Research ArticleSpondyloarthritis
Open Access

Simplified Ankylosing Spondylitis Disease Activity Score (SASDAS) Versus ASDAS: A Post Hoc Analysis of a Randomized Controlled Trial

Emilce E. Schneeberger, Gustavo Citera, Dario Ponce de Leon, Annette E. Szumski, Kenneth Kwok, Mariel Cutri and Maxime Dougados
The Journal of Rheumatology October 2022, 49 (10) 1100-1108; DOI: https://doi.org/10.3899/jrheum.211075

Abstract

Objective To compare the Simplified Ankylosing Spondylitis Disease Activity Score (SASDAS) against the Ankylosing Spondylitis Disease Activity Score (ASDAS) for measuring and categorizing disease activity using data from the EMBARK trial (ClinicalTrials.gov: NCT01258738), a randomized controlled trial of etanercept (ETN) for axial spondyloarthritis (axSpA).

Methods Patients with early active axSpA received ETN 50 mg once weekly (n = 106) or placebo (PBO; n = 109) for 12 weeks in a double-blind manner; they then received open-label ETN for 92 weeks. For this analysis, ASDAS–C-reactive protein (CRP) and SASDAS-CRP were calculated at baseline, week 12, and week 24. The SASDAS was calculated by the linear addition of the ASDAS components without adjustment.

Results A very strong correlation, as determined by the Spearman correlation coefficient, was observed between the ASDAS and SASDAS for continuous variables at baseline and during treatment. For pooled categorical data at baseline, the SASDAS placed 69.9% of patients in the same disease categories as the ASDAS but overestimated for 17.8% of patients and underestimated for 12.2% of patients. A similar pattern was seen postbaseline. Cohen weighted Embedded Image statistics for all individual and pooled treatments and timepoints (0.54-0.73) reflected moderate to substantial agreement. The capacity to differentiate between treatments (ie, ETN and PBO/ETN) was higher with the ASDAS (effect size −0.74, 95% CI −1.03 to −0.46) compared with the SASDAS (effect size −0.51, 95% CI −0.79 to −0.23), but sensitivity to change was generally similar.

Conclusion A very strong correlation between the SASDAS and ASDAS was observed when considering continuous variables; however, moderate to substantial agreement was observed for categorical data, and the SASDAS classified a lower proportion of patients as being in the inactive and low disease activity categories.

Key Indexing Terms:

The Ankylosing Spondylitis Disease Activity Score (ASDAS) is considered the gold-standard assessment tool for assessing axial spondyloarthritis (axSpA). The 2017 treatment recommendations for axSpA state that the ASDAS is the preferred measure for defining the treatment target,1 and it has several advantages over the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI). For instance, the ASDAS is more sensitive, with a better ability to discriminate between high and low disease activity states using a range of cut-off values.2 When the ASDAS is used as an eligibility criterion to decide whether patients should commence anti–tumor necrosis factor (anti-TNF-α) treatment, patients are significantly more likely to show improvements than if the BASDAI is used.3 A higher ASDAS score is also a baseline predictor of both response to anti-TNF-α and determining long-term continuation of therapy,4 and it better predicts structural damage.5 The ASDAS is also associated with various biomarkers of the disease and has good predictive validity.5,6 In addition, the use of the ASDAS was recommended over the BASDAI in the last update of the international guidelines for therapeutic targets in axSpA.1 However, use of the ASDAS may be limited in daily practice, as it requires a scientific calculator or an electronic application for its calculation.

A simple alternative to the ASDAS, called the Simplified ASDAS (SASDAS), consists of the sum of the components of the ASDAS in 2 versions: one with the erythrocyte sedimentation rate (ESR) as part of the calculation, and the other with C-reactive protein (CRP) levels as part of the calculation. It does not require a calculator and is quicker and easier to use in daily clinical practice. The use of the SASDAS has been validated against the ASDAS and/or other standard assessments in previously published manuscripts.7-11 Current daily practice involves using CRP to evaluate the magnitude of biological inflammation, as it is considered a more specific biomarker than ESR, and high CRP levels better predict radiographic progression, response to nonsteroidal antiinflammatory drugs (NSAIDs), and TNF-α inhibition.12-16

This post hoc analysis of data from the EMBARK trial compares the CRP versions of the ASDAS and SASDAS for measuring and categorizing disease activity at baseline and after treatment.

