Effect of Cotreatment with Corticosteroids in Connective Tissue Related Lung Disease (CTD-ILD) Patients on Mycophenolate Mofetil (MMF)

Introduction: Connective Tissue Disease Related Interstitial Lung Disease (CTD-ILD) is often treated with immunosuppressant medications; common among these is Mycophenolate Mofetil (MMF). We hypothesized that co-treatment with corticosteroids would impact disease progression. Methods: We examined a consecutive cohort of CTD-ILD patients followed at Temple University Hospital in Philadelphia, PA since 2015 who had pulmonary function tests (PFTs) performed by American Thoracic Society (ATS)/European Respiratory Society (ERS) Criteria at least one year apart. All patients were treated for CTD-ILD with MMF used either as sole therapy or as combination therapy with prednisone. Univariate logistic analyses were performed revealing the odds ratio (OR) for improvement or worsening of several PFT values (including forced vital capacity (FVC), diffusion capacity of carbon monoxide (DL CO ), and six-minute walk (6MW)) greater than the minimal clinically important difference (MCID) for each value. Results: We included 103 patients (74 women) with an average age of 60 ± 11 years, 49% of our cohort were current or former smokers, and mean BMI was 29 ± 7 kg/m 2 . Patients were observed on treatment for an average of 23 months. CTD distribution included 25% mixed connective tissue disease (MCTD), 24% systemic sclerosis (SSc), 17% rheumatoid arthritis (RA), 14% systemic lupus erythematosus (SLE), 10% other idiopathic in�ammatory myositis (IIM) syndromes, 7% Antisynthetase Syndrome, 5% Sjӧgren’s syndrome. Non-specic interstitial pneumonia (NSIP) was the majority (45%) ILD pattern noted, Usual Interstitial Pneumonia (UIP) 35%, and other types were less prevalent (20%). The majority of patients received corticosteroids as co-treatment with MMF (75 patients (72%)) with a mean daily dose of 15 ± 16 mg of prednisone. Mean daily MMF dose was 1144 ± 675 mg. Glucocorticoid treatment was not associated with signi�cant improvements in PFT values, including FVC, DL CO , and 6MW distance walked. Conclusion: In this small cohort, patients with CTD-ILD receiving MMF did not demonstrate improved lung function when receiving co-treatment with corticosteroids, but larger prospective studies are needed to better elucidate the effect of corticosteroids on this vulnerable group of patients


Introduction
Connective tissue disease-related interstitial lung disease (CTD-ILD) is a form of interstitial lung disease (ILD) common to patients with connective tissue diseases (CTD) including systemic sclerosis (SSc), rheumatoid arthritis (RA), idiopathic in ammatory myositis (IIM) syndromes, Sjӧgren's syndrome, mixed connective tissue disease (MCTD), and other forms of CTD. 1,2][3][4][5] Medications commonly used in CTD-ILD include mycophenolate mofetil (MMF), Tacrolimus, Cyclophosphamide, Azathioprine, Rituximab, and corticosteroids. 4][8][9][10][11] CTD-ILD patients bene t from few published randomized controlled trials. 10,12rticosteroids are often used as co-management with other therapies in CTD-ILD patients.Because clinical trials have focused mainly on SSc-ILD patients, there has been a limited role for corticosteroids at doses > 15mg/d of prednisone due to the risk of renal crisis.For broader populations, corticosteroid use in CTD-ILD has not been studied in detail. 13,14Yamano et al. (2018) evaluated high dose methylprednisolone followed by low dose prednisone and Tacrolimus in a population of 26 CTD-ILD patients, a protocol which produced improvements, 15 but to our knowledge no other research has evaluated the bene t of corticosteroids in patients taking other therapies.
We hypothesized that patients taking MMF for CTD-ILD would gain a small bene t from co-treatment with corticosteroids.

Methods
After approval by the Temple University Institutional Review Board, we retrospectively analysed a consecutive cohort of CTD-ILD patients prescribed MMF.Each patient had a diagnosis of CTD per American College of Rheumatology (ACR) classi cation criteria by a board-certi ed rheumatologist and had ILD con rmed by high-resolution CT chest (HRCT) or by histology.Each patient also had pulmonary function tests including spirometry, diffusion capacity for carbon monoxide (DL CO ), and six-minute walk (6MW).
Each patient was followed continuously at Temple University Hospital (Philadelphia, PA) for an average of 23 months between January 1, 2015 and January 1, 2019, with Pulmonary Function Tests (PFTs) performed according to American Thoracic Society (ATS)/European Respiratory Society (ERS) Criteria measured at least one year apart. 16We excluded patients with inadequate PFTs (n = 52), those who did not meet ACR classi cation criteria for a speci c CTD (n = 82), those who did not have ILD by review of HRCT of chest or by histology (n = 51), those who were followed for less than 26 weeks (n = 11), those who received a lung transplant (n = 9), or those with physician-documented MMF nonadherence (n = 5) (Fig. 1).
103 subjects met the above inclusion and exclusion criteria.The following data were extracted from the medical record for each subject: demographics including age, gender, body mass index (BMI), race, ethnicity, and smoking status, pattern of ILD by HRCT or histology, baseline autoantibodies tested, rheumatologic diagnosis by ACR criteria, systemic manifestations of CTD as documented by a board-certi ed rheumatologist, and treatment dose and duration for both MMF and corticosteroids (Figs.2,3; Table 1).

