Does soursop help with weight loss?

Nobody knows, because as far as we could find, nobody has measured it. We found no randomized trial and no systematic review of soursop and body weight in people. The largest soursop trial gave 143 adults with prehypertension 7.1 oz (200 g) of soursop juice a day for 3 months and measured blood pressure and uric acid, and its abstract reports no weight result. The nearest randomized evidence is about fruit and vegetables in general. Across 8 trials in 1026 people, eating more of them, with no advice to cut other foods, came with 1.2 lb (0.54 kg) less body weight over about 15 weeks.

Unsettled. Soursop for weight loss is an untested claim. The only signal in its favor is a laboratory experiment with rat gut bacteria and rat fat tissue, which has never been repeated in a person.

Not for everyone. Soursop supplements, the capsules and concentrates often sold for weight loss, are a separate question from the fruit. A risk assessment published in the EFSA Journal found strong indications that some soursop preparations may be toxic to nerves and said a safe intake level cannot yet be set.

What the trials found

Soursop itself

Two randomized trials have given soursop to people, and neither was about weight. In the juice trial in Indonesia, the comparison group got nothing at all, and the full text is behind a paywall, so we do not know whether body weight was recorded. The only placebo-controlled trial gave a soursop leaf extract or placebo to 30 people with colorectal cancer whose tumor had been removed, 28 of whom completed 8 weeks. It measured whether their blood killed cancer cells in a dish, and its abstract gives no weight figures.

The one result that sounds relevant is laboratory work. Soursop fruit and peel were fermented with rat gut bacteria, and the products eased the changes that fructose caused in rat fat tissue, more so at higher doses. That is a test tube and tissue result with no body weight in it, and no person in it.

Fruit in general

Two meta-analyses of randomized trials ask what happens when people eat more fruit and vegetables. A 2014 review in the American Journal of Clinical Nutrition looked for trials that advised or delivered more produce for at least 8 weeks. Its main estimate of weight change was a standardized difference of -0.16, with an interval from -0.78 to 0.46, which is not significant. Across 7 trials the figure was 0.04.

A second 2014 review, of 8 trials in 1026 people, found weight 1.2 lb (0.54 kg) lower in the high intake arms, with an interval from 2.3 lb (1.05 kg) to 0.09 lb (0.04 kg) and a lot of disagreement between the trials. Energy intake in those arms was 368 kJ higher, a difference that was not significant. Read together, extra fruit does little to body weight in either direction, and soursop has no data of its own to set it apart.

What a serving holds

The only soursop entry in the US food survey database is the nectar, a sweetened drink, and there is no entry for the raw fruit. A 248 g cup of nectar has about 146 calories and 32.4 g of total sugars, or 59 calories and 13.07 g per 3.5 oz (100 g). A daily glass of that adds energy and sugar to a diet.

Who should be careful

The EFSA Journal assessment covered soursop food supplements sold in the European Union. It found substantial uncertainty about their safe use and a lack of long-term studies. It also records that human studies usable for judging safety do not exist, which is why no safe amount can be named. That is a missing answer: it is not proof of harm, and it is not proof of safety either.

The worry comes from annonacin, a compound in the plant. Laboratory and animal data suggest it reaches the brain: given to rats intravenously over 28 days, it built up in brain tissue, lowered brain energy levels and damaged nerve cells. Injection skips digestion, so how much reaches the brain from eating soursop is unknown. On Guadeloupe, observational data from 160 people with parkinsonism found that those with two unusual forms of it had eaten significantly more soursop than controls or people with ordinary Parkinson's disease. That is an association in people already ill, and it cannot show that soursop caused the disease.

Put together, years of daily soursop or soursop supplements in the hope of losing weight has no proven benefit and an unresolved question about the nervous system. The same assessment records that interactions between soursop preparations and medicines have not been reliably studied, so nobody can say whether they are safe to combine. We found nothing specific on pregnancy or children.

What expert bodies say

No health agency has said anything about soursop and weight that we could find. The closest document is the risk assessment on supplements published in the EFSA Journal, which says a safe intake level cannot be established. It is about safety, not weight.

How we searched

What would change this answer

• A randomized trial of soursop fruit eaten in place of another fruit, with body weight as the outcome. We found none registered or published.
• The full text of the Indonesian juice trial, if it recorded body weight over the 3 months.
• Long-term human safety data on soursop supplements, which is what EFSA says is missing.

More on the fruit, its cancer claims and what has been measured in people, is on the soursop page.

Sources

1. ev-ssw-01 · Meta-analysis or systematic review · systematic review and meta-analysis of randomized controlled trials · n = 2 studies met all criteria; 7 met all or most
   The primary analysis indicated an effect size of weight change (outcome of interest) from baseline [standardized mean difference (SMD) for studies that met all criteria] of -0.16 (95% CI: -0.78, 0.46) (P = 0.60).
   
