Background/Purpose: Rheumatoid arthritis (RA) patients who fail to respond to methotrexate (MTX) can be subjected to an addition of biologic disease-modifying antirheumatic drug (bDMARD). Currently there are 4 modes of action of bDMARDs: inhibitors of tumor-necrosis-factor alpha (TNFi), of interleukin-6 (anti-IL-6), lymphocyte co-stimulation (anti-CTLA4) and B-cell directed therapy (anti-CD20). All bDMARDs show similar response rates at 6 months treatment, while differences in response between patients may relate to variable level of immunomodulation by each drug.

The apathogenic and highly prevalent Torque Teno Virus (TTV) is a potential novel candidate for functional monitoring of immune response. TTV in peripheral blood has been shown to mirror the immunocompetence of its host in patients with HIV, malignancies and solid organ transplantation.

Here we explore levels of TTV replication in patients receiving bDMARDs, and their association with clinical response to therapy.

Methods: The BIOBIO Study is a multicenter randomized open-label trial, including RA patients with insufficient response to MTX. Patients were randomized to either TNFi (infliximab, INF); anti-IL-6 (tocilizumab, TCZ); CTLA4-Ig (abatacept, ABA) or anti-CD20 (rituximab, RTX) in addition to MTX. Serum samples were collected at baseline and 3 months. RT-PCR was used to quantify TTV within these serum samples.

Results: TTV was measured in 95 RA patients randomized to INF (n=23), TCZ (n=22), ABA (n=27) or RTX (n=23); 77% were female, 72% were ACPA and 51% RF positive and 2/3 showed high disease activity. Median TTV at baseline was 4.3*10^4 c/ml (IQR: 7.4*10^3, 1.3*10^5) with no difference between the 4 treatment groups. After 3 months of treatment patients showed increase in TTV levels compared to baseline (INF: p=0.018; ABA: p=0.071; RTX: p≤0.001). There was no change in TTV for TCZ, and therefore omitted from further analyses on the association between TTV and treatment response.

TTV at 3 months after treatment was higher in patients achieving a SDAI85 response at month 6 (p=0.007; Figure 1A). Receiver operating characteristics outlined an area under the curve of 0.756 for prediction of SDAI85 at month 6 after treatment by TTV at month 3 (p<=0.001). A TTV level of 5.6×10^5c/ml at month 3 shows a 67% specificity, 81% sensitivity, relating to a +likelihood ratio of 2.6 (95%CI: 1.6-4.1) for SDAI85 prediction at month 6. Patients in the top tertiles of month 3 TTV had greater changes in SDAI (p=0.043), CDAI (p=0.022) and DAS28 (p=0.022) and higher SDAI85 response rates at month 6 (OR=5.85, CI: 1.4-24.5; p=0.016; Figure 1B. No patient below a TTV value of 2.7*10^4c/ml showed SDAI85 treatment response.

Conclusion:  Our data suggest that TTV level in the peripheral blood corresponds to and predicts clinical response to bDMARDs for RA, except for TCZ. Future prospective trials have to clarify the potential of TTV tailored bDMARD dosing.