Article Info

Author Affiliations: Key Laboratory of Combined Multi-organ Transplantation, Ministry of Public Health, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, the First Affiliated Hospital, Zhejiang University School of Medicine, 79 Qingchun Road, Hangzhou 310003, China (Meng XQ, Chen XH, Sahebally Z, Xu YN, Yin SY, Wu LM and Zheng SS)

Corresponding Author: Shu-Sen Zheng, MD, PhD, FACS, Key Laboratory of Combined Multi-organ Transplantation, Ministry of Public Health, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, the First Affiliated Hospital, Zhejiang University School of Medicine, 79 Qingchun Road, Hangzhou 310003, China (Tel: +86-571-87236570; Fax: +86-571-87236466; Email: shusenzheng@zju.edu.cn)

Contributors: YSY and ZSS conceived and designed the experiments. CXH collected samples. YSY contributed reagents. MXQ and XYN performed the experiments. WLM analyzed the data. MXQ and CXH wrote the manuscript. All authors contributed to the design and interpretation of the study and to further drafts. ZSS is the guarantor.

BACKGROUND: Graft-versus-host disease (GVHD) is associated with high mortality. Early diagnosis is essential to start treatment and to improve outcomes. Because of the inflammatory nature, we hypothesis that cytokine profile of patients with GVHD may serve as diagnostic markers. The present study was to evaluate the role of cytokine profile in the diagnosis of GVHD.

METHODS: An immunoassay was used to detect 29 cytokines simultaneously in the serum; the measuring sensitivity of all cytokines was pg/mL. Healthy subjects undergoing annual routine physical examinations served as negative controls; 23 patients with hepatocellular carcinoma (HCC) who had undergone liver transplantation (the LT group) comprised the test subjects. A total of 22 kidney recipients with biopsy-confirmed GVHD (the RT group) were included for comparison. HCC patients with radical surgery (the HCC group, n=22) served as positive control. The liver contents of the three cytokines, IL-2, IL-18, and IFN-γ, were detected with immunohistochemistry. Serum granzyme B and perforin were measured by flow cytometry.

RESULTS: Of the 29 cytokines, the levels of IL-2 and IL-18 were increased significantly in liver recipients with GVHD compared with healthy controls (P<0.05). The serum levels of these three cytokines in the healthy, HCC, LT, and RT groups were IL-2: 0.90±0.02, 4.14±0.61, 5.10±0.89, and 1.48±0.09 pg/mL; IL-18: 80.61±9.35, 109.51±10.93, 230.11±12.92, and 61.98±7.88 pg/mL; IFN-γ: 24.06±3.88, 24.84±3.21, 40.37±5.88, and 15.33±4.72 pg/mL, respectively. Immunohistochemistry showed that these 3 cytokines expressions in the liver were parallel to the serum cytokine. After standard anti-GVHD treatment, the expressions of IL-2, IL-18, and IFN-γ were decreased in the liver (P<0.05). Serum granzyme B and perforin were significantly increased in GVHD patients (P<0.05).

CONCLUSIONS: IL-2, IL-18 and IFN-γ were from liver and might serve as biomarkers for monitoring GVHD development and the effects of anti-GVHD treatment. Granzyme B and perforin may play a role in increasing IL-2, IL-18, and IFN-γ levels in GVHD patients.

Graft-versus-host disease (GVHD) is a medical complication following the receipt of transplanted tissue from a genetically different person. It is the fifth cause of death following liver transplantation. Data from liver transplant registries^[1-3] show that China has the largest number of patients with HCC on transplant waiting lists; almost half of all liver transplants are performed in patients with HCC. Previously, we reviewed follow-up data on 6012 Chinese patients with HCC who underwent liver transplantation and found that the major causes of death post-transplantation were hemorrhage, infection, graft failure, multiple organ dysfunction syndrome, and GVHD. Of these conditions, GVHD is very challenging for clinicians, as it is difficult to diagnose due to similarities between the clinical presentations of drug reactions or viral infections, especially cytomegalovirus (CMV) disease.^[4] Early non-invasive diagnostic biomarkers for GVHD are urgently needed.

