Abstract
Objective Antimelanoma differentiation–associated gene 5 (anti-MDA5)–positive dermatomyositis with interstitial lung disease (DM-ILD) progresses rapidly and has a poor prognosis. Previously, we reported the efficacy of a combination therapy comprising high-dose glucocorticoids (GCs), calcineurin inhibitors (CNIs), and intravenous cyclophosphamide (IV CYC) in a multicenter clinical trial (UMIN000014344). In the present study, we evaluated the long-term outcomes and effects of induction therapy on the maintenance of remission.
Methods All participants from our previous trial were followed up for > 5 years. Seventy-three other patients with anti-MDA5–positive DM-ILD from our institute were retrospectively integrated into the previous trial for further analysis. Sixty-eight patients achieved remission and survived for > 6 months. Based on the induction treatment, we classified the patients into 2 groups: (1) group T (n = 56), with triple combination therapy (GCs, CNIs, and IV CYC), and (2) group C (n = 12), with monotherapy/dual therapy. The recurrence-free and drug-withdrawal rates of immunosuppressive agents were compared.
Results The overall survival and recurrence-free survival rates at 5 years were 100% for the participants in the previous trial. The 5-year cumulative withdrawal rates for CNIs and GCs were 70% and 53%, respectively. In a comprehensive analysis, the recurrence-free rates in group T were higher than those in group C (90% vs 56%; P < 0.05). The drug-withdrawal rates of CNIs and GCs at 10 years in group T were also higher than those in group C (79% vs 0% and 43% vs 0%, respectively; P < 0.05).
Conclusion Triple combination therapy in the induction phase can reduce the risk of recurrence and facilitate drug withdrawal in anti-MDA5–positive DM-ILD.
Clinically amyopathic dermatomyositis (CADM) is a disease with the typical skin manifestations of dermatomyositis (DM) and few or no features of myopathy.1,2 The antimelanoma differentiation–associated gene 5 (anti-MDA5) antibody is strongly associated with CADM.3,4 Interstitial lung disease (ILD) accompanied by anti-MDA5–positive DM/CADM is often rapidly progressive and associated with a poor prognosis, particularly in Japanese patients.5,6 In the past, approximately 50% of patients died from ILD exacerbation within 6 months of presentation.7 The mortality seemed to be a result of the insufficiency of immunosuppressants, such as the administration of only glucocorticoids (GCs) or GCs and calcineurin inhibitors (CNIs).
Previously, we reported the efficacy of combined immunosuppressive therapy comprising high-dose GCs, CNIs, and intravenous cyclophosphamide (IV CYC) for anti-MDA5–positive DM/CADM-ILD in a single-arm prospective trial.8 With this triple combination therapy, the 6-month survival rates improved to 89% compared to those with conventional therapy.8 The efficacy of plasma exchange (PE) as an additional therapy for anti-MDA5–positive DM/CADM has also been reported.9 These triple combination therapies and additional PE have improved survival rates and prolonged survival.
In accordance with the improvement in the survival rate associated with induction therapy, more patients that are anti-MDA5–positive can be followed up for a longer time. However, no study, to our knowledge, has evaluated long-term maintenance therapy for anti-MDA5–positive DM/CADM-ILD and the management of patients who are anti-MDA5–positive in remission.
In our study, long-term follow-up data for participants from our previous multicenter, single-arm clinical trial (UMIN000014344)8 were analyzed. Subsequently, by combining the participants of the clinical trial with another cohort, we retrospectively investigated the effect of the induction therapy on long-term outcomes and the changes in maintenance immunosuppressive therapy for anti-MDA5–positive DM/CADM-ILD.
METHODS
Study overview. First, we analyzed the long-term outcomes and maintenance therapy of all participants from our previous trial (UMIN000014344)8 who were followed up for > 5 years. Eligible patients were anti-MDA5–positive and diagnosed as having definite, probable, or possible DM or CADM with ILD, without immunosuppressive treatment before admission to one of the registered hospitals, and who were followed up for > 5 years from initial therapy by March 2022. Patients with the following conditions were excluded: deceased patients who did not achieve remission and patients followed up for < 6 months from initial therapy. DM was diagnosed according to the criteria of Bohan and Peter.10 CADM was diagnosed when patients demonstrated typical cutaneous lesions of DM without clinical evidence of myositis, with minimal or no increase in serum creatine kinase.2,10,11 ILD diagnosis was based on respiratory symptoms, physical examination, high-resolution chest computed tomography (CT) findings, and respiratory function tests. Serum anti-MDA5 antibodies were detected by immunoprecipitation using 35S-labeled HeLa cells and ELISA (MESACUP anti-MDA5 test, MBL Co., Ltd.).4,12 Remission was defined as survival for > 6 months after initial treatment without active exacerbation of ILD.
