evipedia.aidev
← Back

Fasting-Mimicking Diet for Health & Longevity

Evidence Review created on 08/31/2026 using AI4L / Grok 4

Also known as: FMD, Fasting Mimicking Diet, Fast-Mimicking Diet, Periodic Fasting-Mimicking Diet, ProLon

Motivation

A fasting-mimicking diet is a short, planned stretch of very low-calorie, plant-based eating meant to copy the body’s responses to going without food while still providing some nourishment. It usually lasts five days, with more calories on day one and very little protein and sugar thereafter. It interests people who want fasting-like repair without a water-only fast.

The protocol came out of longevity laboratories as a practical substitute for prolonged fasting. It is sold as a boxed five-day food kit and is also assembled from ordinary groceries. Clinics later tested monthly cycles in type 2 diabetes, alongside cancer drugs, and in inflammatory bowel disease. Much of the trial food was supplied by L-Nutra, the company that sells the ProLon kit, whose founder Valter Longo holds a financial stake.

This review examines what cyclic fasting-mimicking eating actually changes in people already working to optimize health: body fat, blood pressure, and blood-sugar control, together with the fatigue and adherence costs of repeating a five-day restriction several times a year.

Benefits - Risks - Protocol - Conclusion

High-level overviews of the fasting-mimicking diet (FMD), its origins, and how clinicians and researchers frame cyclic low-protein restriction versus water-only fasting.

No dedicated FMD episode or article was found on Huberman Lab; that platform covers time-restricted eating and fasting more broadly. Life Extension Magazine has brief FMD mentions rather than a dedicated overview, so it was not listed.

Grokipedia

  • Fasting-mimicking diet

    A structured encyclopedia page covering protocol, mechanisms, trial claims, safety, and the commercial ProLon kit, useful as a single map of the topic.

Examine

No dedicated Examine.com article for the fasting-mimicking diet was found. Related pages cover intermittent fasting and prolonged fasting, which are different interventions.

ConsumerLab

No dedicated ConsumerLab article for the fasting-mimicking diet was found.

Systematic Reviews

PubMed systematic reviews and meta-analyses that evaluate FMD for cardiometabolic, diabetes, or cancer-related endpoints.

No systematic review dedicated to FMD adverse events, lean-mass loss, or hypoglycemia was found, so the principal-risk side of the literature is unrepresented here.

Mechanism of Action

FMD keeps calories, protein, and sugars low enough that nutrient sensors treat the state as a fast, while plant fats and micronutrients limit frank starvation. Low amino acids and energy quiet mTOR (mechanistic target of rapamycin, a growth switch) and the growth-hormone–IGF-1 axis; PKA (protein kinase A) signaling also falls. Cells shift from glucose toward fat and ketones, and autophagy (recycling of damaged cell parts) rises. In animals, organs shrink during the cycle and stem-cell programs rebound on refeeding—blood-forming, pancreatic, and neural niches have been mapped. Differential stress resistance is the cancer rationale: healthy cells enter a protected, slow-growth state while many tumor cells remain nutrient-greedy and more drug-sensitive. Human trials reproduce the metabolic signature—lower glucose and IGF-1, higher ketones and IGFBP-1 (IGF-binding protein 1)—during the five days, with partial persistence after refeeding. A competing view holds that time-restricted eating and water-only fasts hit overlapping sensors; whether the packaged macronutrient mix is uniquely required versus any very-low-calorie, low-protein plant menu is not settled. FMD is a food protocol, not a drug: it has no plasma half-life, receptor selectivity, tissue-distribution profile, or CYP-metabolized chemistry (CYP enzymes are liver proteins that break down drugs).

