Dataset Information

Fully automated MR-based virtual biopsy of primary CNS lymphomas.

ABSTRACT:

Background

Primary central nervous system lymphomas (PCNSL) pose a challenge as they may mimic gliomas on magnetic resonance imaging (MRI) imaging, compelling precise differentiation for appropriate treatment. This study focuses on developing an automated MRI-based workflow to distinguish between PCNSL and gliomas.

Methods

MRI examinations of 240 therapy-naive patients (141 males and 99 females, mean age: 55.16 years) with cerebral gliomas and PCNSLs (216 gliomas and 24 PCNSLs), each comprising a non-contrast T1-weighted, fluid-attenuated inversion recovery (FLAIR), and contrast-enhanced T1-weighted sequence were included in the study. HD-GLIO, a pre-trained segmentation network, was used to generate segmentations automatically. To validate the segmentation efficiency, 237 manual segmentations were prepared (213 gliomas and 24 PCNSLs). Subsequently, radiomics features were extracted following feature selection and training of an XGBoost algorithm for classification.

Results

The segmentation models for gliomas and PCNSLs achieved a mean Sørensen-Dice coefficient of 0.82 and 0.80 for whole tumors, respectively. Three classification models were developed in this study to differentiate gliomas from PCNSLs. The first model differentiated PCNSLs from gliomas, with an area under the curve (AUC) of 0.99 (F1-score: 0.75). The second model discriminated between high-grade gliomas and PCNSLs with an AUC of 0.91 (F1-score: 0.6), and the third model differentiated between low-grade gliomas and PCNSLs with an AUC of 0.95 (F1-score: 0.89).

Conclusions

This study serves as a pilot investigation presenting an automated virtual biopsy workflow that distinguishes PCNSLs from cerebral gliomas. Prior to clinical use, it is necessary to validate the results in a prospective multicenter setting with a larger number of PCNSL patients.

SUBMITTER: Parmar V

PROVIDER: S-EPMC10956963 | biostudies-literature | 2024 Jan-Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Fully automated MR-based virtual biopsy of primary CNS lymphomas.

Parmar Vicky V Haubold Johannes J Salhöfer Luca L Meetschen Mathias M Wrede Karsten K Glas Martin M Guberina Maja M Blau Tobias T Bos Denise D Kureishi Anisa A Hosch René R Nensa Felix F Forsting Michael M Deuschl Cornelius C Umutlu Lale L

Neuro-oncology advances 20240101 1

<h4>Background</h4>Primary central nervous system lymphomas (PCNSL) pose a challenge as they may mimic gliomas on magnetic resonance imaging (MRI) imaging, compelling precise differentiation for appropriate treatment. This study focuses on developing an automated MRI-based workflow to distinguish between PCNSL and gliomas.<h4>Methods</h4>MRI examinations of 240 therapy-naive patients (141 males and 99 females, mean age: 55.16 years) with cerebral gliomas and PCNSLs (216 gliomas and 24 PCNSLs), e ...[more]

PMID: 38516329

Similar Datasets

Project description:Primary Central Nervous System Lymphoma (PCNSL) is a lymphoid malignancy of the brain that occurs in ~1500 patients per year in the US. PCNSL can spread to the vitreous and retina, where it is known as vitreoretinal lymphoma (VRL). While confirmatory testing for diagnosis is dependent on invasive brain tissue or cerebrospinal fluid sampling, the ability to access the vitreous as a proximal biofluid for liquid biopsy to diagnose PCNSL is an attractive prospect given ease of access and minimization of risks and complications from other biopsy strategies. However, the extent to which VRL, previously considered genetically identical to PCNSL, resembles PCNSL in the same individual with respect to genetic alterations, diagnostic strategies, and precision-medicine based approaches has hitherto not been explored. Furthermore, the degree of intra-patient tumor genomic heterogeneity between the brain and vitreous sites has not been studied. In this work, we report on targeted DNA next-generation sequencing (NGS) of matched brain and vitreous samples in two patients who each harbored VRL and PCSNL. Our strategy showed enhanced sensitivity for molecular diagnosis confirmation over current clinically used vitreous liquid biopsy methods. We observed a clonal relationship between the eye and brain samples in both patients, which carried clonal CDKN2A deep deletions, a highly recurrent alteration in VRL patients, as well as MYD88 p.L265P activating mutation in one patient. Several subclonal alterations, however, in the genes SETD2, BRCA2, TERT, and broad chromosomal regions showed heterogeneity between the brain and the eyes, between the two eyes, and among different regions of the PCNSL brain lesion. Taken together, our data show that NGS of vitreous liquid biopsies in PCNSL patients with VRL highlights shared and distinct genetic alterations that suggest a common origin for these lymphomas, but with additional site-specific alterations. Liquid biopsy of VRL accurately replicates the findings for PCNSL truncal (tumor-initiating) genomic alterations; it can also nominate precision medicine interventions and shows intra-patient heterogeneity in subclonal alterations. To the best of our knowledge, this study represents the first interrogation of genetic underpinnings of PCNSL with matched VRL samples. Our findings support continued investigation into the utility of vitreous liquid biopsy in precision diagnosis and treatment of PCNSL/VRL.

Dataset Information

Fully automated MR-based virtual biopsy of primary CNS lymphomas.

Background

Methods

Results

Conclusions

Publications

Fully automated MR-based virtual biopsy of primary CNS lymphomas.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets