Larazotide

The mechanism, and why it is specific to coeliac disease. In coeliac disease, gluten peptides have to reach the lamina propria to be deamidated by tissue transglutaminase and trigger the T-cell response that damages the mucosa. One route is paracellular - between the epithelial cells, through tight junctions that gluten itself loosens via zonulin signalling. Larazotide acetate (AT-1001) is an eight-residue peptide that prevents those junctions from opening [1][3]. It is not an immunosuppressant and it does not degrade gluten; it is trying to keep the antigen out.

Gluten challenge trials. In 86 diet-controlled coeliac patients given 2.4 g of gluten daily for 14 days, the primary endpoint was the urinary lactulose/mannitol ratio - a measure of intestinal permeability. It failed. The measurement was highly variable in outpatients, the gluten challenge did not significantly raise it above the gluten-free control, and larazotide did not significantly change it. What the trial did find was that larazotide limited gluten-induced worsening of symptoms at some lower doses but not the higher dose, and that 2.4 g of gluten daily reliably produces symptoms [3]. A second gluten-challenge trial reported similarly [4].

The largest trial, and its strange result. 342 adults with coeliac disease who remained symptomatic after at least 12 months on a gluten-free diet were randomised to larazotide 0.5, 1 or 2 mg three times daily or placebo, with a placebo run-in and run-out. The primary endpoint - average on-treatment coeliac gastrointestinal symptom score - was met by the 0.5 mg dose (P = 0.022 by ANCOVA, P = 0.005 by mixed model). That dose also gave 26% fewer symptomatic days, 31% more improved days, and reduced headache and tiredness. The 1 mg and 2 mg doses were no different from placebo for any endpoint. Safety was comparable with placebo [2].

An inverted dose response is not impossible - a peptide that self-associates at higher concentration, or a receptor that desensitises, could produce it - but no mechanism has been published, and a result that appears only at the lowest of three doses is the kind of result that needs independent replication before it is believed.

Meta-analysis. Four RCTs comprising 626 patients: no significant effect on the lactulose/mannitol ratio regardless of gluten status; significant symptomatic improvement on both the total and coeliac-specific GSRS in the gluten-challenge subgroup but not in those on a gluten-free diet; and reduced gluten-related diarrhoea. The authors concluded larazotide is well tolerated and "seems somehow superior to placebo" in the gluten challenge setting, with more RCTs needed [5].

Where it stands. Larazotide went into a phase 3 programme and there is still no approved non-dietary treatment for coeliac disease [6]. It remains the most clinically advanced attempt.

Tight junctions seal the space between adjacent enterocytes and are dynamically regulated. Zonulin, the mammalian analogue of a Vibrio cholerae toxin, opens them; gluten triggers zonulin release in coeliac disease, and the resulting paracellular leak lets gluten peptides reach the immune cells beneath [1].

Larazotide is a peptide antagonist of that process, acting from the luminal side to prevent junction disassembly. Because it works in the lumen, systemic absorption is unnecessary and undesirable, and the single-dose study confirmed the local mode of action [7].

The trials expose a gap between the mechanism and the clinical effect. If the drug works by keeping junctions shut, permeability should fall - and permeability, measured as the lactulose/mannitol ratio, did not change in any trial or in the pooled analysis [3][5]. Symptoms improved anyway. Either the permeability assay is too noisy to detect the effect - which the 2012 trial explicitly found in outpatients [3] - or the symptom benefit is coming from something other than the stated mechanism. Nobody has resolved which.

An oral capsule taken three times daily before meals. It is designed to be poorly absorbed and to act in the gut lumen, which the proof-of-concept pharmacokinetic study confirmed [1][7]. No marketed product exists.

Consistently unremarkable, which for a chronic oral therapy is a genuine advantage. In the 342-patient 12-week trial, safety was comparable with placebo [2]. The meta-analysis of four RCTs found adverse events comparable between larazotide and placebo, and in fact fewer episodes of gluten-related diarrhoea on larazotide during gluten challenge [5]. The single-dose study established tolerability and the local mode of action [7].

The minimal systemic absorption is the reason for this: a drug that stays in the gut lumen has few opportunities to cause systemic harm [1].

What the safety record does not cover is long-term use. The longest published exposure is 12 weeks [2]. Coeliac disease is lifelong, so a treatment would be taken for decades, and nothing in the published literature speaks to that.

No interaction studies have been published, and minimal systemic absorption makes pharmacokinetic interactions unlikely [7].

The relevant practical point is not an interaction but a caveat: larazotide was tested as an adjunct to a gluten-free diet, never as a replacement for one. Every trial maintained participants on their diet [2], and in the challenge trials the gluten exposure was deliberate and controlled at 2.4 g per day [3].