Children with chronic nonbacterial osteomyelitis (CNO) who had multifocal disease at onset, symmetric bone lesions, or multiple affected body regions were more likely to need second-line treatment than patients without these features, according to findings presented at the annual meeting of the Childhood Arthritis and Rheumatology Research Alliance (CARRA).

CNO is an autoinflammatory condition that results in sterile inflammatory bone lesions and most commonly affects the long bones of people who are skeletally immature. After a first-line treatment of nonsteroidal anti-inflammatory drugs (NSAIDs), second-line treatments per CARRA guidelines typically include methotrexate or sulfasalazine, tumor necrosis factor (TNF)–alpha inhibitors, and bisphosphonates.

“Since it’s common for there to be long delays before diagnosis of CNO, it is important to start an effective treatment promptly,” Katherine D. Nowicki, MD, of Children’s Hospital Colorado, Aurora, Colorado, told attendees. “While we have guidance on which treatments to use, it remains unclear which patients are most likely to respond to NSAIDs and which patients will require a second-line treatment.”

Melissa Oliver, MD, MS, an assistant professor of clinical pediatrics in rheumatology at Riley Children’s Health at Indiana University Health, Indianapolis, was not involved in the research but said the findings of this study are helpful in “counseling families and patients at that initial visit and having a lower threshold to start a second-line agent if NSAID monotherapy is not working well.”

There are no clinical trials on patients with CNO, Oliver said, so very little data exist for guiding clinicians on the best therapy to use and how long to keep patients on therapy while minimizing risk for flare when coming off therapy.

A key clinical takeaway for clinicians is being able to tell patients with unifocal disease that they may not need to be on NSAIDs for a long period and can still do well, Oliver said. For patients with multifocal disease with symmetric bone lesions or multiple regions involved with CNO, “pediatric rheumatologists should have a lower threshold to start a second-line therapy for these patients,” she said.

To better understand how different clinical characteristics predict treatment needs, the researchers conducted a retrospective chart review of 234 patients who received a CNO diagnosis before age 18 and who established care in the Children’s Hospital Colorado’s CNO multidisciplinary clinic between January 2005 and July 2021. After excluding 70 patients, primarily due to inadequate follow-up for assessing treatment response, the researchers included 164 patients whose records they reviewed through January 2022.

The researchers assessed five aspects of disease involvement: Unifocal or multifocal at diagnosis, ever having presence of symmetric bone lesions, regions ever affected by CNO, complications, and disease activity at most recent follow-up. They compared these factors to the start and stop date of each CNO medication, the patient’s treatment response, and the date and reason for discontinuation of treatments.

Among the 164 patients in the study, 32 had a short course of NSAIDs (3-7 months), 62 had a long course of NSAIDs (7 or more months), and 70 received second-line treatment.

Their topline findings revealed that patients with unifocal disease at diagnosis required 47% fewer total days of NSAID monotherapy treatment than those with multifocal disease at diagnosis, Nowicki told attendees. Having symmetric bone lesions increased the likelihood of needing a second-line therapy by 6.86 times compared to those without symmetric bone lesions, and for each additional region affected by CNO, the odds of needing a second-line therapy increased by a factor of 1.94, she said.

There were no significant differences in patient ages or sex or in mean interval from symptom onset to treatment onset across treatment groups. However, patients who received second-line treatment did have a significantly longer average time from symptom onset to diagnosis (324 days) than those who had a short course (119 days) or long course (270 days) of NSAIDs (P = .023). Mean follow-up was also significantly longer for patients with second-line treatment (3.8 years) or long-course NSAIDs (2.7 years) than for those with short-course NSAIDs (1.2 years; P < .001).

Mean erythrocyte sedimentation rate or C-reactive protein did not differ across treatment groups nor did presence of a CNO lesion on x-rays at presentation. But significantly more patients in the second-line group had a biopsy (94%) than in the long-course (74%) or short-course (69%) NSAID groups (P = .0025). They were also more likely to have one or more whole-body MRIs. Most of the patients on short-course (88%) and long-course (82%) NSAIDs did not undergo a whole-body MRI, whereas most patients (59%) on a second-line treatment underwent at least one and 24% underwent three or more MRIs (P < .001).

More patients on short-course NSAIDs had unifocal disease at diagnosis (72%) than those on long-course NSAIDs (47%) or a second-line treatment (41%; P = .015). Patients on a second-line treatment were also more likely to have symmetric involvement in the same bone (73% vs 16% short-course and 23% long-course NSAIDs) and to have more regions of the body affected (P < .001).

There were also significant differences in mean days on NSAID monotherapy and number of NSAIDs trialed. Patients on a second-line treatment had a mean 441 days of NSAID monotherapy compared to 175 days for patients on short-course NSAIDs and 725 for patients on long-course NSAIDs (P < .001). Nearly all the short-course patients (94%) trialed a single NSAID, while more than half the long-course and second-line patients trialed two or more (P < .001).

None of the patients on short-course NSAIDs had complications. More patients on second-line treatments had vertebral height loss (20%) or amplified pain (14%) than long-course patients (13% and 5%, respectively; P = .02).

At the study’s end date, nearly all the patients on short-course NSAIDs were in remission (94%) compared with 71% of patients on long-course NSAIDs and only half of patients (51%) on the second-line treatment (P < .001). None of the patients on short-course NSAIDs had active disease compared with 11% of patients on long-course NSAIDs and 20% of patients on second-line treatments (P = .02).

This study included the largest single-center cohort of patients with CNO in North America, all treated at a multidisciplinary clinic with a protocolized treatment approach, but it remains limited by its retrospective nature and the missing data for 70 patients, Nowicki said. She also noted that whole-body MRI was not systematically performed on all patients, so it was possible patients without an MRI had undetected asymptomatic lesions.

Despite these limitations, Oliver said retrospective studies like these can help pediatric rheumatologists get an idea of reasonable therapies to start, how long to keep patients on them, and when to escalate to the next step.

“I hope one day our CNO research will be able to tell us about which is the optimal second-line therapy for patients, such as bisphosphonates vs TNF inhibitors vs DMARDs [disease-modifying antirheumatic drugs],” Oliver said.

Nowicki and Oliver reported no disclosures. Information on study funding was not provided.

Write A Comment