ABSTRACT
BACKGROUND: Because anti-neutrophil cytoplasmatic antibody (ANCA)-associated vasculitis (AAV) is a rare, life-threatening, auto-immune disease, conducting research is difficult but essential. A long-lasting challenge is to identify rare AAV patients within the electronic-health-record (EHR)-system to facilitate real-world research. Artificial intelligence (AI)-search tools using natural language processing (NLP) for text-mining are increasingly postulated as a solution. METHODS: We employed an AI-tool that combined text-mining with NLP-based exclusion, to accurately identify rare AAV patients within large EHR-systems (>2.000.000 records). We developed an identification method in an academic center with an established AAV-training set (n = 203) and validated the method in a non-academic center with an AAV-validation set (n = 84). To assess accuracy anonymized patient records were manually reviewed. RESULTS: Based on an iterative process, a text-mining search was developed on disease description, laboratory measurements, medication and specialisms. In the training center, 608 patients were identified with a sensitivity of 97.0 % (95%CI [93.7, 98.9]) and positive predictive value (PPV) of 56.9 % (95%CI [52.9, 60.1]). NLP-based exclusion resulted in 444 patients increasing PPV to 77.9 % (95%CI [73.7, 81.7]) while sensitivity remained 96.3 % (95%CI [93.8, 98.0]). In the validation center, text-mining identified 333 patients (sensitivity 97.6 % (95%CI [91.6, 99.7]), PPV 58.2 % (95%CI [52.8, 63.6])) and NLP-based exclusion resulted in 223 patients, increasing PPV to 86.1 % (95%CI [80.9, 90.4]) with 98.0 % (95%CI [94.9, 99.4]) sensitivity. Our identification method outperformed ICD-10-coding predominantly in identifying MPO+ and organ-limited AAV patients. CONCLUSIONS: Our study highlights the advantages of implementing AI, notably NLP, to accurately identify rare AAV patients within large EHR-systems and demonstrates the applicability and transportability. Therefore, this method can reduce efforts to identify AAV patients and accelerate real-world research, while avoiding bias by ICD-10-coding.
Subject(s)
Anti-Neutrophil Cytoplasmic Antibody-Associated Vasculitis , Artificial Intelligence , Humans , Natural Language Processing , Electronic Health Records , Anti-Neutrophil Cytoplasmic Antibody-Associated Vasculitis/diagnosis , SoftwareABSTRACT
INTRODUCTION: Both rituximab (RTX) and cyclophosphamide (CYC) are effectively used in combination with steroids as remission induction therapy for patients with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV). Several studies have shown that the effect on achieving (clinical) remission, frequency and severity of relapses is equivalent for both therapies, but there is accumulating data that the long-term safety profile of RTX might outperform CYC. Combination of RTX with low-dose CYC (LD-CYC) has been investigated in only a few uncontrolled cohort studies, in which clinical remission and a favourable immunological state with low relapse rates was quickly achieved. In this randomised controlled trial, we aim to investigate whether the combination treatment (RTX+LD CYC) is superior in comparison to standard care with RTX only. METHODS AND ANALYSIS: This study is an open-label, multicentre, 1:1 randomised, prospective study for patients with AAV with generalised disease, defined as involvement of major organs, that is, kidneys, lungs, heart and nervous system. In total, 100 patients will be randomised 1:1 to receive either remission induction therapy with standard of care (RTX) or combination treatment (RTX+LD CYC) in addition to steroids and both arms are followed by maintenance with RTX retreatments (tailored to B-cell and ANCA status). Our primary outcome is the number of retreatments needed to maintain clinical remission over 2 years. Secondary outcomes are relevant clinical endpoints, safety, quality of life and immunological responses. ETHICS AND DISSEMINATION: This study has received approval of the Medical Ethics Committee of the Leiden University Medical Center (P18.216, NL67515.058.18, date: 7 March 2019). The results of this trial (positive and negative) will be submitted for publication in relevant peer-reviewed publications and the key findings presented at national and international conferences. TRIAL REGISTRATION NUMBER: NCT03942887.