Reintroduction of immunosuppressive medications in pediatric rheumatology patients with histoplasmosis: a case series

Background Children with rheumatic diseases (cRD) receiving immunosuppressive medications (IM) are at a higher risk for acquiring potentially lethal pathogens, including Histoplasma capsulatum (histoplasmosis), a fungal infection that can lead to prolonged hospitalization, organ damage, and death. Withholding IM during serious infections is recommended yet poses risk of rheumatic disease flares. Conversely, reinitiating IM increases risk for infection recurrence. Tumor necrosis factor alpha inhibitor (TNFai) biologic therapy carries the highest risk for histoplasmosis infection after epidemiological exposure, so other IM are preferred during active histoplasmosis infection. There is limited guidance as to when and how IM can be reinitiated in cRD with histoplasmosis. This case series chronicles resumption of IM, including non-TNFai biologics, disease-modifying anti-rheumatic drugs (DMARDs), and corticosteroids, following histoplasmosis among cRD. Case presentation We examine clinical characteristics and outcomes of 9 patients with disseminated or pulmonary histoplasmosis and underlying rheumatic disease [juvenile idiopathic arthritis (JIA), childhood-onset systemic lupus erythematosus (cSLE), and mixed connective tissue disease (MCTD)] after reintroduction of IM. All DMARDs and biologics were halted at histoplasmosis diagnosis, except hydroxychloroquine (HCQ), and patients began antifungals. Following IM discontinuation, all patients required systemic or intra-articular steroids during histoplasmosis treatment, with 4/9 showing Cushingoid features. Four patients began new IM regimens [2 abatacept (ABA), 1 HCQ, and 1 methotrexate (MTX)] while still positive for histoplasmosis, with 3/4 (ABA, MTX, HCQ) later clearing their histoplasmosis and 1 (ABA) showing decreasing antigenemia. Collectively, 8/9 patients initiated or continued DMARDs and/or non-TNFai biologic use (5 ABA, 1 tocilizumab, 1 ustekinumab, 3 MTX, 4 HCQ, 1 leflunomide). No fatalities, exacerbations, or recurrences of histoplasmosis occurred during follow-up (median 33 months). Conclusions In our cohort of cRD, histoplasmosis course following reintroduction of non-TNFai IM was favorable, but additional studies are needed to evaluate optimal IM management during acute histoplasmosis and recovery. In this case series, non-TNFai biologic, DMARD, and steroid treatments did not appear to cause histoplasmosis recurrence. Adverse events from corticosteroid use were common. Further research is needed to implement guidelines for optimal use of non-TNFai (like ABA), DMARDs, and corticosteroids in cRD following histoplasmosis presentation.

Conclusions: In our cohort of cRD, histoplasmosis course following reintroduction of non-TNFai IM was favorable, but additional studies are needed to evaluate optimal IM management during acute histoplasmosis and recovery. In this case series, non-TNFai biologic, DMARD, and steroid treatments did not appear to cause histoplasmosis recurrence. Adverse events from corticosteroid use were common. Further research is needed to implement guidelines for optimal use of non-TNFai (like ABA), DMARDs, and corticosteroids in cRD following histoplasmosis presentation.

