BACKGROUND

Patients diagnosed with metastatic breast cancer have poor outcome with a median survival of approximately 2 years. While novel therapeutic options are urgently needed, the great majority of breast cancer research has focused on the primary tumor and less is known about metastatic breast cancer and the prognostic impact of the metastatic tumor microenvironment. Here we investigate the immune landscape in unique clinical material. We explore how the immune landscape changes with metastatic progression and elucidate the prognostic role of immune cells infiltrating primary tumors and corresponding lymph node and more importantly distant metastases.

METHODS

Immunohistochemical staining was performed on human breast cancer tissue microarrays from primary tumors (n = 231), lymph node metastases (n = 129), and distant metastases (n = 43). Infiltration levels of T lymphocytes (CD3), regulatory T lymphocytes (Tregs, FOXP3), macrophages (CD68), and neutrophils (NE) were assessed in primary tumors. T lymphocytes and Tregs were further investigated in lymph node and distant metastases.

RESULTS

T lymphocyte and Treg infiltration were the most clinically important immune cell populations in primary tumors. Infiltration of T lymphocytes and Tregs in primary tumors correlated with proliferation (P = 0.007, P = 0.000) and estrogen receptor negativity (P = 0.046, P = 0.026). While both T lymphocyte and Treg infiltration had a negative correlation to luminal A subtype (P = 0.031, P = 0.000), only Treg infiltration correlated to luminal B (P = 0.034) and triple-negative subtype (P = 0.019). In primary tumors, infiltration of T lymphocytes was an independent prognostic factor for recurrence-free survival (HR = 1.77, CI = 1.01-3.13, P = 0.048), while Treg infiltration was an independent prognostic factor for breast cancer-specific survival (HR = 1.72, CI = 1.14-2.59, P = 0.01). Moreover, breast cancer patients with Treg infiltration in their distant metastases had poor post-recurrence survival (P = 0.039). Treg infiltration levels changed with metastatic tumor progression in 50% of the patients, but there was no significant trend toward neither lower nor higher infiltration.

CONCLUSION

Treg infiltration could have clinical applicability as a prognostic biomarker, deciphering metastatic breast cancer patients with worse prognosis, and accordingly, could be a suitable immunotherapeutic target for patients with metastatic breast cancer. Importantly, half of the patients had changes in Treg infiltration during the course of metastatic progression emphasizing the need to characterize the metastatic immune landscape.