Historical Context & Evolution

Prolonged fasting as therapy is old (religious practice, Buchinger-style clinics, early twentieth-century fasting hospitals). Valter Longo’s laboratory, building on yeast and mouse work on calorie restriction and growth-factor signaling, showed that multi-day fasting protected normal cells from chemotherapy while sensitizing tumors. Water-only fasts proved hard to run in oncology and in healthy volunteers. Around 2009–2015 the group formulated a five-day, plant-based, low-protein, high-unsaturated-fat menu to copy fasting’s effects on IGF-1, glucose, ketones, and IGFBP-1 without total starvation. Brandhorst et al., 2015 reported mouse regeneration and lifespan effects plus a human pilot; Wei et al., 2017 was the first randomized trial in generally healthy adults. L-Nutra commercialized the kit as ProLon; the University of Southern California licensed related intellectual property. Longo holds equity in L-Nutra and has stated that his shares are designated for charity—an ongoing financial conflict that runs through much of the literature, because trial food is often supplied by the same company. Independent groups later tested monthly cycles in type 2 diabetes, breast-cancer chemotherapy, and Crohn’s disease. A 2024 international fasting-terminology consensus recognized fasting-mimicking diets as a named class. Opinion has moved from a mouse-only curiosity to a feasible cyclic intervention with biomarker effects, while hard outcomes (cancer events, death) remain unproven and commercial entanglement remains a live criticism.

Expected Benefits

High 🟩 🟩 🟩

Reduction in body weight and abdominal fat ⚠️ Conflicted

Five-day cycles reduce weight and abdominal fat, with relative lean mass often holding. Low protein and calories shift fuel toward stored fat. Three monthly cycles cut about 2.6 kg and 4.1 cm of waist (Wei et al., 2017); 12-month diabetes MRI (magnetic resonance imaging) showed visceral-fat loss (Schoonakker et al., 2025). Net reading: waist and visceral fat fall in trials, mainly in diabetes and higher-BMI (body mass index) groups, but pooled weight change across 11 trials was not significant (Mohammadzadeh et al., 2025).

Magnitude: About 2.6 kg body-weight loss and 4.1 cm waist reduction after three monthly cycles versus usual diet; 12-month diabetes MRI showed 38 cm² less abdominal visceral fat versus control.

Lower blood pressure

Systolic and diastolic pressure fall across cycles, more when baseline values are high. Lower insulin, weight, and sodium during the five days are plausible drivers. Three cycles reduced systolic pressure by about 4.5 mmHg versus control (Wei et al., 2017). A 2025 meta-analysis of 11 randomized trials pooled −4.1 mmHg systolic and −2.3 mmHg diastolic (Mohammadzadeh et al., 2025). Hard cardiovascular events have not been measured.

Magnitude: Pooled −4.1 mmHg systolic and −2.3 mmHg diastolic across 11 randomized trials; larger drops when systolic started above 120 mmHg.

Better glycemic control in type 2 diabetes

Monthly cycles improve HbA1c (a three-month blood-sugar average) and reduce glucose-lowering drugs. Lower liver fat and insulin resistance after refeeding are proposed mechanisms. A 12-month primary-care trial in 100 metformin-treated adults found HbA1c down 3.2 mmol/mol and drug burden down 0.3 versus usual care; glycemic management improved in 53% versus 8% (van den Burg et al., 2024). A cardiovascular-risk meta-analysis also pooled an HbA1c drop (Mohammadzadeh et al., 2025). Insulin-treated groups were largely excluded.

Magnitude: −3.2 mmol/mol HbA1c and −0.3 diabetes-drug burden (how many drugs and at what dose) at 12 months versus usual care; 53% versus 8% improved combined glycemic management.

Medium 🟩 🟩

Lower calculated biological age

Secondary analysis of two clinical studies reported a 2.5-year drop in Levine phenotypic age (a blood composite that predicts illness and death), partly independent of weight loss in the authors’ models (Brandhorst et al., 2024). The same research group, with L-Nutra supplying the diet, produced both datasets. Replication by fully independent labs is limited.

Magnitude: 2.5-year median reduction in calculated biological age after three cycles.

Lower MRI liver fat

The same secondary analyses found lower MRI liver fat and better insulin-resistance markers after three FMD cycles (Brandhorst et al., 2024). Liver fat is a metabolic-risk marker, not a hard clinical event. Both datasets came from the Longo group with L-Nutra food, so independent replication is limited.

Magnitude: MRI hepatic fat was lower after three cycles in two secondary analyses; the reports give no pooled cm² or percent-change figure versus control. HOMA-IR (homeostatic model of insulin resistance) improved in the same analyses.