Background
Over 300,000 children in the USA live with juvenile idiopathic arthritis (JIA) and other chronic rheumatic diseases. Advances in understanding the cytokine-driven pathogenesis of rheumatic diseases have provided potent therapeutics. A growing number of FDA approved biologics targeting cytokines are available, including inhibitors of the pro-inflammatory mediators tumor necrosis factor alpha (TNFai), interleukin (IL)-1b (canakinumab), IL-1R (anakinra), IL-6R (tocilizumab), as well as T-cell co-stimulation (abatacept [ABA]) and B-cells (rituximab) [1]. However, cytokines are also critical to control and clear infections, and autoimmunity itself increases infection risk due to immune system dysregulation [1]. Immunosuppressive medications (IM) heighten risk of developing serious infections, including granulomatous infections like histoplasmosis [1][2][3]. Histoplasmosis is caused by inhaling spores of Histoplasma capsulatum, a fungus found in soil contaminated with bird or bat droppings and endemic to areas such as the Ohio and Mississippi River Valleys [4][5][6]. While typically causing mild or asymptomatic pulmonary infection in immunocompetent patients, H. capsulatum can disseminate systemically in susceptible hosts, leading to poor outcomes and requiring a prolonged antifungal course [6][7][8]. Furthermore, itraconazole, the recommended oral antifungal for H. capsulatum, can result in drug interactions due to inhibiting the CYP3A4 pathway [9][10][11][12][13].
Our center at Nationwide Children's Hospital (NCH) in Columbus, Ohio facilitates over 6,000 visits per year for children with rheumatic diseases (cRD). Standard rheumatic treatments include nonsteroidal antiinflammatory drugs (NSAIDs), corticosteroids, and disease-modifying anti-rheumatic drugs (DMARDs) (methotrexate [MTX], leflunomide, hydroxychloroquine [HCQ], etc.), but many of these patients are or were also managed with TNFa antagonists (etanercept, infliximab, adalimumab, etc.) or other biologics at some point. It is controversial whether low dose MTX increases infection susceptibility, whereas combination IM therapy, corticosteroids, or biologics do, while HCQ does not [1,3,14,16,17]. Indeed, TNFai complicated by infections is well reported [3,8,[17][18][19][20][21][22]. For patients receiving TNFai who present with fever or pulmonary symptoms, extensive infectious workup is undertaken given the broad infectious differential. It is recommended that IM be held during acute infections, but this interruption in treatment may lead to a rheumatic disease flare. Optimal rheumatic disease management is unknown, including the duration of IM discontinuation and, importantly, safety data about reintroducing TNFai or other biologics are largely lacking [1,3,23,24,22]. Meta-analyses suggest some biologics carry a lower infection risk compared to TNFai [25]. For example, an observational study of adult patients with rheumatoid arthritis (RA) noted participants receiving ABA had lower rates of infection-related hospitalization compared with other biologics [26]. However, utilization of these non-TNFai following histoplasmosis diagnosis (FHD) and histoplasmosis clearance (HC) has not been reported and merits further scrutiny. In our report, we describe the treatment course, IM management, and outcomes of cRD diagnosed with histoplasmosis.

Methods
A retrospective chart review of electronic health records identified children followed for rheumatic disease management at NCH, who were diagnosed with and treated for proven, probable, or possible histoplasmosis between January 2011 and January 2018. Primary inclusion criteria included: diagnosis with a pediatric rheumatic disease and a later/concurrent diagnosis of histoplasmosis. Data were collected for IMs prior to and following histoplasmosis diagnosis and histoplasmosis course. IMs were characterized as (1) biologic DMARDs, such as TNFai or non-TNFai, (2) non-biologic DMARDs, and (3) corticosteroids, including intra-articular steroid injections (IASIs) and oral/inhaled/intermuscular/intravenous (systemic) steroids. Systemic steroids were classified as "burst" (high dose > 10 mg/day provided to treat active rheumatic or non-rheumatic disease, like asthma) or "maintenance" (low dose ≤ 10 mg/day). Histoplasmosis resolution was defined as clearance of urine and serum antigens and clinical resolution of signs and symptoms. This study was reviewed and approved by the local Institutional Review Board IRB17-01241.
Histoplasmosis diagnosis and course are shown in Table 2. Three (33 %) patients were able to identify potential exposures (recent bonfire, farm residence, and birds/bats in attic). H. capsulatum was detected by bronchoalveolar lavage, pulmonary wedge biopsy, Histoplasma antigen, and/or serology for a total of 3 proven, 5 probable, and 1 possible histoplasmosis cases [27]. Five (56 %) patients developed disseminated histoplasmosis (DH), and 3 (33 %) developed pulmonary histoplasmosis (PH). The possible case was not further categorized as PH or DH but included in this series due to positive serology and clinical symptoms (data not shown). For the 5 DH patients, 2 were receiving both TNFai and MTX, 1 was receiving TNFai, MTX, and steroid bursts, 1 was receiving DMARDs only, and 1 was receiving DMARD and maintenance steroids. Of the 3 PH patients, 1 was receiving TNFai and MTX and 2 were not receiving IMs. The patient with the possible histoplasmosis infection was receiving DMARDs and burst steroids. At histoplasmosis diagnosis, IM, except HCQ and steroids, were held. Six (67 %) patients (5 DH and 1 PH) required hospitalization for acute histoplasmosis for a median of 14 days [6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22], with 1 DH patient needing admission to the intensive care unit for mechanical ventilation for 14 days. During follow up, 7/9 (78 %) demonstrated complete resolution of antigenemia and clinical symptoms with a median time to clearance of 12.7 months (2.3-37.2) FHD and 2 were clinically asymptomatic at 11 and 25 months FHD while showing low antigenemia. Four of nine (44 %) patients continued prophylactic itraconazole after HC for a median of 23.6 months (5.0-70.0). Median follow-up after HC was 29.3 months (0-70.3), with a total follow-up from histoplasmosis diagnosis of 33 months (11-88.5). There was no observed mortality or histoplasmosis relapse during this time.

