Fig 1.
Elevated expression of 5T4 in NSCLC tumor tissue as compared to normal.
5T4 expression in NSCLC, measured in transcripts per million by RNASeq was obtained from the cancer genome atlas (TCGA) Portal. Differential expression of 5T4 expression level in Lung adenocarcinoma (n = 526) and Lung squamous cell carcinoma tumor types (n = 501) were compared to their matched normal samples (n = 58 and 51 respectively). Significance was determined using ANOVA.
Fig 2.
Expression of 5T4 in characterized cell lines.
A- Western blot of the four control cell lines representing high (H226), moderate (H1975), low (H2122), and no (H460) 5T4 protein levels. The 5T4 protein is associated with a molecular weight of 72 kDa. The 5T4 band densitometry was normalized against GAPDH. Note that H226 cells have the most intensely staining 5T4 band, whereas lower levels can be seen in H1975 and H2122 cells. H460 cells appear to contain no 5T4 protein as assessed using this technique. Data was generated as described in Materials and Methods. B- Immunocytochemistry of the four control cell lines representing high (H226), moderate (H1975), low (H2122), and no (H460) 5T4 protein levels. Note that H226 cells have clear circumferential staining of the membrane, which is diminished in the H1975 and H2122 cells, disappearing completely in the H460 cells. Cells were prepared and stained as described in Materials and Methods.
Table 1.
Expression of 5T4 in control cell lines as measured by qRT-PCR, Western Blot and Immunocytochemistry.
Table 2.
Spearman Correlations of 5T4 expression as measured by Immunocytochemistry and compared to qRT-PCR and Western Blot Analysis.
Fig 3.
Effect of Ischemia and fixation on 5T4 expression in characterized xenografts models.
A- The four control xenografts models representing high (H226), moderate (H1975), low (H2122), and no (H460) 5T4 protein levels, were grown in mice as described in Materials and Methods, resected and subjected to either no ischemia or 6 hours ischemia. Expression levels are indexed to GAPDH as a housekeeping gene. The expression of 5T4 is highest in the H226 cells, and is unaffected by ischemia. H1975 and H2122 cells have reduced expressions of 5T4, which is again unaffected by ischemia. H460 cells appear to have no measurable 5T4 expression as determined by this technique. The expression levels of 5T4 were assessed using qRT-PCR as described in Materials and Methods. B- The four control xenografts models representing high (H226), moderate (H1975), low (H2122), and no (H460) 5T4 protein levels, were grown in mice as described in Materials and Methods, resected and subjected to either no ischemia or 6 hours ischemia. The expression of 5T4 is highest in the H226 cells, and is unaffected by ischemia. H1975 cells have reduced expression of 5T4 as compared to H226 cells, and there does appear to be a slightly significant increase in protein expression under the influence of ischemia. H2122 cells have reduced expressions of 5T4, which is again unaffected by ischemia. H460 cells appear to have no measurable 5T4 expression as determined by this technique. Overall, there was no evidence of a systemic change in 5T4 expression induced by ischemia across the 4 models analyzed. The expression levels of 5T4 were assessed using Western Blot analysis as described in Materials and Methods. Protein levels are indexed to GAPDH as a housekeeping gene. C- The four control xenografts models representing high (H226), moderate (H1975), low (H2122), and no (H460) 5T4 protein levels, were grown in mice as described in Materials and Methods, resected and fixed. The expression levels of 5T4 were then assessed using IHC analysis as described in Materials and Methods. Staining of 5T4 in H226 xenografts is robust, showing strong circumferential staining of the cells in the model. Strong staining is also seen in the H1975 model. Reduced staining is seen in the H2122 xenograft, with fainter staining of the membranes. No overt membrane staining is seen in the H460 xenograft model.
Table 3.
Effect of ischemia and fixation on 5T4 membrane staining as measured by Immunohistochemistry in xenograft models.
Table 4.
Profiling of 5T4 Expression in human NSCLC tumors by western blot, TLDA and IHC.
Table 5.
Spearman Correlations of H-score to TLDA and western blot data for human NSCLC tumors.
Fig 4.
Negative/Low, Moderate and High staining of NSCLC tumor samples for 5T4.
Representative images from the analytical validation sample set of human NSCLC (squamous and adenocarcinoma) tumor samples designated as Negative/Low, Moderate or High for 5T4 expression, processed and stained using the IHC assay as described in Materials and Methods. Samples were selected from a prospectively collected cohort of 24 samples that had been analyzed for 5T4 expression using qRT-PCR and Western Blot analysis, and ranked according to 5T4 expression. The samples identified as Negative/Low using orthogonal techniques have small amount of staining, whereas the samples identified as Moderate and High demonstrate increasing intensity of staining with circumferential staining of cell membranes.
