Fatal C-reactive Protein-less Sepsis with Anti-IL-6 Autoantibody Production after Administration of Durvalumab

Abstract

Durvalumab plus platinum-based chemotherapy is the first-line treatment for extensive-stage small-cell lung cancer. Immune checkpoint inhibitors (durvalumab) can cause immune-related adverse events (irAEs). We herein report the first case of fatal sepsis with anti-interleukin-6 autoantibody production following durvalumab administration. A 62-year-old woman with extensive-stage small-cell lung cancer received carboplatin-etoposide plus durvalumab chemotherapy. Serum C-reactive protein (CRP) levels decreased below the detection sensitivity post-treatment. She developed severe sepsis during maintenance durvalumab therapy; however, her serum CRP level did not increase. The serum tested positive for anti-interleukin-6 autoantibodies, which can cause CRP-less infections. Anti-interleukin-6 autoantibody production and subsequent sepsis without serum CRP elevation are possible irAEs.

Introduction

The CASPIAN trial, a phase 3 randomized controlled study, investigated the efficacy of a combination of durvalumab, an anti-programmed death-ligand 1 antibody, with platinum-based chemotherapy (cisplatin or carboplatin plus etoposide) versus standard platinum-based chemotherapy alone in patients with extensive-stage small-cell lung cancer (ES-SCLC) (1). The trial demonstrated that the addition of durvalumab considerably improved the overall survival compared to standard chemotherapy alone. This combination therapy represents a new standard of care for first-line treatment of ES-SCLC (2). Despite their efficacy, however, immune checkpoint inhibitors, including durvalumab, can induce a unique spectrum of side effects, termed immune-related adverse events (irAEs), due to their mechanism of immune response amplification (3).

We herein report the first case of anti-interleukin-6 (IL-6) autoantibody production and subsequent fatal sepsis after the administration of an immune checkpoint inhibitor.

Case Report

A 62-year-old woman with no history of respiratory, autoimmune, or severe infectious disease was diagnosed with small-cell lung cancer with brain metastasis. The patient was treated with carboplatin-etoposide plus durvalumab. On day 3 of the first cycle of chemotherapy, the patient developed mild pneumonia. Blood tests revealed elevated C-reactive protein (CRP) (2.7 mg/dL) levels. The pneumonia quickly resolved with antimicrobial therapy, and the serum CRP levels decreased to the normal range. After two cycles of chemotherapy, the serum CRP levels decreased to below the measurable range (Fig. 1A). After four cycles of chemotherapy, computed tomography and brain magnetic resonance imaging confirmed a partial response (Fig. 1B-E), and the chemotherapy regimen was switched to maintenance durvalumab.

Figure 1.

(A) Clinical course from the initial visit to the onset of severe sepsis. (B-E) Chest computed tomography (B and C) and brain magnetic resonance imaging (D and E) showing the baseline images before the initiation of chemotherapy (B and D), and a partial response, evidenced by shrinkage of all target lesions (arrows) after four cycles of chemotherapy (C and E).

Seven days after the first maintenance dose of durvalumab, the patient developed a fever and diarrhea. Two days after symptom onset, the patient presented to the emergency room with shock vitals. Computed tomography revealed extensive small and large intestinal wall thickening (Fig. 2A) without lung cancer progression. Blood tests showed elevated leukocyte counts; however, the CRP level was below detection sensitivity (Table). The patient was diagnosed with septic shock due to enteritis and was administered intravenous antibiotics and noradrenaline. As the blood pressure did not increase even with the administration of vasopressor agents, systemic corticosteroids were administered, resulting in an increase in blood pressure. Blood culture revealed Eggerthella lenta, a rare bacterium that causes abdominal infections and severe bacteremia. The isolated E. lenta was susceptible to meropenem, which had already been administered since admission. Owing to the high blood β-D-glucan level, micafungin was administered for possible fungal infection. Despite treatment, multiple organ failure due to sepsis was exacerbated, and the patient died 20 days later.

Figure 2.

