This population-based cohort study examines whether the risk of vitiligo differs in transplant and nontransplant recipients and between patients with and without graft-vs-host disease.
Key Points
Question
Does the risk of vitiligo differ between transplant and nontransplant recipients and between patients with and without graft-vs-host disease (GVHD)?
Findings
In this population-based cohort study of 23 829 transplant recipients and 119 145 age- and sex-matched controls, transplant recipients had a higher risk of vitiligo than controls. The highest risk was observed in patients who had undergone hematopoietic stem cell transplant, and those who received allogeneic grafts or had comorbid GVHD had a higher risk of vitiligo compared with individuals in the control group.
Meaning
Transplant recipients, especially those who underwent hematopoietic stem cell transplant and had GVHD or received allogeneic grafts, had higher risk of vitiligo than controls.
Abstract
Importance
Vitiligo is a multifactorial, depigmenting skin disorder characterized by selective loss of melanocytes. Large-scale studies are lacking to determine the risk of vitiligo in transplant recipients with graft-vs-host disease (GVHD).
Objective
To investigate the incidence rates and risk of vitiligo in patients who had received solid organ transplant (SOT) or hematopoietic stem cell transplant (HSCT) overall and by HSCT graft type and concomitant GVHD.
Design, Setting, and Participants
This population-based cohort study included data from the National Health Insurance Service database of Korea for patients aged 20 years or older who had received a transplant (SOT or HSCT) between January 2010 and December 2017, with follow-up until December 2019. A cohort of age- and sex-matched (1:5) control individuals who did not receive a transplant was included for comparison. Data were analyzed from July 2021 to December 2021.
Exposure
Transplant (SOT or HSCT) and GVHD.
Main Outcomes and Measures
The main outcome was risk of vitiligo, assessed using multivariable Cox proportional hazards regression analyses adjusting for potential confounding factors.
Results
The study included 23 829 patients who had undergone SOT or HSCT (62.78% male; mean [SD] age, 49.58 [11.59] years) and 119 145 age- and sex-matched controls. Patients who had undergone transplant had a significantly higher risk of vitiligo compared with controls (adjusted hazard ratio [AHR], 1.73; 95% CI, 1.35-2.22). Risk of vitiligo was also slightly higher in kidney transplant recipients and liver transplant recipients compared with the controls but was highest in HSCT recipients (AHR, 12.69; 95% CI, 5.11-31.50). Patients who had received allogeneic grafts (AHR, 14.43; 95% CI, 5.61-37.15), those who had received autologous grafts (AHR, 5.71; 95% CI, 1.20-3.18), those with comorbid GVHD (AHR, 24.09; 95% CI, 9.16-63.35), and those without GVHD (AHR, 8.21; 95% CI, 3.08-21.87) had a higher risk of vitiligo compared with controls.
Conclusion and Relevance
In this study, risk of vitiligo was significantly higher in transplant recipients, especially in HSCT recipients and those with allogeneic grafts or comorbid GVHD. These findings provide new insights into the association between the risk of vitiligo and transplant and GVHD. Clinicians should be aware of these risks, implementing a multidisciplinary approach for monitoring.
Introduction
Recently, de novo vitiligo associated with immune disturbances, such as subsequent vitiligo after immune checkpoint inhibitor therapy and transplant, has been reported.1,2 Although an association has been proposed between vitiligo and transplant, including hematopoietic stem cell transplant (HSCT) and solid organ transplant (SOT), the evidence for such an association is based on case reports or case series.2,3,4,5 Of the few population-based studies that have been conducted, most simply analyzed the association between transplant and vitiligo.6,7 To our knowledge, no large-scale studies have quantified the risk of vitiligo according to comorbid graft-vs-host disease (GVHD). We analyzed the risk of subsequent vitiligo in patients who had received SOT or HSCT based on the National Health Insurance Service (NHIS) claims database.
