Dilemma of immunosuppression and infection risk in systemic lupus erythematosus

Abstract

Patients with SLE are at high risk of various infections as evidenced by a number of studies. The main determinants of infection in SLE are disease activity, organ damage, and often inevitable medication. The molecular and cellular mechanisms underlying infection remain unclear. Impaired immunity, immunosuppressants and corticosteroids clearly increase the risk of infection, whereas some medications, such as low-dose IL-2, hydroxychloroquine and IVIG are safe in SLE patients with substantial evidence. It is important to balance the immunosuppression and infection risks in practice. This article focuses on medication-related infections in SLE and discusses the therapeutic options for the disease in clinical practice.

Rheumatology key messages.

• Infection is a common complication in SLE.

• Immunosuppressants and corticosteroids increase the risk of infection in SLE.

• New therapies that don’t increase the risk of infection in SLE are anticipated.

Introduction

Infection in SLE is an important cause of hospitalizations and mortality. Previous studies showed that infection is the leading cause of death in SLE patients [1, 2]. Infection accounts for ∼29.2–43.9% of the morbidity of SLE patients [1–3]. Other factors, including atherosclerosis and nephritis, can also increase morbidity [4]. Opportunistic infections are very common clinically and are related to impaired cellular and humoral immune functions in patients with SLE.

The underlying mechanisms for susceptibility to infection remain unclear, but various explanations have been proposed, including immunosuppressive therapies and immunological changes responsible for infections. On the other hand, pathogens are thought to induce autoimmunity via molecular mimicry and have been shown to stimulate the production of IFN and anti-dsDNA antibodies in SLE patients.

It has been suggested that dysfunction in T cells, B cells, and NK cells, might play a role in the development of infection in SLE [5–7]. Defective phenotypes and functioning of neutrophils, monocytes, macrophages, and dendritic cells (DCs) have been identified in SLE patients. These defects are important in the pathogenesis of SLE, including ineffective apoptotic debris clearance, self-antigen presentation, and inflammatory cytokine production [8].

Moreover, immunosuppressive treatments and corticosteroids (CSs) can impair protective immunity, which increases the risk of infections in these patients. Of note, certain medications such as low-dose IL-2 (Ld-IL2), HCQ and IVIG, emerge as promising therapies for treating autoimmune disorders, including SLE, without interfering with anti-infective immune responses [9–12] (Fig. 1). In this review, we investigate drug-associated infection and potential approaches to improve the outcome of SLE patients.

Figure 1.

Targeting treatment and infection in SLE. Diverse treatments targeting different molecules and immune cells are shown. The black lines with arrows indicate the activation of immune cells; the black dashed lines represent unidentified effects of the treatment. JAK inhibitors: Janus kinase inhibitors; Ld-CSs: low-dose corticosteroids; Ld-IL2: low-dose IL2; MSC: mesenchymal stem cell; DC: dendritic cells; Th17: T-helper type 17; Tfh: T follicular helper; Treg: T regulatory cell; APRIL: a proliferation-inducing ligand; BAFFR: B-cell activating factor receptor; Blys: B lymphocyte stimulator; BCMA: B-cell maturation antigen; mTOR: mechanistic target of rapamycin; TACI: transmembrane activator and calcium modulator cyclophilin ligand interactor; TLRs: Toll-like receptors

Prevalence of infection in SLE

Despite improved management of patients with SLE, infection remains the main cause of morbidity and mortality. The prevalence of infection in SLE varies from study to study, ranging from 29.2% to 43.9% [1–3]. In a multicentre study, Cervera et al. reported that 36.0% of patients with SLE developed infections during a 10-year follow-up [13]. Similar results are shown in another study, with 29.2% of patients suffering at least one major infection, with 13.3% of those infections being nosocomial infections [3].

