Whole-liver intensity-modulated radiation therapy as a rescue therapy for acute graft-versus-host disease after liver transplantation

To the Editor: Graft-versus-host disease (GVHD) is a rare but fatal complication after solid organ transplantation (SOT), with an incidence of 0.2–2.0%. Recent retrospective studies have shown a mortality rate of approximately 85–98% among patients with SOT-GVHD.[1] Various drugs and methods have been shown to improve or alleviate clinical symptoms, but no effective treatments for GVHD currently exist. Recent clinical experience with SOT-GVHD treatment is mixed and limited, including systemic or partial corticosteroid support, the use of Janus kinase (JAK) inhibitors, allogeneic hematopoietic cell transplantation (alloHCT), and donor-targeted serotherapy. Since the number of lymphocytes transplanted with the organ may be a risk factor for GVHD, reducing the number of donor lymphocytes seems beneficial. Intensity-modulated radiation therapy (IMRT) is a type of radiotherapy that uses optimized, non-uniform radiation beam intensities on the patient.[2] Local allograft irradiation has also been used to treat severe resistant rejection after liver transplantation (LT) in previous studies.[3] Based on this, we hypothesized that whole-liver IMRT at a low dosage could be used as a rescue therapy to control acute GVHD after LT by reducing the number of donor-derived immune cells. The donors in both cases were deceased, who donated after brain death (DBD). The study followed the guidelines of the Declaration of Helsinki. Written informed consent was obtained from both participants. The study was approved by the local institutional review boards on human research at the Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology (IRB No. TJ-IRB20221148). One of the patients was a 61-year-old man with diabetes with hepatocellular carcinoma (HCC) secondary to hepatitis B virus (HBV) cirrhosis (Patient No. 1) and the other was a 58-year-old man with diabetes with liver failure caused by chronic alcoholic cirrhosis (Patient No. 2). Postoperative recovery was uneventful, and the patients were discharged on postoperative days (PODs) 13 and 14, with normal liver function. In this study, short tandem repeat (STR)-based analysis was used to evaluate donor chimerism (YUANQI BIO®, Shanghai, China, #CT120017) from blood and/or bone marrow. For patient No. 1, two STR analyses were performed on PODs 36 and 46. For patient No. 2, seven STR analyses were performed between PODs 21 and 44. Patient No. 1 developed an oral ulcer on POD 16, followed by a maculopapular skin rash. STR-based analysis first revealed 4.82% donor chimerism in total blood on POD 36 [Supplementary Table 1, https://links.lww.com/CM9/C211], with normal liver function and white cell count during this period [Figure 1A]. Patient No. 2 developed an extensive rash covering his entire skin on POD 18, followed by oral ulcers and diarrhea. Subsequent STR-based analysis showed donor chimerism rates of 6.79% in total blood on POD 21, 19.33% in the blood cluster of differentiation (CD)3+ T cell population on POD 23, and 9.00% in the bone marrow on POD 23 [Supplementary Table 1, https://links.lww.com/CM9/C211]. The patient No. 1 had normal liver function but somewhat a low white cell count [Figure 1A]. Additionally, the diagnosis of GVHD was confirmed by skin biopsy. Based on clinical symptoms and evidence of donor chimerism, both patients were diagnosed with GVHD.Figure 1: The clinical responses and liver graft biopsies of the patients. (A) The clinical measures of response to IMRT treatment. White blood cell count; ALT; AST; and TBIL in patient No. 1 (left) and patient No. 2 (right). (B–G) Liver graft biopsies for patient No. 1. Light microscopy (Masson’s trichrome stain, Original magification is ×400) of graft biopsies performed on PODs 57 (B), 65 (C), 72 (D), 85 (E), 144 (F), and 228 (G). Only mild liver injury, including slight erythrostasis in the hepatic sinusoid (POD 57), liver cell edema (POD 144), liver cell necrosis (POD 228), and mild acute rejection, occurred during this stage (PODs 65, 72, and 85). ALT: Glutamic-pyruvic transaminase; AST: Glutamic oxaloacetic transaminase; IMRT: Intensity-modulated radiation therapy; POD: Postoperative day; TBIL: Total bilirubin.Computed tomography (CT) scans were taken during inspiration and expiration to track liver motion. The liver was contoured. Since whole-liver IMRT can cause liver injury, a very low radiological dose of 50 cGy was chosen for each treatment based on the advice of the consulting oncologist and oncologist. Patient No. 1 underwent IMRT twice on PODs 43 and 44. Patient No. 2 underwent IMRT nine times between PODs 23–24 and PODs 29–35. The donor chimerism rate in blood CD3+ T cells of patient No.1 was reduced from 4.82% (POD 36) to 0% (POD 46) [Supplementary Table 1, https://links.lww.com/CM9/C211]. Thus, IMRT was concluded. For Patient No. 2, initial IMRT was given on PODs 23 and 24. However, the donor chimerism rate in total blood increased from 19.33% (POD 23) to 33.10% (POD 28) [Supplementary Table 