Efficient estimation of nonparametric genetic risk function with censored data

Wang, Yuanjia; Liang, Baosheng; Tong, Xingwei; Marder, Karen; Bressman, Susan; Orr-Urtreger, Avi; Giladi, Nir; Zeng, Donglin

doi:10.1093/biomet/asv030

Cited by 7 publications

(24 citation statements)

References 33 publications

(52 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most studies have Figure 1 Age-specific cumulative risk of LRRK2 G2019S Parkinson disease in first-degree relatives Estimated age-specific risk of PD in LRRK2 G2019S carriers (solid red line) and noncarriers (solid black line) and their confidence intervals (dashed lines). Estimation obtained from 2,266 first-degree relatives of 473 probands using kin-cohort methods 15 under a Cox proportional hazards model. The hazard ratio of LRRK2 G2019S mutation was estimated to be 2.89.…”

Section: Resultsmentioning

confidence: 99%

“…We estimated penetrance to age 80 years using the kin-cohort method. 14,15 The kin-cohort method uses the LRRK2 G2019S mutation status in the probands to infer unobserved genotypes in relatives and combines this with PD diagnosis and age at onset of PD, current age, or age at death information derived from the FHI in the relatives to estimate penetrance of LRRK2 PD in first-degree relatives. We assumed a 2% prevalence 5 of LRRK2 G2019S mutation in first-degree relatives who had not been genotyped.…”

mentioning

confidence: 99%

“…We assumed a 2% prevalence 5 of LRRK2 G2019S mutation in first-degree relatives who had not been genotyped. We have improved on the traditional kin-cohort methodology 14 in the following ways: (1) when a relative was seen in person, we used the observed LRRK2 genotype and PD status in the analysis to improve precision of the penetrance estimates, 15 (2) we ensured that the cumulative risk function is cumulative and can only increase, (3) we provided hazard ratios (HRs) as a summary measure to compare LRRK2 carrier and noncarrier group using the Cox proportional hazards model, 16 and (4) through a bootstrap resampling method, we treated families as independent, 17 and we did not require first-degree family members within a family to have independent age-at-onset distribution. Thus, we accounted for potential residual familial correlation in addition to carrying an LRRK2 mutation.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Age-specific penetrance ofLRRK2G2019S in the Michael J. Fox Ashkenazi Jewish LRRK2 Consortium

et al. 2015

Self Cite

View full text Add to dashboard Cite

Objective: Estimates of the penetrance of LRRK2 G2019S vary widely (24%-100%), reflective of differences in ascertainment, age, sex, ethnic group, and genetic and environmental modifiers. Methods:The kin-cohort method was used to predict penetrance in 2,270 relatives of 474 Ashkenazi Jewish (AJ) Parkinson disease (PD) probands in the Michael J. Fox LRRK2 AJ Consortium in New York and Tel Aviv, Israel. Patients with PD were genotyped for the LRRK2 G2019S mutation and at least 7 founder GBA mutations. GBA mutation carriers were excluded. A validated family history interview, including age at onset of PD and current age or age at death for each first-degree relative, was administered. Neurologic examination and LRRK2 genotype of relatives were included when available.Results: Risk of PD in relatives predicted to carry an LRRK2 G2019S mutation was 0.26 (95% confidence interval [CI] 0.18-0.36) to age 80 years, and was almost 3-fold higher than in relatives predicted to be noncarriers (hazard ratio [HR] 2.89, 95% CI 1.73-4.55, p , 0.001). The risk among predicted G2019S carrier male relatives (0.22, 95% CI 0.10-0.37) was similar to predicted carrier female relatives (0.29, 95% CI 0.18-0.40; HR male to female: 0.74, 95% CI 0.27-1.63, p 5 0.44). In contrast, predicted noncarrier male relatives had a higher risk (0.15, 95% CI 0.11-0.20) than predicted noncarrier female relatives (0.07, 95% CI 0.04-0.10; HR male to female: 2.40, 95% CI 1.50-4.15, p , 0.001). Conclusion:Penetrance of LRRK2 G2019S in AJ is only 26% and lower than reported in other ethnic groups. Further study of the genetic and environmental risk factors that influence G2019S penetrance is warranted. Neurology ® 2015;85:89-95 GLOSSARY AJ 5 Ashkenazi Jewish; CI 5 confidence interval; FHI 5 family history interview; GBA 5 glucocerebrosidase; HR 5 hazard ratio; PD 5 Parkinson disease.

show abstract

Section: Resultsmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Age-specific penetrance ofLRRK2G2019S in the Michael J. Fox Ashkenazi Jewish LRRK2 Consortium

et al. 2015

Self Cite

View full text Add to dashboard Cite

show abstract

“…Key information for each relative included demographics such as year of birth, current age or age at death, gender, ethnicity, PD status, age at onset of PD, and genotype if known. The PD proband in each family was excluded from the penetrance estimation to avoid ascertainment bias 12 .…”

