BACKGROUND

The TCbHP regimen, consisting of combining docetaxel (T), carboplatin (Cb), trastuzumab (H), and pertuzumab (P), is the preferred neoadjuvant treatment for locally advanced human epidermal growth factor 2 (HER2)-positive breast cancer. However, about 40 % of patients develop resistance to this treatment. Adding TKIs like pyrotinib to anti-HER2 antibodies may enhance efficacy and reduce resistance, but the mechanisms are not fully understood.

METHODS

Imaging mass cytometry (IMC) analyzed tissues from 26 patients treated with NeoPICD (docetaxel, carboplatin, pyrotinib, inetetamab) and 21 patients treated with TCbHP. Cellular changes and spatial relationships were assessed pre- and post-treatment. A co-culture system of tumor cells, fibroblasts, and PBMCs were used to examined cytotoxic T-cell function. A predictive model for treatment outcomes was constructed based on these results.

RESULTS

In TCbHP-sensitive patients, IDOHLA-DR epithelial cells expressing PD-L1 were enriched and interacted with Ki67 T cells and M1 macrophages. In TCbHP-resistant patients, fibroblasts formed a barrier that hindered immune cell access, critical for resistance. NeoPICD disrupted this barrier, enhancing immune cell infiltration and alleviating resistance. Machine learning based on spatial cell architecture can predict treatment outcomes.

CONCLUSION

Spatial organization of cellular interactions in the tumor microenvironment (TME) provides insights into prognosis beyond pathological subtypes. The role of NeoPICD in disruption of fibroblast barriers and enhancement of immune cell function suggests therapeutic advantages in overcoming resistance to anti-HER2 therapies. This research offers new strategies for precision treatment of locally advanced HER2-positive breast cancer.