A meta-analysis comprising 4469 SLE patients showed that infection is the main cause (33.2%) of death in SLE patients [14]. A 30-year long-term follow-up study revealed a similar result of 25.3% deaths due to infection in a cohort of 470 SLE patients [1]. Feng et al. found that nearly half of the mortality (46.2%) in the late-onset elder lupus were caused by infections, and pulmonary infections were the main type of infection [2]. An increased incidence of sepsis was observed among patients with higher disease activity. A recent real-world study also showed that the most common infection site is the respiratory system, followed by the skin and urinary system, while bacteria, viruses and fungi are the most common pathogens [15]. Similar trends were found in a previous study [13].

Impaired immunity increased the risk of infection in SLE

An impaired innate and adaptive immune system in SLE increases the risk of infection. T cells and B cells provide potentially protective effects by removing microbial invaders and neutralizing infectious antigens [16]. Once a patient has been infected, immunologic abnormalities induce the self-reactive immune cells to release proinflammatory cytokines and produce immune complexes, then the pathogens contribute to the deposition of foreign-antigens and self-antigens, leading to inflammation and tissue damage.

Dysfunction of the innate and adaptive immune systems is associated with an increased risk of infection among patients with SLE. In light of the aberrant expression of IFNγ, IL-1 and TNFα, impaired T cell–mediated cytolytic activity participates in the poor response to infections in SLE [17]. As previously described, CD8+ T cells have been shown to be involved in accelerating B cell differentiation with antibody isotype variations [18]. We and others have also found that the cytotoxicity of NK cells is suppressed when they are exposed to the dysregulated immune environment of SLE, not only through reduction in number but also in the reduced response of NK cells to IL-2 [19].

B cell subtypes were altered in SLE patients, especially naïve B, memory B and plasma cells, resulting in autoantibody production. Several studies on antibody response have demonstrated the protective response of B cells against viral infection [20, 21]. Aberrant function in B cells needs further investigation in lupus patients.

As regards the innate immune system, abnormal PMNs incur defective chemotaxis and phagocytosis, accompanied by aberrant membrane recognition and attachment to microorganisms. In addition, acquired deficiency of the early components of the complement system (C1q, C4 and C2) has been identified in SLE and predisposes patients with SLE to infections with encapsulated organisms. The reduced complement receptors CR2 result in the deposition of C3 fragments and attenuate the ability of B cells to stimulate the alternative pathway in SLE [22].

Therefore, the susceptibility to infection is associated with dysregulated immune response to antigens, which is aggravated by SLE disease activity [23]. Our present study illustrated that positive anti-dsDNA and high activity of the disease were associated with an increased risk of infection. The peripheral ratio of CD8+ T cells to CD4+ T cells, and the infiltration of CD8+ T cells in activated tissue, were observed to be dramatically diminished in infected SLE patients and mouse models.

Diverse therapy associated with the susceptibility to infection in SLE

Treatment options in SLE that are now commonly applied are antimalarial drugs, glucocorticoids, immunosuppressive drugs, and biologics [24]. These medications can interfere with cell-mediated and humoral immunity to achieve control of the disease by suppressing inappropriately activated T cells, B cells, and other immune cells (Fig. 1) [25, 26]. Meanwhile, suppression of immunity by medications greatly increases the risk of infection in patients with SLE, including the most common viral infections [27].

Glucocorticoids (GCs) exert powerful anti-inflammatory effects and therefore have been a mainstay in the treatment of SLE, especially in moderate and severe SLE presenting with organ damage. In a cohort study involving 3030 SLE patients and followed for 4 years, GC administration was shown to significantly increase the incidence of infection [28]. It was clearly suggested that prednisone dose was a potent risk factor for increased infection in a multiracial, multiethnic cohort study involving 1243 patients [29]. In addition, GCs also increase the risk of atherosclerotic death [4], and they should not be used at doses above 5 mg of prednisone q.d. in the long term [30].

