Tagatose - Zero Sugar AI

What It Is

Tagatose is a rare sugar: a ketohexose, and a stereoisomer of fructose. Commercially it is made from lactose or galactose, from fructose, or, in the newest process, from maltodextrin by a chain of seven enzymes. In a 34-person sensory study it tasted about 0.90 times as sweet as sugar at every concentration tested, so it swaps for sugar almost one for one by weight. Among the bulk sweeteners on this site, only maltitol and xylitol come as close.

Like allulose, it is a sugar that the US label rules have recently started to treat differently from other sugars. Unlike allulose, it is counted here as part-digestible carbohydrate, it carries real calories, and the EU has authorised it. It has no E number, because the EU regulates it as a novel food, not as a food additive.

Calories: Two Regulators, Two Numbers

FDA: 1.5 kcal per gram. FDA’s letter of 21 November 2025 “recognizes D-tagatose’s caloric value of 1.5 calories per gram”. This is enforcement discretion, the same footing as allulose’s 0.4, not a rule.
EFSA: 3 kcal per gram, from a November 2016 opinion on its energy value for labelling. We have this figure as the UK Food Standards Agency reports it; EFSA’s own opinion could not be opened.

The EU figure is twice the US one. Whichever you use, name whose it is.

Is It Absorbed?

Partly, and more than was once assumed. In 2000 JECFA (the joint FAO/WHO expert committee) worked on the assumption that no more than 20% is absorbed. The one ileostomy study, which measures what leaves the small bowel directly, found 81% apparent absorption of 15 g a day. That study is cited here from its abstract. The absorbed part is handled in the liver by the same pathway as fructose.

That is why this site counts half of tagatose’s grams toward net carbs, not none as for allulose. Tagatose on its own does not raise glucose or insulin: 75 g given to people with and without type 2 diabetes changed neither. But a low glucose response is not the same as no usable carbohydrate, and both regulators give it real energy. We will revisit this if a standardised glycaemic index test is published. None has been; the widely quoted “GI 3” traces only to trade press.

Blood Glucose and HbA1c

A 2026 systematic review in the American Journal of Clinical Nutrition pooled eight tagatose trials. It found lower blood glucose after meals and lower insulin, both at moderate certainty, and a small fall in HbA1c of 0.25 percentage points (95% CI 0.06 to 0.44), also moderate certainty. In the same review allulose showed no HbA1c effect, so this is the one long-run marker on which tagatose did better. We read the abstract, not the list of trials behind the pool.

The largest trial needs to be read with its sponsor. Ensor 2015 randomised 494 people with type 2 diabetes to 15 g of tagatose three times a day for 10 months. It was supported in part by a subsidiary of Spherix, the company then developing tagatose as a diabetes drug, and its corresponding author was Spherix’s president when the data were collected. Much of its effect comes from HbA1c rising in the placebo group: in the US subgroup at month 10, HbA1c fell 0.07 points on tagatose and rose 0.37 on placebo. The only long-term study we found that Spherix did not sponsor was a 12-month pilot, though its lead author also co-wrote Spherix-linked tagatose reviews. Eight people completed it, and it had no control group. HbA1c fell from 10.6% to 9.6%, which was not statistically significant (p = 0.08), and 10 of the 12 people recruited had gut side effects.

FDA’s 2025 letter “acknowledges the different effect D-tagatose has on glycemic response in certain studies”. That is a regulator acknowledging studies, not a finding that tagatose treats diabetes.

The US Label: From Sugar to Not, by Letter

Tagatose’s place on the Nutrition Facts label has changed more than once:

2003: FDA authorised the dental-caries health claim for it and said a “sugar free” claim on tagatose foods would be “false and misleading”, “because D-tagatose is a sugar”.
May 2022: FDA denied Bonumose’s 2018 petition on tagatose labelling, while saying it would not object to 1.5 kcal/g for calories.
December 2023: FDA’s position was still that “D-tagatose must be declared as an added sugar”.
August 2024: a federal court vacated the 2022 denial as arbitrary and capricious and sent it back to FDA.
21 November 2025: FDA said it will not object if tagatose is left out of Total Sugars and Added Sugars, “while FDA considers how to address food labeling for sugars metabolized differently than traditional sugars more broadly”.

