Interest in Genetic Testing in Ashkenazi Jewish Parkinson’s Disease Patients and Their Unaffected Relatives

Abstract

Our objective was to explore interest in genetic testing among Ashkenazi Jewish (AJ) Parkinson’s Disease (PD) cases and first-degree relatives, as genetic testing for LRRK2 G2019S is widely available. Approximately 18 % of AJ PD cases carry G2019S mutations; penetrance estimations vary between 24 and 100 % by age 80. A Genetic Attitude Questionnaire (GAQ) was administered at two New York sites to PD families unaware of LRRK2 G2019S mutation status. The association of G2019S, age, education, gender and family history of PD with desire for genetic testing (outcome) was modeled using logistic regression. One-hundred eleven PD cases and 77 relatives completed the GAQ. Both PD cases and relatives had excellent PD-specific genetic knowledge. Among PD, 32.6 % “definitely” and 41.1 % “probably” wanted testing, if offered “now.” Among relatives, 23.6 % “definitely” and 36.1 % “probably” wanted testing “now.” Desire for testing in relatives increased incrementally based on hypothetical risk of PD. The most important reasons for testing in probands and relatives were: if it influenced medication response, identifying no mutation, and early prevention and treatment. In logistic regression, older age was associated with less desire for testing in probands OR=0.921 95%CI 0.868–0.977, p=0.009. Both probands and relatives express interest in genetic testing, despite no link to current treatment or prevention.

Introduction

Genetic susceptibility plays a role in the development of Parkinson’s disease (PD), however, how, and to what extent remain unclear. The most common mutation associated with PD in Caucasian populations is the G2019S mutation in the leucine-rich repeat kinase-2 (LRRK2) gene (Kachergus et al. 2005). Penetrance is variable, ranging from 24 to 100 % by age 80 (Goldwurm et al. 2011; Trinh et al. 2014) depending on ethnic group and methodology employed.

The goal of this study was to establish baseline knowledge regarding genetic testing, genetics of PD specifically, current diagnosis and treatment of PD, attitudes toward genetic testing for PD and reasons to pursue or decline genetic testing in PD cases and their asymptomatic family members participating in the Michael J Fox Foundation LRRK2 Ashkenazi Jewish (AJ) Consortium (Alcalay et al. 2013). We sought the reasons individuals are more, or less, likely to pursue testing and if those reasons corresponded to desire for genetic testing ’now’, given what is currently known about PD and the options for treatment.

The LRRK2 G2019S mutation is particularly common among AJ with PD. It was identified in 14.8 % of all AJ with PD and in 26 % of those with familial PD; resulting in approximately a 9 fold increased chance of developing PD in mutation carriers compared to age and sex matched controls (Giladi et al. 2011; Orr-Urtreger et al. 2007). In the Michael J Fox AJ Consortium, the largest systematically examined to date, 19.9 % of Ashkenazi PD cases had LRRK2 G2019S mutations, after excluding cases with glucocerebrosidase (GBA) mutations (Alcalay et al. 2013). Given that up to 30 % of AJ PD patients have either LRRK2 G2019S or mutations in GBA, genetic testing may be a consideration for AJ PD cases and their family members. As genetic testing becomes more accessible, not only through health care providers, but also through direct-to-consumer testing, it is important to understand what individuals know about the test, why they opt for testing, and what they believe can be concluded from the results. There have been few papers that explore patient’s and family member’s knowledge and attitudes toward genetic testing for PD. Studies have found that knowledge regarding genetics and PD is correlated with both higher level of education and Jewish ancestry (Falcone, Wood, Xie, Siderowf, and Van Deerlin 2011; Sakanaka et al. 2013; Tan and Jankovic 2006). In contrast, desire to obtain genetic testing results was associated with non-Jewish ancestry but not education, gender or family history of PD (Sakanaka et al. 2013). Therefore, the major research questions investigated in the present study were 1.) What is the understanding of what is currently known about PD and genetic testing? 2.) Is there a desire to obtain testing at this time in those already affected with PD or in their unaffected first-degree relatives? 3.) Is there an association between the desire or lack of desire for testing and reasons individuals deem important in the decision to pursue testing? 4.) Do the above questions vary by G2019S genotype?

