Identifying Glucokinase Monogenic Diabetes in a Multiethnic Gestational Diabetes Cohort: New Pregnancy Screening Criteria and Utility of HbA1c
Victoria L. Rudland,1,2,3 Marcus Hinchcliffe,1,4 Jason Pinner,1,4 Stuart Cole,4 Belinda Mercorella,4 Lynda Molyneaux,2,3 Maria Constantino,1,2,3 Dennis K. Yue,1,2,3 Glynis P. Ross,1,2,3 and Jencia Wong1,2,3
Glucokinase monogenic diabetes (GCK–maturity-onset diabetes of the young [GCK-MODY]) should be differentiated from gestational diabetes mellitus (GDM) because management differs. New pregnancy-speciﬁc screening criteria (NSC) have been proposed to identify women who warrant GCK genetic testing. We tested NSC and HbA1c in a multiethnic GDM cohort and examined projected referrals for GCK testing. RESEARCH DESIGN AND METHODS
Using a GDM database, 63 of 776 women had a postpartum oral glucose tolerance test suggestive of GCK-MODY. Of these 63 women, 31 agreed to undergo GCK testing. NSC accuracy and HbA1c were examined. Projected referrals were calculated by applying the NSC to a larger GDM database (n = 4,415). Four of 31 women were conﬁrmed as having GCK-MODY (prevalence ∼0.5–1/100 with GDM). The NSC identiﬁed all Anglo-Celtic women but did not identify one Indian woman. NSC will refer 6.1% GDM cases for GCK testing, with more Asian/ Indian women referred despite lower disease prevalence. Antepartum HbA1c was not higher in those with GCK-MODY. 1
The NSC performed well in Anglo-Celtic women. Ethnic-speciﬁc criteria should be explored. Glucokinase monogenic diabetes (GCK–maturity-onset diabetes of the young [MODY]) results from heterozygous mutations in the glucokinase (GCK) gene. Fasting hyperglycemia, typically 5.5–8.0 mmol/L (1), is present from birth but often is subclinical and may ﬁrst be detected during routine screening for gestational diabetes mellitus (GDM). It is important to differentiate GCK-MODY from standard GDM because the management of GDM, in particular intensive glycemic control, may adversely affect the fetus of a pregnant woman with GCK-MODY (2,3). In contrast to GDM, which is associated with a greatly increased risk for subsequent diabetes, women with GCK-MODY, and their affected offspring, have a low prevalence of diabetes complications and do not require treatment outside of pregnancy (4–6).
Discipline of Medicine, The University of Sydney, Sydney, Australia 2 Diabetes Centre, Royal Prince Alfred Hospital, Sydney, Australia 3 Department of Endocrinology, Royal Prince Alfred Hospital, Sydney, Australia 4 Department of Medical Genomics, Royal Prince Alfred Hospital, Sydney, Australia Corresponding author: Victoria L. Rudland, [email protected]
Received 12 May 2015 and accepted 4 June 2015. G.P.R. and J.W. contributed equally to this work. © 2015 by the American Diabetes Association. Readers may use this article as long as the work is properly cited, the use is educational and not for proﬁt, and the work is not altered.
Diabetes Care Publish Ahead of Print, published online June 24, 2015
NOVEL COMMUNICATIONS IN DIABETES
Identifying GCK-MODY in Multiethnic GDM Cohort
Performing universal GCK genetic testing in women with GDM is not currently practicable. Standard screening criteria (SSC) to preselect cases that should undergo GCK testing are wellestablished for the nonpregnant population (fasting glucose 5.5–8.0 mmol/L and an increment of ,4.6 mmol/L between fasting and 2-h glucose on an oral glucose tolerance test [OGTT]). The sensitivity of SSC for detecting GCK-MODY outside of pregnancy is .98% (7). New pregnancy-speciﬁc screening criteria (NSC) (fasting glucose $5.5 mmol/L on antepartum OGTT and prepregnancy BMI ,25 kg/m2) have been proposed (4). NSC were derived from a predominantly Anglo-Celtic population (4). Their applicability to other ethnicities is unknown. In addition, HbA1c reference ranges for nonpregnant women with GCK-MODY were recently developed (8). Antepartum HbA1c, a measure of glycemic exposure over time, may differentiate GCK-MODY from standard GDM. The speciﬁc aims of our study of a multiethnic GDM cohort were to 1) estimate the prevalence of GCK-MODY; 2) examine the performance of the NSC; 3) calculate the projected referrals for GCK testing, by ethnicity, using NSC; and 4) examine the utility of HbA1c in differentiating GCK-MODY from standard GDM.
