Abstract
Objective: To describe high-dose biologic use when treating juvenile idiopathic arthritis (JIA). Methods: Patients with JIA enrolled in the Childhood Arthritis and Rheumatology Research Alliance Registry and treated with a biologic after enrollment were eligible. We described the frequency of high-dose biologic use and characteristics of patients receiving high-dose biologics. We used regression modeling to compare 6-month outcomes (using disease activity measures) between those who increased their biologic from standard to high dose (high-dose group) to those who initiated and remained on standard dosing (no-change group), and to those who switched biologic agents (biologic-switch group). We also compared serious adverse events (SAEs) between groups. Results: A total of 5,352 patients with JIA were treated with biologics following enrollment; 1,080 (20%) had ever received a high-dose biologic. There were no significant differences in outcomes between the high-dose group and the biologic-switch group; both improved disease activity measures, including the clinical Juvenile Arthritis Disease Activity Score 10 (–3.53 and –3.95, respectively; P = 0.68). Although the SAE rates in the high-dose group and the biologic-switch group were numerically higher than the no-change group, the event rates were similar, and neither rate was significantly higher than in the no-change group (unadjusted incident rate ratio 2.5 [95% confidence interval (95% CI) 0.7–8.5] and 1.8 [95% CI 0.7–4.6], respectively). Conclusion: Dosing escalation appears to be a reasonable choice to improve disease control, but large, prospective, randomized studies evaluating specific biologic agents are needed.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 1770-1779 |
| Number of pages | 10 |
| Journal | Arthritis Care and Research |
| Volume | 74 |
| Issue number | 11 |
| DOIs | |
| State | Published - Nov 2022 |
Bibliographical note
Funding Information:Supported by the Childhood Arthritis and Rheumatology Research Alliance (CARRA). Dr. Correll's work was supported by CARRA (salary support).
Funding Information:
There were 5,352 patients with JIA in the CARRA registry who received a biologic during their enrollment. Among them, 1,080 (20.1%) had ever received a high-dose biologic (Table 1). Of 506 patients treated with infliximab, 46.6% had been treated with a high dose. Other commonly used biologics that had the highest proportion of patients ever receiving a high dose included adalimumab (20.8%), anakinra (18.4%), and tocilizumab (15.6%). High-dose biologic use varied by >12-fold across CARRA registry sites. Of the 50 sites evaluated, the median proportion of patients receiving high-dose biologics was 17.2%, with a range from 2.6% to 33.2% (Figure 1). CARRA = Childhood Arthritis and Rheumatology Research Alliance; TNF = tumor necrosis factor. High-dose was defined as 40% higher than the upper limits of the standard dose. There were 5,352 patients with JIA in the CARRA registry who received a biologic during their enrollment. Among them, 1,080 (20.1%) had ever received a high-dose biologic (Table 1). Of 506 patients treated with infliximab, 46.6% had been treated with a high dose. Other commonly used biologics that had the highest proportion of patients ever receiving a high dose included adalimumab (20.8%), anakinra (18.4%), and tocilizumab (15.6%). High-dose biologic use varied by >12-fold across CARRA registry sites. Of the 50 sites evaluated, the median proportion of patients receiving high-dose biologics was 17.2%, with a range from 2.6% to 33.2% (Figure 1). CARRA = Childhood Arthritis and Rheumatology Research Alliance; TNF = tumor necrosis factor. High-dose was defined as 40% higher than the upper limits of the standard dose. There were 1,410 patients with JIA who initiated first biologic therapy after enrollment into the CARRA registry and who had complete dosing and follow-up data. Upon initiation of the biologic therapy, most (n = 1,229, 87.1%) started on standard-dose, while 145 (10.3%) started on high-dose. Twenty-five (2.0%) were initiated on a low-dose biologic (<40% of the labeled dose) but were not included in this study. Table 2 compares characteristics of patients at the initiation of standard-dose biologic therapy to high-dose therapy. Etanercept and adalimumab were the most commonly initiated high-dose biologics. Compared to patients started on the standard dose, those who were started on high-dose biologics were younger at JIA diagnosis and upon start of biologic treatment. There were no differences in disease activity measures between these groups. There were also no differences in JIA category, uveitis status, antinuclear