METHODS

Patients and treatment. EMBARK (ClinicalTrials.gov: NCT01258738) was a multicenter, double-blind, placebo (PBO)-controlled, 2-period randomized study that evaluated the efficacy of etanercept (ETN) of 50 mg once weekly for the treatment of patients with active nonradiographic axSpA. Details of the study design have been published previously.17,18

Eligible participants were randomly assigned in a 1:1 ratio to receive either ETN 50 mg once weekly plus a stable background NSAID or PBO plus a background NSAID for 12 weeks. Patients completing the 12-week, controlled, double-blind period entered a 92-week open-label treatment period with ETN 50 mg once weekly and a background NSAID. Clinical assessments included continuous ASDAS-CRP, Bath Ankylosing Spondylitis Functional Index (BASFI), Bath Ankylosing Spondylitis Metrology Index (BASMI), and magnetic resonance imaging (MRI) Spondyloarthritis Research Consortium of Canada (SPARCC) sacroiliac joint (SIJ) scores. Additionally, assessments of patient-reported health-related quality of life (QOL) were administered: the Ankylosing Spondylitis Quality of Life (ASQoL) score, the 36-item Short Form Health Survey (SF-36) mental component summary (MCS) score, and the SF-36 physical component summary (PCS) score. Full details of all the clinical assessments have been published previously.18

Ethics and consent. Ethics approval, the final protocol, any protocol amendments, and informed consent forms for the EMBARK study were reviewed and approved for use by each of the study centers by a duly constituted institutional review board or independent ethics committee (Supplementary Table S1, available with the online version of this article). There was no specific primary ethics committee for the study. Not all 48 sites used reference numbers; therefore, the study protocol number (B1801031) was used as reference. Informed consent forms from all participants were obtained before study activities were initiated.

Endpoints. In the analysis reported here, ASDAS-CRP was calculated at baseline, week 12, and week 24 as follows: (0.121 × total back pain [BASDAI question 2]) + (0.110 × patient global assessment of disease activity) + (0.073 × peripheral pain/swelling [BASDAI question 3]) + (0.058 × duration of morning stiffness [BASDAI question 6]) + (0.579 × Ln(CRP mg/L + 1)). BASDAI questions and patient global assessment of disease activity were scored on a visual analog scale ranging from 0 to 10 cm. The SASDAS was calculated simply by the linear addition of the 5 ASDAS components without adjustment, with CRP measured in mg/dL.

Construct validity: comparison of the ASDAS and SASDAS as continuous variables. For the evaluation of continuous outcomes, the ASDAS and SASDAS, the change in ASDAS and SASDAS, and their standardized versions were analyzed as continuous variables by each timepoint and by pooled treatment. To directly compare the SASDAS and ASDAS, the variables were standardized and placed on the same scale by treatment arm and timepoint. The mean and SD were calculated for the SASDAS and ASDAS separately for each treatment arm and timepoint, and the standardization formula was applied to each SASDAS and ASDAS value by subtracting the corresponding mean and dividing by the corresponding SD.

Spearman correlation coefficient. To evaluate whether the assessment tools were correlated, Spearman correlation coefficients and 95% CIs were calculated for ASDAS vs SASDAS values at baseline, at week 12, and at week 24, as well as for change in ASDAS vs change in SASDAS values.

Intraclass correlation. Intraclass correlation coefficients (ICCs) and 95% CIs were calculated for standardized assessments. ICCs differ from Spearman correlations: ICCs consider both correlation and the difference in values, whereas Spearman correlations consider only the correlation. For the Spearman and ICC coefficients, values of ≥ 0.30 to < 0.50, ≥ 0.50 to < 0.70, ≥ 0.70 to < 0.90, and ≥ 0.90 roughly correspond to low, moderate, high, and very high correlation or agreement, respectively.

Bland-Altman plots. Bland-Altman plots of the difference between standardized ASDAS and standardized SASDAS values vs the average of standardized ASDAS and standardized SASDAS values were generated, allowing for the investigation of possible agreement between measurement error and the true value. Good agreement is indicated in a Bland-Altman plot by a small mean difference, low dispersion around the mean difference line, and no apparent correlation between the average and the difference. Increasing or decreasing trends seen in Bland-Altman plots indicate that the level of agreement differs across the range of instrument scores.

External validity: evaluation of the ASDAS and SASDAS against other disease assessment measures. The ASDAS and SASDAS were also analyzed as continuous variables against other disease assessment measures (ie, BASFI, BASMI, and MRI SPARCC-SIJ) and QOL measures (ie, ASQoL score and SF-36 MCS and PCS scores) at each timepoint and by individual and pooled treatments. The BASFI was selected for use in external validation because at early stages of disease, functional impairment is related to disease activity. Similarly, the BASMI was selected because limitation of spinal mobility is related to disease activity and not structural damage at this stage. Therefore, we expected a high correlation between the ASDAS, the SASDAS, and these variables.

Spearman correlation coefficients and 95% CIs were calculated between each measure vs the ASDAS and SASDAS, and the 2 correlation coefficients were compared using the Steiger Embedded Image statistic, which tests for a significant difference between the 2 Spearman correlation coefficients.

