IBD, anti-TNF drugs and LRRK2

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Inga Peter, Marla Dubinsky and Susan Bressman (Icahn School of Medicine at Mount Sinai) with Anne Wang and three other authors affiliated with AbbVie, which makes the anti-TNF drug adalimumab (the paper discloses that three authors are AbbVie employees and shareholders and that Dubinsky consults for AbbVie and others). Hui et al.: K.Y. Hui (Yale) with senior authors Judy Cho and Inga Peter and many collaborators, using Ashkenazi Jewish IBD cohorts.

What happened

• Peter et al., JAMA Neurology 2018: Using two US insurance-claims databases (MarketScan and Medicare Supplemental; January 2000 to March 2016), 144,018 people with IBD were matched to 720,090 controls; 1,796 had Parkinson's. Incidence was 28% higher with IBD (rate ratio 1.28, 95% CI 1.14–1.44). Within IBD, anti-TNF exposure (drugs blocking the inflammatory messenger TNF, such as infliximab) was linked to a 78% lower rate (0.22, 95% CI 0.05–0.88).
• Hui et al., Sci Transl Med 10(423):eaai7795, 10 Jan 2018: Exome sequencing of Ashkenazi Jewish Crohn's patients, then genotyping of 2,066 cases and 3,633 controls, found LRRK2 N2081D raised Crohn's risk (P = 9.5 × 10⁻¹⁰) and the N551K haplotype (tagging R1398H) protected (P = 3.3 × 10⁻⁸). Across 24,570 Crohn's cases, Parkinson's patients and controls the same LRRK2 region showed shared effects on both diseases. N2081D, like G2019S (the main Parkinson's LRRK2 mutation, in the same kinase domain), raised kinase activity; N551K and R1398H did not, but R1398H raised GTPase activity, which deactivates LRRK2.

Why it moved us toward a cure

It linked peripheral and gut inflammation to Parkinson's through the gene behind the lead drug target (landmark 7), and suggested LRRK2 acts in immune cells as well as neurons. It also framed anti-inflammatory repurposing, alongside landmark 55.

Outcome

Open. The anti-TNF estimate rests on few cases and wide intervals (0.05–0.88), and treated patients may differ in disease severity. Alzforum commenters noted that the excess was driven by men, that Camacho-Soto et al. (2018) found an inverse IBD–Parkinson's link in Medicare data, and that the design cannot show causation. This is not evidence that TNF blockers treat Parkinson's. A registry search (20 Oct 2026) found no anti-TNF trial in Parkinson's; the only TNF-targeted brain trials found were in Alzheimer's.

Sources

• Peter et al., JAMA Neurol 2018;75(8):939–946, DOI 10.1001/jamaneurol.2018.0605, PMID 29710331 (abstract and disclosures via PubMed; an editorial comment, JAMA Neurol 2018;75:917–920, accompanied it) · Alzforum discussion
• Hui et al., Sci Transl Med 2018;10(423):eaai7795, DOI 10.1126/scitranslmed.aai7795, PMID 29321258 (abstract via PubMed)
• Not yet verified: per-ancestry Parkinson's odds ratios in Hui et al. (the abstract gives none); the Lin 2016 and Camacho-Soto 2018 IBD papers beyond Alzforum comments; funders of both studies. No correction or retraction found for either paper.