Enhancing NSCLC recurrence prediction with PET/CT habitat imaging, ctDNA, and integrative radiogenomics-blood insights

Sheeba J Sujit^#¹, Muhammad Aminu^#¹, Tatiana V Karpinets², Pingjun Chen¹, Maliazurina B Saad¹, Morteza Salehjahromi¹, John D Boom^{1

3}, Mohamed Qayati¹, James M George^{1

4}, Haley Allen⁵, Mara B Antonoff⁶, Lingzhi Hong^{1

7}, Xin Hu⁷, Simon Heeke⁷, Hai T Tran⁷, Xiuning Le⁷, Yasir Y Elamin⁷, Mehmet Altan⁷, Natalie I Vokes^{2

7}, Ajay Sheshadri⁸, Julie Lin⁸, Jianhua Zhang², Yang Lu⁹, Carmen Behrens⁷, Myrna C B Godoy¹⁰, Carol C Wu¹⁰, Joe Y Chang¹¹, Caroline Chung^{11

12}, David A Jaffray^{1

12

13}, Ignacio I Wistuba¹⁴, J Jack Lee¹⁵, Ara A Vaporciyan⁶, Don L Gibbons⁷, John Heymach⁷, Jianjun Zhang^{2

7

16

17}, Tina Cascone⁷, Jia Wu^{18

19

20}

Affiliations

¹ Department of Imaging Physics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
² Department of Genomic Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
³ Department of Biomedical Engineering, Duke University, Durham, NC, USA.
⁴ Michigan Center for Translational Pathology, University of Michigan, Ann Arbor, MI, USA.
⁵ Natural Sciences, Rice University, Houston, TX, USA.
⁶ Department of Thoracic and Cardiovascular Surgery, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
⁷ Department of Thoracic/Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
⁸ Department of Pulmonary Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
⁹ Department of Nuclear Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁰ Department of Thoracic Imaging, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹¹ Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹² Institute of Data Science in Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹³ Department of Radiation Physics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁴ Department of Translational Molecular Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁵ Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁶ Lung Cancer Genomics Program, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁷ Lung Cancer Interception Program, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁸ Department of Imaging Physics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA. jwu11@mdanderson.org.
¹⁹ Department of Thoracic/Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA. jwu11@mdanderson.org.
²⁰ Institute of Data Science in Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA. jwu11@mdanderson.org.

^# Contributed equally.

PMID: 38605064
PMCID: PMC11009351
DOI: 10.1038/s41467-024-47512-0

Enhancing NSCLC recurrence prediction with PET/CT habitat imaging, ctDNA, and integrative radiogenomics-blood insights

Sheeba J Sujit et al. Nat Commun. 2024.

. 2024 Apr 11;15(1):3152.

doi: 10.1038/s41467-024-47512-0.

Authors

Affiliations

¹ Department of Imaging Physics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
² Department of Genomic Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
³ Department of Biomedical Engineering, Duke University, Durham, NC, USA.
⁴ Michigan Center for Translational Pathology, University of Michigan, Ann Arbor, MI, USA.
⁵ Natural Sciences, Rice University, Houston, TX, USA.
⁶ Department of Thoracic and Cardiovascular Surgery, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
⁷ Department of Thoracic/Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
⁸ Department of Pulmonary Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
⁹ Department of Nuclear Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁰ Department of Thoracic Imaging, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹¹ Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹² Institute of Data Science in Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹³ Department of Radiation Physics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁴ Department of Translational Molecular Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁵ Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁶ Lung Cancer Genomics Program, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁷ Lung Cancer Interception Program, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁸ Department of Imaging Physics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA. jwu11@mdanderson.org.
¹⁹ Department of Thoracic/Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA. jwu11@mdanderson.org.
²⁰ Institute of Data Science in Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA. jwu11@mdanderson.org.

^# Contributed equally.

PMID: 38605064
PMCID: PMC11009351
DOI: 10.1038/s41467-024-47512-0

Abstract

While we recognize the prognostic importance of clinicopathological measures and circulating tumor DNA (ctDNA), the independent contribution of quantitative image markers to prognosis in non-small cell lung cancer (NSCLC) remains underexplored. In our multi-institutional study of 394 NSCLC patients, we utilize pre-treatment computed tomography (CT) and ¹⁸F-fluorodeoxyglucose positron emission tomography (FDG-PET) to establish a habitat imaging framework for assessing regional heterogeneity within individual tumors. This framework identifies three PET/CT subtypes, which maintain prognostic value after adjusting for clinicopathologic risk factors including tumor volume. Additionally, these subtypes complement ctDNA in predicting disease recurrence. Radiogenomics analysis unveil the molecular underpinnings of these imaging subtypes, highlighting downregulation in interferon alpha and gamma pathways in the high-risk subtype. In summary, our study demonstrates that these habitat imaging subtypes effectively stratify NSCLC patients based on their risk levels for disease recurrence after initial curative surgery or radiotherapy, providing valuable insights for personalized treatment approaches.