Methods

Participants

The Michael J Fox Ashkenazi Jewish Consortium was designed to focus on LRRK2 G2019S probands and their first-degree relatives. In the first stage, 553 PD probands were recruited from 3 sites, Beth Israel Medical Center (n=138), and Columbia University Medical Center (n=143) both in New York, NY, and Tel Aviv University, in Tel Aviv, Israel (n=272) from April 2010 to May 2012 (Alcalay et al. 2013). Institutional review boards at all sites approved the protocols. Probands were recruited if they reported at least two AJ grandparents, and were diagnosed with PD by a movement disorder specialist, based on the United Kingdom Parkinson’s Disease brain bank criteria (with the exception that reporting a family history of PD was not exclusionary as stipulated in the criteria) (Hughes, Daniel, Kilford, and Lees 1992). All NY participants were genotyped for the G2019S mutation but were not informed of the results.

An in-depth assessment was offered to probands who carry G2019S mutations and their family members and to a subset of non-carrier probands. The in-depth analysis consisted of detailed neurological, neuropsychological, psychiatric assessments, medical and family history, non-motor symptomatology, and a genetic attitudes questionnaire that we designed and is the focus of this report. Participants from Tel Aviv were not administered the genetic attitudes questionnaire, since results are offered to all individuals who undergo testing, and will not be further discussed. Probands from the NY sites found to carry G2019S mutations (n=83) and a random sample of non-carrier probands (n=69) were recruited for an in-depth analysis. We attempted to recruit first-degree relatives of all G2019S carriers for the in-depth analyses. Eight probands and five relatives knew their carrier status prior to the study. These 5 relatives came from 3 families (2 relatives from 2 families, 1 from a single family). The proband in each family was a known LRRK2 G2019S carrier.

Assessments

All probands completed a questionnaire regarding demographics and family history of PD. Details of this assessment have been reported elsewhere (Alcalay et al. 2013). Specific scores from the in-depth assessment included in these analyses were the Geriatric Depression Scale (GDS 15), Unified Parkinson’s Disease Rating Scale (UPDRS) Part III, and Montreal Cognitive Assessment (MoCA). Family history of PD was assessed using a reliable, valid interview (Marder et al. 2003).

We designed the Genetic Attitudes Questionnaire (GAQ) that was completed by each participant and reviewed by the interviewer for completeness. The GAQ consisted of four sections: A) Parkinson’s disease and your life (10 items), B) attitudes towards genetic testing (25 items), C) work and retirement (6 items), D) privacy and insurance (22 items). A range of question formats were employed including yes/no/ uncertain, 5-item response scale, and a 1–100 scale. For example, the 5-item response scale regarding desire for testing, response options included definitely want testing, probably want testing, definitely do not want testing probably do not want testing, and cannot decide about testing. All questions in the section on genetic knowledge have been previously published (Falcone et al. 2011). Specific questions about privacy and insurance were derived from the Prospective Huntington At Risk Observational Study (PHAROS) survey (Oster et al. 2008), and will be the subject of a future publication. The goal of the GAQ was to establish the degree of baseline knowledge regarding genetic testing, genetics of PD specifically, current diagnosis and treatment of PD, attitudes toward genetic testing for PD and reasons to pursue or decline genetic testing.

Statistical Analyses

Demographics, genotype, UPDRS, MoCA, GDS, and GAQ, were compared using chi-square tests, and Student’s t-tests as appropriate. Logistic regression models were constructed to test the association between demographic and clinical features in probands (predictors) and interest in genetic testing for PD (outcome) dichotomized as one group being definitely or probably desire testing and the second group included definitely or probably do not want testing and unsure about testing. Models were adjusted for G2019S mutation status, age, gender, disease duration, UPDRS III score, and education (categorized as with or without post-high-school education). Similar analyses were conducted in asymptomatic relatives adjusting for age, gender, G2019S status, and relative class (parent, sibling, child).

