Abstract
Background
Thyroid cancer, the leading endocrine tumor with a rising global incidence, especially in women, is influenced by both genetic and environmental factors. This study examines the relationship between polygenic risk scores (PRS) and thyroid cancer susceptibility in the Han Chinese population, as well as the impact of genetic variants on clinical outcomes.
Methods
Analyzing data from 57 257 participants in the Taiwan Precision Medicine Initiative, the study employed the Affymetrix Genome-Wide TWB 2.0 SNP Array for genotyping. PRS were calculated using single nucleotide polymorphisms (SNPs) from prior genome-wide association studies, specifically PGS000087 and PGS000797, and correlated with clinical parameters like age, sex, comorbidities, and treatment methods.
Results
Among 4063 participants with thyroid tumors (839 malignant, 3224 benign), higher PRS quartiles correlated significantly with increased thyroid cancer incidence. The highest quartile showed a 1.15-fold (PGS000797) and 1.14-fold (PGS000087) greater risk than the lowest quartile. Key findings included an association between higher PRS quartiles and younger onset age, along with a notable link to chronic kidney disease and thyroid hormone levels in specific SNPs.
Conclusion
The study demonstrates PRS's utility in predicting thyroid cancer risk in the Han Chinese population, with higher PRS associated with increased risk and distinct clinical features. While this study focuses on the Han Chinese population, we recognize the importance of comparing PRS performance across different ancestries to fully understand ethnic genetic diversity in cancer risk assessment. Future studies should aim to include such comparative analysis.
Keywords: thyroid cancer, polygenic risk score, genome-wide association study, Han Chinese population, personalized medicine
Thyroid cancer is the most prevalent endocrine tumor. According to Global Cancer Statistics, it ranks ninth among 36 cancer types, with an estimated 567 000 cases worldwide in 2018 [1]. The incidence of thyroid cancer has been increasing over recent decades. In Taiwan, 4932 new cases were diagnosed in 2020, making it the fourth most common cancer in females and 30th in males. A large population-based study indicates a rising overall incidence of thyroid cancer in Taiwan over a 25-year period, across all age groups and major tumor types [2]. Globally, the incidence rate of thyroid cancer in women is 3 times higher than in men [1], a trend also observed in Taiwan.
The exact etiology of thyroid cancer remains elusive, but it is widely accepted that both genetic and environmental factors contribute to its development. Among various cancers, thyroid cancer exhibits the highest heritability, estimated at 53% [3]. Genome-wide association studies (GWAS) have identified numerous single nucleotide polymorphisms (SNPs) significantly associated with thyroid cancer [4-8]. Each genetic variant individually exerts a small effect, but individuals carrying multiple risk variants are considered at higher risk for the disease. Consequently, the concept of a polygenic risk score (PRS) has been introduced. PRS represents the cumulative weighted value of SNPs carried by an individual, allowing researchers to assess the collective contribution of these variants to disease prevalence and incidence in populations [9]. For example, Liyanarachchi et al's study, encompassing 2801 cases and 539 094 controls from 3 large cohorts in Ohio, Iceland, and the United Kingdom, demonstrated that a PRS based on 10 SNPs showed strong predictive power for risk. Individuals in the top decile group had a 6.9-fold greater risk compared to those in the bottom decile [10]. PRS has proven to be a significant predictor of thyroid cancer risk.
A meta-analysis of 2 large cohorts, Genetic Epidemiology Research on Adult Health and Aging and UK Biobank, revealed that a standardized PRS increase of 1 SD was associated with a 1.55-fold increased risk of thyroid cancer [11]. Another study by Kachuri et al from the UK Biobank showed that a standardized PRS increase of 1 SD was associated with a 1.75-fold increased risk of thyroid cancer [12]. These findings predominantly pertain to individuals of European ancestry. The present study aims to investigate the association between PRS and susceptibility to thyroid cancer in the Han Chinese population. Additionally, we assess the impact of genome-wide susceptibility variants on the clinical outcomes of thyroid cancer patients in this population.
Materials and Methods
Study Population
A total of 57 257 participants who volunteered to participate in the Taiwan Precision Medicine Initiative (TPMI) project supervised by Academia Sinica, Taiwan from Taichung Veterans General Hospital (TCVGH) were enrolled between June 2019 and May 2021. Detailed information on participants including medical records, physical examinations, and blood tests were collected. All received genotypes were determined using the Affymetrix Genome-Wide TWB 2.0 SNP Array. The study was approved by the ethics committee of TCVGH Institutional Review Board (IRB No. CE23417A). Written informed consent was obtained from all participants. Clinical parameters were obtained from the dataset and electronic medical records from TCVGH using a deidentification method.
Genotyping and Quality Control
We extracted DNA from the blood samples of the participants and genotyping was performed using Axiom Genome-Wide TWB 2.0 Array Plate (Affymetrix, Santa Clara, CA, USA), which contains 714 431 SNPs and is designed specifically for Taiwan's Han Chinese population [13]. Analysis and quality control were performed using Affymetrix Power Tools software, and markers that failed Hardy–Weinberg equilibrium tests with a P < 1.0 × 10–5, had a minor allele frequency <0.05, or had a genotype missing rate of >5% were excluded. After quality control, a total of 591 048 SNPs were retained for analysis. The use of high-coverage GWAS SNP data from large-scale Han Chinese ancestry in Taiwan using custom arrays has been previously described [14]. The samples with a missingness rate >0.02 and an inbreeding coefficient >0.15 and those with a sex mismatch were removed. Genotype data and selected SNPs were merged according to merged imputation data.
PRS Analysis
PRS was calculated by using the “score” function from plink version 1.9 to aggregate the effects of multiple genetic variants weighted by their effects size [15]. The PRS used in this study was PGS000087 and PGS000797. PGS000087 was derived from a discovery analysis of 12 variants associated with thyroid cancer identified from 2 GWASs [11]. PGS000797 was derived from PGS000087 with inverse variant weights [12]. The list of SNPs and effect sizes were downloaded from the polygenic score catalog.
