Treat IBD, Dodge PD? Epidemiology Ties Parkinson’s to Inflammation

A recent genetic study linked Parkinson’s disease to inflammatory bowel disorders (IBDs), finding the two shared some risk variants. Now, the authors followed on with a large epidemiological study that cements the association. In the April 23 JAMA Neurology, researchers led by Inga Peter at Icahn School of Medicine at Mount Sinai, New York, reported that people with IBD were at elevated risk of developing Parkinson’s disease. Intriguingly, however, those who took anti-inflammatories directed against tumor necrosis factor (TNF) ran a lower risk than did even the general population. “Our results suggest that lowering systemic inflammation could reduce the risk of PD,” Peter told Alzforum. She is now examining other cohorts to determine if anti-inflammatories associate with lower PD incidence in the broader population, as well.

  • People with IBD are more likely to get Parkinson’s.
  • Those taking anti-TNF therapy, however, had a lower risk than did normal controls.
  • The findings suggest that lowering systemic inflammation could help prevent PD.

Others agreed the data add to the picture of Parkinson’s as an inflammatory disease. “If you put all the data together, including human, animal, and microbiome, it strongly suggests that people who have a higher propensity for inflammatory responses have a higher propensity for PD,” Mark Cookson at the National Institutes of Health in Bethesda, Maryland, told Alzforum.

Researchers are increasingly finding links between Parkinson’s and inflammation in the gut. The enteric microbiome changes early in PD, and gut infections trigger α-synuclein release (Dec 2016 news; Apr 2017 news; Jul 2017 news). 

Peter’s previous genetic study identified specific variants in the immune gene LRRK2 that raised or lowered risk for both PD and IBD (Jan 2018 news). The findings suggested that the diseases should tend to occur together, but epidemiological data were scarce, with only one previous study of 23 million people in Taiwan reporting an increased incidence of PD in people with IBD (Lin et al., 2016). 

To confirm the association, Peter and colleagues analyzed data from 2000–2016 from two large medical databases, the Truven Health MarketScan Commercial Database and the Medicare Supplemental Database. Both host insurance claims from across the U.S. and contain millions of records. The authors selected 144,018 adults with IBD and matched them by age and sex to 720,090 unaffected controls. About 60 percent of the IBD patients had ulcerative colitis; the others had Crohn’s disease. The authors excluded people who had PD before their IBD diagnosis.

Among the whole group of cases and controls, 1,796 people had PD. People with IBD were 28 percent more likely to have Parkinson’s. This increased risk was seen mostly in men. Though the effect size was small, it was highly statistically significant. Peter noted that because both diseases are relatively rare, this association can only be detected in large cohorts. She is currently repeating the analysis in a Swedish cohort, and again finding an association.

The authors next looked at how anti-inflammatory treatment affected risk. Nine percent of the IBD patients, or 13,089 people, were taking an anti-TNF immunotherapy such as adalimumab, certolizumab, golimumab, or infliximab. They had a 78 percent lower risk of PD than other IBD patients did. In absolute terms, they had an incidence of 0.08 PD cases per 1,000 person-years, compared with 0.76 in the other IBD patients. Surprisingly, this incidence was also much lower than the 0.57 cases per 1,000 person-years seen in the control population.

Quyen Hoang at Indiana University School of Medicine, Indianapolis, called the lowered risk remarkable. “The link between TNF-associated inflammation and PD pathogenesis deserves further investigation to determine whether or not it is a causal relationship, and to uncover its potential mechanism,” he wrote to Alzforum.

Cookson was intrigued by the underlying biology implied by the TNF finding. TNF is a pro-inflammatory protein and often detrimental, so inhibiting it makes sense, he said. He suggested that future studies examine whether other anti-inflammatory drugs associate with lowered risk as well. “To be able to show that something that modifies the immune system also modifies PD risk is quite exciting,” Cookson said.

