Article Text

Download PDFPDF

Risk factors for the development of ocular graft-versus-host disease (GVHD) dry eye syndrome in patients with chronic GVHD
  1. Jay Ching Chieh Wang1,
  2. Joshua C Teichman2,
  3. Majd Mustafa1,
  4. Heather O'Donnell1,
  5. Raewyn Broady3,
  6. Sonia N Yeung1
  1. 1Department of Ophthalmology and Visual Sciences, University of British Columbia, Vancouver, British Columbia, Canada
  2. 2Department of Ophthalmology, University of Ottawa, Ottawa, Ontario, Canada
  3. 3Bone Marrow Transplant Unit, Department of Medicine, University of British Columbia, Vancouver, British Columbia, Canada
  1. Correspondence to Dr Sonia N Yeung,  Department of Ophthalmology and Visual Sciences, University of British Columbia, 616-2525 Willow Street, Vancouver, British Columbia, Canada V5Z 1K1; sonia.y{at}gmail.com

Abstract

Background To investigate the factors associated with the development of ocular graft-versus-host disease (oGVHD) dry eye syndrome (DES) in patients with chronic GVHD (cGVHD) after receiving allogenic haematopoietic stem cell transplantation (AHSCT)

Methods A retrospective chart review of patients receiving AHSCT between 1998 and 2013 at the Bone Marrow Transplant Unit of the British Columbia Cancer Agency was carried out. Demographic and clinical data from both donors and recipients were obtained. The diagnostic criteria for the development of oGVHD DES from the National Institutes of Health were used to identify patients with the disease. Descriptive and inferential statistics were carried out.

Results A total of 146 patients with a median follow-up time of 24.0 months (range 11.3–249.7 months) were included in this study. Sixty-six (45.2%) patients were women. Seventy-seven (52.7%) patients had oGVHD DES. The median age of patients was 57 years (range 25–71 years). Compared with other ethnicities, Caucasian patients were less likely to develop oGVHD DES, with an OR of 0.29 (p=0.01). Patients who received a transplant from Epstein–Barr-positive donors had a higher prevalence of oGVHD DES (OR=4.39, p=0.01). This was also found in patients with the following systemic involvement of cGVHD: grade 1–3 cGVHD skin involvement (OR=1.57, p=0.01), oral involvement (OR=2.51, p=0.01) and liver involvement (p=0.04). Patients with grade 2–3 overall cGVHD were also more susceptible to oGVHD DES (OR=2.72, p<0.001).

Conclusions This study identified risk factors associated with a higher prevalence of oGVHD DES in post-AHSCT patients with cGVHD.

  • Cornea
  • Ocular surface

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Chronic graft-versus-host disease (cGVHD) is a potentially life-threatening complication that can result from allogeneic haematopoietic stem cell transplantation (AHSCT), where the donor T cells attack the tissues of transplant recipients.1–4 The condition may involve various organ systems, including but not limited to, skin, lung, liver, gastrointestinal (GI) tract and the eyes.5

Ocular involvement of GVHD, termed ocular GVHD (oGVHD) in this study, can develop in approximately 60% of AHSCT patients.6–8 The most common manifestation of oGVHD is dry eye syndrome (DES), which results from lacrimal and conjunctival infiltration of stromal fibroblasts.6–8 Though not life-threatening, oGVHD DES can greatly hinder patients’ quality of life and may be visually debilitating.5 ,7 ,9 ,10 The risk factors for developing oGVHD DES are not well understood. The purpose of this study was to elucidate the factors associated with the development of oGVHD DES in patients with cGVHD.

Materials and methods

Study subjects

This was a retrospective chart review consisting of 146 post-AHSCT patients followed at the Leukemia and Bone Marrow Transplant Unit of the British Columbia Cancer Agency between 1998 and 2013.

The following clinical data from transplant recipients were collected: age, age at transplant, sex, ethnicity, length of follow-up time, ABO blood group, Rh status, infectious disease status, donor-to-recipient human leucocyte antigen results, transplantation indication, transplant tissue source, pretransplant conditioning regimen, donor-to-recipient relationship, overall cGVHD severity and the severity of individual organ systems involved in cGVHD. The infectious diseases screened for included cytomegalovirus (CMV), Epstein–Barr virus (EBV), herpes simplex virus (HSV), herpes zoster virus (HZV), varicella zoster virus and hepatitis C virus (HCV). The organ systems of interest affected in cGVHD included skin, hair and nails, liver, mouth, GI track, lungs, musculoskeletal system, reproductive tract and the haematological system.

