Tongkat Ali Raised Total Testosterone by 24 ng/dL in 105 Men Over 50 and by Nothing in 149 Healthy Men. The Meta-Analysis That Rates It Best Found 87 Percent Heterogeneity, and Europe's Food Safety Regulator Concluded Its Safety Has Not Been Established.
Tongkat ali sits in an odd position among the herbs sold to men for testosterone. The pooled effect size in its only meta-analysis is the largest in the category, several times what ashwagandha or fenugreek can show. It is also the only one of those herbs that a major food-safety regulator has looked at in detail and declined to approve, not because it failed to work but because the genotoxicity file had positive results the applicant could not explain away. Both facts are true at once, and any honest reading of tongkat ali has to hold them together. This article walks through the nine human studies that measured testosterone in men, spends most of its time on the one trial large enough to matter, explains why an effect size of 1.35 from five small trials is a warning rather than a triumph, lays out the European Food Safety Authority's 2021 conclusion in its own words, and finishes with the least discussed problem of all: surveys of what is actually inside the capsules sold under the name.
Key Takeaways: What the Tongkat Ali Evidence Actually Shows
Tongkat ali is the root of Eurycoma longifolia, a Southeast Asian tree, and its standardized water extract is sold as Physta or LJ100 by Biotropics Malaysia Berhad, which funded or co-authored seven of the nine human studies that have measured testosterone in men. The largest trial, published by Chinnappan and colleagues in Food and Nutrition Research in 2021, randomized 105 men aged 50 to 70 with total testosterone under 300 ng/dL to 100 mg, 200 mg or placebo daily for 12 weeks; total testosterone rose from 200.5 to 225.0 ng/dL on 200 mg, from 187.3 to 203.8 on 100 mg, and fell from 183.0 to 177.9 on placebo, with the 200 mg difference significant at p under 0.001. Free testosterone showed no significant difference between groups in that trial, and sex hormone binding globulin did not change in six of the nine studies. Two randomized trials in men with normal testosterone, Ismail 2012 in 109 men and George 2013 in 40 men, both at 300 mg for 12 weeks, found no difference from placebo. A 2022 meta-analysis of five randomized trials and 232 men by Leisegang and colleagues reported a standardized mean difference of 1.352 for total testosterone with a confidence interval of 0.565 to 2.138, heterogeneity of 87 percent, and a Begg's test for publication bias at p equals 0.024. The 2013 Talbott study reported salivary cortisol 16 percent lower and salivary testosterone 37 percent higher than placebo after 4 weeks at 200 mg, in a mixed cohort of 32 men and 31 women. On 27 October 2021 the European Food Safety Authority adopted an opinion concluding that the safety of the Physta extract had not been established under any condition of use, citing a positive in vitro chromosome aberration test and a positive in vivo comet assay in rat stomach and duodenum at 2,000 mg per kg. A 2006 survey of 100 Malaysian tongkat ali products found 26 contained mercury at 0.53 to 2.35 ppm, above the national limit of 0.5 ppm, and a 2018 DNA barcoding study found only 37 percent of tested products were authentic.
Nine Studies, About 480 Men, Four Placebo-Beating Results and Two Clean Nulls
Men enrolled in each human study of tongkat ali that measured serum testosterone, as catalogued by the Leisegang 2022 systematic review. Green marks studies where total testosterone rose relative to a placebo group with a reported p value: Lim 2017, Quin 2021 at two weeks, Leitão 2021 and Chinnappan 2021. Amber marks studies where testosterone rose only against baseline, either because there was no placebo group (Tambi 2012, Henkel 2014) or because no between-group test was reported (Chan 2021). Grey marks the two trials in men with normal testosterone that found no difference from placebo: Ismail 2012 and George 2013, both at 300 mg for 12 weeks. Seven of the nine used the Physta extract from Biotropics Malaysia, and two of the nine were unpublished doctoral dissertations obtained by contacting the authors. Quin's eight-week comparison, not shown separately, found the two-week rise had disappeared.
What is tongkat ali and why would it affect testosterone?