Table 6.
Expression of 5T4 in the tumor and the CTC compartment of samples obtained from patients with NSCLC.
Fig 5.
H-scores of 5T4 in adenocarcinoma and squamous cell carcinoma NSCLC samples.
NSCLC tumor samples from a second cohort were prospectively collected, processed, fixed and stained as described in Materials and Methods. The samples were examined by an anatomic pathologist and each sample was given an H-score which was used to generate a rank ordered distribution. These were then plotted as a continuous distribution.
Table 7.
IHC H-score in adenocarcinoma and squamous cell carcinoma by clinical stage.
Table 8.
IHC score in adenocarcinoma and squamous cell carcinoma by grade.
Fig 6.
Enumeration and characterization of 5$t-expressing CTCs.
A- Characterized control cell lines representing high (H226), moderate (H1975), low (H2122) and negative (H460) 5T4 staining were prepared and spiked into healthy human blood samples as described in materials and methods. The blood was then used to generate slides and stained with 4',6-diamidino-2-phenylindole (DAPI) to identify cell nuclei, and antibodies against CD45, cytokeratin (CK), and 5T4. The individual images were merged to form a composite image. As can be seen, the H226 cells demonstrate the most intense 5T4 staining (magenta), and this staining decreases in the H1975, and the H2122 cells. H460 cells do not appear to have 5T4 staining using this technique. All the control cells have equivalent staining using the cytokeratin (CK) specific antibodies. None of the control cells exhibit CD45 expression. B- Characterized control cell lines representing high (H226), moderate (H1975), low (H2122) and negative (H460) 5T4 staining were prepared and spiked into healthy human blood samples as described in materials and methods. The blood was then used to generate slides and stained with 4',6-diamidino-2-phenylindole (DAPI) to identify cell nuclei, and antibodies against CD45, cytokeratin (CK), and 5T4. The level of 5T4 in individual cells on each slide was quantified and is plotted, which each dot representing an individual cell. 5T4 is measured using fluorescence units. The red line represents the mean 5T4 expression for each cell type. The mean value obtained for H460 cells was used as a cutoff threshold to define negativity for 5T4 expression. Note that even though these control cells are clonal, there is still a range of 5T4 expression seen within each control cell line. C- Cells were processed and stained using the Epic Sciences platform as described in Materials and Methods. Nuclei are visualized in blue using 4',6-diamidino-2-phenylindole (DAPI), CD45 staining is represented in green, cytokeratin staining is represented in red, and 5T4 staining is represented in magenta. The 5T4 channel and the overlay (composite) indicate that 5T4 staining is seen in the membrane and cytoplasm, as is expected. Images are three separate fields from one patient. Note that this patient had CTCs that were either single, or in clusters of 2 or more cells. The presence of clusters of CTCs is seen in many patients.
Fig 7.
Enumeration of 5T4 expressing CTC/mL in adenocarcinoma and squamous cell carcinoma NSCLC samples.
Samples were processed, stained and enumerated as described in Materials and Methods. The number of CTCs per mL ranged from 0 to over 200 per patients in this small study, demonstrating the heterogeneity in CTC numbers between patients with NSCLC. The distribution within adenocarcinoma (●) and squamous cell carcinoma (○) is noted.
Table 9.
CTC metrics in adenocarcinoma and squamous cell carcinoma blood samples derived from CD45-/CK+/DAPI+/5T4+ cells.
Table 10.
CTC metrics by tumor grade.
Table 11.
CTC metrics by tumor clinical stage.
Table 12.
Spearman correlation between IHC H-score and various CTC parameters.
Fig 8.
Concordance between IHC H-score and various CTC parameters.
A- The H score (X-axis) derived for each NSCLC sample, as determined using the 5T4 IHC assay described herein and assessed by a pathologist, was compared to the CTC-H score (Y-axis) as determined by the assay developed to measure 5T4 expression in the CTC compartment. Note the lack of correlation between the degree of expression of 5T4 between the solid tumor and the CTC compartment. B- The H score (X-axis) derived for each NSCLC sample, as determined using the 5T4 IHC assay described herein and assessed by a pathologist, was compared to the 5T4 burden/ml in (Y-axis) as determined by the assay developed to measure 5T4 expression in the CTC compartment. Note the lack of correlation between the degree of expression of 5T4 between the solid tumor and the CTC compartment.