Anorexia Nervosa

1. Disease Information

Overview

Anorexia nervosa is a complex eating disorder primarily characterized by a low body-mass index resulting from persistent restriction of energy intake relative to requirements, an intense fear of gaining weight or becoming fat, and a disturbance in the way one's body weight or shape is experienced. The disorder was first described in medical literature in the 19th century and remains one of the most lethal psychiatric conditions. As stated in the landmark GWAS publication: "Characterized primarily by a low body-mass index, anorexia nervosa is a complex and serious illness" (PMID: 31308545).

Key Identifiers

Synonyms and Alternative Names

• AN
• Anorexia nervosa, restricting type (AN-R)
• Anorexia nervosa, binge-eating/purging type (AN-BP)
• Anorexia athletica (subtype in athletes)
• Mental anorexia

Information Sources

This report integrates data from aggregated disease-level resources (GWAS meta-analyses, systematic reviews, population registries) and individual patient-level cohort studies. Key data sources include the Psychiatric Genomics Consortium (PGC), ENIGMA Eating Disorders Working Group, Danish Psychiatric Central Research Register, and multiple multinational clinical cohort studies.

2. Etiology

Disease Causal Factors

AN is a multifactorial disorder arising from the interaction of genetic, neurobiological, psychological, and environmental factors. It is now recognized as a metabo-psychiatric disorder, reflecting the discovery that genetic risk factors span both psychiatric liability and metabolic/anthropometric traits (PMID: 31308545; PMID: 38431502).

Genetic Risk Factors

Heritability and GWAS findings: Twin studies consistently estimate AN heritability at 48–74%, while SNP-based heritability is approximately 11–17%. As reported: "Genetic studies, particularly genome-wide association studies (GWAS), have identified key loci associated with ED susceptibility, with heritability estimates for these disorders ranging between 48% and 74%" (PMID: 39988782).

The 2019 PGC-ED GWAS meta-analysis of approximately 17,000 AN cases and 55,000 controls identified eight genome-wide significant loci, implicating genes involved in both psychiatric and metabolic pathways (PMID: 31308545).

Key susceptibility loci and candidate genes:

Targeted nanopore sequencing has identified additional poorly characterized variants in GWAS regions, including a polymorphic SINE-VNTR-Alu element (SVA-D) near IP6K2/PRKAR2A and a poly-T short tandem repeat in the 3'UTR of FOXP1, which may affect post-transcriptional processing (PMID: 39741260).

Cross-disorder genetic architecture: AN shares genetic correlations with multiple psychiatric disorders. Genomic structural equation modeling places AN within a "compulsive disorders" factor alongside OCD and Tourette syndrome (PMID: 39009701). Cross-disorder meta-analysis has identified 109 loci associated with at least two psychiatric disorders, with pleiotropic loci showing heightened brain expression beginning prenatally (PMID: 31835028).

AN is negatively genetically correlated with body fat percentage and fat-free mass, and Mendelian randomization identifies AN as potentially causal for decreased fat mass (PMID: 31852892).

Environmental Risk Factors

• Age and sex: Peak onset during adolescence (10–20 years); female:male ratio approximately 8–10:1
• Family history: First-degree relatives have 7–12× increased risk
• Childhood perfectionism and obsessive traits: Significantly associated with AN development; 40.4% of restrictive eaters display obsessive traits (PMID: 20179406)
• Weight/shape teasing: Established risk factor across severity levels (PMID: 38212857)
• Childhood obesity: History of overweight predisposes to eating disorder development
• Relational factors: Unmet emotional needs, feelings of non-belonging, exposure to conditional worth messaging about body/eating (PMID: 41236167)
• Athletic participation: Athletes in aesthetic sports have significantly higher prevalence (PMID: 37528996)
• Cultural factors: Western cultural idealization of thinness plays a salient role in rising incidence

Protective Factors

• Positive family emotional environment and secure attachment
• Healthy body image promotion (non-weight-focused)
• Early intervention for emotional distress
• Genetic protective variants: not well characterized, but some population-specific alleles may confer resilience

Gene-Environment Interactions

Epigenetic mechanisms are increasingly recognized as mediating gene-environment interactions in AN. Exposure to environmental stressors—particularly malnutrition itself—induces changes in DNA methylation, potentially contributing to disease chronification. Epigenome-wide association studies suggest potential reversibility of malnutrition-induced epigenetic changes upon recovery (PMID: 38849516; PMID: 30353170).

3. Phenotypes

Core Behavioral and Psychological Phenotypes

Regarding body image beliefs, a meta-analysis found overvalued ideas in 32.5% and delusional-like beliefs in 23.6% of AN patients, with greater belief rigidity correlating with poorer insight (PMID: 41707619).

Physical Manifestations and Clinical Signs

Laboratory Abnormalities

Electrolyte abnormalities may precede formal ED diagnosis by a median of 386 days, with 18.4% of individuals later diagnosed with an ED having preceding electrolyte abnormalities versus 7.5% of controls (aOR 2.12, 95% CI: 1.86–2.41) (PMID: 36346630).

Quality of Life Impact

AN produces severe impairment across all domains of daily functioning. Role impairment and comorbidity with other mental disorders are highly common (PMID: 19427647). Within the first year of diagnosis, patients are approximately 7× more likely to develop coded depression (HR 7.3, 95% CI: 6.6–8.1) and 9× more likely to engage in self-harm (HR 9.4, 95% CI: 8.2–10.7) compared to matched controls (PMID: 41282513).

4. Genetic/Molecular Information

Causal Genes and Pathogenic Variants

AN is a polygenic disorder without single causal gene mutations. The eight GWAS-significant loci implicate genes involved in neurodevelopment, synaptic function, and metabolic regulation. An estimated ~11,500 genetic variants are thought to contribute to the full heritability landscape (PMID: 31308545).

Candidate genes with robust evidence:

• BDNF (Brain-Derived Neurotrophic Factor): Reduced circulating levels in AN patients; Val66Met polymorphism associated with worse outcomes in humans, though not confirmed in rat ABA models (PMID: 35625351)
• HTR2A (Serotonin 2A Receptor): Meta-analytic support as a susceptibility gene for both AN and bulimia nervosa (PMID: 24202964)
• FOXP1: Transcription factor involved in neurodevelopment; poly-T STR in 3'UTR may affect post-transcriptional processing (PMID: 39741260)

Genetic Correlations with Other Traits

AN shows a unique pattern of genetic correlations that distinguish it from other psychiatric disorders:

Epigenetic Information

Five epigenome-wide association studies (EWASs) have been published on AN, suggesting potential reversibility of malnutrition-induced epigenetic changes upon recovery. Differential DNA methylation may serve as a biomarker for disease status or early diagnosis and may be involved in disease chronification (PMID: 38849516). The field remains nascent, with most studies limited by small sample sizes and candidate gene approaches (PMID: 30353170).

Chromosomal Abnormalities

No specific chromosomal abnormalities are causally linked to AN. Linkage studies have suggested susceptibility loci on chromosomes 1 and 10 for eating disorders broadly (PMID: 24340712).

5. Environmental Information

Environmental Factors

• Sociocultural environment: Western cultural idealization of thinness is consistently implicated in rising incidence. The disorder is more prevalent in higher socio-demographic index countries.
• Prenatal/perinatal factors: Maternal eating disorders are associated with higher risk of wheezing and asthma in offspring (OR: 1.25, 95% CI: 1.06–1.47), suggesting intergenerational effects (PMID: 41330642).

Lifestyle Factors

• Dieting behavior: Dietary restriction is the most proximal behavioral risk factor
• Excessive exercise: Both a symptom and perpetuating factor; athletes in aesthetic sports are at markedly elevated risk (PMID: 37528996)
• Sleep disruption: Bidirectional causal relationships between insomnia/sleep traits and AN identified via Mendelian randomization (PMID: 38703603)

Infectious Agents

AN is not caused by infectious agents. However, gut microbiome dysbiosis is increasingly recognized as a contributing factor in pathophysiology (see Section 6).

6. Mechanism / Pathophysiology

Molecular Pathways

Serotonergic system (GO:0007210 — serotonin receptor signaling pathway): Dysregulated serotonin neurotransmission is central to AN neurobiology. The 5-HT system mediates both aversive/inhibitory signaling and appetite regulation. As described: "Restricted eating may be a means of reducing negative mood caused by skewed interactions between serotonin aversive or inhibitory and dopamine reward systems" (PMID: 23333342).

Dopaminergic reward system (GO:0007212 — dopamine receptor signaling pathway): Altered reward circuitry is implicated in AN pathogenesis. Hyperdopaminergic mice show augmented vulnerability to activity-based anorexia (ABA), and reward deficits may underlie the paradoxical food avoidance seen in AN (PMID: 33768216; PMID: 28475260).

Opioid system: β-endorphin (a POMC gene product) is elevated in female mice undergoing ABA, and mu opioid receptor activation promotes food anticipatory activity, a key ABA feature (PMID: 33661571).