Results

Demographics

We first compared those that did and did not complete the GAQ. One-hundred eleven probands (111/152 [73 %] of eligible probands) and 77 relatives (77/100 [77 %] of eligible asymptomatic first-degree relatives representing 45 families) completed the GAQ. There were no significant differences between those that did and did not complete GAQ (Appendix Table 1), except that probands who completed the GAQ had a lower GDS 15 score than non-responders (p=0.017), and relatives who completed the questionnaire were more likely to have post-high school education (p=0.013). Data presented from this point forward will only pertain to those completing the GAQ.

Demographic characteristics of G2019S carriers and non-carriers who completed the GAQ are shown in Table 1. Similar to the 553 cases that underwent screening at all three sites (Alcalay et al. 2013), probands who were G2019S carriers had longer disease duration (10.39 years [SD=6.73] compared to 7.92 years [SD 6.45] for non-carriers, p=0.054); a higher percent reporting a family history of PD (39.7 % of carriers vs. 21.2 % non-carriers, p=0.041); and carriers were more likely to believe that PD is hereditary (36.4 % of carriers vs. 12 % non-carriers, p=0.007). Among the relatives, there were no significant differences in demographic characteristics seen between G2019S carriers and non-carriers.

Table 1.

Demographics by Genotype for those completing GAQ

	Probands			Relatives
	G2019S Carriers (n=58)	G2019S Non-carriers (n=53)	Significance	G2019S Carriers (n=45)	G2019S Non-carriers (n=32)	Significance
Percent Male	60.3	73.6	0.162	35.8	47.7	0.301
Age, years (SD)	67.74 (10.05)	68.97 (9.62)	0.551	54.96 (17.32)	51.52 (19.51)	0.36
Percent with post high school education	96.4	88.7	0.157	84.6	86.4	1
Number (Percent) with previous PD genetic testing outside of research	7 (12.1 %)	5 (9.4 %)	0.600	3 (7.7 %)	2 (6.5 %)	0.521
Number with genetic testing for another condition outside of research setting	6	6	0.767	6	8	0.293
Percent positive family history in 1st degree relative	39.7	21.2	0.041	31.6	41.9	0.453
Percent that believe PD is hereditary (Yes/No/Unsure)	36.4/25.5/38.2	12/48/40	0.007	40.5/19/40.5	40/20/40	0.995
Age-at-onset, years (SD)	57.05 (12.20)	60.94 (9.78)	0.070	N/A	N/A	N/A
PD duration, years (SD)	10.39 (6.73)	7.92 (6.45)	0.054	N/A	N/A	N/A
Mean UPDRS motor (SD)	10.32 (7.49)	13.31 (9.13)	0.065	N/A	N/A	N/A
MoCA (SD)	24.96 (3.48)	25.51 (3.58)	0.444	26.86 (2.56)	27.58 (2.24)	0.214

Genetic Knowledge

The majority of probands and relatives are aware that scientists have identified genes associated with a higher risk of developing PD, and that a genetic test will not predict age of onset, PD severity, or if one will develop PD with certainty (Fig. 1). This did not change when stratified by genotype. In terms of knowledge regarding genetic testing for other diseases, among probands, a high percentage are aware that genetic testing exists for breast cancer (78.5 %) and Huntington’s Disease (61.9 %), but fewer individuals are aware of genetic testing for Cystic Fibrosis (51 %) and Gaucher Disease (47.2 %). Relatives had similar knowledge to probands (Fig. 1).

Fig. 1.

Performance of probands and relatives on genetic knowledge questions. The answer choices were either true or false; correct answer is given in parenthesis

Desire for Testing

Hypothetically, 32.6 % of the probands and 23.6 % of the asymptomatic relatives responded that they would definitely want to pursue genetic testing if offered “now,” while 41.1 % and 36.1 %, respectively, said they probably want to pursue testing “now.” On the 5-point scale (definite yes, probably yes, unsure, probably no, definitely no) there was no significant difference in responses by G2019S genotype (p=0.826 probands, p=0.07 relatives).

In univariate analyses, probands who did not want testing or who were unsure, were older (73.8 years; SD=9.44) than probands who definitely or probably wanted testing (69.9 years; SD=9.4), p =0.002. Disease duration was significantly longer in those who did not want testing (12.32; SD=8.39, vs. 8.5; SD=6.14), p=0.046; however UPDRS scores were not different. The percentage of G2019S carriers, or those who reported a family history of PD did not differ. Response did not differ by gender, or marital status (married compared to single).