The following clinical data were also obtained from the donor population: sex, age at transplant, ABO blood group, Rh status and infectious disease status as mentioned above.

NIH criteria for the diagnosis of system-specific cGVHD and severity grading

For the purpose of diagnosing and evaluating the cGVHD involvement and severity of each organ system in this study, the National Institutes of Health (NIH) guidelines proposed in 2005 were used.5 According to the guidelines, a diagnosis of ocular GVHD DES associated with cGVHD can be made in post-AHSCT patients when the following signs develop:

  1. “new ocular sicca documented by low Schirmer test values with a mean value of both eyes ≤5 mm at 5 minutes;” OR,

  2. “new onset of keratoconjunctivitis sicca by slit-lamp examination with mean values of 6 to 10 mm on the Schirmer test” that is “accompanied by distinctive manifestations in at least 1 other organ.”

The patients were determined to not have oGVHD DES if they did not meet the NIH criteria by the last follow-up.

Statistical analysis

Descriptive and inferential statistics were performed using Stata (V.13, StataCorp, College Station, Texas, USA). Normally distributed data were analysed with the Student's t-test when there were two groups of continuous variables and with analysis of variance when there were more than two groups of continuous variables. Data that were not normally distributed data were analysed with the Mann–Whitney U (rank sum) test for comparisons between two groups and with the Kruskal–Wallis test for comparisons between more than two groups of variables. Dichotomous data were analysed using the χ2 test. Ordered logistic regression and logistic regression analyses were performed on ordinal and dichotomous data, respectively. Statistical significance was set at p<0.05.

Results

Study population demographics

Of the 146 patients, 77 had oGVHD DES while 69 did not. The median age of the patients was 57 years old (range 25–71 years old), and the median age at transplant was 53.5 years old (range 22–66 years old). There were 80 men (54.8%) and 66 women (45.2%). The median follow-up time for all patients was 24.0 months (range 11.3–249.7 months). The oGVHD DES group had a median follow-up time of 31.9 months (range 2.0–249.7 months) compared with 17.0 months (2.0–67.0 months) for those who did not develop oGVHD DES (p=0.0004) by the time of their most recent follow-up appointment. With respect to ABO blood groups, 65 patients (45.1%) had type A blood group, 55 patients (35.4%) had type O blood group, 20 patients (13.9%) had type B blood group and 8 patients (5.6%) had type AB blood group. In total, 122 (84.7%) patients were Rh-positive while 22 (15.3%) patients were Rh-negative. In terms of ethnicity, 111 (76.0%) were Caucasian, 28 (19.2%) patients were Asian and 7 (4.8%) patients were of other ethnicities. Also, 132 patients (93.6%) were positive for HZV, 129 (90.2%) for EBV, 108 (74.5%) for HSV, 88 (60.3%) for CMV and 2 (1.4%) for HCV.

Comparisons between patients with and without oGVHD DES

Our analysis revealed a number of differences between patients with and without oGVHD DES. It was found that Caucasian patients were less likely to have oGVHD DES compared with patients of either Asian or other ethnicities (OR=0.29, p=0.01, figure 1). χ2 test revealed that Caucasian patients were not more likely to have had transplants from related donors than non-Caucasian patients (p=0.46). Also, patients who received transplants from donors that were EBV-positive were more likely to have oGVHD DES compared with patients who received transplants from donors that were EBV-negative (OR=4.39, p=0.01, figure 2).

Figure 1

Caucasian patients were less likely to develop ocular graft-versus-host disease (oGVHD) dry eye syndrome (DES). In this study, patients with chronic GVHD with and without oGVHD DES were categorised by ethnicity and compared by Pearson's χ2 test. Analysis revealed that Caucasian patients, compared with Asians and other ethnicities, were less likely to have oGVHD DES (p=0.03) with an OR of 0.29 (p=0.01) on logistic regression. KCS, keratoconjunctivitis sicca.