Adrenergic system: Catecholamine dysregulation is involved in AN pathophysiology; catecholamine levels may be higher in AN patients than in healthy controls, with higher norepinephrine concentrations in adipose tissue suggesting local sympathetic nervous system dominance (PMID: 37691603; PMID: 38310530).

Endocannabinoid system: Peripheral cannabinoid signaling is disrupted in AN, affecting both central appetite regulation and peripheral metabolic processes in adipose tissue, liver, pancreas, and skeletal muscle (PMID: 29437028).

Endocrine Dysregulation

The endocrine manifestations of AN are pervasive and represent one of the most clinically significant aspects of the disease. As comprehensively reviewed: "Dysfunction of the hypothalamic-pituitary axis includes hypogonadotropic hypogonadism with relative oestrogen and androgen deficiency, growth hormone resistance, hypercortisolaemia, non-thyroidal illness syndrome, hyponatraemia and hypooxytocinaemia. Serum levels of leptin, an anorexigenic adipokine, are suppressed and levels of ghrelin, an orexigenic gut peptide, are elevated in women with anorexia nervosa; however, levels of peptide YY, an anorexigenic gut peptide, are paradoxically elevated" (PMID: 27811940).

Causal chain:

Chronic energy restriction
    → Reduced adipose tissue → Suppressed leptin (CHEBI:81571)
    → Hypothalamic dysfunction
→ Decreased GnRH → Hypogonadotropic hypogonadism → Amenorrhea + Bone loss
→ Increased CRH → Hypercortisolaemia → Bone loss + Immune suppression
→ GH resistance → Low IGF-1 → Growth impairment
→ Decreased T3 → Non-thyroidal illness → Reduced metabolic rate
    → Gut peptide dysregulation → Elevated ghrelin + Paradoxical elevated PYY
    → Autonomic dysfunction → Sympathetic dysregulation in adipose tissue

Gut Microbiome

AN patients show reduced alpha diversity and lower short-chain fatty acid (SCFA) levels. Dysbiosis affects immune system responses, intestinal permeability, and neurotransmitter production via the gut-brain axis (PMID: 33416044; PMID: 33652962). Interestingly, recent evidence challenges the "leaky gut" concept in adolescent AN: zonulin and lipopolysaccharide-binding protein (LBP) levels are decreased rather than increased, suggesting reduced rather than increased paracellular intestinal permeability, and these alterations persist despite weight recovery (PMID: 40789230).

Microbiota-derived proteins may stimulate the autoimmune system, altering neuroendocrine control of mood and satiety. Microbial richness increases upon weight regain or fecal microbiota transplantation (PMID: 33416044).

Metabolic Changes

• Energy metabolism: Resting energy expenditure is very low in underweight patients but increases dramatically early in refeeding (PMID: 16721178)
• Lipid metabolism: Paradoxical hypercholesterolemia in acute illness; cholesterol levels are higher in females than males and correlate with severity of malnutrition (PMID: 37864342)
• Bone metabolism: Decreased bone formation, increased bone resorption; osteoporosis affects up to 50% of patients and may be irreversible (PMID: 38380189)

Brain Structure and Function

AN is associated with the largest brain structural deficits of any psychiatric disorder investigated by the ENIGMA consortium. A coordinated analysis of 685 AN patients and 963 controls revealed: "In AN, reductions in cortical thickness, subcortical volumes, and, to a lesser extent, cortical surface area were sizable (Cohen's d up to 0.95), widespread, and colocalized with hub regions" (PMID: 36031441).

A comprehensive meta-analysis confirmed: "Results showed significant global brain volume reductions in gray matter (GM), white matter (WM), and increases in cerebrospinal fluid (CSF) in acute AN (N = 1130 patients; N = 40 papers), gradually improving upon weight rehabilitation. However, even after 1.5 years of recovery, significantly lower global GM volume compared to healthy controls was found" (PMID: 41619402).

Key affected brain regions include the bilateral anterior and median cingulate cortex, with decreased both gray matter volume and resting-state functional activity (PMID: 34296492). The hippocampus shows volumetric and functional impairments contributing to memory and learning deficits (PMID: 33176113).

