Intratumoral Th2 predisposition combines with an increased Th1 functional phenotype in clinical response to intravesical BCG in bladder cancer

Abstract

Th1-type immunity is considered to be required for efficient response to BCG in bladder cancer, although Th2 predisposition of BCG responders has recently been reported. The aim was to evaluate the relationship of Th1 and Th2 components in 23 patients undergoing BCG treatment. Peripheral blood, serum and urine samples were prospectively collected at baseline, during and after BCG. Th1 (neopterin, tryptophan, kynurenine, kynurenine-to-tryptophan ratio (KTR), IL-12, IFN-γ, soluble TNF-R75 and IL-2Rα) and Th2 (IL-4, IL-10) biomarkers as well as CD4 expression in T helper (Th), effector and regulatory T cells were determined. Local immune cell subsets were measured on formalin-fixed, paraffin-embedded cancer tissue by immunohistochemistry to examine expression of transcription factors that control Th1 (T-bet) and Th2-type (GATA3) immunity. We confirmed a Th2 predisposition with a mean GATA3/T-bet ratio of 5.51. BCG responders showed significantly higher levels of urinary (p = 0.003) and serum neopterin (p = 0.012), kynurenine (p = 0.015), KTR (p = 0.005), IFN-γ (p = 0.005) and IL-12 (p = 0.003) during therapy, whereas levels of IL-10 decreased significantly (p < 0.001) compared to non-responders. GATA3/T-bet ratio correlated positively with serum neopterin (p = 0.008), IFN-γ (p = 0.013) and KTR (p = 0.018) after the first BCG instillation. We observed a significant increase in CD4 expression in the Th cell population (p < 0.05), with only a modest tendency toward higher frequency in responders compared to non-responders (p = 0.303). The combined assessment of GATA3/T-bet ratio, neopterin and KTR may be a useful biomarker in predicting BCG response. Th2-promoting factors such as GATA3 may trigger Th1-type immune responses and thus contribute to the BCG success.

Keywords: BCG; Bladder cancer; GATA3; Neopterin; T cells; T-bet.

Fig. 1

Prospective study design showing the planned investigations and blood analyses at each visit (baseline, during and after BCG therapy). PBMC peripheral blood mononuclear cells, after (post), BCG Bacillus Calmette–Guérin

Fig. 2

GATA3 and T-bet expression of tumor-infiltrating immune cells prior to BCG therapy. a Distribution of GATA3⁺, T-bet⁺ T cells and GATA3/T-bet ratio in all patients; data represent mean ± SEM. b, c Abundance of GATA3⁺ and T-bet⁺ T cells based on analyzed tumor areas. Data represent mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; Friedman’s tests and Wilcoxon’s signed-rank tests. d Frequency of GATA3⁺, T-bet⁺ tumor-infiltrating T cells and GATA3/T-bet ratio with respect to treatment outcome (response vs. failure). Data represent mean ± SEM; Mann–Whitney U test. e, f Correlation analysis (and confidence bands) of GATA3, T-bet and GATA3/T-bet ratio; *p < 0.05; **p < 0.01; ***p < 0.001. g Superficially invasive bladder cancer with a high count of GATA3⁺ tumor-infiltrating lymphocytes and h low expression of T-bet⁺ T cells; Scale bar ~40 µm; i carcinoma in situ with few GATA3⁺ T cells and j only one intravascular T-bet⁺ T cell (marked with *); Scale bar ~40 µm. LP lamina propria without invasion, IF invasive front, NU neoplastic urothelium, PS papillary stroma, r _s Spearman’s rank correlation coefficient

Fig. 3

Levels of serum neopterin and tryptophan degradation at baseline, during and after BCG therapy. Patients were stratified by treatment outcome (BCG response vs. failure). Data represent mean ± 95% confidence interval (CI); *p < 0.05; **p < 0.01; ***p < 0.001; overall p value from a mixed model analysis for repeated measures between responders and non-responders; Mann–Whitney U tests for single time point comparisons. KTR kynurenine-to-tryptophan ratio

Fig. 4

Levels of urinary neopterin, sTNF-R75 and sIL-2Rα (pro creatinine) at baseline, during and after BCG therapy. Patients were stratified by treatment outcome (BCG response vs. failure). Data represent mean ± 95% confidence interval (CI); *p < 0.05; **p < 0.01; ***p < 0.001; overall p value from a mixed model analysis for repeated measures between responders and non-responders; Mann–Whitney U tests for single time point comparisons. s soluble, TNF tumor necrosis factor, IL interleukin

Fig. 5

Levels of serum IFN-γ, IL-12, IL-10 and IL-4 (pg/ml) at baseline, during and after BCG therapy. Patients were stratified by treatment outcome (BCG response vs. BCG failure). Data represent mean ± 95% confidence interval (CI); *p < 0.05; **p < 0.01; ***p < 0.001; overall p value for changes over time points by Skillings–Mack test; p value from a mixed model analysis for repeated measures between responders and non-responders; Mann–Whitney U tests for single time point comparisons between responders and non-responders. IL interleukin

Fig. 6

CD4 expression in the Th cell population, CD4⁺CD25⁺CD39⁺ T_reg cells and CD4⁺CD25^highCD161⁺ T_eff cells, and CD4⁺/T_reg ratio. a Significant changes in the CD4 expression in Th cells at baseline and during treatment in all patients; Skillings–Mack test; scatter dot plot. b CD4 expression in Th cells at baseline and throughout treatment, depending on therapeutic outcome (response vs. failure); mixed model analysis for repeated measures and Mann–Whitney U tests for single time point comparisons. Data represent mean ± SD; c no significant dynamic changes concerning the CD4 expression in T_reg at baseline and during treatment in all patients; Skillings–Mack test; scatter dot plot. d CD4 expression in T_reg cells at baseline and throughout treatment, depending on therapeutic outcome (response vs. failure); mixed model analysis for repeated measures and Mann–Whitney U tests for single time point comparisons. Data represent mean ± SD; e no significant dynamic changes concerning the CD4 expression in T_eff cells at baseline and during treatment in all patients; Skillings–Mack test; scatter dot plot. f CD4 expression in T_eff cells at baseline and throughout treatment, depending on therapeutic outcome (response vs. failure); mixed model analysis for repeated measures and Mann–Whitney U tests for single time point comparisons. Data represent mean ± SD; g no significant dynamic changes concerning the CD4 expression in total CD4⁺/T_reg ratio at baseline and during treatment in all patients; Skillings-Mack test; Scatter dot plot. h CD4 expression in CD4⁺/T_reg ratio at baseline and throughout treatment, depending on therapeutic outcome (response vs. failure); mixed model analysis for repeated measures and Mann–Whitney U tests for single time point comparisons. Data represent mean ± SD; *p < 0.05; **p < 0.01; ***p < 0.001; MFI mean fluorescence intensity