Pathogenic variants of MODY-genes in adult patients with early-onset type 2 diabetes

Introduction

Type 2 diabetes is a complex disease typically occurring in middle-aged and older adults, which is diagnosed when other forms of diabetes can be reasonably excluded. However, due to the marked clinical heterogeneity, the differential diagnosis of diabetes in adulthood can be challenging. This is especially the case for early-onset type 2 diabetes, which cannot be easily distinguished from monogenic forms of hyperglycemia also occurring at a young age.

It has been recently reported that 2.8% of youth (aged <20 years) with presumed type 2 diabetes carry pathogenic or likely pathogenic (P/LP) variants in genes for maturity-onset diabetes of the young (MODY) [1]. This observation has crucial implications for clinical practice, as these presumed type 2 diabetes patients would benefit from the specific follow-up and treatment strategies that have been developed for MODY [2].

In light of these findings, we investigated whether and to which extent adult patients (>18 years) with early-onset (≤35 years of age) type 2 diabetes also carry P/LP variants in MODY genes.

Methods

We accessed data from a total of 9,493 Italian patients with type 2 diabetes, 626 of whom (6.6%) had an early-onset. All subjects were self-reported whites and were recruited in outpatient diabetic clinics at different research/academic hospitals from Central-Southern Italy, including the Scientific Institute Casa Sollievo della Sofferenza, San Giovanni Rotondo (Apulia), the University Hospital in Foggia (Apulia), the Magna-Graecia University Hospital in Catanzaro (Calabria), the University Hospital in Pisa (Tuscany), the Sapienza University Hospital and several other Hospitals in Rome (Lazio). Diabetes was defined according to American Diabetes Association criteria [3]. Hypertension was defined as a systolic blood pressure >140 mmHg or diastolic blood pressure >90 mmHg, or both, or the presence of antihypertensive treatment. The age at diagnosis of diabetes was based on self-reported information. Family history for type 2 diabetes, referred only to first degree relatives, was self reported.

Of the 626 patients with an early diagnosis, 300 (age=54.4±11.6 years, age at diagnosis 30.5 ±4.4 years, 126 females, BMI=30.7± 6.3 Kg/m², HbA1c=8.2%±1.7), who consented to genetic testing and whose DNA was suitable for NGS, were investigated by means of a targeted panel including 27 monogenic diabetes genes as previously described [4]. We focused the analysis on the same 11 established MODY-genes (HNF1A, GCK, HNF4A, PDX1, HNF1B, NEUROD1, CEL, INS, KCNJ11, ABCC8 and APPL1) that were considered in the previous study in youth with presumed type 2 diabetes, following the same procedures for variants’ selection and classification [1]. Briefly, all rare (i.e., having a minor allele frequency <1% in the Genome Aggregation Database, gnomAD, http://gnomad.broadinstitute.org) coding or splicing high-quality variants (i.e. PASS filter variants with genotype score equal to 99, depth of coverage >30X and mapping quality read values >30) in MODY genes were classified according to the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) guidelines for variant interpretation (https://www.acmg.net/docs/Standards_Guidelines_for_the_Interpretation_of_Sequence_Variants.pdf). All the retrieved variants were confirmed by Sanger Sequencing.

Results

We found that 9 patients (3%) carried eight P/LP variants in MODY genes, broken down as follows: HNF1A (NM_000545.6, c.599G>A; p.R200Q and c.864delG; p.P291QfsTer51, the latter in two patients), GCK (NM_000162.5, c.218A>G; p.D73G and c.760A>C; p.N254H), HNF4A (NM_175914.4, c.334C>T; p.R112W), KCNJ11 (NM_000525.3, c.148C>T; p.R50W), PDX1 (NM_000209.4, c.492G>C; p.E164D), APPL1 (NM_012096.3, c.1270C>T; p.R424X). Details on variants classification are available in Supplementary Table 1. Seven of the nine variants (78%) were in actionable genes (two in GCK, two in HNF1A, one in HNF4A, and one in KCNJ11), possibly allowing a personalized modification of follow-up and/or treatment [2] which, however, did not yet take place in most of our patients.

Clinical features of individuals carrying variants classified as P and LP, were compared to those of non-carriers (Table 1). Similar to what was observed in youth [1], adult patients carrying P/LP variants in MODY genes had a lower rate of hypertension and higher HDL-cholesterol levels than non-carriers (Table 1). They also had lower BMI (Table 1). Altogether, these data suggest less severe insulin resistance in carriers of MODY P/LP variants as compared to non-carriers, which is consistent with the role of MODY genes on insulin secretion rather than insulin sensitivity [5]. Finally, no differences in age at onset or first-degree family history of diabetes were observed between carriers and non-carriers (Table 1).