GO terms: GO:0007268 (chemical synaptic transmission), GO:0048167 (regulation of synaptic plasticity), GO:0007610 (behavior)

Cell types involved: CL:0000540 (neuron), CL:0000127 (astrocyte), CL:0000128 (oligodendrocyte), CL:0000129 (microglial cell)

7. Anatomical Structures Affected

Organ Level

Primary organs: - Brain (UBERON:0000955): Gray matter volume reduction, cortical thinning, white matter changes - Bone (UBERON:0001474): Osteopenia/osteoporosis; up to 50% of patients affected - Heart (UBERON:0000948): Bradycardia, arrhythmias, QTc prolongation - Endocrine glands (hypothalamus, pituitary, thyroid, adrenal, gonads): Global endocrine dysregulation

Secondary organs (complications): - Kidney (UBERON:0002113): Renal failure risk 6× higher in first year (HR 6.0, 95% CI: 4.2–8.5) (PMID: 41282513) - Liver (UBERON:0002107): Liver disease risk 6.7× higher in first year (HR 6.7, 95% CI: 3.8–11.7) - Gastrointestinal tract (UBERON:0005409): Gastric dilatation, delayed gastric emptying, constipation - Skin (UBERON:0002097): Lanugo, xerosis, acrocyanosis, purpura

Body systems involved: Cardiovascular, nervous, digestive, endocrine, musculoskeletal, integumentary, reproductive, renal, hematologic

Tissue and Cell Level

• Nervous tissue: Neurons (CL:0000540), glial cells in cortical and subcortical regions
• Bone tissue: Osteoblasts (CL:0000062), osteoclasts (CL:0000092) — imbalanced remodeling
• Adipose tissue: Adipocytes (CL:0000136) — severe depletion
• Cardiac muscle: Cardiomyocytes (CL:0000746) — atrophy
• Intestinal epithelium: Enterocytes (CL:0000584) — altered permeability

Brain Localization (UBERON terms)

• Cingulate cortex (UBERON:0003027): Both structural and functional deficits
• Hippocampus (UBERON:0002421): Volumetric and memory impairments
• Hypothalamus (UBERON:0001898): Endocrine dysregulation center
• Prefrontal cortex (UBERON:0000451): Enhanced executive inhibition of reward drives
• Precuneus (UBERON:0035150): Cortical thinning

8. Temporal Development

Onset

• Typical age of onset: Adolescence (10–20 years); cumulative lifetime prevalence analysis shows the majority of eating disorders have initial onset in this window (PMID: 19427647)
• Onset pattern: Usually insidious, beginning with dietary restriction that progressively intensifies
• Prolonged emotional distress typically precedes symptom onset, with opportunities for early prevention/intervention often missed (PMID: 41236167)

Progression

• Disease course pattern: Variable — episodic/relapsing-remitting in many; chronic-progressive in 10–20%
• Disease stages:
• Early: Restrictive eating, initial weight loss, psychological symptoms
• Intermediate: Medical complications emerge (amenorrhea, bradycardia, electrolyte disturbances)
• Advanced: Severe malnutrition, multi-organ involvement, psychiatric comorbidity
• End-stage: Organ failure, refractory to treatment, high mortality risk

Long-term Outcomes

In a landmark 21-year follow-up study: "Fifty-one per cent of the patients were found to be fully recovered at follow-up, 21% were partially recovered and 10% still met full diagnostic criteria for anorexia nervosa. Sixteen per cent were deceased, due to causes related to anorexia nervosa. The standardized mortality rate was 9.8" (PMID: 11459385).

A 13-year follow-up of 484 patients found 60.3% recovered, 25.8% relatively good outcome, and 6.4% each with bad and severe outcomes. Recovery rate increased with elapsing relapse-free time (p=0.02) (PMID: 21317006).

Critical Periods

• Adolescence (10–19 years): Primary window of vulnerability and optimal intervention
• Transition from adolescent to adult services: Risk of discontinuity in care (PMID: 33223229)
• First 2 years of illness: Key predictors of long-term outcome established; early response to treatment predicts better outcomes

9. Inheritance and Population

Epidemiology

Inheritance Pattern

AN follows a multifactorial/polygenic inheritance pattern. Key genetic architecture features:

• Heritability: 48–74% (twin studies); moderate heritability confirmed in population-level register data from Denmark and Sweden covering >67,000 individuals with eating disorders (PMID: 40615413)
• Penetrance: Incomplete and age-dependent
• Expressivity: Highly variable — spectrum from subclinical features to life-threatening illness
• Polygenicity: Estimated ~11,500 variants explain 90% of genetic heritability

