Adalimumab in the therapy of uveitis in childhood

Short abstract

“Adalimumab is more effective” …?

Worldwide, around one million patients have been treated with tumour necrosis factor (TNF)‐α antagonists (etanercept, infliximab or adalimumab) for rheumatoid arthritis, juvenile rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis and inflammatory bowel disease. Treatment results in substantial improvement in the signs and symptoms of arthritis (inhibition of progression of radiographic joint damage in psoriatic arthritis, resumption of growth in juvenile idiopathic arthritis (JIA) and attenuation of spinal inflammation in ankylosing spondylitis) as well as improved functional status and quality of life.¹ The use of TNF antagonists in adult uveitis has also been promising in small series.²,³,⁴ Recently, TNF antagonists were also used in paediatric uveitis⁵,⁶ and studies have shown the superiority of infliximab to etanercept in juvenile uveitis.⁷ Vasquez‐Kobian et al⁵ released their results in October 2006 regarding the use of adalimumab in juvenile uveitis. Similarly in this issue, Biester et al⁸ report that the use of adalimumab in refractory juvenile uveitis has good visual outcome (see pages 319). However, since the approval of TNF antagonists, concerns have been raised regarding their safety especially in children. We describe the differences between the three biologic therapies regarding modes of action, visual results, side effects and economic impact on health, and review preliminary evidence suggesting the potential superiority of adalimumab in JIA uveitis.

Adalimumab is a fully human immunoglobulin G1 monoclonal antibody that binds with high affinity and specificity to TNF and neutralises the biological activities of this cytokine by blocking its interaction with the p55 and p75 cell surface TNF receptors. Given the known role of TNF in uveitis, the efficacy and safety of adalimumab in the treatment of uveitis in JIA was analysed by Biester et al.⁸ Chronic asymptomatic anterior uveitis occurs in 10–30% of patients with JIA, usually within 4 years of the onset of arthritis, and is associated with a high frequency of non‐specific low‐titre antinuclear antibodies. Long‐term visual outcome in JIA‐associated uveitis has been described as poor, with one third of patients developing substantial visual impairment and 10% becoming blind.⁶,⁹ Most patients with JIA are already on non‐steroidal anti‐inflammatory drugs because of their arthritis and the drug of choice for polyarthritis is frequently methotrexate. According to several recent reports, low‐dose oral methotrexate is effective in the treatment of chronic non‐infective uveitis.⁹ However, if more effective treatment is needed, systemic glucocorticosteroids and/or low‐dose cyclosporine are added. In patients with refractory chronic uveitis, treatment with a TNF antagonist is indicated.⁶

The three TNF antagonists (etanercept, infliximab and adalimumab) had similar efficacy in rheumatoid arthritis, but that does not appear to be the case with uveitis, where infliximab is more effective than etanercept in both childhood⁷ and adult uveitis.⁴,¹⁰ Both adalimumab and infliximab were effective in reducing uveitis flares in patients with spondylarthropathy but etanercept was not.¹¹ Although infliximab was an effective short‐term immunosuppressive agent with clear benefit, the rate of serious toxic effects was unexpectedly high in a prospective study.² Adalimumab was effective in controlling 80.8% of paediatric uveitis cases,⁵ three cases of Behcet uveitis resistant to infliximab³ and spondyloarthropathy‐related uveitis.¹¹ Ocular response to adalimumab in JIA uveitis occurred within the first 2–6 weeks of therapy.⁵ Arthritis response to adalimumab was much faster with 10 (22.2%) of 45 patients achieving a clinical response within 24 h of dosing.¹² In this issue, Biester et al⁸ found retrospectively that adalimumab was well tolerated and decreased the relapse rate in JIA uveitis cases previously unresponsive to combined therapies (including infliximab), with minimal side effects (absence of anaphylactic reaction or infection).

To explain the therapeutic discrepancy between TNF‐α antagonists, several hypotheses have been put forward relating to differences in molecular structure, mechanism of action, pharmacokinetics (kinetics, route and frequency of administration, type of TNF binding) and pharmacodynamics (apoptosis induction, TNF immunoprecipitation) (table 1).¹,¹³ Etanercept and infliximab have different binding characteristics, with infliximab and adalimumab binding to both soluble and membrane‐bound TNF, while etanercept binds primarily to soluble TNF. These differences in binding may manifest as differing effects on complement activation and apoptosis. Etanercept and infliximab also have different pharmacokinetic profiles that may influence their activity. Because infliximab is administered as bolus injections every 4–8 weeks, there is great variability in concentrations over time (high peaks separated by periods of low levels, with the high peaks possibly contributing to greater tissue penetration), whereas etanercept is administered subcutaneously twice weekly and adalimumab subcutaneously once every 2 weeks. Adalimumab therapy was generally well tolerated¹ and appeared to be less immunogenic than infliximab. The incidence of antibodies against infliximab increased from approximately 45% after the first infusion to 61% after the fifth infusion. Importantly, the duration of the clinical response was shortened in the presence of anti‐infliximab antibodies. Concomitant methotrexate therapy was associated with a reduced incidence of antibody development. Twelve per cent of patients treated with adalimumab alone were antibody‐positive compared to 1% of patients treated with adalimumab plus methotrexate.¹⁴ Unlike infliximab, no demyelinating disorders or lupus‐like syndromes occurred with adalimumab administration. Infliximab was associated with a higher risk than adalimumab of requiring intensification of immunosuppressive therapy than the other anti‐TNF agents and a significant dose escalation over time. Analysis of rheumatic disease activity indicated a reduced therapeutic response to infliximab after the first 6 months of treatment, suggestive of acquired drug resistance.¹⁵

Table 1 Biochemical and clinical profiles of anti‐TNF agents¹,¹³.