Responsiveness to longitudinal change and discrimination between treatments. The between-treatment effect size was used to compare the capacity of the SASDAS vs the ASDAS to differentiate between treatment arms at the primary endpoint visit (ie, week 12). To this end, the treatment effect size and 95% CIs compared the treatment (ie, ETN vs PBO/ETN) discriminant capacity of the standardized ASDAS vs the standardized SASDAS, as well as the standardized change for ASDAS and SASDAS, from baseline to week 12. The effect size was defined as the subtraction of the mean change in PBO/ETN from the mean change in ETN and division by the pooled SD, based on Cohen d.

The within-treatment effect size compares the sensitivity to longitudinal changes within each treatment arm for the SASDAS and ASDAS. For this analysis, the effect size for change from baseline and 95% CI (ie, change in the ASDAS vs change in the SASDAS) was calculated for each treatment. This compared the sensitivity to change from baseline for each pair of variables using mean change divided by mean baseline.

Evaluation of categorical outcomes: classification accuracy. The SASDAS and ASDAS were analyzed as categorical variables for disease activity at each timepoint, as well as by treatment and pooled treatment. As stated by the 2018 nomenclature for disease activity,19 disease activity was categorized as inactive (ASDAS < 1.3; SASDAS ≤ 7.8), low (ASDAS ≥ 1.3 and < 2.1; SASDAS ≥ 7.9 and < 13.8), high (ASDAS ≥ 2.1 and < 3.5; SASDAS ≥ 13.9 and < 27.6), or very high (ASDAS ≥ 3.5; SASDAS ≥ 27.6). To determine the percentage of absolute agreement between 2 indices, 4 × 4 cross-tabulation tables of SASDAS disease categories by ASDAS disease categories were created. SASDAS disease categories were also analyzed by ASDAS split by disease category as < 2.1 (ie, the target to achieve in clinical practice) or ≥ 2.1 (ie, denoting high or very high disease activity).

Cohen weighted Embedded Image statistics were calculated to test for agreement between SASDAS and ASDAS categories. The Embedded Image result can be interpreted as follows: 0 or lower, no agreement; 0.01 to 0.20, no to slight agreement; 0.21 to 0.40, fair agreement; 0.41 to 0.60, moderate agreement; 0.61 to 0.80, substantial agreement; and 0.81 to 1.00, almost perfect agreement.20

In addition, the distribution of each of the other disease assessment measures (ie, BASFI, BASMI, and MRI SPARCC-SIJ) or QOL measures (ie, ASQoL score and SF-36 MCS and PCS scores) was analyzed with boxplots by ASDAS vs SASDAS disease categories at each timepoint and by treatments.

RESULTS

Baseline demographics. In the initial double-blind period of EMBARK, 106 patients were randomized to ETN and 109 were randomized to PBO. Of the 215 patients, the mean age at baseline was 32.0 years, 130 (60.5%) were male, 154 (71.6%) were HLA-B27 positive, and 174 (80.9%) had MRI-confirmed sacroiliitis. The mean BASDAI was 6.0, indicating moderate to severe disease, and the mean duration of disease symptoms was 2.5 years.17

Assessing correlation between the continuous ASDAS and the continuous SASDAS. There were strong linear relationships between the standardized ASDAS and the standardized SASDAS (continuous data) for pooled treatments at baseline (Figure 1A), and for ETN (Figure 1B) and PBO (Figure 1C) at week 12.

Figure 1.
Figure 1.

Standardized ASDAS vs standardized SASDAS for (A) baseline pooled data, (B) week 12 ETN, and (C) week 12 PBO. ASDAS: Ankylosing Spondylitis Disease Activity Score; ETN: etanercept; PBO: placebo; SASDAS: Simplified Ankylosing Spondylitis Disease Activity Score.

Spearman correlation coefficients indicated strong correlations between the ASDAS and SASDAS (≥ 0.82) for pooled treatments and each treatment group at all timepoints (ie, baseline, week 12, and week 24; Supplementary Table S2, available with the online version of this article). There was also good agreement between the 2 scales when observing changes from baseline while on treatment, with Spearman correlation coefficients of 0.92 and 0.91 for the change from baseline to week 12 and week 24, respectively, for pooled treatments. Similar values were observed for individual treatment groups (ie, nonpooled data; Supplementary Table S2).

ICCs: continuous data. For pooled treatments, the standardized ASDAS vs standardized SASDAS ICCs were 0.85 (95% CI 0.81-0.89) at baseline, 0.90 (95% CI 0.87-0.92) at week 12, and 0.90 (95% CI 0.87-0.93) at week 24. Similar values were observed for individual treatment groups (ie, nonpooled data).