Population Demographics

• Sex distribution: Predominantly female; however, male cases are likely underdiagnosed and males with EDs have unmet needs (PMID: 34246009)
• Ethnic/racial disparities: Latinx and Asian patients are half as likely as White patients to receive recommended treatment (aOR 0.49 and 0.55, respectively) (PMID: 36694235)
• Geographic distribution: Higher prevalence in Western/high-income countries, but increasingly recognized globally
• Insurance disparities: Publicly insured patients are one-third as likely to receive recommended treatment (PMID: 36694235)

10. Diagnostics

Clinical Criteria

DSM-5 Diagnostic Criteria (307.1): 1. Restriction of energy intake relative to requirements, leading to significantly low body weight 2. Intense fear of gaining weight or persistent behavior interfering with weight gain 3. Disturbance in the way body weight or shape is experienced

Subtypes: - Restricting type (AN-R) - Binge-eating/purging type (AN-BP)

Severity based on BMI: - Mild: BMI ≥ 17 kg/m² - Moderate: BMI 16–16.99 kg/m² - Severe: BMI 15–15.99 kg/m² - Extreme: BMI < 15 kg/m²

Latent class analysis has empirically identified four phenotypic classes, with obsessive-compulsive features differentiating among restricting AN subgroups (PMID: 14757596).

Laboratory Tests

REMS (Radiofrequency echographic multispectrometry) shows good agreement with DXA for BMD assessment in AN and may be particularly useful during fertile age and pregnancy (PMID: 35896857).

Imaging Studies

• Brain MRI: May reveal gray matter volume reductions and cortical thinning; lower brain volumes of the cerebellum and subcortical gray matter at admission predict worse outcomes (PMID: 37263169)
• PET/SPECT: Research tools for serotonergic and dopaminergic receptor density mapping; clinical utility not established (PMID: 32234640)

Screening Tools

• SCOFF questionnaire: Brief 5-question screening tool
• EAT-26 (Eating Attitudes Test): 26-item self-report measure
• EDI-2 (Eating Disorder Inventory): Comprehensive psychometric assessment
• Electrolyte monitoring: Otherwise unexplained electrolyte abnormalities may identify individuals who benefit from ED screening (PMID: 36346630)

Genetic Testing

Genetic testing is not currently part of standard clinical practice for AN. However, the Eating Disorders Genetics Initiative 2 (EDGI2) is advancing polygenic risk scoring that may eventually enable risk stratification (PMID: 40419993).

Differential Diagnosis

• Bulimia nervosa (binge-purge without low weight)
• Avoidant/restrictive food intake disorder (ARFID) (no body image distortion)
• Binge eating disorder
• Major depressive disorder with appetite loss
• Medical conditions causing weight loss (malignancy, Crohn's disease, hyperthyroidism, Addison's disease)
• Body dysmorphic disorder (significantly higher delusionality than AN) (PMID: 41707619)

11. Outcome/Prognosis

Mortality

AN has the highest mortality rate of all mental disorders:

Eating disorders are independent risk factors for suicide, with bulimia nervosa (HR 2.59) and other eating disorders (HR 2.31) maintaining significance even after adjusting for psychiatric comorbidities. For AN specifically, the association became non-significant after adjustment for comorbid psychiatric conditions (PMID: 40280427).

Mortality rates may be influenced by treatment quality: one Italian center with an integrated multidisciplinary treatment network found SMR of 1.19 (95% CI: 0.79–1.81), not significantly different from the general population (PMID: 36062404).

Recovery and Long-term Outcomes

Prognostic Factors

Poor prognostic factors: - Low BMI at discharge - Low energy and fat intake - High drive for excessive exercise - High perfectionism and interpersonal distrust - High anxiety - Psychiatric comorbidity - Binge/purge subtype transition - Higher levels of depression and compulsivity (PMID: 17628126) - Lower brain volumes at admission (PMID: 37263169) - Older age at onset

Favorable prognostic factors: - Adolescent age at treatment - Early symptom improvement - Maintaining contact with mother - Shorter duration of illness before treatment - Higher pre-treatment BMI (PMID: 37697396)

Complications

Within 10 years of diagnosis, patients experience excess adverse outcomes (PMID: 41282513): - 110 excess renal failure events per 10,000 individuals - 26 excess liver disease events per 10,000 individuals - 95 excess all-cause deaths per 10,000 individuals - 341 excess unnatural deaths per 100,000 individuals

12. Treatment

Psychotherapy (First-line)

Family-Based Treatment (FBT) — MAXO:0000950 (counseling): The leading evidence-based treatment for adolescent AN. "In 5 randomized controlled trials (RCTs) in anorexia nervosa (N=560) remission rates were between 21.2-42% at end of treatment, between 21.8-40% at 6-month follow-up, and between 29-49% at 12-month follow-up" (PMID: 31466116).

