Diverse Obesity Trajectories in a Family Including Identical Twins with a Pathogenic MC4R Variant

Abstract

Introduction

Pathogenic heterozygous melanocortin-4 receptor (MC4R) variants are the most common cause of monogenic obesity, affecting central satiety and appetite regulatory areas of the brain.

Case Presentations

We report a pedigree with a pathogenic MC4R variant (c.380C>T, p.Ser127Leu). In the proband with obesity (BMI 35 kg/m²) and severe insulin resistance, use of combination of semaglutide and naltrexone-bupropion was successful in reducing insulin requirements and weight. His adult monozygotic twin daughters both had childhood-onset obesity; however, weight trajectories differed. Twin 1 had a peak BMI of 29.1 kg/m², which decreased to 19.7 kg/m² with intensive exercise and diet control without weight-lowering medication. Twin 2 had a sedentary lifestyle and epilepsy and had a peak BMI of 30.1 kg/m²; she responded well to naltrexone-bupropion and BMI decreased to 26 kg/m².

Conclusion

The manifestation of obesity, even in cases of monogenic obesity, can vary significantly due to the influence of environmental and lifestyle factors.

Introduction

The pathogenesis of obesity is complex and there is an intricate interplay of genetic, environmental and psychosocial factors which mediate food intake and energy expenditure. While obesogenic dietary patterns and a lack of physical activity are recognized as the primary drivers of the global rise in obesity, genetic factors also strongly contributes to the regulation of body weight [1]. Monogenic causes of obesity are generally rare with an estimate of 5% of severe obesity and includes pathogenic variants in genes coding for leptin, leptin receptor, proprotein convertase 1 (PCSK1), pro-opiomelanocortin (POMC), and melanocortin-4 receptor (MC4R) [1]. Among these, MC4R pathogenic variants are the most common cause of monogenic obesity and have been found in up to 6% of patients with severe childhood obesity [2]. The MC4R gene plays a critical role in the regulation of energy homoeostasis and appetite control [3]. This function is facilitated by a complex communication network involving the gastrointestinal tract, adipose tissue, and the hypothalamus [4].

Here, we report a pedigree with a heterozygous pathogenic variant of MC4R; the proband had severe insulin resistance and young-onset obesity. We described his food cravings using the Food Cravings Questionnaire-Trait (FCQ-T-reduced) [5] and how it changed with various anti-obesity medications. We also reported the contrasting BMI trajectories of his monozygotic twin daughters and discuss the influences of environmental factors. All 3 individuals consented to participate in this observational study and for genetic testing using whole genome sequencing. This study was approved by our local Institutional Review Board.

Case Presentations

Proband

The proband had obesity since childhood and admitted to having a big appetite. He was of Chinese ethnicity, the seventh child from a family of nine. His mother and father were obese, but without known history of diabetes, and died at 48 and 59 years old, respectively, from acute myocardial infarction. Six out of eight of his siblings have diabetes. He was diagnosed with type 2 diabetes mellitus at 17 years old via oral glucose tolerance test at a health screening prior to his 2-year mandatory military service. He was initiated on metformin. He was obese at 80 kg then. His weight gradually increased and peaked at 100 kg with a BMI of 35 kg/m² at age 52 years. He was under the care of multiple endocrinologists for decades for his challenging diabetes control (described below). Workup for secondary causes of obesity was negative with normal thyroid function tests, and he did not have Cushing’s syndrome. A 1 mg overnight dexamethasone suppression test showed adequate normal suppression with 8 am morning cortisol of 44 nmol/L. His lifestyle was sedentary, and he did not exercise regularly. He had high intellectual capabilities and was an educator in a university.

