Cost effectiveness of TNF-alpha inhibitors in RA

With a prevalence of approximately 1%, rheumatoid arthritis (RA) represents a major burden to society. Individuals can be affected at almost any age, but perhaps especially within the ‘prime of life’, where work and home productivity would normally be the highest. Not only does RA affect joints – and often extremely aggressively – it also affects the whole body: with involvement of internal organs that, in the end, leads to a significantly shortened life span if not treated effectively.(1) On the background of such major, systemic and progressive disease, it has been frustrating that, for such a long time, treatment has proven ineffective. Perhaps the biggest reason for this was the lack of understanding of basic pathophysiological processes that underlie this disease.

Advances in understanding the basic science mechanisms that mediate the inflammation in RA have led to the discovery of the key role of the proinflammatory cytokine, tumour necrosis factor alpha (TNFa) in mediating RA. Identification of this cytokine as a potential therapeutic target has led to the rapid development of biologic drugs that could inhibit this cytokine. There are now five licensed TNFa inhibitors (TNFi) for use in this condition: etanercept (Enbrel), adalimumab (Humira), infliximab (Remicade), golimumab (Simponi) and certolizumab (Cimzia).

There is no doubt that these drugs are highly effective, especially in combination with methotrexate. Many clinical trials have demonstrated clear efficacy, both in patients who have failed to respond to standard disease modifying anti- rheumatic drugs (DMARDS), and perhaps even more effectively, in newly diagnosed patients as initial therapy.(2)

There is also no doubt that there is the potential for adverse events. The term ‘tumour necrosis factor’ was coined because, in the laboratory, this cytokine was able to spontaneously kill some cancer cells. Therefore it was initially feared that the chronic inhibition of TNF may lead to the development of new, or recurrence of old cancers. This has not been borne in practice (perhaps with the rare exception of non-melanomatous skin cancers in some individuals). Similarly, in health, TNFa is produced in response to infection and there is no doubt that chronically inhibiting this cytokine leads to an excess of infections.(3) Whereas these are usually easily treated standard bacterial infections, patients taking a monoclonal antibody TNFi (for example, infliximab) have a particular risk of reactivation of latent tuberculosis. This phenomenon is not seen to the same extent with etanercept, which has a significantly different structure.

Clinical trials are essential to gain an understanding of the potential effectiveness of a new agent and also to assess the frequency of common adverse events, for licensing. However, the type of patient recruited to a clinical trial does not necessarily match the typical patient found in a clinic. There is, therefore, a great need to gain a better understanding of the ‘real life’ usage of new drugs. To address this, many countries have set up registries for TNFi (and indeed other biologics) in rheumatic diseases. A good example of this is the British Society for Rheumatology Biologics Registry.(3) This is a large register of patients with RA that includes not only those taking a TNFi but also a control group of patients with severe disease, who, for various reasons, stay on standard treatment without biologics. This is important to help dissect the potential role of a drug, rather than the disease itself. Such registries can also provide important clues as to the cost effectiveness of an agent in real life.

As physicians, we consider value to be proportionate to cost and benefit. While there is no doubt that the acquisition cost of TNFi is high, we must remember that that their benefit in a serious systemic condition such as RA is also extremely high. Within the context of any health economy, where resource is not limitless, difficult decisions often need to be taken. Who can get what drug? These are crucial questions that must be answered according to the merit of the argument, not just who can shout the loudest. It is hard, for example, to compare the value of a high cost drug that may prolong life for just a couple of months, with another high cost drug that may dramatically increase quality of life for decades. The UK National Institute for Health and Clinical Excellence (NICE) attempts to make these difficult decisions by a detailed process, independent of the pharmaceutical industry, that seeks to determine cost effectiveness according to quality-adjusted life years (QALY).(4) NICE has considered TNFi for use in RA on a number of occasions, consistently ruling that they are cost effective, recommending their use for patients who have severe disease activity and who have failed to respond to standard DMARDs such as methotrexate.(5) Comparable bodies in other countries have come to similar conclusions but, unlike the UK, recommend (and reimburse) these agents in patients with even less disease activity and a shorter duration of unresponsive disease – with good data underpinning this.

There is no doubt that all five currently available TNFi are highly effective drugs. There is, however, ongoing debate suggesting that their adverse events may be subtly different.(6) However, it is important to not draw too firm a conclusion from relatively preliminary comparisons.