Tongkat ali is a traditional Malay tonic made from the root of Eurycoma longifolia, and its proposed effect on testosterone runs through a bitter quassinoid called eurycomanone that appears to act on the testis rather than on the hormone already in the blood. The plant is a slow-growing tree of the Simaroubaceae family found across Malaysia, Indonesia, Thailand and Vietnam, and the root has been boiled as a remedy for fever, malaria and flagging vigour for long enough that the Malay name translates roughly as Ali's walking stick. The commercial extract is a hot-water preparation standardized, in the version submitted to European regulators, to 40 to 65 percent glycosaponins and 0.8 to 1.5 percent eurycomanone, with smaller amounts of canthin-6-one alkaloids and the coumarin isoscopoletin. Three mechanisms are put forward. In rat Leydig cells eurycomanone increases expression of CYP17, the enzyme that converts progesterone toward androgens, and raises testosterone output in culture. In vitro it inhibits aromatase, the enzyme that converts testosterone to estradiol, which would in principle leave more testosterone in circulation. The third and most heavily marketed idea is that tongkat ali releases testosterone from sex hormone binding globulin, raising the free fraction. The human trials do not support that third mechanism. Six of the nine studies that measured testosterone also measured SHBG, and none found a change relative to placebo. If tongkat ali does anything in men, the data point to the testis making a little more hormone, not to the hormone being freed from its carrier.
Nine Studies, About 480 Men, Four Placebo-Beating Results and Two Clean Nulls
Men enrolled in each human study of tongkat ali that measured serum testosterone, as catalogued by the Leisegang 2022 systematic review. Green marks studies where total testosterone rose relative to a placebo group with a reported p value: Lim 2017, Quin 2021 at two weeks, Leitão 2021 and Chinnappan 2021. Amber marks studies where testosterone rose only against baseline, either because there was no placebo group (Tambi 2012, Henkel 2014) or because no between-group test was reported (Chan 2021). Grey marks the two trials in men with normal testosterone that found no difference from placebo: Ismail 2012 and George 2013, both at 300 mg for 12 weeks. Seven of the nine used the Physta extract from Biotropics Malaysia, and two of the nine were unpublished doctoral dissertations obtained by contacting the authors. Quin's eight-week comparison, not shown separately, found the two-week rise had disappeared.
Which human studies measured testosterone in men taking tongkat ali?
Nine studies published between 2012 and 2021 measured serum testosterone in men taking tongkat ali, and the 2022 systematic review by Leisegang and colleagues in Medicina is the cleanest inventory of them. Two were uncontrolled before-and-after studies. Tambi 2012 gave 76 men with late-onset hypogonadism 200 mg of Physta for one month and reported total testosterone rising from 163 to 240 ng/dL, with the share of men in the normal range climbing from 35.5 to 90.8 percent, but with no placebo group there is no way to separate the extract from regression to the mean in a cohort selected for low readings. Henkel 2014 followed 13 male cyclists aged 57 to 72 on 400 mg and reported 384 to 442 ng/dL after 5 weeks. Five were double-blind controlled trials in journals: Ismail 2012 in 109 married men aged 30 to 55, George 2013 in 40 healthy men, Chan 2021 in 32 men aged 18 to 30, Leitão 2021 in 45 men with androgen deficiency of the aging male, and Chinnappan 2021 in 105 men aged 50 to 70. Two more were doctoral dissertations, Lim 2017 in 20 trained young men and Quin 2021 in 37 sedentary young men. Seven of the nine used Physta from Biotropics, at doses from 100 to 600 mg a day, for anywhere from 3 days to 6 months. Of those nine, seven reported a within-group rise in total testosterone, but only four reported a rise relative to placebo with a p value attached, and the two largest trials in men with normal testosterone, Ismail and George, found no difference at all. That split, positive in men who started low and null in men who did not, is the single most consistent pattern in the literature.
Nine Studies, About 480 Men, Four Placebo-Beating Results and Two Clean Nulls
Men enrolled in each human study of tongkat ali that measured serum testosterone, as catalogued by the Leisegang 2022 systematic review. Green marks studies where total testosterone rose relative to a placebo group with a reported p value: Lim 2017, Quin 2021 at two weeks, Leitão 2021 and Chinnappan 2021. Amber marks studies where testosterone rose only against baseline, either because there was no placebo group (Tambi 2012, Henkel 2014) or because no between-group test was reported (Chan 2021). Grey marks the two trials in men with normal testosterone that found no difference from placebo: Ismail 2012 and George 2013, both at 300 mg for 12 weeks. Seven of the nine used the Physta extract from Biotropics Malaysia, and two of the nine were unpublished doctoral dissertations obtained by contacting the authors. Quin's eight-week comparison, not shown separately, found the two-week rise had disappeared.