Effect of Anti-Obesity Agents on Weight and T2DM Control

A thorough review of history and clinical notes revealed that he was started on insulin therapy at the age of 45. Since then, he required increasing amount of insulin reaching the point of extremely high doses of insulin in a multi-dose regime (insulin glargine and insulin aspart) of up to 10 units/kg/day (approximately 1,000 units/day), which maintained his HbA1c reasonably well at 7%–8.3%. His severe insulin resistance was evident by the extremely high insulin requirement without hypoglycaemic episodes and an elevated HOMA-IR of 16.3 (normal <2) [6]. His care was transferred from the general endocrinology clinic to our endocrine sub-speciality severe obesity/atypical diabetes clinic at the age of 59. To reduce the insulin resistance, various anti-obesity strategies were tried, including advice on mindful eating behaviours to curb snacking and late-night eating, increase in physical activity and avoiding insulin stacking. Various GLP-1 receptor agonists and doses were tried; subcutaneous liraglutide 1.8 mg (+3 kg), followed by subcutaneous liraglutide 3 mg (−4.7 kg), switched to subcutaneous semaglutide 2 mg weekly (−1.9 kg). However, the combination of subcutaneous semaglutide 2 mg/week and oral naltrexone-bupropion was the most successful in reducing his body weight (−11.5 kg) and insulin requirements, shown in Figure 1a. Subcutaneous liraglutide 1.8 mg (Victoza) once daily was started at age 56 with no difference in weight loss. This was changed to subcutaneous liraglutide 2.4–3 mg once daily (Saxenda) at age 60 and he achieved weight reduction of 4.7 kg over 9 months (from 99.6 kg to 94.9 kg). Liraglutide was stopped and changed to subcutaneous semaglutide 2 mg per week (Ozempic) at age 62 when the medication became available in our clinical practice. With semaglutide, the insulin bolus requirements were successfully reduced to one-third of his original dose. He achieved another 1.9 kg weight loss over 2 months. However, his weight loss subsequently plateaued and he still had impulsive snacking because of persistent cravings for food. He was added on oral naltrexone-bupropion (Contrave) 1 tablet twice a day with weight loss of 11.5 kg in 4 months (from 96 kg to 84.5 kg). This was associated with reduction for insulin requirements to 0.9 units/kg/day (80 units/day), a more than 10-fold decrease from the previous maximum of 1,000 units/day.

Fig. 1.

a Proband’s BMI and HbA1c results during the use of different anti-obesity medication regime. The rating scores of food cravings (scale out of 10), hunger (scale out of 10), and FCQ-T (scale out of 90) were shown. The proband’s continuous glucose profile while on subcutaneous liraglutide 1.8 mg daily (b) and while on subcutaneous semaglutide 2 mg weekly and oral naltrexone-bupropion are shown (c).

A review of his continuous glucose monitoring data demonstrated that combination of semaglutide-naltrexone-bupropion was superior compared to liraglutide in improving his diabetes control with less hyperglycaemic excursions and glycaemic variability (Fig. 1b, c). With the new medication regimen, the mean glucose was lower at 6.0 mmol/L compared to 9.6 mmol/L with a glucose management indicator (GMI) of 5.9% compared to 7.4%, respectively, when compared with the previous medication regimen (liraglutide and insulin); the time-in-range significantly improved at 90% compared to 56%, time above range (10.1–13.9 mmol/L) was 4% compared to 34%, time above range (>13.9 mmol/L) was 0% compared to 9%.

Identification of MC4R Variant and Effect of Anti-Obesity Agents on Cravings

Whole genome sequencing of his blood tests was performed by the Beijing Genomics Institute and analysed for rare variants in the obesity, lipodystrophy, hypertriglyceridemia, diabetes, and insulin resistance genes. This genetic testing confirmed the presence of a pathogenic heterozygous MC4R (c.380C>T, p.Ser127Leu) [7, 8]. This variant c.380C>T results in a serine-to-leucine substitution (p.Ser127Leu). This change disrupts the function of the MC4R, contributing to excessive hunger and reduced satiety. This genetic result explained the persistent cravings that the patient had since childhood and even while on liraglutide. Therefore, to compare the effect of anti-obesity medications on his food cravings, the patient was asked to rate his cravings and hunger using a 10-point Likert scale as well as a short version of the Food Cravings Questionnaire-Trait (FCQ-T-reduced) [5] on liraglutide 1.8 mg/day, liraglutide 3 mg/day, semaglutide 2 mg weekly and subsequently combination of semaglutide 2 mg weekly and naltrexone-bupropion. With each anti-obesity mediation/regimen, the craving score reduced from 10 to 9 to 7 and finally 4, respectively (score rating out of 10). Hunger scores remained at 10 out of 10 on liraglutide and semaglutide, which reduced to 5 out of 10 on combination of semaglutide and naltrexone-bupropion. Using the FCQ-T-reduced questionnaire to assess cravings, he scored 64 out of 90 on initially, and his craving scores reduced to 58 to 47 to 33, respectively, with the different anti-obesity medication regime. The improvement of cravings and hunger scores suggested that the combination therapy was most effective in lowering both cravings and hunger.

Monozygotic Twin Daughters

The patient had monozygotic twin daughters, both working full time and functionally independent (Fig. 2a). Genetic testing revealed that they both were heterozygous for the pathogenic MC4R variant (c.380C>T, p.Ser127Leu). They were born at full-term via normal vaginal delivery 10 min apart. The twins have a brother 8 years younger, born full-term via normal vaginal delivery. He did not have any known medical condition and was slim as a child and teenager. He was a picky eater at childhood and declined to eat certain foods. At adulthood, he had an increased appetite with some food cravings and gained weight to BMI 28 kg/m². His genetic testing was not undertaken. The twins’ mother was well, of normal weight, and did not have diabetes or metabolic disorders.

Fig. 2.

a Pictures of the identical twins at infancy and at age 35 years. b The varying BMI trend of the twins is shown.

Twin 1

At review of this study, twin 1 was at the age of 35 years with a BMI of 19.7 kg/m². She had a normal birthweight of 2.5 kg. She had no known medical history. She was taking levonorgestrel 150 μg and ethinylestradiol 30 μg (Microgynon) for birth control. She recalled that she was an average size compared to her peers until about 10 years old but gained weight from age of 13 years old during puberty and was large to extra-large in clothing size for her age. She lived in the USA from the age of 5 years old and moved back to Singapore at 16 years old with a weight of 60 kg (BMI 24.9 kg/m²). At age 19–25, her weight was around 70 kg with frequent intake of buffets and no exercise. From the age of 25, to lose weight, she then started intermittent fasting with a 5:2 diet and calorie counting. She had an intensive exercise regime which included jogging 8 km three times a week, yoga once a week, elliptical trainers once a week, weightlifting four times a week. Innately, she continued to have cravings for food and remained hungry all the time. She managed to achieve an average weight of 53–57 kg. At the height of COVID-19 restrictions at the age of 30–32 years, she gained 5–7 kg from increased intake of carbohydrates and comfort foods, even though she continued to run 6–8 km five times a week. She managed to lose 3–4 kg when restrictions were lifted with an average weight of 55 kg until the age of 33 years. She then further lost 7 kg to a weight of 48 kg due to emotional stressors with restriction in food intake to 1,500 kcal per day. Her current weight at review of this study was 47.4 kg (BMI 19.7 kg/m²) which she maintained with continued aerobic and resistance exercise (Fig. 2b).

Twin 2

At review of this study, twin 2 was at the age of 35 years with a BMI of 28.8 kg/m² (Fig. 2b). She had low birthweight of 1.55 kg and required to stay in neonatal intensive care unit for a month. Up to the age of 11 years old, she recalled that she was under-weight to normal in size compared to her peers. She was also shorter in height and smaller in weight compared to her twin sister. Twin 2 also had a co-morbidity of epilepsy which itself can predispose to obesity [9]. She was diagnosed with epilepsy at the age of 11 years and depression from 12 years old to 16 years old and was on sertraline. She started to gain weight during this period. On history taking, she reported cravings throughout the day and went for supper with friends at least twice a week. From the age of 24, she frequently snacked on items such as cereal bars, chips, chocolates, cookies, and candy and would drink 1–2 cans of soda a day. She led a sedentary lifestyle and did not exercise. Her current medications include anti-epileptic agents of lamotrigine 100 mg two times a day and valproate sodium chrono 300 mg morning and 500 mg night. Because twin 2 struggled with cravings and continued to gain weight (weight 66 kg; BMI 30.1 kg/m²), she was started on oral naltrexone-bupropion 8/90 mg 1 tablet OM for 2 weeks and increased to 1 tablet BD. Two months post-initiation, her craving score reduced from 9 to 6 out of 10, hunger scores remained at 7 out of 10 and improved from 51 to 44 out of 90 on the FCQ-T-reduced questionnaire. She had achieved a weight loss of 3 kg over 2 months (from 66 kg to 63 kg) and subsequently a further 6 kg over 3 months (from 63 kg to 57 kg) to a BMI of 26 kg/m² (Fig 2b).

Body Composition Analysis

Using DXA, the total body fat percentage were different among the 3 of them with the proband having the highest body fat percentage at 34.2%. The anthropometric, DXA, and metabolic biochemistry are summarized in Figure 3. Compared with twin 2, twin 1 had a lower body fat (19.1% vs. 25.5%), a lower fat mass index (3.81 kg/m² vs. 7.6 kg/m²), a lower android body fat % (16.3% vs. 28%), a lower lean mass index (15.4 kg/m² vs. 21.4 kg/m²), and lower visceral adipose tissue area on DXA (19.2 cm² vs. 134 cm²).

Fig. 3.

DEXA body composition scan image of the proband and his twin daughters with corresponding anthropometric, DXA, and metabolic (fasted) biochemistry result. The genetic tests for all 3 of them were heterozygous for an MC4R pathogenic variant.

Discussion

Here, we presented an interesting pedigree with heterozygous pathogenic MC4R variant causing increased food cravings and hunger. This genetic variant led to childhood-onset obesity in the patient (proband) as well as his monozygotic twin daughters. As far as we are aware, we are the first to report the use of combination of anti-obesity therapy semaglutide with naltrexone-bupropion in a patient with pathogenic MC4R variant, which reduced insulin requirements by 92% and weight by 15.5%. We also reported a few other interesting observations. Unlike the proband who had severe insulin resistance and young-onset type 2 diabetes, his daughters did not have diabetes. Twin 1 followed a different weight trajectory from her identical twin due to an intensive exercise regime, strict dietary restrictions, and an exemplary self-discipline and self-motivation. Twin 2 may have been at increased predisposition for adulthood obesity because of lower birthweight [10], medical history of childhood epilepsy [9], and valproate therapy [11, 12]. Our observation that diet control and healthy lifestyle can significantly modify the obesity trajectory in this case of monogenic obesity, is supported by twin studies which showed 60–90% of the inter-individual variance in BMI is determined by genetic factors and the rest by non-genetic influences [13]. Thus, our report here illustrated that while genetic predisposition is a strong risk factor for childhood-onset obesity, environmental factors play a critical and major role in the obesity phenotype and related complications in adulthood [14].

The genes affected in monogenic disorders all encode ligands and receptors involved in the highly conserved leptin-melanocortin pathway, which regulates food intake and body weight [15]. MC4R deficiency is the most common of human monogenic obesity and typically with an autosomal dominant pattern of inheritance [3, 16]. Other monogenetic obesity syndromes include leptin and leptin receptor deficiency, POMC deficiency syndrome, and PCSK1 mutation [1]. MC4R is primarily expressed in the hypothalamus and plays a crucial role in appetite regulation by interacting with neuropeptides such as orexigenic agouti-related peptide/neuropeptide Y (AgRP/NPY) which enhance appetite and inhibit energy expenditure and anorexigenic pro-opiomelanocortin/cocaine and amphetamine-regulated transcript (POMC/CART) neurons that project to second-order neurons using MC4R for activation, leading to decreased appetite and increased energy expenditure [17]. The Ser127Leu variant of the MC4R gene has been frequently observed in overweight and obese individuals [18–20]. Nevertheless, its presence has also been documented in lean individuals [21, 22], indicating incomplete phenotypic penetrance [22]. This mutation, situated at a highly conserved serine residue within transmembrane helix 3, is strongly suggestive of functional significance. While in vitro analyses have demonstrated high constitutive activity of the receptor [8, 23], markedly diminished cell surface expression of the MC4R Ser127Leu mutant [18] points towards a loss-of-function mechanism. A recent study reported the functional annotations of 369 MC4R variants [16]. Loss-of-function MC4R pathogenic variants promotes obesity [24] and conversely gain-of-function variants are protective against obesity [25]. Typically, individuals exhibit hyperphagia from the first year of life, with increased fat and lean mass and a marked increase in bone mineral density in childhood. Growth is accelerated in early childhood likely caused by disproportionate early hyperinsulinemia. Studies in children with MC4R monogenic obesity showed that children with MC4R variants were taller than obese children without MC4R variants at all ages, particularly during first 5 years [2]. Individuals with homozygosity for MC4R variant exhibited morbid early-onset obesity, while individuals with heterozygosity usually had a milder overweight phenotype [26].

Specifically tailored anti-obesity medications for MC4R variant monogenic obesity are yet to be discovered. Setmelanotide is a MC4R agonist that was FDA-approved for treatment of monogenic forms of obesity such as POMC, PCSK1, or LEP receptor deficiency in the year 2020. Setmelanotide therapy was shown to be effective to induce weight loss >10% in 80% of POMC-related obesity and 45% of LEPR-related obesity [27]. Setmelanotide is more effective as an anti-obesity medication in obesity caused by pathogenic genetic variants upstream of the MC4R pathway than heterozygous genetic variants of MC4 receptor [3]. A recent study by Collet et al. [16] showed that setmelanotide therapy reduced weight only modestly in heterozygous carriers of MC4R variants, and appeared to be more effective in individuals with obesity without MC4R variants. The authors reported that 28 days of continuous infusion of setmelanotide versus placebo resulted in only modest weight loss of 3.5 kg in six individuals with heterozygotes MC4R variants (placebo-corrected weight loss of −2.63 kg) compared with −3.07 kg in controls (placebo-corrected weight loss of −3.97 kg).

While GLP-1 receptor agonists are effective for weight loss, there are only a few reports of use in monogenic obesity. Iepsen et al. [28] identified 14 adults aged 18–65 years old positive for MC4R variants with a BMI above 28 kg/m² and demonstrated 6% weight loss after 16 weeks of treatment with liraglutide 3 mg. In a case report, the use of semaglutide 1 mg/week in a 13-year-old boy with heterozygous variant in MC4R was found to be effective with weight loss of 11% over 12 months [29]. As for the use of naltrexone-bupropion, there was a case report of a 33-year-old patient with heterozygous MC4R gene variant who successfully loss 48.9 kg (−26.7%) in 17 months of treatment following unsuccessful interventions with gastric bypass surgery, liraglutide 3 mg, and metformin treatment [30]. A recently published retrospective study of patients with suspected monogenic obesity in the Netherlands reported that in 2 patients with heterozygous MC4R variant, liraglutide therapy was not useful, instead naltrexone-bupropion was found to be effective with weight loss of −5.1% and −15.8% after 4–6 months of treatment [31]. On a combination therapy of GLP-1 analogue and naltrexone-bupropion, a single retrospective cohort study of 86 adult patients with BMI ≥30 kg/m² showed that addition of naltrexone-bupropion (following liraglutide initiation) was associated with a reduction in weight in both the initial responder (4.3% total body weight loss [p < 0.01]) and non-responder groups (4.0% total body weight loss [p < 0.01]) [32]. There is currently no available literature on the use of combination of GLP-1 analogue and naltrexone-bupropion in monogenic obesity.

In the case of proband reported here, the use of semaglutide and naltrexone-bupropion combination effectively induced weight loss likely through its synergistic effects targeting both gastrointestinal pathways and central nervous system to promote satiety, reduced cravings and food intake. GLP-1 receptor agonists mediate their effects via GLP-1 receptors expressed in the pancreas and extra-pancreatic tissues including gastrointestinal tract, heart, lungs, kidneys, and brain [33]. GLP-1 receptor agonists induce weight loss and improvement in glycaemic control by stimulating insulin secretion, inhibiting glucagon secretion, slowing gastric emptying, and increasing satiety, and reduce hunger by stimulating hypothalamic POMC/CART neurons which inhibits neuropeptide Y and agouti-related peptide [33, 34]. Naltrexone is an opioid antagonist which blocks opioid receptor-mediated POMC auto-inhibition, directly targeting the mesolimbic reward pathway. Bupropion is a dopamine and noradrenaline re-uptake inhibitor. Both naltrexone and bupropion are indicated for treatment of addictive disorders, by inhibiting the reward system (bupropion for depression and smoking cessation and naltrexone for opioid and alcohol addiction). Hence, naltrexone-bupropion combination influence food intake via its action in the reward system consisting of ventral tegmental area, nucleus accumbens, and prefrontal cortex which mediates reward, addiction and craving [35]. Future studies are needed to investigate the role of combination therapy of GLP-1 receptor agonist and naltrexone-bupropion in monogenic obesity disorder.

Conclusion

From this case series of a family with identical MC4R pathogenic variant, we observed that even this form of monogenic obesity arises from a complex interplay of genetics and environmental factors. As with all case reports and case series, limitations include the lack of controls for comparison and small sample size. There was also a potential for recall bias as the proband retrospectively recalled and reported the effects of anti-obesity agents on his cravings and hunger. However, from this case series of 3 individuals of a family, we reported multiple encouraging findings that add to the limited current literature on this form of monogenic obesity. First, the combination use of GLP-1 agonist and naltrexone-bupropion was highly effective for the proband with severe insulin resistance. Second, one of his daughters with the same genetic variant controlled her obesity using naltrexone-bupropion monotherapy. Finally, another daughter (monozygotic twin) avoided obesity and complications through her strong willpower of diet and exercise. As shown in our report, the use of precision medicine via genetics to understand and treat the pathophysiology of obesity in individuals with severe obesity (particularly childhood-onset) may be helpful to aid the multifaceted nature of obesity management.

Acknowledgments

We would like to thank the patient and his two daughters for participating in this study.

Statement of Ethics

Written informed consent was obtained from proband and his two daughters for publication of the details of their medical case and accompanying images. This study protocol was reviewed and approved by our local Institutional Review Board (SingHealth IRB), approval No. 2020/3154. Written informed consent was obtained from the individuals for publication of this manuscript and accompanying images. The CARE Checklist for this case report is available as online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000546795).

Conflict of Interest Statement

W.J.L. received funding from Singapore Ministry of Health’s National Medical Research Council (NMRC) Research Training Fund and has received honoraria from Medtronic, Abbott, DKSH, Roche, Novartis, iNova, Kowa, and Amgen. The other authors have no conflicts of interest to declare related to this study.

Funding Sources

This study was an investigator-initiated study supported by the Changi General Hospital Research Fund.

Author Contributions

W.J.L. was the clinician specialist managing these cases and was involved in the study conception, grant approval, data collection, project administration, and manuscript writing. J.J.X.L. was involved in data collection, writing the original draft preparation, review, and editing. A.J.H. was involved in the genetic test interpretation and manuscript writing. J.K. was involved in the manuscript editing. All authors have read and agreed to the published version of the manuscript.

Funding Statement

This study was an investigator-initiated study supported by the Changi General Hospital Research Fund.

Data Availability Statement

The data supporting the findings of this article are available within the text. Any enquiries regarding the data may be sent to the corresponding author.