Cytokeratin 7 (CK7) Scoring in Patient-Derived Lung Xenografts

Overview

• CK7 as a biomarker in cancer
• CK7 scoring in patient-derived lung xenografts
• Classification of CK7 staining score
• Product
• Literature

CK7 as a biomarker in cancer

Keratins are intermediate filaments of epithelial cells that have multiple functions in anti-apoptosis, protein synthesis, wound healing as well as in the polarization of epithelial cells and signal transduction (Pastuszak et al., 2015). Cytokeratin 7 (CK7) is a type II keratin expressed in simple and glandular epithelia and in specialized cells of mesenchymal organs. In healthy tissues, CK7 immunostaining is detected for example on collecting ducts and scattered epithelial cells in the Bowman capsule of the kidney (Figure 1A) and in epithelial cells of the placenta (Figure 1B), whereas CK7 immunostaining is absent in small intestinal mucosa (Figure 1C) and in keratinizing squamous epithelia of the skin (Figure 1D) (Dum et al., 2022).

CK7, along with cytokeratin 20 (CK20), is one of the most commonly used diagnostic antibodies for the determination of the origin of metastatic cancer tissues (Dum et al., 2022). CK7 and CK20 expression can be used as clinical parameters for the prognosis of different types of cancer. CK7 is usually negative in prostate and colorectal cancer, while CK7 positivity is indicative of urothelial origin or mucinous ovarian cancer. Aberrant expression of CK7 is associated with unfavorable tumor behavior. For example, Dum et al. (2022) correlated detectable expression of CK7 to unfavorable features in adenocarcinomas of the stomach, derived from CK7 negative precursor cells. In contrast, low expression of CK7 was linked to unfavorable features in breast cancer of no special type (NST) and urothelial carcinomas, both derived from CK7 positive precursor cells. In human lung cancer, CK7 positivity depends on the histological subtype (Johansson, 2004). Positive CK7 expression may be an independent clinical parameter for poor prognosis of lung cancer patients (Luo et al., 2017).
 

CK7 scoring in patient-derived lung xenografts

In order to analyze CK7 expression in tumor cells of xenograft models, it is important to distinguish between endogenous murine CK7 and CK7 expressed by human tumor cells. HistoSure anti-human Cytokeratin 7 antibodies discriminate between human and mouse CK7 (Figure 2).

A tissue microarray (TMA) containing 61 annotated patient-derived lung xenograft (PDX) cores was kindly provided by Charles River Laboratories (Freiburg, Germany) and stained using an anti-human Cytokeratin 7 (HS-454 017) antibody on an automated staining platform. Lung cancer xenografts were divided into two types: small cell lung cancer (SCLC; 6/61) and non-small cell lung cancer (NSCLC; 50/61). The latter were further subdivided into NSCLC adenocarcinoma subtype (25/61), NSCLC epidermoid subtype (20/61), and NSCLC large cell subtype (10/61). The CK7 expression level was scored between 0 and 3+ according to staining intensity and percentage of positive cells (Figure 3; Table 1).
 

Classification of CK7 staining score

All SCLC PDX were either negative (5/6) for CK7 or showed only very weak expression (1/6). In contrast, NSCLC PDX of type adenocarcinoma were mostly CK7 positive, with only 3/25 stained negative. 12/25 showed moderate to strong (staining score 1.5 - 2) or even very strong (staining score 2.5 - 3) CK7 expression. NSCLC PDX of type epidermoid were mostly CK7 negative (14/20), and only one specimen showed very strong expression (staining score 2.5). Large cell subtype NSCLC showed more heterogeneous CK7 expression with 4/10 negative specimens and 2/10 specimens with very strong CK7 expression. On average, NSCLC PDX of type adenocarcinoma showed the highest CK7 expression (mean score = 1.7) and SCLC PDX showed the lowest CK7 expression (mean score = 0.1) (Figure 4). 

We then compared expression levels of CK7 by RNA sequencing and by immunohistochemistry. Human CK7 gene expression was evaluated by Charles River in PDX tumor tissues. RNA seq scores ranged from 0 to 12.2. RNA seq scores < 7.2 were translated into CK7 staining scores of 0 (Figure 5). RNA seq scores > 8.4 were usually translated into CK7 staining scores of ≥ 1. PDX with RNA scores > ~10 showed highly variable CK7 staining scores ranging from 1 to 3.

In summary, RNAseq scores and CK7 staining stores yielded comparable results in the present study. High RNAseq scores correlated with medium to high CK7 expression scores in IHC-P. In addition, CK7 expression analysis using immunohistochemistry visualizes tumor heterogeneity. This spatial information is lost in RNAseq. (Figure 6).

Product

Literature

Pastuszak et al., 2015. Cytokeratins in gastroenterology. Systematic review. PMID: 26557935

Dum et al., 2022: Cytokeratin 7 and cytokeratin 20 expression in cancer: A tissue microarray study on 15,424 cancers. PMID: 35390310

Johannson, 2004. Histopathologic Classification of Lung Cancer: Relevance of Cytokeratin and TTF-1 Immunophenotyping. PMID: 15494931

Luo et al., 2017. Identification of relevant prognostic values of cytokeratin 20 and cytokeratin 7 expressions in lung cancer. PMID: 28827446