What did the 2021 trial in 105 men aged 50 to 70 find?
The Chinnappan trial found that 200 mg of Physta a day raised total testosterone by 24.5 ng/dL over 12 weeks, from 200.5 to 225.0, while placebo fell 5.1 ng/dL, from 183.0 to 177.9, a between-group difference that was significant at p under 0.05 at week 4, p under 0.01 at week 8 and p under 0.001 at week 12. The 100 mg arm rose 16.5 ng/dL, from 187.3 to 203.8, reaching significance against placebo only at week 12. The trial ran at three sites in India, screened 113 men, randomized 106 and analysed 105, all of them aged 50 to 70 with a body mass index between 18 and 30 and a baseline total testosterone under 300 ng/dL, which is the cut-off most guidelines use for biochemical hypogonadism. Thirty-five men completed each arm. The size of the effect deserves a plain description. A 12.2 percent rise from 200 ng/dL leaves a man at 225 ng/dL, which is still 75 ng/dL below the threshold that defined him as low in the first place. Free testosterone moved from 14.09 to 16.47 pg/mL on 200 mg and from 16.87 to 17.56 on placebo, and the authors reported no significant between-group difference. DHEA rose slightly from week 2 in both treatment arms, and cortisol fell from 284.6 to 252.8 nmol/L on 200 mg, a within-group change of 11 percent that was significant against baseline. Eight adverse events were recorded in total, seven of them in the treatment arms and six of those gastrointestinal, all mild and resolving within 3 to 5 days without treatment. The study was funded by Biotropics Malaysia Berhad, and two of the authors were company employees.
A 24 ng/dL Rise That Leaves Every Arm Below the Line That Defined Them as Low
Mean total testosterone in ng/dL at baseline and week 12 in the Chinnappan 2021 trial, published in Food and Nutrition Research, in which 105 men aged 50 to 70 with total testosterone under 300 ng/dL were randomized at three sites in India to placebo, Physta 100 mg or Physta 200 mg daily. The 200 mg arm rose from 200.5 to 225.0, a 12.2 percent increase, significant against placebo at week 4 (p under 0.05), week 8 (p under 0.01) and week 12 (p under 0.001). The 100 mg arm rose from 187.3 to 203.8, an 8.8 percent increase, significant against placebo only at week 12. Placebo fell from 183.0 to 177.9. The dashed line marks 300 ng/dL, the entry threshold for the trial and the cut-off most guidelines use for biochemical hypogonadism; no arm reached it. Free testosterone rose from 14.09 to 16.47 pg/mL on 200 mg and from 16.87 to 17.56 on placebo, with no significant difference between groups. The trial was funded by Biotropics Malaysia Berhad and two authors were company employees.
- Baseline
- Week 12
Did the men in that trial feel any different?
The symptom results in the Chinnappan trial were far larger than the hormone results, and that gap is the most interesting thing in the paper. The Aging Males' Symptoms score, a 17-item questionnaire covering sexual, psychological and physical complaints, fell from 58.34 to 26.71 on 200 mg, a 54 percent drop, and from 58.91 to 35.43 on 100 mg, while placebo moved from 58.94 to 58.06. The Fatigue Severity Scale fell from 57.86 to 29.17 on 200 mg, a 50 percent drop, and did not move on placebo. Handgrip strength rose 2.99 kg on 200 mg, from 60.07 to 63.06, against a 0.02 kg fall on placebo. Those are enormous questionnaire changes for a 24 ng/dL rise in testosterone. For comparison, testosterone replacement that lifts a hypogonadal man well into the normal range typically moves the same AMS scale by a handful of points in trials, not 30. Three explanations are possible and the paper cannot separate them. The extract may act on fatigue and mood through pathways unrelated to testosterone, which the cortisol result faintly supports. The questionnaire may have been administered in a way that amplified expectation, which a placebo group flat to the decimal point does not fully rule out because it also means the placebo group reported no improvement of any kind over 12 weeks of trial participation, which is itself unusual. Or the result may not replicate. Leitão 2021, the only other placebo-controlled trial that used the AMS scale, reported significant symptom reductions of 27 and 36 percent in the two exercise arms and did not report one in the extract-only arm.
The Hormone Moved 12 Percent. The Symptom Scores Moved 50.
