Biological therapy in RA: cost-effectiveness

Rheumatoid arthritis (RA) is a chronic disease that has a large impact on patient quality of life and generates great financial and social costs. Starting in the initial phases, it affects daily activities, not only in terms of physical aspects, but also social, psychological and financial aspects. The prevalence of RA is estimated to be between 0.02% and 4.6% of the overall population and more than 2% in people over 60 years of age.¹ This pathology is associated with high levels of comorbidity and reduced life expectancy in five to ten years.² RA patients place a significant burden on health services, similar to coronary heart disease.³ Long-term medical treatment is required to control their disease, and about one-third of patients will require surgery, resulting in substantial indirect costs, estimated at over €65,000 over two years per biotreated patient.⁴ The treatment of RA involves the use of non-steroidal anti-inflammatories, traditional disease-modifying antirheumatic drugs (DMARDs) or a combination of these and physical therapy in their initial stages. However, if there is no response to these treatments, there are tumour necrosis factor alpha (TNF-α) inhibitors (infliximab (IFX), etanercept (ETN), adalimumab (ADA), golimumab and certolizumab) or biological response modifiers (abatacept and tocilizumab) are used alone or in combination with methotrexate (MTX). The use of biological drugs have demonstrated greater efficacy than conventional therapies, but tripled the direct costs of RA.⁵ A systematic review of clinical data for the established TNF blockers and newer biologic response modifiers concluded that in moderate-to-severe RA there were no significant differences in achieving ACR50 response between treatments.⁶

Immunogenicity refers to the ability of protein drugs to provoke an immune response. The immune system can detect small differences in the 3D structure between an introduced foreign molecule and a native protein, leading to the production of anti-drug antibodies (ADAbs).⁷

ETN is a fusion protein comprising a recombinant form of the human TNF soluble receptor p75 linked to the fragment crystallisable (Fc) portion of the human G1 immunoglobulin (IgG1).⁸ This structure is different from other anti-TNF drugs: IFX is an anti-TNF chimeric human-murine IgG1 monoclonal antibody;⁹ ADA and golimumab are recombinant human monoclonal antibodies that bind specifically to TNF.^10,11 Tocilizumab is a humanised IgG1 monoclonal antibody against the human interleukin-6 receptor.¹² Abatacept is a fusion protein composed of the Fc region of IgG1 and the extracellular domain of cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4); inhibiting the costimulation of T cells¹³ and certolizumab pegol is a PEGylated Fab’ fragment of a humanised TNF inhibitor monoclonal antibody¹⁴ (Figure 1).

Antibodies against the biological agents may develop over time.¹⁵ These anti-TNF drug antibodies can interfere with the ability of the antibody to bind to TNF and are referred to as neutralising antibodies.¹⁶ The presence of antibodies against anti-TNF monoclonal antibodies confers the risk of discontinuation of treatment, diminishes the efficacy, increases the adverse events in RA and a risk of development of hypersensitivity reactions in all immune-mediated inflammatory diseases.¹⁷ Neutralising antibodies have not been associated with ETN.¹⁴ By contrast, neutralising antibodies are associated with IFX and ADA and may occur even more frequently when MTX is not given concomitantly, as MTX appears to prevent their development.¹⁵ However, concomitant use of certain DMARDs such as MTX may maintain efficacy and prolong drug survival by reducing ADAb formation to anti-TNFs. There is no evidence of cross-reactivity among different drugs: immunogenicity (the development of specific ADAbs to one TNF inhibitor) does not seem to affect the effectiveness of subsequent anti-TNF agents; therefore, switching to another drug of the same class might be effective in patients who have developed ADAbs to a TNF inhibitor.¹⁸ 

Randomised controlled trials report ADAbs in a small proportion of patients on golimumab and certolizumab, however, the numbers of ADAb-positive patients were insufficient to determine a clear association with impaired therapeutic response.^19,20 Further studies are required to investigate the relationship between clinical response and immunogenicity with golimumab and certolizumab pegol.

TNF-α inhibitors (IFX, ETN, ADA, golimumab and certolizumab) and biologic response modifiers (abatacept and tocilizumab) have transformed the RA treatment paradigm with their superior ability to slow disease progression, reduce joint damage and functional disability when added to DMARDs. Due to the relatively high cost of biological therapies, economic evaluations have emerged in the past decade in attempts to determine the cost efficacy of biological therapy in RA patients. Economic evaluations such as cost-effectiveness analyses (CEAs) and cost-utility analyses (CUAs) compare the costs and consequences (outcomes) of two or more possible courses of action, to help decide which course of action provides best value for money. Several evaluations had demonstrated the favourable pharmacoeconomic profile of biological therapy in RA patients in different countries and perspectives, both in monotherapy or combined with DMARDs and as first-line or after DMARD treatment.^21–25

However, the majority of the ‘classic’ pharmacoeconomic analyses of biological therapies are based on the efficacy data obtained under strictly controlled clinical trials. The manifest difficulty in translating clinical trials into clinical practice has been described,²⁶ where the drug adherence, patient’s condition, comorbidities, concomitant and previous medications play major roles. National Institute for Health and Care Excellence guidance also states the importance of considering all aspects of costs, including administration cost, in determining which anti-TNF therapy to use, “Treatment should normally be initiated with the least expensive drug (taking into account administration costs, required dose, and product price per dose)”.²⁷

In clinical practice, deviations in administered doses have been observed over time, resulting in major deviations in direct associated costs of biological drugs.