Second, to investigate the influence of induction therapy on long-term outcomes and management, we retrospectively enrolled 73 consecutive Japanese patients with anti-MDA5–positive DM/CADM-ILD who were aged ≥ 18 years and had visited or been referred to Kyoto University Hospital from October 2001 to March 2022. We combined them with the participants of the clinical trial and analyzed prognoses such as recurrence-free rate or drug-withdrawal rate. Inclusion criteria were as follows: Japanese patients aged ≥ 18 years with anti-MDA5–positive DM/CADM-ILD who had achieved remission and survived without relapse of ILD. The criteria of DM, CADM, ILD, and remission and the measuring method for anti-MDA5 antibodies were as described above. Relapse of ILD was defined as a deterioration in the respiratory function test or new ground-glass opacities on chest CT, assessed by both a radiologist and a rheumatologist, and the requirement for intensified treatment.13
Patients with the following conditions were excluded: deceased patients who could not achieve remission, patients who had received initial therapy primarily at another hospital, patients who had been followed up for < 6 months from initial therapy, patients who had been diagnosed with malignancy at the same time, and patients who had other myositis-specific autoantibodies, such as antiaminoacyl-tRNA synthetases (ARS) antibodies. Eligible patients for combined analysis are shown in Figure 1. We excluded 28 patients, including 17 who were not in remission and died < 6 months after the initial therapy, 6 who received undetermined induction therapy at another hospital, 3 for whom remission could not be confirmed because the follow-up time was < 6 months, 1 who was diagnosed with malignant lymphoma just after remission of lung disease, and 1 who was positive for both anti-ARS and anti-MDA5 antibodies.
Flow scheme for the analysis of patients with anti-MDA5–positive DM-ILD who achieved remission. Group T: patients treated with GCs, CNIs, and IV CYC. Early group T: patients who received initial IV CYC within 7 days of GC administration. Late group T: patients who received initial IV CYC within 3 months of GC administration. Group C: patients who received monotherapy (only GCs) or dual therapy (GCs with CNIs or GCs with IV CYC). anti-MDA5: antimelanoma differentiation–associated gene 5; ARS: aminoacyl-tRNA synthetase; CADM: clinically amyopathic dermatomyositis; CNI: calcineurin inhibitor; DM: dermatomyositis; ILD: interstitial lung disease; GC: glucocorticoid; IV CYC: intravenous cyclophosphamide.
There were 68 adult Japanese patients with anti-MDA5–positive DM/CADM-ILD who achieved remission and survived for > 6 months. These patients were divided into group T, comprising patients treated with triple combination therapy (GCs + CNIs + IV CYC; n = 56) and group C, comprising patients treated with conventional therapy (only GCs, GCs + CNIs, or GCs + IV CYC; n = 12), although the treatment regimen was decided by each attending physician at the time. Further, to investigate the effect of early initial IV CYC, patients in group T were subdivided into early group T and late group T, based on the timing of IV CYC initiation.
All patients provided written informed consent to participate in this study prior to sample collection. The study was conducted in accordance with the Declaration of Helsinki. The study design was approved by the Ethics Committee of the Kyoto University Graduate School and Faculty of Medicine (approval no. R1540).
Triple combination therapy (group T). The triple combination therapy regimen comprised high-dose GCs, CNIs, and IV CYC as remission induction agents. The regimen was initiated soon after or up to 3 months after the diagnosis of DM or CADM with ILD. Prednisolone (PSL) was used as the main GC and initially administered at 1 mg/kg/day for 4 weeks; thereafter, the existing dose was reduced by 10% every 2 weeks. Tacrolimus or cyclosporine A was used as a CNI. Tacrolimus was adjusted to maintain a 12-hour blood trough level of 10-12 ng/mL. IV CYC was initiated at 500 mg/m2 of body surface area (BSA) biweekly, then gradually increased to a maximum dose of 1000 mg/m2 of BSA according to a nadir leukocyte count from baseline. The intended number of IV CYC administrations was 10-15. Patients who received initial IV CYC within 7 days after GC administration were categorized into the early group T (n = 45), whereas patients who received initial IV CYC > 7 days after GC administration were categorized into the late group T (n = 11).
