Plasma Levels of Retinoids and Carotenoids in Recipients of Allogeneic Hematopoietic Stem Cell Transplantation Given Extracorporeal Photopheresis to Prevent Graft‐Versus‐Host Disease—Analyses From a Randomized Trial

ABSTRACT

Introduction

Whether retinoids and carotenoids impact on graft‐versus‐host disease (GVHD) following allogeneic stem cell transplantation (ASCT) is unknown.

Methods

We conducted a 1:1 randomized controlled trial with extracorporeal photopheresis for GVHD‐prophylaxis as the intervention. The plasma levels of retinoids and carotenoids were determined at randomization and 3 months after ASCT.

Results

We found no significant difference in GVHD occurrence and levels of retinoids between the intervention and control groups at either of the two time points.

Conclusion

Whether routine vitamin A supplementation is warranted for GVHD prevention, needs further study. Our exploratory study highlights the complex role of retinoid homeostasis in immune function during ASCT.

Trial Registration

ClinicalTrials.gov identifier: NCT03204721

1. Introduction

Graft‐versus‐host disease (GVHD) is a major complication of allogeneic stem cell transplantation (ASCT), and caused by donor‐derived T cells’ attack on recipient tissues. Among the factors influencing the differentiation and function of various subsets of T cells are derivatives of vitamin A (i.e., retinoids). Vitamin A and its active metabolites are important in GVHD because they strongly influence immune cell differentiation, tissue homing (especially to the gut), mucosal barrier function, and tissue repair—all processes that determine whether GVHD is amplified or restrained [1, 2]. Notably, recent studies have yielded conflicting results regarding the impact of vitamin A derivatives on GVHD outcomes. For example, Pan et al. demonstrated that supplementation with retinoic acid exacerbated intestinal GVHD‐severity in a murine model [3] whereas Aoyama et al. reported that inhibiting vitamin A signaling in donor T cells augmented intestinal GVHD symptoms in mice [4], indicating a nuanced role of retinoids that may involve both pro‐inflammatory and regulatory aspects in the context of ASCT. The complexity of vitamin A's role in GVHD is further evidenced by human studies, which have demonstrated an association between both high and low plasma levels of vitamin A derivatives and the incidence of GVHD [5, 6]. Moreover, a recent intention‐to‐treat randomized trial reported that supplementation of vitamin A treatment (as retinyl palmitate) prior to ASCT did not reduce the incidence of GVHD in children [7]. In line with this, low plasma levels of vitamin A prior to ASCT did not impact on GVHD risk among adult patients [8]. Collectively, these findings highlight the intricate balance required in retinoids homeostasis for optimal immune function in the ASCTs setting. To our knowledge, previous studies have not assessed in more detail whether plasma retinoids or plasma carotenoids (including pro‐vitamin A carotenoids) are associated with GVHD. We also included analysis of non‐vitamin A carotenoids because they modulate several processes that shape GVHD risk and severity such as oxidative stress, inflammation, immune cell function, epithelial and endothelial integrity, and the gut microbiome [9, 10]. These effects are mostly independent of vitamin A signaling.

Extracorporeal photopheresis (ECP) is a treatment modality for GVHD, with its mechanism of action likely involving the modulation of T cell activity. While it has shown promise in managing established GVHD, its role in GVHD prophylaxis remains largely unexplored. To address this gap in knowledge, we conducted a randomized controlled trial (RCT) investigating the efficacy of ECP as GVHD prophylaxis in patients undergoing ASCT [11]. The intention‐to‐treat analysis did not reveal any difference in GVHD prevalence during the first year post‐ASCT (primary endpoint), but a per‐protocol analysis indicated that the patients who completed the full course of eight ECP cycles experienced about a 50% reduction in GVHD prevalence [11].

We here used data from this RCT to determine the plasma concentrations of a wide range of retinoids and carotenoids at the time of randomization and 3 months later. Such knowledge is important for understanding the possible roles of vitamin A and carotenoids in GVHD.

2. Methods

We conducted a two‐armed open RCT between 2017 and 2020. Adults (age ≥ 18 years) scheduled for an ASCT due to hematological malignancy were eligible for the RCT [11]. Patients in the intervention group were randomized to receive eight prophylactic ECP treatment‐cycles over 6 weeks in addition to standard GVHD prophylaxis, whereas the controls received only standard prophylaxis. Plasma samples available for analyses were collected at the time of randomization and 3 months post‐ASCT. A full trial and methodological description can be found in the report by Ali et al. [11] and in the Supporting Information.