It is certainly interesting to note that, although each TNF blocker binds very selectively to the same target molecule, there are some important clinical differences. For example, the monoclonal antibody TNFi, typified by infliximab, are effective not only in RA but also in Crohn’s disease. Whereas the p75 TNF receptor construct, etanercept, works well in RA, but is ineffective in Crohn’s disease. In addition, there is a tendency for anti-drug antibodies to form against TNF-binding regions in the monoclonal antibody TNFi, interfering with the active binding site for TNFa and correlating with reduction in clinical effectiveness – something which does not seem to occur with etanercept, where anti-drug antibodies are directed at sites distant to those responsible for TNF binding. However, for idiosyncratic reasons, some patients may fail to respond to etanercept, but then respond to a monoclonal antibody TNFi, making the situation much more complicated.

We have investigated real life usage of TNFi in a pragmatic study recently published.(7) We reported that potentially expensive incrementation of biologic drugs (either increasing the dose or decreasing the interval between doses) was found in as many as a third of patients taking infliximab (mouse–human chimeric antibody), some 15% of patients on adalimumab (humanised antibody) but only a handful of patients on etanercept (human p75 receptor construct). This allowed us to calculate the total cost of care for such patients including acquisition cost of drug, infusions when required, cost of training patients to self inject and all lab and hospitalisation costs incurred over the period of a year. We found that, while acquisition cost was, as expected, a major driver of total cost, the need to increase dosing of biologic (dose creep), or other agents (for example, add DMARDs or steroid), contributed significantly to overall costs. Similar findings have recently been reported from a North American study.(8)

As we now have access to a number of biologic drugs, these emerging real life observational studies may help inform us about optimal drug usage to maximise value for money. All things otherwise equal, we certainly would not wish to increment dosing for a biologic drug in cases of secondary inefficacy if we can otherwise swap the agent within the same class or, better, move to a drug with a different mechanism of action.

It is clear that not only does RA represent a major burden to society, but that with effective therapy the induction and maintenance of disease remission is eminently achievable – the target for therapy today. Given this, the benefits of effective treatment not only to the healthcare economy, but also to society as a whole, is very high. Within this context, the TNFi have consistently been found, by many different bodies, to be cost effective.

The future looks even more exciting. The development of biosimilar drugs, with the potential to drive down at least a proportion of acquisition cost, may allow usage of such drugs to change: potentially allowing access to patients with earlier disease or less severe disease activity. Indeed, the paradigm may shift even further so that, as for cancer, RA may be treated initially with combination biologic/small molecule therapy and the high cost biologic drug stopping when remission has occurred and the low cost (likely methotrexate) drug continuing as remission-maintenance therapy. This, together with the exciting potential for personalised or stratified medicine – targeting appropriate treatment more efficiently and yet more cost effectively – makes the future look especially exciting and totally unrecognisable from the experience of most healthcare personnel involved in the care of RA only 10–15 years ago.

1. Vandenbroucke JP, Hazevoet HM, Cats A. Survival and cause of death in rheumatoid arthritis: a 25-year prospective follow up. J Rheumatol 1984;11:158–61.
2. Geiler J, Buch M, McDermott MF. Anti-TNF treatment in rheumatoid arthritis Curr Pharm Des 2011;17(29):3141–54.
3. The British Society of Rheumatology Biologics Register. www.rheumatology.org.uk/bsr_biologics_register/default.aspx (accessed 14 May 2012).
4. National Institute for Health and Clinical Excellence (NICE). www.nice.org.uk/ (accessed 14 May 2012).
5. NICE. Technology Appraisal 130 Rheumatoid arthritis – adalimumab, etanercept and infliximab: guidance. www.nice.org.uk/TA130 (accessed 14 May 2012).
6. Singh JA et al. Adverse effects of biologics: a network meta-analysis and Cochrane overview. http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD008794.pub2/abstract (accessed 14 May 2012).
7. Moots RJ et al. Differences in biologic dose-escalation, non-biologic and steroid intensification among three anti-TNF agents: evidence from clinical practice. Clin Exp Rheumatol 2011;29:26–34.
8. Scabert VF et al. Disability outcomes and dose escalation with etanercept, adalimumab, and infliximab in rheumatoid arthritis patients: a US-based retrospective comparative effectiveness study. Curr Med Res Opin 2012;Jan 11 [Epub ahead of print].