TET, a malignant tumor in the chest, is most common, albeit rare, in adult anterior mediastinal neoplasms, with an incidence rate of only 1.5–3.2 cases per 1000,000 people per year .[37] Among TETs, thymomas are often uniquely associated with autoimmune diseases, whereas thymic carcinoma exhibits more invasive characteristics in the clinic. A study by Radovich et al. reported a strong association of broad histological subtypes (A, AB, B, and thymic carcinoma) with multiple classes of aberrations that occur at different levels.[14] The etiology of TETs is unknown, largely owing to our limited knowledge of the genomic underpinnings of thymoma and thymic carcinoma.
Sequencing whole exons of 15 tumor-normal paired samples revealed some high-frequency mutation driver genes that were previously reported in studies of TETs, including BAP1, ERBB4, HRAS and KIT, among others.[38, 39] This analysis further showed that the highest frequencies of mutations were in ZNF429 (36%) and ZNF208 (29%) genes (Fig. 2). It has been reported that the ZNF429 gene is significantly associated with acute postoperative pain in breast cancer, whereas ZNF208 polymorphisms are associated with ischemic stroke in a southern Chinese Han population.[40, 41] Studies have found that PABPC1 (poly(A) binding protein cytoplasmic 1) is directly involved in the tumorigenesis of gastric cancer through promotion of cell proliferation; consistent with this, down-regulated expression of PABPC1 is associated with tumor progression and a worse prognosis.[42, 43] Our analysis of the TCGA-THYM database further showed that the genes with the highest mutation frequency were GTF2I (50%), HRAS (8%), and MUC16 (7%) (Fig. 5). It has been shown that mutated GTF2I is a thymoma-specific oncogene, and that cells expressing mutant GTF2I highly express genes involved in Wnt and Sonic hedgehog (SHH) signaling pathways.[14] Perplexingly, this latter study did not detect mutations in the GTF2I gene in thymoma or thymic carcinoma samples, possibly owing to insufficient sample size. MUC16, a mucin expressed at high levels on the surface of epithelial ovarian tumor cells, exerts effects similar to those of MUC1, which binds or aggregates neutrophils and other cell types to provide immune protection against tumors.[44, 45]
The expression of PD-L1 has been used as a tumor cell marker in a number of clinical trials and has been approved for this purpose.[46] It is thought that positive expression of PD-L1 in tumors is an indicator of patients who are more likely to respond to treatment.[47] However, some patients who test positive for PD-L1 may not respond to treatment; more importantly, some patients who test negative may still respond, making this an imperfect biomarker.[48] A number of studies have reported conflicting data on the relationship between PD-L1 expression and survival/poor prognosis in TETs.[15, 18, 21, 22] The rates of PD-L1 positivity in thymoma and thymic carcinoma in the current study (33.3% and 58.3%, respectively) are consistent with the reported positivity-rate range.[15 – 21]
High TMB is a characteristic associated with responsiveness to immunotherapy.[49] However, high TMB values (≥20 mutations per Mb) were found to be unsuitable for predicting immunotherapy effects in all types of solid tumors.[9] For different cancer types, establishing the high TMB threshold may require additional clinical studies and information statistics on a larger number of patients. We found that the TMB value in this study was significantly higher than that in the TGA-THYM study, suggesting that patients in our study would be more likely to benefit from immunotherapy (Fig. 3).
Because of differences between our findings regarding TMB and those reported by the TCGA-THYM study, we analyzed the types of mutation, defined by the TI/TV ratio. We found significant differences between our study and TCGA data with respect to C > A, C > T, T > A, and T > G mutations, but found no significant difference for C > G and T > C mutations (Fig. 3). Mutational signatures can be understood as different mutation processes that often generate different combinations of mutation types.[50] Thousands of somatic mutations can be identified in a single cancer sample, making it possible to decipher the mutant signature, even if several mutations are operative.[51] A previous study showed that the C > A mutational signature is associated with tobacco smoking, whereas C > T mutations are induced by ultraviolet light.[52]
Signal pathways are characterized by changes in multiple genes and expression levels, and usually involve simultaneous changes in multiple pathways, such as angiogenesis and notch signaling pathways. Cancers are not driven by single gene mutations or expression changes, but by coordinated changes affecting multiple signaling pathways.[53, 54] ErbB (erythroblastic oncogene B), also called the epidermal growth factor receptor (EGFR),[55] is generally expressed on the surface of epithelial cells, but is often overexpressed in certain tumor cells. Overexpression of EGFR is associated with tumor cell metastasis, invasion, and poor prognosis.[56] In the current study, we found enrichment of ErbB and T cell signaling pathways in a high proportion of samples in the thymic carcinoma group—pathways that were also found at the intersection of gene clusters (Fig. 4B, 4C). In the thymoma group, prominent pathways included longevity-regulating pathways (hsa04213) and central carbon metabolism in cancer (hsa05230) (Fig. 4C). Cell immortalization is an important stage in the tumorigenesis process.[57] The most reliable and repeatable longevity-promoting strategy for mammals is considered to be calorie restriction (CR). Among the pathways involved in regulating the CR effect are the insulin-like growth factor (IGF-1)/insulin signaling pathway, the sirtuin pathway, the adenosine monophosphate (AMP)-activated protein kinase (AMPK) pathway and the target of rapamycin (TOR) pathway.[58] It is thought that these pathways respond to CR via a common mechanism involving activation of autophagy, stress defense mechanisms and survival pathways, while weakening pro-inflammatory mediators and cell growth, thereby promoting cell health and, ultimately, longevity.[59] In the final analysis, the thymoma canceration process likely reflects benign or malignant mutations in multiple genes or changes in gene expression levels.