# Tumour microenvironment and HER2-low status dictate response and resistance to anthracycline-taxane chemotherapy in premenopausal TNBC: a retrospective multicohort study

**Authors:** Shuanglong Cai, Shaohong Yu, Quan Zhou, Fangyuan Kuang, Lingyan Rao, Zhen Fang, Xiaoxin Zheng, Yong Shi, Jingdan Li

PMC · DOI: 10.3389/fphar.2026.1752328 · Frontiers in Pharmacology · 2026-02-09

## TL;DR

This study shows that the tumor environment and HER2-low status affect how premenopausal TNBC patients respond to standard chemotherapy, offering new ways to personalize treatment.

## Contribution

The study identifies sTIL levels and HER2-low status as novel biomarkers for predicting chemotherapy response in premenopausal TNBC.

## Key findings

- Low stromal tumor-infiltrating lymphocytes (sTIL) and HER2-low status are linked to worse disease-free survival in premenopausal TNBC patients.
- A predictive nomogram combining sTIL, HER2 status, and staging achieved high accuracy in predicting chemotherapy outcomes.
- HER2-low tumors may benefit from antibody-drug conjugates, while sTIL-low tumors could respond better to immunomodulatory therapies.

## Abstract

Premenopausal women with triple-negative breast cancer (TNBC) exhibit considerable heterogeneity in their response to standard anthracycline and taxane-based chemotherapy, yet the underlying mechanisms remain poorly understood. We aimed to investigate the combined role of the tumour immune microenvironment and HER2-low status in predicting chemosensitivity and intrinsic resistance in this specific population.

We retrospectively analysed data from 767 premenopausal patients with TNBC across two Chinese medical centres. All patients underwent primary surgery followed by adjuvant chemotherapy based on anthracyclines and taxanes. Patients were randomly assigned to training and internal validation cohorts. Independent predictors of DFS were identified using multivariable Cox proportional hazards regression, and a nomogram was constructed accordingly. The model’s discrimination was assessed using the concordance index (C-index) and time-dependent receiver operating characteristic (ROC) curve analysis, while calibration was evaluated with calibration curves.

Multivariable analysis identified lower stromal tumour-infiltrating lymphocyte (sTIL) expression levels, and HER2 IHC 2+/FISH-negative status as independent factors associated with poorer DFS, besides that T3/T4 staging, higher N staging. A nomogram integrating these four variables demonstrated excellent predictive accuracy for DFS, with a C-index of 0.862 in the training set and 0.861 in the validation set. The area under the ROC curve (AUC) for predicting 3-year DFS was 0.907 and 0.908 in the training and validation sets, respectively.

Our findings reveal that an immune-poor tumour microenvironment and HER2-low biology are key, complementary determinants of intrinsic resistance to standard chemotherapy in premenopausal TNBC. These readily available biomarkers provide a mechanistic rationale for patient stratification, suggesting that sTIL-low tumours might benefit from immunomodulatory strategies, while HER2-low tumours represent a candidate population for novel antibody-drug conjugates. This paradigm shift from empirical to biomarker-informed therapy could help overcome chemoresistance in this high-risk group.

## Linked entities

- **Proteins:** ERBB2 (erb-b2 receptor tyrosine kinase 2)
- **Chemicals:** taxane (PubChem CID 9548828)
- **Diseases:** triple-negative breast cancer (MONDO:0005494)

## Full-text entities

- **Genes:** BRCA1 (BRCA1 DNA repair associated) [NCBI Gene 672] {aka BRCAI, BRCC1, BROVCA1, FANCS, IRIS, PNCA4}, ESR1 (estrogen receptor 1) [NCBI Gene 2099] {aka ER, ESR, ESRA, ESTRR, Era, NR3A1}, COX8A (cytochrome c oxidase subunit 8A) [NCBI Gene 1351] {aka COX, COX8, COX8-2, COX8L, MC4DN15, VIII}, EGFR (epidermal growth factor receptor) [NCBI Gene 1956] {aka ERBB, ERBB1, ERRP, HER1, NISBD2, NNCIS}, PIK3CB (phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit beta) [NCBI Gene 5291] {aka P110BETA, PI3K, PI3KBETA, PIK3C1}, PIK3CA (phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha) [NCBI Gene 5290] {aka CCM4, CLAPO, CLOVE, CWS5, HMH, MCAP}, ESR1 (estrogen receptor 1) [NCBI Gene 396099], ERBB2 (erb-b2 receptor tyrosine kinase 2) [NCBI Gene 2064] {aka CD340, HER-2, HER-2/neu, HER2, MLN 19, MLN-19}, STIL (STIL centriolar assembly protein) [NCBI Gene 6491] {aka MCPH7, SIL}, PGR (progesterone receptor) [NCBI Gene 396198] {aka NR3C3}, PGR (progesterone receptor) [NCBI Gene 5241] {aka NR3C3, PR}, EREG (epiregulin) [NCBI Gene 2069] {aka EPR, ER, Ep}, PARP1 (poly(ADP-ribose) polymerase 1) [NCBI Gene 142] {aka ADPRT, ADPRT 1, ADPRT1, ARTD1, PARP, PARP-1}, AKT1 (AKT serine/threonine kinase 1) [NCBI Gene 207] {aka AKT, PKB, PKB-ALPHA, PRKBA, RAC, RAC-ALPHA}
- **Diseases:** Breast Cancer (MESH:D001943), TNBC (MESH:D064726), toxicity (MESH:D064420), N (MESH:C536108), metastasis (MESH:D009362), death (MESH:D003643), DFS (MESH:D011475), Tumour (MESH:D009369), inflammatory (MESH:D007249)
- **Chemicals:** taxane (MESH:C080625), taxanes (MESH:D043823), capecitabine (MESH:D000069287), deruxtecan (-), trastuzumab (MESH:D000068878), Anthracycline (MESH:D018943)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12926388/full.md

## Figures

14 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12926388/full.md

## References

48 references — full list in the complete paper: https://tomesphere.com/paper/PMC12926388/full.md

---
Source: https://tomesphere.com/paper/PMC12926388