Methods
Data Source, Study Population, and Comorbidities
This cohort study used the NHIS database, which contains data for over 97% of the Korean population; patients aged 20 years or older who had received a transplant between January 2010 and December 2017 were included. We excluded patients with previously diagnosed vitiligo, and a 12-month lag period was applied. The study cohort included SOT (kidney, liver) and HSCT recipients and control individuals (age and sex matched at a ratio of 1:5) (eFigure in the Supplement 1). The participants were followed up for up to 10 years from January 2010 to December 2019 to identify new-onset vitiligo (International Statistical Classification of Diseases and Related Health Problems, Tenth Revision [ICD-10] code L80 diagnosed at least once). The study was approved by the Ethics Committee of Seoul St Mary’s Hospital. Informed consent was waived because of use of deidentified data. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Patients in the study population with matching health checkup data were additionally analyzed. The health checkup database is a sub–data set of the NHIS database, and NHIS subscribers are encouraged to undergo health checkups every 2 years.8 This health checkup database contains information on lifestyle factors, including body mass index (BMI; calculated as weight in kilograms divided by height in meters squared), smoking status, alcohol consumption, physical activity, blood pressure, fasting blood glucose level, and cholesterol levels.
Comorbid diseases were defined on the basis of ICD-10 codes. Information on BMI, smoking status (nonsmoker, ex-smoker, or current smoker), alcohol consumption (nondrinker, mild drinker [<30 g/d], or heavy drinker [≥30 g/d]), physical activity, systolic blood pressure, and fasting plasma glucose and total cholesterol levels was obtained from the health checkup data.
Statistical Analyses
Data were analyzed from July 2021 to December 2021. To evaluate the risk of vitiligo, we performed multivariable Cox proportional hazards regression analyses after adjusting for the following potential confounding factors: age, sex, hypertension, diabetes, dyslipidemia, and previous hematologic disorders (leukemia, lymphoma, multiple myeloma, and bone marrow failure disease). In health checkup database analysis, the following potential confounding factors were adjusted for: age, sex, hypertension, diabetes, dyslipidemia, previous hematologic disorders, alcohol consumption, physical activity, BMI, fasting plasma glucose level, total cholesterol level, and systolic blood pressure.
Statistical analyses were performed using SAS software, version 9.4 (SAS Institute Inc). Two-sided P < .05 indicated statistical significance.
Results
NHIS Database Analysis
The study included 23 829 patients (62.78% male; mean [SD] age, 49.58 [11.59] years) who had undergone SOT (kidney [n = 12 994], liver [n = 7698]) or HSCT (n = 3137) and 119 145 age- and sex-matched controls (Table 1). The mean (SD) observation time was 5.12 (2.15) years (5.18 [2.11] years for the nontransplant group and 4.79 [2.29] years for the transplant group). The risk of vitiligo was higher in the overall transplant group (adjusted hazard ratio [AHR], 1.73; 95% CI, 1.35-2.22), the kidney transplant group (AHR, 1.50; 95% CI, 1.08-2.08), and the liver transplant group (AHR, 1.63; 95% CI, 1.09-2.43) compared with controls and was highest in the HSCT group (AHR, 12.69; 95% CI, 5.11-31.50) (Table 2). Patients with allogeneic HSCT (AHR, 14.43; 95% CI, 5.61-37.15), those who had received autologous grafts (AHR, 5.71; 95% CI, 1.20-3.18), those with comorbid GVHD (AHR, 24.09; 95% CI, 9.16-63.35), and those without GVHD (AHR, 8.21; 95% CI, 3.08-21.87) had a significantly higher risk of vitiligo compared with the control group.
