Prognostic and predictive value of PDL1 expression in breast cancer.

Sabatier R, Finetti P, Mamessier E, Adelaide J, Chaffanet M, Ali HR, Viens P, Caldas C, Birnbaum D, Bertucci F. · 2015

Read Prognostic and predictive value of PDL1 expression in breast cancer. by Sabatier R, Finetti P, Mamessier E, Adelaide J, Chaffanet M, Ali HR, Viens P, Caldas C, Birnbaum D, Bertucci F. (2015) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc4467160.txt.

Introduction

Renaud Sabatier and nine co-authors measured PDL1 in breast cancer and asked what its level predicts, in Oncotarget in 2015. PDL1 is a protein that can shield tumors from immune attack, and drugs blocking its partner have shown promise in several cancers. Its meaning in breast cancer, though, had barely been studied when this work began.

Using gene activity data from thousands of breast cancers and dozens of cell lines, the team found PDL1 raised in a fifth of tumors and more often in the aggressive basal kind. High levels came with large, fast-growing, hormone-insensitive tumors, yet also with signs of a strong local immune attack. In basal tumors, high PDL1 tracked with better survival and with fuller response to chemotherapy given before surgery. The marker, they conclude, flags immune-active basal cancers rather than simply worse disease.

Catalog record

CREATORSabatier R, Finetti P, Mamessier E, Adelaide J, Chaffanet M, Ali HR, Viens P, Caldas C, Birnbaum D, Bertucci F.
YEAR2015
COLLECTIONPAPERS
FORMATTXT
FILE SIZE55 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC4467160