reports of GCK mutations (10) and prediction software. GCK mutations were cross-validated by the Molecular Genetics Laboratory at the Royal Devon and Exeter National Health Service Foundation Trust. Multiplex ligation-dependent probe ampliﬁcation was performed using SALSA MLPA P241-D2 MODY mix-1 probemix and EK1 reagent kit (MRC-Holland, Amsterdam, the Netherlands) to identify partial or whole GCK deletions not identiﬁed through sequencing alone. Data were analyzed using Gene Mapper Software version 5.0 (Applied Biosystems). Statistical Analysis
Statistical analyses were performed with NCSS software (NCSS LLC, Kaysville, UT). Ethnic group comparisons were analyzed using the x2 test. Continuous, nonpaired data were analyzed using a two-sample t test or the Mann-Whitney U test. A P value ,0.05 was considered statistically signiﬁcant. NSC were tested using 1) conﬁrmed GCK-MODY cases to determine their applicability; 2) the GDM cohort (n = 776) to calculate the sensitivity and speciﬁcity of the NSC in differentiating GCKMODY from standard GDM; and 3) a larger GDM database (n = 4,415) to calculate projected referrals for GCK testing. RESULTS
RESEARCH DESIGN AND METHODS
Prevalence of GCK-MODY
Of 776 cases, 4 were conﬁrmed GCKMODY, giving a minimum prevalence of 0.5 in 100 cases of GDM. Four of 31 high-risk women (12.9%) had GCKMODY. Assuming a similar proportion among the remaining high-risk women who were unavailable for testing, the estimated prevalence of GCK-MODY was 1 in 100 cases of GDM. Ethnicspeciﬁc prevalence rates were ;1.4–2.7 per 100 (3–6 per 220) for Anglo-Celtic women and ;1–1.9 per 100 (1–2 per 103) for Indian women. The study group was ethnically diverse: 39% were AngloCeltic, 26% Southeast Asian, and 16% Indian.
GDM diagnostic practices at the Royal Prince Alfred Hospital (RPAH) have previously been described (9). Of 776 women in the RPAH GDM database, 63 had a postpartum OGTT (2008–2012) highly suggestive of GCK-MODY on the basis of SSC. Of those 63 women, 31 agreed to undergo genetic testing. The study was approved by the Sydney Local Health District Ethics Review Committee. Genetic Testing
GCK promoter, coding regions, and intron–exon boundaries of exons 1a–10 were sequenced using BigDye Terminator version 3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA) and analyzed on a 3730xl DNA Analyzer (Applied Biosystems) using Sequencing Analysis Software version 5.2 (Applied Biosystems) and Mutation Surveyor version 4.0.6 (Soft Genetics, State College, PA). The pathogenicity of GCK variants was determined by reviewing published
Performance of NSC
When applied to conﬁrmed GCK-MODY cases, the NSC identiﬁed all Anglo-Celtic women. One Indian woman, however, had a higher BMI and would not have been detected. Prepregnancy BMI was not different between GCK-MODY and GDM cases.