antibody (ANA) or HLA–B27 positivity, clinical sacroiliitis, patient-reported outcomes, or concurrent disease-modifying antirheumatic drug (DMARD) usage. Values are the median (interquartile range) unless indicated otherwise. cJADAS-10 = clinical Juvenile Arthritis Disease Activity Score; JIA = juvenile idiopathic arthritis. Parental education, household income, JIA category, sacroiliitis, uveitis comorbidity, antinuclear antibody positivity, HLA–B27 positivity, use of methotrexate or other nonbiologic disease-modifying antirheumatic drugs within 30 days of initiation of the biologic, and patient-reported outcomes for pain were not significant and were not included in this table. As shown in Table 3, of those who initiated standard biologic dosing after enrollment in the registry, 574 of 827 (69%) continued on standard dosing (no-change group), 63 (7.6%) increased to a high dose (high-dose group), 162 (19.6%) switched to another biologic (biologic-switch group), and 28 (3.4%) decreased to a low dose (not further analyzed). In the high-dose group, most received adalimumab (68.3%), whereas among the biologic-switch group, most were taking etanercept (71.0%), although adalimumab and etanercept were the most commonly used biologics in all groups. The high-dose group tended to be younger at diagnosis (median age 8.0 years) and at the start of a biologic (median age 10.8 years). Those who switched to a new biologic were generally older at diagnosis (median age 11.6 years) and at the start of biologic therapy (median age 13.0). Race, sex, household income, JIA category, nonbiologic DMARD usage within 30 days of biologic initiation, uveitis (ever), and ANA and HLA–B27 positivity were not different between dosing categories. Patients in the high-dose and biologic-switch groups had higher disease activity scores compared to those with no change to their biologic. Patient-reported outcomes, including pain intensity, pain interference, mobility, and global health T scores were better in the no-change group compared to the other groups, although notably, a substantial amount of patient-reported outcomes data was missing. Values are the median (interquartile range) unless indicated otherwise. Missing values are not included in the denominator for percentages. cJADAS-10 = clinical Juvenile Arthritis Disease Activity Score; DMARD = disease-modifying antirheumatic drug; JIA = juvenile idiopathic arthritis; PROMIS = Patient-Reported Outcomes Measurement Information System; RF = rheumatoid factor. Race, sex, household income, antinuclear antibody positivity, HLA–B27, methotrexate and other DMARD use, PROMIS upper-extremity T score, and inadequate disease control were not statistically significant and were not included in this table. Uveitis at the visit closest to initiation of biologic. A higher T score indicates a worse outcome for pain measures, and a lower T score indicates a worse outcome for mobility and global health. There were no significant differences in outcomes between the high-dose group and the biologic-switch group. The sample sizes for these comparisons were small, but the numerical changes in outcomes were similar between these groups in most measures (Table 4). Disease activity scores including cJADAS-10, physician- and parent/patient global assessments, and active joint counts improved in both groups. Patient-reported outcomes including pain intensity and functional mobility also improved in both groups. Although not statistically significant, there was a relatively large difference in the change in PROMIS physical function upper-extremity T score between the high-dose and biologic-switch groups (4.86 versus 0.22, adjusted P = 0.447). –3.53 ± 5.12 (n = 18) –3.95 ± 7.16 (n = 74) –3.78 ± 5.31 (n = 16) –3.28 ± 7.72 (n = 16) 6 (33.3) (n = 18) 23 (31.1) (n = 74) 6 (37.5) (n = 16) 6 (37.50) (n = 16) –1.15 ± 1.58 (n = 24) –1.50 ± 2.16 (n = 96) –1.12 ± 1.65 (n = 21) –1.43 ± 2.61 (n = 21) –1.89 ± 2.55 (n = 27) –1.56 ± 5.59 (n = 104) –2.04 ± 2.57 (n = 23) –0.13 ± 1.60 (n = 23) –0.47 ± 2.22 (n = 19) –0.49 ± 2.72 (n = 81) –0.59 ± 2.32 (n = 17) –1.59 ± 3.04 (n = 17) –0.38 ± 2.83 (n = 16) –1.03 ± 2.79 (n = 59) –0.07 ± 2.73 (n = 14) –0.43 ± 1.16 (n = 14) 2.43 ± 6.18 (n = 10) –0.28 ± 8.67 (n = 40) 2.43 ± 6.18 (n = 10) 0.65 ± 9.32 (n = 10) 2.97 ± 11.29 (n = 15) 2.79 ± 10.47 (n = 50) 2.48 ± 11.84 (n = 13) –0.48 ± 10.09 (n = 13) 4.86 ± 18.21 (n = 7) 0.22 ± 10.15 (n = 20) 4.86 ± 18.21 (n = 7) 0.96 ± 13.65 (n = 7) –1.94 ± 5.31 (n = 14) 1.08 ± 5.57 (n = 57) –2.70 ± 4.71 (n = 12) 2.87 ± 6.79 (n = 12) Values are the mean ± SD unless indicated otherwise. For clinical Juvenile Arthritis Disease Activity Score (cJADAS-10) ≤1 