Bland-Altman: continuous data. Bland-Altman plots were generated to display the difference between standardized ASDAS and standardized SASDAS values vs the average of standardized ASDAS and SASDAS values, along with horizontal lines to denote the mean differences and their 95% limits of agreement. At all the timepoints, the mean difference was 0, indicating that there was no bias, on average, of one assessment over the other. At baseline, for pooled data (Figure 2A), the data points were not scattered evenly around the mean difference horizontal line of 0 but were dispersed above the line, indicating larger standardized ASDAS values compared with standardized SASDAS values. Below the line, the data points had an increasing negative trend across small negative to 0 average values, indicating bias, but there was no trend for average values greater than 0, indicating no bias. For PBO at week 12, the data points appeared evenly scattered around the mean (Figure 2B). For ETN at week 12 (Figure 2C) and pooled data at week 24 (Figure 2D), the interpretation of the Bland-Altman plot was similar to that at baseline, with a smaller increasing negative trend up to an average mean equal to 0.

Figure 2.
Figure 2.

Bland-Altman plots for (A) baseline pooled treatment; (B) week 12 PBO; (C) week 12 ETN; and (D) week 24 pooled treatment. ETN: etanercept; PBO: placebo.

Correlation of the ASDAS and the SASDAS vs other disease assessment measures. BASFI, BASMI, and MRI SPARCC-SIJ data were chosen as other disease assessment measures for comparison (Table 1). For the comparison of the BASFI vs the ASDAS and SASDAS, Spearman correlation coefficients indicated moderate correlations at baseline (0.56 for BASFI vs ASDAS and 0.61 for BASFI vs SASDAS for pooled treatments). Stronger correlations were observed at weeks 12 and 24 (Table 1). Similar trends were observed in the individual treatment arms. Based on the Steiger Embedded Image statistic, significant differences between correlation coefficients were seen only at week 12 for ETN treatment alone (P = 0.002) and at week 24 for pooled treatments (P < 0.001), for the ETN-alone arm (P < 0.001), and for the PBO/ETN arm (P = 0.01).

View this table:
Table 1.

Correlation of the ASDAS and SASDAS against other disease assessment and QOL measures.

Spearman correlation coefficient values were low for the BASMI vs either the ASDAS or SASDAS, with values ranging from 0.02 to 0.25 across all treatments and timepoints, and the difference between the 2 correlation coefficients was not statistically significant at any timepoint (Table 1). This was also true for the MRI SPARCC-SIJ, with Spearman correlation coefficient values ranging from −0.11 to 0.32 across all treatments and timepoints (Table 1). These data reveal that only the BASFI showed a modest correlation with the ASDAS and SASDAS when analyzed as continuous variables.

Correlation of the ASDAS and SASDAS against QOL measures. The ASQoL, SF-36 MCS, and SF-36 PCS scores were assessed in relation to ASDAS and SASDAS measures. For pooled treatments and in the individual treatment arms, the correlation coefficients generally reflected a moderate correlation for the ASQoL score with the ASDAS and SASDAS at all timepoints (Table 1).

Spearman correlation coefficient values were low for the SF-36 MCS score vs either the ASDAS or SASDAS, with values ranging from −0.46 to −0.17 across all treatments and timepoints (Table 1). However, for the SF-36 PCS score, Spearman correlation coefficient values ranged from −0.70 to −0.47 across all treatments and timepoints, reflecting a moderate correlation (Table 1).

Between-treatment effect size and within-treatment effect size. The capacity to discriminate between treatments as evaluated by the between-treatment effect size (ie, treatment effect size) was numerically higher with the ASDAS compared with the SASDAS. The treatment effect sizes for ETN vs PBO at week 12 were −0.74 (95% CI −1.03 to −0.46) and −0.51 (95% CI −0.79 to −0.23) for the ASDAS and SASDAS, respectively (Figure 3A).

Figure 3.
Figure 3.

Comparison of (A) ASDAS vs SASDAS in their capacity to each differentiate between treatments evaluated by the treatment ES, and (B) comparison of the sensitivity to change from baseline for ASDAS vs SASDAS using the ES formula of change divided by baseline. ASDAS: Ankylosing Spondylitis Disease Activity Score; ES: effect size; ETN: etanercept; PBO: placebo; SASDAS: Simplified Ankylosing Spondylitis Disease Activity Score.

The within-treatment effect size (ie, effect size for sensitivity to longitudinal changes) for the ASDAS vs SASDAS was similar for the ETN arm but was slightly larger for the SASDAS (ie, more sensitive) for PBO at week 12 and PBO/ETN at week 24 (Figure 3B).

Agreement between the categorical ASDAS and categorical SASDAS. At baseline, the evaluation of the categorical SASDAS placed 69.9% of patients (0.9% + 6.1% + 39.0% + 23.9%) in the same disease categories as did the ASDAS (Table 2). However, the SASDAS placed slightly more patients in the high or very high disease activity categories compared with the ASDAS. At baseline, 38 out of 213 patients (17.8%) with low or high disease activity as per the ASDAS were categorized as having higher disease activity by the SASDAS (low: 20/213, 0.9%; high: 18/213, 8.5%). In addition, 26 out of 213 patients (12.2%) were categorized as having lower disease activity by the SASDAS than by the ASDAS (low: 3/213, 1.4%; high: 4/213, 1.9%; very high: 19/213, 8.9%; Table 2). A similar pattern was seen postbaseline.