Agent (company)	Mechanism of action	Administration (half life)	Status	Average dose (duration) varies with severity, disease and body weight	Adverse effect	Monitoring
Etanercept	TNF‐α inhibitor	SC twice weekly	FDA approved for	25 mg SC 2×/week	Injection site	CBC
(Amgen Wyeth)	TNF receptor	(4 days)	RA, JA, AS	(24 weeks) or 50 mg	reaction
			psoriasis, psoriatic	SC 2×/week (24 weeks)
			arthritis	in adult; 0.4 mg/kg
				2×/week in paediatric
Infliximab (Centocor)	TNF‐α inhibitor	IV infusion	Phase III trials for	3 mg/kg (0, 2,	Anaphylactic	Baseline PPD;
	chimeric antibody	>120 min	psoriasis, FDA	6 weeks) or	reactions;	liver enzymes
		(9 days)	approved for RA,	5 mg/kg	tuberculosis
			psoriatic arthritis,	(0, 2, 6 weeks)
			Crohn's disease
Adalimumab	TNF‐α inhibitor‐	SC once weekly/	FDA approved for	80 mg week 0, 1	Tuberculosis;	Baseline PPD;
(Abbott Laboratories)	human antibody	every other week	RA phase III trials	then 40 mg q week	injection site	liver enzymes
		(15 days)	for psoriasis,	(12 weeks)	reaction
			psoriatic arthritis	or q 2 weeks in adult;
				20–40 mg q 2 weeks
				in paediatric

AS, ankylosing spondylitis; CBC, complete blood count; FDA, Food and Drug Administration; IV, intravenous; JA, juvenile arthritis; PPD, purified protein derivative; q week, once weekly; q 2 weeks, once every 2 weeks; RA, rheumatoid arthritis; SC, subcutaneous.

There is a risk of reactivation of granulomatous diseases with TNF antagonists, especially tuberculosis, and measures should be taken to detect and treat latent tuberculosis infections.¹⁶ Preliminary data suggest that anti‐TNF therapy may be safe in chronic hepatitis C. However, TNF antagonists have resulted in reactivation of chronic hepatitis B if not given concurrently with antiviral therapy. Solid tumours do not appear to be increased with anti‐TNF therapy. Variable rates of increased lymphoma risk have been described with anti‐TNF therapy compared with the general population, although no increased risk was found compared with a rheumatoid arthritis population. Trials with TNF antagonists in advanced heart failure have shown trends towards a worse prognosis, and TNF antagonists should therefore be avoided in this population. Rare cases of aplastic anaemia, pancytopenia and vasculitis have been described with anti‐TNF therapy. Optic neuritis was reported in six cases (four cases with etanercept and two with adalimumab).¹⁷,¹⁸

Despite their clinical effectiveness, these agents are expensive and the annual cost of treatment ranges from US$12 000 to US$16 000. A recent study reported the mean direct costs of treating patients with TNF inhibitors to be almost three times higher (US$19 016) than treating those not receiving these agents (US$6164).¹⁶ Costs to the patient and insurance/formulary coverage were perceived by rheumatologists as major barriers in prescribing TNF inhibitors.¹⁶ Different routes of administration resulted in different coverage and reimbursement policies in the US and elsewhere. Medicare (and several private insurance companies around the world) covers infliximab, which is administered intravenously, while etanercept and adalimumab are administered subcutaneously and are not covered. In a national survey of practicing United States rheumatologists, adalimumab was perceived to be the least problematic drug, maybe because of its easy dosing schedule (40 mg every other week as compared to 25 mg twice a week for etanercept, and a loading dose at baseline, 2, 6 and every 8 weeks for infliximab) which might result in better outcomes in patient with rheumatoid arthritis.¹⁶ A retrospective study of health plan costs related to rheumatoid arthritis revealed that etanercept was associated with lower drug and outpatient costs to the health plan than infliximab and adalimumab.¹⁹ Compared with etanercept, infliximab was related to 55% higher post‐index rheumatoid arthritis‐related monthly total health care costs paid by the health plan, and adalimumab had 12% higher costs.

We should not forget that periocular long‐acting depot corticosteroids remain one of the most clinically efficient cost‐effective local therapies for adult and some childhood uveitis.²⁰ They can be administered without general anaesthesia in selected cooperative subjects after discussion with the parents and the patient. Unlike the situation in adults, we apply excess topical anaesthesia and use a wire eyelid retractor and forceps to hold Tenon's capsule to ensure immobilisation of the globe during injection. We need to be cautious about anti‐TNF therapy in children (as the long‐term effects are not known) and reserve it for non‐responsive uveitis.

In conclusion, adalimumab has so far demonstrated some preliminary advantages over infliximab, including subcutaneous route of administration, no need for hospitalisation, dosage every 2 weeks, being less immunogenic and less expensive, and having a more a protracted therapeutic response in uveitis.⁵,⁸ However, we need large long‐term prospective randomised series with direct comparison between infliximab and adalimumab. As demonstrated by Biester et al,⁸ the group of TNF inhibitors has enlarged our repertoire of effective treatment modalities in uveitis, in addition to immunosuppressive drugs.