Section: Methodsmentioning

confidence: 99%

“…The genotypes of the remaining 385 relatives were unknown. When most of the genotypes in relatives are not observed, the kin-cohort method 9,12 estimates the probability of missing LRRK2 p.G2019S carrier status in the relatives using the mutation status in PD probands and Mendelian inheritance patterns with the prevalence of mutation in relatives. The method incorporates this estimated probability with age at PD diagnosis in the first-degree relatives to estimate the age-specific cumulative risk of PD in p.G2019S carriers and non-carriers, using an expectation-maximization algorithm 13 developed to handle missing relatives’ genotype data.…”

Section: Methodsmentioning

confidence: 99%

Penetrance estimate of LRRK2 p.G2019S mutation in individuals of non‐Ashkenazi Jewish ancestry

et al. 2017

Self Cite

View full text Add to dashboard Cite

Background Penetrance estimates of the LRRK2 p.G2019S mutation for Parkinson’s disease (PD) vary widely (24%–100%). The p.G2019S penetrance in individuals of Ashkenazi Jewish ancestry has been estimated as 25%, adjusted for multiple covariates. It is unknown whether penetrance varies among different ethnic groups. The objective of this study was to estimate the penetrance of p.G2019S in individuals of non-Ashkenazi Jewish ancestry and compare penetrance between Ashkenazi Jews and non-Ashkenazi Jews to age 80. Methods The kin-cohort method was used to estimate penetrance in 474 first-degree relatives of 69 non-Ashkenazi Jewish LRRK2 p.G2019S carrier probands at eight sites from the Michael J. Fox LRRK2 Cohort Consortium. An identical validated family history interview was administered to assess age at onset of PD, current age, or age at death for relatives in different ethnic groups at each site. Neurological examination and LRRK2 genotype of relatives were included, when available. Results Risk of PD in non-Ashkenazi Jewish relatives who carry a LRRK2 p.G2019S mutation was 42.5% (95% CI: 26.3 – 65.8%) to age 80 which is not significantly higher than the previously estimated 25% (95% CI: 16.7 – 34.2%) in Ashkenazi Jewish carrier relatives. The penetrance of PD to age 80 in LRRK2 p.G2019S mutation carrier relatives was significantly higher than the non-carrier relatives, as seen in Ashkenazi Jewish relatives. Conclusions The similar penetrance of LRRK2 p.G2019S estimated in Ashkenazi Jewish carriers and non-Ashkenazi Jewish carriers confirms that p.G2019S penetrance is 25–42.5% at age 80 in all populations analyzed.

show abstract

Estimation of genetic risk function with covariates in the presence of missing genotypes

et al. 2017

Self Cite

View full text Add to dashboard Cite

In genetic epidemiological studies, family history data are collected on relatives of study participants and used to estimate the age-specific risk of disease for individuals who carry a causal mutation. However, a family member’s genotype data may not be collected due to the high cost of in-person interview to obtain blood sample or death of a relative. Previously, efficient nonparametric genotype-specific risk estimation in censored mixture data has been proposed without considering covariates. With multiple predictive risk factors available, risk estimation requires a multivariate model to account for additional covariates that may affect disease risk simultaneously. Therefore, it is important to consider the role of covariates in the genotype-specific distribution estimation using family history data. We propose an estimation method that permits more precise risk prediction by controlling for individual characteristics and incorporating interaction effects with missing genotypes in relatives, and thus gene-gene interactions and gene-environment interactions can be handled within the framework of a single model. We examine performance of the proposed methods by simulations and apply them to estimate the age-specific cumulative risk of Parkinson’s disease (PD) in carriers of LRRK2 G2019S mutation using first-degree relatives who are at genetic risk for PD. The utility of estimated carrier risk is demonstrated through designing a future clinical trial under various assumptions. Such sample size estimation is seen in the Huntington’s disease literature using the length of abnormal expansion of a CAG repeat in the HTT gene, but is less common in the PD literature.

show abstract

Efficient estimation of nonparametric genetic risk function with censored data

Cited by 7 publications

References 33 publications

Age-specific penetrance ofLRRK2G2019S in the Michael J. Fox Ashkenazi Jewish LRRK2 Consortium

Age-specific penetrance ofLRRK2G2019S in the Michael J. Fox Ashkenazi Jewish LRRK2 Consortium

Penetrance estimate of LRRK2 p.G2019S mutation in individuals of non‐Ashkenazi Jewish ancestry

Estimation of genetic risk function with covariates in the presence of missing genotypes

Contact Info

Product

Resources

About