Percent change from baseline over 12 weeks in the Chinnappan 2021 trial, by arm, for the five outcomes the paper reported in full. Total testosterone rose 12.2 percent on 200 mg and fell 2.8 percent on placebo. Handgrip strength rose 2.99 kg on 200 mg, from 60.07 to 63.06, and fell 0.02 kg on placebo. Cortisol fell 11.2 percent on 200 mg, a within-group change that did not reach significance against placebo. The Aging Males' Symptoms score fell from 58.34 to 26.71 on 200 mg, a 54 percent drop, and from 58.94 to 58.06 on placebo. The Fatigue Severity Scale fell from 57.86 to 29.17 on 200 mg, a 50 percent drop, and rose slightly on placebo. A questionnaire change of that size on a 24 ng/dL rise in testosterone is out of proportion with what testosterone replacement itself produces on the same scale, and a placebo group that reports no change of any kind over 12 weeks of trial participation is itself unusual. Leitão 2021, the only other placebo-controlled trial using the AMS scale, reported significant reductions in its exercise arms and not in its extract-only arm.
- Placebo
- Physta 100 mg
- Physta 200 mg
Does tongkat ali raise testosterone in men with normal levels?
The two largest placebo-controlled trials in men with normal testosterone found no effect, and the positive results in that population come from small, short studies in men under 30. Ismail 2012 randomized 109 married men aged 30 to 55 to 300 mg of Physta or placebo for 12 weeks and found total testosterone unchanged relative to placebo, starting from 476 ng/dL in the treatment group and 542 on placebo; that trial's positive findings were confined to the erectile function domain of a sexual well-being questionnaire, sperm motility and a modest fat loss in the overweight subgroup. George 2013 gave 40 men 300 mg for 12 weeks and found testosterone went from 458 to 484 ng/dL on the extract and from 540 to 542 on placebo, not significant. The results in young men are livelier and shorter. Chan 2021 gave 16 men aged 18 to 30 a large 600 mg dose for two weeks and saw total testosterone rise from 802 to 924 ng/dL while 16 controls drifted from 791 to 769, but the paper provided no between-group statistical test. Quin's 2021 dissertation used the same 600 mg dose in sedentary young men and found a significant rise at two weeks, 871 to 968 against 863 to 790 on placebo, that had vanished by eight weeks. Lim's 2017 dissertation in trained young men found a rise at 3 days and 5 weeks on a weight-based dose. Put together, the healthy-man data show a possible short-lived bump at high doses in men in their twenties, nothing at 300 mg over 12 weeks in men in their thirties and forties, and a consistent signal only in men who began below 300 ng/dL. A man with normal testosterone hoping to push it higher is working against the two best trials available.
Every Controlled Comparison, and the 9-Man Arm That Drives the Pooled Estimate
Percent change in mean total testosterone from baseline on tongkat ali versus placebo in each controlled comparison that reported both arms in the Leisegang 2022 systematic review. George 2013 gave 40 healthy men 300 mg for 12 weeks: 458 to 484 ng/dL against 540 to 542 on placebo, not significant. Chinnappan 2021 gave 105 men aged 50 to 70 with testosterone under 300 ng/dL 100 or 200 mg for 12 weeks: 187.3 to 203.8 and 200.5 to 225.0 against 183.0 to 177.9 on placebo. Quin 2021 gave sedentary men aged 18 to 30 600 mg for 2 weeks: 871 to 968 against 863 to 790, significant, with the difference gone by 8 weeks. Chan 2021 gave 16 men aged 18 to 30 600 mg for 2 weeks: 802 to 924 against 791 to 769 in 16 controls, with no between-group test reported. Leitão 2021 gave 9 men with androgen deficiency of the aging male 200 mg for 6 months: 278.2 to 400.3 against 281.5 to 258.5 in 12 placebo controls, a 44 percent rise that is the largest in the literature and rests on the smallest arm. Ismail 2012 in 109 men reported its post-treatment value as a range and is not plotted; it found no difference from placebo. The pooled standardized mean difference across five randomized trials and 232 men was 1.352, with I-squared of 87.27 percent and a Begg's test for publication bias at p equals 0.024.
- Tongkat ali
- Placebo
What does the 2022 meta-analysis show, and why is an effect size of 1.35 a warning?