   Who: human randomized trials of at least 8 weeks with at least 15 subjects per arm, aimed at weight or fat loss through more minimally processed fruit and vegetables
   Effect: weight change SMD -0.16 (95% CI -0.78 to 0.46), P = 0.60; 7 studies SMD 0.04 (-0.10 to 0.17)
   Certainty: About fruits and vegetables in general; soursop is in none of the RCTs. This is the closest synthesis to the question "will I lose weight if I eat fruit".
   Am J Clin Nutr, 2014 · checked 2026-10-07 · we read the abstract
2. ev-ssw-02 · Meta-analysis or systematic review · systematic review and meta-analysis of randomized controlled trials, random effects · n = 8 RCTs, 1026 participants
   The mean change in body weight was -0.54kg (95% CI: -1.05 to -0.04; n=8; I2 for heterogeneity = 73%, p<0.01) less in the ‘high vegetable and fruit’ intake arms than in the ‘low vegetable and fruit intake’ arms.
   
   Who: randomized trials increasing vegetable and fruit intake without advice to remove other foods; mean extra intake 133 g per day, mean duration 14.7 weeks
   Effect: body weight -1.2 lb (95% CI -2.3 to -0.09; metric: -0.54 kg, -1.05 to -0.04), I2 73%; energy intake +368 kJ (95% CI -27 to 762), not significant
   Certainty: High heterogeneity; the effect is small and not about soursop; supports the thesis "fruit does not harm weight", not "fruit burns fat".
   BMC Public Health, 2014 · checked 2026-10-07 · we read the abstract
3. ev-ssw-03 · Randomized controlled trial(s) · randomized controlled trial, untreated control, open · n = 143
   For a 3-month period, the treatment group was given 2 × 100 g soursop fruit juice per day and the control group was not treated.
   
   Who: adults with essential prehypertension and high-normal serum uric acid from the Mlati Study Database, Indonesia; 2 x 100 g soursop fruit juice daily vs no treatment for 3 months
   Effect: lower mean systolic blood pressure vs untreated control over baseline, week 6 and week 12; weight not reported as an outcome in the abstract
   Certainty: Untreated control, not placebo; full text outside PMC (no_pmc), so whether weight was measured is unknown. Card on the absence of a weight endpoint.
   J Hum Hypertens, 2020 · checked 2026-10-07 · we read the abstract
4. ev-ssw-04 · Randomized controlled trial(s) · randomized double-blind placebo-controlled pre-post trial · n = 30 enrolled, 28 completed
   Thirty outpatients with colorectal cancer who had undergone primary tumor resection were enrolled in a randomized double-blind placebo- controlled pre-post-trial.
   
   Who: outpatients with colorectal cancer after primary tumor resection; A. muricata leaf extract vs placebo daily for 8 weeks
   Effect: higher ex vivo serum cytotoxicity in the extract group; nutritional status monitored but no weight figures in the abstract
   Certainty: Leaf extract, not the fruit; cancer patients; weight as an outcome is not named in the abstract.
   Asia Pac J Clin Nutr, 2017 · checked 2026-10-07 · we read the abstract
5. ev-ssw-05 · Laboratory or animal data · in vitro fermentation and ex vivo tissue model · n = —
   Treatment with soursop-enriched fecal metabolites significantly reversed these alterations in a dose-dependent manner.
   
   Who: rat fecal microbiota fermentation of soursop fruit and peel; ex vivo rat perigonadal white adipose tissue exposed to fructose
   Effect: soursop-derived fecal metabolites reversed fructose-induced changes in glucose metabolism, redox and lipid pathways in a dose-dependent manner (ex vivo)
   Certainty: Mechanistic (D): neither humans nor body mass. In the abstract the authors call soursop a potential functional food; this is a supposition.
   Antioxidants (Basel), 2026 · checked 2026-10-07 · we read the abstract
6. ev-ssw-06 · Position of an expert body · risk assessment of Annona muricata in food supplements (EU-FORA, EFSA Journal) · n = —
   The results of the assessment show that substantial uncertainties exist regarding the safe use of A. muricata-based food supplements. The present data provide strong indications of neurotoxic potential of certain A. muricata preparations. However, the paucity of adequate studies, particularly related to long-term use of A. muricata supplements, currently does not allow the establishment of a safe intake level.
   
