Premium
FoxP3+ T‐cells and PD‐L1 Highlight Immune‐Suppressive Profiles in Vulvar Squamous Cell Carcinoma
Author(s) -
Holthoff Emily R,
Kelly Thomas,
Quick Charles M,
Post Steven R
Publication year - 2016
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.30.1_supplement.439.10
Subject(s) - foxp3 , immune system , stromal cell , stroma , cd8 , t cell , cancer research , population , lymphocyte , metastasis , medicine , biology , pathology , immunology , cancer , immunohistochemistry , environmental health
Suppression of anti‐tumor immune responses is permissive for tumor proliferation and invasion. Increased populations of FoxP3+ T‐regulatory cells and the production of programmed death‐1 ligand (PD‐L1) decrease CD8+ T‐cell mediated immune response and are associated with worse outcomes in several cancers. Recent studies have shown that vulvar squamous cell carcinoma (vSCC) can exhibit two different stromal response patterns. One is a fibromyxoid (FMX) response displaying fibroblasts and immature collagen, and the other is a lymphoplasmacytic (LPC) response rich in T‐lymphocytes and plasma cells. Tumors containing a FMX stroma have worse clinical outcomes, including increased risk for nodal metastasis, when compared to tumors with a LPC stroma. One possible explanation for these poor outcomes is the suppression of an immune response in FMX tumors. The goal of this study was to determine whether FoxP3+ T‐cells and PD‐L1 play a role in immune modulation in vSCC with higher risk for nodal metastasis. T‐cell populations and PD‐L1 expression were analyzed in 37 cases of vSCC. Of the 37 cases, 14 contained a FMX stroma and 7 of these metastasized to the nodes. In contrast, of the 23 cases containing a LPC stroma, only 3 tumors displayed nodal metastasis. Consecutive FFPE sections of each tumor were stained with antibodies to CD3, CD8, FoxP3, and PD‐L1. Aperio ImageScope software was used to quantify T‐lymphocyte populations per 10× field of view at the tumor‐stroma interface. Ratios of CD8+ and FoxP3+ cells to the total CD3+ T‐lymphocyte population were calculated per square micron. For PD‐L1, both tumor and stromal regions in each tumor section were ranked on a 0–3 grading scale based on the number of cells staining positively (0: no cells stained; 1: <10% cells stained, 2: 10–50% cells stained, 3: >50% cells stained positive for PD‐L1). As expected, the total number of CD3+ T‐lymphocytes was significantly higher in tumors with a LPC stroma ( p =0.0218). The proportion of CD8+ cells to the total CD3+ cell population was similar in both tumor groups. Surprisingly, the proportion of FoxP3+ cells to total CD3+ cell population was higher in the tumors with LPC stroma than in tumors with FMX stroma ( p =0.0412); however, there was significant staining variability within the LPC tumor group. PD‐L1 expression was also significantly higher in the LPC tumor group in both the tumor regions ( p =0.0213) and the stromal regions ( p =0.0006) when compared with the FMX tumor group. Within the LPC tumor group, PD‐L1 staining was reflective of the FoxP3+ cell population so that tumors with high PD‐L1 contained higher FoxP3+ populations, and vice versa. Importantly, all 3 of the LPC tumors that displayed nodal metastasis displayed a high FoxP3+ cell population and high PD‐L1 expression, while none of the tumors with a low FoxP3+ population and low PD‐L1 displayed nodal metastasis. These findings suggest that the high risk vSCC with FMX stroma utilize a method of immune suppression that is independent of FoxP3+ T‐regulatory cells or PD‐L1 expression. However, within the lower risk vSCC with LPC stroma, a subset of tumors have an immune‐suppressive profile with increased PD‐L1 expression and an elevated proportion of FoxP3+ cells. These tumors are more likely than other tumors with LPC stroma to exhibit nodal metastases and may be more responsive to targeted immunotherapy. Support or Funding Information UAMS TRI grants UL1TR000039 and KL2TR000063; SPaT institutional training grant T32GM106999