The Leisegang 2022 meta-analysis pooled five randomized trials and 232 men and reported a standardized mean difference of 1.352 for total testosterone, with a 95 percent confidence interval of 0.565 to 2.138 and a p value of 0.001, and that number is more a symptom of the literature than a measure of the herb. A standardized mean difference of 1.35 means the average treated man ended up 1.35 standard deviations above the average placebo man, which in pharmacology is an effect the size of a real drug. Fenugreek's 2026 meta-analysis of 637 men put its effect on total testosterone at 0.25. For a plant extract to score more than five times that, in trials where the absolute change was 24 ng/dL, tells you the pooled standard deviations were small and the trials were tiny. The heterogeneity statistic confirms it: Cochran's Q was 47.1 across seven comparisons with an I-squared of 87.27 percent, meaning nearly nine tenths of the variation between studies reflected genuine differences in what the studies found rather than chance. Begg's test for publication bias returned p equals 0.024, which in a five-study analysis is a loud signal that small negative studies are missing. The Cochrane risk-of-bias assessment rated two trials low risk, two with some concerns in the domain of deviations from intended interventions, and one high risk. Two of the five trials were unpublished PhD dissertations obtained by contacting the authors, and the Leitão comparison contributing the largest single effect involved 9 men against 12. The authors' own conclusion was that more research is required before clinical use, which is a reasonable summary of a pooled estimate with a confidence interval running from a moderate effect to a very large one.
Every Controlled Comparison, and the 9-Man Arm That Drives the Pooled Estimate
Percent change in mean total testosterone from baseline on tongkat ali versus placebo in each controlled comparison that reported both arms in the Leisegang 2022 systematic review. George 2013 gave 40 healthy men 300 mg for 12 weeks: 458 to 484 ng/dL against 540 to 542 on placebo, not significant. Chinnappan 2021 gave 105 men aged 50 to 70 with testosterone under 300 ng/dL 100 or 200 mg for 12 weeks: 187.3 to 203.8 and 200.5 to 225.0 against 183.0 to 177.9 on placebo. Quin 2021 gave sedentary men aged 18 to 30 600 mg for 2 weeks: 871 to 968 against 863 to 790, significant, with the difference gone by 8 weeks. Chan 2021 gave 16 men aged 18 to 30 600 mg for 2 weeks: 802 to 924 against 791 to 769 in 16 controls, with no between-group test reported. Leitão 2021 gave 9 men with androgen deficiency of the aging male 200 mg for 6 months: 278.2 to 400.3 against 281.5 to 258.5 in 12 placebo controls, a 44 percent rise that is the largest in the literature and rests on the smallest arm. Ismail 2012 in 109 men reported its post-treatment value as a range and is not plotted; it found no difference from placebo. The pooled standardized mean difference across five randomized trials and 232 men was 1.352, with I-squared of 87.27 percent and a Begg's test for publication bias at p equals 0.024.
- Tongkat ali
- Placebo
Does tongkat ali lower cortisol or help with stress?
One placebo-controlled study of 63 moderately stressed adults found salivary cortisol 16 percent lower and salivary testosterone 37 percent higher on tongkat ali than on placebo after 4 weeks, and that is the entire human evidence base for the stress claim. Talbott and colleagues published the study in the Journal of the International Society of Sports Nutrition in 2013. Sixty-four adults screened for moderate stress on a validated questionnaire were randomized to 200 mg of Physta or placebo daily; 63 completed, 32 men and 31 women. Saliva was collected morning, afternoon and evening on testing days and analysed for free cortisol and testosterone. Tension fell 11 percent, anger 12 percent and confusion 15 percent on the Profile of Mood States compared with placebo, while depression, vigour and fatigue did not change and global mood showed a non-significant trend at p equals 0.083. Liver enzymes, body weight and body fat did not move. Three cautions apply. The cohort was half women, so the 37 percent testosterone figure is a mixed-sex salivary average and cannot be translated into a serum change in men. Salivary testosterone reflects only the unbound fraction and varies far more than serum, which is why it is rarely used as a primary endpoint. And the study was funded by Biotropics and run by SupplementWatch, with one author employed by the company. The Chinnappan cortisol result, an 11 percent within-group fall on 200 mg that was not significant against placebo, is directionally consistent and equally modest. Whether a modest cortisol reduction is the reason the symptom scores fell so far in the older-men trial is an open question nobody has tested directly.
The Hormone Moved 12 Percent. The Symptom Scores Moved 50.
