Spatial transcriptomics reveals tumor microenvironment-driven subtypes of invasive lobular carcinoma

Area of Science:

• Oncology
• Genomics
• Bioinformatics

Background:

• Invasive lobular carcinoma (ILC) is the second most common breast cancer subtype with unique clinical and biological behaviors.
• Current molecular classifications often neglect the complex spatial organization and tumor microenvironment (TME) of ILC.
• Understanding ILC's TME is crucial for accurate prognostication and targeted therapies.

Purpose of the Study:

• To characterize the spatial organization and TME heterogeneity in hormone receptor-positive, HER2-negative (HR+/HER2-) ILC.
• To develop a multimodal classification system for ILC integrating spatial, morphological, and gene expression data.
• To assess the prognostic value of the novel ILC classification.

Main Methods:

• Spatial transcriptomics was performed on 43 HR+/HER2- ILC tumors.
• Integration of spatial gene expression with histology and single-cell deconvolution.
• Development of a multimodal classification (ILC4TME) based on spatial metrics, morphology, and gene expression.

Main Results:

• High inter- and intratumor heterogeneity was observed in the ILC TME.
• Four distinct ILC subtypes (NSE, P, ARE, MIE) were identified, collectively termed ILC4TME.
• The ILC4TME classification demonstrated prognostic value in external datasets and multivariable models, outperforming existing signatures.

Conclusions:

• The ILC4TME classification captures spatial and cellular heterogeneity missed by current approaches.
• This multimodal classification refines risk stratification and offers a framework for targeted therapies in ILC.
• Specific subtypes (e.g., Proliferative) are linked to prognosis, even with standard endocrine therapy.