Tumor microenvironment heterogeneity in bladder cancer identifies biologically distinct subtypes that predict prognosis and anti-PD-L1 responses.

Tumor microenvironment heterogeneity in bladder cancer identifies biologically distinct subtypes that predict prognosis and anti-PD-L1 responses.

Bladder cancer is the most common cancer of the urinary tract and is histologically and genomically heterogeneous, leading to highly variable outcomes. Although improved strategies, such as the introduction of Bacillus Calmette-Guerin (BCG) and cisplatin-based chemotherapy, have greatly improved the prognosis, high-risk, advanced, and metastatic BCa still remains inadequate. Immune checkpoints (ICs) are central mediators of the TME, and IC inhibitors (ICIs) have emerged as new treatment options for patients with BCa. However, a significant proportion of patients do not respond to ICIs17. Selecting an appropriate population that responds to ICIs and identifying specific biomarkers that can predict the efficacy of ICIs are key to their further application. The TME for BCa varies widely, and it is strongly suggested to explore the mechanisms that lead to distinct responses to ICIs in the context of the complex TME to improve current efficacy and design more efficient therapies.

Immune infiltrating cells are essential among multiple components of the TME and play dual roles in tumor promoting and tumor suppression. Here, we calculated fractions of 22 immune cell types and found that patients with BCa can be classified into C1, C2, and C3 subtypes. The C1 subtype was enriched in CD8+ And CD4 activation+ Memory T cells, activated NK cells, and M1 macrophages. The C2 subtype was enriched in activated mast cells and M0 macrophages, whereas the C3 subtype was enriched in monocytes, B cells, and resting CD4 cells.+ Memory T cells and resting stem cells. Survival was better for C1 than for the other two subtypes. In TME CD8+ T cells play an anti-tumor role by recognizing MHC I molecules and patients with BCa and CD8 are more abundant.+ T cells have better survival18. Th1, Th2, Th9, Th17, and Treg subsets of CD4+ T cells are associated with a better prognosis for BCa19,20. The mechanisms by which T cells and their subtypes coordinate cancer immunity are not clear, but T cell antigen receptor (TCR) signaling plays an important role in T cell activation and transport to the TME. For example, CSK negatively regulates T cell activation by suppressing the initiation of TCR signaling, which is associated with ICI resistance.21. Therefore, ICI immunotherapy can be improved by targeting TCR signaling. M1 and M2 macrophages with extreme phenotypes are essential components of the TME. M1 macrophages inhibit tumor development, progression, and angiogenesis, while M2 macrophages promote tumor initiation, growth, and progression.22. The prognosis of patients with increased numbers of anti-tumor immune cells was improved in this study. This further underscores the importance of infiltrating immune cell profiles in the development and survival of patients with BCa.

Immune checkpoint inhibitors include new treatment approaches for patients with BCa, but only a few respond to them23. The TME is closely associated with response rates to ICIs. For example, existing T-cell immunity correlates with responses to anti-PD-L1 therapy24. Higher antitumor T cell infiltration and macrophage polarization are closely associated with higher response rates24,25. M1 infiltration can predict immunotherapeutic responses among patients with BCa 26. Our results are consistent with these. Furthermore, the C1 subtype with greater TMB had the highest response rate, which confirmed that responders had higher TMB27. Notably, PD-L1 expression on ICs but not on TCs correlates with response11. Here, we found that the response rate was lower in the C3 subtype, which expressed the least amount of PD-L1 in cancer cells. This suggests that the role of PD-L1 expressed in cancer cells requires further exploration. Our findings showed that abnormal metabolism such as fatty acid glycolysis, glycolysis, and PI3K-Akt pathways may reduce response rates. Targeting glycolysis is the most explored strategy to improve immunotherapy. Lactate dehydrogenase (LDH) abundance is associated with poor melanoma response to anti-PD-1 therapy.28. Increased glycolytic activity can impair the response to anti-PD-1 immunotherapy by inhibiting T-cell killing29. Diclofenac, which inhibits glycolysis, also enhances anti-PD-1 immunotherapy30. Based on the current results, targeting glycolysis is a promising strategy to increase the efficacy of ICI. In addition, ECM-related pathways may also hijack antitumor immunity, and ECM can modulate the activity of T cells and macrophages, thus influencing anti-tumor immune responses.31. High collagen level can boost CD8+ T cell exhaustion through leukocyte-associated immunoglobulin receptor 1 (LAIR1), anti-PD-1 therapy, and LAIR1 blockade significantly reduce tumor growth and metastasis32. Therefore, the combination of ICIs and blockade targeting these dysregulated pathways may change the TME landscape. This could enhance anti-tumor potential, increase sensitivity to immunotherapy, and thus may be a promising treatment for non-responders.

Immune cell differentiation and function in the TME are intrinsically linked to metabolic changes, and metabolic reprogramming has profound effects on angiogenesis, development, and progression.33. Competition for substrates between cancer cells and immune cells leads to remodeling of the TME to become immunosuppressive or immunostimulatory.34. Cells need to produce energy to maintain homeostatic processes, while meeting the requirements for manufacturing necessary macromolecules. We examined the metabolic profiles of C1, C2, and C3 subtypes. We found that the metabolic profiles of C1 and C2 were similar and clearly different from those of the C3 subtype. Most of the pathways related to lipid metabolism and some important pathways related to carbohydrate metabolism, such as glycolysis/gluconeogenesis, were enriched in the C3 subtype. Nucleotide metabolism was enriched in C1 and C2 subtypes. Reconfiguration of lipid metabolism, characterized by increased lipid uptake, storage, and synthesis, is a hallmark of cancer35. Tumors with lymph node metastasis accumulate more FAs as fuel and undergo a metabolic shift toward fatty acid oxidation.36. Furthermore, cancer cells can reprogram the physiology of adipocytes, stimulating them to reach distant organs37. These altered lipid metabolic pathways promote cancer cell migration and progression. Among the altered metabolic programs, aerobic glycolysis (Warburg effect) has received the most attention38. The regulation of glycolysis genes is directly related to BCa initiation and progression39. For example, essential glucose uptake transporter-1 (GLUT1) is upregulated in tumors compared to normal tissues, and this is associated with poorer overall and cause-specific survival.40. The present results showed that active cells were enriched in the C1 subtype, while resting cells were mainly enriched in the C3 subtype. The Warburg effect in cancer cells reduces lactate accumulation, inhibiting the activity of anti-tumor cells, including CD8.+ T and NK cells41,42. Competitive TME impairs immune cell activation by reducing their glycolytic capacity. Taken together, these altered metabolic pathways in the TME of BCa may lead to distinct levels and types of immune cells in different subtypes, and directly or indirectly lead to the variable outcomes associated with BCa.

We developed a TME epigenetic signature to predict survival of patients with BCa and anti-PD-L1 response rates. Although PD-L1 expression correlates with response rates, it has not been demonstrated that it can be used as a predictor of therapeutic response43. The predictive ability of our signature exceeded that of PD-L1 mRNA expression and immunohistochemistry results at staging. We found higher response rates among high-risk patients with BCa. Therefore, anti-PD-L1 therapy should be recommended for this type of patients. This would help in developing personalized approaches for patients with BCa. However, this study has several limitations. For example, we applied only bioinformatics methods to detect BCa heterogeneity. Therefore, further large-scale investigations in vivo and in vitro are needed to confirm our findings.

You may also like...

Leave a Reply

Your email address will not be published. Required fields are marked *