Prediction of ineffectiveness of biological drugs using machine learning and explainable AI methods: data from the Austrian Biological Registry BioReg

OBJECTIVES

Machine learning models can support an individualized approach in the choice of bDMARDs. We developed prediction models for 5 different bDMARDs using machine learning methods based on patient data derived from the Austrian Biologics Registry (BioReg).

METHODS

Data from 1397 patients and 19 variables with at least 100 treat-to-target (t2t) courses per drug were derived from the BioReg biologics registry. Different machine learning algorithms were trained to predict the risk of ineffectiveness for each bDMARD within the first 26 weeks. Cross-validation and hyperparameter optimization were applied to generate the best models. Model quality was assessed by area under the receiver operating characteristic (AUROC). Using explainable AI (XAI), risk-reducing and risk-increasing factors were extracted.

RESULTS

The best models per drug achieved an AUROC score of the following: abatacept, 0.66 (95% CI, 0.54-0.78); adalimumab, 0.70 (95% CI, 0.68-0.74); certolizumab, 0.84 (95% CI, 0.79-0.89); etanercept, 0.68 (95% CI, 0.55-0.87); tocilizumab, 0.72 (95% CI, 0.69-0.77). The most risk-increasing variables were visual analytic scores (VAS) for abatacept and etanercept and co-therapy with glucocorticoids for adalimumab. Dosage was the most important variable for certolizumab and associated with a lower risk of non-response. Some variables, such as gender and rheumatoid factor (RF), showed opposite impacts depending on the bDMARD.

CONCLUSION

Ineffectiveness of biological drugs could be predicted with promising accuracy. Interestingly, individual parameters were found to be associated with drug responses in different directions, indicating highly complex interactions. Machine learning can be of help in the decision-process by disentangling these relations.

A Prospective Study to Evaluate the Impact of Golimumab Therapy on Work Productivity and Activity, and Quality of Life in Patients With Rheumatoid Arthritis, Psoriasis Arthritis and Axial Spondyloarthritis in a Real Life Setting in AUSTRIA. The GO-ACTIVE Study

Objective

To analyze real-world evidence on work productivity and daily activity impairment (WPAI) and health-related quality of life (HRQoL) in rheumatoid arthritis (RA), psoriatic arthritis (PsA), and axial spondyloarthritis (axSpA) patients treated with golimumab in Austria.

Methods

This was a prospective, non-interventional, multi-center study conducted in RA, PsA and axSpA patients initiating golimumab between April 2016 and May 2020 in 40 centers in Austria. WPAI, HRQoL (RAQoL, ankylosing spondylitis (AS)QoL and PsAQoL) questionnaires and disease activity (Clinical Disease Activity Index, CDAI, in RA and PsA; Bath Ankylosing Spondylitis Disease Activity Index, BASDAI, in axSpA) were assessed at baseline and months 3, 6, 12, 18, and 24. Association between WPAI and disease activity was tested using linear regression.

Results

We enrolled 233 patients (RA, n = 95; axSpA, n = 69; PsA, n = 69), 110 patients were followed up to month 24. Mean age was 50.2 ± 14.2 years; 64% were female. Disease activity decreased from baseline to month 24 (RA: CDAI −24.3 ± 13.5; axSpA: BASDAI −4.4 ± 2.1, and PsA: CDAI −21.7 ± 8.5, p < 0.0001, each). Total work productivity impairment (TWPI), activity impairment and presenteeism subscores continuously decreased throughout month 24 in all indications: RA (−58.3 ± 23%, −62.6 ± 23.8% and −61.7 ± 23.3%, respectively as compared to baseline; p < 0.0001, each), axSpA (−34.4 ± 38.3%, p = 0.0117; −60.9 ± 25.9%, and −43.8 ± 26.6%, respectively, p ≤ 0.0001 both) and PsA (−35.8 ± 43.7%, p = 0.0186; −52.3 ± 25.4%, p < 0.0001; and −43.3 ± 33.5%, p = 0.0007, respectively). Absenteeism scores decreased only in RA patients (−9.2 ± 24.9%, p = 0.0234). HRQoL improved between baseline and month 24 (RAQoL: −12.6 ± 7.5; ASQoL: −8.0 ± 4.3; PsAQoL; −8.3 ± 6.4, p < 0.0001, each). TWPI, presenteeism and activity impairment strongly associated with disease activity throughout the study.

Conclusions

This real-world study confirms the benefit of golimumab on work productivity/daily activity impairment in Austrian RA, PsA, and axSpA patients.