GVHD after orthotopic liver transplantation (OLT) is a severe complication. The incidence is 1%-2% and the mortality rate 85%-90%. Recent studies have found that cytokines modulate both T-helper type 1 (Th1) and Th2 responses. Cytokine storms are evident in patients with GVHD.^{[5, 6]} The serum concentrations of many cytokines (e.g., IL-18) are elevated in experimental models of GVHD.^[5] However, the cytokine profile of the cytokine storm has not been defined. Most relevant studies have used dedicated sample processing, time-consuming flow cytometry, and complicated data analysis;^[7-9] hindering early diagnosis and treatment. In this study, we established a cytokine examination protocol using routine blood samples. We described a high-throughput method that simultaneously measures the levels of multiple cytokines. This is especially valuable for early non-invasive diagnosis of GVHD in patients after transplantation. We also analyzed an HCC cohort that had not undergone transplantation, and healthy controls. We describe serum diagnostic platform for GVHD after organ transplantation; this is not organ-specific and is not compromised by any underlying disease.

We screened the medical records of patients who underwent organ transplantation in our center and included 87 patients for whom long-term follow-up data were available. Forty-five transplantation patients [23 HCC patients who underwent liver transplantation (the LT group) and 22 patients who underwent renal transplantation (the RT group)] with biopsy-confirmed GVHD were included. HCC patients who had not undergone transplantation (HCC group, n=22) and healthy subjects (healthy control, n=20) served as positive and negative controls, respectively. GVHD was confirmed histologically after physical examination and virus and microbiology screening to exclude differential diseases such as continuing bacterial and fungal infections. Treatment for GVHD: methylprednisolone (0.5 g/d) for 3 days and 2 doses of Zenapax (50 mg).

This study was approved by the Ethics Committee of the First Affiliated Hospital, Zhejiang University School of Medicine. Written informed consent for blood sample collection was obtained in advance. All clinical information was retrieved from the Transplant Recipients Database described in previous publications.^{[11, 12]}

Cytokine levels were measured using the Procarta Plex™ Multiplex Immunoassay Kit (Affymetrix; Foster City, CA, USA) as described in the Procarta manual and previous publications.^[13-15] In this assay, multianalyte-profiling beads are used to detect multiple cytokines employing fluorescent dye technology. The use of two lasers and digital signal processing effectively allows multiplexing of up to 50 unique assays within a single sample. Blood samples were centrifuged at 1000 g at 4 �� for 10 minutes, and the sera were stored below -20 �� prior to simultaneous one-time cytokine measurements. Twenty-five µL of serum from each patient were incubated with antibodies against human cytokines in 96-well plates. After rinsing, the plates were incubated with multiple antibodies and the reactions detected using a streptavidin-phycoerythrin combination; data were quantified using the Luminex²⁰⁰ system (Bio-Rad; Shanghai, China). Quality control featured plotting standard curves plotted using dilutions of the reconstituted antigens; the curves were constructed using Certificate Analysis Software.

Pathological slides containing biopsy material collected at the time of the initial diagnosis (GVHD onset) were compared with those containing material collected at the time of the second biopsy (after standard anti-GVHD therapy). Tissues were fixed in 10% (v/v) buffered formalin overnight and embedded in paraffin. Antibodies against IL-2, IL-18, and IFN-γ (Santa Cruz Biotechnology, Santa Cruz, CA, USA) were used for immunohistochemistry. The quality control was samples from patients with inflammatory bowel disease, in which IL-2, IL-18, and IFN-γ are expressed at stable levels. Staining was quantified using Image J software (the Chinese version developed by the National Institutes of Health).

After liver or renal transplantation, GVHD generally develops when donor lymphocytes mount an alloreactive response against host histocompatibility antigens. Patients develop fever, rash, diarrhea, and pancytopenia. Blood tests are run and biopsies are performed when GVHD is suspected. Blood samples were collected from patients who developed fever or a rash after liver or renal transplantation; GVHD was confirmed by biopsy.^[16] We found that GVHD developed several months post-transplantation. Thus, control blood samples from HCC patients were collected at a similar time after radical surgery, ensuring that surgical intervention in the controls would have the same impact as that in the transplantation groups.