Adjunct to cancer therapy ⚠️ Conflicted

Cyclic FMD is proposed to protect healthy cells and sensitize tumors via lower glucose, insulin, and IGF-1. In DIRECT, radiological response was more common with FMD (odds ratio 3.17, about three times more likely, versus control); pathology improved in finishers but was weaker as randomized, and toxicity did not differ (de Groot et al., 2020). A 101-patient safety study found metabolic and immune shifts without proving tumor-control superiority (Vernieri et al., 2022). Net reading: mixed efficacy signals, not a proven anticancer therapy.

Magnitude: Odds ratio 3.17 (about three times more likely) for radiological complete or partial response in DIRECT; tumor-pathology benefit held in those who finished the diet (per-protocol) rather than in everyone as randomized (intention-to-treat).

Clinical response in mild-to-moderate Crohn’s disease

Three monthly cycles improved clinical response and remission versus continued usual diet. Lower systemic and gut inflammatory mediators are proposed. In a 97-person randomized trial, 69% on FMD versus 44% of controls met clinical response, and fecal calprotectin (a stool marker of gut inflammation) fell (Kulkarni et al., 2026). This is a single open-label trial.

Magnitude: Clinical response 69.2% versus 43.8%; clinical remission 64.6% versus 37.5% after three cycles.

Multiple sclerosis quality-of-life scores

A fasting-mimicking cycle reduced autoimmune damage and remyelinated axons (rebuilt nerve insulation) in a mouse multiple-sclerosis model. A 60-person randomized three-arm pilot reported clinically meaningful quality-of-life gains after one FMD cycle plus a Mediterranean diet, similar to a ketogenic-diet arm (Choi et al., 2016). It remains a single small trial without a larger confirmatory study.

Magnitude: Clinically meaningful HRQOL (health-related quality of life) summary-scale improvements at 3 months in a 60-person randomized pilot; no larger confirmatory effect size exists.

Low 🟩

Lower circulating IGF-1

IGF-1, a growth signal tied observationally to some cancers, falls during FMD cycles. Three cycles cut it about 22 ng/ml versus control (Wei et al., 2017); a 2025 meta-analysis pooled −19 ng/ml (Mohammadzadeh et al., 2025). It is an indirect lab marker, not a validated event surrogate.

Magnitude: About −19 to −22 ng/ml IGF-1 in pooled and trial analyses; about −55 ng/ml when baseline IGF-1 exceeded 225 ng/ml.

Speculative 🟨

Autophagy activation in circulating immune cells

A small randomized pilot reported higher autophagic flux in blood immune cells on ProLon versus usual diet (Espinoza et al., 2025). Flux is a mechanistic marker, not a clinical endpoint; L-Nutra sponsored the trial.

Shift toward a more lymphoid blood profile

Three cycles raised the lymphoid-to-myeloid ratio, a younger-appearing immune blood pattern, in secondary analyses of the original randomized cohort (Brandhorst et al., 2024). No infection or vaccine-response outcomes were shown.

Pancreatic beta-cell regeneration

Mouse FMD cycles regenerated insulin-producing beta cells; human type 1 islets in dishes made more insulin under fasting-like conditions (Cheng et al., 2017). No controlled human trial has shown recovery.

Lifespan extension

Bi-monthly FMD from midlife extended median lifespan and cut tumors in mice (Brandhorst et al., 2015). No human longevity trial exists.

Benefit-Modifying Factors

  • Starting risk-factor levels: Weight, blood pressure, glucose, IGF-1, triglycerides, LDL (low-density lipoprotein) cholesterol, and CRP (C-reactive protein, an inflammation marker) moved most in people already above clinical cutoffs in post-hoc analyses of Wei et al., 2017.

  • Type 2 diabetes on metformin: Twelve-month primary-care cycles improved HbA1c and reduced medication need versus usual care; insulin-treated diabetes was not the studied population (van den Burg et al., 2024).

  • Sex: Both sexes appear in the major trials. Dropouts in the original program were disproportionately female, so completed-case estimates may over-represent men who tolerated the kit.

  • Age: Effects on blood pressure, glucose, and calculated biological age were measured in adults 18–70; older, higher-risk participants drove more of the biomarker change. Frail adults were excluded.

