Clinical and real-world evidence indicates that switching between reference and biosimilar etanercept does not impact the safety and efficacy of treatment [25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42]. None of the reviewed switching studies reported any new or unexpected safety signals, or major changes in treatment efficacy, following a treatment switch [25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42]. Although data on PROs are limited, the available evidence suggests that PROs remain consistent after a treatment switch [29, 30, 34, 40]. A potential nocebo effect was reported in only two of the 14 studies reviewed [31, 37]. These promising data should bolster confidence in both healthcare professionals and patients that switching between reference and biosimilar etanercept does not have an impact on treatment safety and efficacy. This may help to reduce the incidence of the nocebo effect and could improve global biosimilar adoption rates.

The different etanercept biosimilars showed similar effectiveness and safety outcomes across real-world studies; however, treatment retention rates were generally slightly lower for SB4 than SDZ-ETN [29,30,31,32,33,34,35,36,37,38,39,40,41,42]. It is important to note that these studies are not directly comparable, and observed differences in retention rates may be due to variability in several factors, including study length, recruitment time periods and geographical locations (Table 2). Another explanation may be varying patient satisfaction with different autoinjector devices. In one retrospective study, 5% of patients reported difficulty with using the autoinjector device as the reason for withdrawal from SB4 treatment [37]. In line with these data, in a multinational survey, both patients and nurses displayed a higher preference for the SDZ-ETN autoinjector compared with the autoinjectors for reference etanercept and SB4 [43]. This preference was mainly due to the buttonless injection feature, visual feedback after injection and the convenient shape of the device [43].

Retention rates can also be affected by the incidence of the nocebo effect, necessitating a switch back to the less cost-effective reference etanercept [44]. In two real-world studies of SB4, patients who discontinued treatment after switching to the biosimilar had statistically significant increases in subjective measures of disease activity (e.g. DAS-28 or tender joint count) but no or only small changes in objective measures of inflammation (e.g. CRP level or the number of swollen joints), suggesting a potential nocebo effect [31, 37]. Notably, in these two studies, patients with RA had lower retention rates compared with those with PsA, AS or AxSpA, suggesting that the incidence of the nocebo effect might be higher in this patient population [31, 37, 45]. Interestingly, in one of these studies, patients with RA had the longest prior duration of treatment with reference etanercept [31]. Similarly, in the other study, longer baseline duration on reference etanercept among patients with RA was associated with an increased rate of biosimilar withdrawal [37]. This finding may, at least partly, be attributed to heightened anxiety regarding a change in treatment after a long period on reference etanercept, making the patients more vulnerable to the nocebo effect. However, development of the nocebo effect is also influenced by various other factors, including the individual patient’s psychosocial disposition [23]. As such, because of the observational design of the two switching studies [31, 37], no definitive causal relationships between biosimilar discontinuation and the nocebo effect were confirmed.

In the absence of validated diagnostic criteria, the nocebo effect remains difficult to accurately quantify and is, therefore, understudied [46]; none of the 18 studies reviewed in this article quantified its influence [25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42]. In the broader literature, the reported incidence of the nocebo effect in patients with RA switching to biosimilars has been highly variable, with some studies finding no evidence of such effect and others reporting rates of up to 13.1% [47,48,49,50]. Although one retrospective study in patients with RA or AxSpA found that shared decision-making can reduce the rate of the nocebo effect compared with a systematic switch carried out without patient consultation [47], a systematic review conducted across a range of different biologic drugs concluded that the current quality of evidence on the biosimilar nocebo effect is insufficient to draw any meaningful conclusions [51]. Therefore, further retrospective switching studies designed to assess PROs and subjective disease activity measures are needed to evaluate the real-world prevalence of a nocebo effect following a biosimilar treatment switch across different indications. Such studies may help to identify those patient populations that will benefit most from strategies known to mitigate the nocebo effect, such as educational schemes and shared decision-making [52], with the aim to reduce back-switching to reference bDMARDs and maximise cost savings for healthcare systems.

Biosimilars have the potential to introduce price competition and make bDMARDs more affordable. The benefits of biosimilar adoption have already been observed in Europe, where the entry of biosimilar etanercept into the market led to price reductions and increased treatment utilisation [12]. In France, 5 years after biosimilar etanercept introduction, the market share of reference etanercept had reduced from 100% to 26%, whereas the market share of SDZ-ETN had increased from 0 to 57% (translating to cost savings of almost €137 million) [16]. However, percentage changes in anti-TNF price and market share after biosimilar introduction have been shown to differ by up to 71% and 92%, respectively, across different European countries [4]. The international differences in cost savings can be attributed, at least in part, to varying biosimilar penetration rates, because of distinct regional-level biosimilar quotas, quota adherence monitoring mechanisms, insurer–manufacturer discount contracts and insurer–prescriber gain-sharing arrangements [53]. For example, in Norway, where physicians must use the cheapest available drug unless there is a clinical reason to do otherwise, the market share of biosimilar etanercept is above 82% [24]; however, similar strategies have been implemented in only approximately half of European countries [24].

Differences in biosimilar knowledge among physicians can also affect prescription rates [45]. Many patients find the switch between a reference and biosimilar medicine acceptable when it is recommended by their treating physician [54,