We used flow cytometry to measure the levels of monoclonal antibodies binding to granzyme B and perforin (BD Pharmingen, San Jose, CA, USA). Blood samples (100 µL) were added to tubes containing 5 µL of antibody solutions against granzyme B or perforin, mixed in the dark, and incubated at room temperature for 15 minutes. Flow cytometry using a Beckman Coulter platform employed both live and logical gating strategies. Data on a minimum of 10 000 fluorescent dots were analyzed.

All data are expressed as means±SD. Comparisons were performed using Student’s t test. A P value of <0.05 was considered statistically significant. One-way ANOVA was used to compare among-group differences. All analyses were performed using SPSS software, version 17.0 (Chicago, IL, USA).

After reviewing medical histories, we found that GVHD manifested not as liver or kidney problems but rather as other-organ features such as rash, fever, or diarrhea. The clinical symptoms are critical in terms of differential diagnosis. We found that GVHD developed after recovery from surgery at 253±39 and 192±51 days post-liver and post-renal transplantations, respectively. Accurate diagnoses were made using classical histological and immunohistochemical criteria.^[16]

Compared with healthy controls, the levels of IL-2 and IL-18 were significantly elevated in liver recipients with GVHD (Fig. 1). The levels of these 3 cytokines in the healthy controls, HCC, LT, and RT groups were IL-2: 0.90±0.02, 4.14±0.61, 5.10±0.89, and 1.48±0.09 pg/mL; IL-18: 80.61±9.35, 109.51±10.93, 230.11±12.92, and 61.98±7.88 pg/mL; IFN-γ: 24.06±3.88, 24.84±3.21, 40.37±5.88, and 15.33±4.72 pg/mL, respectively. The assay sensitivities ranged from 0.5 pg/mL (IL-2) to 1.22 pg/mL (IL-18).

GVHD was histologically confirmed using routine methods. Hematoxylin and eosin (H&E)-stained slides of liver biopsy material revealed lymphocytic infiltration, damage to vessel ducts, portal space congestion, and duct epithelial destruction (Fig. 2). The levels of IL-2, IL-18, and IFN-γ were significantly higher in the GVHD onset patients. After standard treatment for GVHD, the quantitative evaluation of IL-2, IL-18, and IFN-γ were decreased significantly (P<0.05, Fig. 3).

In efforts to define early diagnostic biomarkers predicting GVHD development, granzyme B and perforin levels were determined in peripheral blood CD8 T cells from healthy controls, HCC patients, and GVHD patients receiving liver/renal transplants. Both granzyme B and perforin levels were elevated in GVHD patients compared with healthy controls (flow counts are shown in Fig. 4A, and numerical data from three repeat experiments are shown in Fig. 4B). These suggested that the granzyme B/perforin pathway is involved in GVHD development.

In our liver and renal transplantation practice, the recognization of GVHD is based on specific clinical and biopsy findings, rather than onset time. The clinical signs in skin, gastrointestinal tract, and liver are always correlated with pathological features. The routine pathological exam (H&E, Masson trichrome stain and periodic acid-schiff) can provide the fundamental proof of the hepatic damage and the involvement of biliary epithelium.

As emphasized above, GVHD diagnosis should be based on characteristic clinical symptoms of the target organ and pathological data on biopsy. In clinical practice, however, the most frequent GVHD symptoms (rash, fever, and diarrhea) are so nonspecific that it becomes very difficult to differentiate GVHD from a drug rash, general infection, venous occlusive disease, viral resurrection, or treatment-associated toxicity. Laboratory assays for human leukocyte antigen (HLA), DNA studies, and invasive biopsies require time, specialized equipment, and professional expertise, compromising a timely diagnosis and delaying treatment. It is thus urgent to develop an objective, non-invasive and quantitative method for GVHD diagnosis.

