2022年 · 第102卷第42期

Lung cancer originates from the bronchial mucosa or glands of the lungs, and is a malignant tumor with the highest morbidity and mortality in China and even in the world. Non-small cell lung cancer (NSCLC) accounts for 85% of the total number of lung cancer cases and is a serious threat to people's health. While smoking is a major risk factor for lung cancer, genetic factors can also contribute to the development of lung cancer. In the past 20 years, genome-wide association studies (GWASs) have strongly linked 45 genetic loci to NSCLC. However, most studies have been conducted in European populations. Recent studies have found another six NSCLC loci in the Chinese population, but there is still a lack of a high-quality reference dataset for the Chinese population. In addition, traditional GWASs mainly studies common and low-frequency variants by attributing gene arrays with haplotype references. Although this method can study more than 90% of common variants, it is difficult to use to study rare variants. In order to provide a high-quality reference dataset for lung cancer research in China, improve the efficiency of array-based genetic research, and reveal the possible genetic causes of lung cancer on a larger scale, this study performed large-scale whole genome sequencing on 2 984 NSCLC patients and 3 020 non-cancer control patients from the Nanjing lung cancer cohort. As a result, 91 942 226 (91.43%) single nucleotide variants (SNVs) and 8 623 324 (8.57%) indels (indels) of 50 bp or less were found. Of these, only 8 079 426 SNVs (8.79%) and 1 082 986 indels (12.56%) were considered common and low-frequency [minor allele frequency (MAF) ≥0.5%], and 20 common and low-frequency loci (MAF ≥0.5%) were identified, including five newly reported loci, SRC, DSP, RGL2, BTN3A2, and CCDC116. For rare variants with loss of function (MAF<0.5%), the study found that BRCA2 and 18 other cancer susceptibility genes affected 5.29% of NSCLC individuals, and 98.91% (181/183) of newly identified loss-of-function variants were associated with NSCLC risk. These association study findings were validated in an independent case-control study including 4 410 individuals and a prospective cohort study including 23 826 individuals.

Lynch syndrome (LS) is the most common hereditary colorectal cancer (CRC) syndrome characterized by germline pathogenic variants in mismatch repair (MMR) -related genes that cause microsatellite instability. Patients who meet the clinical criteria for LS and MMR deficiencies and do not have any identified germline pathogenic variants are generally considered to have Lynch-like syndrome (LLS). These patients are at higher risk of colorectal cancer and extracolonic tumors, but there are limited studies on their underlying genetic etiology. Recent studies have found that there is a biallelic germline variation of the MUTYH gene in LLS cases, and MUTYH-associated polyposis can overlap with the LS phenotype through somatic inactivation of the MMR gene. In addition to this, LLS patients carrying POLE and POLD1 germline variants have also been identified. Germline variants in DNA repair genes, such as MCM8, MCM9, WRN, MCPH1, BARD1, REV3L, EXO1, POLD1, Rfc1, RPA1, and MLH3, have also been reported in patients with LLS, but the results of the potential germline mutation profile of LLS in the Brazilian population are unknown. The study included 20 patients with LLS diagnosed from the Barretus Cancer Hospital in Brazil and performed whole exome sequencing (WES). It was found that among these 20 patients, 60% were female patients, and the age at which the tumor was first diagnosed was (48±8) years old. 75% of patients (n=15) with colorectal cancer as the first tumor, followed by endometrial cancer (n=2), ovarian cancer (n=2) and gastric cancer (n=1）。 Five patients were diagnosed with a second type of tumor; These tumors include colorectal cancer, endometrial cancer, breast cancer, and non-melanoma. A total of 90% of patients had a family history of cancer and 75% had LS-associated tumors in their families. In addition, 319 high-quality germline variants were screened in 2 389 gene analyses. Variants were classified according to the American Society for Medical Genetics and Genomics (ACMG) criteria, of which, 33.5% of the variants were classified as benign or probably benign (107/319), 63.9% of the variants were of uncertain significance (204/319), and 2.5% of the variants were classified as pathogenic or probably pathogenic (8/319). Among the known cancer genes MUTYH and ATM, pathogenic or likely pathogenic variants were found in 35% (7/20) of patients. These patients carrying the pathogenic variant compared with those who did not carry the pathogenic variant at the age at which the tumor was first diagnosed [mean age 48.3 vs 48.2 years,P=0.974], number of tumor families (mean number of tumors 3.7 vs 6.7,P=0.193), tumor grade (P=0.650) or tumor stage (P=0.854) None of the differences were statistically significant. In addition, rare potentially pathogenic variants have also been identified in the DNA repair gene POLN and other cancer-related genes PPARG, CTC1, DCC, and ALPK1.