Abstract 175P
Background
EA5163 is a phase 3 randomized trial investigating first-line mono immunotherapy (IO) or IO with chemotherapy in advanced non-small cell lung cancer (NSCLC). Previous work has retrospectively validated a CT based radiomic signature of tumor associated quantitative vessel tortuosity (QVT) for prognosticating outcomes in IO treated NSCLC (Alilou et al. Sci Adv 2022 ). In this study, we report initial results of evaluating QVT in assessing IO related treatment changes in EA5163, specifically associating changes in QVT (Δ-QVT) with changes in tumor volume (Δ-TV) between baseline (B) and 6 week post-treatment scans (TP1).
Methods
CT scans from 29 patients enrolled in EA5163 with measurable lung lesions at B and TP1 were included. Vasculature was automatically segmented in a 25mm region around manually annotated tumors in CT using MATLAB R2022b. 84 QVT features measuring tortuosity, curvature and angle statistics were extracted. No information on outcomes, survival, or treatment arm assignment was available due to the ongoing nature of the trial. Spearman correlation and false discovery rate corrected p-values were computed to associate Δ-QVT with Δ-TV in an unsupervised setting. A risk score (QRS) was generated as the output of a classifier previously trained on 5 Δ-QVT features prognostic of RECIST v1.1 response.
Results
14 Δ-QVT features were significantly (p<0.05) correlated with Δ-TV with an absolute range of 0.45 and 0.76. 4 features in the QRS signature were significantly correlated with Δ-TV (Table). QRS based stratification showed net decrease in TV in low risk patients, whereas a net increase in high-risk patients at TP 1 (p<0.001). The Δ-QVT features were not associated with baseline tumor volume. Table: 175P
Feature | Correlation | p-value |
Vessel volume | 0.71 | <0.001 |
1st bin of tortuosity histogram | 0.65 | 0.002 |
12th bin of angles histogram | 0.61 | 0.004 |
2nd bin of curvature histogram | 0.65 | 0.002 |
Vessel volume to tumor volume ratio | -0.48 | 0.053 |
Conclusions
This study demonstrated an association between CT derived radiomic biomarker Δ-QVT and Δ-TV in an ongoing prospective clinical trial. Validation of these results upon interim analyses or trial completion is warranted.
Clinical trial identification
NCT03793179.
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
This study was coordinated by the ECOG-ACRIN Cancer Research Group (Peter J. O'Dwyer, MD and Mitchell D. Schnall, MD, PhD, Group Co-Chairs) and supported by the National Cancer Institute (NCI) of the National Institutes of Health (NIH) under the following award numbers: U10CA180820, U10CA180794, UG1CA233234, UG1CA233270, and UG1CA233247. This study was also supported by an NCI grant NIH 1 R01CA257612-01A1. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Disclosure
V. Velcheti: Financial Interests, Personal, Advisory Board: ITeos Therapeutics, Bristol Myers Squibb, Merck, AstraZeneca/MedImmune, GSK, Amgen, Elevation Oncology, Merus, Taiho Oncology; Financial Interests, Institutional, Research Funding: Genentech, Trovagene, Eisai, OncoPlex Diagnostics, Alkermes, NantWorks, Genoptix, Altor BioScience, Merck, Bristol Myers Squibb, Atreca, Heat Biologics, Leap Therapeutics, RSIP Vision, GSK. A. Gupta: Financial Interests, Personal, Other, Consultancy for future development of AI tools on chest radiographs: GE Healthcare; Financial Interests, Personal, Stocks/Shares, Own shares of the company presently valued at 56 dollars: Picture Health; Financial Interests, Institutional, Funding, Research support for ongoing AI projects around chest radiography: GE Healthcare. S.S. Ramalingam: Financial Interests, Personal, Other, Editor in Chief, Cancer journal: American Cancer Society; Financial Interests, Research Grant: Merck, AstraZeneca, Advaxis, BMS, Amgen, Takeda, Genmab, GSK. K. Kelly: