AI-powered single-cell imaging and morphology-guided label-free enrichment reveal tumor heterogeneity, genomic relatedness, and divergence in malignant effusions
- 1University of California San Francisco, San Francisco, CA
Abstract
Background: Malignant effusions are clinically accessible yet underutilized sources of metastatic tumor cells. However, abundant background immune cells complicate tumor cell isolation and molecular characterization. We integrate AI-powered single-cell imaging and morphology-guided label-free tumor cell enrichment with genomic profiling to characterize phenotypic and genomic heterogeneity in malignant effusions. Methods: We applied a pre-trained deep learning model to brightfield images of single cells from malignant effusions of 10 patients with metastatic breast cancer to extract morphological features, quantify tumor cell fraction (TCF), and guide label-free microfluidic enrichment. We profiled copy-number alterations and somatic mutations in enriched tumor cell fractions and compared them with available matched effusion cell-free DNA and solid tumors. Results: AI-powered single-cell morphological analysis identified putative tumor cells in all 10 patients (median TCF, 5.4%; range, 0.51-15.8%; n = 25 samples). AI estimates of major immune-cell populations varied across patients and samples and correlated with flow cytometry measurements (n = 21 paired samples, Pearson r = 0.69-0.75). Unsupervised analysis identified four tumor cell morphotypes, revealing intra- and inter-patient morphological heterogeneity and longitudinal variation in morphotype composition. Genomic profiling of enriched tumor cell fractions by low-pass whole-genome sequencing (n = 26 samples) and whole-exome sequencing (n = 24 samples) identified copy-number alterations and somatic mutations, with tumor fraction influencing genomic signal strength. Comparisons of genomic profiles from tumor cell-enriched fractions with matched effusion cell-free DNA and solid tumors revealed shared genomic alterations, providing orthogonal support for tumor origin and relatedness, while differences across analytes revealed genomic divergence and tumor heterogeneity. Conclusion: AI-powered analysis leverages single-cell morphology for tumor cell identification and label-free enrichment, while integration with genomic profiling enables simultaneous interrogation of phenotypic and genomic heterogeneity in malignant effusions. This approach expands the potential of malignant effusions for liquid biopsy applications and biomarker development and discovery.
Related
- Genetic Relatedness of Cambodian Plasmodium falciparum Isolates Was Associated with Geography and Occupation
- Rethinking networks through relatedness - a qualitative study among inpatients in forensic psychiatry.
- Genomic Epidemiology and Genetic Relatedness of Antimicrobial-Resistant Morganella morganii in Community and Hospital Settings in Bangladesh
- Genetic relatedness between clinical and environmental Legionella isolates reveals limited within-system diversity and differences between hot- and cold-water systems.
- Breadth of ACE2 receptor usage predicts host range and antigenic relatedness across bat sarbecoviruses
BibTeX
@misc{magbanua2026aipowered,
title = {{AI-powered single-cell imaging and morphology-guided label-free enrichment reveal tumor heterogeneity, genomic relatedness, and divergence in malignant effusions}},
author = {M. J. M. Magbanua and L. A. Huppert and J. Cruz and J. Goudreau and S. Zhou and K. Yin and P. Stejskal and t. Gill and R. Carelli and J. Kim and L. Hsie and M. Wang and Z. Lian and S. Manepalli and M. Phelan and M. Ray and C. Johnson and J. Nieto and E. C. Schneider and A. Mavropoulos and A. L. Delson and C. Simmons and A. Parizadeh and J. Crawford and M. Bruck and L. Her and V. Barnett and A. M. Cabrera and M. Melisko and R. Balassanian and H. S. Rugo and L. van 't Veer and A. Jovic and W. P. Devine and R. W. Sayaman and M. Salek and M. Masaeli and J. Rosenbluth and H. Goodarzi},
howpublished = {medRxiv},
year = {2026},
month = nov,
doi = {10.64898/2026.11.03.26367106},
url = {https://www.medrxiv.org/content/10.64898/2026.11.03.26367106v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - AI-powered single-cell imaging and morphology-guided label-free enrichment reveal tumor heterogeneity, genomic relatedness, and divergence in malignant effusions
AU - M. J. M. Magbanua
AU - L. A. Huppert
AU - J. Cruz
AU - J. Goudreau
AU - S. Zhou
AU - K. Yin
AU - P. Stejskal
AU - t. Gill
AU - R. Carelli
AU - J. Kim
AU - L. Hsie
AU - M. Wang
AU - Z. Lian
AU - S. Manepalli
AU - M. Phelan
AU - M. Ray
AU - C. Johnson
AU - J. Nieto
AU - E. C. Schneider
AU - A. Mavropoulos
AU - A. L. Delson
AU - C. Simmons
AU - A. Parizadeh
AU - J. Crawford
AU - M. Bruck
AU - L. Her
AU - V. Barnett
AU - A. M. Cabrera
AU - M. Melisko
AU - R. Balassanian
AU - H. S. Rugo
AU - L. van 't Veer
AU - A. Jovic
AU - W. P. Devine
AU - R. W. Sayaman
AU - M. Salek
AU - M. Masaeli
AU - J. Rosenbluth
AU - H. Goodarzi
PY - 2026
DA - 2026/11/03
PB - medRxiv
DO - 10.64898/2026.11.03.26367106
UR - https://www.medrxiv.org/content/10.64898/2026.11.03.26367106v1
AB - Background: Malignant effusions are clinically accessible yet underutilized sources of metastatic tumor cells. However, abundant background immune cells complicate tumor cell isolation and molecular characterization. We integrate AI-powered single-cell imaging and morphology-guided label-free tumor cell enrichment with genomic profiling to characterize phenotypic and genomic heterogeneity in malignant effusions. Methods: We applied a pre-trained deep learning model to brightfield images of single cells from malignant effusions of 10 patients with metastatic breast cancer to extract morphological features, quantify tumor cell fraction (TCF), and guide label-free microfluidic enrichment. We profiled copy-number alterations and somatic mutations in enriched tumor cell fractions and compared them with available matched effusion cell-free DNA and solid tumors. Results: AI-powered single-cell morphological analysis identified putative tumor cells in all 10 patients (median TCF, 5.4%; range, 0.51-15.8%; n = 25 samples). AI estimates of major immune-cell populations varied across patients and samples and correlated with flow cytometry measurements (n = 21 paired samples, Pearson r = 0.69-0.75). Unsupervised analysis identified four tumor cell morphotypes, revealing intra- and inter-patient morphological heterogeneity and longitudinal variation in morphotype composition. Genomic profiling of enriched tumor cell fractions by low-pass whole-genome sequencing (n = 26 samples) and whole-exome sequencing (n = 24 samples) identified copy-number alterations and somatic mutations, with tumor fraction influencing genomic signal strength. Comparisons of genomic profiles from tumor cell-enriched fractions with matched effusion cell-free DNA and solid tumors revealed shared genomic alterations, providing orthogonal support for tumor origin and relatedness, while differences across analytes revealed genomic divergence and tumor heterogeneity. Conclusion: AI-powered analysis leverages single-cell morphology for tumor cell identification and label-free enrichment, while integration with genomic profiling enables simultaneous interrogation of phenotypic and genomic heterogeneity in malignant effusions. This approach expands the potential of malignant effusions for liquid biopsy applications and biomarker development and discovery.
ER -