Could having the autoimmune disease vitiligo increase a person’s risk of dementia? Researchers led by Mu-Hong Chen at the Taipei Veterans General Hospital, Taiwan, said as much in the November 12 Journal of Dermatology. They scoured health records of 1,320 people with vitiligo and 5,280 healthy controls to track how many had been diagnosed with dementia. The correlation was stunning: People with vitiligo had five times the risk of developing any type of dementia than did those with normal skin. The risk for vascular dementia quadrupled, and for Alzheimer’s shot up a whopping 12-fold—almost as bad as having two APOE4 genes.

  • In Taiwan, people with vitiligo have a higher risk of vascular dementia and Alzheimer’s.
  • Sex, age, and phototherapy are not to blame.
  • Results contrast a Korean study that found no association.

But is this association real? A larger, nationwide study in South Korea found no association between vitiligo and AD. If this association holds true, it would add another link between autoimmune diseases and dementia risk.

Epidemiologists were cautious about concluding too much, given the contrasting Korean study. “It is very rare to see associations of this magnitude that are not spurious or biased in some way, so I’m a bit skeptical,” Katrina Abuabara at the University of California, San Francisco, told Alzforum.

Vitiligo. In people with this progressive disorder, patches of skin lose pigmentation. [Courtesy of Tamara Abrantes.]

Vitiligo is an autoimmune skin disease where melanin-producing cells called melanocytes slowly die off, leaving lighter patches of skin. Although physicians consider it cosmetic, the overreactive immune system and loss of melanin may indicate something more than skin-deep.

Could melanin begin to explain any link between vitiligo and dementia? Mice with familial AD presenilin mutations have lighter fur than wild-type animals, hinting at impaired melanin production (Jan 2006 news). While BACE2 has weaker ties to AD than its homolog, BACE1, it processes the melanocyte protein PMEL to form a functional amyloid that binds to melanin. BACE2 knockout mice have a silvery coat and rats, rabbits, and dogs given a BACE inhibitor develop lighter skin and patchy fur (Dec 2013 conference news; Nov 2016 conference news). Some people taking BACE inhibitors developed some kind of skin reaction (Feb 2015 news). Therein lies a conundrum—skin lightening associated with reduced BACE activity might be expected to protect against AD.

“It is premature to associate vitiligo with AD pathogenesis,” Matthew Farrer at the University of Florida, Gainesville, told Alzforum (full comment below). He noted that vitiligo and other pigmentation loss is rare in people with Alzheimer’s.

Still, some other autoimmune diseases, including lupus, Type 1 diabetes, and psoriasis, another skin disorder, have been tied to increased dementia risk (Li et al., 2018; Wotton and Goldacre, 2017). Chen wondered if vitiligo, which unleashes neuroinflammatory cytokines, might play a role (Gholijani et al., 2020). 

To find out, first author Tsung-Hsien Chang and colleagues combed through health records from the Taiwan National Health Insurance Research Database, a compilation of health information from almost all Taiwanese residents. They identified 1,320 people 45 years or older diagnosed with vitiligo and 5,280 healthy controls. Four controls were matched to every vitiligo participant based on potential confounding factors, including sex, age, socioeconomic status, and certain dementia-related comorbidities, including cerebrovascular diseases, traumatic head injury, high blood pressure, diabetes, and high plasma cholesterol.

Chang and colleagues checked the participants’ medical records from 1998 on, to see who was diagnosed with dementia during the next 15 years. For every 100,000 person-years, 503 people with vitiligo developed dementia compared to just 102 healthy controls. After adjusting for the confounding factors, the results remained striking: People with vitiligo had five times the risk of developing any form of dementia, four times the risk of developing vascular dementia, and 12 times the risk of developing Alzheimer’s. The authors, and other scientists contacted by Alzforum, noted that these dementia-specific risks should be interpreted carefully because diagnoses listed in medical records are often dubious, and coding of dementia types poor or missing.

What could explain this data?

One possibility is baseline health differences between people with and without vitiligo. Although the researchers did control for some comorbidities, other potential confounding factors were missing from the medical records, including smoking history, alcohol consumption, body mass index, and family medical history.

Notably, people with vitiligo had higher baseline Charlson comorbidity index scores, a general indicator of health measuring 19 conditions. Higher CCI scores indicate that vitiligo participants were more likely to have cardiovascular problems, cancer, and other chronic diseases. Vitiligo is known to be associated with other autoimmune diseases, including Type 1 diabetes, but data on other cardiovascular risk factors are sparse (Chen et al., 2015; Dahir et al., 2018). “The increased risk may be due to another disease that vitiligo is associated with, not vitiligo itself,” said Abuabara.

The vitiligo-dementia association seemed strong in people aged 45-64, which Eric Larson at the Kaiser Permanente Washington Health Research Institute, Seattle, considered noteworthy. However, he stressed that the quality of data from retrospective studies relying on medical claim data is unknown. “This highlights the need for more research,” he said.

