Abstract PS3-12-07: Tertiary Lymphoid Structures as Histopathologic Biomarkers in Invasive Breast Cancer: Subtype-Specific Patterns and Clinical Implications

Abstract Invasive breast cancer encompasses a range of tumor subtypes with variable immune landscapes that influence prognosis and treatment response. Tertiary lymphoid structures (TLS), organized immune aggregates resembling secondary lymphoid organs, may serve as histologic indicators of local antitumor immunity. We evaluated clinicopathologic and immune features associated with TLS presence and maturity using histology-based scoring and gene expression–based modeling. TLS and tumor-infiltrating lymphocytes (TILs) were scored on H&E-stained whole slide images (N = 842) from the Carolina Breast Cancer Study (CBCS). Associations with clinicopathologic features and progression-free survival (PFS) were assessed using multinomial logistic and Cox models. Immune expression profiles of TLS groups informed development of weighted centroid classifiers: a 10-gene TLS classifier based on Tfh genes enriched in TLS-positive tumors and B cell genes elevated in tumors with mature (mTLS) vs. only immature (iTLS), and an 18-gene TIL signature derived from remaining lymphocyte markers upregulated in TLS-positive cases. Classifiers were trained and tested on non-overlapping subsets of the CBCS cohort, then applied to the I-SPY2 neoadjuvant trial dataset to assess association with pathologic complete response (pCR). TLS were identified in 38% of tumors and varied in number and location. Subjects were assigned to one of three TLS groups: No TLS, iTLS Only, or mTLS Present (>93% of which also contained iTLS). mTLS Present tumors harbored approximately 2-fold the total TLS compared to iTLS Only and were over twice as likely to contain intratumoral TLS (54.2% vs. 26.3%). TLS presence and maturation were significantly associated with younger age (<50), Black race, and high grade (mTLS ORs 1.66-6.28) as well as ER-negativity and basal-like or ER−/HER2+ subtypes (mTLS ORs 6.70-17.00, all p < 0.001). TLS were also enriched in tumors with genomic instability (mutant-like vs. wild type-like p53, OR = 5.57; high vs. low homologous recombination deficiency scores, OR = 4.21) and high-risk ROR-PT scores (OR = 6.50; all p < 0.001). Stromal TIL scores were strongly correlated with TLS group and total TLS burden (mTLS OR = 21.06, p < 0.001; ρ = 0.54, p < 0.001, respectively). In unadjusted analyses of ER− tumors, both high TIL strength (HR = 0.52, 95% CI: 0.27–0.99, p = 0.047) and presence of mTLS (HR = 0.34, 95% CI: 0.15–0.79, p = 0.012) were significantly associated with improved PFS. However, in multivariable models adjusting for both histologic TLS group and TIL score, only mTLS remained independently associated with improved PFS (HR = 0.39, p = 0.036), while TIL score was not (HR = 0.67, p = 0.27). Our expression-based classifiers mirrored these findings: predicted mTLS was significantly associated with improved PFS in univariate models (HR = 0.42, p = 0.007) and retained near-significance after adjusting for predicted TIL score (HR = 0.46, p = 0.057), while the TIL predictor was no longer significant (HR = 0.90, p = 0.75). In the I-SPY2 neoadjuvant trial (N = 987), predicted mTLS tumors had the highest pCR rates across TLS groups (39% overall, p = 0.055; 52% in HR− subset, p = 0.028). After adjustment for treatment arm, compared to TLS-low tumors, predicted mTLS remained associated with greater odds of pCR in the overall cohort (OR = 1.36, 95% CI: 0.99–1.86, p = 0.054) and the HR− subset (OR = 1.50, 95% CI: 0.96–2.33, p = 0.074), though these relationships did not reach statistical significance. These findings demonstrate that both histologically-defined and gene-predicted mTLS are independently associated with improved survival in ER− breast cancer and identify tumors more likely to respond to neoadjuvant therapy. These results support incorporating TLS assessment into future breast cancer classification and treatment frameworks. Citation Format: A. J. Cozzo, S. C. Van Alsten, L. T. Olsson, Q. Wang, K. A. Hoadley, M. A. Troester. Tertiary Lymphoid Structures as Histopathologic Biomarkers in Invasive Breast Cancer: Subtype-Specific Patterns and Clinical Implications [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS3-12-07.

