Background/Purpose: A major challenge in the biologic era is to predict clinical response. A large variability in the level of TNFα expression has been recognized which may influence the outcome of TNF antagonism. Thus far, treatment decisions are solely based on clinical disease activity. In this way, only 40% of rheumatoid arthritis (RA) and spondyloarthritis (SpA) patients achieve a clinically important response (ACR50 or ASAS40). We hypothesized that in vivo assessment of TNFα by scintigraphy with 99mTc-radiolabeled anti-TNFα antibodies may be helpful to optimize and monitor the effect of TNFα blockade. This technique may facilitate ‘evidence-based biological therapy’ by in vivo measurement in inflamed joints of a target cytokine, prior to therapeutic administration of a biologic.

Objective: Predict response to therapy by baseline immunoscintigraphy before starting anti-TNF treatment in active RA and SpA patients.

Methods: Certolizumab pegol (CZP)was radiolabeled with Tc99m. Whole body images and static images of hands and feet were acquired immediately following administration, 4-6 hours and 24 hours post injection. Immunoscintigraphic findings were scored as either negative (no or faint uptake) or positive (clear uptake). All patients were treated with CZP for 6 months and standard clinical assessments were performed. Statistical analysis for the joint based response (tender and swollen) was based on a logistic regression model including patient as random effect to accommodate for clustering of the joints in the patient and using the odds ratio (OR) as summary statistic.

Results: 20 patients were included (RA n=5, SpA n=15), scanned and treated according to the protocol. Images 4-6 hours post-injection yielded the best discriminatory results of radiolabeled uptake. Immunoscintigraphy was a good predictor for a joint being swollen at baseline as only 2.2% of the scintigraphic negative joints were swollen compared to 63.5% of the scintigraphic positive joints.The mid scintigraphic evaluation had a significant predictive value on the tender (OR=18.7, P<0.0001) and swollen joint count (OR=44.4, P<0.0001) at baseline, implying that clear uptake increases the odds of having a problematic joint. After 24 weeks, the odds of joints remaining painful was significantly smaller in joints with a positive scintigraphic result as compared to joints with a negative scintigraphic result (OR=0.41, P=0.04):32 out of 114 (28.1%) painful negative scintigraphic joints at baseline remained painful at week 24, whereas only 18 out of 58 (13.8%) painful positive scintigraphic joints at baseline remained painful at week 24. No significant results were found for the swollen joints.

Conclusion: Baseline scintigraphic detection of TNFα with radiolabeled anti-TNFα has a significant predictive value at baseline and after 24 weeks treatment with CZP on the tender joint count. Tender joints not identified by immunoscintigraphy respond to a lesser degree to anti-TNF treatment, potentially indicating another biological reason for a painful joint. Therefore TNF-immunoscintigraphy could offer a new tool to identify good clinical responders to biological treatment.

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/baseline-scintigraphic-detection-of-tnfa-as-a-predictor-of-therapy-response-after-treatment-with-certolizumab-pegol-in-rheumatoid-arthritis-and-spondyloarthritis-patients/