Figure 2

Patients who received transplants from Epstein–Barr virus (EBV)-positive donors were at higher risk for developing ocular graft-versus-host disease (oGVHD) dry eye syndrome (DES). Patients with chronic GVHD with and without oGVHD DES were categorised by the EBV infectious status of their transplant donors and compared by Pearson's χ2 test. Analysis revealed that patients who received transplants from EBV-positive donors were more likely to develop oGVHD DES (p=0.01), with an OR of 4.39 (p=0.01) on logistic regression. KCS, keratoconjunctivitis sicca.

With respect to cGVHD, it was found that patients with grade 2–3 overall cGVHD were more likely to develop oGVHD DES in comparison to patients with no cGVHD or grade 1 overall cGVHD (OR=2.72, p<0.001, figure 3). In addition, patients with grade 1–3 cGVHD oral (OR=2.51, p=0.01, figure 4) or skin (OR=1.57, p=0.01, figure 5) involvement were more likely to develop oGVHD DES compared with patients without involvement in the respective sites. This was also found for patients with cGVHD liver involvement compared with those without liver involvement (p=0.04, figure 6).

Figure 3

Patients with moderate-to-severe chronic graft-versus-host disease (cGVHD) were at risk for developing ocular GVHD (oGVHD) dry eye syndrome (DES). Patients with cGVHD with and without oGVHD DES were categorised by the severity of their cGVHD and compared by the Pearson's χ2 test. The severity was graded by the National Institutes of Health criteria. Analysis revealed that patients with moderate and severe cGVHD were more likely to develop oGVHD DES compared with patients without or with grade 1 cGVHD (p<0.001), with an OR of 2.72 (p<0.001) on logistic regression. KCS, keratoconjunctivitis sicca.

Figure 4

Patients with mouth involvement were at risk for developing ocular graft-versus-host disease (oGVHD) dry eye syndrome (DES). Patients with chronic GVHD (cGVHD) with and without oGVHD DES were categorised by the severity of their cGVHD mouth involvement and compared by the Pearson's χ2 test. The severity was graded by the National Institutes of Health criteria. Analysis revealed that patients with cGVHD mouth involvement were more likely to develop oGVHD DES compared with patients without mouth involvement (p<0.01), with an OR of 2.51 (p=0.01) on logistic regression. KCS, keratoconjunctivitis sicca.

Figure 5

Patients with skin involvement were at risk for developing ocular graft-versus-host disease (oGVHD) dry eye syndrome (DES). Patients with chronic GVHD (cGVHD) with and without oGVHD DES were categorised by the severity of their cGVHD skin involvement and compared by the Pearson's χ2 test. The severity was graded by the National Institutes of Health criteria. Analysis revealed that patients with cGVHD skin involvement were more likely to develop oGVHD DES compared with patients without skin involvement (p=0.04), with an OR of 1.57 (p=0.01) on logistic regression. KCS, keratoconjunctivitis sicca.

Figure 6

Patients with liver involvement were at risk for developing ocular graft-versus-host disease (oGVHD) dry eye syndrome (DES). Patients with chronic GVHD (cGVHD) with and without oGVHD DES were categorised by the severity of their cGVHD liver disease and compared by Pearson's χ2 test. The severity was graded by the National Institutes of Health criteria. Analysis revealed that patients with cGVHD liver disease were more likely to develop oGVHD DES compared with patients without liver involvement (p=0.04). KCS, keratoconjunctivitis sicca.

Interestingly, the types of pretransplant conditions received by patients were not different between patients with and without oGVHD DES (p=0.416).

Discussion

The present study identified risk factors for the development of oGVHD in a population suffering from chronic systemic GVHD after AHSCT. By identifying those patients at risk for developing oGVHD, the findings of this study may assist bone marrow transplant clinicians and ophthalmologists who evaluate and monitor post-AHSCT cGVHD patients for the development of oGVHD DES.

Our findings support previous work showing that systemic cGVHD involvement is associated with the development of oGVHD DES. Our study found that patients with cGVHD who had involvement of the skin (OR=1.57), mouth (OR=2.51) or liver were at increased risks for developing oGVHD DES. This is in agreement with a previous report by Westeneng et al,11 who found that skin and mouth involvement were both risk factors for oGVHD DES. We also found that patients with moderate-to-severe systemic cGVHD were at an increased risk for developing oGVHD DES (OR=2.72), which is also consistent with previous reports.11–13 Of note, no significant OR was found for patients with liver involvement.