Clinical Parameters
Patients diagnosed with thyroid cancer were identified based on International Classification of Diseases, Ninth Revision code 193, along with pathological proof. Additionally, benign thyroid tumor was identified using International Classification of Diseases, Ninth Revision codes 226, 239.7, 241.0, 241.9, and 246.2. The incidence of thyroid cancer among participants was analyzed. Subsequently, we examined PRS associations with clinical parameters such as age at diagnosis, sex, comorbidities [chronic kidney disease (CKD), hypertension, diabetes mellitus, ischemic heart disease, cerebrovascular disease, hyperlipidemia, family history, and other cancer], biomarkers (TSH, thyroglobulin, T3, T4, and free T4), and surgery.
Statistical Analysis
The demographic data are shown as mean ± SD for continuous variables. An ANOVA was conducted for continuous variables, while categorical variables were presented as number (percent) and analyzed using Chi-square test. The PRS was assessed as a categorical variable, categorized into 4 groups based on quartiles of PRS values, namely Q1 (0-25%), Q2 (26-50%), Q3 (51-75%), and Q4 (76-100%). Odds ratio (OR) and 95% confidence interval (CI) of PRS were calculated by univariate logistic regression analysis to adjust potential confounders. The use of quartile comparison was an initial approach to categorize risk levels. Significance of difference was set as P-value smaller than .05. All statistical analyzes were conducted using SAS version 9.4. (SAS Institute Inc., Cary, NC, USA) and IBM SPSS statistical software for Windows, version 22.0 (IBM Corp., Armonk, NY, USA).
Results
Among the 57 257 participants from the TPMI project in TCVGH, a total of 4063 participants were included in the analysis, with 839 patients diagnosed with malignant thyroid tumor and 3224 patients with benign thyroid tumor. The participants were categorized into 4 groups (Q1 to Q4) by quartile of PRS. The incidence of thyroid cancer in PGS000797 group was significantly higher in Q4 compared to Q1 (OR = 1.151, 95% CI = 1.053-1.258, P < .0001). In the PGS000087 group, incidence was significantly higher in Q4 compared to Q1 (OR = 1.141, 95% CI = 1.049-1.254, P < .0001), as shown in Table 1.
Table 1.
Odd ratios for thyroid cancer in study population by PRS
| Comparison | OR | 95% CI | P-valuea | |
|---|---|---|---|---|
| PGS000797 | ||||
| Q4|Q1 | 1.151 | 1.053 | 1.258 | <.001 |
| Q3|Q1 | 1.019 | 0.93 | 1.116 | .3958 |
| Q2|Q1 | 1.01 | 0.922 | 1.107 | .2596 |
| PGS000087 | ||||
| Q4|Q1 | 1.147 | 1.049 | 1.254 | <.001 |
| Q3|Q1 | 1.028 | 0.938 | 1.127 | .656 |
| Q2|Q1 | 0.997 | 0.91 | 1.093 | .126 |
Abbreviations: CI, confidence interval; OR, odds ratio; PRS, polygenic risk score.
a Comparisons of categorical variables were analyzed using logistic regression.
The characteristics of the 4 quartiles of PRS among the 4063 patients with thyroid tumor in the PGS000797 group are presented in Table 2. The onset age was found to be statistically significantly younger in Q3 (Q1, 52.85 ± 12.82 years; Q2, 53.03 ± 13.15 years; Q3, 51.49 ± 13.05 years; Q4, 52.16 ± 13.6 years, P = .0325). There was no statistically significant association between the 4 quartiles of PRS, comorbidities and biochemical analysis. When looking at surgery received, there was a statistically significantly higher proportion of patients undergoing partial thyroidectomy (0.29%), lobectomy (15.13%), and total thyroidectomy (8.45%) in Q4.
Table 2.
Characteristics of the study subjects (PGS000797)
| Variable | Quartile of polygenic risk score (PGS000797) | P | |||
|---|---|---|---|---|---|
| Q1 (n = 1019) | Q2 (n = 1014) | Q3 (n = 1016) | Q4 (n = 1018) | ||
| Demography | |||||
| Age (mean/SD) | 59.75 (13.35) | 60.11 (13.41) | 58.50 (13.31) | 59.26 (13.98) | .0444 |
| Onset age (mean/SD) | 52.85 (12.82) | 53.03 (13.15) | 51.49 (13.05) | 52.16 (13.56) | .0325 |
| Sex (%) | .3579 | ||||
| Female | 768 (75.37) | 779 (76.82) | 776 (76.38) | 750 (73.67) | |
| Male | 251 (24.63) | 235 (23.18) | 240 (23.62) | 268 (26.33) | |
| Medical history (comorbidities) (%) | |||||
| Chronic kidney disease | .2032 | ||||
| No | 955 (93.72) | 944 (93.10) | 928 (91.34) | 942 (92.53) | |
| Yes | 64 (6.28) | 70 (6.90) | 88 (8.66) | 76 (7.47) | |
| Hypertension | .3734 | ||||
| No | 841 (82.53) | 818 (80.67) | 844 (83.07) | 849 (83.40) | |
| Yes | 178 (17.47) | 196 (19.33) | 172 (16.93) | 169 (16.60) | |
| Diabetes mellitus | .7550 | ||||
| No | 907 (89.01) | 891 (87.87) | 907 (89.27) | 900 (88.41) | |
| Yes | 112 (10.99) | 123 (12.13) | 109 (10.73) | 118 (11.59) | |
| Ischemic heart disease | .9471 | ||||