Immunosuppressive drugs such as MMF and AZA inhibit cell proliferation, increasing the risk of infection while reducing the disease activity (Table 1) [35]. In an open-label randomized controlled clinical study, patients being treated with MMF and AZA had a significantly higher incidence of infection, up to 48.7% [36]. CYC, an alkylating drug still commonly used in severe SLE patients, is also generally associated with infectious complications [51]. CSA and tacrolimus have been found to increase the risk of infection in SLE patients [36, 52, 53]. The calcineurin-mediated signalling pathways have been shown to contribute significantly to fungal virulence, and inhibiting calcineurin is a possible antifungal strategy [54, 55]. In a randomized, double-blind, phase II study of 314 patients with SLE, patients being treated with baricitinib were found to have a higher infection incidence than the placebo control group [56]. In patients being treated with a combination of rituximab and belimumab, up to 73.3% increase in infection was found [57]. A randomized, placebo-controlled phase IIb clinical study reported a higher incidence of infection in patients being treated with high-dose atacicept [58]. Treatment with anifrolumab has a good response in SLE, but an increased rate of infection. In a randomized controlled trial, herpes zoster and bronchitis occurred in 7.2% and 12.2% of patients, respectively, who received anifrolumab [59]. In addition, the clinical trial of ocrelizumab was terminated prematurely owing to increased infection in patients when used in combination with MMF [49]. As for targeted T cell therapy, ustekinumab was shown to enhance infection in a clinical study among active SLE patients [60].

Table 1.

