NF-κB associated markers of prognosis in early and metastatic triple negative breast cancer

Abstract

Background: Triple negative breast cancer (TNBC) is the most aggressive subtype of breast cancer. While PD-1 based immunotherapies overall have led to improved treatment outcomes for this disease, a diverse response to frontline chemotherapy and immunotherapy still exist in TNBC, highlighting the need for more robust prognostic markers.

Methods: Tumor-intrinsic immunotranscriptomics, serum cytokine profiling, and tumor burden studies were conducted in two syngeneic mouse models to assess differential effects in both the early-stage and metastatic setting. Bioinformatic analyses of both early and metastatic TNBC patient data were performed to assess if identified NF-κB-associated factors are associated with improved patient clinical outcomes.

Results: NF-κB signaling driven by lymphotoxin beta expression is associated with tumor regression in TNBC mouse models. Furthermore, lymphotoxin beta expression in patient TNBC cohorts is prognostic of improved survival outcomes.

Conclusions: This study highlights the potential role for NF-κB-associated factors, specifically lymphotoxin beta to be used as prognostic markers in TNBC, which could ultimately provide insight for improved targeted treatment approaches in the clinic.

Keywords: Biomarkers; Immune checkpoint inhibitors; Immunotherapy; NF-κB pathway; Triple negative breast cancer.

Fig. 1

E0771 ICI monotherapy experimental timeline. Created with BioRender.com

Fig. 2

PD-1 inhibition is associated with enhanced NF-κB signaling in an E0771 TNBC in vivo model. On-treatment NanoString PanCancer IO360 cell type score analysis (n = 3 per group) demonstrates (A) increased infiltration of cytotoxic cells and (B) an elevated NF-κB signaling pathway scores in anti-PD-1 treated tumors. NF-κB signaling score differences were found to be driven by differential expression of Ltb when compared with the (C) anti-TIM-3 group and (D) anti-LAG-3 group. (F) Western blot analysis of on-treatment tumors shows increased levels of p52 and p50, markers for noncanonical and canonical NF-κB signaling, respectively. *p < 0.05, as indicated

Fig. 3

Serum cytokine analysis for E0771 ICI monotherapy study. Post-treatment serum levels of (A) GM-CSF, (B) IFNγ, (C) IL-2, (D) IL-17, (E) IL-9, (F) IL-13, (G) CCL3, (H), CCL4, (I) CXCL1, (J) CXCL10, and (K) CXCL5. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, as indicated

Fig. 4

Pearson correlation analysis of post-treatment serum cytokine levels with tumor burden in E0771 ICI monotherapy study. *p < 0.05, as indicated

Fig. 5

Ltb is associated with increased cytotoxic levels and decreased tumor size. (A) Correlation between Ltb and CTL levels from the TCGA METABRIC TNBC cohort analyzed by the TIDE query gene function. (B) Correlation between Ltb (mRNA expression, Illumina HT-12 V3 microarray) and tumor size (mm) (n = 187). Ltb expression stratified by (C) quartile (n = 56 lower quartile, n = 47, upper quartile) and (D) median tumor size (n = 107 low, n = 80 high). All TNBC subtype data abstracted from the TCGA METABRIC Nature 2012 & Nat Commun 2016 cohort. ***p < 0.0005. ****p < 0.00005, as indicated. Error bars denote standard deviation

Fig. 6

Ltb is associated with improved survival outcomes in TCGA TNBC METABRIC cohort. Correlation analysis between Ltb and (A) RFS (n = 187) and (B) OS (n = 161). Kaplan Meier curve analysis of Ltb’s association with (C) RFS stratified by lower and upper quartile Ltb expression and OS stratified by (D) lower and upper quartile and (E) median Ltb expression. All TNBC subtype data abstracted from the TCGA METABRIC Nature 2012 & Nat Commun 2016 cohort. TCGA, The Cancer Genome Atlas; TNBC, triple negative breast cancer; RFS, relapse free survival; OS, overall survival

Fig. 7

Ltb is associated with improved survival outcomes in TNBC patients treated with anti-PD-1 therapy. Kaplan Meier curve analysis revealed upon stratification of Ltb by upper and lower quartile that higher Ltb expression was significantly associated with improved (A) OS in TNBC patients and (B) TOT in pembrolizumab-treated TNBC patients (n = 1038 patients each in top and bottom quartile). TNBC, triple negative breast cancer; TOT, time on treatment; OS, overall survival

Fig. 8

E0771 and 4T1 TNBC standard-of-care frontline treatment regimen in vivo study. (A) Experimental timeline. Created with BioRender.com. (B) Tumor burden analysis following combinatorial PD-1 inhibition and chemotherapy (carboplatin and paclitaxel) treated mice in the E0771 and 4T1 models (n = 5 per group). Error bars denote standard deviation. *p < 0.05, **p < 0.01, as indicated

Fig. 9

Serum cytokine analysis for E0771 and 4T1 TNBC standard-of-care frontline treatment regimen in vivo study. E0771 serum levels of (A) CXCL1 (KC) and (B) CCL2 (MCP-1). 4T1 serum levels of (C) CCL11 (Eotaxin) and (D) G-CSF. Error bars denote standard deviation. *p < 0.05, **p < 0.01, as indicated

Fig. 10

E0771 and 4T1 TNBC standard-of-care frontline treatment regimen. Pearson correlations of post-treatment serum cytokine levels with tumor burden, with statistically significant values outlined in red