  • Growth-pathway genetics: People with lower growth-hormone/IGF-1 signaling (FOXO3 (a stress-response gene) or rare GHR (growth-hormone-receptor) variants) already sit on the pathway FMD suppresses; extra IGF-1 drop may be smaller. No FMD-specific genotype-dosing rule exists.

  • Between-cycle diet: A pescatarian, 12-hour time-restricted pattern (Longo’s “longevity diet”) is the usual refeeding template; a high-protein refeed can reverse the IGF-1 dip faster.

Potential Risks & Side Effects

High 🟥 🟥 🟥

Transient fatigue, weakness, and headache

The most common trial adverse events are fatigue, weakness, and headache during the five days. They are usually grade 1–2, reverse with refeeding, and track the calorie drop. Across the original randomized program, no grade 3 or higher events were reported; 54–100% of participants reported no given event depending on the symptom (Wei et al., 2017). Similar mild effects recur in diabetes and oncology feasibility cohorts.

Magnitude: Predominantly grade 1–2 fatigue, weakness, and headache; no grade ≥3 events in the 100-person randomized program.

Medium 🟥 🟥

Transient absolute lean-mass loss

Absolute lean mass dips by cycle end and can remain slightly lower 5–7 days after refeeding, while relative lean percentage often rises because fat falls more (Wei et al., 2017). Force production did not drop in healthy young men after three cycles (Nardon et al., 2022). In type 2 diabetes, 12-month monthly cycles left abdominal muscle area unchanged on MRI (Schoonakker et al., 2025). Sarcopenic older adults (those with age-related muscle loss) are under-studied.

Magnitude: Absolute lean mass fell in the randomized FMD arm (P = 0.004 (the probability of this difference if there were no real effect)) while relative lean percentage did not; 12-month MRI abdominal muscle area −1.6 cm² versus control (not significant).

Incomplete adherence and dropout

About 25% of volunteers left the original three-cycle program versus 10% on a usual diet, often citing schedule, dislike, or nonadherence (Wei et al., 2017). Oncology trials report incomplete cycle completion around chemotherapy (de Groot et al., 2020). Feasibility, not a medical injury, is the practical limiter.

Magnitude: 25% dropout during FMD cycles versus 10% on usual diet; 71% completed three cycles in the combined arms.

Gastrointestinal symptoms

Nausea, bloating, and constipation appear in kit-based cycles, reflecting very-low-calorie, high-fiber soups and a sudden drop in food volume. They are usually mild and time-limited. Systematic rates vary by trial (Wei et al., 2017).

Magnitude: Grade 1–2 gastrointestinal complaints in trial CTCAE (Common Terminology Criteria for Adverse Events) tables; no pooled prevalence across studies.

Hypoglycemia, especially on insulin or sulfonylureas

People on insulin or sulfonylureas (insulin-releasing diabetes drugs) face hypoglycemia during a multi-day calorie cut. In a 101-person oncology FMD cohort, grade 3/4 hypoglycemia occurred in 5% and syncope (fainting) in 1% (Vernieri et al., 2022). The primary-care diabetes trial enrolled metformin-only patients and excluded insulin (van den Burg et al., 2024).

Magnitude: Grade 3/4 hypoglycemia in 5% and syncope in 1% of a 101-person oncology FMD cohort; insulin-treated diabetes was not studied.

Low 🟥

Speculative 🟨

Gallstones after repeated rapid weight-loss cycles

Very-low-calorie diets can raise gallstone risk via gallbladder stasis. FMD trials have not reported gallstones; the basis is mechanistic and from other low-calorie programs, not FMD-specific events.

Relapse or harm in eating-disorder history

A prescribed five-day restriction can cue restriction–binge cycles in susceptible people. Trials excluded eating disorders; the concern is clinical, not a measured FMD adverse-event rate.

Risk-Modifying Factors

  • Diabetes drug class: Metformin-only use was compatible with monthly cycles in primary care. Insulin or sulfonylureas raise hypoglycemia risk and were excluded from that trial (van den Burg et al., 2024).

  • Baseline BMI and frailty: Low BMI and sarcopenia leave less fat buffer; absolute lean-mass dip matters more. Trials used BMI floors around 18–27 kg/m² depending on the protocol.

  • Sex: Completed-case cohorts in the original program were less female than the enrolled set; women may be more likely to stop a cycle, which can bias apparent safety.