| No | 944 (92.64) | 934 (92.11) | 941 (92.62) | 938 (92.14) | |
| Yes | 75 (7.36) | 80 (7.89) | 75 (7.38) | 80 (7.86) | |
| Cerebrovascular disease | .6794 | ||||
| No | 961 (94.31) | 944 (93.10) | 949 (93.41) | 949 (93.22) | |
| Yes | 58 (5.69) | 70 (6.90) | 67 (6.59) | 69 (6.78) | |
| Hyperlipidemia | .7209 | ||||
| No | 869 (85.28) | 848 (83.63) | 861 (84.74) | 867 (85.17) | |
| Yes | 150 (14.72) | 166 (16.37) | 155 (15.26) | 151 (14.83) | |
| Family history | .1696 | ||||
| No | 108 (70.59) | 110 (79.14) | 122 (71.76) | 103 (67.76) | |
| Yes | 45 (29.41) | 29 (20.86) | 48 (28.24) | 49 (32.24) | |
| Other cancer | .1723 | ||||
| No | 868 (85.18) | 874 (86.19) | 898 (88.39) | 874 (85.85) | |
| Yes | 151 (14.82) | 140 (13.81) | 118 (11.61) | 144 (14.15) | |
| Laboratory data (mean/SD) | |||||
| TSH | 5.83 (22.10) | 8.05 (28.15) | 7.94 (27.37) | 8.96 (25.61) | .1157 |
| Thyroglobulin | 118.68 (760.22) | 120.83 (578.94) | 221.18 (2091) | 180.71 (1856.36) | .8829 |
| T3 | 91.17 (53.42) | 87.64 (52.15) | 88.42 (45.47) | 82.60 (43.73) | .2132 |
| T4 | 8.15 (2.88) | 8.55 (2.35) | 7.83 (2.63) | 8.08 (3.11) | .3197 |
| Free T4 | 8.59 (6.45) | 8.57 (6.70) | 8.83 (7.25) | 8.72 (7.68) | .8876 |
| Surgery (%) | |||||
| Partial thyroidectomy | 2 (0.2) | 1 (0.10) | 1 (0.10) | 3 (0.29) | .0258 |
| Lobectomy | 116 (11.38) | 108 (10.65) | 129 (12.70) | 154 (15.13) | .0128 |
| Near-total/subtotal thyroidectomy | 7 (0.69) | 1 (0.10) | 8 (0.79) | 5 (0.49) | .0888 |
| Total thyroidectomy | 51 (5.00) | 60 (5.92) | 79 (7.78) | 86 (8.45) | .0063 |
| Neck lymph node dissection | 7 (0.69) | 6 (0.59) | 12 (1.18) | 12 (1.18) | .3413 |
Among the 4063 patients with thyroid tumor in the PGS000087 group, as Table 3, there was no significant difference in onset age (Q1, 52.92 ± 12.94 years; Q2, 52.95 ± 13.24 years; Q3, 51.81 ± 13.24 years; Q4, 51.83 ± 13.32 years, P = .0622). In terms of comorbidity, we observed a statistically significant difference association between the 4 quartiles of PRS and the incidence of CKD. A higher proportion of CKD was found in Q3 [Q1, 65 (6.43%); Q2, 79 (7.61%); Q3, 93 (9.32%); Q4, 61 (59.8%), P = .0196]. We also found mean TSH level was significant high in Q4 (Q1, 5.38 µIU/mL; Q2, 7.69 µIU/mL; Q3, 7.83 µIU/mL; Q4, 9.80 µIU/mL, P = .0119). Patients in Q4 had a significantly higher proportion of patients undergoing total thyroidectomy [Q1, 50 (4.95%); Q2, 68 (6.55%); Q3, 76 (7.62%); Q4, 82 (8.04%), P = .0277].
Table 3.
Characteristics of the study subjects (PGS000087)
| Variable | Quartile of polygenic risk score (PGS000087) | P | |||
|---|---|---|---|---|---|
| Q1 (n = 1011) | Q2 (n = 1038) | Q3 (n = 998) | Q4 (n = 1020) | ||
| Demography | |||||
| Age (mean/SD) | 59.88 (13.51) | 59.96 (13.44) | 58.82 (13.43) | 58.94 (13.68) | .1081 |
| Onset age (mean/SD) | 52.92 (12.94) | 52.95 (13.09) | 51.81 (13.24) | 51.83 (13.32) | .0622 |
| Sex (%) | .4307 | ||||
| Female | 754 (74.58) | 797 (76.78) | 764 (76.55) | 758 (74.31) | |
| Male | 257 (25.42) | 241 (23.22) | 234 (23.45) | 262 (25.69) | |
| Medical history (comorbidities) (%) | |||||
| Chronic kidney disease | .0196 | ||||
| No | 946 (93.57) | 959 (92.39) | 905 (90.68) | 959 (94.02) | |
| Yes | 65 (6.43) | 79 (7.61) | 93 (9.32) | 61 (5.98) | |
| Hypertension | .6760 | ||||
| No | 842 (83.28) | 859 (82.76) | 811 (81.26) | 840 (82.35) | |
| Yes | 169 (16.72) | 179 (17.24) | 187 (18.74) | 180 (17.65) | |
| Diabetes mellitus | .8751 | ||||
| No | 902 (89.22) | 914 (88.05) | 885 (88.68) | 904 (88.63) | |
| Yes | 109 (10.78) | 124 (11.95) | 113 (11.32) | 116 (11.37) | |
| Ischemic heart disease | .4639 | ||||
| No | 938 (92.78) | 948 (91.33) | 929 (93.09) | 942 (92.35) | |
| Yes | 73 (7.22) | 90 (8.67) | 69 (6.91) | 78 (7.65) | |
| Cerebrovascular disease | .7825 | ||||
| No | 949 (93.87) | 964 (92.87) | 933 (93.49) | 957 (93.82) | |
| Yes | 62 (6.13) | 74 (7.13) | 65 (6.51) | 63 (6.18) | |
| Hyperlipidemia | .8283 | ||||
| No | 863 (85.36) | 880 (84.78) | 837 (83.87) | 865 (84.80) | |
| Yes | 148 (14.64) | 158 (15.22) | 161 (16.31) | 155 (15.20) | |
| Family history | .2787 | ||||
| No | 105 (68.63) | 112 (78.32) | 117 (70.91) | 109 (71.24) | |
| Yes | 48 (31.37) | 31 (21.68) | 48 (29.09) | 44 (28.76) | |
| Other cancer | .6299 | ||||
| No | 0.6299 | 860 (82.85) | 847 (84.87) | 859 (84.22) | |
| Yes | 167 (16.52) | 178 (17.15) | 151 (15.13) | 161 (15.78) | |
| Laboratory data (mean/SD) | |||||
| TSH | 5.38 (19.47) | 7.69 (26.67) | 7.83 (27.26) | 9.80 (28.91) | .0119 |
| Thyroglobulin | 125.93 (781.67) | 150.59 (705.48) | 200.07 (2045) | 168.17 (1829.46) | .9699 |
| T3 | 91.16 (53.24) | 85.01 (38.20) | 89.73 (44.54) | 83.69 (56.17) | .2154 |