When NSC were applied to the GDM cohort, 33 of 776 women proceeded to genetic testing to identify ;3–6 patients with GCK-MODY. Using the minimum prevalence, the sensitivity of the NSC was 75% and the speciﬁcity was 96.1%. Projected Referrals for GCK Genetic Testing Using the NSC
When the NSC were applied to the larger GDM database, 6.1% of patients would be eligible for genetic testing, reduced from 14.2% using the SSC (Table 1). Only 3.2% of Anglo-Celtic women would be tested, reduced from 15.3%. At least twice as many Southeast Asian (8.0%) and Indian (7.5%) women would be referred. Utility of HbA1c
Antepartum HbA1c was not signiﬁcantly different between the groups with GCKMODY (5.6 6 0.4% [38 6 4 mmol/mol]) and GDM (5.3 6 0.4% [34 6 4 mmol/mol]; P = 0.9). Only 2 of 4 women with GCKMODY had an antepartum HbA1c $5.6% (38 mmol/mol), the lower limit of the nonpregnant GCK-MODY reference range (8). CONCLUSIONS
This study demonstrated that, in this multiethnic GDM cohort, 1) the prevalence of GCK-MODY was ;0.5–1 in 100; 2) the NSC would have detected all Anglo-Celtic women with GCK-MODY but not the single Indian woman with GCK-MODY; 3) the application of the NSC will halve the need for genetic testing, with a preferential reduction for referred Anglo-Celtic women; and 4) antepartum HbA1c has limited value in differentiating GCK-MODY from standard GDM. The estimated prevalence of GCKMODY in GDM, and the sensitivity and speciﬁcity of the NSC, are consistent with those in previous studies (4,11). In this study, one Indian woman with GCK-MODY did not satisfy the BMI criterion and would not have been detected. Unlike the study by Chakera et al.(4), we did not ﬁnd a difference in prepregnancy BMI between the GCK-MODY and GDM groups. While BMI has historically helped differentiate MODY from other types of diabetes, it may become less useful as background rates of obesity increase among all ethnicities (12). Our study assessed the projected impact of the NSC on various ethnic groups using an ethnically diverse database. We
Rudland and Associates
Table 1—Projected referrals for GCK genetic testing in a multiethnic GDM cohort using the NSC compared with the SSC applied in pregnancy Projected referrals for GCK testing Projected referrals for GCK testing Reduction in referrals using NSC using SSC, n (%) using NSC, n (%) compared with SSC, % GDM cohort (n = 4,415)
Ethnic breakdown Anglo-Celtic (n = 960)
European (n = 449) Southeast Asian (n = 1,796)
71 (15.8) 185 (10.3)*
16 (3.6) 143 (8.0)*
Indian (n = 453)
Arabic (n = 234)
Aboriginal (n = 52)
Islander (n = 126)
Other (n = 345)
*P , 0.003 for the difference between these ethnic groups and Anglo-Celtic women.
demonstrated that the NSC would reduce referrals of Anglo-Celtic women for GCK testing by ;80% but would refer substantially more Southeast Asian and Indian women despite lower disease prevalence. This is, to our knowledge, the ﬁrst study to investigate antepartum HbA1c in GCK-MODY. No difference between the GCK-MODY and GDM groups was demonstrated. Given that only half of the GCK-MODY group had an antepartum HbA1c within the nonpregnant GCKMODY reference range, this reference range does not seem to discriminate adequately between GCK-MODY and standard GDM during pregnancy. A major strength of our study is its unique design. A difﬁculty in studying GCK-MODY is that, because of its low population prevalence, a prospective study would take several years to identify a sufﬁcient number of women with GCK-MODY (13). Our study design maximized the yield of GCK-MODY by preselecting from a GDM database women whose postpartum OGTT was highly suggestive of GCK-MODY based on well-substantiated SSC. Our detection rate (12.9%) is consistent with another multiethnic GDM cohort (14) and validates our approach. Our study has important clinical implications. One of the main issues with MODY is that .80% of cases are undiagnosed, in part because of the ﬁnancial cost of genetic testing (13). The targeted application of the NSC is predicted to halve the number of GDM cases tested for GCK-MODY, reducing the economic burden of GCK testing. Our data support adoption of the NSC for use in Anglo-