So a label printed after November 2025 may show no sugars from tagatose, and an older label may legitimately show it as added sugar. By regulation it is still a sugar, and the letter does not mention “sugar free” claims. It also says nothing about Total Carbohydrate, which is calculated by difference, so tagatose stays in it.

Tolerance and Laxative Labels

JECFA: “ingestion of 30g or more of D-tagatose on a single occasion may cause gastrointestinal effects in humans”. That is a caution for setting use levels, not a recommended maximum. In a screening study of 73 young men given about 30 g, 15.1% had nausea and 31.5% diarrhoea, and wind did not settle over 15 days. Twenty grams in chocolate did not significantly increase watery stools. At 75 g, every subject had diarrhoea, nausea or wind. Most of these studies come from one 1999 journal supplement of tagatose safety studies, and we could not see who sponsored them.

In the EU and UK, any product with more than 15 g of tagatose per serving, and any drink over 1% tagatose, must say “excessive consumption may produce laxative effects”. The US has no such requirement. FDA’s own exposure estimate puts the 90th-percentile US eater at 39.8 g a day, above the 30 g caution if eaten at once.

Safety and Regulatory Status

Tagatose is one of the few sweeteners JECFA once refused. In 2000 it allocated no ADI “because of concern about the potential of D-tagatose to induce glycogen deposition and hypertrophy in the liver and to increase the concentrations of uric acid in serum”. It set 0 to 80 mg/kg in 2001 and a temporary 0 to 125 in 2003, and in 2004, once two-year rat studies were in, an ADI “not specified”. That ADI “does not apply to individuals with hereditary fructose intolerance”. If you have that diagnosis, treat tagatose as you treat fructose.

The uric-acid question has not entirely gone away. In a five-person liver-scan study, 30 g of tagatose raised serum uric acid by 17% at 50 minutes, still above baseline nearly four hours later, while fructose did not. Eight weeks of 75 g a day did not raise fasting uric acid (a co-author of that study invented the Biospherics tagatose process), and the 2026 review found no pooled effect on it. If you manage gout or high uric acid, this is worth raising with your clinician before regular use.

US: GRAS by notification, with three FDA “no questions” letters (2001, 2011 and 2021). It is not affirmed in FDA’s regulations.
EU: an authorised novel food since December 2005, with no maximum levels. The maltodextrin process was assessed as safe for Great Britain in 2025; we found no EU authorisation of that route.

Baking and Teeth

Tagatose is a reducing sugar, so it browns like sugar does, unlike every polyol here. Cookies made with it instead of sucrose came out “harder and darker with a lower spread”. It is one of only two sugars covered by the US dental-caries health claim, which must describe it as “a sugar that, unlike other sugars, does not promote the development of dental caries”. The EU allows a tooth-mineralisation claim and a “lower blood glucose rise” claim, both only when tagatose replaces other sugars. The two dental claims rest on plaque-acid and fermentation evidence, not on trials counting cavities.

Comparison Note

Against allulose: both are rare sugars kept off the US sugars lines by FDA enforcement discretion. This site subtracts allulose in full; tagatose was up to 81% absorbed in the one study that measured it, carries 1.5 to 3 kcal/g, and counts half. Tagatose is authorised in the EU; allulose is not. Against erythritol and the other polyols: tagatose is sweeter than erythritol and browns, which no polyol does by the same route, but it is a sugar in law and is handled like fructose in the liver.

What would change this page: FDA rulemaking on sugars metabolised differently, which would make the 2025 discretion a rule or end it; an independently funded long-term HbA1c trial; or a standardised glycaemic index test.

// Sourced record

Tagatose — the checked figures

Supported but limited Reviewed 27 Oct 2026 Next review 30 Apr 2027

A sugar substitute cannot treat a low blood glucose

Low blood glucose is below 70 mg/dL. The ADA's stated rule is 15 grams of FAST-ACTING carbohydrate, then recheck after 15 minutes — none of the substitutes on this site qualify, because not raising blood glucose is the entire point of them. Severe hypoglycaemia is an emergency needing help from others; if someone is unconscious and glucagon is unavailable or nobody knows how to use it, call 911. Your own care plan takes priority over anything written here.