View this table:
Table 2.

ASDAS by SASDAS categories at baseline.

Cohen weighted Embedded Image statistics ranged from 0.54 to 0.73 for all individual and pooled treatments and timepoints, reflecting moderate to substantial agreement.

Distribution of standard disease assessments by ASDAS and SASDAS disease categories. At baseline, disease activity, defined by the ASDAS or SASDAS, increased in severity in line with increasing median BASFI scores (Supplementary Figure S1A, available with the online version of this article). This trend was observed for all treatments and timepoints. There was no trend for an increase in disease activity, defined by the ASDAS or SASDAS, in line with increasing median BASMI (Supplementary Figure S1B) or MRI SPARCC-SIJ scores (Supplementary Figure S1C). The number of patients with inactive disease was very low (n = 2 for ASDAS and n = 4 or 5 for SASDAS), so it is not possible to draw any meaningful conclusions from the group of patients with inactive disease.

Distribution of QOL measures by ASDAS and SASDAS disease categories. At baseline (pooled data), disease activity, defined by ASDAS or SASDAS disease categories, increased in severity in line with increasing (ie, worsening) median ASQoL scores (Figure 4A). This trend was observed for all treatments and timepoints. At baseline (pooled data), disease activity, defined by the ASDAS or SASDAS, increased in severity in line with decreasing (ie, worsening) median SF-36 MCS scores (Figure 4B) and SF-36 PCS scores (Figure 4C). This trend was observed for all treatments and timepoints.

Figure 4.
Figure 4.

Pooled treatments by baseline ASDAS/SASDAS disease categories: (A) baseline ASQoL, (B) baseline SF-36 MCS, and (C) baseline SF-36 PCS scores. The bottom and top of each box represent quartile 1 (Q1) and quartile 3 (Q3), respectively. The lower whisker is the maximum value of minimum value or (Q1 − 1.5 × IQR), while the upper whisker is the minimum value of maximum value or (Q3 + 1.5 × IQR). ASDAS: Ankylosing Spondylitis Disease Activity Score; ASQoL: Ankylosing Spondylitis Quality of Life; MCS: mental component summary; PCS: physical component summary; SASDAS: Simplified Ankylosing Spondylitis Disease Activity Score; SF-36: 36-item Short Form Health Survey.

DISCUSSION

Despite the simplicity of the SASDAS compared with the ASDAS, the data presented here show overall agreement between these 2 measures for classifying patients with active disease. For categorical data, Cohen weighted Embedded Image statistics ranged from 0.54 to 0.73 for all individual and pooled treatments and timepoints, reflecting moderate to substantial agreement. The Spearman correlation coefficient values of approximately 0.80 to 0.90 for all treatments and timepoints represent a strong correlation between the ASDAS and SASDAS for continuous variables when assessing disease severity at baseline and postbaseline. These data on the agreement between SASDAS-CRP and ASDAS-CRP are in line with previous analyses, such as a report of 84 patients with axSpA (New York 1984 and Assessment of SpondyloArthritis international Society 2009 criteria) from the ESPAXIA cohort (Estudio de Espondiloartritis Axial IREP Argentina)9 and a study including 397 consecutive patients with axSpA showing a strong correlation between the 2 scoring systems.7 A study of 254 patients with axSpA attending the rheumatology outpatient department of Sir Ganga Ram Institute, New Delhi, India, also reported good correlation for SASDAS-CRP with ASDAS-CRP,21 and a report involving 97 patients found moderate agreement between SASDAS-CRP and ASDAS-CRP.10

In the exploratory study reported here, an analysis of ICC also found high concordance between the 2 scales when considering both the correlation and value differences. When observing changes from baseline while on treatment, there was also good agreement between the 2 scales. The discriminant capacity evaluated by the treatment effect size was numerically higher with the ASDAS compared with the SASDAS, but the sensitivity to change was similar. Bland-Altman plots revealed good agreement between the 2 scales, indicated by a small mean difference and by low dispersion around the mean difference line, and no apparent correlation between the average and the difference.