PubMed Disclaimer

Conflict of interest statement

X.L. reports Consulting/advisory fees from Eli Lilly, EMD Serono (Merck KGaA), AstraZeneca, Spectrum Pharmaceutics, Novartis, Regeneron, Boehringer Ingelheim, Hengrui Therapeutics, Bayer, Teligene, Taiho, Daiichi Sankyo, Janssen, Blueprint Medicines, Sensei Biotherapeutics, SystImmune, ArriVent, Abion, and Abbvie Research Funding to Institution from Eli Lilly, EMD Serono, ArriVent, Dizal, Teligene, Regeneron, Janssen, ThermoFisher, Takeda, and Boehringer Ingelheim. Travel Support from EMD Serono, Janssen, and Spectrum Pharmaceutics. Stock options from BlossomHill. T.C. reports speaker fees and honoraria from The Society for Immunotherapy of Cancer, Bristol Myers Squibb, Roche, Medscape, and PeerView; having an advisory role or receiving consulting fees from AstraZeneca, Bristol Myers Squibb, EMD Serono, Merck & Co, Genentech, and Arrowhead Pharmaceuticals; and institutional research funding from AstraZeneca, Bristol Myers Squibb, and EMD Serono. N.I.V. receives consulting fees from Sanofi, Regeneron, Oncocyte, and Eli Lilly; and research funding from Mirati outside the submitted work. J.Y.C. reports research funding from BMS-MDACC, Siemens Healthcare, and consultation fees from Legion Healthcare Partners. L.Y. has grant support from Lantheus Inc. M.C.B.G. has received research funding from Siemens Healthcare. I.W. has received honoraria from Genentech/Roche, Astra Zeneca, Merck, Guardant Health, Flame, Novartis, Sanofi, Daiichi Sankyo, Dava Oncology, Amgen, GlaxoSmithKline, HTG Molecular, Jansen, Merus, Imagene, G1 Therapeutics, Abbvie, Catalyst Therapeutics, Genzyme, Regeneron, Oncocyte, Medscape, Platform Health, Pfizer, Physicians’ Education Resource, HPM Education, and Aptitute Health; Additionally, I.W. has received research support from Genentech, Merck, Bristol-Myers Squibb, Medimmune, Adaptive, Adaptimmune, EMD Serono, Pfizer, Takeda, Amgen, Karus, Johnson & Johnson, Bayer, Iovance, 4D, Novartis, and Akoya. D.L.G. has served on scientific advisory committees for Menarini Ricerche, 4D Pharma, Onconova, and Eli Lilly and has received research support from Takeda, Astellas, NGM Biopharmaceuticals, Boehringer Ingelheim and AstraZeneca. J.V.H. reports being on scientific advisory boards for AstraZeneca, Boehringer Ingelheim, Genentech, GlaxoSmithKline, Eli Lilly, Novartis, Spectrum, EMD Serono, Sanofi, Takeda, Mirati Therapeutics, BMS, and Janssen Global Services; receiving research support from AstraZeneca, Takeda, Boehringer Ingelheim, and Spectrum; and receiving licensing fees from Spectrum. C.C.W reports research support from the Medical Imaging and Data Resource Center from NIBIB/University of Chicago and royalties from Elsevier. J.Z. reports serving on the consulting/advisory board of Bristol-Myers Squibb, AstraZeneca, Novartis, Johnson & Johnson, GenePlus, Innovent, Varian, and Catalyst; receiving research grants to institutions from Merck, Novartis, and Johnson & Johnson. J.W. reports research funding from Siemens Healthcare. The remaining authors declare no competing interests.

Figures

**Fig. 1. Study cohorts and their characteristics.**
a The discovery cohort included 75 patients from the PROSPECT database and 124 patients from the TCIA database. b The validation cohorts consisted of 133 patients from the ICON database and 62 patients from ACRIN database.

**Fig. 2. Study design of habitat imaging–based radiomic analysis of NSCLC patients.**
a Overview of the data collected and pre-processing stages of the CT and ¹⁸F-FDG PET images. The pre-processing involved ¹⁸F-FDG PET to CT registration and fusion of segmented tumor regions from ¹⁸F-FDG PET and CT images along with their local entropy maps. b The habitat imaging analysis framework consisted of a 2-stage clustering process: Individual-level clustering, where tumor regions of each patient were over-segmented into superpixels; and population-level clustering, where clustering was performed on superpixels pooled from all patients to discover distinct tumor subregions. The multiregional spatial interaction (MSI) matrix summarizes the co-occurrence statistics among different tumor subregions. The 92 MSI features extracted from the MSI matrix identified patient subtypes. c Radiogenomic analysis along with ctDNA metrics confirm the clinical and biological relevance of the identified imaging subtypes. Created with BioRender.com.