Table 1.

Baseline clinical characteristics of patients carrying or not P/LP MODY variants

	Non-Carriers (n=291)	Carriers (n=9)	p-Value
Female (%)	41.2	66.7	0.24
Age at recruitment (yrs)	54.4 ± 11.7	55.0 ± 6.4	0.88
Age at diagnosis (yrs)	30.5 ± 4.4	30.2 ± 4.0	0.84
Disease duration (yrs)	24.0 ± 12.3	25.2 ± 6.3	0.76
Diabetes family history (%)	72.7	85.7	0.74
BMI (kg/m 2 )	30.9 ± 6.2	25.1 ± 4.2	<0.01
HbA1c (%); (mmol/mol)	8.2 ± 1.7; 66.0 ± 18.6	8.6 ± 2.0; 70.0 ± 21.9	0.48
SBP (mmHg)	132.3 ± 15.1	129.3 ± 16.9	0.66
DBP (mmHg)	78.4 ± 9.5)	76.3 ± 9.3)	0.52
Total cholesterol (mg/dl)	176.8 ± 42.8)	207.2 ± 49.9)	0.06
HDL cholesterol (mg/dl)	46.3 ± 14.1	58.9 ± 11.4	<0.01
Triglycerides (mg/dl)	149.0 ± 110.0	113.8 ± 92.0	0.09
Insulin treatment (%)	57.0	44.4	0.51
Lipid-lowering treatment (%)	55.0	55.6	1
Hypertension (%)	73.5	33.3	0.02

Continuous variables were reported as mean ± SD whereas categorical variables as total frequencies and percentages.

Two-sided p values refer to unpaired t-tests or Mann-Whitney for continuous variables and to Pearson chi-squared or Fisher Exact test for categorical ones.

P/LP: Pathogenic/Likely Pathogenic; BMI: body mass index; HbA_1c: Glycated haemoglobin; SBP: systolic blood pressure; DBP: diastolic blood pressure.

Discussion

In this study we found that 3% (9/300) of Italian patients with early-onset type 2 diabetes carry P/LP variants in MODY genes. Whether these individuals can be defined as MODY patients, regardless of features traditionally used for the clinical diagnosis of MODY such as family history of autosomal dominant diabetes, early-life diagnosis, and lack of obesity or other signs of insulin-resistance, remains a matter of debate [6]. Overall, our data, along with those reported by other groups, clearly indicate need for better training at continuing medical education courses in the scope of monogenic diabetes, so as to improve the skills of clinical diabetologist, especially those who take care of adult patients.

However, with this caveat in mind, it is striking that the proportion of patients with clinically diagnosed type 2 diabetes carrying P/LP variants in MODY-genes that we observed among adults is almost identical to that observed in youth (3% vs. 2.8%) [1]. This means that this phenomenon is consistent across the first several decades of life rather than being limited to the pediatric population. Of note, given the caution suggested by the ACMG/AMP guidelines in classifying novel or recently discovered variants as P/LP, one cannot exclude that some of the variants currently classified as of uncertain significance (VUS, n=10) will be upgraded to P/LP in the future, thereby increasing the proportion of patients with early-onset type 2 diabetes carrying such variants.

It is also notable that the differences in clinical characteristics between P/LP variant carriers and non-carriers closely resembled those described in youth [1], even if, the low number of carriers individuals in this study did not allow robust cut-offs suggesting when genetic analysis should be requested. Overall, these findings have important practical implications, since in both studies approximately 80% of carriers of MODY-gene P/LP variants were likely to benefit from specific follow-up and/or treatment strategies. In addition, the discovery of specific genetic defects underlying diabetes in young adult patients with presumed type 2 diabetes makes it possible to provide genetic testing and counseling to their relatives as well.

Among the limitations of our study is the relatively small sample size and the lack of information on vertical multigenerational diabetes transmission. Therefore, larger and more informative studies are needed to generate more precise estimates of the prevalence of P/LP variants in MODY genes among adult patients with presumed early-onset type 2 diabetes.

In conclusion, developing tools for the identification of patients with presumed type 2 diabetes who would benefit from MODY genetic testing is timely not only for youth [1] but also for patients who develop diabetes in early adulthood. In addition, a better knowledge of monogenic diabetes might certainly improve adults diabetes services and also decrease the percentage of possible misdiagnoses.

Abbreviations

ACMG/AMP

the American College of Medical Genetics and Genomics and the Association for Molecular Pathology

MODY

Maturity-onset diabetes of the young

NGS

Next generation Sequencing

VUS

Variants of uncertain significance

Data availability.

The datasets generated and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.