| T4 | 8.31 (2.66) | 8.43 (2.36) | 7.85 (3.01) | 7.99 (2.95) | .4051 |
| Free T4 | 8.69 (6.42) | 8.52 (6.57) | 8.90 (7.55) | 8.60 (7.53) | .7774 |
| Surgery (%) | |||||
| Partial thyroidectomy | 2 (0.20) | 2 (0.19) | 0 () | 3 (0.29) | .5109 |
| Lobectomy | 3 (0.20) | 121 (11.66) | 126 (12.63) | 150 (14.71) | .0535 |
| Near-total/subtotal thyroidectomy | 4 (0.20) | 3 (0.29) | 7 (0.70) | 5 (0.49) | .6062 |
| Total thyroidectomy | 5 (0.20) | 68 (6.55) | 76 (7.62) | 82 (8.04) | .0277 |
| Neck lymph node dissection | 6 (0.20) | 10 (0.96) | 9 (0.90) | 13 (1.27) | .3243 |
The characteristics of the 4 quartiles of PRS among the patients with malignant thyroid tumor and benign thyroid tumor are presented independently for comparison. Among the 839 patients with malignant thyroid tumor in the PGS000797 group (Table 4), the onset age was found to be statistically significantly younger in Q4 (Q1, 51 ± 12.64 years; Q2, 51.25 ± 12.99 years; Q3, 49.89 ± 12.50 years; Q4, 48.11 ± 12.78 years, P = .0344). There was statistically significant association among the 4 quartiles of PRS in terms of hypertension, diabetes mellitus, and hyperlipidemia. Patients in Q2 had the highest rate of hypertension [Q1, 33, (21.29%); Q2, 41 (21.35%); Q3, 30 (13.70%); Q4, 25 (9.16%), P = .0004], diabetes mellitus [Q1, 19 (12.26%); Q2, 24 (12.50%); Q3, 19 (8.68%); Q4, 12 (4.40%), P = .0068], and hyperlipidemia [Q1, 28 (18.06%); Q2, 30 (18.75%); Q3, 28 (12.79%); Q4, 22 (8.06%), P = .0026] among the 4 quartiles of PRS. However, among the patients with malignant thyroid tumor in the PGS000087 group, there was no statistically significant association between the 4 quartiles of PRS and onset age, sex, comorbidities, biochemical analysis, and surgery. The detailed values of each item of malignant thyroid tumor in the PGS000087 group are listed in Table 5.
Table 4.
Characteristics of the patients with malignant thyroid tumor (PGS000797)
| Variable | Quartile of polygenic risk score (PGS000797) | P | |||
|---|---|---|---|---|---|
| Q1 (n = 155) | Q2 (n = 192) | Q3 (n = 219) | Q4 (n = 273) | ||
| Demography | |||||
| Age (mean/SD) | 58.25 (12.92) | 58.82 (12.92) | 56.81 (12.51) | 55.68 (13.23) | .0458 |
| Onset age (mean/SD) | 51 (12.64) | 51.25 (12.99) | 49.89 (12.50) | 48.11 (12.78) | .0344 |
| Sex (%) | .6546 | ||||
| Female | 120 (77.42) | 146 (76.04) | 160 (73.06) | 212 (77.66) | |
| Male | 35 (22.58) | 46 (23.96) | 59 (26.94) | 61 (22.34) | |
| Medical history (comorbidities) (%) | |||||
| Chronic kidney disease | .9506 | ||||
| No | 147 (94.84) | 183 (95.31) | 207 (94.52) | 261 (95.60) | |
| Yes | 8 (5.16) | 9 (4.69) | 12 (5.48) | 12 (4.40) | |
| Hypertension | .0004 | ||||
| No | 122 (78.71) | 151 (78.65) | 189 (86.30) | 248 (90.84) | |
| Yes | 33 (21.29) | 41 (21.35) | 30 (13.70) | 25 (9.16) | |
| Diabetes mellitus | .0068 | ||||
| No | 136 (87.74) | 168 (87.50) | 200 (91.32) | 261 (95.60) | |
| Yes | 19 (12.26) | 24 (12.50) | 19 (8.68) | 12 (4.40) | |
| Ischemic heart disease | .0963 | ||||
| No | 144 (92.90) | 179 (93.23) | 205 (93.61) | 266 (97.44) | |
| Yes | 11 (7.10) | 13 (6.77) | 14 (6.39) | 7 (2.56) | |
| Cerebrovascular disease | .0958 | ||||
| No | 145 (93.55) | 180 (93.75) | 206 (94.06) | 267 (97.80) | |
| Yes | 10 (6.45) | 12 (6.25) | 13 (5.94) | 6 (2.20) | |
| Hyperlipidemia | .0026 | ||||
| No | 127 (81.94) | 156 (81.25) | 191 (87.21) | 251 (91.94) | |
| Yes | 28 (18.06) | 36 (18.75) | 28 (12.79) | 22 (8.06) | |
| Family history | .2902 | ||||
| No | 20 (66.67) | 31 (81.58) | 26 (70.27) | 41 (82.00) | |
| Yes | 10 (33.33) | 7 (18.42) | 11 (29.73) | 9 (18.00) | |
| Other cancer | .8442 | ||||
| No | 134 (86.45) | 161 (83.85) | 182 (83.11) | 229 (83.88) | |
| Yes | 21 (13.55) | 31 (16.15) | 37 (16.89) | 44 (16.12) | |
| Laboratory data (mean/SD) | |||||
| TSH | 19.02 (45.69) | 23.60 (49.05) | 22.91 (48.02) | 21.29 (39.28) | .8051 |
| Thyroglobulin | 128.19 (897.93) | 98.74 (608.44) | 68.38 (468.30) | 193.89 (2021.40) | .7870 |
| T3 | 76.08 (23.88) | 77.12 (25.53) | 76.73 (23.53) | 72.71 (24.54) | .4269 |
| T4 | 8.52 (3.38) | 9.73 (2.67) | 8.93 (3.96) | 8.28 (3.37) | .448 |
| Free T4 | 8.29 (6.64) | 9.08 (6.62) | 8.48 (6.84) | 9.15 (6.63) | .5126 |
| Surgery (%) | |||||
| Partial thyroidectomy | 2 (1.29) | 0.00 (0) | 0.00 (0) | 3 (1.10) | .1398 |
| Lobectomy | 52 (33.55) | 62 (32.29) | 72 (32.88) | 95 (34.80) | .9462 |
| Near-total/subtotal thyroidectomy | 4 (2.58) | 0.00 (0) | 2 (0.91) | 3 (1.10) | .1302 |
| Total thyroidectomy | 36 (23.23) | 39 (20.31) | 67 (30.59) | 71 (26.01) | .1031 |
| Neck lymph node dissection | 7 (4.52) | 6 (3.13) | 12 (5.48) | 12 (4.40) | .7172 |
Table 5.