**Fig. 3. Systematic benchmarks of unsupervised clustering analysis of tumor region.**
Visualization in UMAP: a distribution of superpixels across the discovery (PROSPECT, TCIA) and validation ICON cohorts. b the 8 clusters identified using Louvain clustering by dimension reduction. c Imaging interpretation of the eight cluster regions using High, Low, Intermediate levels. d Examples of habitat maps of patients with disease recurrence and no recurrence from the discovery set. Rows 1, 2 and 3 show habitat maps of patients who has recurrence of disease after 5, 47 and 37.5 months, respectively. They show high volume of clusters 7 and 8. Rows 4, 5 and 6 show habitat maps of patients who has no recurrence of disease showing high volume of clusters 3 and 4.

**Fig. 4. Imaging subtypes were prognostic for recurrence-free survival in NSCLC patients.**
Kaplan-Meier curve comparing RFS of individuals with low risk (green), high risk (red), and intermediate risk (purple) subtypes with P = 1e-05 by log-rank test in the discovery cohort (a) and P = 0.024 by log-rank test in the validation cohort (b). Kaplan-Meier curve comparing OS of individuals with low risk (green), high risk (red), and intermediate risk (purple) subtypes with P = 0.037 by log-rank test in the training cohort (c) P = 0.0017 by log-rank test in the internal validation cohort (d) and P = 0.0201 by log-rank test in the external validation cohort (e). Source data are provided as a Source Data file.

**Fig. 5. Workflow of habitat imaging–based intratumor partitioning on chest CT and PET images.**
Example patients with similar baseline clinical characteristics in the low-risk, intermediate-risk, and high-risk subtype groups. a CT and b PET scans. c, d are the fused 3D tumor volume oversegmented to superpixels and integrated by clustering using the Louvain algorithm to identify the habitat regions. Created with BioRender.com.

**Fig. 6. Circulating tumor DNA (ctDNA) and gene set enrichment analysis.**
a Bar plot showing the percentage of patients with ctDNA detected before surgery, stratified across different risk groups. b Bar plot showing the percentage of patients with ctDNA clearance status (Persistent, Cleared, Never Detected) after SOC therapy, stratified across different risk groups. c C-index comparison of three different RFS prediction models using combinations of clinical features, tumor volume, imaging subtypes, and serial ctDNA (ctDNA detection at pre-surgery and ctDNA clearance status after SOC therapy). d Top 10 Hallmark Pathways differentially active in high- versus low-risk group according to the meta-analysis of the studied cohorts. Source data are provided as a Source Data file.

See this image and copyright information in PMC

References

1. Forde PM, et al. Neoadjuvant nivolumab plus chemotherapy in resectable lung cancer. N. Engl. J. Med. 2022;386:1973–1985. doi: 10.1056/NEJMoa2202170. - DOI - PMC - PubMed
1. Cascone T, et al. Neoadjuvant chemotherapy plus nivolumab with or without ipilimumab in operable non-small cell lung cancer: the phase 2 platform NEOSTAR trial. Nat. Med. 2023;29:593–604. doi: 10.1038/s41591-022-02189-0. - DOI - PMC - PubMed
1. Heymach JV, et al. Perioperative durvalumab for resectable non–small-cell lung cancer. N. Engl. J. Med. 2023;389:1672–1684. doi: 10.1056/NEJMoa2304875. - DOI - PubMed
1. Mu W, et al. Non-invasive decision support for NSCLC treatment using PET/CT radiomics. Nat. Commun. 2020;11:5228. doi: 10.1038/s41467-020-19116-x. - DOI - PMC - PubMed
1. Vaidya P, et al. CT derived radiomic score for predicting the added benefit of adjuvant chemotherapy following surgery in stage I, II resectable non-small cell lung cancer: a retrospective multicohort study for outcome prediction. Lancet Digital Health. 2020;2:e116–e128. doi: 10.1016/S2589-7500(20)30002-9. - DOI - PubMed

MeSH terms

Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions

Save citation to file

Email citation

Add to Collections

Add to My Bibliography

Your saved search

Create a file for external citation management software

Your RSS Feed

Enhancing NSCLC recurrence prediction with PET/CT habitat imaging, ctDNA, and integrative radiogenomics-blood insights

Affiliations

Enhancing NSCLC recurrence prediction with PET/CT habitat imaging, ctDNA, and integrative radiogenomics-blood insights

Authors

Affiliations

Abstract

Conflict of interest statement

Figures

References

MeSH terms

Substances

Grants and funding

LinkOut - more resources

Full Text Sources