American Diabetes Association clinical guidance checked 31 Aug 2026

Sweetness: 90% of sugar Measured, not a spec: in a 34-panellist sensory study across 3-20% sucrose-equivalent solutions, tagatose ran at about 0.90 of sucrose AT EVERY CONCENTRATION — its sweetness curve had almost exactly sucrose's slope, unlike erythritol, which the same study put at 0.50-0.78. So a 1:1 swap by weight comes close to matching sweetness, which no polyol in this file does except xylitol. The UK FSA quotes "75-92%" with no primary source; the "92%" figure common in trade press was not found in any primary document. Abstract read; full text not opened. Fujimaru T, Park JH, Lim J; Journal of Food… peer reviewed checked 27 Oct 2026
Calories: 1.5 kcal per gram ENFORCEMENT DISCRETION, not a rule — the same footing as allulose's 0.4. FDA's letter of 21 November 2025 "recognizes D-tagatose's caloric value of 1.5 calories per gram" and says FDA will not object to its use when declaring calories. The figure is older than the letter: FDA said in 2022 it would not object to 1.5 kcal/g, and the GRN 977 notice traces it to a 1999 FDA labelling letter that this pass did not locate. THE EU DISAGREES BY A FACTOR OF TWO: EFSA concluded in November 2016 that the energy value for labelling is 3 kcal/g (as reported by the UK Food Standards Agency, fsa_rp1652; EFSA's own opinion was blocked). Whether the EU wrote that into law was not checked. Quote the authority with the number. US FDA, Human Foods Program — signed Kyle Diamantas,… regulatory checked 27 Oct 2026
Net carbs: 50% of its grams still count CONSERVATIVE, and deliberately not subtract_full like allulose. Tagatose taken ALONE does not raise glucose or insulin (75 g in people with and without type 2 diabetes, Donner 1999) — but it is not an unabsorbed sugar. The one ileostomy study measured 81% apparent small-bowel absorption of 15 g/day (Normen 2001, abstract), against JECFA's 2000 assumption of no more than 20%; the absorbed fraction is metabolised in the liver by the fructose pathway; and both regulators assign it real energy (FDA 1.5 kcal/g, EFSA 3 kcal/g). Low glycaemic effect is not the same thing as no metabolisable carbohydrate, so this record counts half. Revisit if a standardised GI test is published — none was found (the widely quoted "GI 3" traces only to trade press). US FDA, Human Foods Program — signed Kyle Diamantas,… regulatory checked 27 Oct 2026 "Net carbs" is not a recognised measure — start from Total Carbohydrate The American Diabetes Association is explicit: the term "does not have a legal definition and is not used by the Food and Drug Administration or recognized by American Diabetes Association", and the FDA recommends using total carbohydrates on the label. Their reasoning applies directly to the subtraction this site performs: it assumes fibre and sugar alcohols are not absorbed or metabolised, "but this is not always true, and some are partially digested and therefore still provide calories as well as impact blood glucose" — and because the label does not say WHICH sugar alcohol is present, the effect on blood glucose "cannot be determined precisely". Treat the figure above as a formulation aid, not a carbohydrate count to dose insulin against. American Diabetes Association clinical guidance checked 31 Aug 2026
Glycemic response: No rise from tagatose alone; lowers the postprandial response to other carbohydrate; small pooled HbA1c reduction (-0.25 points), leaning on industry-run trials Osborn et al. 2026 (AJCN; PROSPERO CRD42024567433; the same review session 14 used for allulose): across 8 tagatose trials, postprandial glucose SMD -1.03 (-1.36 to -0.71) and insulin SMD -1.05 (-1.61 to -0.49), both moderate certainty; HbA1c mean difference -0.25 (-0.44 to -0.06), moderate certainty — the one long-run marker where tagatose beat allulose, which showed no HbA1c effect in the same review; fasting insulin lower at very low certainty; "No other significant pooled effects". Abstract only — the tagatose trial list, and therefore how much weight the next study carries in the pool, was not seen. READ THE LARGEST TRIAL WITH ITS CONFLICT: Ensor 2015 (phase 3, 494 randomised, people with type 2 diabetes not on glucose-lowering drugs, 15 g three times a day for 10 months) was "supported in part by the Biospherics subsidiary of Spherix Incorporated", the company developing tagatose as a diabetes drug, and its corresponding author "was president of Spherix at the time the clinical data were collected". Much of its placebo-adjusted effect is HbA1c RISING in the placebo arm (US subgroup at month 10: tagatose -0.07, placebo +0.37). The only independent long-term study found, a 12-month uncontrolled pilot (Donner 2010, n=8 completers), saw HbA1c fall from 10.6% to 9.6%, NOT significant (p=0.08), with GI side effects in 10 of 12 recruited. FDA's 2025 letter "acknowledges the different effect D-tagatose has on glycemic response in certain studies" — a regulator acknowledging studies, not an efficacy finding. Osborn L, DuPuis K, Liu S, Della Corte D,… peer reviewed checked 27 Oct 2026