Conventional therapy (group C). Patients in group C received monotherapy (only GCs) or dual therapy (GCs with CNIs or IV CYC) as remission induction therapy. Although methods of using GCs, CNIs, and IV CYC have not been strictly defined, we considered late addition of CNIs to be when patients had CNIs added > 3 months from GC administration. Other immunosuppressants were not strictly restricted at initial therapy.
The choice of treatment regimen was dependent on the attending physicians. In both groups, patients underwent PE as additional therapy when they worsened clinically and required oxygen administration during remission induction therapy.
Endpoints. The primary endpoint was the comparison of the recurrence-free rate at 120 months after the initial therapy between groups T and C. Recurrence-free time was defined as the interval between the initiation of induction therapy for ILD and the exacerbation of ILD or the last follow-up. Secondary endpoints included the drug-free rate of GCs and/or CNIs at 60 or 120 months from initial therapy and the median PSL dose at 36 months after initial induction therapy. We defined drug-free as complete withdrawal of GCs and/or CNIs (temporal withdrawal of drugs was not defined as drug-free). We also collected the clinical data of remission, adjusting for the time from initial therapy and serious adverse events after initial therapy. Serious adverse events were defined as unexpected admissions, such as infection or emergency surgery. We also analyzed the differences in these endpoints among the 3 groups.
Statistics. Recurrence-free and drug-free rates were estimated by Kaplan-Meier analysis. The log-rank test was also used for comparison. The Wilcoxon signed-rank test and Fisher exact test were used to assess the associations of clinical features between the groups. A P value of < 0.05 was considered statistically significant. The P values for multiple comparisons were adjusted using Bonferroni correction. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan) for R (The R Foundation for Statistical Computing).14
RESULTS
Long-term follow-up data from the clinical trial. The overall survival and recurrence-free survival rates of the participants who achieved remission in our previous multicenter, single-arm clinical trial (UMIN000014344)9 are demonstrated in Figure 2A. Twenty-three patients were enrolled and all patients survived without relapse of ILD. The overall survival and recurrence-free survival rates at 60 months after the initial therapy were both 100% (the overall survival rate is not shown in the figure). Drug-withdrawal rates for immunosuppressants in these patients are exhibited in Figure 2B-D. The 60-month cumulative withdrawal rates for CNIs and GCs were 70% and 53%, respectively. The cumulative achievement rate for being drug-free for both CNIs and GCs at 60 months was 38%. The changes in clinical variables from the induction phase to the remission phase are shown in Table 1. The median time of analysis was 2330 days from initial therapy. Respiratory function and Krebs von den Lungen-6 (KL-6) were significantly improved in the remission phase. Disease activity markers, such as serum ferritin, KL-6, and anti-MDA5 antibody titers, improved 12 months after the induction of initial therapy and remained within the normal range during the remission phase (Supplementary Figure S1, available with the online version of this article). All patients enrolled in our previous trial maintained remission and had improved disease activity markers and respiratory functions.
Clinical course of participants from the previous clinical trial. (A) Recurrence-free rate. (B) Drug-free rate for CNIs. (C) Drug-free rate for GCs. (D) Drug-free rate for both CNIs and GCs. Twenty-three patients were enrolled. The cumulative rates were analyzed using the Kaplan-Meier test. CNI: calcineurin inhibitor; GC: glucocorticoid.
Clinical variable changes of participants in our previous clinical trial.
Characteristics of patients for the combined retrospective analysis. A comparison of the clinical characteristics and laboratory tests between the groups is displayed in Table 2. Fifty-six patients were treated with triple combination therapy (group T) and 12 patients with conventional therapy (group C). There were no significant differences in age or sex between the 2 groups. Although the time from disease onset to initial treatment was significantly longer in group C (P < 0.001), the severity of ILD, such as saturation of peripheral oxygen and oxygen therapy at initial therapy, was not different between the 2 groups. Serum ferritin and KL-6 levels were elevated in both groups, with no significant differences between the groups.
Comparison of clinical characteristics at baseline between group T and group C.
The predicted percent of diffusing lung capacity for carbon monoxide (%DLCO) was significantly lower in group C than in group T (P < 0.01). However, the predicted percent of forced vital capacity did not differ between the groups. GCs were used for all patients in both the groups and the initial median dose of PSL was significantly higher in group T (P < 0.01). CNIs were used for all patients in group T, and none of the patients in group C were treated with triple combination therapy due to the protocol. CNIs were started at approximately the same time as the GCs in all patients in group T, which was significantly earlier than group C. Immunosuppressants other than GCs, CNIs, or IV CYC were not used at initial therapy in either of the groups (data not shown). PE was used more frequently for patients in group T than in group C (18% vs 0%), although this difference was not significant. The median follow-up time was not significantly different between the groups.
