Machine Learning Model for Predicting Outcomes of Biologic Therapy in Psoriasis
Bibliographic record
Abstract
Abstract Background Biological agents used for the therapy of psoriasis lose efficacy over time, which leads to discontinuation of the drug. Optimization of long-term biologic treatment is an area of medical need but there are currently no prediction tools for biologic drug discontinuation. Objective To compare the accuracy of the risk factor-based frequentist statistical model to machine learning to predict the 5-year probability of biologic drug discontinuation. Methods The national Danish psoriasis biologic therapy registry, Dermbio, comprising 6,172 treatment series with anti-TNF (Etanercept, Infliximab, Adalimumab), Ustekinumab, Guselkumab and anti-IL17 (Secukinumab and Ixekizumab) in 3,388 unique patients was used as data source. Hazard ratios (HR) were computed for all available predictive factors using Cox regression analysis. Different machine learning (ML) models for the prediction of 5-year risk of drug discontinuation were trained using the 5-fold cross validation technique and using 10 clinical features routinely assessed in psoriasis patients as input variables. Model performance was assessed using the area under the receiver operating characteristic curve (AUC). Results The lowest 5-year risk of discontinuation was associated with therapy with ustekinumab or ixekizumab, male sex and no previous exposure to biologic therapy. The predictive model based on those risk factors had an AUC of 0.61. The best ML model (gradient boosted tree) had an AUC of 0.85. Conclusions A machine learning-based approach, more than a statistical model, accurately predicts the risk of discontinuation of biologic therapy based on simple patient variables available in clinical practice. ML might be incorporated into clinical decision making.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.010 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".