Infection rates with different medications in SLE

Medications	Source (author, year）	Subject (n)	Design	Infection rate	Pathogen
CSs	Singh JA et al. 2016 [31]	32 RCTS, n = 2611	Systematic review	Serious infection:	NA
				CSs 18.1%
Immunosuppressives
CYC	Zheng Z et al. 2022 [32]	CYC (n = 156);	Open-label, phase III trial	Severe infection:	Herpes zoster, virus, bacterium, tuberculosis
		TAC (n = 158)		CYC 16.2%; TAC 8.9%
	Shao Miao et al.. 2022 [33]	SILD CYC (n = 506);	Retrospective, multicentre	SILD CYC 13.04%;	NA
		HD CYC (n = 256)		HD CYC 22.27%
MMF	Appel GB et al. 2009 [34]	MMF (n = 185)	Open-label	MMF 68.5%; CYC 61.7%	Herpes virus, influenza virus, bacterium, fungus, candida, tuberculosis, etc.
		CYC (n = 185)
AZA	Dooley MA et al. 2011 [35]	MMF (n = 116);	RCT	Infection:	NA
		AZA (n = 111)		MMF 79.1%; AZA 78.4%
TAC	Mok CC et al. 2016 [36]	MMF (n = 76);	Open-label	Major infection:	Herpes zoster, virus
		TAC (n = 74)		TAC 5.4%
				Minor infection:
				TAC 20.1%
CSA	Fervenza FC et al. 2019 [37]	RTX (n = 65);	Open-label, multicentre	RTX 26.2%; CSA 30.8%	NA
		CSA (n = 65).
	Moroni G et al. 2006 [38]	CSA (n = 36);	RCT	CSA 19.4%; AZA 42.4%	NA
		AZA (n = 33)
Voclosporin	Rovin BH et al. 2021 [39]	Voclosporin (n = 179);	Multicentre, RCT,	Infection:	NA
		Placebo (n = 178)	Phase III trial	Voclosporin 65%;
				Placebo 57%
LEF	Zhang M et al. 2019 [40]	LEF (n = 48);	Prospective, multicentre, RCT	LEF 33%; CYC 35%	Virus, bacterium
		CYC (n = 52).
	Cui T et al. 2005 [41]	LEF (n = 35);	Multicentre, controlled	LEF 22.9%; CYC 12.5%	Virus, bacterium
		CYC (n = 16)
MTX	William S et al. 2012 [42]	MTX+Placebo (n = 207);	RCT,	Infection:	NA
		MTX+Ocrelizumab (n = 398)	Phase III trial	MTX+Placebo 51.2%;
				MTX+Ocrelizumab 51.5/52%
Anti-malarial drugs
HCQ	Sakai R et al. 2020 [43]	HCQ (n = 1095);	Retrospective, longitudinal study	Hospitalized infection:	NA
				HCQ 4.5%;
		Non-HCQ (n = 1095)
				Non-HCQ 5.6%
Biologics
Rituximab	Merrill JT et al. 2010 [44]	RTX (n = 169);	Multicentre, RCT	Infection:	Bacterium, virus, fungus
		Placebo (n = 88)	Phase II/III trial	RTX 82.2%;
				Placebo 83.0%
	Rovin BH et al. 2012 [45]	Rituximab (n = 73);	RCT	Infection:	Bacterium, virus
		Placebo(n = 71)	Phase III trial	RTX 84.9%;
				Placebo 90.1%
Belimumab	Zhang F et al. 2018 [46]	Belimumab (n = 471);	Multicentre, RCT	Severe infection:	Virus, bacterium
				Belimumab 5.5%;
			Phase III trial
		Placebo (n = 236)
				Placebo 5.3%
	Navarra SV et al. 2011 [47]	Belimumab (n = 578);	Multicentre, RCT	Belimumab 67/68%;	NA
				Placebo 64%
			Phase III trial
		Placebo (n = 287)
Ld IL-2	He J et al. 2020 [9]	Ld-IL2 (n = 30);	RCT	Ld-IL2 6.9%;	NA
				Placebo 20.0%
		Placebo (n = 30)
	Humrich JY et al. 2022 [12]	Ld-IL2 (n = 50);	Multicentre, RCT	Serious infection:	NA
			Phase II trial	Ld-IL2 2%
		Placebo (n = 50)
	Zhou P et al. 2021 [15]	LD-IL2 (n = 219);	Retrospective study	Ld-IL2 7.3%;	Virus, bacterium, fungus
				Placebo 25.1%
		Placebo (n = 446)
Blisibimod	Merrill JT et al. 2018 [48]	Blisibimod (n = 245);	RCT	Infection:	Herpes zoster virus and other viruses, bacterium
				Blisibimod 43.7%;
		Placebo (n = 197)
				Placebo 43.9%
Ocrelizumab	Mysler EF et al. 2013 [49]	Ocrelizumab (n = 253);	Multicentre, RCT	Infection:	Herpes zoster virus, bacterium, Pneumocystis jiroveci, cytomegalovirus, cryptococcus
			Phase III trial	Ocrelizumab 59.1/68.3%,
		Placebo (n = 125)
				Placebo 56.0%
Anifrolumab	Chia YL et al. 2022 [50]	Anifrolumab:	Post-hoc of RCTs	Non-opportunistic serious infection:	Virus, herpes zoster virus, bacterium
				Anifrolumab 150 mg 2.2%;
		150 mg, n = 91;
		300 mg, n = 356;
				Anifrolumab 300 mg 3.9%;
		Placebo (n = 366)
				Placebo 4.9%

CSs: corticosteroids; HD: high dose; LD-IL2: low-dose interleukin 2; LEF: leflunomide; MMF: mycophenolate mofetil; MTX: methotrexate; RCT: randomized controlled trial; RTX: rituximab; SILD: short-interval, lower-dose; TAC: tacrolimus.

Several medications (e.g. MTX, Ld-IL2 and HCQ) are considered to be safe in SLE patients in terms of not increasing infection risk, and Ld-IL2 therapy has emerged as a new therapeutic option in a variety of autoimmune diseases, aiming to reduce the risk of infection (Fig. 2, Table 1) [15, 61–65]. It has been suggested that Ld-IL2 improves the immune response to the hepatitis-B vaccine in uremic patients [66]. A similar notion was supported by a clinical trial that demonstrated that Ld-IL2 ameliorated HCV-induced vasculitis without increasing viral load [67]. Recently, we have shown that Ld-IL2 treatment reduced the incidence of infection in SLE compared with the control group during the 6-month follow-up period (7.3% vs 25.1%) [15]. However, the relevant mechanisms need to be further studied.