Table 1. Baseline Characteristics of the Study Population.
| Characteristic | Patientsa | ||||||
|---|---|---|---|---|---|---|---|
| No transplant (n = 119 145) | Transplant (n = 23 829) | KT (n = 12 994) | LT (n = 7698) | HSCT | |||
| All (n = 3137) | Allogeneic graft (n = 1733) | Autologous graft (n = 1404) | |||||
| Sex | |||||||
| Female | 44 345 (37.22) | 8869 (37.22) | 5368 (41.31) | 2118 (27.51) | 1383 (44.09) | 801 (46.22) | 582 (41.45) |
| Male | 74 800 (62.78) | 14 960 (62.78) | 7626 (58.92) | 5580 (72.49) | 1754 (55.91) | 932 (53.78) | 822 (58.55) |
| Age, mean (SD), y | 49.58 (11.58) | 49.58 (11.59) | 48.40 (11.99) | 52.66 (9.73) | 46.89 (12.47) | 43.34 (12.61) | 51.27 (10.80) |
| Age group, y | |||||||
| 20-39 | 23 565 (19.78) | 4713 (19.78) | 3112 (24.04) | 715 (9.29) | 886 (28.24) | 674 (38.89) | 212 (15.10) |
| 40-59 | 73 225 (61.46) | 14 645 (61.46) | 7669 (59.25) | 5247 (68.16) | 1729 (55.12) | 879 (50.72) | 850 (60.54) |
| ≥60 | 22 355 (18.76) | 4471 (18.76) | 2113 (16.32) | 1736 (22.55) | 522 (16.64) | 180 (10.39) | 342 (24.36) |
| Comorbidities | |||||||
| Type 2 diabetes | 7128 (5.98) | 2650 (11.12) | 1272 (9.83) | 1187 (15.42) | 191 (6.09) | 79 (4.56) | 112 (7.98) |
| Hypertension | 16 436 (13.79) | 2388 (10.02) | 1440 (11.12) | 623 (0.08) | 325 (10.36) | 129 (7.44) | 126 (8.97) |
| Dyslipidemia | 2687 (2.26) | 168 (0.71) | 87 (0.67) | 37 (<0.01) | 44 (1.40) | 17 (0.98) | 27 (1.92) |
| Leukemia | 31 (0.03) | 1386 (5.82) | 4 (<0.01) | 4 (<0.01) | 1378 (43.92) | 1232 (71.09) | 146 (10.40) |
| Lymphoma | 59 (0.05) | 668 (2.80) | 6 (<0.01) | 27 (<0.01) | 635 (20.24) | 88 (5.08) | 547 (38.96) |
| Multiple myeloma | 10 (0.01) | 769 (3.23) | 4 (<0.01) | 3 (<0.01) | 762 (24.29) | 27 (1.56) | 735 (52.35) |
| Bone marrow failure | 22 (0.02) | 805 (3.38) | 59 (<0.01) | 68 (<0.01) | 678 (21.61) | 630 (36.35) | 48 (3.42) |
Abbreviations: HSCT, hematopoietic stem cell transplant; KT, kidney transplant; LT, liver transplant.
Data are presented as number (percentage) of patients unless otherwise indicated.
Table 2. Risks and IRs of Vitiligo in Patients With Transplant and Subgroup Analysis on the Risk of Vitiligo.