   Who: consumers of Annona muricata food supplements in the European Union
   Effect: no safe intake level can be established; strong indications of neurotoxic potential of certain preparations; long-term data lacking
   Certainty: Concerns supplements and concentrates (often it is these that are sold "for weight loss"), not the fruit as food. Does not mean demonstrated harm: there is no data for a limit.
   EFSA Journal (EU-FORA fellowship report), 2020 · checked 2026-10-07 · we read the fulltext
7. ev-ssw-07 · Observational data · cross-sectional study · n = 160
   Consumption of soursop was significantly greater in both PSP-like and PDC patients than in controls and Parkinson's disease patients.
   
   Who: 160 parkinsonian patients in Guadeloupe compared with controls
   Effect: soursop consumption significantly greater in PSP-like and parkinsonism-dementia patients than in controls and Parkinson's disease patients
   Certainty: Association, not causality; the mechanism is annonacin (a mitochondrial complex I inhibitor). A reason not to eat a lot of soursop daily for many years "for weight loss".
   Brain, 2007 · checked 2026-10-07 · we read the abstract
8. ev-ssw-08 · Position of an expert body · food composition database · n = —
   Soursop, nectar (FNDDS 64221010): Energy 59 kcal per 100 g, 146.3 kcal per 248 g (quantity not specified); Total Sugars 13.07 g per 100 g, 32.4 g per 248 g.
   
   Who: USDA FNDDS food composition data
   Effect: per 3.5 oz (100 g): energy 59 calories, total sugars 13.07 g; per 248 g portion: 146.3 calories, 32.4 g sugars
   Certainty: FNDDS has only nectar, no raw fruit; nectar is a sweetened drink, which is the opposite of "for weight loss".
   USDA FNDDS, food code 64221010, Soursop, nectar (fdc_id 2709350) · checked 2026-10-07 · we read the dataset
9. srs-f-04 · Position of an expert body · risk assessment following the BfR and EFSA botanical guidance, with data arranged by composition, human studies, animal toxicology, in vitro studies and interactions · n = literature searched across PubMed, Medline, Scopus, Google Scholar and Web of Science to mid-March 2020, including gray literature
   Regarding the use of A. muricata in food supplements, scientific information from human intervention studies, which might be used for safety evaluation of A. muricata, is currently lacking. This represents a major data gap that impedes risk assessment, especially in the case when long-term use of high supplemental A. muricata doses is intended...There is also a lack of reliable in vitro and in vivo studies with respect to potential interactions between A. muricata preparations and conventional drugs.
   
   Who: Annona muricata food supplement products
   Effect: the assessment records that scientific information from human intervention studies usable for safety evaluation is currently lacking, calls this a major data gap that impedes risk assessment particularly for long-term use at high supplemental doses, and separately records a lack of reliable in vitro and in vivo studies on interactions between Annona muricata preparations and conventional drugs
   Certainty: this is axis 9 in its purest form: the authority does not evade, but names exactly what is lacking. The second half of the quote matters most in practice: 80.7% of cancer patients in the Trinidad survey use graviola (card srs-f-08), and almost all of them receive chemotherapy, while there are no data on interaction at all. The absence of interaction data does NOT mean there is no interaction: it means no one has looked for it
   EFSA Journal (EU-FORA fellowship report), 2020 · checked 2026-09-21 · we read the fulltext
10. srs-f-07 · Laboratory or animal data · animal toxicology evidence summarized in the EFSA risk assessment, from intravenous administration of annonacin to rats over 28 days · n = animal studies only, no human toxicokinetic data identified
    For instance, after i.v. application of annonacin over 28 days to rats, annonacin accumulated in the brain parenchyma, decreased brain ATP levels, induced neuropathological abnormalities in the basal ganglia and loss of nigral and striatal neurons in exposed animals (Champy et al., 2003).
    
    Who: rats, plus wild-type and human tau transgenic mice in a separate one-year oral study of soursop fruit juice
    Effect: after intravenous annonacin over 28 days, annonacin accumulated in the brain parenchyma, decreased brain ATP levels, induced neuropathological abnormalities in the basal ganglia and caused loss of nigral and striatal neurons; a separate one-year oral exposure to soursop fruit juice increased numbers of neurons with phosphorylated tau protein in several brain regions in mice
    Certainty: level D, and the limits here are strict: annonacin was administered INTRAVENOUSLY, that is, bypassing all digestion, and how much of it reaches the blood after eating the fruit nobody has measured: EFSA says directly that there are no in vivo toxicokinetic data. So this card is about WHAT the substance does once it is in the blood, not about how much of it gets there from food. The mouse study on JUICE is oral and one year long, and it is closer to everyday life, but these are mice. Together with the empty axis 2 in humans (none of the five mandatory queries of the constitution gave anything), this is our entire state of knowledge about absorption
    EFSA Journal (EU-FORA fellowship report), 2020 · checked 2026-09-21 · we read the fulltext