Background Anti-TNF biological therapy has revolutionised management of Ankylosing Spondylitis (AS) however, variability in treatment response within the AS population exists and may possibly be influenced by ethnicity.
Objectives To investigate the effect of ethnicity on treatment response to anti-TNF biologics by analysis of patient disease activity scores of severe AS patients by ethnicity.
Methods Severe AS patients defined by a persistent Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) of >4 and poor response to non-steroidal anti-inflammatory drugs (NSAIDs) were identified retrospectively from an anti-TNF therapy clinic database of patients treated by the Rheumatology Department between 2004 and 2014. Prescribed biologics were Infliximab, Etanercept, Adalimumab, Golimumab and Certolizumab. Patient clinical demographics and baseline BASDAI and spinal pain Visual Analogue Scale (VAS) scores from electronic records were collected pre-biologic treatment initiation and from 12 weeks post treatment follow-up. Data was analysed by non parametric descriptive statistics for changes in disease activity levels and compared by ethnicity. Adequate treatment effect was determined as a follow-up BASDAI score reduction with respect to the baseline of ≥2 units or 50% of the pre-treatment baseline score (BASDAI50) with a reduction of the spinal pain VAS by ≥2 cm. 99 of 256 available electronic records were complete at baseline and follow-up for BASDAI and corresponding VAS scores and were selected for analysis.
Results Overall 59/99 (59.6%) of the 99 selected records achieved an adequate response. Interrogation of records by database ethnicity classifications showed the following groups to be present in the data set:
49: Caucasian – English, Scottish, Welsh, Irish, Any Other White Background
44: South Asian – Bangladeshi, Indian, Pakistani
3: Other Asian
1: Other Mixed
By ethnicity an adequate response was achieved in:
36/49 (73.5%) of Caucasians,
20/44 (45.5%) of South Asians,
2/3 (66.7%) of the Asian other group,
1/2 (50%) of the Other group and 0/1 (0%) of the other mixed group.
Kruskal-Wallis multiple comparison testing showed statistical differences between baseline and follow-up mean BASDAI scores within Caucasian and South Asian ethnicities (Figure 1). Within the Caucasian group the baseline median=7.28 (IQR 6.3, 8.35), mean=7.31 (CI 6.92–7.70) versus follow-up median=3 (IQR 2, 5.1), mean=3.79 (CI 3.05–4.54). Within the South Asian subgroup the baseline median=7.25 (IQR 6.45, 8.06), mean=7.25 (CI 6.88–7.63) versus follow-up median=4.04 (IQR 2.45, 6.8) mean=4.62 (CI 3.8–5.44), both comparisons P<0.001. No significance was returned for comparisons between all ethnic groups at baseline and all at follow-up. “Other” group data subsets were too small to draw conclusions.
Conclusions Therapy was efficacious in 59.6% of patients analysed. By ethnicity statistically significant BASDAI score improvements were achieved within Caucasian and South Asian populations respectively. Baseline scores between these ethnicities were similar however the patient proportion within each of these groups that achieved an adequate response was higher among Caucasians 73.5% versus 45.5% of South Asians. This preliminary study may suggest as yet undiscovered ethnic influences affecting treatment response that merits further investigation.
Disclosure of Interest None declared