Peter plans to analyze large cohorts of people with rheumatoid arthritis and cirrhosis, who also take TNF inhibitors, to see if they have lower rates of PD, too. Because of the close association between the gut and PD risk, it is possible the therapy would only protect people with IBD, she noted. TNF inhibitors do not cross the blood-brain barrier, so Peter believes they would be most effective as preventatives to lower systemic inflammation and disease risk.

In an accompanying editorial, Clemens Scherzer at Brigham and Women’s Hospital, Boston, praised the strategy of using large data sets to analyze how existing drugs affect disease risk, calling it a “virtual repurposing” trial. “To conduct a prevention trial for PD from scratch would be incredibly difficult and expensive,” he wrote. “Virtual repurposing trials are incredibly powerful for generating novel hypotheses based on patients instead of animal models.”—Madolyn Bowman Rogers

Comments

  1. A number of recent studies have shown intriguing links between the etiology of inflammatory bowel disease (IBD) and Parkinson’s disease (PD) in terms of shared genetic risk and pathophysiology. However, this new study by Peter et al. now provides convincing clinical data that support the association of these two diseases. Indeed, they provide evidence from a large retrospective cohort of individuals with IBD showing an increased incidence of PD among these individuals compared with healthy controls.

    Specifically, the incidence of PD among individuals with IBD was 28 percent higher compared to that among unaffected individuals. Perhaps more striking was that the incidence of PD among IBD-affected individuals decreased by 78 percent in those exposed to anti-TNFα therapy compared with unexposed individuals.

    All together, these findings highlight the importance of inflammation, possibly stemming from the gut, as a driver in both disease settings. Moreover, it also suggests that anti-inflammatory therapies, including anti-TNFα treatment, could be an option to treat PD. Finally, this clinical study further underscores the need for continued basic research that will uncover common disease mechanisms between IBD and PD and identify novel therapeutic targets.

    View all comments by Dana Philpott

  2. This is a really nice study. While it does not suggest causality, the robust association helps us formulate new hypotheses for what the similarities and shared pathways between IBD and PD could be, i.e., chronic dysregulation of inflammation.

    This work reminds me of the association study of treated versus untreated rosacea vis-à-vis risk of PD in a large Danish cohort (Egeberg et al., 2016). The two papers are “composed” in similar ways. They convey an intriguing scientific message, are brief, and inspire new research directions. All three diseases—IBD, rosacea, and PD—are complex, i.e., requiring the interaction between environmental triggers and genetic susceptibility to initiate tissue changes, which is further modified by the factor of time, and often gender. In IBD and rosacea, a contribution of microbial triggers has been established; in PD, the field has emphasized neurotoxicant exposure history, but this is slowly changing, and research into xenobiotic exposure history is on the rise.

    The strengths of Peter et al. are: size of cohorts; integration of a treatment modulator for IBD, i.e., anti-TNF-α therapy; and a balanced discussion of their findings.

    New questions that arise include: What other established immunotherapies in IBD treatment could possibly show a positive or negative association with PD incidence rates? What about effects of smoking and caffeine on the PD risk modulation? Both habits have a known effect in the expressivity of IBD. What about risk association by other systemic inflammatory conditions (and their immunomodulation by biologics), such as rheumatoid arthritis and psoriasis, to name just two? How much of anti-TNF-α reaches the brain during chronic IBD treatment?

    Change is coming to the field of exposome research in typical PD. Patients will welcome it, as we have a growing arsenal of drugs for modulating inflammation in systemic medical illnesses.

    References:

    Egeberg A, Hansen PR, Gislason GH, Thyssen JP. Exploring the Association Between Rosacea and Parkinson Disease: A Danish Nationwide Cohort Study. JAMA Neurol. 2016 May 1;73(5):529-34. PubMed.

    View all comments by Michael Schlossmacher

  3. This study examined whether Parkinson’s disease was more prevalent among individuals with inflammatory bowel disease (IBD) and whether anti-tumor necrosis factor (TNF) treatment for IBD affected PD incidence. The authors reported that the incidence of PD in IBD patients in their study was 28 percent higher than in individuals without IBD. This result was driven by the effect in men with minimal effect of IBD on PD incidence observed in women.