Patient Use of Immunosuppressive Medications Following Histoplasmosis Infection was Necessary
Patients initially managed their underlying rheumatic conditions during antifungal therapy with NSAIDs or HCQ, but while receiving these medications all patients experienced rheumatic disease flare requiring further intervention.  (Table 3).

Biologics and DMARDs
Active rheumatic disease led to initiating biologics and DMARDs in 7 patients (Table 4) including: 3 ABA (with   (Table 4).

Discussion
In this case series, we describe treatment and disease course of cRD diagnosed with histoplasmosis. Outcomes were favorable, with non-TNFai biologic and DMARD medications, including ABA and HCQ, being safely reinitiated in patients requiring rheumatic disease alleviation. Although TNFai therapy complicated by granulomatous infections like histoplasmosis has been well-recognized [7,[17][18][19][20]23], few published reports address reinstitution of IM treatments following histoplasmosis diagnosis [8,19,22,23]. Moreover, there is a lack of guidance on underlying disease management and long-term followup [19,22]. In this series, rheumatic diseases were initially controlled by NSAIDs and HCQ, but all cRD eventually needed additional treatments. Moreover, we observed that Cushingoid symptoms developed in almost half (4/9) of patients receiving corticosteroids during antifungal treatment. Thus, managing rheumatic disease during recovery from histoplasmosis poses challenges as IM are halted upon serious infections.

Itraconazole and Steroids
The often-chronic nature of fungal and rheumatic therapies make selecting compatible medications critical. In severe DH, initial therapy may begin with intravenous amphotericin transitioning to oral itraconazole, with a total antifungal duration of at least 12 months [7,21]. Itraconazole is an effective antifungal, but it potently inhibits the CYP3A4 pathway, thereby decreasing steroid metabolism [8,9,11,12]. Nevertheless, adjunctive corticosteroids may still be indicated for fungal infections and other inflammatory states, like rheumatic flares [1,7,8]. Furthermore, systemic or local corticosteroid use by itself can also cause iatrogenic Cushing's syndrome, although drug interactions increase risk [9,28,29]. Shortly following IASIs, 44 % of our cRD developed nonallergic adverse reactions with prominent Cushingoid features. To our knowledge, this is the first description of Cushing's syndrome in cRD undergoing itraconazole treatment following IASI and systemic steroids. However, previous reports associating itraconazole and steroid therapy with hypothalamic-pituitary-adrenal (HPA) axis suppression and, more infrequently, overt Cushing's syndrome have been published, usually for patients with cystic fibrosis taking inhaled steroids [9,10,13,15]. The somewhat surprisingly high fraction of patients presenting with overt Cushingoid features in our report could perhaps be due to differences in systemic steroid levels, route of steroid administration, type of steroid, underlying disease, and/or drug interactions. For example, a study comparing the effects of itraconazole on a single dose of methylprednisolone versus prednisone in healthy volunteers indicated prednisone's pharmacokinetics were unaffected while methylprednisolone concentration increased [12]. In our limited cohort, is difficult to draw firm conclusions about what/if combinations of corticosteroids and itraconazole place patients at greatest risk for HPA axis-related syndromes. High dose steroidal bursts combined with IASI perhaps contribute, but more research is needed, for example, to determine if route of steroid administration matters and if underlying comorbidities affect risk. Regardless, there is an urgency to determine optimal strategies of IM usage in the setting of concurrent histoplasmosis that facilitate histoplasmosis recovery and rheumatic disease management. Moreover, perhaps HPA axis monitoring should be considered in patients needing itraconazole and steroids [13]. Table 4 Reintroduction of immunosuppressive medications and patient histoplasmosis course. Patient history was examined for timing of systemic immunosuppressive medication reinitiation (localized steroid use was not included). Timing of reinitiation was compared with state of histoplasmosis, with "still +" denoting patient had not cleared their histoplasmosis infection based on antigenemia and/or clinical symptoms and "HC for #" indicating how many months it had been since histoplasmosis clearance to beginning the immunosuppressive medication. Median months of follow-up since histoplasmosis clearance was 45.6 (14.7-78.9)