In a naturalistic Finnish study, 61.5% of adolescents achieved full weight restoration (EBW ≥95%) with FBT, and 42.3% required no further treatment (PMID: 37697396).

Cognitive Behavioral Therapy-Enhanced (CBT-E) — MAXO:0000376 (cognitive behavioral therapy): Recommended as a second-line approach for adolescents when FBT is not effective or applicable, and as a primary approach for adults. CBT-E shows moderate to large effect sizes regardless of previous FBT failure (PMID: 30829421; PMID: 33223229).

Pharmacotherapy

No medications are currently FDA-approved specifically for AN. Key pharmacological considerations:

Nutritional Rehabilitation (MAXO:0001298 — dietary modification)

Weight gain of 2.2–4.4 lb per week stabilizes cardiovascular health (PMID: 33382560). Refeeding syndrome risk requires careful monitoring of phosphate, magnesium, and potassium levels.

Hospitalization (MAXO:0010363 — inpatient care)

Hospitalization effectively overcomes the acute phase and promotes lasting changes. A multidisciplinary approach comprising clinical/nutritional, psychotherapeutic, family, occupational, and body therapy components is recommended (PMID: 12452251).

Treatment Strategy and Future Directions

• Integrated multidisciplinary networks may significantly reduce mortality (PMID: 36062404)
• GLP-1 receptor agonists are being investigated via Mendelian randomization for potential effects on AN (PMID: 40141382)
• Psychedelic-assisted therapy, novel monoaminergic drugs, and hormone analogues are emerging areas of investigation (PMID: 32858054)
• Probiotics and microbiome-targeted therapies show early promise for restoring gut-brain axis function (PMID: 33652962)

13. Prevention

Primary Prevention

• Media literacy programs to counter thin-ideal internalization
• Body image education emphasizing positive body focus rather than weight/dieting (PMID: 33382560)
• Family support programs for building healthy emotion regulation skills
• Athlete screening: Targeted prevention in aesthetic sports populations (PMID: 37528996)

Secondary Prevention (Early Detection)

• Electrolyte screening: Unexplained electrolyte abnormalities in adolescents should prompt ED assessment (PMID: 36346630)
• BMI trend monitoring: Subtle weight changes should be investigated in primary care (PMID: 33382560)
• Web-based parent interventions: The E@T program showed significant increase in expected body weight at 12-month follow-up (Cohen d=0.42), though parent adherence was low (PMID: 30552078)

Tertiary Prevention

• Long-term psychiatric monitoring, as 6% of treated patients attempt suicide even years after initial treatment (PMID: 35142161)
• Ongoing bone density monitoring and treatment
• Relapse prevention through continued therapeutic support; relapse risk declines after 4 years symptom-free (PMID: 9054777)

Genetic Counseling

While genetic testing is not standard, families with affected members should be informed of the elevated familial risk (7–12× for first-degree relatives). The EDGI2 initiative may eventually enable polygenic risk-based screening (PMID: 40419993).

14. Other Species / Natural Disease

Animal Models of Relevance

While AN does not occur naturally in animals in the same form as in humans, self-starvation behaviors have been documented in various species:

• Activity-based anorexia (ABA) in rodents: The most widely used animal model, in which rats or mice with unlimited running wheel access and restricted food availability develop paradoxical hypophagia, hyperactivity, and life-threatening weight loss, recapitulating key features of human AN including hypothermia and anhedonia (PMID: 28475260; PMID: 38103992)

Comparative Biology

The POMC/opioid system and dopaminergic reward circuitry involved in AN are evolutionarily conserved across mammals, supporting the translational relevance of rodent models. However, the cognitive and psychological components of AN (body image distortion, fear of weight gain) cannot be modeled in animals, representing a fundamental limitation.

15. Model Organisms

Activity-Based Anorexia (ABA) Model

Species: Primarily Mus musculus (NCBI Taxon: 10090) and Rattus norvegicus (NCBI Taxon: 10116)

Protocol: Combining limited food access with unlimited running wheel access produces paradoxical decrease in food intake, hyperactivity, and life-threatening weight loss (PMID: 34124309).