(A) Abdominal computed tomography showing diffuse wall thickening from the jejunum to the colon. (B) Detection of anti-IL-6 antibodies levels in the serum obtained from the patient in this report (I), healthy control (II and III), and the patient control with small-cell lung cancer treated with platinum–etoposide plus durvalumab (IV) using a commercially available enzyme-linked immunosorbent assay kit (MBS946170; MyBioSource, San Diego, USA). The sample was considered positive for anti-IL-6 autoantibodies if its OD value was ≥2 × the OD value of a control sample. (C) Clinical course after the onset of severe sepsis. After intravenous hydrocortisone administration, serum IL-6 autoantibody levels gradually decreased and serum CRP levels increased accordingly. CRP: C-reactive protein, IL: interleukin, OD: optical density

Table.

Laboratory Tests at Admission.

Complete blood cells				Alkaline phosphatase	211	U/L
White blood cells	25,100 ↑	/μL		C-reactive protein	<0.1 ↓	mg/dL
Neutrophils	84.9 ↑	%		Procalcitonin	18.1 ↑	ng/mL
Lymphocytes	7.4	%		Serum immunology
Monocyte	5.5	%		Rheumatoid factor	negative
Eosinophils	1.8	%		Anti-nuclear antibody	negative
Red blood cells	3.25×106	/μL		Anti-cyclic citrullinated peptide	negative
Hemoglobin	12.0	g/dL		PR3-ANCA	negative
Hematocrit	34.9	%		MPO-ANCA	negative
Platelets	31.7×104	/μL		Anti-parasitic antibodies	negative
Blood biochemistry				T-SPOT	negative
Total protein	5.7	g/dL		β-D-glucan	46.1 ↑	pg/mL
Albumin	3.5	g/dL		IgG	1,040	mg/dL
Blood urea nitrogen	22.0	mg/dL		IgG1	587	mg/dL
Creatine	0.69	mg/dL		IgG2	360	mg/dL
Total bilirubin	0.49	mg/dL		IgG3	22.6	mg/dL
Aspartate transaminase	237 ↑	U/L		IgG4	70.1	mg/dL
Alanine aminotransferase	113 ↑	U/L		IgA	177	mg/dL
Lactate dehydrogenase	752 ↑	U/L		IgM	48 ↓	ng/mL

PR3-ANCA: proteinase 3-anti-neutrophil cytoplasmic antibodies, MPO-ANCA: myeloperoxidase-anti-neutrophil cytoplasmic antibodies, Ig: immunoglobulin

Anti-IL-6 autoantibody production is a cause of CRP-free infection (4-6). Anti-IL-6 autoantibodies were detected in the serum of the patient (Fig. 2B). After hydrocortisone administration, the levels of autoantibodies decreased, whereas CRP levels increased (Fig. 2C).

Discussion

We encountered a case of CRP-less sepsis with anti-IL-6 autoantibody production after immune checkpoint inhibitor administration. IL-6 is produced during the inflammatory response to infection (7). Secreted IL-6 enters the bloodstream and promotes CRP production. Tocilizumab, an anti-IL-6 receptor antibody, has been approved for the treatment of systemic inflammatory diseases such as Castleman's disease and rheumatoid arthritis (8). Tocilizumab reduces serum CRP levels to almost zero by inhibiting IL-6 signaling (9). IL-6 is a central cytokine in CRP production, and inhibition of IL-6 signaling reduces serum CRP levels, even during systemic inflammation.

Serum anti-IL-6 autoantibody production in infectious diseases without serum CRP elevation was first reported in 2008 (4). To date, only four cases of anti-IL-6 autoantibody production during infectious diseases have been reported, all of which were idiopathic (4-6). In our case, serum CRP gradually decreased after durvalumab administration and did not increase during sepsis, suggesting that anti-IL-6 antibody production might have occurred as a side effect of durvalumab administration. The finding that serum CRP was elevated in pneumonia patients during one cycle of chemotherapy also supports this possibility.