    This study was quite similar in structure to Lin et al. (2016). Both studies evaluated PD incidence within an IBD cohort, excluded any patients who had a PD diagnosis prior to an IBD diagnosis, included a wide age range of subjects, and matched IBD patients and controls on the basis of age, sex, and year of index date. Like this study, Lin et al. also reported an increased risk of PD (35 percent) among IBD patients, with the effect primarily observed in males. These findings are intriguing, as they substantiate the findings regarding overlapping genetic risk factors for IBD and PD as well as the predicted gut-brain mechanism of PD pathogenesis proposed by Braak and colleagues in which it is hypothesized that intestinal inflammation could promote the development of neuropathology and ultimately neurodegeneration in the central nervous system (CNS) (for a review, see Houser and Tansey, 2017). The sex differences noted in these studies are also interesting and in keeping with the known male sex bias in PD.

    The relationship between IBD and PD is not quite as straightforward, however. Another recently published study evaluated the incidence of IBD in recently diagnosed PD patients and found apparently contradictory results—an inverse association between PD and IBD (Camacho-Soto et al., 2018). There are many potential explanations for these conflicting findings, the primary one being numerous differences in study design. The studies by Peter et al. and Lin et al. differ from Camacho-Soto et al. in whether researchers looked for incidence of prior IBD among PD patients or PD development among IBD patients, in the age of subjects, the timeline of IBD and PD diagnoses, type of clinical records used, criteria for classification of PD and IBD, whether cases and controls were matched or not, and potential confounding factors accounted for (degree of use of healthcare services, smoking history, etc.). In any human study, these kinds of variables can have enormous impacts on the results. It is also possible that these disparate findings reveal different facets of the connections between IBD and PD. Perhaps a diagnosis of IBD in early adulthood (included in Peter et al. and Lin et al.) increases the risk of developing PD, while treatment of IBD in later life does not (as seen in Camacho-Soto et al.). Furthermore, the inclusion criteria for all of these studies focus on individuals who had an IBD diagnosis before a PD diagnosis, and if these disorders do share overlapping mechanisms, there may be interesting effects of PD on the development of IBD that have yet to be explored epidemiologically. As research on the gut-brain involvement in PD as well as IBD progresses, we may be better able to tease apart such distinctions and develop a more detailed understanding of the mechanisms linking the disorders. There is no doubt, however, that neurologists and movement disorder specialists must now include questions regarding IBD and other autoimmune conditions when taking family histories of their patients.

    The most novel and exciting aspect of this study is its evaluation of the association between anti-TNF therapy for IBD and the incidence of PD. The authors reported that the incidence of PD was reduced by 78 percent in IBD patients treated with anti-TNF therapy compared with those who did not receive this treatment. This suggests that the relationship between IBD and PD goes beyond overlapping genetic susceptibility and identifies intestinal inflammation as a key mechanism of PD pathogenesis. This is consistent with reports of elevated indicators and mediators of inflammation in stool and intestinal tissue from PD patients compared with controls (Schwiertz et al., 2018; Houser et al., 2018; Devos et al., 2012). 

    It would have been even more informative if this study had undertaken additional probing of anti-TNF effects, including a comparison of the incidence of PD in anti-TNF-treated IBD patients versus controls without IBD, and a comparison of PD incidence in subjects without IBD, who received anti-TNF therapy for other conditions, versus non-IBD controls who did not. These kinds of comparisons could combat the argument that there may have been substantial clinical differences in disease activity in IBD patients prescribed anti-TNF therapy and those who were not, and that it was these differences rather than the immunosuppressant regimen that distinguished these two groups within the IBD cohort.