Characteristics of the patients with malignant thyroid tumor (PGS000087)
| Variable | Quartile of polygenic risk score (PGS000087) | P | |||
|---|---|---|---|---|---|
| Q1 (n = 152) | Q2 (n = 191) | Q3 (n = 210) | Q4 (n = 286) | ||
| Demography | |||||
| Age (mean/SD) | 58.38 (13.28) | 58.63 (12.76) | 56.65 (12.50) | 55.93 (13.16) | .0812 |
| Onset age (mean/SD) | 51.26 (12.81) | 51.12 (12.51) | 49.63 (12.79) | 48.35 (12.81) | .0511 |
| Sex (%) | .9389 | ||||
| Female | 115 (75.66) | 143 (74.87) | 159 (75.71) | 221 (77.27) | |
| Male | 37 (24.34) | 48 (25.13) | 51 (24.29) | 65 (22.73) | |
| Medical history (comorbidities) (%) | |||||
| Chronic kidney disease | .5826 | ||||
| No | 145 (95.39) | 183 (95.81) | 196 (93.33) | 274 (95.80) | |
| Yes | 7 (4.61) | 8 (4.19) | 14 (6.67) | 12 (4.20) | |
| Hypertension | .5565 | ||||
| No | 124 (81.58) | 161 (84.29) | 177 (84.29) | 248 (86.71) | |
| Yes | 28 (18.42) | 30 (15.71) | 33 (15.71) | 38 (13.29) | |
| Diabetes mellitus | .2076 | ||||
| No | 132 (86.84) | 175 (91.62) | 195 (92.86) | 263 (91.96) | |
| Yes | 20 (13.16) | 16 (8.38) | 15 (7.14) | 23 (8.04) | |
| Ischemic heart disease | .1142 | ||||
| No | 140 (92.11) | 179 (93.72) | 205 (97.62) | 270 (94.41) | |
| Yes | 12 (7.89) | 12 (6.28) | 5 (2.38) | 16 (5.59) | |
| Cerebrovascular disease | .1893 | ||||
| No | 141 (92.76) | 181 (94.76) | 205 (97.62) | 271 (94.76) | |
| Yes | 11 (7.24) | 10 (5.24) | 5 (2.38) | 15 (5.24) | |
| Hyperlipidemia | .5440 | ||||
| No | 127 (83.55) | 164 (85.86) | 181 (86.19) | 253 (88.46) | |
| Yes | 25 (16.45) | 27 (14.14) | 29 (13.81) | 33 (11.54) | |
| Family history | .1278 | ||||
| No | 20 (62.50) | 30 (83.33) | 26 (72.22) | 42 (82.35) | |
| Yes | 12 (37.50) | 6 (16.67) | 10 (27.78) | 9 (17.65) | |
| Other cancer | .9577 | ||||
| No | 129 (84.87) | 162 (84.82) | 177 (84.29) | 238 (83.22) | |
| Yes | 23 (15.13) | 29 (15.18) | 33 (15.71) | 48 (16.78) | |
| Laboratory data (mean/SD) | |||||
| TSH | 15.72 (39.69) | 22.76 (46.64) | 24.12 (49.79) | 22.79 (43.12) | .3464 |
| Thyroglobulin | 133.21 (919.61) | 145.60 (773.99) | 38.90 (160.44) | 177.15 (1962.50) | .7548 |
| T3 | 76.67 (22.91) | 75.89 (26.97) | 77.28 (23.55) | 72.92 (23.77) | .4722 |
| T4 | 9.71 (2.36) | 9.66 (2.58) | 8.08 (4.28) | 8.29 (3.45) | .1919 |
| Free T4 | 8.64 (6.51) | 9.00 (6.65) | 8.42 (6.96) | 9.02 (6.62) | .7734 |
| Surgery (%) | |||||
| Partial thyroidectomy | 2 (1.32) | 1 (0.52) | 0.00 (0) | 2 (0.70) | .4437 |
| Lobectomy | 48 (31.58) | 66 (34.55) | 69 (32.86) | 98 (34.27) | .9273 |
| Near-total/subtotal thyroidectomy | 4 (2.63) | 1 (0.52) | 2 (0.95) | 2 (0.70) | .2949 |
| Total thyroidectomy | 33 (21.71) | 50 (26.18) | 61 (29.05) | 69 (24.13) | .4116 |
| Neck lymph node dissection | 5 (3.29) | 10 (5.24) | 9 (4.29) | 13 (4.55) | .8538 |
Among the 3224 patients with benign thyroid tumor in the PGS000797 group (Table 6), the onset age was found to be statistically significantly younger in Q3 (Q1, 53 ± 12.82 years; Q2, 53.45 ± 13.16 years; Q3, 51.94 ± 13.18 years; Q4, 53.64 ± 13.54 years, P = .0448). In terms of comorbidity, there was a statistically significant association among the 4 quartiles of PRS in other cancer (P = .015). Patients in Q3 had the lowest proportion of concomitant other cancer [Q1, 157 (18.19%); Q2, 145 (17.66%); Q3, 102 (12.83%); Q4, 120 (16.11%), P = .0152]. Among the patients with malignant thyroid tumor in the PGS000087 group (Table 7), there was a statistically significant association between the 4 quartiles of PRS and CKD (P = .0406). Patients in Q3 had the highest proportion of CKD [Q1, 58(6.76%); Q2, 71 (8.38%); Q3, 79 (10.06%); Q4, 49 (6.68%), P = .0406].
Table 6.