Financial Interests, Personal, Royalties, Primary reviewer for the small cell lung cancer chapter: UPTODATE. H. Borghaei: Financial Interests, Personal, Advisory Board: BMS, Genentech, Eli Lilly, Merck, EMD-serono, AstraZeneca, Novartis, Genmab, Regeneron, Amgen, Takeda, PharmaMar, Jazz Pharma, Mirati, Daiichi, Guardant, Natera, Oncocyte, BeiGene, iTeo, Boehringer Ingelheim, Puma, BerGenbio, Janssen; Financial Interests, Personal, Other, Training discussion: Pfizer; Financial Interests, Personal, Other, DSMB: Novartis; Financial Interests, Institutional, Other, Clinical trial support: BMS, Amgen; Financial Interests, Personal, Stocks/Shares, Options for scientific advisory role: Sonnetbio; Financial Interests, Personal, Stocks/Shares, Options for scientific advisory board: Nucleai, Inspira (Rgenix); Financial Interests, Institutional, Trial Chair, Investigator initiated trial support: BMS; Financial Interests, Institutional, Coordinating PI, Investigator initiated trial support: Amgen, Lilly; Financial Interests, Personal and Institutional, Trial Chair, Chair steering committee: Miratti; Financial Interests, Personal and Institutional, Steering Committee Member: Amgen, AstraZeneca; Financial Interests, Personal, Steering Committee Member, Also trial support: BMS; Other, DSMB: Novartis, Takeda, Incyte, Springworks. A. Madabhushi: Financial Interests, Personal, Advisory Board, Serve on SAB and consult: SimbioSys; Financial Interests, Personal, Advisory Board: Aiforia, Picture Health; Financial Interests, Personal, Full or part-time Employment: Picture Health; Financial Interests, Personal, Ownership Interest: Picture Health, Elucid Bioimaging, Inspirata Inc; Financial Interests, Personal, Royalties: Picture Health, Elucid Bioimaging; Financial Interests, Institutional, Funding: AstraZeneca, Bristol Myers Squibb, Boehringer Ingelheim, Eli Lilly. All other authors have declared no conflicts of interest.
Resources from the same session
176P - Enhancing immunotherapy response prediction via multimodal integration of radiology and pathology deep learning models
Presenter: Marta Ligero
Session: Poster session 01
177P - Revealing differences in radiosensitivity of advanced non-small cell lung cancer (NSCLC)through single-cell sequencing data
Presenter: Peimeng You
Session: Poster session 01
178P - Explainable radiomics, machine and deep learning models to predict immune-checkpoint inhibitor treatment efficacy in advanced non-small cell lung cancer patients
Presenter: Leonardo Provenzano
Session: Poster session 01
179P - Molecular tumor board directed treatment for patients with advanced stage solid tumors: A case-control study
Presenter: Dhruv Bansal
Session: Poster session 01
180P - An HLA-diet-oriented system unveiling organ-specific occurrence of multiple primary cancers (MPC) with prevention strategy: A large cohort study of 47,550 cancer patients
Presenter: Zixuan Rong
Session: Poster session 01
181P - GeNeo: Agnostic comprehensive genomic profiling versus limited panel organ-directed next-generation sequencing within the Belgian PRECISION initiative
Presenter: Philippe Aftimos
Session: Poster session 01
182P - ALK fusion detection by RNA next-generation sequencing (NGS) compared to DNA in a large, real-world non-small cell lung cancer (NSCLC) dataset
Presenter: Wade Iams
Session: Poster session 01
183P - Frequency of actionable fusions in 7,735 patients with solid tumors
Presenter: Kevin McDonnell
Session: Poster session 01
184P - Patient-specific HLA-I genotypes predict response to immune checkpoint blockade
Presenter: Kyrillus Shohdy
Session: Poster session 01
185P - Real-world data analysis of genomic profiling-matched targeted therapy outcomes in patients with fusion-positive NSCLC
Presenter: Jyoti Patel
Session: Poster session 01