Immune-Mediated Adverse Effects of Checkpoint Inhibitors: A Clinical Experience

Systemic anti-tumour immunotherapy with antibodies to CTLA-4, PD-1 and PD-1L was one of the most significant breakthroughs in solid tumour pharmacotherapy in decades. Unfortunately, their superior efficacy and broader use are limited by rare but often life-threatening adverse effects that differ from the previously known spectrum of adverse effects of anti-cancer chemotherapy. The mechanisms of these immune-related adverse effects (irAEs) and their epidemiology, prognostic significance regarding the further course of the disease, and sometimes their management have been insufficiently investigated so far. The incidence of any irAE can be up to 60% for checkpoint inhibitors and 10–30% for more severe grades (G3-4) (Khan & Gerber, 2020). These adverse effects do not appear to be dose-dependent (Wang et al., 2017). The kinetics of their onset vary widely between organs and organ systems. It is not yet known why some irAEs appear predominantly with some latency and others immediately after initiation of therapy. The ‘natural’ pharmacodynamics of checkpoint inhibitors play a role in the mechanism of irAE development. The costimulatory signal on antigen-presenting cells is disinhibited upon binding to the target receptor, and the cellular and humoral immune response is triggered. Unfortunately, the expansion of some T-lymphocytes subsets is also probably involved in the irAE development.

We aimed to review the irAEs in terms of frequency, severity, and type according to the tissue/organ affected in patients treated with ICIs at MMCI. The secondary aim was to present a clinical case of severe irAE and its management.

Using the hospital information system, we performed a retrospective analysis of the irAEs of ICIs used in the Masaryk Memorial Cancer Institute, a tertiary comprehensive oncology centre, from 2011 to 2020.

The ICIs were administered to 648 patients, with the most frequent indications being malignant melanoma, renal carcinoma, and lung carcinoma (235, 144 and 114 cases, respectively). The most common ICI was nivolumab (316 patients), followed by pembrolizumab and ipilimumab (93 and 68 patients). An overview of the ICIs and indications is presented in Table 1.

Severe immune-related adverse effects requiring immunosuppressive treatment were pronounced in 83 patients (12.8%), which is not different from the literature data (Khan & Gerber, 2020). The exact numbers of irAEs, in particular ICIs, are presented in Table 2.

The most frequent irAEs across all treatments were pneumonitis, followed by hepatitis, colitis, and dermatitis (21, 15, 13, 12 cases, respectively). As apparent in Table 2, if low numbers are omitted, the nivolumab + ipilimumab combination probably possesses the highest risk of irAE.

A 52-year-old woman (88 kg bodyweight, 175 cm height) was diagnosed with renal clear cell carcinoma. She underwent randomisation in the clinical study with nivolumab/placebo vs nivolumab/ipilimumab every three weeks. Two weeks after the first cycle, the patient reported mild adverse effects (headache, flushes, fluctuating blood pressure), and laboratory findings revealed G2 hyperthyreosis, which was managed with metoprolol. The study treatment continued. Hyperthyreosis turned into hypothyreosis, and thereafter the patient’s treatment was substituted with levothyroxine. After four cycles, colitis grade 2 manifested, and thus, prednisone 1 mg/kg was administered. The patient was unblinded from the study protocol, and it was revealed that she was treated with nivolumab (3 mg/kg) + ipilimumab (1 mg/kg). Because the patient’s status improved, the prednisone was tapered sequentially. The treatment continued with only nivolumab 480 mg every four weeks. After six cycles of nivolumab, colitis grade 2 manifested again and was transiently managed with methylprednisolone 62.5 mg, followed by prednisone 20 mg/day and mesalazine 3 g/day. The ICI treatment stopped promptly. Subsequently, colitis grade 4 relapsed. Due to insufficient response to glucocorticoids, the following treatment was sequentially administered over four weeks, with insufficient effect: infliximab 5 mg/kg, octreotide 100 mcg every 8 hours, vedolizumab 300 mg and mycophenolate mofetil (1000 mg q/d). Most of the time during the colitis treatment, the situation was complicated by urinary tract infection resolved with sulfamethoxazole and amikacin. Partial response (alleviation from colitis symptoms) was achieved only by high dose glucocorticoid pulses. Hemicolectomy and ileostomy were performed after two months due to pneumoperitoneum, and vertebroplasty was performed after another two months due to the fracture of L1-L3, with a satisfactory result. Eighteen months after the episode of G4 colitis, symptomatic treatment with mesalazine and hydrocortisone was provided, and the patient was held in therapy due to residual symptoms of colitis.

The occurrence of irAEs in the patients treated at Masaryk Memorial Cancer Institute was comparable with the literature data (Khan & Gerber, 2020). Although ICI treatment offers superior efficacy over classical chemotherapy in some solid tumours, unpredictable irAEs may occur. Early diagnosis and appropriate management of irAEs are essential in the perspective of subsequent treatment. Even if early and proper treatment is provided, the irAEs may progress and disable further oncological treatment.