Characteristics of the patients with benign thyroid tumor (PGS000797)
| Variable | Quartile of polygenic risk score (PGS000797) | P | |||
|---|---|---|---|---|---|
| Q1 (n = 863) | Q2 (n = 821) | Q3 (n = 795) | Q4 (n = 745) | ||
| Demography | |||||
| Age (mean/SD) | 60.03 (13.42) | 60.40 (13.51) | 58.95 (13.50) | 60.57 (14.02) | .0820 |
| Onset age (mean/SD) | 53.21 (12.82) | 53.45 (13.16) | 51.94 (13.18) | 53.64 (13.54) | .0448 |
| Sex (%) | .0747 | ||||
| Female | 647 (74.97) | 632 (76.98) | 615 (77.36) | 538 (72.21) | |
| Male | 216 (25.03) | 189 (23.02) | 180 (22.64) | 207 (27.79) | |
| Medical history (comorbidities) (%) | |||||
| Chronic kidney disease | .1098 | ||||
| No | 807 (93.51) | 760 (92.57) | 719 (90.44) | 681 (91.41) | |
| Yes | 56 (6.49) | 61 (7.43) | 76 (9.56) | 64 (8.59) | |
| Hypertension | .5721 | ||||
| No | 718 (83.20) | 667 (81.24) | 653 (82.14) | 601 (80.67) | |
| Yes | 145 (16.80) | 154 (18.76) | 142 (17.86) | 144 (19.33) | |
| Diabetes mellitus | .1664 | ||||
| No | 770 (89.22) | 723 (88.06) | 705 (88.68) | 639 (85.77) | |
| Yes | 93 (10.78) | 98 (11.94) | 90 (11.32) | 106 (14.23) | |
| Ischemic heart disease | .3182 | ||||
| No | 799 (92.58) | 754 (91.84) | 734 (92.33) | 672 (90.20) | |
| Yes | 64 (7.42) | 67 (8.16) | 61 (7.67) | 73 (9.80) | |
| Cerebrovascular disease | .1544 | ||||
| No | 815 (94.44) | 763 (92.94) | 741 (93.21) | 682 (91.54) | |
| Yes | 48 (5.56) | 58 (7.06) | 54 (6.79) | 63 (8.46) | |
| Hyperlipidemia | .3761 | ||||
| No | 741 (85.86) | 692 (84.29) | 668 (84.03) | 616 (82.68) | |
| Yes | 122 (14.14) | 129 (15.71) | 127 (15.97) | 129 (17.32) | |
| Family history | .0505 | ||||
| No | 88 (71.54) | 79 (78.22) | 96 (72.18) | 62 (60.78) | |
| Yes | 35 (28.46) | 22 (21.78) | 37 (27.82) | 40 (39.22) | |
| Other cancer | .0152 | ||||
| No | 706 (81.81) | 676 (82.34) | 693 (87.17) | 625 (83.89) | |
| Yes | 157 (18.19) | 145 (17.66) | 102 (12.83) | 120 (16.11) | |
| Laboratory data (mean/SD) | |||||
| TSH | 2.61 (6.96) | 3.26 (14.20) | 2.61 (9.53) | 2.93 (10.62) | .6944 |
| Thyroglobulin | 96.15 (202.32) | 181.16 (489.30) | 910.77 (4797.57) | 111.22 (286.70) | .2600 |
| T3 | 98.68 (61.76) | 94.76 (63.27) | 96.97 (54.85) | 93.10 (55.70) | .8494 |
| T4 | 8.02 (2.72) | 8.18 (2.13) | 7.50 (1.99) | 7.97 (2.97) | .3429 |
| Free T4 | 8.66 (6.41) | 8.39 (6.72) | 8.96 (7.39) | 8.49 (8.17) | .5979 |
| Surgery (%) | |||||
| Partial thyroidectomy | 0 | 1 (0.12) | 1 (0.13) | 0 | .7316 |
| Lobectomy | 64 (7.42) | 46 (5.60) | 57 (7.17) | 59 (7.92) | .2971 |
| Near-total/subtotal thyroidectomy | 3 (0.35) | 1 (0.12) | 6 (0.75) | 2 (0.27) | .2247 |
| Total thyroidectomy | 15 (1.74) | 21 (2.56) | 12 (1.51) | 15 (2.01) | .4695 |
Table 7.
Characteristics of the patients with benign thyroid tumor (PGS000087)
| Variable | Quartile of polygenic risk score (PGS000087) | P | |||
|---|---|---|---|---|---|
| Q1 (n = 858) | Q2 (n = 847) | Q3 (n = 785) | Q4 (n = 734) | ||
| Demography | |||||
| Age (mean/SD) | 60.16 (13.54) | 60.26 (13.58) | 59.37 (13.63) | 60.11 (13.71) | .5439 |
| Onset age (mean/SD) | 53.24 (12.94) | 53.36 (13.19) | 52.41 (13.33) | 53.19 (13.27) | .4540 |
| Sex (%) | .1841 | ||||
| Female | 638 (74.36) | 654 (77.21) | 603 (76.82) | 537 (73.16) | |
| Male | 220 (25.64) | 193 (22.79) | 182 (23.18) | 197 (26.84) | |
| Medical history (comorbidities) (%) | |||||
| Chronic kidney disease | .0406 | ||||
| No | 800 (93.24) | 776 (91.62) | 706 (89.94) | 685 (93.32) | |
| Yes | 58 (6.76) | 71 (8.38) | 79 (10.06) | 49 (6.68) | |
| Hypertension | .3161 | ||||
| No | 717 (83.57) | 698 (82.41) | 632 (80.51) | 592 (80.65) | |
| Yes | 141 (16.43) | 149 (17.59) | 153 (19.49) | 142 (19.35) | |
| Diabetes mellitus | .3902 | ||||
| No | 769 (89.63) | 739 (87.25) | 688 (87.64) | 641 (87.33) | |
| Yes | 89 (10.37) | 108 (12.75) | 97 (12.36) | 93 (12.67) | |
| Ischemic heart disease | .4654 | ||||
| No | 0.4654 | 769 (90.79) | 721 (91.85) | 672 (91.55) | |
| Yes | 61 (7.11) | 78 (9.21) | 64 (8.15) | 62 (8.45) | |
| Cerebrovascular disease | .4567 | ||||
| No | 0.4567 | 783 (92.44) | 725 (92.36) | 686 (93.46) | |
| Yes | 51 (5.94) | 64 (7.56) | 60 (7.64) | 48 (6.54) | |
| Hyperlipidemia | .5151 | ||||
| No | 0.5151 | 716 (84.53) | 654 (83.31) | 612 (83.38) | |
| Yes | 123 (14.34) | 131 (15.47) | 131 (16.69) | 122 (16.62) | |
| Family history | .3802 | ||||
| No | 0.3802 | 82 (76.64) | 91 (70.54) | 67 (65.69) | |
| Yes | 36 (29.75) | 25 (23.36) | 38 (29.46) | 35 (34.31) | |
| Other cancer | .4665 | ||||
| No | 0.4665 | 698 (82.41) | 667 (84.97) | 621 (84.60) | |
| Yes | 144 (16.78) | 149 (17.59) | 118 (15.03) | 113 (15.40) | |
| Laboratory data (mean/SD) | |||||
| TSH | 2.88 (7.69) | 3.20 (13.81) | 2.28 (5.82) | 3.02 (13.06) | .4926 |
| Thyroglobulin | 108.52 (234.27) | 164.15 (477.73) | 767.51 (4324.96) | 110.09 (305.09) | .3786 |
| T3 | 98.00 (61.41) | 91.39 (43.36) | 98.34 (52.90) | 96.06 (76.61) | .6986 |
| T4 | 7.97 (2.63) | 8.05 (2.16) | 7.76 (2.45) | 7.80 (2.60) | .8838 |
| Free T4 | 8.70 (6.40) | 8.37 (6.55) | 9.06 (7.76) | 8.37 (7.99) | .3593 |
| Surgery (%) | |||||
| Partial thyroidectomy | 0 | 1 (0.12) | 0 | 1 (0.14) | .4813 |
| Lobectomy | 62 (7.23) | 55 (6.49) | 57 (7.26) | 52 (7.08) | .9210 |
| Near-total/subtotal thyroidectomy | 2 (0.23) | 2 (0.24) | 5 (0.64) | 3 (0.41) | .5557 |
| Total thyroidectomy | 17 (1.98) | 18 (2.13) | 15 (1.91) | 13 (1.77) | .9658 |
Discussion
Our study aimed to utilize the PRS method to investigate the association between disease-associated variants identified in GWAS and thyroid cancer susceptibility in Han Chinese individuals from the TCVGH-TPMI cohort. We observed that individuals in the top quartile of PRS had a significantly higher risk of thyroid cancer compared to those in the bottom quartile, which is consistent with previous studies [10-12].