Counts in Total Carbohydrate (US): Yes By default, not by an FDA statement. The 2025 letter covers Total Sugars, Added Sugars and calories only and says nothing about Total Carbohydrate, which 21 CFR 101.9(c)(6) calculates by difference — so tagatose stays in it, as allulose does. US FDA, Human Foods Program — signed Kyle Diamantas,… regulatory checked 27 Oct 2026
Counts in Total Sugars (US): No ENFORCEMENT DISCRETION since 21 November 2025, "while FDA considers how to address food labeling for sugars metabolized differently than traditional sugars more broadly". By regulation tagatose is still a sugar: FDA said so in 2002-03 when it refused to exclude it from "sugars" and called a "sugar free" claim on tagatose foods "false and misleading", and that reasoning has not been withdrawn. Route here: Bonumose petition (2018), FDA denial (May 2022), a court vacating that denial as arbitrary and capricious (D.D.C., 28 Aug 2024), then the 2025 letter. The letter does not mention "sugar free" or other nutrient content claims. US FDA, Human Foods Program — signed Kyle Diamantas,… regulatory checked 27 Oct 2026
Counts in Added Sugars (US): No Same letter, same discretion. Until late 2025 FDA's position was that "D-tagatose must be declared as an added sugar" (FDA, December 2023), so older labels can legitimately still show it there. US FDA, Human Foods Program — signed Kyle Diamantas,… regulatory checked 27 Oct 2026
Acceptable daily intake: JECFA: not specified (2004). No EU ADI; authorised as a novel food with no maximum levels A history worth knowing, because tagatose is one of the few sweeteners JECFA once refused. 2000: NO ADI, "because of concern about the potential of D-tagatose to induce glycogen deposition and hypertrophy in the liver and to increase the concentrations of uric acid in serum". 2001: 0-80 mg/kg bw, from a human no-effect level of 15 g three times a day. 2003: temporary 0-125. 2004: "not specified", once two-year rat histopathology was in. JECFA attached two conditions: the 30 g single-occasion GI caution (see tolerance), and that the ADI "does not apply to individuals with hereditary fructose intolerance" — anyone with that diagnosis should treat tagatose as they treat fructose. Joint FAO/WHO Expert Committee on Food Additives (INCHEM) regulatory checked 27 Oct 2026
Digestive tolerance: GI effects from about 30 g in one sitting; EU/UK labels warn above 15 g per serving JECFA (2004): "ingestion of 30g or more of D-tagatose on a single occasion may cause gastrointestinal effects in humans" — a caution about use levels, NOT a recommended maximum, though the GRN 977 dossier paraphrases it as one. In a screening study of 73 healthy young men given about 30 g, nausea ran at 15.1% and diarrhoea at 31.5%, and flatulence did not settle over 15 days of 30 g a day (Buemann 1999); 20 g in chocolate did not significantly increase watery stools (Lee & Storey 1999); 75 g at once caused diarrhoea, nausea and/or flatulence in every subject (Donner 1999). In the EU and UK "a tolerable single dose of no more than 15 g D-tagatose per serving was agreed", and the novel-food authorisation requires "excessive consumption may produce laxative effects" on any product above 15 g per serving and on drinks above 1% tagatose. The US has no such label requirement. FDA's own exposure estimate in GRN 977 puts the 90th-percentile US eater at 39.8 g a day — above the 30 g single-sitting caution if eaten at once. Most tolerance studies are abstract-only here and sit in a 1999 journal supplement of tagatose safety studies; sponsorship was not visible. Joint FAO/WHO Expert Committee on Food Additives (INCHEM) regulatory checked 27 Oct 2026
Dental evidence: Non-cariogenic (regulatory finding); clinical trial evidence thin The only sugar besides isomaltulose that the US dental-caries health claim covers: 21 CFR 101.80 lists "The sugars D-tagatose and isomaltulose", and requires the claim to identify it "as a sugar that, unlike other sugars, does not promote the development of dental caries". The EU authorises a comparative tooth-mineralisation claim for it as a sugar replacer (Reg 432/2012). Both rest on plaque-pH and fermentation evidence, not caries-incidence trials: a 2025 systematic review found only 3 RCTs out of 1,139 records, with heterogeneous quality. Electronic Code of Federal Regulations regulatory checked 27 Oct 2026
