Commentary: ILD and other issues in RA treatment, March 2023

Two recent studies examined interstitial lung disease (ILD) in patients with rheumatoid arthritis (RA). Albrecht and colleagues examined the prevalence of ILD in German patients with RA using a nationwide claims database from 2007 to 2020. Using diagnosis codes for seropositive and seronegative RA (along with disease-modifying antirheumatic drug prescriptions) as well as ILD, they found the prevalence of ILD to be relatively stable from 1.6% to 2.2%, and that incidence was stable (reported as 0.13%-0.21% per year, rather than per patient-year) over the course of the study. There is likely some misclassification with the primary reliance on diagnosis codes (of the included patients with RA only 44% were seropositive). They also excluded drug-induced ILD by diagnosis code, which may not be sufficient. Overall, the prevalence of ILD seems on the low end of what might be expected and may reflect a need for earlier evaluation to detect subclinical ILD.

Kronzer and colleagues performed a case-control study of 84 patients with incident RA-ILD compared with 233 patients with RA without ILD to evaluate the risk associated with specific anticitrullinated protein antibodies (ACPA) for the development of ILD. Compared with the clinical risk factors of smoking, disease activity, obesity, and glucocorticoid use, six "fine-specificity" ACPA were better able to predict ILD risk, with immunoglobulin (Ig) A2 to citrullinated histone 4 associated with an odds ratio (OR) of 0.08, and the others (IgA2 to citrullinated histone 2A, IgA2 to native cyclic histone 2A, IgA2 to native histone 2A, IgG to cyclic citrullinated filaggrin, and IgG to native cyclic filaggrin) were associated with OR of 2.5-5.5 for ILD. In combination with clinical characteristics, the authors developed a risk score with 93% specificity for RA-ILD that should be validated in other populations.

Suh and colleagues examined the association of RA with another less well-studied organ complication, end-stage renal disease (ESRD), using a large national insurance database. Once again, the accuracy of diagnosis is not fully clear using International Classification of Diseases, Tenth Edition (ICD-10), codes for classification. Overall, people with RA had a higher risk for ESRD than did people without RA, regardless of sex or smoking status. Because no immediate mechanistic connection between RA and ESRD is evident, it is possible that part of the increased risk is due to medications used in RA treatment, such as nonsteroidal anti-inflammatory drugs, but this hypothesis remains to be tested.

Finally, a footnote to the success of the treat-to-target strategy (T2T) in RA comes in a study by Ramiro and colleagues of the RA-BIODAM cohort, which, along with other studies, has shown the success of T2T in achieving and maintaining long-term clinical remission in RA. The effect of T2T on radiographic progression, however, is less clear. In this study, over 500 patients were followed for 2 years and a comparison between the T2T strategy and radiographic damage was made. The T2T strategy consisted of intensification of treatment if the Disease Activity Score (DAS-44) did not achieve a goal of < 1.6. This was compared with the radiographic damage (based on the change in Sharp-van der Heijde score[SvdH]) over a 6-month period. Overall, the change in progression was not different among patients who were treated with a stricter adherence to T2T (ie, a higher proportion of T2T) compared with those who were not, suggesting that a looser application of T2T will not necessarily cause a worsening of radiographic progression. It is possible, given the intervals of assessment in this study, that a longer follow-up after T2T is necessary to detect progression, or, given that patients were not randomly assigned, patients who were more strictly treated with T2T were already at higher risk for radiographic progression. However, this study is also helpful in understanding how insights from controlled trials may play out in usual clinical practice.

Two recent studies examined interstitial lung disease (ILD) in patients with rheumatoid arthritis (RA). Albrecht and colleagues examined the prevalence of ILD in German patients with RA using a nationwide claims database from 2007 to 2020. Using diagnosis codes for seropositive and seronegative RA (along with disease-modifying antirheumatic drug prescriptions) as well as ILD, they found the prevalence of ILD to be relatively stable from 1.6% to 2.2%, and that incidence was stable (reported as 0.13%-0.21% per year, rather than per patient-year) over the course of the study. There is likely some misclassification with the primary reliance on diagnosis codes (of the included patients with RA only 44% were seropositive). They also excluded drug-induced ILD by diagnosis code, which may not be sufficient. Overall, the prevalence of ILD seems on the low end of what might be expected and may reflect a need for earlier evaluation to detect subclinical ILD.

Kronzer and colleagues performed a case-control study of 84 patients with incident RA-ILD compared with 233 patients with RA without ILD to evaluate the risk associated with specific anticitrullinated protein antibodies (ACPA) for the development of ILD. Compared with the clinical risk factors of smoking, disease activity, obesity, and glucocorticoid use, six "fine-specificity" ACPA were better able to predict ILD risk, with immunoglobulin (Ig) A2 to citrullinated histone 4 associated with an odds ratio (OR) of 0.08, and the others (IgA2 to citrullinated histone 2A, IgA2 to native cyclic histone 2A, IgA2 to native histone 2A, IgG to cyclic citrullinated filaggrin, and IgG to native cyclic filaggrin) were associated with OR of 2.5-5.5 for ILD. In combination with clinical characteristics, the authors developed a risk score with 93% specificity for RA-ILD that should be validated in other populations.