Immune system analyses post-transplantation are effective to this end but are difficult to perform. Analyses of multiple immune markers are very time-consuming and require many reagents, decreasing the accuracy of parallel comparisons. Here, we offered a solution by identifying cytokine biomarkers of early GVHD and providing reference baseline data. Serum cytokine levels and pathological data afford comprehensive evaluation, making it unnecessary to explore organ specificity or underlying disease.

Previous studies in mice suggested that IL-18 induces the cytokine storm and is critical in terms of GVHD-induced injury,^{[5, 6]} as well as injuries associated with other inflammatory diseases.^[17-23] Thus, we measured the levels of IL-18 and other cytokines in human patients with GVHD. To optimize the use of small amounts of serum, a multiplex immunoassay was used to measure many cytokines in a single run. This minimizes blood sample requirements (only 25 µL of serum are required), avoids among-group variation, and maximizes quality control.

IL-2 and IL-18 levels were all significantly higher in liver recipients with GVHD compared with those in healthy controls (P<0.05). Pathology showed that the cytokines were expressed principally in inflammatory cells infiltrating the portal space of GVHD patients. After treatment of GVHD, IL-2, IL-18 and IFN-γ levels decreased significantly (P<0.05). Our study sheds light on the roles played by IL-2, IL-18, and IFN-γ in the context of liver GVHD.

From an immunological perspective, GVHD is the rejection of the donor organ by host T-cells; cytotoxic T-lymphocytes also play an important role. In addition, the failure of regulatory mechanisms may play a role in GVHD. However, the role played by CD4 is relatively clear: CD4 T-cells exert a beneficial immunosuppressive effect in GVHD patients. Thus, we focused only on CD8. Because granzyme B and perforin mediate rejection, we also measured the levels of these two key proteins. Granzyme B and perforin expression levels were elevated in GVHD patients, supporting the role of T-cell-mediated cytotoxicity in GVHD and explaining the increased cytokine levels.

As the diagnosis of GVHD depends principally on nonspecific clinical symptoms, prompt identification and treatment of GVHD are often delayed. Laboratory biomarkers are crucial for early GVHD identification. Currently, biomarker investigations are limited principally to patients receiving blood stem cell transplants. Many clinically relevant biomarkers of biological, pathogenic, or pharmacological responses have been defined.^[24] Ideal biomarker for GVHD should be specific, sensitive to the stage of GVHD, capable of being measured noninvasively, and rapid, simple, accurate, and inexpensive to assay in a standardized manner.^[25-29] The same biomarker should be useful for not only early diagnosis but also defining prognostic outcomes.

Our study extends exploratory research on GVHD biomarkers from patients undergoing blood stem cell transplantation to those receiving solid organs. We found that GVHD development in solid organs was associated with elevated levels of the cytokines IL-2, IL-18, and IFN-γ, perhaps predisposing the patient to GVHD. Sampling at a time close to GVHD onset may provide clues as to which biological events initiate GVHD development. Due to the low occurrence, there were few reports comparing cytokines in GVHD post solid organ transplantation. Our current study provides some initial insights in the matter of cytokines with different background and baselines. As shown in Fig. 1, serum IL-2 and IL-18 increased in liver recipients with GVHD compared with healthy controls. We are carrying out further HLA matching degree analysis in larger cohort to investigate the underlying mechanism. Most importantly, an early non-invasive blood test may permit timely therapeutic intervention.

There are some limitations of the study. GVHD has been found most often after allogenic hematopoietic stem cell transplantation and much less frequently after transplantation of immunologically active solid organs such as the liver, kidney, and small intestine.^[25-29] We provide first-hand data on GVHD after solid organ transplantation. The formation of appropriate control groups was very challenging. We did not include other parallel baseline controls such as renal transplantation patients lacking GVHD, liver transplantation patients without GVHD, small intestine transplantation patients with/without GVHD, or hematopoietic stem cell transplantation patients with/without GVHD. Further investigations on reasonable numbers of patients with GVHD of various organs, after careful randomization and matching, are required.

