Introduction

The therapeutic options available for the management of rheumatoid arthritis (RA) have been considerably diversified these past years, regarding both their mechanism of action but also their route of administration. Particularly, abatacept (ABA) and tocilizumab (TCZ) that were initially exclusively administered intravenously (IV) were developed for subcutaneous (SC) administration with efficacy and tolerance profiles similar to the IV formulation [1, 2]. The switch from the IV route to the SC route of administration has shown sustained efficacy [3]. ABA has been available in its subcutaneous formulation in France since April 2014, and SC TCZ was made accessible in May 2015.

From a medico-economic point of view, SC biologics are generally cost-saving compared with IV biologics [4]. In France, people undergoing IV infusions have an open choice to switch to a self-administered SC route or to keep with their hospital-administered infusions. Little is known about what motivates this choice; understanding the underlying reasons could be helpful in guiding patients towards the most cost-effective choice.

From a clinical point of view, adherence is a key issue for any successful treatment, including the use of biologics for the treatment of RA [5]. Patients’ belief in their own medication is an important factor for their adherence, leading rheumatologists to take into account patients’ preferences when choosing a treatment [6–8]. Now that available treatments for RA seem to have reached comparable efficacy, the most important factor influencing patients’ preferences is the route of administration [9]. Understanding and discussing with patients, their preferences for a given route of administration would be a step towards improving the adherence, and hopefully the efficacy, of TCZ and ABA treatments.

The objective of this study was to examine patients’ reasons for choosing to keep with their IV infusions of ABA or TCZ or to switch to SC injections.

Methods

This was an observational, cross-sectional, multicentre study performed between April 2014 and December 2014 among patients receiving IV infusions (ABA or TCZ) for the treatment of RA in three rheumatology departments of tertiary care hospitals.

Participants

All patients were above 18 years of age and fulfilled the 1987 ACR criteria for the diagnosis of RA. Patients were excluded if were presenting with cognitive impairment, with French language difficulties or who were changing their biologic treatment in the day of data collection.

Data on the following were collected: age, sex, occupational activity, date of the disease onset, rheumatoid factor (RF) and/or ACPA positivity, presence of bony erosions, presence of rheumatoid nodule, duration of ABA or TCZ treatment, combination with conventional synthetic disease-modifying anti-rheumatic drugs (cs-DMARDs) or corticosteroids and number of previous or ongoing SC treatments. RA activity at inclusion was also evaluated (tender and swollen joint counts, patient’s pain assessment scale (0–100), ESR, CRP, DAS28-ESR, DAS28-CRP).

The study was approved by an institutional review board. Data were anonymised and written consent was obtained.

Questionnaire

This study was based upon an easily understandable self-administered questionnaire completed by consecutive RA patients attending outpatient or day-care hospital units for their IV infusion of ABA or TCZ. Patients filled in their questionnaire alone, without the presence of a physician or a nurse. The attending physician did not have access to patients’ responses. The aim of the questionnaire was to determine what reasons had motivated or would motivate patients to decide to switch to the SC route of administration or to keep with the IV route. A variety of reasons guiding this choice were proposed for patients to select. Patients were questioned on how and by whom injections would be performed if the SC route were chosen (self-administration, nurse, third party). Finally, there was an item concerning the person to contact in case of a problem with the SC treatment.

Statistical analyses

Quantitative data are reported as mean ± standard deviation. Qualitative data are reported as number and percentage. The Chi-square test, or Fisher’s exact test as appropriate, was used for comparative analysis of proportions. Continuous variables were compared using Student’s t test for normal data, or Mann-Whitney-Wilcoxon test. Multivariate analysis was then applied to search for factors affecting the choice of route of administration. A binary logistic regression model was constructed with the three variables exhibiting p < 0.2 at bivariate analysis: other ongoing SC treatment, experience of previous IV biologics and leflunomide intake. ROC area under the curve analysis was used to assess model quality. As it was not satisfactory (0.64), the model relevance was dubious, and the coefficients of the model were not interpreted. Computations were performed with R software (version 3.2.5). A p value less than 0.05 was considered statistically significant.

