Elevated baseline power Doppler discriminates an RA subgroup highly responsive to therapy

Rheumatology key message

• RA patients with low-power Doppler US score, high DAS28 and multiple co-morbidities have less modifiable disease activity.

Sir, Recent systematic literature reviews and expert panel recommendations by the ACR and the EULAR support the use of musculoskeletal US (MSUS) in monitoring RA disease activity [1, 2]. MSUS is inexpensive and enables multiple point-of-care assessments. Power Doppler US (PDUS) score measures synovitis and may be useful in assessing response to therapy, where studies have reported suppression of PDUS signal after administration of RA treatment [3, 4]. This study evaluates the predictive value of baseline PDUS measures on response to drug by the DAS28 and clinical disease activity index (CDAI) in RA. Few studies have evaluated the association between MSUS and co-morbidities, and we investigate this concept as well. This report describes the results of a pilot 12 month open-label s.c. abatacept study of 25 RA patients naive to biologics.

The study was approved by the University of California, Los Angeles institutional review board, registered on clinicaltrials.gov (NCT01299961), and all patients signed an informed consent form. Inclusion criteria were as follows: ACR 1987 RA diagnostic criteria; age ≥18 years; stable DMARDs; no prior exposure to biologics; DAS28/ESR >3.2; prednisone ≤10 mg; and total PDUS ≥1 for at least two MCP joints. Patients completed detailed questionnaires regarding their demographics, function and co-morbidities. MSUS assessment of PDUS was performed at baseline, 3 weeks, 3, 6 and 12 months. A GE Logic E9 machine with ML6-15 probe (GE Healthcare) was used. Seven joints were scanned by MSUS of the most affected side (wrist, MCP joint 2/3, PIP joint 2/3 and MTP joint 2/5) according to Backhaus et al. [5]. PDUS was scored semi-quantitatively according to published consensus definitions [6]. The clinical assessor was blinded to the US data and vice versa. In addition, when scoring images, the ultrasonographer was blinded to the sequence of the visits and patient.

All 25 patients enrolled in the study completed at least 3 months of therapy with s.c. abatacept. Nineteen of the 25 patients completed the 12 month visit. Six patients dropped out for the following reasons: lack of efficacy (two patients); adverse events [three patients (recurrent oral ulcers, erythema nodosum, chronic obstructive pulmonary disease exacerbation), no serious adverse events]; and lost to follow-up (one patient).

Patients were separated into two groups based on median baseline PDUS of complete cases: patients with baseline PDUS <5 and patients with baseline PDUS ≥5 (Table 1). Interestingly, the changes between baseline and 12 months for DAS28/ESR, CDAI and the HAQ disability index (HAQ-DI) for the PDUS ≥5 group were significantly greater than seen in the PDUS <5 group, where patients in the PDUS ≥5 group experienced more than twice the magnitude of improvement (DAS28 change of 1.1 vs change of 2.7, CDAI change of 12.8 vs change of 28.0, and HAQ-DI change of 0.3 vs change of 0.8). Baseline PDUS was significantly correlated with change in DAS28/ESR at 12 months (DAS28/ESR ρ = 0.65, P < 0.01). We also performed a multivariate linear regression for change in DAS28/ESR as the outcome, with baseline PDUS and baseline DAS28/ESR as the predictors. Baseline PDUS was almost significantly associated with change in DAS28/ESR (P = 0.06), after accounting for baseline DAS28/ESR (P = 0.52). The number of co-morbidities was significantly higher in those with PDUS < 5 (P = 0.02). Lastly, we found that lower baseline PDUS was significantly correlated with higher co-morbidities (ρ = −0.48, P = 0.04).

Table 1.

Baseline power Doppler US score and disease activity measures over 12 months

	PDUS <5 cohort			PDUS ≥5 cohort			P-value difference in change
	Baseline, n = 10	Month 12, n = 10	Change	Baseline, n = 9	Month 12, n = 9	Change
DAS28/ESR, mean (s.d.)	6.0 (0.61)	4.8 (1.1)	1.1 (1.21)*	6.6 (0.78)	3.8 (1.2)	2.7 (0.96)**	**
CDAI, mean (s.d.)	33.7 (6.5)	19.4 (10.0)	12.8 (13.4)*	39.5 (5.8)	11.7 (7.9)	28.0 (7.5)**	**
HAQ-DI, mean (s.d.)	1.0 (0.51)	0.7 (0.6)	0.3 (0.62)	1.5 (0.85)	0.6 (0.79)	0.8 (0.52)**
PDUS, mean (s.d.)	2.6 (1.37)	2.5 (1.9)	0.07 (2.08)	8.3 (2.14)	1.9 (2.3)	6.3 (1.55)**	**
GSUS, mean (s.d.)	7.4 (2.28)	8.3 (2.37)	−0.9 (3.46)	11.9 (2.92)	8.4 (3.2)	3.5 (2.45)**	**
ESR, mean (s.d.)	36.4 (14.79)	31.7 (22.21)	4.7 (16.68)	55.6 (29.10)	28.8 (19.32)	23.1 (33.39)
OA on X-rays, %	57			36
Number of co-morbidities, mean (s.d.)	5.1 (3.7)			1.8 (1.3)*
BMI, mean (s.d.)	28.4 (7.9)			24.4 (6.5)

*P < 0.05, **P < 0.01. CDAI: Clinical Disease Activity Index; GSUS: grey-scale synovial hypertrophy; HAQ-DI: HAQ-disability index; PDUS: power Doppler US Score.

There are MSUS studies reporting early changes or baseline PDUS predicting a variety of RA outcome measures. We cite two studies that use baseline PDUS to predict other 1 year outcomes. Ellegaard et al. [7] demonstrated that MSUS colour fraction quantification of the wrist in 109 RA patients starting a TNF inhibitor predicted which patients would remain on therapy in 1 year (P = 0.02). Kamishima et al. [8] evaluated 29 RA patients starting i.v. tocilizumab and performed MSUS at baseline, 2 and 5 months. The authors reported that the quantitative PDUS may be predictive of structural damage progression at 1 year.

There are limitations to our study. This is a small open-label study, and we performed post hoc analyses to determine whether baseline PDUS can predict the 12 month response to therapy. As a result of the limited sample size, we were unable to perform multivariate regression models with more than two variables. Yet, in this small cohort, we were able to demonstrate a significant difference between patients with baseline PDUS >5 and <5.

In conclusion, we found that those patients with severe disease activity had a wide range of baseline PDUS and that patients with high baseline PDUS had a more robust response to therapy. Perhaps high PDUS may identify patients with an RA phenotype more amenable to biologic intervention. In addition, patients with low baseline PDUS and high DAS28/CDAI may have other factors (co-morbidities, obesity, OA, fibromyalgia, etc.) besides RA disease activity that resulted in their high scores. We posit that PDUS rather than DAS28 identifies those patients with modifiable disease activity. Thus, PDUS assessment may aid in treatment decisions concerning whether escalation or switching therapy is needed in RA patients with multiple co-morbidities. Additional studies are required to validate these findings.

Funding: This work was supported by a Bristol Meyer Squibb investigator initiated grant to the University of California, Los Angeles, California; National Institutes of Health/National Center for Advancing Translational Science University of California Los Angeles CTSI Grant Number UL1TR000124.

Disclosure statement: V.K.R. received an investigator initiated grant from Bristol Meyer Squibb (conduct and analyses were performed independent of Bristol Meyer Squibb). She has also consulted for Crescendo Biosciences and received an investigator initiated grant from Genentech and Pfizer. All other authors have declared no conflicts of interest.