Percent change from baseline over 12 weeks in the Chinnappan 2021 trial, by arm, for the five outcomes the paper reported in full. Total testosterone rose 12.2 percent on 200 mg and fell 2.8 percent on placebo. Handgrip strength rose 2.99 kg on 200 mg, from 60.07 to 63.06, and fell 0.02 kg on placebo. Cortisol fell 11.2 percent on 200 mg, a within-group change that did not reach significance against placebo. The Aging Males' Symptoms score fell from 58.34 to 26.71 on 200 mg, a 54 percent drop, and from 58.94 to 58.06 on placebo. The Fatigue Severity Scale fell from 57.86 to 29.17 on 200 mg, a 50 percent drop, and rose slightly on placebo. A questionnaire change of that size on a 24 ng/dL rise in testosterone is out of proportion with what testosterone replacement itself produces on the same scale, and a placebo group that reports no change of any kind over 12 weeks of trial participation is itself unusual. Leitão 2021, the only other placebo-controlled trial using the AMS scale, reported significant reductions in its exercise arms and not in its extract-only arm.
- Placebo
- Physta 100 mg
- Physta 200 mg
Is tongkat ali safe? What did the European Food Safety Authority conclude?
On 27 October 2021 the European Food Safety Authority's Panel on Nutrition, Novel Foods and Food Allergens adopted an opinion concluding that the safety of Eurycoma longifolia root extract has not been established under any condition of use, and that sentence is the most important safety statement in the tongkat ali literature. The application was submitted by Biotropics Malaysia Berhad in August 2016, originally to the United Kingdom Food Standards Agency, seeking approval for the Physta extract as a food supplement at up to 200 mg a day for adults. The file included a full genotoxicity battery. The bacterial mutation test and the mouse micronucleus test were negative. An in vitro chromosome aberration test in human lymphocytes was positive, which the Panel described as indicating clastogenic properties, meaning the extract could break chromosomes in cultured cells. EFSA requested a follow-up in vivo comet assay, which measures DNA strand breaks in living tissue. That assay was positive at the highest dose of 2,000 mg per kg body weight in the stomach and duodenum of rats, the tissues that first contact a swallowed extract, and histopathology indicated the result reflected genotoxicity rather than cell death. The Panel concluded the extract has the potential to induce DNA damage, of concern particularly for first-contact tissues. It also noted that an earlier unpublished chronic toxicity study had shown adverse effects in liver and kidney, and that the human trials submitted were designed to detect benefit and addressed only a limited number of safety endpoints. Two things should be kept in proportion. The positive comet dose was roughly 700 times a 200 mg human dose on a per-kilogram basis, and a positive comet assay at a maximum tolerated dose does not establish harm at supplement doses. But regulators treat genotoxicity as a threshold hazard, one with no safe dose that can be assumed, which is why the opinion refused to set one. Human trials up to 12 weeks have been unremarkable, with gastrointestinal complaints the only pattern. Long-term safety data in humans does not exist. This article is educational and is not medical advice.
A Quarter Failed a Mercury Limit, a Quarter Were the Wrong Plant, Two Fifths Had No Marker Compound
Findings from three surveys of retail tongkat ali products, expressed as a percentage of products tested in each survey. Ang and colleagues analysed 100 tongkat ali hitam preparations sold in Malaysia and reported in 2006 that 26 contained mercury at 0.53 to 2.35 parts per million, above the national limit of 0.5 ppm for traditional medicines; only 4 of the 26 were registered with the drug control authority. A 2018 DNA barcoding study in Pharmaceutical Biology found 37 percent of tested herbal products were authentic Eurycoma longifolia and 27 percent were adulterated with other species, with the remainder unresolved, and at least one authentic product contained no eurycomanone on chromatography. A Malaysian survey of 41 products cited in the Leisegang 2022 review found 24 contained eurycomanone: 11 inside the recommended 0.8 to 1.5 percent range, 9 above it at 1.6 to 8.48 percent, and 4 below it, with 17 products containing none. Red marks findings that indicate a contaminated, substituted or inert product, amber a product outside the standardized range, and green a product that matched its label. The trials and the EFSA opinion concern the standardized Physta extract; none of these surveys describes it.
What is actually in the bottle?
