Expression of the Cell Adhesion Molecule CD44 in Human Lung Tumors and Cell Lines.

Background: The purpose of this study was to examine the expression of the cell adhesion molecule CD44 in normal lung, primary and metastatic lung tumors, and cell lines derived from primary lung carcinomas. Methods and Results: A total of 68 lung specimens including normal tissue and primary and metastatic tumors, as well as 28 cell lines cultured from primary lung tumors with high recurrence, were examined for CD44 expression by semiquantitative reverse transcription polymerase chain reaction. Variant exon expression was confirmed by Southern blotting and hybridization of particular samples. In tumor tissues, loss of CD44 variant expression correlated with increasing tumor stage; a smaller percentage of more aggressive and poorly differentiated tumors expressed CD44v. Tumors metastatic to the lung were negative for CD44 variant expression. In primary lung cell lines, as in tumor tissue, tumors of higher histologic grade were characterized by loss of CD44 variant expression. Conclusion: CD44 isoform expression in normal lung and tumor tissues and cell lines revealed an overall decrease in CD44 alternative splicing in lung neoplasms of increased malignancy.

CD44 Expression in Benign and Neoplastic Human Prostates.

Background: CD44, a major cell surface receptor for hyaluronic acid, is a family of ubiquitous cell surface glycoproteins. Altered levels of CD44 expression, seen in many epithelial neoplasms, have prognostic implications. Expression of standard and variant isoforms of CD44 was assessed in normal and neoplastic human prostate tissue and culture cells to evaluate as a marker for malignant transformation. Methods and Results: Expression of CD44s, CD44R, v5, v6, v7/8 and v10 was assessed in prostate tissue (benign and malignant) and cell lines (DU-145, PC-3, LNCaP, p69) and primary cultures of normal prostates and adenocarcinoma cells obtained from prostatectomies using reverse transcriptor polymerase chain reaction, Western blotting, and immunofluorescence. No CD44 expression was seen in LNCaP cells. p69, DU-145, and PC-3 cells expressed CD44s and CD44R. p69, cells demonstrated a 1000-bp-long form of CD44 mRNA, unique to this normal cell line. Both normal and neoplastic prostatic tissue demonstrated CD44s on Western blotting. Conclusions: In agreement with previous studies, prostatic adenocarcinoma cells, except LNCaP, expressed CD44s. Different patterns of CD44 expression were seen in benign and neoplastic prostate. Benign prostate exhibited higher v5 protein levels, whereas neoplastic prostates demonstrated higher CD44s expression. CD44s expression was identified in all neoplastic prostates as compared with only 50% of the benign prostates. No significant difference in expression of the other variants assessed (v6, v7, v7/8, and v10) was observed in the benign and neoplastic prostates.

Prediction of Biologic Aggressiveness in Colorectal Cancer by p53/K-ras-2 Topographic Genotyping.

Background: Topographic genotyping is a system of solid tissue molecular analysis designed to correlate microscopic alterations with specific forms of gene damage. In this system, microscopic targets are selected on the basis of cellular and immunohistochemical features. Minute tissue samples corresponding to these targets are precisely removed and analyzed for the presence and specific type of oncogene/tumor suppressor gene damage by means of polymerase chain reaction (PCR) followed by DNA sequencing. In this study, topographic genotyping was used to investigate the prognostic value of p53 and K-ras-2 mutational damage in 204 patients with colorectal cancer from two tertiary care centers. Methods and Results: The intensity and distribution of p53 immunohistochemical staining were correlated with the presence and specific type of mutational change, which resulted in a better understanding of the highly variable nature of p53 immunostaining patterns. Molecular genotype was correlated with the depth and extent of colorectal cancer spread (Dukes B and C) and with survival for follow-up periods of up to 10 years. An algorithm for p53/K-ras-2 genotyping was formulated to include p53 immunohistochemistry and DNA sequencing both of p53 exons 5-8 and of K-ras-2 exons 1 and 2. By using this algorithm, survival was shown to be significantly better in those patients whose tumors manifested normal p53 and K-ras-2 genes (P >.01). Patients with p53-mutated tumors composed a poor prognostic group, characterized by a high rate of intra-abdominal recurrence in the form of peritoneal seeding. Patients with K-ras-2-mutated tumors also composed a poor prognostic group, marked by a tendency for distant hematogenous metastasis involving lung, bone, and brain. Conclusions: Topographic genotyping's molecular diagnostic approach to colorectal cancer combines immunohistochemistry, PCR, and DNA sequencing. It is informative, cost-effective, timely, and yet fully integrated with standard histopathology. The use of this approach by pathologists as a model system for molecular diagnosis of colorectal cancer and other forms of solid tumor malignancy is recommended. As new prognostic molecular lesions are documented for tumor progression and metastasis, topographic genotyping will be well suited to facilitate their clinical application.