Lethal Lung Disease and Biologics: Is There a Link?

A new, potentially life-threatening parenchymal lung disease is increasingly being detected in patients with systemic juvenile idiopathic arthritis (sJIA), according to recent reports.

Systemic JIA is distinct from other forms of JIA in that the arthritis is accompanied by quotidian spiking fevers, rash, and serositis. In many patients, the diagnosis has been a dismal one. In approximately 10% to 15% of patients, a severe, often lethal complication known as macrophage activation syndrome (MAS) develops, which can result in hemophagocytosis and release of cytokines such as interferon (IFN)-γ, hepatic dysfunction, and coagulopathy. However, pulmonary involvement in sJIA typically has been mild, most often manifesting as pleuritis or pleural effusion.

Traditionally, sJIA had been treated with corticosteroids and NSAIDs, but the introduction of cytokine-targeting biologics some 15 years ago led to rapid, life-changing improvements in up to two-thirds of patients. These agents include the interleukin (IL)-1 inhibitors anakinra (Kineret) and canakinumab (Ilaris) as well as the IL-6 inhibitor tocilizumab (Actemra).

Clinical trials of canakinumab and tocilizumab for sJIA were published in 2012 in the New England Journal of Medicine, accompanied by an editorial co-authored by Elizabeth D. Mellins, MD, of Stanford University. At the time, she noted that the new agents "signal a new era in the treatment of systemic JIA," but warned that "cytokine blockade may create an imbalance in the cytokine network, thereby driving or unmasking previously uncommon complications."

An early warning of a possible new disease complication was a study published in 2013 of 25 patients with lung manifestations including pulmonary arterial hypertension in 64%, interstitial lung disease in 28%, and alveolar proteinosis in 20%. Mortality at 10 months following the onset of pulmonary disease approached 70%.

Mellins then organized a meeting in 2016 where rheumatologists, pulmonologists, and pathologists discussed 33 cases of sJIA involving interstitial lung disease; the high mortality prompted her to begin reporting the concerning findings at conferences. Questions were being raised linking possible exposure to biologics with the unusual lung disease because of the temporal association.

"Given that sJIA and its adult counterpart disease, adult-onset Still's disease, have been recognized for over 120 years, it is striking that this complication was not recognized and described until the past decade," said Michael A. Ombrello, MD, who heads the Translational Genetics and Genomics Unit at the National Institute of Arthritis and Musculoskeletal and Skin Diseases in Bethesda, Md.

In a recent editorial accompanying a study of 18 patients with this complication, Peter A. Nigrovic, MD, of Brigham and Women's Hospital in Boston, cautioned that the task today is to "seek with urgency to understand this dark cloud over the otherwise dramatically successful advent of biologic treatment in sJIA."

"The storm is here, and navigating sJIA treatment has become even more challenging than before," Nigrovic wrote.

The multicenter cohort

Mellins and her colleagues have now published a retrospective study of 61 patients, in which 46 of 61 exhibited a constellation of unusual clinical findings, including acute erythematous clubbing. This subgroup was notable for exposure to the IL-1 and 6 inhibitors, with the predominant pathology being pulmonary alveolar proteinosis/endogenous lipoid pneumonia (PAP/ELP).

Among affected children, 5-year survival was only 42%, Mellins' group reported on the recently established pre-peer review platform medRxiv.

They analyzed clinical, radiologic, pathologic, and genetic data from their cohort compared with 471 sJIA patients in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) registry.

The clinical picture at presentation included few if any respiratory symptoms, although hypoxia was detected in 43% and pulmonary hypertension in 30%. Other clinical findings included pruritic rash in more than half of patients and eosinophilia in more than one-third. In 38%, anaphylaxis to tocilizumab was reported, compared with 0.6% of those in the comparator cohort. Unexplained abdominal pain also was observed.

High levels of serum ferritin and lymphopenia were observed during the year prior to the onset of lung disease, suggesting a possible "extended incubation phase associated with smoldering inflammation," Mellins and colleagues observed.

The most common pattern seen on chest CT involved peripheral septal thickening with or without ground glass opacities, although less frequent patterns included peripheral consolidation and peribronchovascular consolidation as well as hyper-enhancing lymph nodes. Radiologic evidence of fibrosis was uncommon, and pleural effusion was rare.

Among patients who had biopsy or autopsy tissue analysis, the patchy PAP/ELP pattern was accompanied by vascular abnormalities. On electron microscopy, macrophage accumulation was seen, but genetic analysis found no association with genes that cause primary hereditary PAP.