The clinical characteristics of the 3 groups (early group T, late group T, and group C) are shown in Supplementary Table S1 (available with the online version of this article). Group C had a significantly greater time from disease onset to initial therapy compared with the early and late group T (P < 0.05, with Bonferroni correction). The %DLCO and initial median dose of PSL in group C were significantly lower compared to the early group T (both P < 0.01, with Bonferroni correction). The frequency of tacrolimus use, total dose, and IV CYC count in group C was significantly lower than in both the early and late group T (both P < 0.01, with Bonferroni correction).
Recurrence-free rate of the patients in the combined retrospective analysis. The overall survival rates for group T and group C were both 100% (data not shown). The overall recurrence-free survival rate is shown in Figure 3A. The cumulative recurrence-free rates at 120 months were 90% and 56% in group T and group C, respectively, and these were significantly different (log-rank test, P = 0.02). All patients who required oxygenation during the induction phase in both groups recovered to a stage where supplemental oxygen could be withdrawn. Group T was divided into the early and late groups. The overall recurrence-free rates at 120 months from initial therapy of the early T, late T, and C groups were 92%, 83%, and 56%, respectively (Supplementary Figure S2, available with the online version of this article). Although there were no significant differences (log-rank test, P = 0.06), the early group T demonstrated the highest recurrence-free survival rate among the 3 groups.
Clinical course of participants included in the comprehensive analysis. (A) Comparison of recurrence-free rates between group T (triple combination therapy group, n = 56) and group C (conventional therapy group, n = 12). (B) Drug-free rate for CNIs. (C) Drug-free rate for GCs. (D) Drug-free rate for both CNIs and GCs. The recurrence-free rate and drug-free rates for CNIs and GCs are significantly higher in group T than in group C (P < 0.05, respectively). The cumulative rates were calculated using the Kaplan-Meier method. The log-rank test was used to compare the rates. CNI: calcineurin inhibitor; GC: glucocorticoid.
In terms of clinical characteristics and laboratory tests performed during the remission phase, KL-6 was significantly higher in group C, although respiratory function tests were not significantly different between both groups. However, due to the small number of patients in group C, the statistical significance may not be meaningful (Supplementary Table S2, available with the online version of this article).
Drug-withdrawal rate of immunosuppressants. The cumulative withdrawal rates of immunosuppressants, as well as drug-free rates, are exhibited in Figure 3B-D. The 60-month cumulative withdrawal rates of CNIs were 59% and 0% in group T and group C, respectively, which were significantly different (log-rank test, P < 0.01). Moreover, the 120-month cumulative withdrawal rate of CNIs in group T was 79%. The 60-month cumulative withdrawal rates of GCs were 32% and 0% in group T and group C, respectively, and were significantly different (log-rank test, P < 0.05). The cumulative drug-free rates for GCs at the 120-month point were 43% and 0% in group T and group C, respectively. Although there was no significant difference (P = 0.13), only the patients in group T achieved drug-free (withdrawal of both CNIs and GCs) remission (17% at 60 months and 36% at 120 months).
The cumulative withdrawal rates for CNIs or GCs and drug-free rates for the early T, late T, and C groups are demonstrated in Supplementary Figure S3 (available with the online version of this article). The 60-month cumulative withdrawal rate for CNIs in group T was 68%, which was significantly higher than that in group C (P < 0.001, with Bonferroni correction). The 60-month cumulative withdrawal rate for GCs in the early group T was 38%, which was also significantly higher than that in group C (P < 0.05, with Bonferroni correction). The 60-month and 120-month drug-free rates in the early T and late T groups were 22% and 43%, respectively, which were the highest among the values for the 3 groups. However, the difference was not significant.
The maintenance dose of PSL at 36 months. The PSL dose required 36 months after the initial induction therapy is shown in Figure 4. The median dose of PSL at 36 months in group T was significantly lower than that in group C (3.0 mg/day and 6.0 mg/day, respectively; P < 0.01). The median doses of PSL at 36 months in the early T, late T, and C groups were 3.0 mg/day, 3.5 mg/day, and 6.0 mg/day, respectively (Supplementary Figure S4, available with the online version of this article). The median dose of PSL in the early group T was significantly lower than that in group C (P < 0.05, with Bonferroni correction).