3. Results and Discussion

We randomized 76 patients to the intervention group, whereas 77 were controls. Relevant characteristics of the total study cohort are given in the Table S1 (see also Ali et al. [11]). Three months post‐transplantation, five patients in the intervention group and four controls had died (p = 0.72), whereas 21 patients in the intervention group and 23 controls had developed acute GVHD (Grade II–IV; p = 0.76), none developed overlap syndrome or clinical signs indicating chronic GVHD.

Figure 1 shows the plasma concentrations of the retinoids at the two data collection time points. A marked increase in the plasma concentration of all‐trans‐retinol was identified at 3 months in both study groups. The plasma concentrations of all three retinoids were not significantly different between the intervention and control groups at either of the two time points. Figure 2 shows the plasma concentrations of carotenoids (lutein, zeaxanthin, β‐cryptoxanthin, α‐carotene, β‐carotene, and lycopene). None of the provitamin A carotenoids increased after 3 months in the intervention group, while β‐carotene increased among the controls. The non‐provitamin A carotenoid lycopene increased in both study groups.

FIGURE 1.

Plasma concentrations of retinoids at the time of randomization and 3 months after ASCT. Data are from all‐trans‐retinol (a), all‐trans‐retinoic acid (b), and 13‐cis‐retinoic acid (c). Values are shown as box plots in green (intervention group) and red color (control group). Unadjusted (Bonferroni‐adjusted) p‐values are shown above the box plots.

FIGURE 2.

Plasma concentrations of carotenoids at the time of randomization and 3 months after ASCT. Data are from lutein (a), zeaxanthin (b), β‐cryptoxanthin (c), α‐carotene (d), β‐carotene (e), and lycopene (f). Values are shown as box plots in green (intervention group) and red color (control group). Unadjusted (Bonferroni‐adjusted) p‐values are shown above the box plots.

Since the two study groups were quite similar in most characteristics, and both the prevalence of deaths and GVHD were similar in the two study groups 3 months after ASCT, we pooled the two study groups. Figures S1 and S2 show similar plasma concentrations of retinoids and carotenoids, respectively, in those who had been diagnosed with GVHD and those who were GVHD‐free after 3 months.

To the best of our knowledge, this is the first report of plasma concentrations of retinoids and carotenoids among ASCT recipients participating in a RCT with occurrence of GVHD as the primary endpoint. Among the retinoids, all‐trans‐retinol increased from the time of randomization to 3 months post‐ASCT in the intervention group. The carotenoids showed an inconsistent pattern with time and between the two study groups. This may reflect the diverse functions of each of these carotenoids. Notably, we did not detect any impact of whether the patients had developed GVHD on their plasma concentrations of either retinoids or carotenoids. Also, there was no difference in the immune reconstitution up to 3 months, as earlier described [11].

A study limitation is that the original RCT with prophylactic ECP was not primarily designed to investigate plasma retinoids or carotenoids. We also acknowledge the relatively small sample size and that blood samples from only two time points were available for analyses. Given the nature of (i) a secondary analysis, (ii) inclusion of only two time points, and (iii) the limited sample size, we cannot rule out that factors, such as altered dietary intakes, impaired organ function (i.e., of the liver and/or gut), concurrent inflammation/infection and treatments with immunomodulatory drugs such as steroids, may have impacted the plasma concentrations of retinoids and carotenoids. A major strength of this study is the RCT‐design, which reduces confounding and enhances interpretability. In addition, a wide range of retinoids and carotenoids were included.

In conclusion, we could not identify any effect of ECP on the plasma concentrations of retinoids 3 months after ASCT. In contrast, the plasma concentrations of carotenoids showed more variability and thus a possible effect of ECP cannot be excluded for some of them (i.e., β‐cryptoxanthin and α‐carotene). Importantly, neither the plasma concentrations of retinoids nor of carotenoids seemed to depend on whether the patients developed GVHD or not. Whether or not routine supplementation of vitamin A derivatives should be given to prevent GVHD warrants further study.

Author Contributions

P.O.I. analyzed the data and drafted the manuscript. M.M.A. collected and analyzed the data. N.B. measured the vitamin A derivatives. R.B., G.E.T., and T.G.D. analyzed the data. All authors designed the study, interpreted the data, provided inputs on manuscript drafts, and approved the submitted version of the manuscript.

Funding

This study received funding from the Foundation at Oslo University Hospital and the Throne Holst Foundation.

Ethics Statement

The study was approved by the Regional Committee for Medical and Health Research Ethics of South‐East Norway (2016/1130), the Data Protection Officer, Oslo University Hospital (2016/9408), and the Norwegian Medicines Agency (EUDRACT 2017‐000614‐39), and was performed according to the principles of Good Clinical Practice.

Conflicts of Interest

Rune Blomhoff is a shareholder of Vitas Ltd, Oslo, Norway.

Supporting information

Supporting Information

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.