Patients in the lung disease group were significantly younger than CARRA registry controls at sJIA diagnosis (2.3 vs 5.2 years) and also more often had trisomy 21 (9.8% vs 0.2%), which can increase susceptibility to adverse drug events.

Survival was "drastically lower" among patients with lung disease, with mortality rates of 159 per 1,000 patient-years compared with a U.K. cohort of patients with systemic JIA treated with biologics (3.9 per 1,000), and the most common causes of death were diffuse lung disease and MAS.

Most patients (19 of 23 biopsied) with PAP/ELP had been exposed to IL-1/IL-6 inhibitors, with a median time from initiation of treatment to the onset of lung disease of 1.2 years. "Thus, IL-1/IL-6 inhibitor exposure may promote development of a PAP/ELP-like disease in a subset (apparently small and perhaps genetically susceptible) of sJIA patients, among the substantially larger group of patients who derive striking benefit from these inhibitors," Mellins and colleagues wrote.

There are a number of possible explanations for an association between lung disease and IL-1/IL-6 inhibition, she said. One is that a subset of patients met the criteria for drug-related eosinophilic systemic syndrome, which is a potentially life-threatening hypersensitivity reaction characterized by skin eruptions and hematologic abnormalities including high eosinophil counts.

Another hypothesis is related to an earlier study of children with sJIA, in which some children developed evidence of a type I interferon response induced by anakinra. A second study found that anakinra responders showed an exaggerated response to interferons (both type I and type II), suggesting that the pathology may be driven by interferons. Notably, an underlying type I interferonopathy has recently been described in trisomy 21. In addition, a mouse model has demonstrated that interferon-γ upregulation can cause a similar pattern of lung inflammation.

It's also possible that cytokine inhibitors can influence the larger cytokine network, making the child more responsive to other cytokines, or can affect homeostasis in the lung. "There are multiple ways these inhibitors might increase vulnerability to PAP lung disease," Mellins explained in an interview. "At the moment I'm in favor of the notion that multiple factors may be involved, all of which need to be explored."

She and her colleagues noted that more than 50 additional cases of lung disease have been observed in patients with sJIA, and there also have been reports in the FDA adverse event website of unusual lung disease in adults with rheumatoid arthritis and adult-onset Still's disease treated with IL-1/IL-6 inhibitors.

However, these anti-cytokine agents also are used in the treatment of autoinflammatory diseases, including the hereditary periodic fever syndromes, and there have been no reports to date of the development of an analogous form of lung disease in patients with these conditions, Ombrello noted.

"Therefore, if anti-cytokine treatment contributes to the development of sJIA lung disease, it seems to do so in a way that is specific to sJIA pathophysiology (or the pathophysiology of a subset of sJIA)," Ombrello told MedPage Today.

Mellins' group noted that their analysis was limited by the retrospective data and may be influenced by channel bias, and emphasized that a causal relationship between IL-1/IL-6 targeting and lung disease cannot be presumed.

The Cincinnati 18

"We first started seeing this in 2010," said Alexei A. Grom, MD, of Cincinnati Children's Hospital Medical Center, and through 2018, the center has treated 74 patients with systemic JIA, which has long been a referral center for sJIA.

He and his colleagues published the 18-patient study from their center, in which lung-disease cases were matched with 36 sJIA patients who had no apparent pulmonary complications. Those with the lung disease had predominantly systemic symptoms, were younger at the time sJIA was diagnosed, often in the first year or two of life (OR 6.5), had previous episodes of MAS (80% vs 20%, OR 14.5), and had very high serum levels of IL-18, at 27,612 versus 5,413 pg/mL, his group reported in Arthritis & Rheumatology.

Respiratory symptoms such as tachypnea were subtle and sometimes even absent at presentation. Digital clubbing was common, and affected patients had similar pain and disability as controls. During approximately 1 year of observation, 14 of the patients remained stable or improved, and none died.

A "striking" number of adverse events to cytokine-targeted biologic agents was observed, particularly with tocilizumab. All but one of the 18 patients had been exposed to at least one biologic agent.

CT scan findings in this group of patients included pleural, septal, and peribronchovascular thickening, "ground glass" or "tree-in-bud" opacities, and lymphadenopathy.

In eight of the patients, open lung biopsies were done and all showed distinctive histopathologic findings characteristic of PAP/ELP, with foamy alveolar macrophages, eosinophilic proteinaceous material, and lipid accumulation. There also was a CD4+ T cell infiltrate, fibrosis, and vasculopathy. However, despite some similarities with what is seen in primary PAP, which results from the inability of alveolar macrophage to clear the protein and lipids, the number and appearance of the macrophages and degree of inflammation were more prominent than in autoimmune or hereditary PAP.

