Abstract
Introduction: Cytokine release syndrome (CRS) is a potentially life-threatening systemic disease that has been observed after treatment with antibodies and adoptive T cell therapies. In this case, we observed nivolumab-induced CRS in a patient with gastric cancer. Case Presentation: A 43-year-old male with advanced gastric cancer was treated with nivolumab as a third-line chemotherapy. He had no history of allergies. Eight days after the first administration of nivolumab, fever, tachycardia, appetite loss and increases in liver and biliary enzymes were observed. Computed tomography revealed neither bile duct obstruction nor progression of liver metastases but showed that there was edema of the Gleason sheath. Histopathological analysis of the liver revealed cholestatic liver injury with CD8+ T lymphocyte and macrophage infiltration. Neither viral infection nor autoimmune disease was revealed. His symptoms were similar to those of CRS observed after T cell therapy. We diagnosed his disease as nivolumab-induced liver injury and cholangitis accompanied by CRS based on his serum cytokine levels. Discussion/Conclusion: To the best of our knowledge, this is the first report of nivolumab-induced CRS in a patient with gastric cancer.
Introduction
Cytokine release syndrome (CRS) is a potentially life-threatening systemic disease that has been observed after treatment with antibodies such as CD28 antibodies and adoptive T cell therapies such as CD19-specific chimeric antigen receptor (CAR) T cell therapy. CRS causes various symptoms, such as fever, hypotension, hypoxemia, tachycardia, liver injury and neurological disorders. In the case of CD19-CAR-T-related CRS, treatment with the anti-IL-6 receptor antibody may be an option, but there are no currently available methods of resolving immune checkpoint inhibitor-related CRS [1]. Nivolumab is a well-known immune checkpoint inhibitor that inhibits the binding between PD-1 and its ligands PD-L1 and PD-L2 [2]. There have been some reports of checkpoint inhibitor-related CRS, but not in patients with gastric cancer [3, 4].
Case Report/Case Presentation
A 43-year-old male who had no history or family history of allergies was diagnosed with human epidermal growth factor receptor 2 (HER2)-negative gastric adenocarcinoma. He had been treated with 5 courses of SOX (S-1 40, 50, or 60 mg according to body-surface area, orally administered twice a day for 14 days with oxaliplatin 100 mg/m2 on day 1, every 3 weeks) as a first-line chemotherapy and 2 course of paclitaxel + ramucirumab (paclitaxel 80 mg/m2 on days 1, 8, 15 with 8 mg/kg ramucirumab on days 1 and 15, every 4 weeks) as a second-line chemotherapy. However, both treatments failed. He was referred to our department for third-line chemotherapy. He was 178 cm tall and weighed 78.3 kg. He had an Eastern Cooperative Oncology Group (ECOG) performance status of 2, body temperature of 36.5°C, blood pressure of 138/92 mm Hg, pulse rate of 67 beats per minute and SpO2 of 98% (atmospheric air). Physical examination revealed abdominal pain, back pain and leg oedema. He had taken vonoprazan fumarate, loxoprofensodium, rebamipide, adenine, naldemedine, magnesium oxide, fentanyl patches and a continuous infusion of oxycodone for pain. The baseline laboratory findings are shown in Table 1. A computed tomography (CT) scan showed left adrenal metastasis, peritoneal dissemination and suspected liver metastases. Eight days after the first administration of nivolumab, he developed a high fever (39.0°C), tachycardia, appetite loss, malaise, and elevated levels of bilirubin, liver enzyme, biliary enzyme and C-reactive protein (CRP) (Table 1, Fig. 1). A CT scan revealed oedema of the Gleason sheath (Fig. 2a). Neither bile duct obstruction nor liver metastasis progression was revealed (Fig. 2b). Histopathological analysis of the liver revealed cholestatic liver injury (Fig. 3a). Immunohistochemical examination revealed CD8+ T lymphocyte and macrophage infiltration into the intrahepatic bile duct (Fig. 3b-g). There was no evidence of Epstein-Barr virus infection, cytomegalovirus infection or autoimmune disease (Table 1). Although his blood culture was negative, sulbactam/cefoperazone was started empirically given the possibility of a biliary tract infection. As the levels of bilirubin, biliary enzyme and CRP increased on day 9 (Fig. 2), we started prednisolone (PSL) 80 mg (1 mg/kg/day) as a treatment for nivolumab-induced liver injury and cholangitis. His symptoms, including high fever, tachycardia, appetite loss and malaise, without any obvious infection, were similar to the symptoms of CRS that we have observed after TCR-Gene Transduced T Cell Transfer therapy [5]. The examination of his serum revealed marked elevation of the level of interferon (IFN)-γ elevation in the early phase and bimodal elevation of the level of TNF-α (Fig. 1b). His bilirubin level decreased temporarily but subsequently increased (Fig. 1a). On day 22, 1 g of methyl prednisolone was started for three days. However, his levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were increased. After adding mycophenolate mofetil 2 g daily to the PSL on day 27, T-bil, his levels of ALT and ALT decreased, but his levels of alkaline phosphatase (ALP) and gamma-glutamyl transferase (γ-GTP) continued to increase. Although his bilirubin level was improved after the addition of mycophenolate mofetil, his levels of AST and ALT increased again due to the progression of the liver metastases. He died of gastric cancer on day 105.
