The best magazine
Results
TIM-3 Expression on TILs and Tumor Cells in Primary Pulmonary Carcinoma
It was reported that TIM-3 is expressed not only on various immunocytes but also on melanoma cells. We sought to detect whether TIM-3 is also expressed on pulmonary carcinoma cells. As shown in Image 1, immunohistochemical analysis showed TIM-3 protein as a brown-stained product (Images 1C and 1D), while TIM-3 was not stained in the normal serum control (Image 1A) and isotype IgG control (Image 1B), confirming the specificity of TIM-3 staining in tissue sections. Interestingly, we detected TIM-3 expression in lung cells from patients with NSCLC (Image 1D) but not from patients with bronchodilation (regarded as normal lung tissue control) (Image 1C), strongly suggesting enhanced TIM-3 expression in malignant lung cells. As far as we know, this is the first report which demonstrated TIM-3 expression in malignant pulmonary cells. As shown in Table 2, TIM-3–stained pulmonary carcinoma cells could be detected in 86.7% (26/30) of NSCLC specimens, of which 57.7% (15/26) showed moderate/strong (++/+++) Image 2C and 42.3% (11/26) showed weak (+) TIM-3 expression Image 2D. Although no TIM-3–stained lung cells could be detected in tissue samples from 3 patients with bronchodilation, TIM-3 staining was identified in infiltrated macrophages in bronchodilation tissues (Image 1C).
(Enlarge Image)
Image 1.
TIM-3 expression in primary pulmonary carcinoma. A, Normal serum, and B, isotope IgG were used as negative controls (×40). C, Very few macrophages are noted in bronchodilation patients with weak TIM-3 expression (arrow) (×20). D, Tumor cells in primary pulmonary carcinoma were clearly TIM-3 stained (arrows) (×20) (insets, ×100).
(Enlarge Image)
Image 2.
Representative immunohistochemical staining of TIM-3 expression in both infiltrated immunocytes and tumor cells from patients with non–small cell lung cancers. A, Macrophages (arrows), and B, lymphocytes (arrows) are evident in a primary pulmonary carcinoma with strong TIM-3 staining (×20). C, Tumor cells (arrows) are evident with strong TIM-3 staining (×100). D, Tumor cells (arrows) are evident with weak TIM-3 staining (×100) (insets, ×100).
TILs were observed in 93.3% (28/30) of tested NSCLC tissue specimens without preoperative therapy. As expected, TIM-3 was positively stained on 92.9% (26/28) of TILs and most (data not shown) infiltrated tumor-associated macrophages (TAMs) in lung cancer tissues Image 2A and Image 2B, which was consistent with its expression in both normal and inflammatory immunocytes. The expression of TIM-3 on TILs and TAMs (Images 2A and 2B) seems to be stronger than that on infiltrated macrophages in tissue specimens from 3 patients with bronchodilation (Image 1C).
Correlation of TIM-3 Expression With Clinicopathologic Parameters and Host Immune Response
To explain the role of TIM-3 expression in malignant pulmonary cells from patients with NSCLC, the correlation between TIM-3 expression in lung cells and various clinicopathologic parameters, such as histologic types, pathologic T classification, nodal status, and stages of lung cancer, were analyzed. As shown in Table 1, significant positive correlations were observed between TIM-3 expression in malignant pulmonary cells and the parameters of histologic types and pathologic T classification (all values of P < .05). But we failed to find significant correlations between TIM-3 expression and the parameters of age (P = .7588), gender, nodal status (P = .1395), and disease stage (P = .1773). However, the frequency of strong TIM-3 expression was higher in patients with nodal status and in III/IV disease stages than those without nodal status and in I/II disease stages (Table 1).
Among patients with NSCLC, 93.3% (28/30) contained lymphocyte infiltrations that had almost exclusively lymphoid morphologic characteristics, indicating that they were TILs. Although the positive correlation between TIM-3 expression in lung cells and the host immune response was not significant (P > .05), a moderate/strong host immune response occurred more frequently in patients with a positive TIM-3 expression in lung cells than in those with a negative TIM-3 expression (65.3% [17/26] vs 50% [2/4]; Table 2). This is consistent with the point of view that TIM-3 expresses on activated T cells but not naïve T cells. However, in a comparison among host immune responses of patients with TIM-3 positive expression tumor, a stronger immune response was demonstrated in patients with weak TIM-3 expression than those with moderate/strong TIM-3 expression (72.7% [8/11] vs 60% [9/15]; Table 2). This indicated that strong TIM-3 expression, both in immune cells and in tumor cells, correlates with low immune response. This phenomenon is further illustrated in Image 3, which shows a weak host immune response in patients with a strong TIM-3 expression (Image 3A), whereas a strong immune response was associated with weak TIM-3 expression (Image 3B).
(Enlarge Image)
Image 3.
Representative immunohistochemical staining of TIM-3 expression with host immune response. A, Patients with strong TIM-3 expression in tumor cells and few TILs (×20). B, Patients with weak TIM-3 expression in tumor cells and more TILs (×20).
Correlation of TIM-3 Expression and Clinicopathologic Parameters With Patients' Survival Rate
In a postassay follow-up period of more than 60 months, 16 patients died of cancer. Of these deaths, 12 were patients with moderate/strong TIM-3 expression tumors, while 4 were patients with weak/negative TIM-3 expression. In the entire cohort, the 5-year survival rate of patients with moderate/strong TIM-3 expression tumor cells was significantly lower than those with weak/negative TIM-3 expression (40% vs 6.7%; P = .0014, log-rank test) Table 3. The clinicopathologic parameters of age, gender, histologic type, pathologic T classification, nodal status, and disease stage were also analyzed to determine their effects on the patients' survival. As indicated in Table 3, the parameters of nodal status and disease stage, together with TIM-3 status, had a significant effect on the overall survival rate. Univariate survival analysis was performed to investigate possible prognostic effects of TIM-3 on NSCLC. The expression of TIM-3 in pulmonary cells had a statistically significant correlation with a worsening of the survival probability (P < .001) Figure 1. This was also confirmed with a multivariate survival analysis (Table 3) using factors that were statistically significant in the univariate analysis, namely nodal status, disease stage, and TIM-3 status. All of these analyses suggest that TIM-3 expression can be used as an independent prognostic factor for patients with NSCLC (relative risk, 4.481; 95% confidence interval, 1.790 - 11.22; P = .0005).
(Enlarge Image)
Figure 1.
Negative correlation between TIM-3 expression and survival rate. Kaplan-Meier estimates of the overall survival curves for primary pulmonary carcinoma patients with TIM-3 negative/weak expression and TIM-3 moderate/strong expression for the entire cohort (P < .001).
Source: ...