Discussion
The majority of HNC was diagnosed at stage III-IV, and its high fatality rate and poor quality of life become significant concerns to social and public health (35). HPV, heavy use of tobacco and alcohol can increase the risk of HNC(36), which has been proved by various studies. Recent studies have witnessed the development of immunotherapy of HNC. PD-1 was first discovered and associated with suppressing T cells, downregulating the immune response, and apoptosis(37). PD-L1(38), one of the ligands of PD-1, is expressed in T cells and antigen-presenting cells; PD-L2(39), another ligand of PD-1, is present in antigen-presenting cells, stromal, tumor endothelial cells, can be induced by other cells as well(40).
To investigate the effect of PD-1/PD-L1 inhibitors alone or in combination with traditional therapy, we performed the systematic review and meta-analysis included five RCTs and nine single-arm clinical trials, which contained 2,911 HNC patients. The 5 RCTs revealed that immunotherapy improved survival rates by 63% compared with SOC. The reasons we inferred are as follows, HNC is characterized by inadequate immune surveillance, more immunosuppressive cytokines, and impairment of immune cells both in function and number(41); immunotherapy may enhance the capacity of the immunity system to inhibit tumor cell infiltration. However, DNA repair defects associated with migratory and invasive behavior may induce the poor prognosis of HNC patients treated with chemotherapy may induce (42). Of note, conventional therapies are lack specificity.
We observed patients with HPV+ had better overall survival (P <0.000001), and participants with PD-L1 CPS ≥1 were more suitable for PD-1/PD-L1 immunotherapy (P<0.0001). Viral antigens exist as exosomes or integrate into host cells throughout carcinogenesis (43), which may strengthen the immunity of HPV+ patients (44). HPV+HNSCC enriched more B cells (44), and PD-1 expression was lower in CD8+ T cells, while the PD-1 frequency negatively correlated with IFN-γ secretion capacity (45). The frequency of PD-1 positive tumor-infiltrating lymphocytes (TIL) was higher in HPV+patients, but their frequency of CD4+ regulatory T cells (Tregs) cells was slightly lower than HPV-TILs(44). The enrichment area of B cells in HPV+ TIL was found primarily in the germinal center, in which B cells experienced germinal center reactions and could appear at different stages of development; while the majority of B cells in HPV TILs is mainly plasma cells and switched memory cells (44). Additionally, TP53 mutations (46), amplification, and overexpression of CCND1(the gene that encodes Cyclin D1) induced drug resistance, which was more common on HPV- tumors (47).
PD-L1 overexpression was considered to influence prognosis of HNSCC(48). PD-L1 on exosomes was associated with disease progression of HNSCC patients, including patients’ disease activity and the lymph node status(49). Overexpression PD-L1 on circulating tumor cells (CTC) indicated worse OS and PFS at the end of treatment for locally advanced HNSCC(50). In contrast, it was relevant to favorable survival when overexpression on tumor cells together with tumor-associated immune cells (TAICs)(47), which seems to be paradoxical and needs more investigation. Besides, some studies revealed that PD-L2 expression was the potential prognostic biomarker of PFS and response to pembrolizumab in HNSCC patients (40), PD-L1 and IFN- γ related gene signatures impacted effects of PD1/PD-L1 immunotherapy in recurrent or metastatic HNSCC(47). Intriguingly, patients with high PD-1 expression also have a better response to PD-1 inhibitors(45).
Although PD-1/PD-L1 immunotherapy influences less on the digestive system and blood system, it might cause a higher risk of hypothyroidism. In a report about nivolumab-related thyroiditis, some of the patients endured transient thyrotoxic phases followed by thyroiditis, part of them with antithyroglobulin (anti-Tg) and antithyroid peroxidase (TPO) antibodies. However, thyrotropin binding inhibitory immunoglobulins (TBII) of all of them were absent, while the other patients developed hypothyroidism without the thyrotoxic phase, anti-Tg, and anti-TPO antibodies were positive(51). Some patients with hypothyroidism but without thyrotoxicosis might endure dormant thyrotoxic stage, and the mechanism for PD-1/PD-L1 immunotherapy related-thyroid dysfunction was unclear(52). We assumed PD-1/PD-L1 immunotherapy induced thyroiditis, thus caused hypothyroidism and the most common fatigue symptoms (52). In a nutshell, it is necessary to test thyroid function after receiving PD-1/PD-L1 immunotherapy.