In Finland, early arthritis patients are mainly treated by rheumatologists. The Current Care Guideline advises general practitioners to refer all patients with suspected RA to specialist clinics. In addition, a rheumatologist’s certificate is needed to apply special reimbursement for antirheumatic medication, by which we identified our cases. Consequently, our cohort includes those arthritis patients, who had been examined by rheumatologists and prescribed DMARDs and glucocorticoids. Obviously, all patients with UA are not included. These facts explain, why within one month from the index date more than 90% of the RA patients purchased DMARDs and nearly 70% MTX. In seropositive patients, the percentages were higher, and within one year, only 1.4% of the seropositive and 2.6% of the seronegative RA patients had not purchased any DMARDs. For the UA patients the DMARD coverage was slightly less. These numbers also suggest a good drug adherence among Finnish arthritis patients.
Our results may not be comparable with studies in other settings. In a Canadian cohort studying the DMARD treatment of 24,942 early RA patients during the year following the diagnosis in 1997–2006, only 21% of patients treated by a general practitioner received any DMARDs, but 67% of those treated by a rheumatologist did so [12]. In a Danish cohort of 1516 early RA patients studying the initiation of MTX between 1996 and 2006, only 21% of the patients received MTX within 90 days; though, in 13% of the patients another DMARD had been initiated [13]. In studies based on large RA cohorts from US commercial and Medicare claims databases, the initiation of DMARD treatment within one year after diagnosis decreased from 63 to 56% between the cohorts 2004–08 and 2009–12 [14, 15]. Another US study based on claims databases found that over half of the 63,101 RA patients identified did not receive DMARD treatment within 90 days after diagnosis [16]. In a smaller Canadian cohort of 204 early RA patients in 2003–06, only 23% were prescribed a DMARD within 3 months and 47% within 6 months [17]. Also, in a recent Italian cohort of 1336 RA patients, less than 40% of the patients had started treatment with MTX within 3–6 months from the diagnosis [18].
Better coverages are found in contemporary materials treated by rheumatologists. In the French ESPOIR cohort of 775 early inflammatory arthritis patients, 77% received at least one DMARD after a median of four months [19]. A Canadian study of 339 RA patients found that 92% of the patients began DMARD therapy within three months [20]. In a multicenter ERAN cohort in UK and Eire, DMARDs were prescribed to 97% of the 808 early RA patients; however, the median time of DMARD initiation was 8 months after the symptom onset [21]. An Italian study reported that 83% of 10,401 patients were prescribed a csDMARD at RA diagnosis but only 6% of them received combination therapy [22].
Our previous analysis of Finnish early RA patients between 2000 and 2007 showed that although at the beginning of the study period SSZ was the most commonly prescribed DMARD during the first 3 months after the diagnosis, at the end of the observation period (2006–07) it had given way to MTX (69%) and combination DMARDs (53%) as the initial treatment [23]. Our earlier results also demonstrated that the proportion of patients starting triple combination within the first month increased from 6 to 16% between 2000 and 07. Our current results confirm that there has been a further increase. However, only 22% of early seropositive RA patients commenced the triple therapy recommended in the latest 2015 Current Care Guideline [8]. The Finnish recommendations from 2009 favored the start of the triple therapy and low dose prednisolone only for patients with very active RA, but not automatically as the first choice in all patients. Also, real life patients diagnosed with RA have seldom as active disease as patients in clinical trials [24], and further, they may have comorbidities and polypharmacy contraindicating certain medications, thus it is possible that rheumatologists base their decisions more on T2T principle than on slavishly following certain recommendations [25]. Either way, since we are lacking data of the levels of activity of the patients’ disease, further conclusions on whether only one-fifth of the patients had active disease requiring triple therapy cannot be drawn.
There is a distinction between guidelines encouraging a T2T strategy on the one hand and what actually happens in practice. In an ESPOIR cohort, where adherence to three of the EULAR recommendations concerning the start and early adjustment of DMARDs was studied, the adherence rate for all three recommendations was only 23% among early arthritis patients [26]. Still, among those patients whose treatment adhered to given recommendations, the risk of clinical and radiographic progression was lower. Knowing that the fulfilment of treatment guidelines in real life is always suboptimal, the strict national treatment recommendations, such as the recommendation of triple therapy initiation in early RA in Finland, will probably lead to optimal outcomes.
