Dataset Information

Machine Learning Approaches for Predicting Bisphosphonate-Related Osteonecrosis in Women with Osteoporosis Using VEGFA Gene Polymorphisms.

ABSTRACT:

Objective

This nested case-control study aimed to investigate the effects of VEGFA polymorphisms on the development of bisphosphonate-related osteonecrosis of the jaw (BRONJ) in women with osteoporosis.

Methods

Eleven single nucleotide polymorphisms (SNPs) of the VEGFA were assessed in a total of 125 patients. Logistic regression was performed for multivariable analysis. Machine learning algorithms, namely, fivefold cross-validated multivariate logistic regression, elastic net, random forest, and support vector machine, were developed to predict risk factors for BRONJ occurrence. Area under the receiver-operating curve (AUROC) analysis was conducted to assess clinical performance.

Results

The VEGFA rs881858 was significantly associated with BRONJ development. The odds of BRONJ development were 6.45 times (95% CI, 1.69-24.65) higher among carriers of the wild-type rs881858 allele compared with variant homozygote carriers after adjusting for covariates. Additionally, variant homozygote (GG) carriers of rs10434 had higher odds than those with wild-type allele (OR, 3.16). Age ≥ 65 years (OR, 16.05) and bisphosphonate exposure ≥ 36 months (OR, 3.67) were also significant risk factors for BRONJ occurrence. AUROC values were higher than 0.78 for all machine learning methods employed in this study.

Conclusion

Our study showed that the BRONJ occurrence was associated with VEGFA polymorphisms in osteoporotic women.

SUBMITTER: Kim JW

PROVIDER: S-EPMC8230421 | biostudies-literature | 2021 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine Learning Approaches for Predicting Bisphosphonate-Related Osteonecrosis in Women with Osteoporosis Using <i>VEGFA</i> Gene Polymorphisms.

Kim Jin-Woo JW Yee Jeong J Oh Sang-Hyeon SH Kim Sun-Hyun SH Kim Sun-Jong SJ Chung Jee-Eun JE Gwak Hye-Sun HS

Journal of personalized medicine 20210610 6

<h4>Objective</h4>This nested case-control study aimed to investigate the effects of <i>VEGFA</i> polymorphisms on the development of bisphosphonate-related osteonecrosis of the jaw (BRONJ) in women with osteoporosis.<h4>Methods</h4>Eleven single nucleotide polymorphisms (SNPs) of the <i>VEGFA</i> were assessed in a total of 125 patients. Logistic regression was performed for multivariable analysis. Machine learning algorithms, namely, fivefold cross-validated multivariate logistic regression, e ...[more]

PMID: 34200782

Similar Datasets

Project description:We have explored the potential role of genetics in the development of osteonecrosis of the jaw (ONJ) in multiple myeloma (MM) patients under bisphosphonate therapy. A genome wide association study was performed using 500.568 single nucleotide polymorphisms (SNPs) in two series of homogeneously treated MM patients: one with ONJ (22 MM cases) and another without ONJ (65 matched MM controls). Four SNPs (rs1934951, rs1934980, rs1341162 and rs17110453) mapped within the Cytochrome P450-2C gene (CYP2C8) showed a different distribution between cases and controls with statistically significant differences (p=1.07x10-6, p=4.231x10-6, p=6.22x10-6 and p=2.15x10-5, respectively). SNP rs1934951 was significantly associated with a higher risk of ONJ development even after Bonferroni correction (P corrected value=0.02). Genotyping results displayed an overrepresentation of the T allele in cases as compared with controls (48% vs. 12%). Thus, individuals homozygous for the T allele had an increased likelihood of developing ONJ (Odds ratio 12.75, 95% confidence interval 3.7 to 43.5). We studied 22 cases (MM with ONJ) and 65 controls (MM without ONJ), matched for age, gender and ethnicity (note: all of the cases and controls are Caucasian). All patients were enrolled in the GEM-00 protocol, which consists on polychemotherapy and autologous transplantation. All received BPs therapy, either Pamidronate (16 cases, 57 controls) or Zoledronic Acid (6 cases, 8 controls) planned for 2 years (median 22 months, range 9-24 months). Clinical characteristics were similar between controls and cases. Study protocols were approved by the ethics committee and written informed consent was obtained from all participants. Each patient was genotyped using the Affymetrix GeneChip Mapping 500K set of microarrays (Affymetrix, Santa Clara, CA) according to the manufacturer’s recommendations. Genotypes were determined using the BRLMM algorithm with cases and controls undergoing joint cluster analysis, after ensuring a robust association test through quality filtering tests. Based on stringent quality control criteria a total of 339.972 SNPs were selected for subsequent analyses. Criteria for exclusion were: 1) call rate<90%, 2) minor allele frequency <5% and 3) deviations from Hardy-Weinberg equilibrium with a p<0.00001. Sexual chromosomes were excluded for analysis. To test for allelic associations between SNPs and ONJ, we constructed 2x2 contingency tables and compared using the two-sided Fisher’s exact or Chisquare tests through SPSS software (SPSS 14.0, Inc. Chicago, IL, USA). P-values were corrected (Pc) using the Bonferroni correction. The strength of association was estimated by the odds ratio (OR), and their 95% confidence intervals (CI) were calculated by Cornfield methods. Linkage disequilibrium between SNPs was analyzed using the Arlequin Software (http://anthro.unige.ch/arlequin).