  • Age: Older adults have less physiologic reserve for a 700 kcal day. Participants over 70 and frail patients were generally not enrolled.

  • Oncology status: Cachexia (disease-related wasting), BMI near 18, or poor intake make a five-day deficit riskier; DIRECT required BMI over 18 and no diabetes (de Groot et al., 2020).

  • Growth-pathway genetics: Rare growth-hormone insensitivity already suppresses IGF-1; stacking FMD is biologically redundant and untested. No actionable pharmacogene panel exists.

Key Interactions & Contraindications

  • Insulin and sulfonylureas (glipizide, glyburide): Absolute caution. Multi-day calorie and carbohydrate cuts can cause severe hypoglycemia; the studied diabetes FMD program excluded these drugs. Severity: contraindication without specialist supervision and dose holds.

  • Other antihypertensives (ACE (angiotensin-converting enzyme) inhibitors such as lisinopril, ARBs (angiotensin receptor blockers) such as losartan, diuretics): Caution. ACE inhibitors and ARBs plus lower sodium can cause lightheadedness. Severity: monitor; temporary dose reduction is common.

  • GLP-1 (glucagon-like peptide-1) receptor agonists (semaglutide, tirzepatide; drugs that mimic a gut satiety hormone): Caution. Stacked calorie deficit, nausea, and lean-mass loss can compound. Severity: monitor intake and muscle; spacing cycles away from dose increases is typical.

  • Warfarin: Caution. A sudden shift to packaged plant soups changes vitamin K intake and can move INR (international normalized ratio, a clotting-time measure). Severity: extra INR checks around the cycle.

  • Chemotherapy plus high-dose dexamethasone: Monitor. DIRECT omitted dexamethasone on FMD days without increasing recorded toxicity; this is protocol-specific, not a general license to drop steroids (de Groot et al., 2020).

  • Protein supplements, collagen, and BCAAs (branched-chain amino acids): They blunt the low-amino-acid signal that FMD uses to drop IGF-1 and mTOR. Severity: they cancel the intended mechanism if used on cycle days.

  • Electrolyte and magnesium supplements: Potentially additive and useful against headache and cramps on 700 kcal days. Severity: supportive, not a drug interaction.

  • NSAIDs (non-steroidal anti-inflammatory drugs; ibuprofen, naproxen) and acetaminophen: Caution. A 700 kcal day can worsen NSAID gastric irritation. Severity: take with remaining food; acetaminophen is the usual headache alternative if the liver is healthy.

Populations who should avoid Fasting-Mimicking Diet:

  • Pregnancy or breastfeeding
  • BMI below about 18.5 kg/m², active eating disorder, or recent bariatric surgery
  • Type 1 diabetes, or type 2 diabetes on insulin or sulfonylureas without close medical supervision
  • Advanced frailty, recent myocardial infarction (<90 days), or unstable heart failure (NYHA Class III–IV (a severe symptom-based heart-failure grade))
  • Children and adolescents still growing
  • Known allergy to nuts, soy, or other kit ingredients when using a commercial box

Risk Mitigation Strategies

  • Screen before cycle one: Trial protocols confirm BMI, diabetes-drug class, pregnancy status, and eating-disorder history so hypoglycemia, underweight, and restriction-binge risk are not first seen mid-cycle.

  • Insulin and sulfonylurea holds: Dose holds coordinated with the prescribing clinician before day one reduce hypoglycemia risk; metformin-only trials did not cover insulin-treated fasting.

  • Electrolytes and ≥2 L fluid: Sodium, potassium, and magnesium plus water reduce headache, cramps, and orthostatic drops (blood pressure falling on standing) on 700 kcal days.

  • Resistance work off-cycle: Two to three weekly strength sessions on feeding days offset the absolute lean-mass dip seen in Wei et al., 2017.

  • Slow refeed (1–3 days): Small, low-fiber, moderate-protein meals after day five limit bloating and the rebound overeating that erases fat loss.

  • Stop rules: Studied programs stop a cycle for persistent dizziness, syncope (fainting), glucose under about 70 mg/dL with symptoms, or inability to drink.

  • Stacked prolonged fasts: Consecutive water fasts plus FMD extend the calorie deficit and raise untested hypoglycemia, lean-mass, and dropout risk without extra trial safety data.