Results

Patients’ characteristics

A total of 238 patients were screened, 226 questionnaires collected and 201 questionnaires analysed. Among the patients who completed analysable questionnaires, 127 patients were treated by TCZ and 74 by ABA (Fig. 1).

Fig. 1
figure 1

Study flow-chart

Full size image

Patients were predominantly females (81.1%), aged on average 58.4 ± 12.6 years, and generally with a low disease activity (mean DAS28-CRP 3.1 ± 1.3; mean DAS28-ESR 2.8 ± 1.3). Search for RF or anti-CCP antibodies was positive in 94.7% of the patients, and 80% had radiographic erosions (Table 1). A csDMARDS was associated in 43.3% of the cases and prednisone in 32.5%. Disease duration was 16.8 ± 11.4 years. Patients had been treated by ABA or TCZ for 36.1 ± 24.9 months.

Table 1 Demographics and disease characteristics of study participants compared according to routes of administration inside each treatment group
Full size table

Overall, 45.8% (92/201) of the patients chose to keep the IV route of administration (IV group). The only significant difference compared with patients who chose to switch to SC injections (SC group) was another ongoing SC treatment: 15.9% in the SC group versus 4.3% in the IV group (p = 0.016).

Patients receiving TCZ had previously received more SC biologics in the SC group (1.3 versus 0.9, p = 0.023). More patients receiving leflunomide chose the IV route (12.7 versus 2.8%) (p < 0.05). Patients choosing the SC route had previously received a greater number of SC biologics (1.3 ± 0.9 versus 0.9 ± 0.9) (p < 0.05). Particularly, patients having previously received etanercept chose more predominantly the SC route (72.2 versus 50.9%) (p < 0.05). None of the other types of SC biologics previously used differed significantly between groups.

Patients receiving ABA had more active diseases with a mean DAS28-CRP of 3.7 ± 1.1 in the SC group and 3.1 ± 1.0 in the IV group (p < 0.05). Patients with other ongoing SC treatments favoured more easily the SC route (22.9 versus 5.4%) (p < 0.05), whereas patients having experienced previous IV biologics favoured the IV route (75.0 versus 45.9%) (p < 0.05). The quality of the model constructed in multivariate analysis and based on three variables (other ongoing SC treatment, previous experience of other IV biologics and leflunomide combination therapy) was not satisfactory. Thus, none of these data, nor any other, is sufficient to predict this choice.

Patients’ preferences

Patients’ preferences towards SC biologic treatment are summarized in Fig. 2. Personal and/or occupational problems related to going to the hospital for the IV infusions was the first reason mentioned by 72% of the patients, followed by the possibility of greater autonomy (38.7% of the patients) and economic considerations (21.5% of the patients). Technical problems related to the IV infusions were pointed out by 14% of the patients. Only 1.1% of the patients mentioned their fear of the hospital (Fig. 2).

Fig. 2
figure 2

Reasons guiding patients to switch to the subcutaneous (SC) route of administration of their treatment with abatacept or tocilizumab

Full size image

Worries about a lack of follow-up or medical assistance during the SC injections were the most common reasons for maintaining IV treatment, cited by 72.1 and 61.2% of the patients, respectively (Fig. 3). Other factors included maintaining social relationships with other patients developed at the hospital, mentioned by 40.5% of the patients, as well as the frequency of the treatments (monthly versus weekly), fear of adverse events and fear of SC injections, mentioned by 32.9, 27.7 and 17.9% of the patients, respectively. 3.6% of the patients preferred to pursue their IV treatment, because they already had another SC treatment (Fig. 3).