Metabolic research: Large single doses briefly raise uric acid; weeks of daily use did not raise fasting levels The absorbed fraction is handled like fructose: in a 31P-MRS study (n=5) 30 g tagatose trapped tagatose-1-phosphate in the liver, liver ATP fell about 12%, and serum uric acid was 17% higher at 50 minutes and still not back to baseline at 230 minutes — fructose showed no such change in that study (Buemann 2000, abstract). Against that: 75 g a day for 8 weeks did not raise FASTING uric acid (Saunders 1999, abstract; a co-author is the Biospherics tagatose inventor), JECFA judged 15 g in hyperuricaemic men to have "no biologically significant effect" while noting there was no control group, and Osborn 2026 found no significant pooled uric-acid effect. Relevant mainly to anyone managing gout or high uric acid, and the reason the 2000 JECFA meeting withheld an ADI. Buemann B, Gesmar H, Astrup A, Quistorff B; Metabolism… peer reviewed checked 27 Oct 2026
Health claims: US: dental caries claim. EU: lower blood-glucose rise and tooth mineralisation, both only when replacing other sugars US: 21 CFR 101.80 since 2003 (petition 02P-0177), reconfirmed in FDA's 2025 letter. EU, Reg (EU) 432/2012: "Consumption of foods/drinks containing [D-tagatose] instead of [other sugars] induces a lower blood glucose rise after their consumption compared to sugar-containing foods/drinks", on condition that tagatose replaces other sugars to the extent a REDUCED SUGARS claim requires; and "contributes to the maintenance of tooth mineralisation" where plaque pH does not fall below 5.7. Both EU claims are COMPARISONS WITH SUGAR, not a stand-alone benefit — the same footing as the polyols' claims. The EFSA opinions behind them (2011) were not opened (Wiley 403); wording is from the regulation as adopted, not the consolidated text. European Commission (text as adopted, via legislation.gov.uk) regulatory checked 27 Oct 2026
Browning: high A reducing ketohexose, so Maillard-active like allulose and unlike every polyol here. Cookies made with tagatose instead of sucrose were "harder and darker with a lower spread" (Taylor 2008, abstract); panellists liked the 100%-tagatose colour but not its sweetness, and a 50% swap matched the control on liking. The GRN 977 notifier calls it "a very effective browning agent" — a manufacturer's words, consistent with the bake data. Taylor TP, Fasina O, Bell LN; Journal of Food… peer reviewed checked 27 Oct 2026
Bulk contribution: bulking Used at sugar-like weights; at about 0.9 sucrose sweetness, close to a 1:1 swap. JECFA lists its functions as sweetener, texturizer, stabilizer, humectant and formulation aid. Joint FAO/WHO Expert Committee on Food Additives (INCHEM mirror… regulatory checked 27 Oct 2026
United States: GRAS by notification Three FDA "no questions" letters: GRN 78 (Arla Foods Ingredients, lactose route, 2001), GRN 352 (CJ CheilJedang, 1 Feb 2011) and GRN 977 (Bonumose, from maltodextrin by a seven-enzyme cascade, 21 Oct 2021). GRN 78's date differs by a day between FDA's inventory (25 Oct 2001) and FDA's own GRN 977 letter (24 Oct 2001). Not affirmed in 21 CFR 184. As with every GRAS letter, FDA states it "is not an affirmation that D-tagatose is GRAS under 21 CFR 170.35". FDA's exposure estimate in GRN 977: 15.9 g per person per day at the mean, 39.8 g at the 90th percentile. US FDA, Office of Food Additive Safety regulatory checked 27 Oct 2026
European Union: Authorised novel food (no E number) On the Union list of novel foods (Reg (EU) 2017/2470) as "D-Tagatose", first authorised in December 2005 under Reg 258/97 Art 4.2 for Arla — by notification, with no Commission Decision (2005/581/EC, sometimes cited for it, is isomaltulose). Production by isomerisation of galactose or epimerisation of fructose; food categories and maximum levels "Not specified"; laxative statement above 15 g per serving (see tolerance). The newer maltodextrin route was assessed as safe for GREAT BRITAIN by the FSA/FSS in 2025 (RP1652); no EU authorisation of that route was found, and the current consolidated EU entry was not opened (EUR-Lex bot-check) — the wording here is the 2017 text as adopted. Unlike allulose, which the EU has not authorised at all. European Commission (text as adopted, via legislation.gov.uk) regulatory checked 27 Oct 2026