outcome, frequency (%) is presented and logistic regression is used to compare groups. PROMIS = Patient-Reported Outcomes Measurement Information System. Adjusted for age at visit, sex, age at diagnosis, juvenile idiopathic arthritis category, race, and outcome assessment at visit. A higher T score indicates a worse outcome for pain measures, and a lower T score indicates a worse outcome for mobility and global health. When we performed propensity score 1:1 matching to compare 6-month outcomes between those who increased medication dosage to the high-dose group and the no-change group, the sample sizes were again small. As shown in Table 4, no statistically significant differences were found except that the mean active joint count decreased more in the high-dose group compared to the matched patients in the no-change group, and the PROMIS global health assessment score worsened in the high-dose group and improved in the no-change group. Other measured outcomes including the cJADAS-10, physician global assessment, patient/parent global assessment, and other patient-reported outcomes were not significantly different between groups. There were 1,138 patients included in the discontinuation and SAE analysis. The medication discontinuation rate per 100-patient years was higher for those who increased to a high-dose biologic (33.9) and for those who switched biologics (43.6) compared to those who had no change to their standard dose (19.2) (Table 5). Overall, SAEs were uncommon in all treatment groups. Although the SAE rates in the high-dose biologic group and the biologic-switch group were numerically higher than the no-change group, the event rates were at a similar level, and neither rate was significantly higher than the in no-change group (unadjusted incidence rate ratio 2.5 [95% CI 0.7–8.5] and 1.8 [95% CI 0.7–4.6], respectively). Event rates are presented per 100 patient-years. 95% CI = 95% confidence interval; IQR = interquartile range; Ref. = reference. There were 5,352 patients with JIA in the CARRA registry who received a biologic during their enrollment. Among them, 1,080 (20.1%) had ever received a high-dose biologic (Table 1). Of 506 patients treated with infliximab, 46.6% had been treated with a high dose. Other commonly used biologics that had the highest proportion of patients ever receiving a high dose included adalimumab (20.8%), anakinra (18.4%), and tocilizumab (15.6%). High-dose biologic use varied by >12-fold across CARRA registry sites. Of the 50 sites evaluated, the median proportion of patients receiving high-dose biologics was 17.2%, with a range from 2.6% to 33.2% (Figure 1). CARRA = Childhood Arthritis and Rheumatology Research Alliance; TNF = tumor necrosis factor. High-dose was defined as 40% higher than the upper limits of the standard dose. There were 1,410 patients with JIA who initiated first biologic therapy after enrollment into the CARRA registry and who had complete dosing and follow-up data. Upon initiation of the biologic therapy, most (n = 1,229, 87.1%) started on standard-dose, while 145 (10.3%) started on high-dose. Twenty-five (2.0%) were initiated on a low-dose biologic (<40% of the labeled dose) but were not included in this study. Table 2 compares characteristics of patients at the initiation of standard-dose biologic therapy to high-dose therapy. Etanercept and adalimumab were the most commonly initiated high-dose biologics. Compared to patients started on the standard dose, those who were started on high-dose biologics were younger at JIA diagnosis and upon start of biologic treatment. There were no differences in disease activity measures between these groups. There were also no differences in JIA category, uveitis status, antinuclear antibody (ANA) or HLA–B27 positivity, clinical sacroiliitis, patient-reported outcomes, or concurrent disease-modifying antirheumatic drug (DMARD) usage. Values are the median (interquartile range) unless indicated otherwise. cJADAS-10 = clinical Juvenile Arthritis Disease Activity Score; JIA = juvenile idiopathic arthritis. Parental education, household income, JIA category, sacroiliitis, uveitis comorbidity, antinuclear antibody positivity, HLA–B27 positivity, use of methotrexate or other nonbiologic disease-modifying antirheumatic drugs within 30 days of initiation of the biologic, and patient-reported outcomes for pain were not significant and were not included in this table. As shown in Table 3, of those who initiated standard biologic dosing after enrollment in the registry, 574 of 827 (69%) continued on standard dosing (no-change group), 63 (7.6%) increased to a high dose (high-dose group), 162 (19.6%) switched to another biologic (biologic-switch group), and 28 (3.4%) decreased to a low dose (not further analyzed). In the high-dose group, most received adalimumab (68.3%), whereas among the biologic-switch group, most were taking etanercept (71.0%), although adalimumab and etanercept were the most commonly used biologics in all groups. The high-dose group tended to be younger at diagnosis (median age 8.0 years) and at the start of a biologic (median age 10.8 years). Those who switched to a new biologic were generally older at diagnosis (median age 11.6 years) and at the start of biologic therapy (median age 13.0). Race, sex, household income, JIA category, nonbiologic DMARD usage within 30 days of biologic initiation, uveitis (ever), and ANA and HLA–B27 positivity were not different between dosing categories. Patients in the high-dose and biologic-switch groups had higher disease activity scores compared to those with no change to their biologic. Patient-reported outcomes, including pain intensity, pain interference, mobility, and global health T scores were better in the no-change group compared to the other groups, although notably, a substantial amount of patient-reported outcomes data was missing. Values are the median (interquartile range) unless indicated otherwise. Missing values are not included in the denominator for percentages. cJADAS-10 = clinical Juvenile Arthritis Disease Activity Score; DMARD = disease-modifying antirheumatic drug; JIA = juvenile idiopathic arthritis; PROMIS = Patient-Reported Outcomes Measurement Information System; RF = rheumatoid factor. Race, sex, household income, antinuclear antibody positivity, HLA–B27, methotrexate and other DMARD use, PROMIS upper-extremity T score, and inadequate disease control were not statistically significant and were not included in this table. Uveitis at the visit closest to initiation of biologic. A higher T score indicates a worse outcome for pain measures, and a lower T score indicates a worse outcome for mobility and global health. There were no significant differences in outcomes between the high-dose group and the biologic-switch group. The sample sizes for these comparisons were small, but the numerical changes in outcomes were similar between these groups in most measures (Table 4). Disease activity scores including cJADAS-10, physician- and parent/patient global assessments, and active joint counts improved in both groups. Patient-reported outcomes including pain intensity and functional mobility also improved in both groups. Although not statistically significant, there was a relatively large difference in the change in PROMIS physical function upper-extremity T score between the high-dose and biologic-switch groups (4.86 versus 0.22, adjusted P = 0.447). –3.53 ± 5.12 (n = 18) –3.95 ± 7.16 (n = 74) –3.78 ± 5.31 (n = 16) –3.28 ± 7.72 (n = 16) 6 (33.3) (n = 18) 23 (31.1) (n = 74) 6 (37.5) (n = 16) 6 (37.50) (n = 16) –1.15 ± 1.58 (n = 24) –1.50 ± 2.16 (n = 96) –1.12 ± 1.65 (n = 21) –1.43 ± 2.61 (n = 21) –1.89 ± 2.55 (n = 27) –1.56 ± 5.59 (n = 104) –2.04 ± 2.57 (n = 23) –0.13 ± 1.60 (n = 23) –0.47 ± 2.22 (n = 19) –0.49 ± 2.72 (n = 81) –0.59 ± 2.32 (n = 17) –1.59 ± 3.04 (n = 17) –0.38 ± 2.83 (n = 16) –1.03 ± 2.79 (n = 59) –0.07 ± 2.73 (n = 14) –0.43 ± 1.16 (n = 14) 2.43 ± 6.18 (n = 10) –0.28 ± 8.67 (n = 40) 2.43 ± 6.18 (n = 10) 0.65 ± 9.32 (n = 10) 2.97 ± 11.29 (n = 15) 2.79 ± 10.47 (n = 50) 2.48 ± 11.84 (n = 13) –0.48 ± 10.09 (n = 13) 4.86 ± 18.21 (n = 7) 0.22 ± 10.15 (n = 20) 4.86 ± 18.21 (n = 7) 0.96 ± 13.65 (n = 7) –1.94 ± 5.31 (n = 14) 1.08 ± 5.57 (n = 57) –2.70 ± 4.71 (n = 12) 2.87 ± 6.79 (n = 12) Values are the mean ± SD unless indicated otherwise. For clinical Juvenile Arthritis Disease Activity Score (cJADAS-10) ≤1 outcome, frequency (%) is presented and logistic regression is used to compare groups. PROMIS = Patient-Reported Outcomes Measurement Information System. Adjusted for age at visit, sex, age at diagnosis, juvenile idiopathic arthritis category, race, and outcome assessment at visit. A higher T score indicates a worse outcome for pain measures, and a lower T score indicates a worse outcome for mobility and global health. When we performed propensity score 1:1 matching to compare 6-month outcomes between those who increased medication dosage to the high-dose group and the no-change group, the sample sizes were again small. As shown in Table 4, no statistically significant differences were found except that the mean active joint count decreased more in the high-dose group compared to the matched patients in the no-change group, and the PROMIS global health assessment score worsened in the high-dose group and improved in the no-change group. Other measured