We also tested the validity of the SASDAS and ASDAS against other widely used disease assessment measures, such as the BASMI, the BASFI, and the MRI SPARCC-SIJ. The BASFI showed a moderate correlation with the ASDAS and SASDAS when analyzed as continuous variables. When observing ETN-treated patients only, the difference between the 2 correlation coefficients (ie, BASFI vs ASDAS and BASFI vs SASDAS) was statistically significant at week 12 and week 24, but not at baseline. When analyzing PBO/ETN-treated patients and pooled treatment data, the difference between the 2 correlation coefficients for the BASFI vs ASDAS and the BASFI vs SASDAS was statistically significant at week 24 only. Generally, the SASDAS appeared to show a slightly stronger correlation with the BASFI than the ASDAS, and the difference between the 2 Spearman correlation coefficient values became significant as the correlation between the BASFI and SASDAS became stronger postbaseline. When considering disease categories, again, only the BASFI showed a good degree of agreement with the SASDAS and ASDAS, with median BASFI scores increasing in line with increasing SASDAS or ASDAS disease severity categories. The association between disease activity and functional capacity in patients with axSpA has been reported in several studies and, to a lesser extent, with spinal mobility.22-25 Changes in the BASMI and MRI SPARCC-SIJ are usually smaller and slower,26-28 which could explain the lack of agreement of the ASDAS and SASDAS with these 2 measures. In addition, the agreement between spinal inflammation and spinal mobility measures has been shown to be fair29; a possible explanation is that inflammation in axSpA can be transient, with periods of remission and relapse and with little influence on axial mobility. Our data also reveal a trend for increasing ASDAS or SASDAS values in line with worsening ASQoL, SF-36 MCS, and SF-36 PCS scores.

Unweighted indices are frequently used as simplified measures to assess disease activity in other rheumatic diseases, including the Simplified Disease Activity Index in rheumatoid arthritis and the Disease Activity Index for Psoriatic Arthritis.30,31 The SASDAS, an unweighted index, has previously been compared with the ASDAS in patients with ankylosing spondylitis.9 Here, we have compared the SASDAS with the ASDAS, the weighted gold-standard scale for assessing disease activity in patients with active nonradiographic axSpA, with longitudinal follow-up allowing comparison of the performance of these measures in a manner not previously seen. While results indicate strong correlations between the 2 measures, the SASDAS tended to categorize more patients into higher disease categories and had a somewhat diminished capacity to discriminate between treatments. While the SASDAS is not intended to replace the ASDAS as the standard measure of disease activity in axSpA, it may be useful as a simplified measure in some situations. To our knowledge, this analysis is the first in attempting to simplify the evaluation of axSpA according to the ASDAS composite index.

ACKNOWLEDGMENT

Medical writing support was provided by David Sunter, PhD, CMPP, and Lorna Forse, PhD, of Engage Scientific Solutions, and was funded by Pfizer.

Footnotes

  • This study was sponsored by Pfizer.

  • EES provides services such as speaker, investigator, and medical advisor for AbbVie, Amgen, BMS, Genzyme, Janssen, Eli Lilly, Novartis, and Pfizer. GC provides services such as speaker, investigator, and medical advisor for AbbVie, Amgen, BMS, Gema Biotech, Genzyme, Janssen, Eli Lilly, Pfizer, and Roche. DPdL, KK, and MC are employees of Pfizer and own stock in Pfizer. AES is an employee of Syneos Health, which was contracted by Pfizer for the development of this publication. MD has received personal honoraria for his participation at symposium and/or advisory boards organized by Pfizer, AbbVie, Merck, UCB, Novartis, and Eli Lilly, and his department has received research grants from Pfizer, AbbVie, Merck, UCB, Novartis, and Eli Lilly.

  • Accepted for publication April 11, 2022.

This is an Open Access article, which permits use, distribution, and reproduction, without modification, provided the original article is correctly cited and is not used for commercial purposes.