The number of DMARD purchases could reflect drug adherence although some patients may not always buy medication for the next three months as usually happens. In our analysis, the patients initiating the FIN-RACo combination had twice the number of DMARD purchases during the first year compared to those patients who did not start with the FIN-RACo combination. Frequent purchases suggest a regular drug usage and good survival rate of the FIN-RACo combination.
Two Finnish studies based on real life early arthritis patient cohorts have been published. The first one included 406 early RA or UA patients between 2008 and 11 [27]. Of the RA patients, at three months 20% were using triple therapy, 33% other MTX based combination, 36% MTX monotherapy, and 8% other DMARD monotherapy; for the UA patients the respective percentages were 6, 28, 43, and 17%, respectively. At one year, the proportions of RA patients using various medications had not changed markedly. In a more recent (2011–14) FIN-ERA cohort of 611 DMARD naïve early arthritis patients (506 RA and 105 UA patients) recruited in five Finnish outpatient rheumatology clinics, MTX-based combination therapy was initiated to 68% of the patients and the proportion of triple combination (MTX, SSZ and HCQ) was 31% [28]. These results, in line with our results, show that DMARD initiation for early arthritis patients is generally comprehensive in Finland.
In Finland there are no separate treatment recommendations for UA, but the active T2T principle is widely used in clinical practice regardless of the diagnosis. The European League of EULAR recommendations for early arthritis in 2007 recommended for patients at risk of developing persistent and/or erosive arthritis a DMARD as early as possible, preferably MTX, even if no classification criteria for a specific disease are fulfilled [29]. The latest update of the recommendation in 2016 presented no major changes to these principles [9]. The fulfillment of the EULAR recommendations for the treatment of early arthritis was studied in 813 patients from the ESPOIR cohort between 2002 and 05; 78% of patients started a DMARD, 67% MTX and 52% reached remission [30]. In our material the DMARD initiation was more comprehensive, but the proportion on MTX lower; this might be explained by the fact that traditionally SSZ has been prescribed in seronegative oligoarthritis in Finland.
In our study, the use of self-injected biological DMARDs was more common among seronegative RA patients than seropositive RA or UA patients. Seronegative RA may be a heterogenous group of diseases, as shown in a recent 10-year observational study [31], thus explaining poorer treatment outcomes with csDMARDs and a greater need for switching to bDMARDs.
A shortcoming of this study is the lack of clinical data; we do not have information of the disease’s activity at diagnosis, nor at follow-up. Further, we may miss patients that the physician intends not to treat. Nevertheless, the incidence of seropositive RA has remained stable throughout this millennium, that of seronegative RA has decreased slightly supposedly due to changed diagnostics, and for the same reason, the incidence of UA has increased [32]. Thus, it seems unlikely that we are missing many patients. However, since this is a register-based study and we lack the data on the duration of symptoms before the diagnosis, it is possible that the current study includes some patients that are diagnosed with a time lag; in these cases, DMARDs used represent patients’ initial treatment rather than the treatment of early rheumatic disease. Also, we do not know how high a proportion of UA patients received a more specific diagnosis later; this could offer an interesting area for further research. Further, in the lack of clinical data it is possible that a certain proportion of the UA patients in our study may not be comparable to patients in so called early arthritis clinics, but have a chronic inflammatory arthritis requiring specific anti-rheumatic drug therapy.
Even though we do not have any clinical outcome measures, the initiation of self-injected biologics served as a surrogate marker of treatment failure. We were expecting to see that patients having received combination DMARDs as their first treatment, and thus judged by their treating rheumatologist to have an active disease to be the ones to end up starting a biologic earlier and more often than other patients, but at least during the first year that was not the case. Thus, at least in the early phase, combination DMARD treatment appears to be effective.
Although we are lacking the data on infusion based biological drugs, only a small proportion of patients are initiating them during the first year after the diagnosis, the time period of interest in our study. According to the Finnish ROB-FIN register study, in 2004–14 the first TNF-inhibitors were initiated to RA patients after a median (IQR) of 8.2 (2.4–17) years of disease duration, most often with adalimumab (39%) or etanercept (39%), while infliximab was a rarer choice (12%) [33]. According to the most recent, yet unpublished, ROB-FIN register data from 2010 to 15, the use of infliximab as the first biologic had decreased from 7.5 to 4.2% for RA patients (Kalle Aaltonen, personal communication). Rituximab was the first choice for as many as 20–23% of the RA patients. Still, the majority started self-injected drugs, and the median (IQR) point of starting the first biologic was after 10 (4.4–18) years of disease duration. Consequently, our results of the first year treatment would hardly have changed markedly, were the infusion-based drugs included.