Pt
Immunosuppressive Medication Resumption (months following histoplasmosis diagnosis, histoplasmosis status) Months to HC

Histoplasmosis and Immunosuppressants
Acute DH develops in about 1 of every 2000 acute histoplasmosis cases, and most acute DH occurs in immunosuppressed patients [30]. Markedly, immunocompromised patients (e.g., AIDS) have a 10-fold increased risk for developing DH compared to immunocompetent persons, and DH mortality rates in severely immunocompromised patients can reach 30 % [31][32][33]. Chu [19,20,22,23]. Notably, Vergidis et al. described outcomes of 98 patients with various autoimmune diseases diagnosed with histoplasmosis, wherein 3/98 patients continued non-combination TNFai, 11/58 patients continued DMARDs, and 13/23 patients continued corticosteroids. The 3 histoplasmosis cases maintaining TNFai, which is not the standard of care, were described as "mild" and all patients achieved remission after 6-10 months of itraconazole. During follow-up (median 32 months), 2/25 patients resuming TNFai (and 1/49 who had not) developed histoplasmosis recurrence with 1 fatality (taking TNFai). Follow-up data were unavailable for the other patients, and resumption information on DMARDs and corticosteroids was not reported [22]. Furthermore, after TNFai was held in 14/15 adult patients with RA, 4 resumed them at least 5 months FHD. The patient continuing TNFai, whose lung biopsy showed inactive disease, did well, while one patient who resumed TNFai developed clinical recurrence, permanently halted it, and recovered [23]. In our case series, persistent rheumatic disease activity led to resumption of biologics and non-HCQ DMARDs in 1 cSLE and 6 JIA patients FHD. However, no patients were restarted on TNFai due to safety concerns even if histoplasmosis disease was mild. One JIA patient did not require IM and the patient with MCTD proceeded on their pre-histoplasmosis regimen of HCQ and maintenance steroids. The 3 patients resuming MTX had favorable outcomes. HCQ was continued without issue for the cSLE and MCTD patients, and HCQ was an alternative (or additive) treatment option for JIA patients. Unlike some evidence on MTX, HCQ does not increase infection risk, but all 3 patients starting either medicine cleared their histoplasmosis without recurrence.
Considering timing of biologic reinitiation FHD, 5 cRD began the non-TNFai ABA before [2] or following [3] HC, while the initiations of tocilizumab or ustekinumab were only after HC. Although no patients were receiving ABA, tocilizumab, or ustekinumab prior to their histoplasmosis diagnosis, all 4 cRD beginning them have remained free of histoplasmosis recurrence and neither patient starting ABA while still antigen positive experienced worsening of their histoplasmosis. Moreover, 1 of these 2 ABA patients has achieved HC to date. Despite the small numbers, the use of ABA in these cRD resulted in good rheumatologic outcomes without exacerbating the infection, suggesting that ABA may be an option for treating rheumatic patients with histoplasmosis. The inability to support some cRD without higher tier medications underscores the need to promote infection safety while controlling rheumatic disease. For example, considering lower risk (non-TNFai) biologics, like ABA, for patients in endemic fungal areas such as the Ohio and Mississippi River Valleys where histoplasmosis is the most frequent TNFai-associated endemic fungal infection [6]. While screening for histoplasmosis prior to beginning biologics is not routinely recommended, nevertheless physicians prescribing IM must be aware of histoplasmosis presentations and risk factors [1].

Conclusions
Histoplasmosis is an important complication of the immunosuppressive therapy utilized to treat cRD in endemic areas. This small case series proposes corticosteroids, non-HCQ DMARDs, and non-TNFai biologic treatments initiated months FHD or after HC did not appear to induce histoplasmosis recurrence, and, for a majority of patients, permitted histoplasmosis antigen clearance. While receiving antifungal therapy, necessary intra-articular or systemic steroid usage may potentially merit HPA axis monitoring. However, firm conclusions from our report are limited due to the small number of patients and varying IM therapies. Given the crucial role of IMs like biologics and DMARDs in cRD care, additional studies are needed to examine the long-term safety and efficacy of ABA, tocilizumab, ustekinumab, MTX, HCQ, and corticosteroids during histoplasmosis treatment and recovery.