Immune checkpoint inhibitors, such as durvalumab, have shown substantial efficacy in treating various malignancies, including small-cell lung cancer. However, their mechanism of enhancing immune activity can also lead to irAEs, which are distinct from those observed with traditional chemotherapy (3). Immune checkpoint inhibitors can elicit irAEs in the form of autoimmune diseases (10). There have been no reports of anti-IL-6 autoantibody production induced by immune checkpoint inhibitors.

However, the regulatory mechanism of serum anti-IL-6 autoantibody production remains unknown. Corticosteroids have been shown to be effective adjunctive therapies in the management of septic shock. Multiple studies have indicated that corticosteroid administration may reduce the duration of septic shock (11). In the present case, the patient was treated with systemic steroids for septic shock and responded poorly to vasopressor agents. After steroid administration, the serum anti-IL-6 autoantibody levels decreased, whereas the serum CRP levels increased. Although the anti-IL-6 antibody production in this case is more likely drug-induced than idiopathic, the steroid-induced decrease in anti-IL-6 antibodies provides a new insight into the regulatory mechanisms of anti-IL-6 autoantibody production.

E. lenta is a Gram-positive anaerobic bacterium that naturally inhabits the human gastrointestinal tract. It is considered part of the normal intestinal flora but can be an opportunistic pathogen that induces abdominal sepsis, especially in immunocompromised conditions (12,13). Sepsis due to E. lenta is rare but can be severe and life-threatening in immunocompromised patients (14,15). Our patient was undergoing treatment with carboplatin and etoposide, chemotherapeutic regimens known to induce neutropenia. However, no grade ≥3 neutropenia events were observed throughout the treatment course. In addition, at the time of the sepsis onset, the patient was receiving maintenance therapy with durvalumab monotherapy, and no chemotherapy-induced neutropenia was present. As shown in Table, the patient's blood test results indicated no significant abnormalities in the immune profile.

However, the role of anti-IL-6 autoantibody production in immunosuppression remains controversial. Severe sepsis has been reported in patients with no immune abnormalities other than the presence of anti-IL-6 autoantibodies (6). Inhibition of the IL-6 pathway by tocilizumab is associated with a significantly increased risk of serious bacterial infections (16,17). Thus, it is plausible that suppression of the IL-6 inflammatory pathway by anti-IL-6 autoantibodies may have contributed to the onset of severe sepsis induced by E. lenta in this patient.

Several limitations associated with the present study warrant mention. First, we did not assess the production of anti-IL-6 autoantibodies in patients before the onset of sepsis. Measuring anti-IL-6 antibodies in the serum collected before the sepsis onset might help provide insights into the production of autoantibodies without the influence of sepsis on the immune profile. However, in this case, no serum was available before the onset of sepsis, which prevented such measurements. If the reduction in the patient's serum CRP levels is indicative of anti-IL-6 autoantibody production, it suggests that autoantibodies were likely generated during the third cycle of chemotherapy with carboplatin, etoposide, and durvalumab prior to the onset of sepsis. Second, we did not evaluate whether or not the anti-IL-6 autoantibodies in the patient's serum possessed a neutralizing capacity against IL-6. It would have been possible to assess the neutralizing ability of IL-6 using in vitro experiments with serum-containing autoantibodies. However, in this case, the amount of serum preserved from the patient prior to systemic steroid administration was insufficient for the experiments, preventing us from evaluating the neutralizing capacity against IL-6. Nevertheless, given that the serum CRP levels did not increase during the clinical course despite the onset of sepsis, it can be inferred that the anti-IL-6 antibodies in the serum may have neutralized serum IL-6 induced by sepsis.

This is the first report of CRP-less sepsis with anti-IL-6 autoantibody production following immune checkpoint inhibitor administration. Anti-IL-6 autoantibody production as an irAE should be considered in the differential diagnosis of CRP-less sepsis after immune checkpoint inhibitor administration.

Written informed consent was obtained from the patients' family for publication of this case report and accompanying images.

The authors state that they have no Conflict of Interest (COI).