In our study, we utilized PRS from 2 distinct sources, namely PGS000087 and PGS000797, both of which comprised an identical set of 12 genetic variants. PGS000087, developed by Graff et al, was derived from a meta-analysis of the Genetic Epidemiology Research on Adult Health and Aging and the UK Biobank within European ancestry populations [11]. Their study reported that an increase of 1 SD in the standardized PRS was associated with a 1.55-fold elevated risk of thyroid cancer. Conversely, PGS000797, developed by Kachuri et al, was constructed from a European ancestry dataset [12]. Their investigation revealed that each 1 SD increase in the standardized PRS was associated with a 1.75-fold heightened risk of thyroid cancer. They also showed that higher PRS was association with elevated risk of thyroid cancer. Both studies are robust in their methodologies, but Kachuri et al provided a more focused analysis on the UK Biobank with a detailed assessment of risk stratification and the impact of genetic and modifiable risk factors.
In our study, we found that the incidence of thyroid cancer in the PGS000797 group was significantly higher in Q4 compared to Q1 (OR = 1.151, 95% CI = 1.053-1.258, P < .0001). In the PGS000087 group, the incidence was also significantly higher in Q4 compared to Q1 (OR = 1.141, 95% CI = 1.049-1.254, P < .0001). In comparison to the studies of Graff et al and Kachuri et al, our research demonstrates a lower level of risk of thyroid cancer. This discrepancy may be attributed to the ethnic diversity among study populations. Our cohort study comprises Han Chinese individuals, while the other 2 studies predominantly focused on European ancestry populations. Further investigation and additional research are needed to thoroughly explore these observed differences.
In our study, we found a statistical difference in the analysis of different surgical procedures. The partial thyroidectomy in the PGS000797 group has a statistical difference; however, the sample size for this subgroup is limited. This reason is that most patients received thyroid lobectomy or total thyroidectomy. It aims to avoid the potential need for a secondary surgery post initial pathological findings malignancy. The 2015 American Thyroid Association Management Guidelines mentioned that partial lobectomy and subtotal thyroidectomy are deemed inappropriate operations for individuals with suspected malignancy [16].
The SNP rs12129938 is an intronic mutation in the PCNXL2 gene at 1q42.2 [8]. Runmei Hao et al reported that rs12129938 enhanced susceptibility to thyroid cancer at age > 45 years, but this SNP decreased thyroid cancer risk at age ≤ 45 years [17]. In our study, we did not explore the correlation between age, PGS, and the incidence of thyroid cancer. Subsequent research should aim to elucidate whether there is an association between age, PGS, and the incidence of thyroid cancer. Rs11693806, rs2466076, rs1588635, and rs116909374 are associated with serum levels of thyroid function-related hormones (ie, TSH, T3, or free T4). Rs116909374 is associated not only with thyroid cancer but also with low TSH levels [6]. Rs56062135 is not just associated with thyroid cancer but also exhibits associations with asthma and coronary artery disease [18, 19]. Some SNPs are associated with other cancers. The intergenic variant rs7902587 in 10q24.33 is significantly associated with lung cancer and ovarian cancer, and rs11693806 in 2q35 is associated with breast cancer. Rs6793295 is a missense mutation in the LRRC34 gene in 3q26.2 and has been previously reported to associate with thyroid cancer, interstitial lung disease, bladder cancer, multiple myeloma, and telomere length. Rs73227498 in the EPB41L4A gene in 5q22.1 has the second-highest expression in the thyroid. Some SNPs used in PGS000087 and PGS000797 are associated with cancer, but only the benign thyroid tumor in the PGS000797 group showed a statistically significant association with other cancers in our study. This may be attributed to the potential lack of correlation between these SNPs and cancer within our population. This requires further research to confirm.
The use of PGS000087 and PGS000797 in the current study was based on their specific relevance to thyroid cancer in Asian populations, as detailed in prior GWAS studies. However, the authors recognize the merit in using a more diverse, multiethnic PRS for a comprehensive analysis and will consider this for future research. The choices were based on preliminary correlations observed in the dataset and the common chronic diseases in our region. This study has several other limitations that should be acknowledged. First, the study was limited by a small case number and no radiation exposure history was accounted. With an increasing patient numbers in TPMI, a more detailed analysis of genetic variants and their association with the disease subtypes may be presented. Second, the focus on East Asian ancestry was due to the study's specific context within a Han Chinese population. The authors agree that a comparison with other ancestries could provide broader insights, and PRS performance can vary across populations. Future studies should aim to include such comparative analysis. Finally, the use of quartile comparison was an initial approach to categorize risk levels. For a continuous variable such as PRS, the multivariate regression analysis adjusted for covariates would provide a more nuanced understanding of the association and reduce potential confounding.