Where this leaves it

A rare sugar that bakes and sweetens almost like sugar and does not raise glucose on its own — with a small HbA1c signal no other sweetener here has, but one that rests heavily on trials run by its would-be drug developer. Its US "not a sugar" label treatment is a 2025 enforcement-discretion letter, not law.

Thirty grams in one sitting gives a meaningful share of people GI symptoms, and the EU and UK require a laxative warning above 15 g per serving. The absorbed part is handled like fructose: not for anyone with hereditary fructose intolerance.

If you're managing diabetes

Does not raise glucose on its own and may blunt the rise from other carbohydrate — but it is partly absorbed, so this record counts half its grams, and the HbA1c benefit leans on industry-run trials.

The pooled HbA1c reduction (-0.25 points, moderate certainty, Osborn 2026) is real but small, and the largest trial behind it was sponsored by the company developing tagatose as a diabetes drug, with much of its effect coming from HbA1c rising in the placebo arm. The one independent 12-month pilot did not reach significance. Net carbs: up to 81% may be absorbed in the small bowel and it carries 1.5 (FDA) to 3 (EFSA) kcal/g, so the conservative half-count applies. On a US label printed after November 2025 it may no longer appear under Total Sugars or Added Sugars, but it is still in Total Carbohydrate. Tolerance: 30 g in one sitting is where GI symptoms become common. A gout or high uric acid history is worth mentioning to a clinician before regular use; hereditary fructose intolerance rules it out.

Osborn L, DuPuis K, Liu S, Della Corte D,… peer reviewed checked 27 Oct 2026

What would change this: FDA rulemaking on "sugars metabolized differently than traditional sugars", which would turn the 2025 discretion into a rule or withdraw it; an independently funded long-term trial of HbA1c; or a standardised GI test.

No sweetener here is established to improve long-term health ADA Standards of Care 2026, Section 5: "NNS do not appear to have a significant effect on glycemic management," and evidence on weight is mixed — some reviews find modest weight-loss benefit, others an association with weight gain, likely reflecting reverse causation and confounding; added to an eating plan without energy restriction, NNS show no weight benefit. Substituting can reduce sugar and calories. That is a different claim from improving an outcome, and only the first is supported. CORROBORATED 2026-08-31 at much larger scale: Jeong et al.'s umbrella review (29 meta-analyses, 24 observational studies plus 6 RCTs, 50M+ participants, 135 associations examined) found observational studies link higher sweetener intake to INCREASED risk of weight gain, metabolic syndrome, type 2 diabetes and hypertension, while RCTs show only modest weight-loss benefit versus sugar specifically. The two study designs are not measuring the same comparison — see the general-health page for why — but neither supports a durable, independent long-term health benefit, which is the ADA position this guardrail already states. American Diabetes Association Professional Practice Committee for Diabetes; Diabetes… clinical guidance checked 31 Aug 2026

Every figure above links to the document it came from and carries the date we last opened that document. Where a figure has no source it is not shown at all, rather than stated without one. Tell us if you find one that is wrong.

Get told when the sweetener data changes.

This page is educational information, not medical or dietary advice. Nutrition and regulatory details change over time — verify against the cited primary sources before relying on them for formulation or health decisions.