Biomarkers and pharmacoeconomics in rheumatoid arthritis: a good marriage

The histopathological analysis of synovial samples in the ADACTA study showed the evidence of four major phenotypes in the Rheumatoid Arthritis (RA) synovium: lymphoid, myeloid, low inflammatory, and fibroid. Each phenotype has a distinct underlying gene expression signature¹. The identification of two ICAM-1 and CXCL-13 biomarkers elucidated how patients with a myeloid phenotype had a better response in terms of ACR50 to Adalimumab (ADA), an anti- tumor necrosis factor α (TNFα) monoclonal antibody, as well as the patients with lymphoid phenotype had a better response to Tocilizumab (TCZ), an interleukin 6 receptor antagonist. Recent studies have well investigated MRP-8/14 (calprotectin) as a promising myeloid serum marker of inflammation and responsiveness to therapies in particular to anti-TNFα; the molecule is involved in the recruitment of inflammatory cells by interaction with endothelial cells and directly reflects the leukocyte counts in the inflamed joints rather than systemic inflammatory activity². Several studies demonstrated an association between high baseline titers of Rheumatoid Factor (RF) and a reduced response to anti-TNFα treatment^3,4, whereas there is no consensus about the usefulness of basal anti-citrullinated protein antibodies (ACPAs) levels as predictor of clinical response to anti-TNFα therapy⁵. Recent literature showed decreased RF titers during anti-TNFα treatment, mainly in patients who responded to treatment^6,7. Controversial data are reported on the decrease of the ACPA levels, mainly in responder patients⁸. Most of the studies concluded that RF positivity may predict a better response to Rituximab (RTX)⁹ and TCZ¹⁰ but not to Abatacept (ABT) in RA patients. ABT therapy was able to rescue immune function and led to an effective and safe clinical outcome, suggesting that RA patients, in whom anti-TNF-α failed, are immunologically prone to benefit from an agent targeting a different pathway^11,12. In an observational cohort of patients with RA treated with RTX, seropositive patients achieved significantly greater reductions in Disease Activity Score based on 28-joint count (DAS28) at 6 months compared to the seronegative patients^13,14. These concepts could be translated in terms of pharmacoeconomics with the advent of the biosimilars of Infliximab, Adalimumab, Etanercept, Rituximab which have shown a saving of 35–50%¹⁵. In the subanalysis of the AMPLE study 36.4% of patients in the highest quartile (Q4) of ACPA concentration who received ABT achieved Clinical Disease Activity Index (CDAI) ≤2.8 after one year of treatment. In the same study, 22.7% of Q4 ACPA patients treated with ADA achieved CDAI ≤2.8. After 2 years of treatment, 51.2% of the Q4 ACPA patients on ABT achieved CDAI ≤2.8 and 30.8% of those on ADA¹⁶. The data was confirmed using the Corrona RA registry in the United States (US) in real life. ACPA positive patients showed a better response to ACPA negative in terms of CDAI when treated with ABT¹⁷. In RA, the spectrum of seronegativity is gradually decreasing on the basis of the novel autoantibodies correlating with the disease activity and joint erosivity: anti carbamylated proteins (anti-CarP) antibodies¹⁸, anti-malondialdehyde-acetaldehyde (anti-MAA) antibodies¹⁹, anti-BRAF antibodies²⁰, anti-peptidyl-arginine deaminase 3/4 (PAD3/PAD4) antibodies²¹, anti-14-3-3ƞ antibodies whose titles appear to have been changed during treatment with TCZ²² and anti-CEP-1 antibodies²³. In the June 2020 issue of Journal of Medical Economics, Park et al.²⁴ evaluated a budget impact model in choosing ABT therapy in 100% of patients with APCA positivity. The ACPA test cost increased by $20,684 and treatment cost increased by $1,60,467 totaling an overall budget increase of $181,151. Testing RA patients with ACPA to increase the use of ABT led to a small increase in total budget costs. The differences in rheumatoid phenotype biomarkers highlighted in the ADACTA study¹ can also influence the pharmacoeconomic analysis. Among the 163 patients treated with TCZ and 162 patients treated with ADA, mean total drug and administration costs per patient over 24 weeks were $18,290.60 and $25,623.10, respectively. Mean drug and administration costs per each clinical response achieved were lower with TCZ than with ADA (DAS28 < 2.6: $45,868 vs $2,44,174; ACR20: $28,127 vs $51,887; ACR50: $38,720 vs $92,244; ACR70: $56,253 vs $1,43,136). The total hospital days were 32 days with TCZ and 43 days with ADA; mean hospital costs per patient were $484.50 with TCZ and $651.10 with ADA²⁵. Cost-consequences data reported in an Italian cost scenario, have been evaluated starting from the AMPLE study²⁶. The data have showed efficacy in favor of ABT compared to ADA . The data in favor of ABT was due to a difference with lower costs for adverse events (–€237,246 or -€237per patient)²⁷. Data from the AMPLE study was also evaluated in a scenario in Germany, Spain, Italy, US and Canada. Patients were classified by the presence or absence of ACPA and their response to ABT or ADA. The cost per response in ACPA positive patients was in favor of ABT compared to ADA according to ACR20, ACR90 and HAQ-DI. The remission costs according to DAS28were in favor of ABT in ACPA negative patients according to CDAI (Clinical Disease Activity Index) and SDAI (Simplified Disease Activity Index) in ACPA positive patients²⁸. In another study patients starting treatment with ABT or ADA plus methotrexate were evaluated after 6 months²⁹. The continuation or the discontinuation of the therapy was evaluated with European League Against Rheumatism treatment response; the progression of the disease and disability was evaluated with Health Assessment Questionnaire Disability Index score. Quality-adjusted life-years (QALYs) and incremental cost per QALY gained were calculated by baseline ACPA groups (Q1, 28–234 AU/mL; Q2, 235–609 AU/mL; Q3, 613–1045 AU/mL; and Q4, 1060–4894 AU/mL). Incremental cost per QALY for ABT (vs. ADA) was the lowest in the high ACPA titer group (Q4, £6200/QALY), followed by the next lowest titer group (Q3, £26,272/QALY)²⁹. In another study, the response as the first biological line of treatment was evaluated in patients with inadequate MTX response in a German scenario. The response in terms of remission was evaluated with DAS28 < 2.6. The study also considered the direct costs and the drug costs. The response to ABT or ADA was evaluated in terms of cost-effectiveness analysis as a function of the presence of ACPA in patients. Treatment sequences starting with ABT resulted in lower costs per day in remission (mean 330 €/day, range 328–333 €/day) compared to sequences starting with ADA (mean 384 €/day, range 378–390 €/day)³⁰. Pharamacoeconomic review of published data shows a favorable effect of ABT in ACPA positive patients³¹. The right therapeutic choice based on biomarkers can reflect on some therapeutic strategies, an early remission can determine a dose escalation and a tapering towards drug free remission which leads to further savings³². Recent data applied to biosimilars have shown that the therapeutic tapering strategy can be cost-saving even compared to the non-medical switch from the originator to the biosimilar when it is complicated by adverse events or loss of efficacy³³. Personalized medicine is still bethroted in RA but there are the premises for present and future biomarkers and pharmacoeconomics to fly to the altar in a future marriage.

Transparency

Declaration of funding

No funding to report.

Declaration of financial/other interests

No declarations of interest.

Author contributions

All authors contributed equally.

Acknowledgements

None reported.