| Variable | Patients, No. | Duration, person-years | IR, per 1000 population | AHR (95% CI)a | |
|---|---|---|---|---|---|
| Total | With vitiligo | ||||
| Transplant | |||||
| No | 119 145 | 272 | 617 454.2 | 0.44 | 1 [Reference] |
| Yes | 23 829 | 119 | 114 054.40 | 1.04 | 1.73 (1.35-2.22) |
| Transplant type | |||||
| None | 119 145 | 272 | 617 454.2 | 0.44 | 1 [Reference] |
| Kidney | 12 994 | 42 | 65 554.7 | 0.64 | 1.50 (1.08-2.08) |
| Liver | 7698 | 27 | 36 524.0 | 0.74 | 1.63 (1.09-2.43) |
| HSCT | 3137 | 50 | 11 975.7 | 4.18 | 12.69 (5.11-31.50) |
| Transplant method | |||||
| None | 119 145 | 272 | 617 454.2 | 0.44 | 1 [Reference] |
| Kidney | 12 994 | 42 | 65 554.7 | 0.64 | 1.50 (1.08-2.08) |
| Liver | 7698 | 27 | 36 524.0 | 0.74 | 1.63 (1.09-2.43) |
| HSCT | |||||
| Allograft | 1733 | 46 | 6516.8 | 7.06 | 14.43 (5.61-37.15) |
| Autograft | 1404 | 4 | 5458.9 | 0.73 | 5.71 (1.20-3.18) |
| Transplant type and GVHD | |||||
| None | 119 145 | 272 | 617 454.2 | 0.44 | 1 [Reference] |
| Kidney | 12 994 | 42 | 65 554.7 | 0.64 | 1.50 (1.08-2.08) |
| Liver | 7698 | 27 | 36 524.0 | 0.74 | 1.63 (1.09-2.43) |
| HSCT | |||||
| Without GVHD | 2242 | 17 | 8808.5 | 1.93 | 8.21 (3.08-21.87) |
| With GVHD | 895 | 33 | 3167.1 | 10.42 | 24.09 (9.16-63.35) |
| HSCT | |||||
| None | 15 685 | 39 | 77 267.6 | 0.50 | 1 [Reference] |
| Without GVHD | 2242 | 17 | 8808.5 | 1.93 | 5.72 (2.01-16.28) |
| With GVHD | 895 | 33 | 3167.1 | 10.42 | 16.42 (5.84-46.14) |
| Allograft | |||||
| Without GVHD | 869 | 13 | 3440.6 | 3.78 | 6.05 (2.04-17.91) |
| With GVHD | 864 | 33 | 3076.2 | 10.73 | 16.81 (5.94-47.57) |
| Autograft | |||||
| Without GVHD | 1373 | 4 | 5368 | 0.75 | 4.45 (0.90-21.97) |
| With GVHD | 31 | 0 | 90.9 | 0 | NA |
Abbreviations: AHR, adjusted hazard ratio; GVHD, graft-vs-host disease; HSCT, hematopoietic stem cell transplant; IR, incidence rate; NA, not applicable.
Adjusted for age, sex, and comorbidities (type 2 diabetes, hypertension, hyperlipidemia, leukemia, lymphoma, multiple myeloma, and bone marrow failure disorders).
In subgroup analyses according to comorbid GVHD, HSCT recipients with comorbid GVHD had a significantly higher risk of vitiligo (AHR, 16.42; 95% CI, 5.84-46.14) compared with the control group without HSCT. The risk of vitiligo was significantly higher in the allogeneic graft HSCT group with comorbid GVHD than in the control group (AHR, 16.81; 95% CI, 5.94-47.57). Kaplan-Meier curves of the cumulative incidence rate of vitiligo showed significantly higher risk in the HSCT group, particularly among patients who underwent allogeneic HSCT or had comorbid GVHD compared with the control group (log-rank test, P < .001) (Figure).
Figure. Kaplan-Meier Estimates of the Cumulative Incidence Rates of Vitiligo.
GVHD indicates graft-vs-host disease; HSCT, hematopoietic stem cell transplant; KT, kidney transplant; LT, liver transplant; SOT, solid organ transplant.
Results of NHIS Health Checkup Data
Among the study population, a total of 84 586 individuals (transplant [n = 10 355]; control [n = 74 231]) with matching health checkup data were additionally analyzed. In analyses using the health checkup database (eTable in Supplement 1), the HSCT group (AHR, 10.32; 95% CI, 2.69-39.60), patients with an allogeneic graft (AHR, 12.80; 95% CI, 3.10-52.92), and patients with comorbid GVHD (AHR, 25.09; 95% CI, 6.08-103.60) had a significantly higher risk of subsequent vitiligo compared with the control group. However, no increased risk was observed in patients who underwent SOT, HSCT with autologous graft, or HSCT without GVHD compared with the control group.