Surveys of tongkat ali products on the market have found mercury contamination, species substitution and wildly variable amounts of the marker compound, and this matters because the trials all used one standardized extract that most retail products are not. Ang and colleagues analysed 100 tongkat ali hitam products sold in Malaysia and published the results in 2006: 26 contained mercury at 0.53 to 2.35 parts per million, above the Malaysian limit of 0.5 ppm for traditional medicines, and only 4 of those 26 were registered with the national drug authority. A 2018 study in Pharmaceutical Biology used DNA barcoding to check whether tongkat ali herbal products actually contained Eurycoma longifolia and found 37 percent authentic, 27 percent adulterated with other species, and the rest unresolvable; at least one product that passed the DNA test contained none of the expected eurycomanone when checked by chromatography. A Malaysian survey cited in the 2022 meta-analysis tested 41 products and found 24 contained any eurycomanone at all; of those, 11 fell inside the recommended 0.8 to 1.5 percent range and 9 were above it, some as high as 8.48 percent, which is a six-fold spread in the active compound between bottles that share a label. The practical reading is simple. A man buying tongkat ali is buying a product whose contents may bear no relation to the extract in the Chinnappan trial, and the EFSA genotoxicity finding applied to the standardized, quality-controlled version, not to whatever a marketplace listing happens to contain.
A Quarter Failed a Mercury Limit, a Quarter Were the Wrong Plant, Two Fifths Had No Marker Compound
Findings from three surveys of retail tongkat ali products, expressed as a percentage of products tested in each survey. Ang and colleagues analysed 100 tongkat ali hitam preparations sold in Malaysia and reported in 2006 that 26 contained mercury at 0.53 to 2.35 parts per million, above the national limit of 0.5 ppm for traditional medicines; only 4 of the 26 were registered with the drug control authority. A 2018 DNA barcoding study in Pharmaceutical Biology found 37 percent of tested herbal products were authentic Eurycoma longifolia and 27 percent were adulterated with other species, with the remainder unresolved, and at least one authentic product contained no eurycomanone on chromatography. A Malaysian survey of 41 products cited in the Leisegang 2022 review found 24 contained eurycomanone: 11 inside the recommended 0.8 to 1.5 percent range, 9 above it at 1.6 to 8.48 percent, and 4 below it, with 17 products containing none. Red marks findings that indicate a contaminated, substituted or inert product, amber a product outside the standardized range, and green a product that matched its label. The trials and the EFSA opinion concern the standardized Physta extract; none of these surveys describes it.
Who paid for the trials, and does it change the reading?
Biotropics Malaysia Berhad, the maker of Physta and LJ100, funded or supplied the extract for seven of the nine human testosterone studies, employed authors on the two most cited papers, and submitted the failed European safety application, and that concentration of interest is unusual even by supplement-industry standards. The Chinnappan trial listed the company as funder with two employee authors. The Talbott stress study was funded by the company and included a company author. Ismail, George, Tambi, Henkel and Chan all used company-supplied Physta. Industry funding does not make a trial wrong, and the Chinnappan trial was multi-centre, double-blind and reported its raw values in full, which is more than many supplement trials manage. What funding does is shape which questions get asked and which results get published, and the publication-bias test in the meta-analysis suggests the second effect is operating. The single independent randomized trial in the set, Leitão 2021 in Brazil, used a non-Physta dry extract at 200 mg for 6 months in 45 men and found a testosterone rise in the extract-only arm of 9 men, alongside its main finding that concurrent training did most of the work for strength, fitness and erectile function. One well-run independent trial, in 100 or more men over 45 with low testosterone, using a verified extract, for 6 months, with a symptom scale and a hormone panel and a prespecified analysis plan, would settle more than the nine existing studies combined. Nobody without a product to sell has funded one.
A 24 ng/dL Rise That Leaves Every Arm Below the Line That Defined Them as Low
Mean total testosterone in ng/dL at baseline and week 12 in the Chinnappan 2021 trial, published in Food and Nutrition Research, in which 105 men aged 50 to 70 with total testosterone under 300 ng/dL were randomized at three sites in India to placebo, Physta 100 mg or Physta 200 mg daily. The 200 mg arm rose from 200.5 to 225.0, a 12.2 percent increase, significant against placebo at week 4 (p under 0.05), week 8 (p under 0.01) and week 12 (p under 0.001). The 100 mg arm rose from 187.3 to 203.8, an 8.8 percent increase, significant against placebo only at week 12. Placebo fell from 183.0 to 177.9. The dashed line marks 300 ng/dL, the entry threshold for the trial and the cut-off most guidelines use for biochemical hypogonadism; no arm reached it. Free testosterone rose from 14.09 to 16.47 pg/mL on 200 mg and from 16.87 to 17.56 on placebo, with no significant difference between groups. The trial was funded by Biotropics Malaysia Berhad and two authors were company employees.