Two risk factors identified in this study have not been previously reported. First, Caucasian patients were less likely to develop oGVHD DES in comparison to Asian and other ethnicities (OR=0.29). The cause for this finding was unclear from the current literature. However, a previously published report suggested that Caucasians may exhibit lower prevalence and severity of DES.14 Furthermore, Eckrich et al15 previously found that Caucasians had lower incidence of GVHD and mortality after AHSCT for aplastic anaemia compared with African-Americans. Future studies on the relationship between ethnicity and GVHD may be helpful in further clarifying this issue.

This study also found that patients who received transplants from EBV-positive donors were more likely to develop oGVHD DES compared with patients who received transplants from EBV-negative donors (OR=4.39). No relevant literature was found that specifically discussed the relationship between EBV and oGVHD DES. However, previous studies provided evidence to suggest the potential role of EBV in the pathogenesis of Sjögren’s syndrome (SS).16–19 The salivary gland biopsies of patients with SS contained higher levels of EBV DNA compared with their healthy counterparts. This may trigger an exaggerated immune response and cause salivary gland destruction.16–19 It is possible that patients who received AHSCT from EBV-positive donors may be at risk for developing oGVHD DES through similar mechanisms. Again, further studies may help to clarify this relationship.

It was previously reported that male patients who received transplants from female donors were at increased risk for oGVHD DES.12 ,20 This study did not replicate those findings. Furthermore, the role of pretransplant conditioning on the development of oGVHD had not previously been investigated. This study did not find a significant difference in the incidence of oGVHD DES in relation to the patient's indication for transplantation, specific pretransplant conditioning regimen and donor–recipient relation. No empiric ocular therapies were used in pretransplant or post-transplant regimens.

Previous studies have examined the length of time for the onset of oGVHD DES. Ogawa et al21 reported a median onset time of 171 days, or approximately 5.7 months, in a group of 53 patients. Tichelli et al22 reported a median onset time of 13.8 months in a group of 48 patients. Most recently, Shikari et al23 reported a median time of onset of 293 days, which is approximately 9.8 months, in a group of 200 patients. The median onset time of oGVHD DES was not examined in this study.

One weakness of this study is the retrospective nature of its design, which could lead to selection bias. Furthermore, ophthalmic histories prior to transplants were not available for all subjects. Therefore, the influence of ocular diseases prior to bone marrow transplantation on the incidence and severity of oGVHD DES was unclear. Further studies are warranted to clarify this. In addition, the p value of a number of patient characteristics was close to the cut-off, and a larger sample size may help to clarify the significance of these factors. It can be argued that the use of the NIH criteria to diagnose oGVHD DES is a weakness of this study since it does not take into account tear break-up time or ocular surface stains. Recently, the International Chronic GVHD Consensus Group proposed new diagnostic criteria for oGVHD. It took into account the ocular surface disease index, Schirmer's score without anaesthesia, corneal staining and conjunctival injection.10 However, according to the paper by Ogawa et al10, the newly proposed criteria have yet to be validated for clinical use. As such, the NIH criteria were arguably the most appropriate guidelines available at the time the study was carried out.

Conclusion

The present study compared the characteristics between post-AHSCT cGVHD patients with and without oGVHD DES. The results of this study suggest that Caucasians are at lower risk for the development of oGVHD DES. Moreover, patients who received transplants from EBV-positive donors and patients who suffer from skin, liver, mouth or moderate-to-severe cGVHD are at increased risk for the development of oGVHD DES.

Acknowledgments

We would like to thank Janet Nitta of the Leukemia and Bone Marrow Transplant Program of British Columbia for her help in collection of patient data.

References

Footnotes

  • Contributors JCCW: conception and design, analysis and interpretation of data, writing of article, critical revision of the article, final approval, data collection and literature search. JCT: conception and design of study, analysis and interpretation, and statistical expertise. HO: data collection and paper revision. MM: article revision and data collection. RB: provisions of patient charts and administrative and logistical support. SY: primary investigator, conception of study, analysis and interpretation, writing of article, data collection and provision of patient charts.

  • Competing interests None declared.

  • Ethics approval University of British Columbia (UBC) Research Ethics Board. All research methodology followed the tenets of the Declaration of Helsinki.

  • Provenance and peer review Not commissioned; externally peer reviewed.