuracy and specificity • The analytical accuracy was measured by Precision- Sensitivity plot (25ng LBx-Low) • The false positive rates were computed by FP/kb region. • Once different VAF threshold increases, FP rates dropped further. Source: “Evaluating the analytical validity of circulating tumor DNA sequencing assays for precision oncology“, Nature Biotechnology, Apr 2021 “To compare the accuracy of the participating ctDNA assays, we generated precision recall curves, ranking known variants and FPs according to their observed VAFs. For Lbx-low samples at 25ng input, BRP was the most accurate assay, with roughly equivalent sensitivity but superior precision to IDT (Fig. 4b and Supplementary Fig. 4c). ”
Page 52
52 Performance – Sensitivity • LBx-low (25 ng input) replicates in each participating assay in different expected VAF bin. Source: “Evaluating the analytical validity of circulating tumor DNA sequencing assays for precision oncology“, Nature Biotechnology, Apr 2021 “The most sensitive assays (IDT and BRP) achieved sensitivity greater than 0.90 for variants with 0.3–0.5% VAF; however, no assays reached this mark for variants with 0.2–0.3% or 0.1– 0.2% VAF (Fig. 4a).” “The performance characteristics of the assays evaluated here were broadly similar to what has been reported by several ctDNA sequencing providers (based on internal testing) that did not participate in this study. During validation of the Guardant360 CDx hybrid capture assay, variants were detected with high sensitivity (~94%) at VAF ≥ 0.4%, declining to ~64% among variants with VAF ranging from 0.05% to 0.25%.” FoundationACT showed ~99% sensitivity for SNVs with VAF > 0.5%, ~95% for 0.25%– 0.5% VAF and ~70% for 0.125– 0.25% VAF.”