    Based on the PD rates reported in this study, though, it does appear that in addition to reduced incidence of PD compared with IBD patients not treated with anti-TNF drugs, anti-TNF-treated IBD patients had a reduced incidence of PD compared with controls. If this were true, it would indicate that not only is TNF-associated inflammation responsible for much of the increased frequency of PD among IBD patients but also that suppressing this inflammation could be protective in the general population. Our group and several others have validated soluble TNF as a therapeutic target in multiple preclinical models of neurodegeneration; but clinical trials in humans have not yet been performed. While Camacho-Soto et al. did not specifically evaluate effects of anti-TNF treatments in their IBD cohort, they did report significantly reduced incidence of PD among individuals treated with immunosuppressant drugs of various kinds, including biologics, regardless of whether the subjects had IBD, lending support to the idea of peripheral inflammation as a key risk factor for PD. As it is, these studies provide strong rationale for investigating the capacity of anti-inflammatory drugs and in particular TNF-targeted interventions to slow the progress of PD and even to prevent the onset of clinical symptoms in individuals at higher risk for the development of PD.

    References:

    Lin JC, Lin CS, Hsu CW, Lin CL, Kao CH. Association Between Parkinson's Disease and Inflammatory Bowel Disease: a Nationwide Taiwanese Retrospective Cohort Study. Inflamm Bowel Dis. 2016 May;22(5):1049-55. PubMed.

    Houser MC, Tansey MG. The gut-brain axis: is intestinal inflammation a silent driver of Parkinson's disease pathogenesis?. NPJ Parkinsons Dis. 2017;3:3. Epub 2017 Jan 11 PubMed.

    Camacho-Soto A, Gross A, Searles Nielsen S, Dey N, Racette BA. Inflammatory bowel disease and risk of Parkinson's disease in Medicare beneficiaries. Parkinsonism Relat Disord. 2018 May;50:23-28. Epub 2018 Feb 9 PubMed.

    Schwiertz A, Spiegel J, Dillmann U, Grundmann D, Bürmann J, Faßbender K, Schäfer KH, Unger MM. Fecal markers of intestinal inflammation and intestinal permeability are elevated in Parkinson's disease. Parkinsonism Relat Disord. 2018 Feb 12; PubMed.

    Houser MC, Chang J, Factor SA, Molho ES, Zabetian CP, Hill-Burns EM, Payami H, Hertzberg VS, Tansey MG. Stool immune profiles evince gastrointestinal inflammation in Parkinson's disease. Mov Disord. 2018 Mar 23; PubMed.

    Devos D, Lebouvier T, Lardeux B, Biraud M, Rouaud T, Pouclet H, Coron E, Bruley des Varannes S, Naveilhan P, Nguyen JM, Neunlist M, Derkinderen P. Colonic inflammation in Parkinson's disease. Neurobiol Dis. 2012 Sep 24;50C:42-48. PubMed.

    View all comments by Madelyn Houser View all comments by Malu G. Tansey

  4. These findings were was novel to me, and show how the immune system is always surprising us.

    Systemic inflammation needs to be taken into account when considering cell-based therapy. There are multiple clinical trials planned for treatment of Parkinson’s disease using transplants of dopamine neurons derived from allogeneic pluripotent stem cells. Would peripheral/systemic inflammatory signals have an effect on allogeneic transplants of DA neurons, such as those derived from human embryonic stem cells? I don't know...but given that it’s immunology, I expect to be surprised.

    View all comments by Jeanne Loring

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Do Microbes in the Gut Trigger Parkinson’s Disease?
  2. Rumblings of Parkinson’s: Gut Microbiome Shifts in Early Stage of Disease
  3. Put ’Em Up: Does α-Synuclein Help Fight Microbes in the Gut?
  4. Genetic Findings Link Parkinson’s to Crohn’s Disease, Inflammation

Paper Citations

  1. Lin JC, Lin CS, Hsu CW, Lin CL, Kao CH. Association Between Parkinson's Disease and Inflammatory Bowel Disease: a Nationwide Taiwanese Retrospective Cohort Study. Inflamm Bowel Dis. 2016 May;22(5):1049-55. PubMed.

Further Reading

Primary Papers

  1. Peter I, Dubinsky M, Bressman S, Park A, Lu C, Chen N, Wang A. Anti-Tumor Necrosis Factor Therapy and Incidence of Parkinson Disease Among Patients With Inflammatory Bowel Disease. JAMA Neurol. 2018 Apr 23; PubMed.

Annotate

To make an annotation you must Login or Register.