Figure 2.

Ld-IL2 reduced the risk of infections. Ld-IL2 reduces the incidence of bacterial and viral infections, as shown in previous studies. The red line in the figure indicates a defined reduction of infection; the black dashed line represents potential as suggested by experimental data but with insufficient clinical evidence. URTV: upper respiratory tract virus; LD-IL2: low-dose IL 2

Antimalarial drugs such as HCQ, the cornerstone drug for SLE treatment, provide safe and favourable therapeutic effects in SLE patients, even in those with infection [11, 68] (Table 1). Numerous large cohort studies have demonstrated that antimalarial drugs cause no increased risk of infection in SLE patients and might even play a role in preventing infections [28, 29, 69].

Therapies restoring immune homeostasis without interfering with anti-infective immunity in SLE

Immune homeostasis is important in maintaining remission and protecting against infection in SLE. It has been suggested that Ld-IL2 therapy has a beneficial effect by restoring immune tolerance without imposing immunosuppression (Fig. 1), thereby reducing the risk of infection. In addition to regulating various subsets of CD4+ T cells, IL-2 is essential for the proliferation of NK cells and CD8+ T cells, which enhance the anti-infection function [70]. We and others have demonstrated that NK cells, especially CD56^bright NK cells, are impaired in active SLE. The number and function of NK cells were preferentially increased with Ld-IL2 administration, potentially increasing anti-infectious immunity [9]. IL-2 treatment also improved CD8+ T cell proliferation and accelerated viral clearance in mice infected with influenza [15]. Not only the immune cells, but serum complements increased significantly after Ld-IL2 therapy [71, 72], which could also be the reason for decreasing infection levels in patients with SLE.

Clinical trials have provided evidence that mTOR blockade by rapamycin can enhance responsiveness to vaccination and reduce infections with the influenza virus in healthy elderly subjects [73]. It has been shown that HCQ was associated with lower infection rates in several studies [74–76]. The main functions of HCQ are exerted by pH-dependent iron deprivation and increasing lysosomal pH. An increase in pH caused by HCQ inhibits invariant chain clipping by proteases. HCQ selectively inhibits the binding of low-affinity self-antigen peptides to the MHC-II binding site, but not of high-affinity foreign-antigen peptides, which is probably associated with the lower infection rates in patients receiving HCQ treatment [77]. More recent studies have shown that HCQ can inhibit endosomic disassembly of the internalized virus and thus reduce the release of viral RNA to the cytoplasm for replication, and inhibit Zika virus (ZIKV) RNA replication by preventing ZIKV-induced autophagy [78].

Early studies showed that IVIG exerts immunomodulatory effects in SLE, decreasing the risk of infection in patients after bone marrow transplantation [79]. However, recent data has shown that the role of IVIG in reducing infection needs to be investigated further [80–82]. Therefore, it remains a challenge to find more therapeutic approaches to controlling disease activity, and meanwhile balance the impaired immune regulation of SLE (Fig. 1). In addition, it seems that human umbilical cord mesenchymal stem cells and T cell vaccine may have immune homeostasis effects in SLE [83–86].

Conclusion

Infection is still a significant challenge causing increased morbidity and mortality in SLE. A dysregulated immune response, disease severity, and some medications profoundly exacerbate the risk of infection in SLE. In contrast to many immunosuppressants and CSs, Ld-IL2, IVIG, HCQ and rapamycin might be safe to be used in SLE patients with infection. Further study of the underlying mechanism of infection with aberrant immunity and novel therapies are expected to facilitate control of SLE with long-term remission.

Funding

This study was supported by the National Natural Science Foundation of China (NSFC 32141004). This paper was published as part of a supplement financially supported by Janssen Medical Affairs Global Services, LLC, a part of the Janssen pharmaceutical companies of Johnson & Johnson.

Disclosure statement: The authors have declared no conflict of interest.