Results
We included 103 patients who were observed on treatment with MMF for an average of 23 months.The baseline characteristics of the 103 consecutive patients are summarized in Table 1 and compared between groups.Patients were predominantly female had an average age of 60 ± 11 years, a mean BMI of 29 ± 7, and 49% were current or former smokers.The racial breakdown of our cohort was 52% Caucasian, 31% African American, 2% Asian, and 15% other race or mixed race.15% of patients identi ed as being of hispanic ethnicity (Table 1).
Signi cant differences were noted between patients receiving MMF as sole therapy when compared to patients receiving corticosteroids as co-treatment with MMF.Patients receiving co-treatment with corticosteroids were more likely to have diagnoses of RA or MCTD, were more likely to report arthropathy, and received lower starting doses of MMF (1007 ± 469 milligrams (mg) per day vs 1539 ± 926 mg/day (p = 0.009) (Table 1).
For the entire cohort, decreases were noted in mean FVC, mean DL CO , and mean DL CO % predicted over the follow-up interval, but ambulatory oxygen requirement in liters/minute via nasal cannula and distance walked on 6MW testing did not change signi cantly.The mean prescribed prednisone dose decreased signi cantly (p < 0.001) from 15.7 to 7.9 mg and the mean MMF dose increased signi cantly (p < 0.001) from 1275 to 2143 mg/day over the course of the study period (Table 2).
We hypothesized that patients taking MMF for CTD-ILD would gain a small bene t from cotreatment with corticosteroids.There were no signi cant differences in changes in PFT values between those treated with corticosteroids in addition to MMF versus those treated only with MMF.Over the observation period, there was no between-groups difference in change in FVC, DL CO , or 6MW distance walked over the MCID for each test (Table 3).

Interpretation
Corticosteroids are a mainstay of treatment in CTD-ILD, but there are few studies investigating their e cacy and there are presently no guidelines directing their use.In our study, there was no effect of steroid cotreatment in CTD-ILD patients taking MMF.
Corticosteroids are frequently prescribed as both monotherapy and cotherapy for CTD-ILD due to antiin ammatory and immunosuppressive effects on endothelial, broblast, and leukocyte cell function. 17he use of chronic low-dose steroids as co-management in the maintenance phase of disease has been shown to improve CTD-related arthropathy and myopathy. 16Although long-term steroid use in ILD patients has been implicated in the development of dose-dependent steroid-induced myopathy, it has not been shown to have an overall effect on exercise capacity or ability to tolerate activities of daily living (ADL) in CTD patients. 18eroid-sparing agents such as MMF may reduce steroid requirements while stabilizing DL CO and FVC. 7r cohort as a whole suffered a small decrease in FVC and DL CO over the follow up interval, but had no change in exercise capacity while taking decreasing steroid dosages and increasing MMF dosages over the course of the study period.This nding is re ective of the progressive impact of CTD-ILD on pulmonary physiology, but suggests a discrepancy between physiological effects and overall clinical signi cance to patients.These ndings highlight the need for further study to better correlate effects on pulmonary physiology to functional status and overall quality of life.
This study had several limitations.Our cohort of only 103 patients was small and recruited from a single center.However, this cohort is valuable for study, as most research on CTD-ILD patients thus far includes a large number of patients with systemic sclerosis. 6,9,12Our cohort had a diverse distribution of CTD and may be more representative of the diversity of disease encountered in clinical practice outside of systemic sclerosis specialty centers.Larger studies across multiple treatment centers are needed to avoid confounding variables and to increase con dence in the results.Our cohort was recruited at a specialty lung center, which may confer referral bias.
As there are no guidelines for corticosteroid treatment in patients with CTD-ILD, more research is needed to elucidate the optimal dosing and duration of therapy, and to clarify the goals of therapy.It remains unclear whether the addition of corticosteroids to steroid-sparing therapy bene ts patients with CTD-ILD, and if so, in which subgroups and to what degree.Our patients received a minimum of 26 weeks and average of 23 months of follow-up.Longer term analyses may further uncover corticosteroid side effects which may affect their long-term tolerability.

Conclusion
In this small cohort, patients with CTD-ILD receiving treatment with MMF did not demonstrate improved lung function when receiving co-treatment with corticosteroids.Larger prospective studies are needed to better elucidate the effect of corticosteroids on this vulnerable group of patients.
Abbreviations List

Declarations
This research was conducted after IRB approval and was deemed exempt from ethics committee approval by Temple University, and in compliance with all relevant guidelines and regulations.
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Figures Figure 1 Inclusion and Exclusion Criteria Figure 2
Figures

Table 1
Demographic Data of CTD-ILD Patients at Time of MMF Treatment Initiation Legend: Connective Tissue Disease-Associated Interstitial Lung Disease (CTD-ILD), Mycophenolate Mofetil (MMF), American College of Rheumatology (ACR), Connective Tissue Disease (CTD), Standard Deviation (SD), Mixed Connective Tissue Disease (MCTD), Rheumatoid Arthritis (RA), Systemic Lupus Erythematosus (SLE), milligrams (mg) Patients were then divided into two groups, those who received only MMF, and those who received MMF and a corticosteroid.Chi-squared or paired t-tests were used to compare demographic and baseline data between the two groups, and to describe the change in PFT values and medication doses over time in the entire cohort (Table1,2).

Table 2 Pulmonary
Function Tests, Exercise Capacity, Oxygen Requirement, and Corticosteroid Dose Over 23 Months of Observation on MMF, Entire Cohort Univariate logistic analyses were then performed revealing the odds ratio (OR) for improvement or worsening of several PFT values between the two groups (including forced vital capacity (FVC), DL CO , and 6MW distance greater than the minimal clinically important difference (MCID) for each value.Given the small number of patients in the study, improvement of FVC, DL CO , and/or 6MW distance greater than the MCID for each test was analysed as a combined variable.Data are presented as percent (%), mean ± standard deviation (SD) or OR ± con dence interval.P-values < 0.05 were considered statistically signi cant.