A different immunologic profile characterizes patients with HER-2-overexpressing and HER-2-negative locally advanced breast cancer: implications for immune-based therapies

Abstract

Introduction: The clinical efficacy of trastuzumab and taxanes is at least partly related to their ability to mediate or promote antitumor immune responses. On these grounds, a careful analysis of basal immune profile may be capital to dissect the heterogeneity of clinical responses to these drugs in patients with locally advanced breast cancer undergoing neoadjuvant chemotherapy.

Methods: Blood samples were collected from 61 locally advanced breast cancers (36 HER2- and 25 HER2+) at diagnosis and from 23 healthy women. Immunophenotypic profiling of circulating and intratumor immune cells, including regulatory T (Treg) cells, was assessed by flow cytometry and immunohistochemistry, respectively. Serum levels of 10 different cytokines were assessed by multiplex immunoassays. CD8+ T cell responses to multiple tumor-associated antigens (TAA) were evaluated by IFN-γ-enzyme-linked immunosorbent spot (ELISPOT). The Student's t test for two tailed distributions and the Wilcoxon two-sample test were used for the statistical analysis of the data.

Results: The proportion of circulating immune effectors was similar in HER2+ patients and healthy donors, whereas higher percentages of natural killer and Treg cells and a lower CD4+/CD8+ T cell ratio (with a prevalence of naïve and central memory CD8+ T cells) were observed in HER2- cases. Higher numbers of circulating CD8+ T cells specific for several HLA-A*0201-restricted TAA-derived peptides were observed in HER2+ cases, together with a higher prevalence of intratumor CD8+ T cells. Serum cytokine profile of HER2+ patients was similar to that of controls, whereas HER2- cases showed significantly lower cytokine amounts compared to healthy women (IL-2, IL-8, IL-6) and HER2+ cases (IL-2, IL-1β, IL-8, IL-6, IL-10).

Conclusions: Compared to HER2- cases, patients with HER2-overexpressing locally advanced breast cancer show a more limited tumor-related immune suppression. This may account for the clinical benefit achieved in this subset of patients with the use of drugs acting through, but also promoting, immune-mediated effects.

Figure 1

Immunophenotypic characterization of peripheral blood lymphocytes. Percentages of (a) natural killer cells (NK; CD3^-CD16⁺CD56⁺; neg = 20, pos = 17, ctrl = 17), (b) B cells (CD3^-CD19⁺; neg = 14, pos = 15, ctrl = 15), (c) T lymphocytes (CD3⁺CD19^-; neg = 20, pos = 17, ctrl = 17), (d) CD4⁺/CD8⁺ ratio (CD3⁺CD4⁺/CD3⁺CD8⁺; neg = 20, pos = 17, ctrl = 17), and (e) regulatory T cells (Treg; CD3⁺CD4⁺CD25^highCD127^lowFoxP3⁺; neg = 20, pos = 15, ctrl = 17) assessed in HER2^- (neg), HER2-overexpressing (pos) patients and age-matched healthy women (ctrl). (f) Differentiation (memory) status of CD3⁺CD4⁺ and CD3⁺CD8⁺ lymphocytes was investigated through CCR7 and CD45RA expression (neg = 12; pos = 10; ctrl = 10). Statistical analysis was performed with t Student test. *, P<0.05 (refer to text for exact P value). HER-2, human epidermal growth factor receptor-2.

Figure 2

Composite figure showing CD8 and FoxP3 expression in lymphoid cells infiltrating representative HER2^- or HER2⁺ breast carcinomas. (a) Few CD8⁺ cells are present within the tumor, and infiltrate tumor nests in a HER2^- case. (b) Some FoxP3⁺ cells infiltrate a HER2^- breast carcinoma. (c) CD8⁺ cells are numerous and surround tumoral cords in a HER2⁺ breast carcinoma. (d) Higher numbers of FoxP3⁺ cells infiltrate a HER2⁺ case. (a to d) Immunohistochemical stain; paraffin section; Hematoxylin counterstain; (a and c) 20× original magnification, (b and d) 40× magnification. HER2: human epidermal growth factor receptor-2.

Figure 3

CD8⁺ T cell responses to multiple breast cancer-associated antigenic epitopes assessed by IFN-γ-ELISPOT (interferon-γ-Enzyme Linked Immunosorbent Spot). All tests were performed using CD8⁺ purified T cells as effectors and autologous peptide-loaded monocytes as antigen presenting cells (APCs; effector:target ratio of 1:1). The number (enumerated as SFC, spot forming cells) of TAA-specific (or FluM1-specific, flu matrix protein1-derived epitope) circulating CD8⁺ T cells was investigated in HER2^- (neg = 7) and HER2⁺ (pos = 6) breast cancer patients, whereas antigen-specific responses of healthy women were used as controls (ctrl = 5). PHA-loaded and empty monocytes (EMPTY MONO) were used as positive and negative controls, respectively. For peptides amino acid sequences, refer to Table 2. HER-2, human epidermal growth factor receptor-2; TAA, tumor-associated antigens.

Figure 4

Serum cytokine profile. Interleukin (IL)-2, IL-12p70, IL-1α, IL-1β, IL-8, IL-6, IL10 (neg = 36, pos = 25, ctrl = 23), Tumor necrosis factor-α (TNF-α; neg = 33, pos = 23, ctrl = 23), granulocyte-macrophage colony-stimulating factor (GM-CSF; neg = 33, pos = 23, ctrl = 19) and transforming growth factor-β (TGF-β; neg = 27, pos = 21, ctrl = 9) levels were evaluated in serum samples from HER2^- (neg), HER2⁺ (pos) patients and age-matched healthy women (ctrl). Statistical analysis was performed with the Wilcoxon two-sample test. *, P<0.05 (refer to text for exact P value). HER2, human epidermal growth factor receptor-2.