outcomes including the cJADAS-10, physician global assessment, patient/parent global assessment, and other patient-reported outcomes were not significantly different between groups. There were 1,138 patients included in the discontinuation and SAE analysis. The medication discontinuation rate per 100-patient years was higher for those who increased to a high-dose biologic (33.9) and for those who switched biologics (43.6) compared to those who had no change to their standard dose (19.2) (Table 5). Overall, SAEs were uncommon in all treatment groups. Although the SAE rates in the high-dose biologic group and the biologic-switch group were numerically higher than the no-change group, the event rates were at a similar level, and neither rate was significantly higher than the in no-change group (unadjusted incidence rate ratio 2.5 [95% CI 0.7–8.5] and 1.8 [95% CI 0.7–4.6], respectively). Event rates are presented per 100 patient-years. 95% CI = 95% confidence interval; IQR = interquartile range; Ref. = reference. Biologic agents are integral to the treatment of juvenile idiopathic arthritis (JIA) and associated uveitis (1,2). Pediatric rheumatologists may increase the dosage of biologics beyond the standard ranges to achieve better disease control. Recent treatment guidelines for JIA-associated uveitis recommend increasing monoclonal antibody tumor necrosis factor inhibitors (TNFis) beyond the standard dose prior to changing biologics in inadequately controlled chronic anterior uveitis (3). Although high-dose regimens of biologics are widely discussed in the treatment of JIA and uveitis, there have been few published studies demonstrating the efficacy or safety of high-dose biologic use (4–9). These published studies have been small, single-center retrospective studies or case reports, and each evaluated a single biologic. This study describes the use of high-dose biologics in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) registry and assesses their effectiveness and safety in the treatment of JIA. SIGNIFICANCE & INNOVATIONS Anecdotally, pediatric rheumatologists are known to prescribe high-dose biologics to treat inadequately controlled juvenile idiopathic arthritis (JIA), but few published data exist on the safety and efficacy of this practice. Our study found that within the Childhood Arthritis and Rheumatology Research Alliance registry, 20% of children with JIA received high-dose biologics, although the use of high-dose biologics varied widely by site. Patients with JIA whose biologic dose escalated from standard to high dose showed improved disease activity scores, similar to patients who switched from one standard-dose biologic to another; therefore, increasing biologics to a high dose may be a reasonable option for patients not adequately controlled on standard-dose therapy. There was no significant increase in serious adverse events in patients receiving high-dose biologics compared to those receiving standard doses. Anecdotally, pediatric rheumatologists are known to prescribe high-dose biologics to treat inadequately controlled juvenile idiopathic arthritis (JIA), but few published data exist on the safety and efficacy of this practice. Our study found that within the Childhood Arthritis and Rheumatology Research Alliance registry, 20% of children with JIA received high-dose biologics, although the use of high-dose biologics varied widely by site. Patients with JIA whose biologic dose escalated from standard to high dose showed improved disease activity scores, similar to patients who switched from one standard-dose biologic to another; therefore, increasing biologics to a high dose may be a reasonable option for patients not adequately controlled on standard-dose therapy. There was no significant increase in serious adverse events in patients receiving high-dose biologics compared to those receiving standard doses. Patients with JIA enrolled in the Childhood Arthritis and Rheumatology Research Alliance Registry and treated with a biologic after enrollment were eligible. We described the frequency of high-dose biologic use and characteristics of patients receiving high-dose biologics. We used regression modeling to compare 6-month outcomes (using disease activity measures) between those who increased their biologic from standard to high dose (high-dose group) to those who initiated and remained on standard dosing (no-change group), and to those who switched biologic agents (biologic-switch group). We also compared serious adverse events (SAEs) between groups.
Publisher Copyright:
© 2021 American College of Rheumatology.
PubMed: MeSH publication types
- Journal Article
- Research Support, Non-U.S. Gov't