Suh and colleagues examined the association of RA with another less well-studied organ complication, end-stage renal disease (ESRD), using a large national insurance database. Once again, the accuracy of diagnosis is not fully clear using International Classification of Diseases, Tenth Edition (ICD-10), codes for classification. Overall, people with RA had a higher risk for ESRD than did people without RA, regardless of sex or smoking status. Because no immediate mechanistic connection between RA and ESRD is evident, it is possible that part of the increased risk is due to medications used in RA treatment, such as nonsteroidal anti-inflammatory drugs, but this hypothesis remains to be tested.

Finally, a footnote to the success of the treat-to-target strategy (T2T) in RA comes in a study by Ramiro and colleagues of the RA-BIODAM cohort, which, along with other studies, has shown the success of T2T in achieving and maintaining long-term clinical remission in RA. The effect of T2T on radiographic progression, however, is less clear. In this study, over 500 patients were followed for 2 years and a comparison between the T2T strategy and radiographic damage was made. The T2T strategy consisted of intensification of treatment if the Disease Activity Score (DAS-44) did not achieve a goal of < 1.6. This was compared with the radiographic damage (based on the change in Sharp-van der Heijde score[SvdH]) over a 6-month period. Overall, the change in progression was not different among patients who were treated with a stricter adherence to T2T (ie, a higher proportion of T2T) compared with those who were not, suggesting that a looser application of T2T will not necessarily cause a worsening of radiographic progression. It is possible, given the intervals of assessment in this study, that a longer follow-up after T2T is necessary to detect progression, or, given that patients were not randomly assigned, patients who were more strictly treated with T2T were already at higher risk for radiographic progression. However, this study is also helpful in understanding how insights from controlled trials may play out in usual clinical practice.

Two recent studies examined interstitial lung disease (ILD) in patients with rheumatoid arthritis (RA). Albrecht and colleagues examined the prevalence of ILD in German patients with RA using a nationwide claims database from 2007 to 2020. Using diagnosis codes for seropositive and seronegative RA (along with disease-modifying antirheumatic drug prescriptions) as well as ILD, they found the prevalence of ILD to be relatively stable from 1.6% to 2.2%, and that incidence was stable (reported as 0.13%-0.21% per year, rather than per patient-year) over the course of the study. There is likely some misclassification with the primary reliance on diagnosis codes (of the included patients with RA only 44% were seropositive). They also excluded drug-induced ILD by diagnosis code, which may not be sufficient. Overall, the prevalence of ILD seems on the low end of what might be expected and may reflect a need for earlier evaluation to detect subclinical ILD.

Kronzer and colleagues performed a case-control study of 84 patients with incident RA-ILD compared with 233 patients with RA without ILD to evaluate the risk associated with specific anticitrullinated protein antibodies (ACPA) for the development of ILD. Compared with the clinical risk factors of smoking, disease activity, obesity, and glucocorticoid use, six "fine-specificity" ACPA were better able to predict ILD risk, with immunoglobulin (Ig) A2 to citrullinated histone 4 associated with an odds ratio (OR) of 0.08, and the others (IgA2 to citrullinated histone 2A, IgA2 to native cyclic histone 2A, IgA2 to native histone 2A, IgG to cyclic citrullinated filaggrin, and IgG to native cyclic filaggrin) were associated with OR of 2.5-5.5 for ILD. In combination with clinical characteristics, the authors developed a risk score with 93% specificity for RA-ILD that should be validated in other populations.

Suh and colleagues examined the association of RA with another less well-studied organ complication, end-stage renal disease (ESRD), using a large national insurance database. Once again, the accuracy of diagnosis is not fully clear using International Classification of Diseases, Tenth Edition (ICD-10), codes for classification. Overall, people with RA had a higher risk for ESRD than did people without RA, regardless of sex or smoking status. Because no immediate mechanistic connection between RA and ESRD is evident, it is possible that part of the increased risk is due to medications used in RA treatment, such as nonsteroidal anti-inflammatory drugs, but this hypothesis remains to be tested.

Finally, a footnote to the success of the treat-to-target strategy (T2T) in RA comes in a study by Ramiro and colleagues of the RA-BIODAM cohort, which, along with other studies, has shown the success of T2T in achieving and maintaining long-term clinical remission in RA. The effect of T2T on radiographic progression, however, is less clear. In this study, over 500 patients were followed for 2 years and a comparison between the T2T strategy and radiographic damage was made. The T2T strategy consisted of intensification of treatment if the Disease Activity Score (DAS-44) did not achieve a goal of < 1.6. This was compared with the radiographic damage (based on the change in Sharp-van der Heijde score[SvdH]) over a 6-month period. Overall, the change in progression was not different among patients who were treated with a stricter adherence to T2T (ie, a higher proportion of T2T) compared with those who were not, suggesting that a looser application of T2T will not necessarily cause a worsening of radiographic progression. It is possible, given the intervals of assessment in this study, that a longer follow-up after T2T is necessary to detect progression, or, given that patients were not randomly assigned, patients who were more strictly treated with T2T were already at higher risk for radiographic progression. However, this study is also helpful in understanding how insights from controlled trials may play out in usual clinical practice.