The use of immunological biomarkers to improve individualization of postoperative radiotherapy in breast cancer

Abstract

Radiotherapy (RT) forms the cornerstone of most curative breast cancer treatment due to its well-established risk-reducing effect on local recurrences at the population level. However, there is heterogeneity regarding treatment benefits at the individual level, and research currently aims to better tailor treatment decisions based on tumor biology. This thesis aimed to investigate if immunological biomarkers from the primary tumor can be used to predict the benefit from RT at the individual level and improve treatment individualization.

Study I: Tumor blocks were collected from the randomized SweBCG91RT cohort, and tumor-infiltrating lymphocytes (TILs) were assessed on whole sections. High levels of TILs were associated with a reduced risk of local recurrence (HR 0.61, CI 95% 0.39-0.96, p=0.033) and a non-significant benefit from RT (HR 0.58, CI 95% 0.28-1.19, p=0.138). Study II: We evaluated CD8+ and FOXP3+ T cells on tissue microarrays (TMAs) to further characterize the lymphocytic immune infiltrate. We found that unirradiated patients with high levels of CD8 and low levels of FOXP3 had a reduced risk of a local recurrence (HR: 0.41, CI 95% 0.19-0.86, p= 0.018 compared to patients without an immune infiltrate (HR 1.0, reference) and appeared to derive a reduced benefit from RT (HR 0.60, CI 95% 0.18-2.0, p=0.41) compared to patients with immune-depleted tumors (HR 0.37, CI 95% 0.24-0.57, p<0.001). Gene expression analyses of anti- and protumoral immune cells yielded similar results. Studies III: We hypothesized that tumor-intrinsic factors modulate an immune infiltrate's biological implications and that immunological biomarkers' predictions would be improved by incorporating this information. We used gene expression analysis to develop two models capturing tumor-intrinsic and immunological characteristics, respectively. We found that the integration of these two dimensions improved the identification of patients with differing RT benefits. We refer to the unpublished manuscript for additional details. Study IV: Expanding on the notion that an enhanced patient stratification for RT benefit is achieved by integrating markers of tumor aggressiveness and immunological activation, we used the clinically available biomarkers to characterize the state of the immune infiltrate and the aggressiveness of the tumor. Again, we observed that this had the potential to improve RT individualization. We refer to the unpublished manuscript for additional details.

Immunological biomarkers from the primary tumor provide independent information on the risk of local recurrences, which can be used to stratify patients according to the need for postoperative radiotherapy. An immune infiltrate's implications depend on tumor-intrinsic characteristics, and successful implementation of immunological biomarkers in clinical practice, therefore, requires a co-analysis of such factors. Tumors with an activated immune response may have a low risk of a local recurrence and constitute a group where de-escalation of RT may be feasible.