Session Item

High level of tumor infiltrating lymphocytes has been associated with improved survival benefit after postmastectomy radiotherapy (PMRT) in breast cancer (BC) patients. The aim of this study was to examine, if the association is related to specific subsets of T-lymphocytes

The DBCG82bc cohort constitutes high-risk BC patients, diagnosed 1983-90, treated with total mastectomy and partial axillary lymph-node dissection, randomized to +/-PMRT and followed by adjuvant systemic treatment. Multiplex chromogen immunohistochemistry was performed on tissue micro arrays (TMA) constructed from formalin-fixed, paraffin-embedded, treatment-naïve tumor tissue from 1,024 patients. Subsets of T-lymphocytes (e.g., T_{helper cells}-, T_cytotoxic- cells) was visualized by CD4, CD8, and FOXP3, and tumor epithelium by CK7/19. Digital image analysis was used to determine the area fraction of the immune cells (area of immune cells/area of TMA core). A competing risk model, Kaplan-Meier analysis, and multivariate Cox regression analysis were used for correlating area fractions and clinical outcome.

In general, the area fraction of CD8+ lymphocytes in the tumors was low (median: 0.1%, range: 0-6%). When separating patients based on the median value, the group of patients with CD8+ cells < median (“no CD8”) did not benefit from PMRT with respect to risk of distant metastasis (DM) (adjusted Hazard Ratio (adj. HR) = 0.87 (95% confidence intervals (CI): 0.69-1.10)) or risk of death (adj. HR = 0.94 (CI: 0.77-1.15)) (Figure 1). On the contrary, patients with CD8+ cells over the median value (“CD8+”) were found to derive a significant benefit from PMRT in terms of reduced risk of DM (adj. HR = 0.60 (CI: 0.47-0.76)) and reduced risk of death (adj. HR = 0.6 (CI: 0.51-0.77)). The reduction in risk of DM after 20 years in “CD8+” patients treated with PMRT was 16% (from 64% to 48%), and the increase in overall survival (OS) was 16% (from 22% to 38%). The effect on reduced risk of death in the “CD8+” group was further found to be strengthend with increasing levels of CD8 (adj. HR (51-75% percentile) = 0.76 (CI: 0.57,1.00); adj.HR (76-100% percentile) = 0.53 (CI: 0.40,0.71)). PMRT significantly reduced risk of loco-regional recurrence, but the effect was independent of CD8 (adj. HR “no CD8” = 0.25 (CI: 0.16,0.39); adj.HR “CD8+” = 0.21 (CI: 0.13,0.32)). Evaluation of the other markers (CD4, FOXP3) and combinations of markers did not show similar predictive information as for CD8.

No or low level of cytotoxic CD8+ cells may identify a group of BC patients with no significant risk reduction of DM or improvement of OS after PMRT. On the other hand, presence of CD8+ cells in treatment-naïve tumor predict significant benefit of PMRT, and the effect size may vary with increasing level of CD8+ cells in the tumor. The findings may indicate that RT triggers a local immune response that induces a systemic effect, which improves survival through eradication of distant tumor deposits.