Purpose
To demonstrate the usefulness of the BioRx.si registry for monitoring effectiveness (hypotheses 1A and 1B) and safety (hypothesis 2) of biological disease modifying anti-rheumatic drugs (bDMARDs) used for treating rheumatic diseases.
Hypotheses
1A. In patients with rheumatoid arthritis (RA) who failed treatment with first tumor necrosis factor inhibitor (TNFi) due to ineffectiveness or intolerance, the survival of the subsequent bDMARD depends on its mechanism of action.
1B. In patients RA, ankylosing spondylitis (AS) and psoriatic arthritis (PsA) there is no difference in the persistence of one TNFi, i.e., golimumab, when compared to other TNFis (oTNFi).
2. Two-step national screening for latent tuberculosis (TB) infection (LTBI) in patients with RA, AS and PsA treated with TNFis, which in the first step includes medical history, tuberculin skin test (TST), chest X-ray and, in the event of pathological findings in the first step, of the examination at a pulmonologist, who, if necessary, orders additional tests and prescribes TB chemoprophylaxis, ensures a comparably small incidence of infection with mycobacteria tuberculosis in this group of patients, as is described in the available literature.
Study concept, methodology, subjects
All Slovenian rheumatologists contribute data to the Slovenian national registry BioRx.si on-line. Effectiveness and safety data have been collected for RA since 2008, and for AS and PsA since 2010. For research purposes, data from the registry can be accessed using various parameters and can then be statistically analyzed.
In all analyses we used appropriate descriptive statistical methods to characterize the currently studied groups of patients. In the analyses pertaining to hypotheses 1A and 1B we assessed the persistence of bDMARDs, a composite marker of effectiveness, and safety, using the Kaplan-Meier and Cox methods. To test hypothesis 2, we assumed Poisson distribution to determine the TB IR. We determined the age and sex standardized TB IR using the direct method of standardization, with the Slovenian general population divided by sex into 5-year brackets as the standard population. Additionally, we estimated the standardized TB IR (SIR) in comparison with the general Slovenian population.
Results
1A. Two hundred thirty-eight out of 688 patients who received a TNFi as the first bDMARD were switched to another bDMARD by December 2012: 130 to a second TNFi and 108 to either rituximab (31.5%) or tocilizumab (68.5%) (non-TNFi). Disease activity at starting the second bDMARD and stopping the first TNFi due to either lack or loss of effectiveness were identified as potential confounders. There appeared to be a statistically significant retention advantage of the non-TNFi over the second TNFi (log rank test, p=0.000). This advantage was retained even after adjusting for measured confounders using the inverse probability-weighted Cox model (hazard ratio (HR) 4.39; 95% CI 2.62–8.01, p<0.001).
1B. During the 8-year observation period, from 1 January 2010 to 31 July 2018, 24 Slovenian rheumatologists from eight centers contributed data on 368, and 1654 patients treated for 849, and 3321 person-years with golimumab and oTNFis, respectively. The overall proportions of RA, AS and PsA patients being persistent on golimumab vs. oTNFis at 2 years after starting the therapy did not differ significantly and were 53%, 67% and 59% vs. 47%, 65% and 59%, respectively. The crude and adjusted hazard ratios for golimumab discontinuation did not differ significantly between bDMARD-naïve and bDMARD-experienced patients for any of the indications. In contrast, bDMARD-experienced AS and PsA patients treated with oTNFis were significantly more likely to discontinue treatment.
2. Among the 2429 patients exposed to at least one TNFi for a total of 10,445 (49% RA, 33% AS and 18% PsA) person-years (PY), 99% completed LTBI screening and 6% required TB chemoprophylaxis. Six RA (three adalimumab, three certolizumab), two PsA (two golimumab) and zero AS patients developed TB. Five out of eight had miliary TB, three out of eight had pulmonary TB and two patients died. The age-standardized and sex-standardized TB IR (95% CI) per 100,000 PYs/SIRs (95% CI) compared with the general Slovenian population for the current TNFi exposure were 52 (0–110)/6.7 (0.6–80), 47 (0–110)/6.1 (0.3–105), 45 (0–109)/5.8 (0.3–112) overall, in RA and PsA, respectively.
Conclusions
1A. In RA patients that discontinued treatment with the first TNFi due to ineffectiveness or side effects, the persistence of the second bDMARD was better if a non-TNFi agent, rather than a second TNFi was used.
1B. The persistence of golimumab in patients with RA, AS, and PsA in Slovenia was comparable with its persistence in more affluent Western European countries. We observed a better persistence of golimumab compared to other TNFis in bDMARD-experienced AS and PsA patients.
2. The TB IR in the Slovenian RA, AS, and PsA patients treated with TNFi was comparable to TB IRs in TB non-endemic countries with less than a tenth of the patients requiring TB chemoprophylaxis.
|