Therapeutic Protocol

  • Standard five-day cycle: Day 1 about 1,090–1,100 kcal (roughly 10% protein, 45–55% fat, 35–45% carbohydrate); days 2–5 about 700–800 kcal with ~9% protein, plant-based, low sugar, high unsaturated fat (Brandhorst et al., 2015).

  • Frequency used in trials: Three consecutive monthly cycles, then every 3–4 months for maintenance in generally healthy adults; monthly for 12 months in the metformin-treated diabetes program (van den Burg et al., 2024).

  • Commercial kit versus grocery version: ProLon is the studied boxed menu (L-Nutra). Attia and others have used self-assembled 700 kcal low-protein days; those variants are less standardized and less present in randomized trials.

  • Time of day: The commercial kit spreads soups, bars, and olives across the day. Some practitioners take the day’s calories as one evening meal; neither schedule is proven superior for IGF-1 or ketones.

  • Not a drug (no half-life): Fasting markers (glucose, ketones, IGF-1) typically reverse within 1–3 days of refeeding; there is no plasma half-life, CYP pathway (liver enzymes that break down drugs), or receptor occupancy to dose against.

  • Split versus single eating: Kit instructions split intake; collapsing calories into one sitting is a comfort choice, not an efficacy rule, and was not the trial method.

  • Genetics: No FMD-specific pharmacogene changes the five-day calorie targets. FOXO3 (a stress-response gene) and growth-hormone/IGF-1 pathway variants may alter how much IGF-1 can still fall.

  • Sex: The same calorie boxes are issued to both sexes in commercial kits; larger bodies may be relatively more restricted. Menstrual disruption is possible during the deficit week.

  • Age: Adults 18–70 dominate the trials. Older users often start with fewer annual cycles and closer blood-pressure and glucose checks.

  • Baseline biomarkers: High BMI, blood pressure, glucose, IGF-1, or CRP predicted larger moves in Wei et al., 2017; people already in the healthy band gained less.

  • Pre-existing conditions: Type 2 diabetes on metformin, overweight hypertension, and mild Crohn’s were the disease groups with dedicated randomized data. Cachexia and type 1 diabetes were not.

  • Competing approaches: Water-only fasts (TrueNorth-style) and Buchinger fasting are the main clinical alternatives; a Mediterranean diet matched some cardiometabolic changes in one FMD comparison trial (Mishra et al., 2023) without being framed here as the default.

Discontinuation & Cycling

  • Cyclic, not a daily lifelong diet: FMD is a five-day pulse. Between cycles, trials returned people to their usual or a pescatarian pattern; it is not meant as a permanent 700 kcal menu.

  • No classic drug withdrawal: Stopping FMD does not produce a pharmacologic withdrawal syndrome. Hunger and fatigue end with refeeding.

  • Refeed instead of a taper: Calories step up over 1–3 days (soups and cooked vegetables, then legumes and fish or eggs) rather than tapering a milligram dose.

  • Cycling is the method: Repeating cycles is how trials maintained weight, blood pressure, and HbA1c effects; a single cycle’s changes partly fade over months (Wei et al., 2017).

  • When not to repeat: New pregnancy, falling BMI, progressive lean-mass loss, or escalating diabetes drugs are practical stop signals used in clinic protocols.

Sourcing and Quality

  • What the trials used: Most randomized trials used L-Nutra kits (ProLon or research-identical boxes): soups, nut bars, olives, teas, and a micronutrient drink. The University of Southern California licensed related intellectual property; Longo holds L-Nutra equity.

  • Matching the studied macros: Day 1 ~1,100 kcal and days 2–5 ~700–800 kcal, ~9–11% protein, low sugar, high unsaturated fat, plant-based. Grocery versions that miss the protein cap may not copy IGF-1 or ketone effects.

  • Allergen and additive check: Commercial boxes commonly contain nuts, soy, and sesame. Allergy to those ingredients rules out the stock kit; clinic grocery versions have been used instead.

  • Third-party testing: This is a food protocol, not a supplement. Lot labeling and the same calorie/protein math as published trials matter more than a USP (United States Pharmacopeia) seal.