REFERENCES

  1. 1.↵
    1. Smolen JS,
    2. Schols M,
    3. Braun J, et al.
    Treating axial spondyloarthritis and peripheral spondyloarthritis, especially psoriatic arthritis, to target: 2017 update of recommendations by an international task force. Ann Rheum Dis 2018;77:3-17.
    OpenUrlAbstract/FREE Full Text
  2. 2.↵
    1. Fernández-Espartero C,
    2. de Miguel E,
    3. Loza E, et al.
    Validity of the ankylosing spondylitis disease activity score (ASDAS) in patients with early spondyloarthritis from the Esperanza programme. Ann Rheum Dis 2014;73:1350-5.
    OpenUrlAbstract/FREE Full Text
  3. 3.↵
    1. Fagerli KM,
    2. Lie E,
    3. van der Heijde D, et al.
    Selecting patients with ankylosing spondylitis for TNF inhibitor therapy: comparison of ASDAS and BASDAI eligibility criteria. Rheumatology 2012;51:1479-83.
    OpenUrlCrossRefPubMed
  4. 4.↵
    1. Arends S,
    2. Brouwer E,
    3. van der Veer E, et al.
    Baseline predictors of response and discontinuation of tumor necrosis factor-alpha blocking therapy in ankylosing spondylitis: a prospective longitudinal observational cohort study. Arthritis Res Ther 2011;13:R94.
    OpenUrlCrossRefPubMed
  5. 5.↵
    1. Ramiro S,
    2. van der Heijde D,
    3. van Tubergen A, et al.
    Higher disease activity leads to more structural damage in the spine in ankylosing spondylitis: 12-year longitudinal data from the OASIS cohort. Ann Rheum Dis 2014;73:1455-61.
    OpenUrlAbstract/FREE Full Text
  6. 6.↵
    1. Pedersen SJ,
    2. Sørensen IJ,
    3. Garnero P, et al.
    ASDAS, BASDAI and different treatment responses and their relation to biomarkers of inflammation, cartilage and bone turnover in patients with axial spondyloarthritis treated with TNFalpha inhibitors. Ann Rheum Dis 2011;70:1375-81.
    OpenUrlAbstract/FREE Full Text
  7. 7.↵
    1. Salaffi F,
    2. Ciapetti A,
    3. Carotti M,
    4. Gasparini S,
    5. Citera G,
    6. Gutierrez M.
    Construct validity and responsiveness of the simplified version of Ankylosing Spondylitis Disease Activity Score (SASDAS) for the evaluation of disease activity in axial spondyloarthritis. Health Qual Life Outcomes 2014;12:129.
    OpenUrl
  8. 8.
    1. Sommerfleck FA,
    2. Schneeberger EE,
    3. Buschiazzo EE,
    4. Maldonado Cocco JA,
    5. Citera G.
    A simplified version of Ankylosing Spondylitis Disease Activity Score (ASDAS) in patients with ankylosing spondylitis. Clin Rheumatol 2012;31:1599-603.
    OpenUrlPubMed
  9. 9.↵
    1. Schneeberger EE,
    2. Zamora N,
    3. Citera G.
    SASDAS (simplified version of ankylosing spondylitis disease activity score)-ESR performance and development of SASDAS-CRP and their agreement with ASDAS-ESR and ASDAS-CRP in patients with ankylosing spondylitis. Clin Rheumatol 2016;35:2865-6.
    OpenUrl
  10. 10.↵
    1. Solmaz D,
    2. Yildirim T,
    3. Avci O,
    4. Tomas N,
    5. Akar S.
    Performance characteristics of the simplified version of ankylosing spondylitis disease activity score (SASDAS). Clin Rheumatol 2016;35:1753-8.
    OpenUrl
  11. 11.↵
    1. Sundaram TG,
    2. Muhammed H,
    3. Aggarwal A,
    4. Gupta L.
    A prospective study of novel disease activity indices for ankylosing spondylitis. Rheumatol Int 2020;40:1843-9.
    OpenUrl
  12. 12.↵
    1. Sieper J,
    2. van der Heijde D,
    3. Dougados M, et al.
    Efficacy and safety of adalimumab in patients with non-radiographic axial spondyloarthritis: results of a randomised placebo-controlled trial (ABILITY-1). Ann Rheum Dis 2013;72:815-22.
    OpenUrlAbstract/FREE Full Text
  13. 13.
    1. Poddubnyy D,
    2. Sieper J.
    Radiographic progression in ankylosing spondylitis/axial spondyloarthritis: how fast and how clinically meaningful? Curr Opin Rheumatol 2012;24:363-9.
    OpenUrlCrossRefPubMed
  14. 14.
    1. Pedersen SJ,
    2. Sørensen IJ,
    3. Lambert RGW, et al.
    Radiographic progression is associated with resolution of systemic inflammation in patients with axial spondylarthritis treated with tumor necrosis factor α inhibitors: a study of radiographic progression, inflammation on magnetic resonance imaging, and circulating biomarkers of inflammation, angiogenesis, and cartilage and bone turnover. Arthritis Rheum 2011;63:3789-800.
    OpenUrlCrossRefPubMed
  15. 15.
    1. Benhamou M,
    2. Gossec L,
    3. Dougados M.
    Clinical relevance of C-reactive protein in ankylosing spondylitis and evaluation of the NSAIDs/coxibs’ treatment effect on C-reactive protein. Rheumatology 2010;49:536-41.
    OpenUrlCrossRefPubMed
  16. 16.↵
    1. Blachier M,
    2. Canouï-Poitrine F,
    3. Dougados M, et al.
    Factors associated with radiographic lesions in early axial spondyloarthritis. Results from the DESIR cohort. Rheumatology 2013;52:1686-93.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Dougados M,