Phenotype recapitulation: - Severely restricted food intake - Excessive exercise/hyperactivity - Dramatic weight loss - Loss of reproductive cycles - Hypothermia - Anhedonia - Elevated POMC mRNA and beta-endorphin - Body image distortion — NOT modeled - Fear of weight gain (cognitive component) — NOT modeled - Voluntary nature of food restriction — NOT modeled

Genetic Models

Research applications: The ABA model enables investigation of neurobiological mechanisms, pharmacological interventions, genetic susceptibility factors, and the distinction between vulnerable and resilient phenotypes (PMID: 38103992).

Key Findings — Detailed Evidence

Finding 1: AN is a Complex Metabo-Psychiatric Disorder with the Highest Psychiatric Mortality

The reconceptualization of AN as a metabo-psychiatric disorder represents a paradigm shift in understanding this illness. The 2019 PGC-ED GWAS meta-analysis demonstrated that genetic risk factors for AN span both psychiatric liability (correlations with OCD, schizophrenia, depression) and metabolic traits (negative correlations with BMI, body fat, type 2 diabetes; positive correlations with HDL cholesterol and physical activity). This dual nature means that AN cannot be adequately understood through either a purely psychiatric or purely metabolic lens. The mortality data are stark: the most current meta-analysis confirms AN has the highest all-cause mortality ratio (RR=5.52) and suicide-related mortality (RR=9.86) of all eating disorders and indeed all mental illnesses.

Finding 2: Global Endocrine Dysregulation

The pervasive endocrine dysfunction in AN extends across virtually every hormonal axis and represents both an adaptive response to chronic starvation and a self-perpetuating pathological mechanism. The paradoxical elevation of peptide YY (an anorexigenic hormone) in the context of severe underweight may contribute to the maintenance of food restriction. The endocrine changes have direct consequences for bone health (osteoporosis via estrogen deficiency and hypercortisolism), cardiovascular health, and reproductive function. Critically, most endocrine disturbances are reversible with sustained weight restoration, providing both therapeutic targets and indicators of recovery.

Finding 3: Brain Structural Deficits

The ENIGMA consortium's finding that AN produces the largest cortical thickness deficits of any psychiatric disorder (Cohen's d up to 0.95) has profound implications for understanding cognitive function, treatment response, and long-term outcomes in AN. The deficits are widespread, colocalize with cortical hub regions, and are directly associated with BMI. While partially reversible with weight restoration, global gray matter volume remains significantly lower even after 1.5 years of recovery, suggesting some degree of lasting neural impact. Lower brain volumes at admission predict worse clinical outcomes, supporting neuroimaging as a potential prognostic tool.

Finding 4: Family-Based Treatment Efficacy and Treatment Landscape

FBT remains the treatment with the strongest evidence base for adolescent AN, though remission rates of 29–49% at 12-month follow-up highlight the substantial proportion of patients who do not respond. The absence of any FDA-approved pharmacological treatment for AN is a critical gap. Olanzapine shows the most promising data as an adjunctive agent (significantly greater weekly weight gain at approximately 9 mg/day), while novel bone-targeted therapies (teriparatide, romosuzumab) address the skeletal complications. The finding that long-term outcomes may be improved by integrated multidisciplinary treatment networks — potentially normalizing mortality rates — underscores the importance of care delivery models.

Mechanistic Model

GENETIC VULNERABILITY (48-74% heritability, 8+ GWAS loci)
    |
    Polygenic risk across psychiatric + metabolic pathways
    (Serotonin, dopamine, BDNF, FOXP1, CADM1)
    |
ENVIRONMENTAL TRIGGERS
    |-- Sociocultural pressure (thin ideal)
    |-- Psychological traits (perfectionism, anxiety, OCD)
    |-- Relational factors (unmet needs, conditional worth)
    +-- Dieting / caloric restriction
 |
DISEASE INITIATION
    |-- Reward circuit dysregulation (decreased dopamine signaling)
    |-- Enhanced executive inhibition of feeding drives
    +-- Serotonin-mediated anxiety reduction through restriction
 |
SELF-PERPETUATING MECHANISMS
    |-- Starvation --> Endocrine dysregulation
    |     |-- Decreased Leptin --> Hypothalamic dysfunction
    |     |-- Increased Cortisol --> Bone loss + immune suppression
    |     |-- Decreased GnRH --> Hypogonadism --> Amenorrhea
    |     +-- Paradoxical increased PYY --> Maintained anorexia
    |-- Starvation --> Brain volume loss (d=0.95)
    |     +-- Cognitive rigidity --> Impaired treatment response
    |-- Starvation --> Gut dysbiosis
    |     +-- Altered gut-brain signaling
    |-- Starvation --> Epigenetic changes
    |     +-- Possible disease chronification
    +-- Beta-endorphin elevation --> Exercise reward
 +-- Hyperactivity reinforcement
      |
CLINICAL MANIFESTATIONS
    |-- Psychiatric: Depression, anxiety, OCD, suicidality
    |-- Skeletal: Osteoporosis (50%), fractures
    |-- Cardiac: Bradycardia, arrhythmias
    |-- Metabolic: Electrolyte disturbances, hypercholesterolemia
    |-- Renal/Hepatic: Organ damage (HR 6.0-6.7)
    +-- Neurological: Persistent GM volume reduction
 |
OUTCOMES (without adequate treatment)
    |-- Recovery: ~50-60%
    |-- Chronic illness: ~10-20%
    +-- Death: SMR 5.52 (highest of all psychiatric disorders)