Certainly, people with vitiligo should not worry about dementia based on this study. The South Korean study of 15,000 people found no association between this skin disorder and dementia (Kim et al., 2019). Researchers led by Sang Ho Oh at the Yonsei University College of Medicine, Seoul, identified 1,500 people ages 40-79 with vitiligo and 13,500 healthy controls from the Korean National Health Insurance Service health screening cohort. After controlling for age, sex, and smoking history, the incidence of AD among vitiligo and healthy participants was similar. In fact, the risk in the vitiligo group was 16 percent lower. Notably, the two studies controlled for different comorbidities and relied on different codes in the medical record.

The drastically different outcomes of the two nationwide studies raises questions. Abuabara wanted to know more about ascertainment bias—the more you look for something, the more likely you are to find it. “The Taiwanese scientists did not seem to make any adjustments for ascertainment bias. Overall, I would put more stock in the Korean study,” she said.

Farrer believes the strong association seen in the Taiwanese study needs to be replicated before conclusions can be drawn. “While lack of replication is not evidence against the association, the findings will need positive replication,” he said.—Chelsea Weidman Burke


  1. It’s an interesting hypothesis, but more especially for Parkinson’s disease (PD) than AD. The pathogenesis of vitiligo encompasses both melanocyte vulnerability and autoimmunity by interferon-γ-producing, melanocyte-specific CD81 T cells, and melanized neurons in the substantia nigra are preferentially lost in PD.

    There is a recent report of a novel PSEN1 nonsense mutation, chr14: 73673106 c.881G>A (p.W294*) in acute encephalopathy with retinitis pigmentosa (You et al., 2020). Nevertheless, there are a great many more patients with AD caused by PSEN1/2 missense mutations without vitiligo, retinal, or nigral depigmentation/loss reported. Melanosomes are a good model to study intracellular protein transport, and presenilins have a role, but it’s premature to associate vitiligo with AD or PD pathogenesis (Wang et al., 2006). 

    The Taiwanese study, drawn from a national database of 23 million, has marginally greater power than the Korean study, with 39 and 16 vitiligo cases with dementia respectively. Nevertheless, associations need replication, and the results in the first attempt from Korea are not supportive. While lack of replication is not evidence against, the findings will need positive replication.

    I am not aware of vitiligo being noted as a clinical feature in either neurodegenerative disorder, nor of any synthesis from GWAS that’s suggestive of this etiology. It would be helpful to have more detailed clinical and pathologic data from the double-positive cases to further examine the potential relationship between these disorders.


    . Identification and Clinical Analysis of the First Nonsense Mutation in the PSEN1 Gene in a Family With Acute Encephalopathy and Retinitis Pigmentosa. Front Neurol. 2020;11:319. Epub 2020 May 5 PubMed.

    . Regulation of tyrosinase trafficking and processing by presenilins: partial loss of function by familial Alzheimer's disease mutation. Proc Natl Acad Sci U S A. 2006 Jan 10;103(2):353-8. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Mouse of a Different Color: Presenilin Mutations Affect Melanin
  2. Blocking BACE—Do Adult Mouse Phenotypes Predict Side Effects?
  3. Will Next-Gen BACE Inhibitors Dodge Side Effects?
  4. BACE Inhibitor Trial Put on Hold

Paper Citations

  1. . Dementia and Alzheimer's disease risks in patients with autoimmune disorders. Geriatr Gerontol Int. 2018 Sep;18(9):1350-1355. Epub 2018 Jul 25 PubMed.
  2. . Associations between specific autoimmune diseases and subsequent dementia: retrospective record-linkage cohort study, UK. J Epidemiol Community Health. 2017 Jun;71(6):576-583. Epub 2017 Mar 1 PubMed.
  3. . Predominant role of innate pro-inflammatory cytokines in vitiligo disease. Arch Dermatol Res. 2020 Mar;312(2):123-131. Epub 2019 Oct 16 PubMed.
  4. . Comorbidity profiles in association with vitiligo: a nationwide population-based study in Taiwan. J Eur Acad Dermatol Venereol. 2015 Jul;29(7):1362-9. Epub 2014 Dec 16 PubMed.
  5. . Comorbidities in vitiligo: comprehensive review. Int J Dermatol. 2018 Oct;57(10):1157-1164. Epub 2018 May 28 PubMed.
  6. . Risk for neurodegenerative disorders in Korean patients with vitiligo: A nationwide retrospective cohort study. J Am Acad Dermatol. 2019 Aug;81(2):621-623. Epub 2019 Jan 14 PubMed.

Further Reading

Primary Papers

  1. . Association between vitiligo and subsequent risk of dementia: A population-based cohort study. J Dermatol. 2021 Jan;48(1):28-33. Epub 2020 Nov 12 PubMed.