Macrophage signature-based prediction of cancer treatment response using MIL-attention
- 1IBM Research
Abstract
Predicting immunotherapy response from single-cell data remains difficult due to patient-level labels, extreme class imbalance, and highly heterogeneous macrophage states. We present a Multiple Instance Learning (MIL) framework that treats each patient as a bag of macrophage embeddings derived from a single-cell RNA foundation model. The architecture incorporates an attention-based pooling mechanism with reduced model complexity, dropout-enhanced regularization and explicit attention penalties to improve stability in small-sample regimes. To address imbalanced clinical datasets, MIL outputs are optimized with a combined focal loss and supervised contrastive objective that simultaneously sharpens class boundaries and improves representation clustering. Across three cancer datasets, this approach outperforms pseudobulk aggregation, embedding baselines and standard MIL variants. Attention-weighted attribution and transcriptional regulatory analysis reveal distinct macrophage programs, interferon and antigen-presentation networks in responders versus hypoxia-linked regulatory modules in non-responders. This shows the potential of MIL to uncover predictive and mechanistically interpretable immune states.
Related
- PET Imaging of Tumor-Associated Macrophage Dynamics and Response to Clinical Treatment for Potential Translation
- Personalized prediction of response to selective serotonin reuptake inhibitors (SSRIs) treatment for major depression disorder using deep learning networks from connectivity maps of EEG
- Signature Definition Shapes Associations Between Exenatide-Related Transcriptional Programs and Macrophage Senescence Scores in Colorectal Cancer
- Menstrual cycle and hormonal contraceptive intake alter the electrophysiological signature of selective auditory spatial attention
- Development and External Validation of an M2 Macrophage-Related Prognostic Signature in Breast Cancer Using Multi-Omics Data and Machine Learning.
BibTeX
@misc{madgwick2026macrophage,
title = {{Macrophage signature-based prediction of cancer treatment response using MIL-attention}},
author = {M. Madgwick and S. Witham and M. Occhetta and M. Haneklaus and B. Camanzi and M. Smyrnakis and L.-J. Gardiner},
howpublished = {bioRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.28.747791},
url = {https://www.biorxiv.org/content/10.64898/2026.08.28.747791v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Macrophage signature-based prediction of cancer treatment response using MIL-attention
AU - M. Madgwick
AU - S. Witham
AU - M. Occhetta
AU - M. Haneklaus
AU - B. Camanzi
AU - M. Smyrnakis
AU - L.-J. Gardiner
PY - 2026
DA - 2026/08/28
PB - bioRxiv
DO - 10.64898/2026.08.28.747791
UR - https://www.biorxiv.org/content/10.64898/2026.08.28.747791v1
AB - Predicting immunotherapy response from single-cell data remains difficult due to patient-level labels, extreme class imbalance, and highly heterogeneous macrophage states. We present a Multiple Instance Learning (MIL) framework that treats each patient as a bag of macrophage embeddings derived from a single-cell RNA foundation model. The architecture incorporates an attention-based pooling mechanism with reduced model complexity, dropout-enhanced regularization and explicit attention penalties to improve stability in small-sample regimes. To address imbalanced clinical datasets, MIL outputs are optimized with a combined focal loss and supervised contrastive objective that simultaneously sharpens class boundaries and improves representation clustering. Across three cancer datasets, this approach outperforms pseudobulk aggregation, embedding baselines and standard MIL variants. Attention-weighted attribution and transcriptional regulatory analysis reveal distinct macrophage programs, interferon and antigen-presentation networks in responders versus hypoxia-linked regulatory modules in non-responders. This shows the potential of MIL to uncover predictive and mechanistically interpretable immune states.
ER -