1, https://links.lww.com/CM9/C211]. Continuous IMRT was then performed from POD 29 to POD 35. We measured donor chimerism rates in total blood, blood CD3+ T cells, and bone marrow. Results showed a continuous reduction in donor chimerism. By POD 44, patient No. 2’s donor chimerism rate was 1.61% in total blood and 3.61% in blood CD3+ T cells [Supplementary Table 1, https://links.lww.com/CM9/C211]. Additionally, the clinical symptoms of both patients significantly improved during treatment. For patient No. 1, oral ulcers, skin rash, and diarrhea disappeared, and the patient recovered, although the white cell count was slightly high during this period. Patient No. 2 had more severe symptoms than patient No. 1. After nine rounds of IMRT, patient No. 2’s white cell count returned to normal, and the extensive skin rash and diarrhea resolved. In addition to IMRT, the conventional treatments for the patients are shown in Supplementary Table 2, https://links.lww.com/CM9/C211. Although IMRT significantly suppressed GVHD progression, both patients developed notable liver dysfunction after IMRT treatment. Patient No. 1 showed elevated liver function indices (glutamic-pyruvic transaminase [ALT]: 65 U/L, glutamic oxaloacetic transaminase [AST]: 40 U/L, and total bilirubin [TBIL]: 4.9 μmol/L) 9 days after stopping IMRT on POD 53. Liver biopsy revealed mild acute rejection (PODs 65, 72, and 85; Figure 1) with negative panel reactive antibodies (PRA). Despite treatment with methylprednisolone (500 mg/day for 3 days), artificial liver therapy, and anti-human T lymphocyte globulin (ATG), the liver injury persisted for a prolonged period. Even with only two rounds of IMRT, liver injury was evident, including slight erythrostasis in the hepatic sinusoid (POD 57, Figure 1B), liver cell edema (POD 144, Figure 1F), and liver cell necrosis (POD 228, Figure 1G). Fortunately, liver function gradually returned to normal level more than 2 years after IMRT (ALT: 43 U/L, AST: 43 U/L, and TBIL: 99.8 μmol/L on POD 995, Figure 1G). Patient No. 2 developed acute and severe liver dysfunction with elevated ALT, AST, TBIL, thrombopenia, and other coagulation disorders about 1 week after nine rounds of IMRT (ALT: 82 U/L, AST: 198 U/L, and TBIL: 38 μmol/L on POD 41, Figure 1A). Despite immediate platelet transfusion (2 U/day) and oral avatrombopag (40 mg/day) to improve coagulation function, along with intravenous polyene phosphatidylcholine to improve liver function, the patient’s condition deteriorated. Patient No. 2’s liver function worsened over the next 3 days, and he died on POD 45 due to liver failure. A liver biopsy was not performed on patient No. 2 because of poor coagulation function. Patient No. 1 survived and has remained healthy with regular weekly follow-ups to date (the date of report completion). To date, there is few completely effective approaches for curing LT-GVHD. Our patients were treated with systemic corticosteroids and ruxolitinib once GVHD was diagnosed. Although clinical symptoms such as oral ulcers, skin rash, and diarrhea improved, the rate of donor chimerism did not decrease after treatment with corticosteroids and ruxolitinib. Therefore, we were unsure whether corticosteroids or JAK inhibitors could fully suppress GVH immune responses. Facing a therapeutic dead end, we used IMRT, which has been employed to suppress tumor cells in the liver, to target GVH-associated immune cells in liver grafts. Local allograft irradiation has been used to suppress acute rejection in previous studies.[3] In our study, IMRT was an effective strategy for both patients, as evidenced by significant decreases in donor chimerism rates to 0 in patient No. 1 and to 1.61% in patient No. 2, along with recovery of the recipient immune response indicated by increased white cell counts. IMRT is essentially a donor-targeted therapy. In 2020, Zuber et al[4] reported a novel strategy for treating GVHD after kidney transplantation using the adoptive transfer of anti-human leukocyte antigen (HLA) donor-specific antibodies (DSAs) through the transfusion of highly selected plasma. This is also a donor-targeted therapy that effectively eliminates the donor-specific immune response. Therefore, IMRT has the potential to successfully treat refractory LT-GVHD patients. On the contrary, we must consider the adverse effects on the liver that can occur with IMRT therapy. A reasonable explanation is that IMRT can suppress not only GVH-associated immune cells but also normal liver cells such as parenchymal hepatic and bile duct cells. Additionally, this liver injury from IMRT could persist for years (as seen in patient No. 1) and even be fatal (as seen in patient No. 2). Therefore, choosing an appropriate IMRT dose for treating GVHD by simultaneously preserving liver function is crucial. Based on our observations, future studies are needed to determine the precise and individualized dosage and frequency of IMRT for patients with LT-GVHD. Funding This study was supported by a grant from the Key Research and Development Plan of Hubei Province (No. 2022BCA015). Conflicts of interest None.