Discussion
In this study, the risk of vitiligo was significantly higher in transplant recipients, with a higher risk observed in kidney and liver transplant recipients and an even greater risk in the HSCT group. Among HSCT recipients, the allogeneic HSCT group and those with comorbid GVHD had a higher risk compared with the autologous HSCT group and the group without GVHD. A previous population-based study in Korea found an association between HSCT and subsequent vitiligo compared with controls (odds ratio [OR], 3.130; 95% CI, 1.859-5.271) and demonstrated that the type of graft (allograft vs autograft) (OR, 5.593; 95% CI, 1.628-19.213) might be an independent factor associated with subsequent vitiligo, consistent with our study.6 However, another population-based study found that the risk of vitiligo was lower after SOT (OR, 0.305; 95% CI, 0.148-0.630).7 In our study, patients who underwent SOT had only a slightly higher risk of vitiligo and there was no significant difference in the health checkup data analysis. Unlike after HSCT, cases of GVHD are rare after SOT, which may contribute to the observed difference in vitiligo risk.9,10,11 The results of health checkup data analysis may differ from the initial analysis due to additional adjustments for lifestyle factors and inclusion of only patients who underwent a health checkup, potentially differing from the initial analysis that included all transplant recipients. Further research is needed to determine the risk of vitiligo after SOT and to identify the underlying mechanism.
Although the nature of the association still needs to be clarified, several hypotheses can be made. First, adoptive passive transfer of melanocyte autoantibodies from a donor to a recipient has been suggested, and several previous reports on new-onset vitiligo after allogeneic HSCT from a donor with vitiligo are available.12,13 Second, chronic GVHD might induce an immune response selective to melanocyte destuction.14 It is noteworthy that allogeneic HSCT recipients are more prone to developing GVHD compared with autologous HSCT recipients, which could contribute to the observed association.
Strengths and Limitations
Our study benefits from a nationally representative, large study population. In particular, because transplant recipients are covered by the Individual Copayment Beneficiaries Program for rare and intractable disorders, diagnosis is relatively reliable. Furthermore, we comprehensively analyzed the risk of vitiligo, including subgroup analysis of almost all transplant types and methods and host reactions, and we tried to adjust for various potential confounding factors, including lifestyle characteristics using the health checkup database, which was not previously examined, to our knowledge. Last, patients were followed up for a relatively long period of 10 years. This study has limitations, including reliance on NHIS claims data without medical record review; potential selection bias due to undiagnosed patients or those who did not visit the hospital; lack of detailed information on donors, recipients, disease severity, treatment history, and GVHD in the NHIS database; difficulty in distinguishing vitiligo from pigmentary changes related to cutaneous GVHD; potential overlap with previous studies conducted in Korea; and relatively wide 95% CIs, suggesting some uncertainty in the findings and emphasizing the need for careful interpretation.
Conclusions
This study found that risk of vitiligo was significantly higher in patients who had undergone HSCT and in those with comorbid GVHD. The findings suggest that early detection and management of vitiligo lesions can be improved by estimating the likelihood of its development in transplant recipients and implementing a multidisciplinary approach for monitoring.