Discussion/Conclusion
Although immune checkpoint blockade is typically described as being well tolerated, it sometimes causes serious immune-related adverse events (irAEs) [6]. In this case, the patient had cholestatic liver injury. Because the examination of the patient’s serum revealed marked elevation of the level of IFN-γ in the early phase and bimodal elevation of the level of TNF-α (Fig. 1b), we diagnosed his disease as nivolumab-induced liver injury and cholangitis accompanied by CRS.
Non-tumour tissue injury by PD-1 inhibitor-activated T cells is a cause of autoimmune-like disorders. Five cases of nivolumab-induced cholangitis have been reported [7-9]. The nivolumab-induced cholangitis reported by Kawakami and Kashima resulted in bile duct dilation and moderate elevation of the level of bilirubin (peak T-bil 0.7–3.8 mg/dL, bil was not reported by Kashima) [7, 8]. Histopathological analysis of the bile duct revealed infiltration of neutrophils into the mucosa [8]. In contrast, the nivolumab-induced cholangitis reported by Glesomino did not result in bile duct dilation but did cause marked elevation of the level of bilirubin (peak T-bil 10.5 mg/dL). Histopathological analysis of the liver revealed the presence of T cells, with a predominance of CD8-positive cells, as part of bile duct aggression and intraductal microabscesses, and mild sinusoidal expansion without fibrosis. Glesomino suggested that there were 2 types of immune-mediated cholangitis, namely, large-duct cholangitis and small-duct cholangitis [9]. In our case, we observed no bile duct dilation and marked elevation of the level of bilirubin (peak T-bil 13.4 mg/dL, peak D-bil 9.9 mg/dL), which was similar to small-duct cholangitis. Hypercytokinemia may contribute to small-duct cholangitis. IFN-γ, TNF-α, IL-2 and IL-6 were assessed by enzyme-linked immunosorbent assay (ELISA) using ELISA kits (Thermo Fisher Scientific Inc., Japan) in this case. It seems that the levels of IFN-γ, IL-6 and TNF-α were related to the onset of fever and liver injury in the early phase. After the administration of PSL, the levels of IFN-γ and IL-6 decreased immediately, while TNF-α increased. Although mycophenolate mofetil was added to the PSL, it was not able to completely suppress TNF-α (Fig. 1b). These findings suggest that TNF-α might not be secreted by only CD8+ T cells. We know that not only activated T cells secrete IFN-γ and TNF-α. In our analysis, after PSL administration, the levels of IFN-γ and TNF-α decreased immediately, and only the level of TNF-α increased soon thereafter. TNF-α was temporarily suppressed by steroid pulses on days 22–24 (Fig. 1b). These results suggest that nivolumab-activated T cells are suppressed by PSL and that other cells are less sensitive to steroid-secreted TNF-α after PSL administration. Macrophages are well-known major TNF-α-producing cells and are associated with PD-1/PD-L1 signalling. Macrophages might secrete TNF-α. Further investigation is needed. The use of an anti-TNF-α antibody at the time of hepatic injury after the administration of an immune checkpoint inhibitor is not recommended to avoid idiosyncratic drug-induced liver injury [10, 11]. However, in the case of liver injury and cholangitis related to an increase in the level of TNF-α, the use of anti-TNF-α antibody might be an option.
HLA-B8, -DR3, and -DRw 52a are thought to be related to the onset of primary sclerosing cholangitis [12], and HLA-DR8 is thought to be related to the onset of primary biliary cirrhosis [13-15]. In this case, the patient’s HLAs were HLA-B18, -B51, -DR3, and -DR8. Although Gelsomino stated that HLA-DR11 is characteristic of nivolumab-induced cholangitis, our observations did not support that conclusion [9]. More cases are needed to clarify whether HLA is associated with nivolumab-induced cholangitis.
In our case, the patient received palliative radiotherapy for the cancer pain caused by the left adrenal metastasis (30 Gy in 10 fractions from day 12 to 25). Radiation therapy may increase damage to the radiation field during or after the administration of immune checkpoint inhibitors. However, the first increase in T-bil occurred on day 8. The mean radiation dose to the liver was 27%. This result suggested that there was little association between liver damage and radiotherapy in this case.
To the best of our knowledge, this is the first report of nivolumab-induced CRS in a patient with gastric cancer. In the case of nivolumab-induced liver injury with TNF-α, the use of an anti-TNF-α antibody might be an option. Further investigation is warranted.
Statement of Ethics
This article does not contain any studies with human participants performed by any of the authors. Written informed consent was obtained from the patient using the hospital default informed consent form.
Disclosure Statement
The authors have no conflicts of interest to declare.
Funding Sources
We do not have any funding source to disclose concerning to this report.
Author Contributions
H. Oda wrote the initial draft of the manuscript. M. Ishihara contributed to interpretation of data and assisted in the preparation of the manuscript. Y. Miyahara and J. Nakamura contributed to analysis and interpretation of data. M. Iwasa contributed to diagnosis of liver injury. Y. Kozuka contributed to pathological diagnosis. A. Tsunoda, Y. Yamashita, K. Saito and T. Mizuno contributed to clinical treatment. N. Katayama supervised the patient treatment. All other authors critically reviewed the manuscript and approved the final version of the manuscript.