Low-dose blinatumomab produced strong short-term responses in heavily pretreated RA, but its most provocative effect emerged later, when some patients once again responded to drugs that had previously failed.
A bispecific T-cell engager may be able to restore responsiveness to previously failed rheumatoid arthritis drugs, according to a small case series that raises the prospect of partially “resetting” multidrug-resistant disease.
The study of 15 patients with active multidrug-resistant RA found that low-dose blinatumomab produced rapid clinical improvement despite an exceptionally refractory cohort.
Every patient had failed methotrexate and at least three biologic or targeted synthetic DMARDs, while 10 of the 15 had also been refractory to rituximab.
But the most provocative finding emerged after the initial effect of blinatumomab wore off.
Fourteen of the 15 patients eventually flared, generally after about five months, and required biologic or targeted synthetic DMARDs to be reintroduced.
Six patients were given the same drug they had previously failed and three received another agent from a previously unsuccessful drug class.
Seven of the 15 achieved a clinical response lasting more than three months after retreatment with previously used drugs or drug classes, including JAK inhibitors, abatacept, and TNF inhibitors.
The researchers said RA appeared to become easier to manage after blinatumomab, with post-treatment flares resembling a “de-escalated phenotype” that was again responsive to standard therapies.
The Annals of the Rheumatic Diseases case series included patients with a median age of 55 years and median disease duration of 13 years. Baseline disease was highly active, with a median DAS28-CRP of 5.03 and CDAI of 28.
Blinatumomab is a CD3xCD19 bispecific T-cell engager that redirects T cells against CD19-positive B cells. The patients received a deliberately low-dose regimen, with cumulative doses of 77-189µg compared with doses of up to 784µg per treatment cycle in haematological indications.
Despite the low exposure, all 15 patients showed a rapid clinical response. By week 12, median tender joint count had fallen from seven to two and swollen joint count from five to one. Median DAS28-CRP fell from 5.03 to 2.4 and CDAI from 28 to eight.
Three patients achieved CDAI remission and nine achieved low disease activity or better. All achieved at least an ACR20 response, 11 reached at least ACR50, and eight reached at least ACR70.
After relapse, five patients received abatacept, five received JAK inhibitors, and three received TNF inhibitors. Two of the five abatacept-treated patients, four of the five JAK inhibitor-treated patients, and all three TNF inhibitor-treated patients had previously failed the corresponding mechanism of action.
“Notably, clinical efficacy of the reintroduced b/tsDMARDS was even seen in patients who had failed on the same class of drug before the introduction of blinatumomab,” the researchers wrote.
The finding comes with an important qualification. Secondary loss of response, rather than primary failure, had been the most common reason for those earlier treatment failures, and the researchers said the concept of pharmacological re-sensitisation was therefore most relevant to secondary resistance.
Supplementary data nevertheless suggested some heterogeneity, including primary failure in a patient later treated with abatacept and secondary failure in patients later treated with baricitinib and etanercept.
The biological findings offered a possible explanation for the apparent reset. Follow-up synovial biopsies were available from five patients, with CD20-positive B cells falling to zero in four. In contrast, lymph-node B cells were reduced but not fully depleted.
Ultrasound showed reduced synovitis, while FAPI-PET/CT in five patients showed an overall reduction in synovial fibroblast activation. Lower post-treatment FAPI activity in the hands was associated with a longer period before additional therapy was required.
Related
The researchers hypothesised that synovial B-cell depletion may disrupt local B-cell-fibroblast interactions, leaving subsequent inflammation more responsive to conventional treatments. They also raised T-cell exhaustion and a treatment-free “drug holiday” as alternative explanations for the apparent re-sensitisation.
Peripheral B-cell aplasia was relatively short lived, with a median recovery time of 21 days. Disease-associated autoantibodies also fell, with rheumatoid factor decreasing by an average 62.1%, anti-mutated citrullinated vimentin antibodies by 54%, and anti-CCP2 antibodies by 45% at three months, before rising again alongside disease activity.
The lack of durable drug-free remission may reflect incomplete depletion in secondary lymphoid organs.
The researchers said residual B cells remained in lymph nodes, and three patients did not achieve complete peripheral B-cell depletion, raising the possibility that higher doses or longer-acting T-cell engagers could produce more durable effects.
Safety signals included three cases of grade 1 cytokine release syndrome, five mild infections in four patients, and no cases of immune effector cell-associated neurotoxicity syndrome.
One patient with substantial pre-existing cardiovascular risk died from a cardiovascular event a year after treatment. The researchers said it was unlikely to be related, although it was not independently adjudicated.
They stressed that the findings remained exploratory. The study was uncontrolled, dexamethasone premedication may have contributed to early improvement, biopsy timing varied considerably, and within-class switching could itself explain some subsequent responses.
“Nevertheless, the sustained clinical benefit over months, the responsiveness to previously failed therapies, and the tissue-level changes (B-cell depletion in synovium, reduced FAPI uptake) are unlikely to be explained by dexamethasone exposure alone,” the researchers wrote.
“Biopsy timing heterogeneity (lymph node 29-97 days; joints 29-242 days, including 1 biopsy after 3 cycles at day 242) constitutes another limitation, as it might sample active B-cell repopulation rather than residual posttreatment tissue burden, and precludes definitive comparisons between depletion efficacy in synovium and lymph nodes.
“The qualitative depletion differences between tissues, while an intriguing hypothesis, should be interpreted in this light. Notably, the dynamics of lymphocyte infiltration, retention, residence, and egress may differ between the lymph node and the synovium.”
Rather than needing to induce permanent drug-free remission, a short course might eventually prove useful if it can convert otherwise multidrug-resistant disease back into a state that responds to established DMARDs, the researchers said.
“Future strategies must now balance dose escalation to achieve deep tissue depletion against the safety profile, but our findings suggest that total eradication of the pathogenic B-cell clone is not the only path to achieve long-term clinical responses,” they concluded.