- Baseline
- Week 12
How should a man over 40 use this evidence?
A man over 40 should read tongkat ali as a herb with a plausible but small testosterone effect in men who already test low, no demonstrated effect in men who test normal, a symptom signal too large to trust until it replicates, and a safety file that a serious regulator declined to sign. Three practical points follow. First, the effect that exists is roughly 25 ng/dL over 12 weeks in men starting around 200 ng/dL, which is a real change and not a clinically transformative one; a man at that level has a medical question about why his testosterone is low, and a supplement that moves him to 225 has not answered it. Second, the only trial evidence for how a man feels on it comes from a study whose questionnaire results outran its hormone results by an order of magnitude, and until an independent group reproduces a 30-point AMS drop, the honest position is that we do not know whether men feel different on tongkat ali. Third, the choice to take it is a choice about an unresolved genotoxicity signal and an unregulated supply chain, not about whether it is a strong or weak testosterone booster. Anyone who decides to try it should use a product that discloses eurycomanone content, take it for a defined period rather than indefinitely, get a testosterone panel before and after so the decision rests on their own numbers, and understand that nobody has followed men on it for longer than 6 months. Anyone with a history of cancer, liver or kidney disease, or on prescription medication should talk to a physician first. The gaps that would matter most to men over 40, long-term safety, effects on prostate-specific antigen and haematocrit, and any interaction with the hypothalamic-pituitary axis that would show up on LH, have not been studied at all.
Every Controlled Comparison, and the 9-Man Arm That Drives the Pooled Estimate
Percent change in mean total testosterone from baseline on tongkat ali versus placebo in each controlled comparison that reported both arms in the Leisegang 2022 systematic review. George 2013 gave 40 healthy men 300 mg for 12 weeks: 458 to 484 ng/dL against 540 to 542 on placebo, not significant. Chinnappan 2021 gave 105 men aged 50 to 70 with testosterone under 300 ng/dL 100 or 200 mg for 12 weeks: 187.3 to 203.8 and 200.5 to 225.0 against 183.0 to 177.9 on placebo. Quin 2021 gave sedentary men aged 18 to 30 600 mg for 2 weeks: 871 to 968 against 863 to 790, significant, with the difference gone by 8 weeks. Chan 2021 gave 16 men aged 18 to 30 600 mg for 2 weeks: 802 to 924 against 791 to 769 in 16 controls, with no between-group test reported. Leitão 2021 gave 9 men with androgen deficiency of the aging male 200 mg for 6 months: 278.2 to 400.3 against 281.5 to 258.5 in 12 placebo controls, a 44 percent rise that is the largest in the literature and rests on the smallest arm. Ismail 2012 in 109 men reported its post-treatment value as a range and is not plotted; it found no difference from placebo. The pooled standardized mean difference across five randomized trials and 232 men was 1.352, with I-squared of 87.27 percent and a Begg's test for publication bias at p equals 0.024.
- Tongkat ali
- Placebo
Frequently Asked Questions
Have more questions? Contact us or explore our full peptide intelligence platform.
Related Articles
Is NAD+ a Peptide? No. It Is a Coenzyme Built From Two Nucleotides, and the Peptide Label Comes From the Clinic, Not the Chemistry
Is NAD+ a peptide? No. NAD+ is a coenzyme built from two nucleotides. What NAD+ is used for, what IV NAD+ studies found, and how it differs from SS-31.
Epitalon vs Pinealon: One Lab Built Both. Epitalon Has an FDA Review and Three Small Human Studies. Pinealon's Case Rests Mostly on Cell Dishes and Rats.
Epitalon vs pinealon compared: structure, targets, sleep and melatonin data, brain studies, human evidence, FDA status, and whether stacking them is tested.
Epithalon vs Epitalon: Same Peptide, Two Spellings. The Name That Means Something Different Is Epithalamin, and Most of the Human Data Belong to It.
Epithalon vs epitalon: one AEDG tetrapeptide, two spellings. Why epithalamin is a different substance and what FDA found in its 2026 epitalon review.