Authors

Institutions

Publication Details

Journal
Clinical Cancer Research
Published
2026-02-17
DOI
https://doi.org/10.1158/1557-3265.sabcs25-ps3-12-07
Primary Topic
Cancer Immunotherapy and Biomarkers
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Abstract PS3-12-07: Tertiary Lymphoid Structures as Histopathologic Biomarkers in Invasive Breast Cancer: Subtype-Specific Patterns and Clinical Implications

K. A. Hoadley, S. C. Van Alsten, A. J. Cozzo, L. T. Olsson et al.
Clinical Cancer Research
Cancer Immunotherapy and Biomarkers
article

Abstract PS3-12-07: Tertiary Lymphoid Structures as Histopathologic Biomarkers in Invasive Breast Cancer: Subtype-Specific Patterns and Clinical Implications

K. A. Hoadley, S. C. Van Alsten, A. J. Cozzo, L. T. Olsson, M. A. Troester, Q. Wang
article en

Abstract

Abstract Invasive breast cancer encompasses a range of tumor subtypes with variable immune landscapes that influence prognosis and treatment response. Tertiary lymphoid structures (TLS), organized immune aggregates resembling secondary lymphoid organs, may serve as histologic indicators of local antitumor immunity. We evaluated clinicopathologic and immune features associated with TLS presence and maturity using histology-based scoring and gene expression–based modeling. TLS and tumor-infiltrating lymphocytes (TILs) were scored on H&E-stained whole slide images (N = 842) from the Carolina Breast Cancer Study (CBCS). Associations with clinicopathologic features and progression-free survival (PFS) were assessed using multinomial logistic and Cox models. Immune expression profiles of TLS groups informed development of weighted centroid classifiers: a 10-gene TLS classifier based on Tfh genes enriched in TLS-positive tumors and B cell genes elevated in tumors with mature (mTLS) vs. only immature (iTLS), and an 18-gene TIL signature derived from remaining lymphocyte markers upregulated in TLS-positive cases. Classifiers were trained and tested on non-overlapping subsets of the CBCS cohort, then applied to the I-SPY2 neoadjuvant trial dataset to assess association with pathologic complete response (pCR). TLS were identified in 38% of tumors and varied in number and location. Subjects were assigned to one of three TLS groups: No TLS, iTLS Only, or mTLS Present (>93% of which also contained iTLS). mTLS Present tumors harbored approximately 2-fold the total TLS compared to iTLS Only and were over twice as likely to contain intratumoral TLS (54.2% vs. 26.3%). TLS presence and maturation were significantly associated with younger age (<50), Black race, and high grade (mTLS ORs 1.66-6.28) as well as ER-negativity and basal-like or ER−/HER2+ subtypes (mTLS ORs 6.70-17.00, all p < 0.001). TLS were also enriched in tumors with genomic instability (mutant-like vs. wild type-like p53, OR = 5.57; high vs. low homologous recombination deficiency scores, OR = 4.21) and high-risk ROR-PT scores (OR = 6.50; all p < 0.001). Stromal TIL scores were strongly correlated with TLS group and total TLS burden (mTLS OR = 21.06, p < 0.001; ρ = 0.54, p < 0.001, respectively). In unadjusted analyses of ER− tumors, both high TIL strength (HR = 0.52, 95% CI: 0.27–0.99, p = 0.047) and presence of mTLS (HR = 0.34, 95% CI: 0.15–0.79, p = 0.012) were significantly associated with improved PFS. However, in multivariable models adjusting for both histologic TLS group and TIL score, only mTLS remained independently associated with improved PFS (HR = 0.39, p = 0.036), while TIL score was not (HR = 0.67, p = 0.27). Our expression-based classifiers mirrored these findings: predicted mTLS was significantly associated with improved PFS in univariate models (HR = 0.42, p = 0.007) and retained near-significance after adjusting for predicted TIL score (HR = 0.46, p = 0.057), while the TIL predictor was no longer significant (HR = 0.90, p = 0.75). In the I-SPY2 neoadjuvant trial (N = 987), predicted mTLS tumors had the highest pCR rates across TLS groups (39% overall, p = 0.055; 52% in HR− subset, p = 0.028). After adjustment for treatment arm, compared to TLS-low tumors, predicted mTLS remained associated with greater odds of pCR in the overall cohort (OR = 1.36, 95% CI: 0.99–1.86, p = 0.054) and the HR− subset (OR = 1.50, 95% CI: 0.96–2.33, p = 0.074), though these relationships did not reach statistical significance. These findings demonstrate that both histologically-defined and gene-predicted mTLS are independently associated with improved survival in ER− breast cancer and identify tumors more likely to respond to neoadjuvant therapy. These results support incorporating TLS assessment into future breast cancer classification and treatment frameworks. Citation Format: A. J. Cozzo, S. C. Van Alsten, L. T. Olsson, Q. Wang, K. A. Hoadley, M. A. Troester. Tertiary Lymphoid Structures as Histopathologic Biomarkers in Invasive Breast Cancer: Subtype-Specific Patterns and Clinical Implications [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS3-12-07.

Clinical Cancer ResearchVol. 32(4_Supplement)
University of North Carolina at Chapel Hill (US)
Good health and well-being
Openalex Percentile: Top 52%
Cancer Immunotherapy and Biomarkers
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.