"Our data suggest that although PAP-like features are indicative of macrophage dysfunction, it is not likely to be caused by intrinsic macrophage defects as in primary PAP. Alternatively, alveolar macrophage function may be altered by the external inflammatory milieu of sJIA and MAS, promoting the deviation of macrophage differentiation away from the phenotype required for surfactant recycling," Grom and colleagues wrote.

"The lung disease that is described in these patients is most similar to PAP, but the pathologic and radiographic features are heterogeneous," said Ombrello, who was not involved in the studies.

"Nonetheless, the lung disease observed in these sJIA cohorts seems to be distinct from the other forms of lung disease that can develop in the context of inflammatory or autoimmune diseases, as well as distinct from the primary conditions that lead to PAP," Ombrello said.

Because the histologic picture suggested that macrophage dysfunction could be implicated in the lung disease, Grom's group analyzed bronchoalveolar lavage fluid, and found significant increases in IL-18 levels, a cytokine that contributes to the pathophysiology of MAS, possibly through induction of interferon-γ. Moreover, gene expression analysis found substantial upregulation in genes such as CXCL10 and CXCL9 that are associated with interferon-γ and MAS, further supporting a possible link between interferon-γ and the lung disease.

In his editorial commenting on this study, Nigrovic stated, "The fact that most sJIA lung disease patients have a history of MAS and high IL-18 suggests that they are particularly prone to IFN-γ excess, and thus potentially to IFN-γ-mediated activation of alveolar macrophages."

Biologics to blame?

Grom, however, does not believe that biologic exposure is to blame for the new lung disease. "I think we need to be very careful with that [concept]. Certainly it remains a major concern that we may be inducing this disease with biologics, but I've been to China recently, specifically to see very similar patients who have never been exposed to a biologic. They did not do biopsies but the patients I saw looked very similar and their chest CTs looked similar as well," he told MedPage Today.

"Also, in December I was in Israel for a rheumatology society meeting with other pediatric rheumatologists. I showed them images from our patients and asked if they were seeing this in Israel, and there had not been a single case. They have a similar approach to management as we do although they use more steroids, and at least half of them had been trained in the States," he said.

In addition, other areas of Europe where biologics are commonly used, the lung disease remains very rare. In a very recent series of 42 sJIA patients from Utrecht whose median age was 7.1 years and who were treated with first-line anakinra, only one developed pulmonary disease, and that occurred during a fatal episode of MAS.

"These observations underscore the critical need for well designed multicenter epidemiologic studies including different geographic areas with different patterns of medication utilization to define risk factors and support strategies to prevent development of this disorder," Grom's team concluded.

"It's really important to underscore how powerfully useful these drugs are in sJIA, and the number of children affected by these lung manifestations is very small," Mellins emphasized to MedPage Today.

"It is premature to make treatment recommendations solely on the basis of our findings," she and her colleagues wrote, stressing that management should be individualized and that urgent efforts should focus on improved detection and prevention.

Ombrello made a similar point. "These studies have not demonstrated any evidence that there is a causal relationship between sJIA lung disease and anti-cytokine therapies in sJIA," he said. "As a result, these studies should not change the therapeutic approach to sJIA."

As for treatment of the lung complications, Mellins said that is still a work in progress: some children have resolved their lung disease, but the numbers are too few to assess any treatment strategies. "But clinicians should pay attention to red flags, such as young age at sJIA onset, lymphopenia, and the loss of response to therapy," she said. In her cohort, some children had been well controlled but flared and then lung disease was detected.

"But we don't know which is the cart and which is the horse -- whether developing lung disease stimulates systemic inflammation or whether systemic inflammation is a serious risk factor for lung disease," she said.

Prophylaxis for pneumocystis pneumonia is recommended, she noted.

"Another difficulty is that our ability to diagnose this condition is not great," Grom told MedPage Today. "We pick it up late, when there are already irreversible fibrotic changes in the lung. There is a dissociation between the relatively mild clinical presentation and the severity of the histopathologic findings or massive lung disease on chest CT," he said.

He also argued that patients with the clinical pattern his group described -- early onset, recurrent MAS, and very high IL-18 -- should be carefully monitored for signs of lung disease and have a low threshold for obtaining a chest CT.

"Given the surprisingly high mortality of sJIA lung disease reported in these studies, it is critically important to incorporate screening tests for pulmonary disease into the routine management of children with sJIA," Ombrello concluded.

Comment