Inter- and Intra-Observer Agreement of PD-L1 SP142 Scoring in Breast Carcinoma-A Large Multi-Institutional International Study

Establishment and verification of a prognostic immune cell signature-based model for breast cancer overall survival

Background

Breast cancer (BRCA) is a prevalent and aggressive disease. Despite various treatments being applied, a significant number of patients continue to experience unfavorable prognoses. Accurate prognosis prediction in BRCA is crucial for tailoring individualized treatment plans and improving patient outcomes. Recent studies have highlighted the significance of immune cell infiltration in the tumor microenvironment (TME), but predicting survival remains challenging due to the heterogeneity of BRCA. The aim of this study was thus to produce an immune cell signature-based framework capable of predicting the prognosis of patients with BRCA.

Methods

The GSE169246 dataset was from the Gene Expression Omnibus (GEO) database, comprising single-cell RNA sequencing (scRNA-seq) data from 95 individuals with BRCA. Seurat, principal component analysis (PCA), the unified matrix polynomial approach (UMAP) algorithm, and linear dimensionality reduction were used to determine the heterogeneity of T cells. Overlapping analysis of differentially expressed genes (DEGs), genes associated with prognosis, and T-cell pharmacodynamics-related genes were used to obtain the T-cell core pharmacodynamics-related genes. The dimensionality of the T-cell core pharmacodynamics-related genes was reduced employing the least absolute shrinkage and selection operator (LASSO) Cox regression model and the LASSO model. The prognostic model was built via a Cox analysis of the overall survival (OS) information. The clinical sample included 95 patients with BRCA who underwent surgical treatment from October 2018 to October 2021 at the Second Affiliated Hospital of Qiqihar Medical University. Patients were divided into a good prognosis group and a poor prognosis group based on their prognostic outcomes. The predictive value of tumor characteristics and immune responses was validated through correlation analysis, logistic regression analysis, and receiver operating characteristic (ROC) analysis.

Results

A group of 95 genes was used to establish a prognostic model. In the GEO clinical sample, with a high-risk group demonstrating shorter median survival times (2,447 vs. 6,498 days, P=4.733e-12). Area under the curve (AUC) values of 0.75, 0.75, and 0.72 were obtained for 2-, 4-, and 6-year OS predictions, respectively. Clinical validation found that the 6-year OS of the favorable prognosis group was significantly higher than that of the unfavorable prognosis group (92.06% vs. 65.62%; P=0.005). Poor prognosis was positively correlated with age, tumor size, B-cell level, and CTLA4 level and negatively correlated with tumor stage (T1/T2), lymph node metastasis stage (N0), clinical stage I-II, CD3+T-cell, CD4+T-cell, CD8+T-cell, neutrophil, lymphocyte, natural kill cell, TIGIT expression and OS. The combined model of clinical parameters had an AUC value of 0.898.

Conclusions

This study established a prognostic model that demonstrated excellent predictive value for OS of BRCA. The predictive model developed offers valuable insights into prognosis and treatment planning, emphasizing the importance of tumor characteristics and immune cell infiltration.

Reproducibility of c-Met Immunohistochemical Scoring (Clone SP44) for Non-Small Cell Lung Cancer Using Conventional Light Microscopy and Whole Slide Imaging

Emerging therapies for non-small cell lung cancer targeting c-Met overexpression have recently demonstrated promising results. However, the evaluation of c-Met expression can be challenging. We aimed to study the inter and intraobserver reproducibility of c-Met expression evaluation. One hundred ten cases with non-small cell lung cancer (40 biopsies and 70 surgical specimens) were retrospectively selected in a single laboratory (LPCE) and evaluated for c-Met expression. Six pathologists (4 seniors and 2 juniors) evaluated the H-score and made a 3-tier classification of c-Met expression for all cases, using conventional light microscopy (CLM) and whole slide imaging (WSI). The interobserver reproducibility with CLM gave global Cohen Kappa coefficients (ƙ) ranging from 0.581 (95% CI: 0.364-0.771) to 0.763 (95% CI: 0.58-0.92) using the c-Met 3-tier classification and H-score, respectively. ƙ was higher for senior pathologists and biopsy samples. The interobserver reproducibility with WSI gave a global ƙ ranging from 0.543 (95% CI: 0.33-0.724) to 0.905 (95% CI: 0.618-1) using the c-Met H-score and 2-tier classification (≥25% 3+), respectively. ƙ for intraobserver reproducibility between CLM and WSI ranged from 0.713 to 0.898 for the c-Met H-score and from 0.600 to 0.779 for the c-Met 3-tier classification. We demonstrated a moderate to excellent interobserver agreement for c-Met expression with a substantial to excellent intraobserver agreement between CLM and WSI, thereby supporting the development of digital pathology. However, some factors (scoring method, type of tissue samples, and expertise level) affect reproducibility. Our findings highlight the importance of establishing a consensus definition and providing further training, particularly for inexperienced pathologists, for c-Met immunohistochemistry assessment in clinical practice.

