AI-driven multi-omics identifies SPP1⁺ tumor-associated macrophages as a therapeutic target and predicts lymph node metastasis in papillary thyroid cancer

Artificial intelligence (AI) is transforming drug design by enabling the systematic discovery of therapeutic targets and the prediction of drug response from complex biomedical data. In papillary thyroid carcinoma (PTC), lymph node metastasis remains a key determinant of poor outcome, yet targetable drivers and predictive models for metastasis are lacking. Here, we integrate single-cell RNA sequencing, spatial transcriptomics, and machine learning to identify SPP1⁺ tumor-associated macrophages (TAMs) as a functionally validated driver of lymph node metastasis in PTC. Using single-cell and spatial analyses, we show that SPP1⁺ TAMs engage TM4SF1⁺ malignant epithelial cells via the SPP1-CD44 axis within a hypoxic niche, driving tumor stemness, invasion, and metastatic progression. SPP1⁺ TAMs exhibit a hypoxic, M2-polarized phenotype and their abundance correlates with worse overall and progression‑free survival in metastatic PTC patients. We further develop an integrative machine-learning model based on SPP1⁺ TAM‑derived molecular features that predicts lymph node status with high accuracy. Preclinically, pharmacological targeting of SPP1 attenuates tumor cell migration, invasion, and lung metastasis, supporting its candidacy as a therapeutic target. Collectively, this study demonstrates how AI‑driven multi‑omics analysis can jointly identify a functionally validated target and generate a clinically deployable predictive model, illustrating a framework for bridging computational discovery with translational oncology.

Authors

Institutions

Publication Details

Journal
npj Precision Oncology
Published
2026-11-20
DOI
https://doi.org/10.1038/s41698-026-01774-7
Primary Topic
Immune cells in cancer
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

AI-driven multi-omics identifies SPP1⁺ tumor-associated macrophages as a therapeutic target and predicts lymph node metastasis in papillary thyroid cancer

Yinde Huang, Zhaokai Zhou, 方申存, Tianyu Zhao et al.
1 links
npj Precision Oncology
Immune cells in cancer
article

AI-driven multi-omics identifies SPP1⁺ tumor-associated macrophages as a therapeutic target and predicts lymph node metastasis in papillary thyroid cancer

Yinde Huang, Zhaokai Zhou, 方申存, Tianyu Zhao, Weixiong Zhu, Run Shi, Jiaxuan Huai, Shaohui Xu, Xiqiao Zhou, Jing Sun, Yongjie Wang, Yingjian Wang, Jianpeng Zhang
article en

Abstract

Artificial intelligence (AI) is transforming drug design by enabling the systematic discovery of therapeutic targets and the prediction of drug response from complex biomedical data. In papillary thyroid carcinoma (PTC), lymph node metastasis remains a key determinant of poor outcome, yet targetable drivers and predictive models for metastasis are lacking. Here, we integrate single-cell RNA sequencing, spatial transcriptomics, and machine learning to identify SPP1⁺ tumor-associated macrophages (TAMs) as a functionally validated driver of lymph node metastasis in PTC. Using single-cell and spatial analyses, we show that SPP1⁺ TAMs engage TM4SF1⁺ malignant epithelial cells via the SPP1-CD44 axis within a hypoxic niche, driving tumor stemness, invasion, and metastatic progression. SPP1⁺ TAMs exhibit a hypoxic, M2-polarized phenotype and their abundance correlates with worse overall and progression‑free survival in metastatic PTC patients. We further develop an integrative machine-learning model based on SPP1⁺ TAM‑derived molecular features that predicts lymph node status with high accuracy. Preclinically, pharmacological targeting of SPP1 attenuates tumor cell migration, invasion, and lung metastasis, supporting its candidacy as a therapeutic target. Collectively, this study demonstrates how AI‑driven multi‑omics analysis can jointly identify a functionally validated target and generate a clinically deployable predictive model, illustrating a framework for bridging computational discovery with translational oncology.

npj Precision OncologyVol. 10(1)
Nanjing University of Chinese Medicine (CN), Chongqing University (CN), German Cancer Research Center (DE), Medical University of Graz (AT), Soochow University (CN), Nanjing Brain Hospital (CN), Lanzhou University Second Hospital (CN), Berlin Institute of Health at Charité - Universitätsmedizin Berlin (DE), Nanjing Chest Hospital (CN), Second Xiangya Hospital of Central South University (CN), First Affiliated Hospital of Zhengzhou University (CN), First People's Hospital of Chongqing (CN), Nanjing Medical University (CN), Chongqing Medical University (CN)
Chongqing Municipal Health Commission, National Natural Science Foundation of China, Natural Science Foundation of Chongqing, Science and Technology Planning Project of Yuzhong District of Chongqing City
Good health and well-being
Immune cells in cancer
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon