Body Mass Index Fluctuation and Variability

By the ExomeDNA Research Team

Body Mass Index (BMI) variance refers not to mean body weight, but to how much an individual's BMI fluctuates across time and contexts. While most BMI genetic research focuses on identifying variants associated with higher average weight, a distinct body of work examines the genetic basis of intra-individual variability in BMI. Research by Wang et al. (2019) and Ahmad et al. (2016) has identified loci, including BDNF, ADCY3, FANCL, and BCDIN3D, where genetic variants are associated with greater or lesser fluctuation in BMI, independent of the variants that determine a person's weight set point. Understanding the genetics of BMI variability offers a different lens on weight biology: not just where your weight tends to settle, but how stably it settles there.

What is BMI variability?

BMI variability refers to the degree to which a person's body mass index fluctuates over time. It is distinct from mean BMI: two people with the same average BMI over a decade may have very different variance, one remaining stable at a consistent value, the other oscillating substantially. This variability is itself a heritable phenotype, shaped by genetic variants that influence biological mechanisms governing weight stability and adaptability.

Research base: Moderate.

High BMI variance may reflect greater biological responsiveness to environmental fluctuations in diet, activity, or stress. Low BMI variance suggests a more tightly regulated weight set point. The genetic architecture underlying BMI variance is partially distinct from that underlying mean BMI, meaning that studying this phenotype reveals biological mechanisms not visible in traditional genome-wide analyses of average weight.

The genetics of weight fluctuation

Genome-wide approaches to studying BMI variance have employed methods that detect associations with the degree of fluctuation around the mean, rather than the mean itself. These variance-GWAS and genotype-by-environment interaction frameworks have identified a partially distinct gene set from standard BMI analyses, while sharing some loci such as ADCY3 and BDNF.

Wang et al. (2019) applied a genotype-by-environment interaction framework to UK Biobank data, inferring genetic effects on phenotypic variability across complex traits including BMI. This approach identified variants where the effect on BMI is not constant across environmental contexts, in other words, variants that make BMI more or less environmentally responsive (Wang et al., 2019).

Ahmad et al. (2016) examined gene-environment interactions for BMI in a study of 14,131 Pakistani adults, identifying a novel interaction locus associated with BMI variability in conjunction with smoking exposure (Ahmad et al., 2016). This work illustrated that the genetic architecture of BMI variability also has components that vary by ancestry and environmental exposure, reinforcing the value of diverse and interaction-aware study designs.

Key genes: BDNF, ADCY3, and FANCL

BDNF, brain-derived neurotrophic factor, is one of the most robustly replicated single-gene associations with BMI in human genetics. In the context of BMI variance, BDNF is particularly compelling: it regulates the plasticity of hypothalamic circuits governing appetite and satiety, and BDNF levels in the brain fluctuate in response to diet, exercise, stress, and sleep. This responsiveness to environmental signals may translate to greater BMI variability for people with variants that alter BDNF expression or signaling efficiency. Common variants near BDNF are associated with both higher mean BMI and with appetite dysregulation phenotypes in multiple population studies.

ADCY3 encodes adenylyl cyclase 3, the enzyme that generates cyclic AMP in response to hormonal signals. Cyclic AMP mediates lipolytic signaling in fat cells, appetite regulation in the hypothalamus, and insulin sensitivity in the pancreas. Variants that reduce ADCY3 signaling efficiency may impair the tight homeostatic feedback that ordinarily stabilizes body weight, contributing to greater BMI variability in addition to influencing mean BMI. The appearance of ADCY3 in both mean BMI and BMI variance analyses suggests it operates at a central regulatory node relevant to multiple dimensions of weight biology.

FANCL encodes the E3 ubiquitin ligase subunit of the Fanconi anemia core complex, a DNA damage repair pathway. Its presence in a BMI variance cohort reflects growing evidence that DNA damage response pathways influence adipogenesis, the differentiation of precursor cells into mature fat cells, and that metabolic stress can activate these repair mechanisms in adipose progenitor cells. Variants in FANCL may alter the pace or fidelity of adipocyte renewal, contributing to variability in fat mass regulation over time.

BCDIN3D encodes an RNA methyltransferase belonging to the Rossmann-fold methyltransferase family that acts as a 5'-methylphosphate capping enzyme specific for cytoplasmic histidyl tRNA. RNA modification enzymes are increasingly recognized as regulators of translational efficiency, affecting how much protein is produced from a given transcript. Downstream effects on metabolic gene expression may contribute to the BMI variability signal at this locus, though the precise mechanism requires further investigation.

FAIM2 (Fas apoptosis inhibitory molecule 2) is a transmembrane protein involved in regulating cell survival signals. Its presence in the BMI variance gene set adds to the emerging picture that cellular survival and renewal pathways in adipose tissue may contribute to the biological variability in body weight observed across individuals.

What the research says

BMI variance genetics is a younger field than mean BMI genetics, and the evidence base is correspondingly smaller. Wang et al. (2019) made a methodological contribution by demonstrating that genotype-by-environment interactions inferred from phenotypic variability in the UK Biobank could identify loci not detectable by standard mean-trait analysis, validating variance GWAS as a complementary approach to traditional association studies (Wang et al., 2019).

Stat block: Wang et al. (2019) used a variance-based genome-wide analysis framework applied to UK Biobank participants to identify genetic variants associated with phenotypic variability in BMI and other complex traits, demonstrating that variance GWAS reveals biological signal not captured by mean-trait analysis alone (Wang et al., 2019).

Ahmad et al. (2016) contributed evidence of gene-environment interaction for BMI in a non-European ancestry population of 14,131 Pakistani adults, illustrating that gene-by-environment interactions involving BMI variability have both ancestry-specific and environmentally-contingent components (Ahmad et al., 2016).

Stat block: The gene candidates in the BMI variance cohort include both shared loci with mean-BMI genome-wide studies (BDNF, ADCY3) and loci that appear specifically in variance analyses (FANCL, BCDIN3D), suggesting that distinct biological mechanisms underlie weight stability versus weight level, a finding with implications for how genetic BMI profiles should be interpreted in personalized contexts.

How BMI variability genetics affects you

People with genetic variants associated with higher BMI variability may find that their weight responds more readily to changes in diet, activity level, stress, or sleep. This responsiveness can function as both a biological asset and a challenge, depending on the context. Greater genetic susceptibility to BMI fluctuation means that positive lifestyle changes may produce more pronounced weight responses, and that periods of metabolic stress may also produce larger deviations from baseline.

This profile is qualitatively different from a high mean-BMI genetic profile. A person with a stable high-set-point genotype may tend toward a consistently higher weight that is resistant to environmental change. A person with a high-variance genotype may show wider oscillation but also respond more readily to structured interventions. The distinction matters for how lifestyle strategies are designed and evaluated.

Weight cycling, the pattern of repeated cycles of weight loss and regain, has independently documented biological effects on metabolic function, lean mass, and hormonal regulation. For people with genetically higher BMI variability, minimizing the amplitude of weight fluctuations is a worthwhile goal in its own right.

Working with your BMI variability profile

Variants associated with BMI variance point toward weight stability as an explicit health goal, not just weight level management. Strategies that reduce the amplitude of weight fluctuations, rather than targeting average weight alone, may be particularly aligned with the biology highlighted by this genetic profile.

Practical starting points:

• Aim for dietary consistency rather than cycling between highly restrictive and unrestricted eating patterns; irregular dietary behavior is more likely to amplify BMI variability for this genetic profile
• Monitor weight trends over weeks and months rather than daily, to distinguish meaningful biological signals from normal day-to-day fluctuation; people with higher genetic BMI variance may see more short-term variability
• Prioritize sleep regularity, as circadian disruption independently promotes BMI variability via effects on appetite hormones, and BDNF activity is particularly sensitive to sleep quality
• Regular aerobic and resistance exercise supports hypothalamic BDNF levels, which may contribute to more stable appetite regulation for people with variants in this gene
• For people with FANCL-pathway variants, minimizing chronic metabolic stress through blood sugar stability, anti-inflammatory dietary patterns, and consistent exercise may be relevant to reducing oxidative load on adipose progenitor cells

Frequently asked questions

Q: What is the difference between BMI variance and mean BMI genetics? A: Mean BMI genetics identifies variants associated with higher or lower average body weight. BMI variance genetics identifies variants associated with how much BMI fluctuates over time. The two overlap partially: genes like ADCY3 and BDNF influence both dimensions of weight biology. Other genes, including FANCL and BCDIN3D, appear specifically in variance analyses, suggesting distinct biological mechanisms underlying weight stability versus weight level.

Q: Does high BMI variability mean my weight is unhealthy? A: Not necessarily. BMI variability describes how much weight fluctuates, not whether it falls in a particular range. People with genetically higher BMI variability may respond more readily to both positive and negative changes in diet, activity, and lifestyle. The health implications depend on many factors beyond the variance characteristic itself.

Q: Why is BDNF relevant to weight variability? A: BDNF regulates the plasticity of brain circuits governing appetite and energy balance. Because BDNF levels respond to environmental factors including diet, exercise, stress, and sleep, variants that alter BDNF signaling may make hypothalamic appetite circuits more environmentally responsive, contributing to greater BMI fluctuation alongside any effect on average weight.

Q: What does FANCL have to do with body weight? A: FANCL is primarily known for its role in the Fanconi anemia DNA damage repair pathway. Emerging research suggests that DNA repair mechanisms are relevant to the renewal of adipose progenitor cells, and that metabolic stress can activate these pathways in fat tissue. Variants in FANCL may influence adipocyte biology in ways that affect fat mass stability over time.

Q: Can lifestyle changes reduce genetic BMI variability? A: Lifestyle choices do not alter underlying genetic variants, but they can modify the environmental context in which those variants operate. Consistent dietary patterns, regular sleep, and structured exercise can reduce the amplitude of BMI fluctuation by reducing the environmental variability that interacts with these loci.

References

Ahmad S, et al. (2016). A novel interaction between the FLJ33534 locus and smoking in obesity: a genome-wide study of 14,131 Pakistani adults. International Journal of Obesity. PMID: 26278006.

Wang H, et al. (2019). Genotype-by-environment interactions inferred from genetic effects on phenotypic variability in the UK Biobank. Science Advances. PMID: 31453325.

Data sources: GWAS Catalog, Open Targets, ClinVar, ClinGen, NCBI Gene, dbSNP, PheGenI.