Nivolumab Induced Encephalopathy in a Man With Metastatic Renal Cell Cancer

A Case Report

Jindřich Kopecký; Ondřej Kubeček; Tomáš Geryk; Birgita Slováčková; Petr Hoffmann; Miroslav Žiaran; Peter Priester


J Med Case Reports. 2018;12(262) 

In This Article


Immunotherapy is a treatment modality that has experienced a renaissance in the treatment of solid tumors during the last decade. Despite numerous attempts to utilize the immune system to fight cancer, including the depletion of T-regulatory lymphocytes,[2] and the use of cancer vaccines or cytokines,[3] previous results have been unsatisfactory. However, the discovery and understanding of the function and regulatory role of several receptors and their ligands, that is, cytotoxic T-lymphocyte antigen-4 (CTLA-4), PD-1, and programmed death-ligand 1 (PD-L1) in T cell activation have encouraged the further development and clinical use of immunotherapy.[4,5] Currently, we can affect these regulatory mechanisms with a new class of drugs called immune checkpoint inhibitors. There is a growing body of evidence that these drugs represent effective treatments with the possibility of achieving a durable response in a wide spectrum of solid tumors, including melanoma,[6,7] non-small cell lung cancer,[8,9] urothelial carcinoma,[10,11] and renal cell carcinoma (RCC).[1]

The pathogenesis of irAEs is closely related to immune regulation affecting the activity of cytotoxic T cells. The mechanism of action of checkpoint inhibitors is to disrupt an interaction between regulatory receptors and their ligands, which normally produce an inhibitory signal with subsequent attenuation of immune activity. If the binding of ligands to regulatory receptors is prevented, for example, by anti-PD-L1 or anti-PD-1 antibodies, the activated T cells can contribute to an antitumor response.[12–14] On the other hand, this leads to an imbalance and possible disruption of immunological tolerance, which might cause an uncontrolled immune response directed against self-tissue antigens. This may result in undesired inflammatory activity involving various organ systems, referred to as irAEs. The most commonly affected sites are the gastrointestinal tract, skin, and endocrine system, but any organ or system of the human body can be affected.

In our case the initial presentation of adverse symptoms related to nivolumab was poor but unusual with severe chorea-like dyskinesia symptoms. There is only sparse information in the literature about similar cases; that is why the diagnostic and therapeutic algorithm is more difficult than, for example, for immune-related pneumonitis. Our case shows and emphasizes the urgent need of up front resolute immunosuppressive therapy in cases of immune-related encephalitis. Based on our case we propose the paraneoplastic antigen Ma2 as a helpful biomarker in order to make a correct diagnosis and to distinguish immune-related encephalitis from other possible causes.

According to the literature, irAEs affecting the central nervous system (CNS) induced by anti-PD-1 antibody treatment are rare.[15] The vast majority of observed irAEs affecting the CNS have been related to anti-CTLA-4 antibodies.[16,17] Most cases of immune checkpoint inhibitor-related CNS toxicities were of lower grade, with only approximately 0.1–3% of all events being grade 3 or 4.[18] There are a limited number of anecdotal reports regarding anti-PD-1 antibody-induced autoimmune CNS toxicities.[19–21]

Immune-mediated encephalitis can be difficult to distinguish from encephalitis related to other causes. It can produce a wide range of symptoms, including headache, fever, weakness, fatigue, impaired memory, hallucinations, and convulsions. The diagnosis is usually made per exclusionem (a diagnosis of exclusion). Immune-mediated encephalitis induced by checkpoint inhibitors usually arises at the very beginning of therapy.[15,19] However, as we demonstrate here, it can occur at any time during therapy.

The ratio of CD4+/CD8+ T cells in aseptic meningitis is typically approximately 3:1;[22] however, it was unusually low in our case (approximately 1:1). This result is probably because the PD-1 receptor is strongly expressed on CD8+ T cells,[23] resulting in the proliferation of the CD8+ subpopulation. Another explanation is that the CD4+/CD8+ ratio naturally decreases during inflammation, so the result might be affected by the time of examination.[24]

The exact mechanism of immune-mediated encephalitis remains unclear. It is believed that the majority of irAEs are caused predominantly by T cells, although some proportion of irAEs may be mediated by other immune cells. N-methyl-D-aspartate (NMDA) receptor antibodies seem to play an important role in the pathogenesis of immune-mediated encephalitis.[25] These antibodies have been found in patients who developed immune-mediated encephalitis related to immune checkpoint inhibitor therapy.[19] NMDA receptors are expressed on the surface of melanocytes and are coded by the gene GRIN2A, which tends to be highly mutated in patients with melanoma.[26] The formation of antibodies against aberrant NMDA receptors in melanoma is supposed to induce encephalitis, as these antibodies may cross-react with NMDA receptors in the brain.[15] This mechanism has not been observed in mRCC yet. Antibodies against the NMDA receptor were not present in our patient, suggesting that other mechanisms are probably involved in the pathogenesis of checkpoint inhibitor-induced immune encephalitis.

Anti-Ma2-associated paraneoplastic neurological syndromes, such as brain encephalitis, usually present as isolated or combined limbic, diencephalic, or brainstem dysfunction.[27,28] The presence of anti-Ma2 antibodies has been associated with neoplasms in approximately 96% of described cases,[28] mainly cases of testicular cancer and small cell lung carcinoma. There is a growing body of evidence that anti-Ma2-associated encephalitis differs from typical paraneoplastic limbic or brainstem encephalitis and may therefore be unrecognized. The association of anti-Ma2 antibodies and mRCC has not been proven yet. However, the association between anti-Ma2 antibodies and immune encephalitis might be similar to that between anti-NMDA antibodies and melanoma. We cannot exclude the presence of anti-Ma2 antibodies prior to the start of checkpoint inhibitor therapy, but this finding undoubtedly contributed to the diagnosis of immune-mediated encephalitis in our patient.

The mainstay of the management of immune-mediated encephalitis is therapy with intravenously administered corticosteroids, similarly to that of other irAEs. However, therapy can be successful only if absolute compliance of the patient and his family can be guaranteed. The dose of corticosteroids depends on toxicity grade, ranging from 0.5 to 2 mg/kg per day.[29] The response to corticosteroid therapy in immune encephalitis seems to be slower and less pronounced than that in other irAEs. Therefore, higher initial doses of corticosteroids may be required. Alternatively, other immunosuppressive drugs such as infliximab can be used.

It must be emphasized that patients treated with intravenously administered corticosteroids are especially prone to infectious complications, and adequate care should be paid to this issue. The presented patient was administered trimethoprim/sulfamethoxazole as prevention against Pneumocystis pneumonia. Unfortunately, early signs of infection could not be recognized as he refused to remain in our hospital. It should therefore be recommended that patients with severe irAEs are treated in an in-patient setting in order to recognize early signs of infectious complications and start antibiotic treatment as soon as possible.

Unfortunately, we were not able to convince either our patient or his family to comply, so there was no way to extend our therapy (for example, with intravenously administered immunoglobulins or plasmapheresis) to prevent his death.