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Prognostic and predictive value of PDL1 expression in breast cancer By Renaud Sabatier, Pascal Finetti, Emilie Mamessier, José Adelaide, Max Chaffanet, Hamid Raza Ali, Patrice Viens, Carlos Caldas, Daniel Birnbaum, François Bertucci Published in: Oncotarget DOI: 10.18632/oncotarget.3216 ABSTRACT Expression of programmed cell death receptor ligand 1 (PDL1) has been scarcely studied in breast cancer. Recently PD1/PDL1-inhibitors have shown promising results in different carcinomas with correlation between PDL1 tumor expression and responses. We retrospectively analyzed PDL1 mRNA expression in 45 breast cancer cell lines and 5,454 breast cancers profiled using DNA microarrays. Compared to normal breast samples, PDL1 expression was upregulated in 20% of clinical samples and 38% of basal tumors. High expression was associated with poor-prognosis features (large tumor size, high grade, ER-negative, PR-negative, ERBB2-positive status, high proliferation, basal and ERBB2-enriched subtypes). PDL1 upregulation was associated with biological signs of strong cytotoxic local immune response. PDL1 upregulation was not associated with survival in the whole population, but was associated with better metastasis-free and overall specific survivals in basal tumors, independently of clinicopathological features. Pathological complete response after neoadjuvant chemotherapy was higher in case of PDL1 upregulation (50% versus 21%). In conclusion, PDL1 upregulation, more frequent in basal breast cancers, was associated with increased T-cell cytotoxic immune response. In this aggressive subtype, upregulation was associated with better survival and response to chemotherapy. Reactivation of dormant tumor-infiltrating lymphocytes by PDL1-inhibitors could represent promising strategy in PDL1-upregulated basal breast cancer. INTRODUCTION Despite recent progresses, nearly 20% of patients with breast cancer still develop metastases and die from disease progression. During the last decades, molecular alterations involved in mammary oncogenesis and metastatic progression have been identified, leading to major therapeutic progresses such as hormone therapy targeting the estrogen receptor (ER) and targeted therapies directed against oncogenic proteins (ERBB2, EGFR, VEGF, and PI3K/AKT/mTOR pathway). Nevertheless, if not present initially, resistant clones emerge in most of cases because of the high mutagenic and adaptable capacity of cancer cells, making the tumor responses temporary. Thanks to the adaptability of the immune response, cancer immunotherapy can theoretically address this issue. Breast cancer is less immunogenic than melanoma or renal cell carcinoma and the results of adoptive immunotherapy (interleukin 2, interferons, and vaccines) have been relatively disappointing. However, the role of immunity has emerged during the last decade with the demonstration of a favorable prognostic impact of the presence of tumor-infiltrating lymphocytes (TILs) [1-3] and of gene expression signatures of immune response, notably for ER-negative, highly proliferative tumors [4-8]. Immune response represents a complex phenomenon based on a balance between activator and inhibitor pathways that regulate TILs activity. This balance may be disturbed in certain pathological conditions such as cancer where the inhibition of the immune system will favor tumor progression. One key inhibitor is the PD1-PDL1 pathway. PD1 (Programmed cell Death 1) is a cell surface membrane protein expressed by various immune cells including T-cells; it is activated by its ligands PDL1 and PDL2, which are expressed by antigen-presenting cells such as macrophages or B-cells. After engagement by its ligands, PD1 attenuates lymphocyte activation [9-13] and promotes T-regulatory cell development and function, allowing to terminate the immune response. Recent works have suggested that it could be a main actor in cancer progression through anti-cancer immune response inhibition [10, 14, 15]. Indeed, tumor cells from different locations express PDL1 and thus can inhibit the immune response. Clinical trials testing anti-PD1 or anti-PDL1 drugs to restore anti-cancer immunity have shown very promising results with durable responses, notably in melanoma and renal, lung, prostate and bladder carcinomas [16-18], and phase III studies are ongoing. Furthermore, relationship between PDL1 expression on tumor and/or immune cells and objective response has been reported [16, 17, 19-21]. PDL1 expression has been studied in different cancers such as kidney, lung, pancreas, esophagus, ovary, colorectal, head and neck and squamous cell carcinomas, melanomas and gliomas [22-32], with evidence of correlations with clinicopathological tumor features in several studies. In breast cancer, the PD1-PDL1 pathway has been very scarcely studied [33-39]; only two prognostic studies, including 650 cases analyzed at the protein level (immunohistochemistry, IHC) [39] and 398 cases at the mRNA level [36], recently addressed the prognostic issue, but provided divergent results. Here, we have analyzed PDL1 mRNA expression in 45 breast cancer cell lines and 5,454 breast cancers profiled using DNA microarrays. We searched for correlations between PDL1 expression and genomic and clinicopathological data, including survival and response to chemotherapy. RESULTS PDL1 expression and copy number alterations in breast cancer PDL1 expression was measured by using probe sets whose identity and specificity showed 100% accuracy (Supplementary Table 1). We evaluated PDL1 expression in 45 breast cancer cell lines. Luminal cell lines (N=16) showed lower PDL1 expression level than basal cell lines (N=11; p=1.46E-03, Tukey test) and mesenchymal cell lines (N=8; p=2.0E-04, Tukey test). PDL1 expression was similar between basal and mesenchymal cell lines (p=0.34, Turkey test; Figure 1 and Supplementary Figure 1). PDL1 expression was analyzed in 5,454 clinical breast cancer samples pooled from 18 data sets (Supplementary Tables 2-3): 1076 tumors (20%) showed PDL1 upregulation when compared to normal breast (ratio T/NB ≥2; “PDL1-up” group), and 4378 (80%) did not show upregulation (ratio <2; “PDL1-no up” group). [Figure: PD-L1 mRNA expression across molecular subtypes of breast cancer cell linesPDL1 expression level reported as a box plot according to the molecular subtype of cell lines. The p-values are indicated (Tukey test) are indicated as follows: **, p<0.01; ***, p<0.001; NS, p>0.05.] Array-CGH data were available for 3,140 tumors. PDL1 copy number alterations were rare: 134 tumors (4%) presented losses, including 13 with homozygous deletion (0.4%), whereas 163 (5%) showed gains, including 39 (1%) with amplification. Of note, 74% of amplified tumors were basal subtype. Basal tumors presented more gains when compared to the other molecular subtypes (17% vs from 1 to 4% for the other subtypes; p<1E-04, Fisher's exact test), but similar levels of losses (p=0.53). Correlation existed between DNA copy number and mRNA expression since tumors with PDL1 gains displayed a higher PDL1 expression level (p=2.18E-07, Student t-test; Supplementary Figure 2). PDL1 expression and clinicopathological features We searched for correlations between PDL1 mRNA expression and clinicopathological features. As shown in Table 1, PDL1 expression was generally associated with poor-prognosis features: pathological type (with more ductal and medullary carcinoma in the “PDL1-up” group), large pathological tumor size, high tumor grade, negative ER status, negative PR status, positive ERBB2 status, and positive Ki67 status. No correlation was found with patients' age and pathological axillary lymph node status. Regarding the molecular subtypes, we observed more basal and ERBB2-enriched cases and less luminal and normal-like cases in the “PDL1-up” group than in the “PDL1-no up” group (p=5.0E-04, Student t-test). [Table: PDL1 expression and clinicopathological features] PDL1 expression and immune features Next, we investigated whether PDL1 expression was associated with immunity-related parameters in clinical samples of the whole data set (Supplementary Table 4). First, we found a correlation between PDL1 expression and several immune prognostic gene expression signatures of basal breast cancer [4-7]. Breast cancer samples predicted by these classifiers as having a higher expression of immune response genes (good-prognosis) indeed overexpressed PDL1. Second, we found that the probability of activation [40] of immune-related pathways such as IFNα, IFNγ, STAT3 and

Read the complete text