    2. van der Heijde D,
    3. Sieper J, et al.
    Symptomatic efficacy of etanercept and its effects on objective signs of inflammation in early nonradiographic axial spondyloarthritis: a multicenter, randomized, double-blind, placebo-controlled trial. Arthritis Rheumatol 2014;66:2091-102.
    OpenUrlPubMed
  18. 18.↵
    1. Maksymowych WP,
    2. Dougados M,
    3. van der Heijde D, et al.
    Clinical and MRI responses to etanercept in early non-radiographic axial spondyloarthritis: 48-week results from the EMBARK study. Ann Rheum Dis 2016;75:1328-35.
    OpenUrlAbstract/FREE Full Text
  19. 19.↵
    1. Machado PM,
    2. Landewé R,
    3. van der Heijde D; Assessment of SpondyloArthritis international Society
    . Ankylosing Spondylitis Disease Activity Score (ASDAS): 2018 update of the nomenclature for disease activity states. Ann Rheum Dis 2018;77:1539-40.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. McHugh ML.
    Interrater reliability: the kappa statistic. Biochem Med 2012;22:276-82.
    OpenUrlPubMed
  21. 21.↵
    1. Bansal N,
    2. Duggal L,
    3. Jain N.
    Validity of simplified ankylosing spondylitis disease activity scores (SASDAS) in Indian ankylosing spondylitis patients. J Clin Diagn Res 2017;11:OC06-09.
    OpenUrl
  22. 22.↵
    1. Landewé R,
    2. Dougados M,
    3. Mielants H,
    4. van der Tempel H,
    5. van der Heijde D.
    Physical function in ankylosing spondylitis is independently determined by both disease activity and radiographic damage of the spine. Ann Rheum Dis 2009;68:863-7.
    OpenUrlAbstract/FREE Full Text
  23. 23.
    1. Poddubnyy D,
    2. Fedorova A,
    3. Listing J, et al.
    Physical function and spinal mobility remain stable despite radiographic spinal progression in patients with ankylosing spondylitis treated with TNF-alpha inhibitors for up to 10 years. J Rheumatol 2016;43:2142-8.
    OpenUrlAbstract/FREE Full Text
  24. 24.
    1. Molto A,
    2. Tezenas du Montcel S,
    3. Wendling D,
    4. Dougados M,
    5. Vanier A,
    6. Gossec L.
    Disease activity trajectories in early axial spondyloarthritis: results from the DESIR cohort. Ann Rheum Dis 2017;76:1036-41.
    OpenUrlAbstract/FREE Full Text
  25. 25.↵
    1. Marengo MF,
    2. Schneeberger EE,
    3. Gagliardi S,
    4. Maldonado Cocco JA,
    5. Citera G.
    [Determinants of functional disability in patients with ankylosing spondylitis in Argentina]. [Article in Spanish] Rev Arg Reumatol 2009;20:20-5.
    OpenUrl
  26. 26.↵
    1. Konca S,
    2. Keskin D,
    3. Ciliz D,
    4. Bodur H,
    5. Sakman B.
    Spinal inflammation by magnetic resonance imaging in patients with ankylosing spondylitis: association with disease activity and outcome parameters. Rheumatol Int 2012;32:3765-70.
    OpenUrlCrossRefPubMed
  27. 27.
    1. Machado P,
    2. Landewé R,
    3. Braun J,
    4. Hermann KGA,
    5. Baker D,
    6. van der Heijde D.
    Both structural damage and inflammation of the spine contribute to impairment of spinal mobility in patients with ankylosing spondylitis. Ann Rheum Dis 2010;69:1465-70.
    OpenUrlAbstract/FREE Full Text
  28. 28.↵
    1. Rudwaleit M,
    2. Schwarzlose S,
    3. Hilgert ES,
    4. Listing J,
    5. Braun J,
    6. Sieper J.
    MRI in predicting a major clinical response to anti-tumour necrosis factor treatment in ankylosing spondylitis. Ann Rheum Dis 2008;67:1276-81.
    OpenUrlAbstract/FREE Full Text
  29. 29.↵
    1. Castro MP,
    2. Stebbings SM,
    3. Milosavljevic S,
    4. Bussey MD.
    Construct validity of clinical spinal mobility tests in ankylosing spondylitis: a systematic review and meta-analysis. Clin Rheumatol 2016;35:1777-87.
    OpenUrl
  30. 30.↵
    1. Smolen JS,
    2. Breedveld FC,
    3. Schiff MH, et al.
    A simplified disease activity index for rheumatoid arthritis for use in clinical practice. Rheumatology 2003;42:244-57.
    OpenUrlCrossRefPubMed
  31. 31.↵
    1. Schoels M,
    2. Aletaha D,
    3. Funovits J,
    4. Kavanaugh A,
    5. Baker D,
    6. Smolen JS.
    Application of the DAREA/DAPSA score for assessment of disease activity in psoriatic arthritis. Ann Rheum Dis 2010;69:1441-7.
    OpenUrlAbstract/FREE Full Text

DATA AVAILABILITY

Upon request, and subject to review, Pfizer will provide the data that support the findings of this study. Subject to certain criteria, conditions, and exceptions, Pfizer may also provide access to the related individual deidentified participant data. See https://www.pfizer.com/science/clinical-trials/trial-data-and-results for more information.

Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

 Request Permissions

Bookmark this article

Jump to section

Keywords

CLINICAL TRIALS
ETANERCEPT
OUTCOME
AXIAL SPONDYLOARTHRITIS

Keywords