In a multivariate model, including G2019S, gender, education, age, belief that PD was hereditary, and UPDRS score, only age was associated with definite or probable desire for testing (OR=0.921 95%CI 0.868–0.977, p=0.009), such that older age was associated with less interest in desire for testing now. Seventy asymptomatic relatives who did not know their genetic status were included in an analysis to predict interest in genetic testing. In a multivariate model, adjusting for relationship to the proband, gender, education, age, belief that PD is hereditary and genotype, there were no significant predictors of interest in testing. However, all interviewed asymptomatic parents of PD probands (6 from 4 families) expressed interest in testing. Thirty percent of siblings and 70.6 % of children were interested in testing.

Interest in genetic testing among relatives based on hypothetical risk of PD is shown in Fig. 2. The higher the likelihood that a test predicts the development of PD, the more likely an individual says he/she would want to be tested. There is equipoise at 25 % hypothetical risk, with approximately 25 % in each of the four groups.

Fig. 2.

Percent of asymptomatic relatives likely to choose testing based on hypothetical risk of PD

Reasons for Testing

Figures 3a and b show the relative importance of reasons in deciding to be tested for probands and relatives, respectively. The most important reasons cited by both groups were learning they do not carry a mutation, and being identified early for prevention or treatment. Additionally, 93 % of probands and 93 % of relatives reported they would definitely pursue testing if it determined response to medications. Reproductive decision-making was the least important reason for wanting testing for both probands and relatives.

Fig. 3.

a Probands’ reasons for testing b Relatives’ reasons for testing

Figures 4a and b explore some negative consequences of being tested as related to insurance and privacy. The interview was administered prior to initiation of the Affordable Care Act. A total of 21.8 % of probands were somewhat or very concerned about losing health insurance, compared to 28.6 % of relatives. A total of 19.3 % of probands and 23.5 % of relatives were very or somewhat worried about losing disability insurance.

Fig. 4.

a Proband’s reasons for testing related to insurance and privacy b Relatives reasons for testing related to insurance and privacy

Discussion

Interest in genetic testing in AJ PD families depends on how likely a test would predict the development of PD, medication response, and the relationship to early detection. The majority of probands (73.7 %) and relatives (59.7 %) express interest in genetic testing now, despite the fact that there is no known link to treatment or prevention. We established that in this population there is a basic understanding of PD genetics, the importance of various reasons to consider testing, and that most affected individuals and unaffected first-degree relatives indicate that they would choose to be tested now.

Knowledge

Other studies have reported low scores on genetic knowledge questions in both PD patients and caregivers (Falcone et al. 2011; Sakanaka et al. 2013; Tan and Jankovic 2006). Though we did not calculate a genetic knowledge score, the majority of participants answered general PD genetics questions correctly. Sakanaka et al. reported 51.6 % of PD patients and 55.6 % of non-blood related caregivers answered correctly that “scientists have identified genes associated with a higher risk of developing PD,” with a positive correlation of those with Jewish ancestry responding correctly (Sakanaka et al. 2013, p. 116, Fig. 1). Our findings further support this increased knowledge in the Ashkenazi Jewish population, with 77.5 % of probands and 88.3 % of relatives answering correctly. However, other than for breast cancer, relatively few individuals were aware of genetic testing for other conditions. It is possible participants’ greater knowledge of PD genetics motivated them to participate in a PD genetics study. They may have also received information regarding PD genetics from their clinicians or through participation in research.

The high percentage knowing about genetic testing for breast cancer is expected given a high prevalence of BRCA1/2 mutations in the AJ population (Struewing et al. 1997). Particularly surprising though, is the low percentage aware of genetic testing for Gaucher Disease, associated with a GBA mutation in this population and linked to PD (Sidransky et al. 2009). GBA genetic testing is included in the recommended prenatal testing panel for the AJ population, though it was added several years after testing for Tay Sachs. Knowledge of both BRCA mutations and GBA homozygosity could lead to earlier detection and intervention for breast cancer and Gaucher Disease, respectively, as compared to PD, for which there is no current intervention based on genetic mutation status. Lack of an intervention has been cited as reducing interest in genetic testing in PD (Dahodwala et al. 2007).