Celtic women with GDM. Ethnic-speciﬁc screening criteria should be explored. Acknowledgments. The authors gratefully
acknowledge all of the diabetes educators who have collected data and maintained the RPAH GDM database. Funding. V.L.R. was supported by a University of Sydney Postgraduate Award and a National Health and Medical Research Council Scholarship. Duality of Interest. No conﬂicts of interest relevant to this article were reported. Author Contributions. V.L.R. wrote the manuscript, conceived of and designed the study, performed recruitment and genetic testing, acquired and analyzed data, and interpreted the results. M.H. reviewed and edited the manuscript, conceived of and designed the study, assisted with genetic testing, acquired and analyzed data, and interpreted the results. J.P. and D.K.Y. reviewed and edited the manuscript, conceived of and designed the study, and interpreted the results. S.C. and B.M. reviewed the manuscript and assisted with genetic testing. L.M. and M.C. reviewed the manuscript and analyzed data. G.P.R. and J.W. reviewed and edited the manuscript, conceived of and designed the study, assisted with recruitment and genetic testing, acquired and analyzed data, and interpreted the results. V.L.R., G.P.R., and J.W. are the guarantors of this work and, as such, had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
References 1. Byrne MM, Sturis J, Cle´ ment K, et al. Insulin secretory abnormalities in subjects with hyperglycemia due to glucokinase mutations. J Clin Invest 1994;93:1120–1130 2. Chakera AJ, Carleton VL, Ellard S, et al. Antenatal diagnosis of fetal genotype determines if maternal hyperglycemia due to a glucokinase mutation requires treatment. Diabetes Care 2012;35:1832–1834 3. Spyer G, Macleod KM, Shepherd M, Ellard S, Hattersley AT. Pregnancy outcome in patients with raised blood glucose due to a heterozygous glucokinase gene mutation. Diabet Med 2009;26:14–18
4. Chakera AJ, Spyer G, Vincent N, Ellard S, Hattersley AT, Dunne FP. The 0.1% of the population with glucokinase monogenic diabetes can be recognized by clinical characteristics in pregnancy: the Atlantic Diabetes in Pregnancy cohort. Diabetes Care 2014;37:1230–1236 5. Stride A, Shields B, Gill-Carey O, et al. Crosssectional and longitudinal studies suggest pharmacological treatment used in patients with glucokinase mutations does not alter glycaemia. Diabetologia 2014;57:54–56 6. Steele AM, Shields BM, Wensley KJ, Colclough K, Ellard S, Hattersley AT. Prevalence of vascular complications among patients with glucokinase mutations and prolonged, mild hyperglycemia. JAMA 2014;311:279–286 7. Ellard S, Bellann´e-Chantelot C, Hattersley AT; European Molecular Genetics Quality Network (EMQN) MODY group. Best practice guidelines for the molecular genetic diagnosis of maturity-onset diabetes of the young. Diabetologia 2008;51:546–553 8. Steele AM, Wensley KJ, Ellard S, et al. Use of HbA1c in the identiﬁcation of patients with hyperglycaemia caused by a glucokinase mutation: observational case control studies. PLoS One 2013;8:e65326 9. Rudland VL, Pech C, Harding AJ, et al. Zinc transporter 8 autoantibodies: what is their clinical relevance in gestational diabetes? Diabet Med 2015;32:359–366 10. Osbak KK, Colclough K, Saint-Martin C, et al. Update on mutations in glucokinase (GCK), which cause maturity-onset diabetes of the young, permanent neonatal diabetes, and hyperinsulinemic hypoglycemia. Hum Mutat 2009;30:1512–1526 11. Allan CJ, Argyropoulos G, Bowker M, et al. Gestational diabetes mellitus and gene mutations which affect insulin secretion. Diabetes Res Clin Pract 1997;36:135–141 12. Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999–2008. JAMA 2010;303:235–241 13. Shields BM, Hicks S, Shepherd MH, Colclough K, Hattersley AT, Ellard S. Maturity-onset diabetes of the young (MODY): how many cases are we missing? Diabetologia 2010;53:2504–2508 14. Kousta E, Ellard S, Allen LI, et al. Glucokinase mutations in a phenotypically selected multiethnic group of women with a history of gestational diabetes. Diabet Med 2001;18:683–684