These real-world deviations had to be taken into account when studying cost-effectiveness profiles, changing from a theoretical to a practical approach.

RA is associated with a substantial economic burden, both in direct and indirect costs.²⁸ Predicting the real annual costs associated with biological therapies is complex due to differences in dosing schedules, treatment gaps, switching between biological drugs, treatment administration and dose modifications over time. For this reason, and due to the diverse biologic treatment options available to treat RA, real-world data regarding the cost and efficiency of these therapies may have implications in the economics of RA management, and are of major importance for physicians and healthcare systems, aiming to prescribe the most efficient treatments. Several recent studies determined the cost of biological therapy in real-world settings.

Schabert et al estimated the annual cost per treated patient from the payer perspective for ETN, ADA and IFX in adults with RA, psoriasis, psoriatic arthritis or ankylosing spondylitis. Of the 30,107 patients included in the analysis: 16,280 RA patients received ETN (47.6%), followed by ADA (29.7%) and IFX (22.7%). The annual TNF-blocker cost per treated RA patient was lower for ETN ($16,787) followed by ADA ($19,308) and then IFX ($22,939). For new and continuing RA patients, the same rank order for cost (ETN<ADA<IFX) was observed. The authors concluded that based on real-world drug utilisation data in adult patients with RA, ETN had a lower cost per treated patient than ADA or IFX.²⁹

Ramírez-Herráiz et al assessed mean doses of ADA, ETN and IFX in RA patients under clinical practice, correlated their doses with clinical effectiveness and evaluated the cost implications in two Spanish hospitals. Among patients clinically controlled, 36.73% of ADA cases and 52.83% of those treated with ETN remained in reduced doses. By contrast, IFX required an increased dose to achieve clinical control in 76.32% of cases, reporting mean annual costs of ADA of €11,963, of €10,095 in IFX patients and of €9595 in ETN-treated patients.³⁰

Bonafede et al estimated the annual biologic response modifier cost per treated patient with RA, psoriasis, psoriatic arthritis and/or ankylosing spondylitis receiving etanercept, abatacept, adalimumab, certolizumab, golimumab, infliximab, rituximab or ustekinumab. Included in the study were 69,349 patients; 38,026 RA patients were included. ETN was the most commonly used (41.6% of patients), followed by ADA 27.2%, IFX 19.2%, abatacept 6.6%, rituximab 3.5%, golimumab 1.2% and certolizumab <1%. The annual cost per RA-treated patient was $19,295 for abatacept, $19,537 for rituximab, $21,329 for ETN, $21,761 for golimumab, $21,819 for certolizumab, $23,078 for IFX and $23,351 for ADA. This study showed that among patients with RA, the cost per treated patient for abatacept, certolizumab, golimumab or rituximab ranged from 90–102% relative to ETN, whereas patients treated with IFX and ADA ranged from 102–108% relative to ETN.³¹

Borrás-Blasco et al evaluated the economic impact associated with a biological therapy prioritisation protocol supported by a multidisciplinary Hospital Commission of Biological Therapies for RA patients in a Spanish tertiary hospital. In this protocol, ETN was selected as preferred therapy for both naive and previously biotreated patients based on its cost-effectiveness advantages. A total of 186 patients were included in the analysis. In the pre-protocol period (2009–2010), total expenses were increased by €110,000, up to €1,761,000 in 2010 (€11,362 patient/year).

After protocol implementation, total expenses decreased by €53,676 on the 2010–2011 period, and €149,200 on the 2011–2012 period. On the 2010–2011 period the cost of biological therapy per patient-year decreased €355 (€11,007 patient/year) and additional €653 (up to €10,354 patient/year) by 2012, with a cumulative effect of the protocol implementation of €1008 per patient-year. In the pre-protocol period, the annual cost/patient was €10.812 with ETN, €10.942 with IFX, €12.961 with ADA and €12.739 with ABA. By 1 January 2013, the annual cost per patient was €9469 with ETN, €10,579 with IFX, €11,117 with ADA and €13,540 with ABA. The authors concluded that the creation of a Multidisciplinary Commission of Biological Therapies is key to rational management of RA patients and optimisation of resources, allowing €200,000 to be saved after a two-year efficiency protocol implementation.³²

Real-world data in RA patients treated with biological therapies allow observing the effects of real biological drug dosing in clinical practice, rather than recommended dosing, which should be need to take into consideration when estimating the cost to the payer for TNF-blocker or biologic response modifiers treatment. Moreover, it would be desirable to determine real-world cost data of biological drugs not only in RA patients, but in other inflammatory diseases (ankylosing spondylitis or psoriatic arthropathy). This real-world drug cost information would help payers, rheumatologists and formulary managers to make more efficient decisions when treating patients with biological therapies.