Maintenance dose of PSL at 36 months after induction of initial therapy. The median doses of PSL were 3.0 mg/day and 6.0 mg/day in group T (n = 45) and group C (n = 8), respectively, which are significantly different based on the Mann-Whitney U test (P < 0.01). Data are presented as box plots, where the boxes represent the IQR, the lines within the boxes represent the median values, and the lines outside the boxes represent the minimum and maximum values. PSL: prednisolone.
Details of relapsed patients. There were 2, 1, and 4 patients who experienced ILD relapse in the early T, late T, and C groups, respectively. The detailed clinical course of the patients with lung relapse is shown in Supplementary Table S3 (available with the online version of this article). One patient in the early group T was newly diagnosed with breast cancer several months before the relapse of ILD, and another patient experienced ILD relapse approximately 1 year after tacrolimus discontinuation due to liver dysfunction. Most of the relapsed patients could be reinduced to remission by increasing GCs to moderate or high doses and restarting IV CYC and/or CNI; however, 2 patients in group C experienced a second relapse of skin manifestations or ILD after reinduction therapy.
Adverse events. During the observation period, there were 14 (10 patients) and 6 (3 patients) serious adverse events in group T and group C, respectively. Most of the events were bacterial or viral infections. There was no malignancy, such as leukemia, myelodysplastic syndromes, malignant lymphoma, or bladder cancer, which are thought to be induced by CYC (Supplementary Table S4, available with the online version of this article), in either group. Serious adverse events in the 3 groups (early group T, late group T, and group C) are shown in Supplementary Table S5.
DISCUSSION
We previously reported the importance of early intervention with triple combination therapy for remission induction and 6-month survival of patients with anti-MDA5–positive DM-ILD in a multicenter prospective trial.8 However, there is no reported evidence regarding management during the remission phase.To our knowledge, this is the first study to demonstrate the effects of induction therapy on long-term outcomes or maintenance treatment for anti-MDA5–positive DM/CADM-ILD.
In the present study, the long-term survival and relapse-free rates of patients with anti-MDA5–positive DM-ILD treated with triple combination therapy remained high. Some patients achieved complete drug-free status after the withdrawal of both GCs and CNIs. Disease-monitoring markers of interstitial pneumonia, such as KL-6 and respiratory function tests, demonstrated significant improvement 12 months after the initial treatment, and these improvements were maintained for an extended period of time. These findings suggested that the efficacy of triple combination therapy, including IV CYC, affected not only survival in the remission induction phase but also disease stability with minimal immunosuppressant agents in the maintenance phase. Based on these results, the present study investigated whether the difference in remission induction therapies affects long-term prognosis and stability during the maintenance phase in patients with anti-MDA5–positive DM-ILD who survived the acute phase of their disease.
In order to collect data from and analyze patients with anti-MDA5–positive DM-ILD who had achieved remission, we integrated another retrospective cohort of 73 patients from our institution into the previous clinical trial8 and divided them into 2 groups for retrospective analysis, based on the remission induction treatment received. There were no significant differences between the 2 groups in terms of age, lactate dehydrogenase, C-reactive protein, and ferritin levels, which have been reported as poor prognostic factors.15
However, the time from onset to treatment was significantly longer in group C, and respiratory function tests, such as %DLCO, were also significantly poorer in group C. Group C included more patients who were treated during the previous time period than group T. This suggested that the diagnosis and initiation of treatment had been delayed, and interstitial pneumonia may have progressed because of the lack of established testing methods for anti-MDA5 antibodies. However, after remission, patients in group C demonstrated improvement in respiratory function, which was not significantly different from that of group T (Supplementary Table S2, available with the online version of this article). This suggested that the respiratory function of patients who are anti-MDA5 positive can be improved to some extent after remission induction, even if their ILD is somewhat advanced before treatment. Therefore, it is unlikely that differences in diagnosis delay and respiratory function prior to induction therapy affected relapse or drug-withdrawal rates during long-term observation.
Some patients were treated with PE in group T, although there are conflicting reports of the effectiveness of PE.9,16 PE in this study was only used in the induction phase and the number of patients treated with PE was low. Thus, this was not thought to influence the duration of remission achievement. Considering this, we do not believe that there were any major differences in patient background between the 2 groups in this study, other than differences in treatment protocols. Comparison between the 3 groups after subdividing group T into 2 groups exhibited the same trend.