Conclusions
In this cohort study conducted in a hospital setting, we observed that individuals in the top quartile of PRS were more susceptible to thyroid cancer. In the PGS000797 group, it tended to be diagnosed at a younger age, whereas in the PGS000087 group, no such correlation was observed. We found associations between patients with thyroid cancer in PGS000797 and hypertension, diabetes mellitus, or hyperlipidemia. To further validate our findings, PRS scoring derived from a cohort with a longer follow-up, such as the TPMI cohort, would be preferable.
Acknowledgments
The authors would like to thank the Taiwan Precision Medicine Initiative, a partnership between Academia Sinica and medical centers in Taiwan.
Contributor Information
Yi-Hao Chen, Department of Surgery, Taichung Veterans General Hospital, Taichung City 407219, Taiwan.
I Chieh Chen, Department of Medical Research, Taichung Veterans General Hospital, Taichung City 407219, Taiwan.
Chia-Man Chou, School of Medicine, National Yang Ming Chiao Tung University, Taipei City 11221, Taiwan; Division of Pediatric Surgery, Department of Surgery, Taichung Veterans General Hospital, Taichung City 407219, Taiwan; Department of Post-Baccalaureate Medicine, College of Medicine, National Chung Hsing University, Taichung 402202, Taiwan.
Sheng-Yang Huang, Email: drugholic@vghtc.gov.tw, School of Medicine, National Yang Ming Chiao Tung University, Taipei City 11221, Taiwan; Division of Pediatric Surgery, Department of Surgery, Taichung Veterans General Hospital, Taichung City 407219, Taiwan; Department of Post-Baccalaureate Medicine, College of Medicine, National Chung Hsing University, Taichung 402202, Taiwan.
Funding
The study was supported by research project TCVGH-1125401B from Taichung Veterans General Hospital (to S.Y.H.). The study sponsor was not involved in study design; data collection, analysis, and interpretation; or writing or submitting this manuscript.
Disclosures
The authors declare that they have no disclosures relevant to the subject matter of this article.
Data Availability
All datasets generated during and analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.
References
- 1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394‐424. [DOI] [PubMed] [Google Scholar]
- 2. Tsai M-C, Hsieh C-T, Hsu H-Y, et al. Association between thyroid cancer and cardiovascular disease risk: a nationwide observation study. Sci Rep. 2022;12(1):18438. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Czene K, Lichtenstein P, Hemminki K. Environmental and heritable causes of cancer among 9.6 million individuals in the Swedish Family-Cancer Database. Int J Cancer. 2002;99(2):260‐266. [DOI] [PubMed] [Google Scholar]
- 4. Hwangbo Y, Park YJ. Genome-wide association studies of autoimmune thyroid diseases, thyroid function, and thyroid cancer. Endocrinol Metab (Seoul). 2018;33(2):175‐184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Northcott P, Nakahara Y, Wu X, et al. Multiple recurrent genetic events converge on control of histone lysine methylation in medulloblastoma. Nat Genet. 2009;41(4):465––4472.. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gudmundsson J, Sulem P, Gudbjartsson DF, et al. Discovery of common variants associated with low TSH levels and thyroid cancer risk. Nat Genet. 2012;44(3):319‐322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Köhler A, Chen B, Gemignani F, et al. Genome-wide association study on differentiated thyroid cancer. J Clin Endocrinol Metab. 2013;98(10):E1674‐E1681. [DOI] [PubMed] [Google Scholar]
- 8. Gudmundsson J, Thorleifsson G, Sigurdsson JK, et al. A genome-wide association study yields five novel thyroid cancer risk loci. Nat Commun. 2017;8:14517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Torkamani A, Wineinger NE, Topol EJ. The personal and clinical utility of polygenic risk scores. Nat Rev Genet. 2018;19(9):581‐590. [DOI] [PubMed] [Google Scholar]
- 10. Liyanarachchi S, Gudmundsson J, Ferkingstad E, et al. Assessing thyroid cancer risk using polygenic risk scores. Proc Natl Acad Sci U S A. 2020;117(11):5997‐6002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Graff RE, Cavazos TB, Thai K, et al. Cross-cancer evaluation of polygenic risk scores for 16 cancer types in two large cohorts. Nat Commun. 2021;12(1):970. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Kachuri L, Graff RE, Smith-Byrne K, et al. Pan-cancer analysis demonstrates that integrating polygenic risk scores with modifiable risk factors improves risk prediction. Nat Commun. 2020;11(1):6084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Fan C-T, Lin J-C, Lee C-H. Taiwan biobank: a project aiming to aid Taiwan's transition into a biomedical island. Pharmacogenomics. 2008;9(2):235‐246. [DOI] [PubMed] [Google Scholar]
- 14. Wei C-Y, Yang J-H, Yeh E-C, et al. Genetic profiles of 103,106 individuals in the Taiwan biobank provide insights into the health and history of Han Chinese. NPJ Genom Med. 2021;6(1):10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Marees AT, de Kluiver H, Stringer S, et al. A tutorial on conducting genome-wide association studies: quality control and statistical analysis. Int J Methods Psychiatr Res. 2018;27(2):e1608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Haugen BR, Alexander EK, Bible KC, et al. 2015 American Thyroid Association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid. 2016;26(1):1‐133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Hao R, Han P, Zhang L, et al. Genetic polymorphisms in the PCNXL2 gene are risk factors for thyroid cancer in the Chinese population. Future Oncol. 2021;17(34):4677‐4686. [DOI] [PubMed] [Google Scholar]
- 18. Zhu Z, Lee PH, Chaffin MD, et al. A genome-wide cross-trait analysis from UK Biobank highlights the shared genetic architecture of asthma and allergic diseases. Nat Genet. 2018;50(6):857‐864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. van der Harst P, Verweij N. Identification of 64 novel genetic loci provides an expanded view on the genetic architecture of coronary artery disease. Circ Res. 2018;122(3):433‐443. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All datasets generated during and analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.