Fig. 3
figure 3

Reasons guiding patients to maintain the intravenous (IV) route of administration of their treatment with abatacept or tocilizumab

Full size image

To the question about who to contact in case of a problem with the SC biologic, 74.4% of the patients cited the hospital rheumatology department, 61.1% their general practitioner, 17.8% their private-practitioner rheumatologist and 3.3% their pharmacist. None of the patients mentioned the pharmaceutical company marketing the product.

In the SC group, 49% of the patients wanted to perform their injections themselves, 43.9% opted for a nurse and 7.1% for a family member.

Discussion

This study shows that when given the choice of the route of administration of their non-TNFi, almost half of the patients tend to prefer keeping with IV infusions rather than switching to SC injections. Indeed, 43.3% of the TCZ users wished to keep with their IV administrations, and 50% of the ABA users declined the SC route. Reasons for this choice included fear of the treatment itself and also worries about a lack of medical assistance and follow-up. Patients choosing the SC route pointed out the advantage of greater autonomy, and the problems related to repeated hospital visits for IV infusions as well as economic considerations.

A significant number of patients favoured maintaining the IV route. The proportion was similar to that reported in patients receiving TNF is in the Italian RIVIERA study, 50.2% of whom preferred infliximab over SC TNF blockers [10]. At the time of that study, SC TCZ was not yet available, and patients may have felt that the choice was only hypothetical, unlike ABA users who more largely rejected the available SC route. Despite greater experience with biologics several years after the RIVIERA study, our patients still expressed the need for reassuring medical assistance and hospital administration. Available evidence demonstrating the similar safety profile of these two routes of administration for both TCZ and ABA could be expected to be sufficient to reassure patients. This raises the question of how sufficiently this information is delivered [1, 2]. For the vast majority of patients, the rheumatology department remains their leading source of information, despite the fact that those receiving SC biologics attend more consultations with their private-practitioner rheumatologist. More active involvement of the private-practitioner rheumatologist in patient follow-up and decision-making concerning biologic treatments might help improve patient confidence in home treatments.

Our findings suggest that neither age nor being occupationally active influence significantly the choice for one or the other route of administration. We were unable to find any other study taking into account occupational activity status for comparison with our results, but there has been one report that patients in the occupational activity age range prefer SC TNF is over infliximab [11]. In this study by Chilton et al., younger patients favoured SC treatment, but nothing but a trend in the TCZ group corroborates this finding.

Irrespective of the disease involved, patients tend to prefer their current treatment due to a lack of knowledge about alternative routes of administration. Two oncology studies also addressed the issue, notably with HER2+ breast cancer treatments (trastuzumab), and found that 60% of women would choose the SC route for reasons of convenience [12, 13]. Unlike rheumatology patients who are used to their longstanding disease and clearly report seeing the hospital as a place of social interaction, cancer patients feel the need to break away from the hospital [12]. Interestingly, after having experienced SC trastuzumab, 84.5% of the patients initially favouring the IV route changed their mind for the SC route [13]. In our study, past experience of routes of administration had an impact on their further choice. Indeed, patients having experienced previous biologics with an IV route of administration further favoured keeping with their IV ABA. Furthermore, patients experiencing concomitant SC treatments (mainly methotrexate) were more inclined to choose the SC particularly when they had previously been treated with etanercept, the only TNF inhibitor having the same frequency of administration (weekly) than the one of ABA and TCZ. A cross-sectional survey of 500 TNFi users for various conditions showed that 89.9% of the patients undergoing SC TNFi treatment preferred the SC route, and 71.8% of the infliximab users favoured the IV route [14]. Similar results from a survey of 107 RA patients in Denmark showed that 85% of the patients receiving IV bDMARDs preferred the IV route, and 71% of the patients receiving SC TNFi favoured the SC route [15].

A majority of the patients still fear the switch to the SC route especially due to a lack of confidence in the tolerance profile of ABA and TCZ as well as their (in) ability to cope with potential reactions without medical assistance. Dispensing improved information could increase this confidence and encourage patients to choose the SC route with the objective of improving the medico-economic effectiveness of biologics.