  • Cost versus DIY: A branded five-day box typically costs on the order of $175–250. A grocery cycle can be far cheaper but is not the formulation in most randomized controlled trials (RCTs).

Practical Considerations

  • Time to effect: Weight, ketones, and blood pressure move during the five days; waist, IGF-1, and HbA1c shifts are scored after three monthly cycles or at 12 months in diabetes care.

  • Common pitfalls: Eating extra protein or snacks on cycle days; stacking a water fast; skipping electrolytes; bingeing on refeed day one; treating FMD as a weekly habit rather than a pulse.

  • Regulatory status: ProLon is marketed as food, not as an FDA (U.S. Food and Drug Administration)-approved drug or disease treatment. Oncology and diabetes uses in trials are experimental relative to labeled food status.

  • Cost and access: Boxed cycles (~$175–250) are rarely insured; payers have little incentive to favor the kit over usual diet or a grocery version, a structural bias in coverage and guidelines.

  • Conflict of interest in the evidence: L-Nutra supplied food in many trials; the protocol’s inventor holds company equity. Independent replication still exists (van den Burg et al., 2024, de Groot et al., 2020, Kulkarni et al., 2026) but does not erase the conflict.

Interaction with Foundational Habits

  • Sleep: Direct, mixed. Hunger and nocturia (waking at night to urinate) from extra fluids can fragment sleep on cycle nights; some people report deeper sleep once ketones rise. Clinic protocols often move caffeine earlier and keep the bedroom cool; new sleep drugs are not part of the five-day trial methods.

  • Nutrition: Direct and defining. FMD is itself a nutrition protocol. Between cycles, adequate protein and micronutrients rebuild lean mass; a chronic low-protein diet plus FMD stacks IGF-1 suppression with sarcopenia risk. Refeed with protein and plants, not ultra-processed surplus.

  • Exercise: Direct, time-dependent. Heavy lifting on 700 kcal days is poorly tolerated and was not the trial model. Strength work on feeding days kept force and abdominal muscle in Nardon et al., 2022 and Schoonakker et al., 2025. Easy Zone 2 walking (conversation still possible) is the usual on-cycle compromise.

  • Stress management: Indirect. Cortisol and perceived stress can rise with hunger; people with high baseline anxiety drop out more. Short walks, light social meals (even if the plate is small), and not stacking work travel with day 3–4 are the practical buffers.

Monitoring Protocol & Defining Success

Baseline testing before cycle one establishes starting BMI, blood pressure, fasting glucose, HbA1c, lipids, IGF-1, high-sensitivity CRP, and a basic metabolic panel so later changes have a personal reference. People with diabetes add a home glucose plan. Dual-energy X-ray absorptiometry (DEXA) or another body-composition method is useful when lean-mass protection is a goal. Ongoing monitoring is at the end of the first cycle (or 5–7 days after refeed), after three monthly cycles, then every 6–12 months if cycling continues — matching the timepoints used in Wei et al., 2017 and the 12-month diabetes program. Home weight, waist, and blood pressure on day 1 and day 6 of each cycle catch acute drops. Success is a lower waist and blood pressure or HbA1c without a progressive lean-mass slide, not a lower scale weight at any cost.