As seen in other studies (Falcone et al. 2011; Sakanaka et al. 2013; Tan and Jankovic 2006), there is a desire to pursue genetic testing for PD with 73.7 % of probands answering they would definitely or probably desire testing now, and 59.7 % of relatives in this study responding similarly. However, when questioned how likely individuals were to pursue testing if a positive result meant they had a 25 % risk of developing PD, which is closest to the estimated lifetime penetrance of G2019S (Clark et al. 2006), there was equipoise.

Among PD probands, older individuals were less likely to definitely/probably want to pursue testing “now.” We found no single factor that predicted desire for testing in relatives including LRRK2 genotype, demographic features, or belief that PD is hereditary. This may reflect that each individual has different reasons for choosing to be tested or that individuals think they would want to be tested, but may not actually follow through when given the opportunity, especially following genetic counseling. Giladi et al. (2011) found that out of 46 Ashkenazi first-degree relatives, of PD patients known to be carriers of the G20919S mutation, only five wanted to pursue genetic testing following clinical counseling.

In a previous study (Falcone et al. 2011) the most important themes cited by PD patients to pursue testing were personal information/curiosity (30.6 %), to help research (26.5 %), and information for family (22.4 %). Similarly, we found in probands, personal information (i.e., finding out they do not carry the mutation and participation in research for early identification for treatment) were the most frequently cited reasons for wanting to be tested.

Practice Implications

Genetic counselors advising the Ashkenazi Jewish population regarding testing for G2019S, should be aware that penetrance of the G2019S mutation has been reported from approximately 24–100 %, (Goldwurm et al. 2011), but that penetrance among Ashkenazi Jews may be at the lower end of the spectrum (Clark et al. 2006). It is currently unknown whether the range of penetrance estimates is due to differing methodology, the inclusion of both clinic and community based samples or whether there are genetic modifiers in specific ethnic groups. The reduced penetrance could influence the decision to undergo testing. Currently, mutation status would not result in answers regarding age of onset of PD or treatment options. There are no reports of preimplantation genetic diagnosis being utilized for G2019S. As we learn more about the manifestations, progression and targeted interventions related G2019S carrier status; decisions based on knowledge of G2019S carrier status may be informative.

Study Strengths

The strengths of our study include that we were able to explore an educated population of the same ethnicity with a high prevalence of the G2019S mutation using a comprehensive survey, which highlighted various aspects of perceived knowledge and consequences of genetic testing. We had both symptomatic probands and asymptomatic at-risk family members. All individuals were genotyped, but the majority did not know their genotype. This allowed us to analyze if mutation status alone affected belief of PD being hereditary.

Study Limitations and Research Recommendations

Limitations of our study include generalizability outside of this population. The individuals (probands and relatives) knew that they were participating in a genetic study of PD, though they were not informed of results. Not everyone completed the GAQ. We did not perform test retest reliability on the GAQ, though several questions were derived from other studies. We did not include specific knowledge questions regarding LRRK2. Currently, to our knowledge, there is no information available regarding possible psychological stress following identification of G2019S mutation status for LRRK2. As mutation status was not revealed to individuals in our study, we could not answer this question. The focus of future research could include follow up of individuals who are informed about their mutation status. Our findings should be replicated in a larger sample of family members of PD pro-bands who carry LRRK2 mutations. Given the wide penetrance estimates, comparison of the GAQ in Ashkenazi Jewish families and other ethnic groups who carry LRRK2 mutations would be important. Attitudes toward genetic testing for PD mutations may change, given the health care legislation. In addition, were there a disease modifying therapy for PD or specifically for LRRK2 PD, attitudes and desire for testing might change.

Conclusion

In conclusion, our study explores the knowledge, attitudes, and desires toward the complexities of genetic testing in PD, in a high-risk population, and finds a high desire for genetic testing with no clear predictive factor other than age. As genetic screening for PD becomes more widely available, it will be imperative to ensure proper counseling regarding expectations of consequences of results. Further studies should be pursued in other populations and consider other PD associated genes to provide generalizability.