Evidence Base

Landmark Studies

Limitations and Knowledge Gaps

1. Genetic architecture: Only 8 GWAS-significant loci identified to date; the EDGI2 initiative aims to dramatically expand sample sizes across diverse ancestral backgrounds, which is essential given that most genetic studies have been conducted in European-ancestry populations.

2. Pharmacological treatment gap: No approved medications for core AN symptoms; olanzapine and other agents show promise but large-scale RCTs are lacking. The network meta-analysis protocol (EfaNosa, PMID: 41366465) will provide the first systematic comparison of all available treatments.

3. Male AN is understudied: Males represent 10–13% of clinical samples but are likely underdiagnosed. Mortality and morbidity data for males are limited due to small sample sizes (PMID: 34246009).

4. Epigenetic studies remain in infancy: Sample sizes are small, methods are heterogeneous, and longitudinal data are limited. Whether epigenetic changes are causes, consequences, or biomarkers of illness remains unclear (PMID: 38849516).

5. Brain recovery trajectory: While acute brain volume deficits are well-documented, the timeline and completeness of neurological recovery remain uncertain, with evidence suggesting persistent deficits beyond 1.5 years of recovery.

6. Treatment matching: No validated predictors exist for matching individual patients to optimal treatment modality (FBT vs. CBT-E vs. other approaches).

7. Microbiome causality: While gut dysbiosis is documented in AN, whether it is cause, consequence, or both remains unclear. The unexpected finding of reduced rather than increased intestinal permeability challenges prevailing hypotheses.

8. Racial/ethnic and socioeconomic disparities: Significant disparities in treatment access exist; culturally adapted interventions are lacking.

Proposed Follow-up Experiments/Actions

1. Expand GWAS in diverse populations: Support EDGI2 recruitment across non-European ancestries to identify population-specific and shared risk loci, potentially doubling the number of significant loci.

2. Longitudinal epigenome-wide association studies: Design EWAS with >500 participants at multiple time points (acute, weight-restored, 2+ years recovered) to distinguish state vs. trait epigenetic markers and identify biomarkers for relapse risk.

3. RCT comparing FBT vs. CBT-E in adolescents: Directly compare the two leading psychotherapies with standardized outcome measures and treatment matching analysis to identify which patients benefit from each approach.

4. Placebo-controlled multi-site olanzapine trial: Conduct definitive phase III trial with sufficient power to establish olanzapine as adjunctive therapy for AN, including both weight and psychological outcomes.

5. Neuroimaging prognostic biomarker validation: Prospective multi-site study using standardized MRI protocols to validate brain structural measures as prognostic tools for treatment response and long-term outcome.

6. Microbiome intervention trials: Randomized trials of fecal microbiota transplantation or targeted probiotic supplementation in AN patients during refeeding to assess impact on gut-brain axis function, mood, and treatment response.

7. Polygenic risk score clinical utility: Test whether polygenic risk scores can identify high-risk individuals for targeted prevention, particularly in families with affected members and in athletic populations.

8. GLP-1 receptor agonist safety monitoring: Given the widespread use of GLP-1 receptor agonists for obesity, establish systematic monitoring for emergence or exacerbation of eating disorder symptoms in treated populations.

9. Bone therapy optimization: Conduct larger RCTs of romosuzumab and teriparatide in AN-related osteoporosis to establish evidence-based treatment protocols for this common and potentially irreversible complication.

10. Culturally adapted interventions: Develop and test ED treatment programs specifically designed for underserved populations (Latinx, Asian, publicly insured youth) to address documented disparities in treatment access and outcomes.

Report generated from systematic analysis of 89 publications spanning genetics, neurobiology, clinical outcomes, and therapeutics of anorexia nervosa. All citations verified against PubMed abstracts.