eTable 1. Risks and Incidence Rates of Vitiligo in Transplant Patients According to a Health Check-Up Database
eFigure 1. Flowchart of Included Patients
Data Sharing Statement
References
- 1.Teulings HE, Limpens J, Jansen SN, et al. Vitiligo-like depigmentation in patients with stage III-IV melanoma receiving immunotherapy and its association with survival: a systematic review and meta-analysis. J Clin Oncol. 2015;33(7):773-781. doi: 10.1200/JCO.2014.57.4756 [DOI] [PubMed] [Google Scholar]
- 2.Nguyen J, Singh N, Afifi S, et al. Vitiligo following autologous hematopoietic stem cell transplantation. Clin Lymphoma Myeloma Leuk. 2020;20(4):e171-e173. doi: 10.1016/j.clml.2019.12.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Sanli H, Akay BN, Arat M, et al. Vitiligo after hematopoietic cell transplantation: six cases and review of the literature. Dermatology. 2008;216(4):349-354. doi: 10.1159/000117705 [DOI] [PubMed] [Google Scholar]
- 4.Bradley V, Kemp EH, Dickinson C, Key T, Gibbs P, Clatworthy MR. Vitiligo following a combined liver-kidney transplant. Nephrol Dial Transplant. 2009;24(2):686-688. doi: 10.1093/ndt/gfn661 [DOI] [PubMed] [Google Scholar]
- 5.Cho SB, Roh MR, Chung KY, Lee KH, Park YK. Generalized vitiligo after allogeneic bone marrow transplantation. Acta Derm Venereol. 2005;85(1):82-83. doi: 10.1080/00015550410021556 [DOI] [PubMed] [Google Scholar]
- 6.Bae JM, Choi KH, Jung HM, et al. Subsequent vitiligo after hematopoietic stem cell transplantation: a nationwide population-based cohort study from Korea. J Am Acad Dermatol. 2017;76(3):459-463. doi: 10.1016/j.jaad.2016.08.064 [DOI] [PubMed] [Google Scholar]
- 7.Bae JM, Choi KH, Choi JY, et al. Decreased risk of vitiligo in organ transplant recipients: a population-based cohort study. J Am Acad Dermatol. 2018;78(1):211-212. doi: 10.1016/j.jaad.2017.08.015 [DOI] [PubMed] [Google Scholar]
- 8.Rhee EJ, Han K, Ko SH, Ko KS, Lee WY. Increased risk for diabetes development in subjects with large variation in total cholesterol levels in 2,827,950 Koreans: a nationwide population-based study. PLoS One. 2017;12(5):e0176615. doi: 10.1371/journal.pone.0176615 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sheu J, Saavedra AP, Degar BA, et al. Bone marrow engraftment and associated dermatologic sequelae in a three-yr-old after liver transplantation. Pediatr Transplant. 2015;19(2):E41-E46. doi: 10.1111/petr.12412 [DOI] [PubMed] [Google Scholar]
- 10.Akbulut S, Yilmaz M, Yilmaz S. Graft-versus-host disease after liver transplantation: a comprehensive literature review. World J Gastroenterol. 2012;18(37):5240-5248. doi: 10.3748/wjg.v18.i37.5240 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Sharma A, Armstrong AE, Posner MP, et al. Graft-versus-host disease after solid organ transplantation: a single center experience and review of literature. Ann Transplant. 2012;17(4):133-139. doi: 10.12659/AOT.883704 [DOI] [PubMed] [Google Scholar]
- 12.Neumeister P, Strunk D, Apfelbeck U, Sill H, Linkesch W. Adoptive transfer of vitiligo after allogeneic bone marrow transplantation for non-Hodgkin’s lymphoma. Lancet. 2000;355(9212):1334-1335. doi: 10.1016/S0140-6736(00)02120-6 [DOI] [PubMed] [Google Scholar]
- 13.Alajlan A, Alfadley A, Pedersen KT. Transfer of vitiligo after allogeneic bone marrow transplantation. J Am Acad Dermatol. 2002;46(4):606-610. doi: 10.1067/mjd.2002.117215 [DOI] [PubMed] [Google Scholar]
- 14.Totani A, Amin H, Bacchi S, Lewis I. Vitiligo following stem-cell transplant. Bone Marrow Transplant. 2020;55(2):332-340. doi: 10.1038/s41409-019-0626-x [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Risks and Incidence Rates of Vitiligo in Transplant Patients According to a Health Check-Up Database
eFigure 1. Flowchart of Included Patients
Data Sharing Statement