The relapse-free rate was significantly higher for patients in group T than in group C. There was no significant difference between the rates in the early and late T groups. However, relapses in the early group T included patients who had discontinued immunosuppressive agents due to adverse effects and a patient who had experienced comorbid malignancy during the course of their disease (Supplementary Table S3, available with the online version of this article). In contrast, relapses in the late group T and group C did not include patients with unintended treatment changes or coexisting malignancy. Therefore, early administration of a sufficient amount of CYC might be beneficial for maintaining long-term remission.
The drug-withdrawal rates were also significantly higher in group T than in group C. Although there was no statistical significance, a substantial number of patients in group T achieved complete discontinuation of both GCs and CNIs, whereas none of the patients in group C achieved drug-free remission. The decision to withdraw drugs, such as GCs or immunosuppressive agents, was dependent on the discretion of the attending physician. Considering the medical record information, the attending physician decided to reduce or discontinue the drug based on the titer of anti-MDA5 antibodies, length of the relapse-free period, and presence of clinical symptoms related to dermatomyositis, such as skin, joint, and respiratory symptoms, although there is no predefined protocol for decreasing and discontinuing drugs. Although the PSL dose was significantly higher in group T than in group C during induction therapy, the PSL dose at 36 months was significantly lower in group T. This suggests that the disease activity in group T was stabilized earlier than that in group C, which resulted in a steady reduction in the GC dose. When group T was further divided into 2 groups, patients who were able to discontinue both GCs and CNIs were observed only in the early group T. This could be attributed to the fact that triple combination therapy, including early IV CYC, led to long-term deep remission. Regarding adverse events, although we did not compare the results statistically, there did not seem to be differences between serious adverse events in group T and group C. Moreover, group T was treated with a high total dose of CYC, and there were no malignancies that could be induced by CYC.
There have been several reports on the long-term prognosis of anti-MDA5–positive DM-ILD. Isoda et al17 reported a significantly lower relapse rate in anti-MDA5–positive DM-ILD than in anti-ARS–positive DM-ILD over a 2-year observation period. With regard to long-term prognostic data, there are reports of a case involving anti-MDA5–positive DM-ILD that remained in remission for approximately 7 years after initial treatment and a case where ILD recurrence was only observed after 9 years.18,19 However, these were relatively short-term studies with a small number of patients. Hence, these studies were insufficient to evaluate the long-term prognosis and management of anti-MDA5–positive DM-ILD during the remission period.
In our study, we included a substantial number of patients with anti-MDA5–positive DM-ILD, standardized the induction of remission therapy, and analyzed the data obtained from long-term observation (median 5.0 [IQR 0.5-20.3] years). The survival and relapse-free rates of anti-MDA5–positive DM-ILD were clearly higher during the remission maintenance period in the initial triple combination therapy group. Even in relapse cases, reinduction therapy, including IV CYC, demonstrated a high rate of remission reinduction.
This study had several limitations. First, there was selection bias regarding the inclusion of patients. We analyzed only patients who had survived for > 6 months, and patients with severe illness at initial therapy were excluded if they died within 6 months. Further, patients in group C were diagnosed in the previous time period (about 2000-2010 vs about 2010-2020 for group T), so most patients had delayed diagnosis because the recently established commercial base method for detecting anti-MDA5 was not used. As a result, in this study we analyzed patients with generally mild disease activity. Second, this study included indication bias because this was a retrospective study conducted by reviewing the charts and the therapy, including the prescription of drugs, which were dependent on the attending physician. Third, we only examined the activity of pulmonary lesions and did not assess other organ activities, such as skin or joint lesions. Since residual skin and joint lesions could affect the dose adjustment of GCs and CNIs, we preferred to evaluate overall disease activity using an established scoring system.
In conclusion, the induction of remission with triple combination therapy reduced the relapse risk of ILD and led to early dose reduction and discontinuation of immunosuppressive agents in the maintenance phase in many cases. Further investigation and validation are needed to establish a prospective study in which a drug-tapering protocol during the remission phase is defined and a multifaceted and standardized disease activity core set is used.
Footnotes
The authors declare no conflicts of interest relevant to this article.
- Accepted for publication August 23, 2023.
- Copyright © 2023 by the Journal of Rheumatology
REFERENCES
DATA AVAILABILITY
The data underlying this article will be shared on reasonable request to the corresponding author.
ONLINE SUPPLEMENT
Supplementary material accompanies the online version of this article.