A total of 45 GVHD patients that developed after liver or renal transplantation were evaluated. We focused on baseline rather than time-course data. The findings require confirmation, and data from other patients treated in other centers are required. In the absence of a cohort differing from our current cohort in terms of baseline cytokine levels, diagnostic performance is difficult to evaluate. We thus present initial data on baseline cytokine levels. We suggest that a cytokine panel, rather than a single cytokine, should be measured to explore the etiology of GVHD.

In conclusion, IL-2 and IL-18 levels were increased significantly in the livers of patients with GVHD,^[2] paralleling their increased levels in serum.^[3] Such expression was associated with portal space congestion, epithelial cell destruction, and bile duct damage.^[4] After anti-GVHD treatment, the liver expression levels decreased significantly, suggesting that IL-2, IL-18 and IFN-γ are involved in liver GVHD, likely contributing to tissue damage.

5 Min CK, Maeda Y, Lowler K, Liu C, Clouthier S, Lofthus D, et al. Paradoxical effects of interleukin-18 on the severity of acute graft-versus-host disease mediated by CD4+ and CD8+ T-cell subsets after experimental allogeneic bone marrow transplantation. Blood 2004;104:3393-3399. PMID: 15280194

7 Grönberg H, Adolfsson J, Aly M, Nordström T, Wiklund P, Brandberg Y, et al. Prostate cancer screening in men aged 50-69 years (STHLM3): a prospective population-based diagnostic study. Lancet Oncol 2015;16:1667-1676. PMID: 26563502

10 Peng W, Liu G, Xie W, Huang H, Wu J, Shou Z, et al. Interleukin-2 receptor antagonist compared with antithymocyte globulin induction therapy in kidney transplantation from donors after cardiac death. Int J Clin Pract Suppl. 2015;(183):23-28. PMID: 26177071

11 Zhou J, Hu Z, Zhang Q, Li Z, Xiang J, Yan S, et al. Spectrum of de novo cancers and predictors in liver transplantation: analysis of the scientific registry of transplant recipients database. PLoS One 2016;11:e0155179. PMID: 27171501

14 Pranzatelli MR, Tate ED, McGee NR, Colliver JA. Pediatric reference ranges for proinflammatory and anti-inflammatory cytokines in cerebrospinal fluid and serum by multiplexed immunoassay. J Interferon Cytokine Res 2013;33:523-528. PMID: 23659672

15 de Koning L, Liptak C, Shkreta A, Bradwin G, Hu FB, Pradhan AD, et al. A multiplex immunoassay gives different results than singleplex immunoassays which may bias epidemiologic associations. Clin Biochem 2012;45:848-851. PMID: 22537455

16 Murali AR, Chandra S, Stewart Z, Blazar BR, Farooq U, Ince MN, et al. Graft versus host disease after liver transplantation in adults: a case series, review of literature, and an approach to management. Transplantation 2016 Aug 3. PMID: 27495762

23 Macgregor JN, Li Q, Chang AE, Braun TM, Hughes DP, McDonagh KT. Ex vivo culture with interleukin (IL)-12 improves CD8(+) T-cell adoptive immunotherapy for murine leukemia independent of IL-18 or IFN-gamma but requires perforin. Cancer Res 2006;66:4913-4921. PMID: 16651448

24 Gatza E, Braun T, Levine JE, Ferrara JL, Zhao S, Wang T, et al. Etanercept plus topical corticosteroids as initial therapy for grade one acute graft-versus-host disease after allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant 2014;20:1426-1434. PMID: 24892263

27 Choi SW, Braun T, Chang L, Ferrara JL, Pawarode A, Magenau JM, et al. Vorinostat plus tacrolimus and mycophenolate to prevent graft-versus-host disease after related-donor reduced-intensity conditioning allogeneic haemopoietic stem-cell transplantation: a phase 1/2 trial. Lancet Oncol 2014;15:87-95. PMID: 24295572

29 Choi SW, Stiff P, Cooke K, Ferrara JL, Braun T, Kitko C, et al. TNF-inhibition with etanercept for graft-versus-host disease prevention in high-risk HCT: lower TNFR1 levels correlate with better outcomes. Biol Blood Marrow Transplant 2012;18:1525-1532. PMID: 22469883