Biomarker Optimal Functional Range Why Measure It? Context/Notes
Fasting glucose 70–85 mg/dL Tracks the FMD glucose dip and diabetes risk Conventional lab “normal” often goes to 99 mg/dL; draw fasting, not during day 3 if symptomatic
HbA1c 4.8–5.3% Three-month glycemic average Conventional diabetes cutoff 6.5%; recheck after ≥3 cycles, not after one week
IGF-1 Track change from own baseline (no single longevity target) Confirms the growth-signal drop FMD is built to cause Draw 5–7 days after refeed to match trial timing; high baseline (>~200–225 ng/mL) predicted larger drops
Fasting insulin / HOMA-IR Insulin 2–6 μIU/mL; HOMA-IR <1.5 Insulin resistance improved in secondary analyses Fasting draw; pair with glucose
LDL cholesterol <100 mg/dL functional; <70 mg/dL if high ASCVD risk Lipid shifts were inconsistent across trials Conventional <130 mg/dL is looser; check fasting. ASCVD = atherosclerotic cardiovascular disease.
Triglycerides <80–100 mg/dL Fell more when baseline was high Fasting draw; sensitive to the refeed meal
hs-CRP <0.5–1.0 mg/L Inflammation marker that fell in at-risk subjects hs-CRP = high-sensitivity C-reactive protein. Conventional <3 mg/L; avoid drawing during acute illness.
eGFR / creatinine >90 mL/min/1.73 m² when possible Safety during a low-calorie week eGFR = estimated glomerular filtration rate (kidney-filter score). Conventional staging uses 60; hydration affects creatinine.
Electrolytes (Na, K, Mg) Stay in lab normal; Mg often targeted ~2.0–2.3 mg/dL Headache and cramp prevention Recheck if on diuretics
Body composition (DEXA or MRI) Fat mass down, lean mass stable versus own baseline Distinguishes fat loss from muscle loss Wei used DXA (dual-energy X-ray absorptiometry); diabetes trial used abdominal MRI
  • Energy and hunger on days 3–5
  • Sleep continuity during the cycle
  • Strength in usual lifts in the feeding weeks
  • Dizziness on standing
  • Mood and any restriction–binge cues
  • For diabetes: home glucose logs, especially mornings
  • If DEXA is unavailable: waist circumference and a strength log

Emerging Research

  • Prostate cancer FMD trial: NCT05832086 is a recruiting phase 2 study (target 138) of intermittent FMD for prostate-cancer control and metabolic outcomes (Freedland). Positive results would strengthen the oncology adjunct case; null results would weaken it.

  • Large colorectal FMD cohort: NCT05384444 (Fudan, target 602, active) tests FMD in colorectal cancer. Size makes it able to move the cancer-adjunct claim either way.

  • NSCLC (non-small cell lung cancer) plus immunotherapy: NCT06671613 (U.S. Department of Veterans Affairs, recruiting, n=66) pairs FMD with checkpoint inhibitors. Could support or contradict the immune-remodeling story from Vernieri et al., 2022.

  • Longevity-diet follow-up: NCT07255300 follows the Varapodio longevity and FMD program for age-correlated risk factors (Fondazione Valter Longo). Sponsor overlap with the FMD inventor is a bias risk if biomarkers move.

  • Autophagy and aging pilots: NCT06115551 and related Cedars-Sinai work (Espinoza et al., 2025) ask whether FMD raises human autophagic flux. Confirmation would support a cellular-repair claim; failure would leave autophagy mouse-only.

  • What could weaken the case: No completed trial has shown fewer heart attacks, cancer deaths, or longer life. DIRECT’s pathology result was weaker when everyone as randomized was counted than among finishers (de Groot et al., 2020). Ongoing L-Nutra-linked programs cannot, by themselves, erase that conflict.

Conclusion

The fasting-mimicking diet is a cyclic, five-day, low-calorie plant pattern built to copy a prolonged fast’s metabolic signature without total food withdrawal. In people who accept inconvenient protocols, the most consistent human findings are fat loss with relative muscle preservation, lower blood pressure, and better blood-sugar control in type 2 diabetes—especially when starting numbers are high. Calculated biological-age and liver-fat signals move favorably when the same trial bloodwork is re-examined, and one trial in mild Crohn’s disease showed symptom and gut-inflammation gains. Use as a companion to cancer treatment remains mixed: metabolic and immune shifts appear; proof that tumors shrink more in routine care does not.

The literature is unusually concentrated. Many cycles used food from L-Nutra, which sells the boxed ProLon program, and Valter Longo, who designed the diet, holds a financial interest in that company. Mild fatigue, headache, and weakness during the five days are common. Absolute lean mass can dip and then recover. People on insulin or certain diabetes drugs face a blood-sugar-too-low risk that the main diabetes study did not test. Underweight, pregnancy, eating disorders, and frailty sit outside the studied safety envelope.

For a health-optimizing adult, the evidence describes a periodic tool with real blood-marker movement, not a proven lifespan therapy. The boxed product is expensive and rarely covered; insurers have little reason to prefer it over usual eating. A grocery version can match calorie and protein targets but is less studied. Neither replaces sleep, protein-adequate feeding between cycles, or resistance training.

Top - Benefits - Risks - Protocol