Divergent Cellular Expression Patterns of PD-L1 and PD-L2 Proteins in Breast Cancer

PD-L1 immunohistochemistry (IHC) has become an established method for predicting cancer response to targeted anti-PD1 immunotherapies, including breast cancer (BC). The alternative PD-1 ligand, PD-L2, remains understudied but may be a complementary predictive marker. Prospective analysis of 32 breast cancers revealed divergent expression patterns of PD-L1 and PD-L2. PD-L1-positivity was higher in immune cells than in cancer cells (median = 5.0% vs. 0.0%; p = 0.001), whereas PD-L2-positivity was higher in cancer cells than immune cells (median = 30% vs. 5.0%; p = 0.001). Percent positivity of PD-L1 and PD-L2 were not correlated, neither in cancer cells nor immune cells. Based on a cut-point of ≥1% positivity, ER+ tumors (n = 23) were frequently PD-L2-positive (73.9%), whereas only 40.9% were PD-L1-positive. These data suggest differential control of cellular PD-L1 and PD-L2 expression in BC and a potential role for PD-L2 IHC as a complementary marker to PD-L1 to improve selection of aggressive ER+ BC that may benefit from anti-PD-1 therapy.

PD-L1 testing in metastatic triple-negative breast cancer: Interobserver and interplatform reproducibility of CE-IVD assays for CPS and IC scores

PD-L1 test is recommended in different types of tumors to select patients eligible for immune checkpoint inhibitors (ICI) therapy. Several factors make this test challenging in metastatic triple-negative breast cancer (mTNBC). Different assays and platforms are available, each associated with distinct scoring systems and threshold values specific to the ICI compound used, i.e. CPS≥10 for pembrolizumab and IC ≥ 1 % for atezolizumab. Our objective was to assess the consistency of PD-L1 testing in mTNBC by examining interobserver and interassay reproducibility. We assessed n = 60 mTNBC samples for PD-L1 testing using 22C3 pharmDx assay on a Dako Autostainer Link 48 and VENTANA PD-L1 (SP263) on a Ventana BenchMark Ultra. Additionally, a subset of n = 19 samples was tested using the SP142 assay, also on the Ventana BenchMark Ultra. CPS with both 22C3 and SP263 was independently evaluated by five pathologists, all certified PD-L1 trainers. The IC with SP142 was assessed by three of these pathologists, who have particular expertise in breast pathology. Following the computation of the intraclass correlation coefficient (ICC) for each assay and their respective thresholds, we assessed the agreement between different raters and assays using Fleiss's κ, with a 95 % confidence interval (CI). Overall, we observed a significant (p < 0.001) ICC with both CPS assays [22C3 = 0.939 (CI:0.913-0.96); SP263 = 0.972 (CI:0.96-0.982); combined 22C3-SP263 = 0.909 (CI:0.874-0.938)]. Fleiss's κ confirmed an almost perfect agreement among pathologists and assays: 22C3 = 0.938 (CI:0.857-1.018); SP263 = 0.972 (CI:0.890-1.052); combined 22C3-SP263 = 0.907 (CI:0.869-0.945). Perfect inter-rater agreement was reached considering IC. This study establishes the reliability of assessing CPS in mTNBC using either the 22C3 pharmDx, as employed in the KEYNOTE studies, or the VENTANA SP263 assay. Each assay must be used on its designated platform, namely the Dako for 22C3 pharmDx and the Ventana for VENTANA SP263. It is important to remark that CPS and IC identify different patient cohorts and, therefore, are not interchangeable.

Triple-negative breast cancer: from none to multiple therapeutic targets in two decades

Triple-negative breast cancers (TNBCs) are more likely to occur in younger patients and have a poor prognosis. They are highly heterogeneous tumors consisting of different molecular subtypes. The only common characteristic among them is the absence of targets for endocrine therapy and human epidermal growth factor receptor 2 (HER2) blockade. In the past two decades, there has been an increased understanding of these tumors from a molecular perspective, leading to their stratification according to new therapeutic strategies. TNBC has ushered breast carcinomas into the era of immunotherapy. The higher frequency of germline BRCA mutations in these tumors enables targeting this repair defect by drugs like PARP inhibitors, resulting in synthetic lethality in neoplastic cells. Additionally, we have the identification of new molecules to which this generation of smart drugs, such as antibody-drug conjugates (ADCs), are directed. In this review, we will discuss the trajectory of this knowledge in a systematic manner, presenting the molecular bases, therapeutic possibilities, and biomarkers.

Reproducibility in PD-L1 Immunohistochemistry Quantification through the Tumor Proportion Score and the Combined Positive Score: Could Dual Immunostaining Help Pathologists?

We studied the pathologists' agreements in quantifying PD-L1 expression through the tumor proportion score (TPS) and the combined positive score (CPS) using single PD-L1 immunohistochemistry (S-IHC) and double immunohistochemistry (D-IHC) combining PD-L1 staining and tumor cell markers. S-IHC and D-IHC were applied to 15 cancer samples to generate 60 digital IHC slides (30 whole slides images and 30 regions of interest of 1 mm²) for PD-L1 expression quantification using both TPS and CPS, twice by four pathologists. Agreements were estimated calculating intraclass correlation coefficients (ICC). Both S-IHC and D-IHC slides analyses resulted in excellent (for TPS, ICC > 0.9) to good (for CPS, ICC > 0.75) inter- and intra-pathologist agreements with slightly higher ICC with D-IHC than with S-IHC. S-IHC resulted in higher TPS and CPS than D-IHC (+5.6 and +6.1 mean differences, respectively). High reproducibility in the quantification of PD-L1 expression is attainable using S-IHC and D-IHC.