Identification of clinically actionable secondary genetic variants from whole‐genome sequencing in a large‐scale Chinese population

Clinical DNA sequencing is increasingly being chosen as a diagnostic test for Mendelian disorders in genomic medicine. Besides the primary findings, clinically actionable secondary genetic variants could be detected in the DNA sequencing. The genetic variants from genes proposed by American College of Medical Genetics and Genomics (ACMG) should be reported to clinician as secondary findings if the annotation suggested pathogenic or likely pathogenic.1 With the increasing application of DNA sequencing in the clinic, the ACMG updated the SF v3.0 list to 73 genes in 2021.2 Ethnic disparities exist in allele frequency of pathogenic variants. From the NHLBI Exome Sequencing Project (ESP), 0.7% and 0.5% of adults of European and African ancestry, respectively, were expected to have highly actionable penetrant pathogenic variants.3 Approximately 7% of 196 Korean individuals exhibited pathogenic variants,4 and at least one pathogenic variant was reported in 21% of 2049 Japanese individuals.5 The carrier frequency of secondary findings was highly variable among populations, but the prevalence of pathogenic or likely pathogenic variants (P/LP) in Chinese population remains unclear. We analysed 4480 individuals' whole-genome sequencing data from Westlake BioBank for Chinese pilot project (WBBC)6, 7 to evaluate the prevalence of pathogenic genetic variants in the Chinese population for the 73 genes recommended by ACMG, and further investigated the ethnic differences among worldwide populations. A total of 9373 variants were found in the coding region, splicing site, intron and UTR in the WBBC samples, with 97.3% of these being missense and synonymous variants (Table S1). Following the variant classification standard (Figure 1 and Supporting Information), we identified 295 P/LP variants (99 pathogenic and 196 likely pathogenic variants, Table S2), accounting for 3.15% of the variants. For autosomal dominant inheritance (AD), the ratio of the P/LP variants was highest for TNNT2 (24.14%), LDLR (21.65%) and SCN5A (14.69%) genes (Table S3). The highest ratio of the P/LP variants was shown by MUTYH (24.07%), ATP7B (23.93%) and GAA (12.93%) for the autosomal recessive inheritance (AR). Additionally, 20% (3/15) of the variants were P/LP variants in GLA (X-linked inheritance) gene. At the population level, approximately 17.37% (778/4480) of Chinese individuals carried at least one reported P/LP variant, whereas 4.2% (186/4480) of individuals had the pathogenic (P) variants. Because the 4480 samples also included individuals with Parkinson's disease (PD), we estimated a population frequency of 16.6% for P/LP variants in the PD patients and 18% in relatively healthy individuals. The proportion of P/LP carriers showed no significant differences between the PD patients and relatively healthy individuals (p = .297). Excluding the autosomal recessive condition carriers, the prevalence of P/LP variants was 10.9% (488/4480) compared to 1.4% (62/4480) for pathogenic variants in the WBBC cohort. For the autosomal dominant cardiovascular and cancer diseases, we found that 7.32% and 2.67% of the individuals carried P/LP variants in 31 cardiovascular and 27 cancer genes, respectively. A closer look at the single gene, MUTYH (3.15%, AR), ATP7B (2.86%, AR), SCN5A (1.96%, AD), LDLR (1.72%, AD) and GAA (1.03%, AR) showed a relatively high population frequency of the P/LP variants in the Chinese population (Table S3). Our study observed significant ethnic differences in allele frequency of likely pathogenic or pathogenic variants between Chinese and European populations (Table 1 and Figure 2). We found that 24 P/LP variants from 15 genes exhibited relatively remarkable ethnic differences (Table 1). The minor allele frequencies of variants p.Pro5Gln (MSH2, Figure 2A), c.850-2A>G (MUTYH, Figure 2B) and p.Ala1180Val (SCN5A, Figure 2D) in the WBBC were relatively higher than in non-East Asian populations (Supporting Information). Contrastingly, p.Gly382Asp (MUTYH, Figure 2C), c.-32-13T>G (GAA, Figure 2E) and p.Asp444His (BTD, Figure 2F) showed a significantly high allele frequency in European population. We found an unusual difference in the pathogenic variant p.Asp444His in the BTD gene where the allele frequency exceeded 2% in South Asian, European and Admixed American populations (MAF_SAS = 0.035, MAF_EUR = 0.043 and MAF_AMR = 0.019). However, this variant was very rarely detected in the East Asian population (MAF_WBBC = 0.0006 and MAF_EAS = 0). In fact, the prevalence of biotinidase deficiency in East Asian (1/15 000 in Japanese and 1/620 400 in Chinese8) was lower than other ethnic groups (e.g., 1/9000 in Brazil9; please refer to the Supporting Information for more details). To access the full list of the variants, we provided a user-friendly website to search for the annotation and frequency of variants in Chinese and other populations (https://wbbc.westlake.edu.cn/). Considering the ethnic discrepancies in incidence of diseases, the recommendation list should include highly penetrant phenotypes and genes in the East Asian population. Citrin deficiency, an inherited autosomal recessive metabolic disease, was initially reported and found mostly in individuals of East Asian ancestry.10 We found four heterozygous pathogenic variants of SLC25A13, c.550C>T (p.Arg184*), c.615+5G>A, c.852_855del and c.1180+1G>A in 1.5% (66/4480) in the individuals from WBBC. The c.852_855del variant in SLC25A13 gene was the most common variants among East Asians (MAF_WBBC = 0.006 and MAF_EAS = 0.004) but rarely detected in other populations. In conclusion, we found that approximately 17.37% (778/4480) of Chinese individuals carried at least one reported P/LP variant in the 73 genes recommended by ACMG, and 295 P/LP genetic variants were detected in our WBBC pilot cohort. We observed ethnic differences in allele frequency of P/LP variants between Chinese and European populations, 24 P/LP variants from 15 genes exhibited relatively remarkable ethnic differences (such as rs13078881 on BTD for biotinidase deficiency). We also suggested that high-penetrance genes (e.g., SLC25A13 gene for citrin deficiency) in the East Asians should be included in the recommendation list. Prevention and early intervention could reduce the risk of potentially severe consequences of genetic disorders for the undiagnosed carriers; therefore, secondary findings should be incorporated in clinical DNA sequencing reports appropriately. This study was supported by a grant from National Natural Science Foundation of China (32061143019). We are grateful to staffs from the High-Performance Computing Center at Westlake University for technical support. The authors declare that there is no conflict of interest. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.