Predicting abatacept retention using machine learning
Abstract Background The incorporation of machine learning is becoming more prevalent in the clinical setting. By predicting clinical outcomes, machine learning can provide clinicians with a valuable tool for refining precision medicine approaches and improving treatment outcomes. Methods This was a...
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-02-01
|
| Series: | Arthritis Research & Therapy |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13075-025-03484-0 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571479156326400 |
|---|---|
| author | Rieke Alten Claire Behar Pierre Merckaert Ebenezer Afari Virginie Vannier-Moreau Anael Ohayon Sean E. Connolly Aurélie Najm Pierre-Antoine Juge Gengyuan Liu Angshu Rai Yedid Elbez Karissa Lozenski |
| author_facet | Rieke Alten Claire Behar Pierre Merckaert Ebenezer Afari Virginie Vannier-Moreau Anael Ohayon Sean E. Connolly Aurélie Najm Pierre-Antoine Juge Gengyuan Liu Angshu Rai Yedid Elbez Karissa Lozenski |
| author_sort | Rieke Alten |
| collection | DOAJ |
| description | Abstract Background The incorporation of machine learning is becoming more prevalent in the clinical setting. By predicting clinical outcomes, machine learning can provide clinicians with a valuable tool for refining precision medicine approaches and improving treatment outcomes. Methods This was a post hoc analysis of pooled patient-level data from the global, real-world ACTION and ASCORE trials in patients with rheumatoid arthritis (RA) initiating abatacept. Patient demographic and disease characteristics were input across 10 machine learning models used to predict 12-month treatment retention. Retention was defined as treatment for > 365 days or ≤ 365 days in patients who achieved remission or major clinical response (based on European Alliance of Associations for Rheumatology response criteria). The pooled dataset was split into a training/validation cohort for model development and a test cohort for an unbiased evaluation of performance. SHapley Additive exPlanation (SHAP) values determined the level of importance and directionality for key patient features predicting abatacept retention. Results The pooled ACTION and ASCORE dataset included 5320 patients with RA (mean [standard deviation] age 57.7 [12.7] years; 79% female). The 12-month abatacept retention rate was 61% (n = 3236) with a discontinuation rate of 39% (n = 2037). In the training set (n = 4218), the gradient-boosting classifier model demonstrated the best performance (testing accuracy: 62%). This model had an area under the receiver operating characteristic curve (95% confidence interval) of 0.620 (0.586, 0.653) and F1 score of 0.659 (0.625, 0.689) in the test set of patients (n = 1055). Using this model, the five most important variables predicting 12-month abatacept retention were low body mass index (BMI), low American College of Rheumatology functional status class, anti-citrullinated protein antibody (ACPA) positivity, low Patient Global Assessment, and younger age. Conclusions The gradient-boosting classifier model identified key patient features predictive of abatacept retention from this large, real-world study population. The SHAP values conveyed the directionality and importance of BMI, functional status, ACPA serostatus, Patient Global Assessment, and age for abatacept retention. Findings are consistent with previous observations and help validate the machine learning approach for predictive modelling in RA treatment, and may help inform clinical decision making. Trial registration NCT02109666 (ACTION), NCT02090556 (ASCORE). |
| format | Article |
| id | doaj-art-d2260f5733a0425f85c3460893b2331e |
| institution | Kabale University |
| issn | 1478-6362 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | BMC |
| record_format | Article |
| series | Arthritis Research & Therapy |
| spelling | doaj-art-d2260f5733a0425f85c3460893b2331e2025-02-02T12:35:41ZengBMCArthritis Research & Therapy1478-63622025-02-0127111110.1186/s13075-025-03484-0Predicting abatacept retention using machine learningRieke Alten0Claire Behar1Pierre Merckaert2Ebenezer Afari3Virginie Vannier-Moreau4Anael Ohayon5Sean E. Connolly6Aurélie Najm7Pierre-Antoine Juge8Gengyuan Liu9Angshu Rai10Yedid Elbez11Karissa Lozenski12Schlosspark-Klinik UniversityTulsyData Revenue GmbHPrivate PracticeBristol Myers SquibbBristol Myers SquibbBristol Myers SquibbSchool of Infection and Immunity, College of Medical Veterinary and Life Sciences, University of GlasgowUniversité de Paris, AP-HP, Hôpital Bichat Claude-BernardBristol Myers SquibbBristol Myers SquibbSignifienceBristol Myers SquibbAbstract Background The incorporation of machine learning is becoming more prevalent in the clinical setting. By predicting clinical outcomes, machine learning can provide clinicians with a valuable tool for refining precision medicine approaches and improving treatment outcomes. Methods This was a post hoc analysis of pooled patient-level data from the global, real-world ACTION and ASCORE trials in patients with rheumatoid arthritis (RA) initiating abatacept. Patient demographic and disease characteristics were input across 10 machine learning models used to predict 12-month treatment retention. Retention was defined as treatment for > 365 days or ≤ 365 days in patients who achieved remission or major clinical response (based on European Alliance of Associations for Rheumatology response criteria). The pooled dataset was split into a training/validation cohort for model development and a test cohort for an unbiased evaluation of performance. SHapley Additive exPlanation (SHAP) values determined the level of importance and directionality for key patient features predicting abatacept retention. Results The pooled ACTION and ASCORE dataset included 5320 patients with RA (mean [standard deviation] age 57.7 [12.7] years; 79% female). The 12-month abatacept retention rate was 61% (n = 3236) with a discontinuation rate of 39% (n = 2037). In the training set (n = 4218), the gradient-boosting classifier model demonstrated the best performance (testing accuracy: 62%). This model had an area under the receiver operating characteristic curve (95% confidence interval) of 0.620 (0.586, 0.653) and F1 score of 0.659 (0.625, 0.689) in the test set of patients (n = 1055). Using this model, the five most important variables predicting 12-month abatacept retention were low body mass index (BMI), low American College of Rheumatology functional status class, anti-citrullinated protein antibody (ACPA) positivity, low Patient Global Assessment, and younger age. Conclusions The gradient-boosting classifier model identified key patient features predictive of abatacept retention from this large, real-world study population. The SHAP values conveyed the directionality and importance of BMI, functional status, ACPA serostatus, Patient Global Assessment, and age for abatacept retention. Findings are consistent with previous observations and help validate the machine learning approach for predictive modelling in RA treatment, and may help inform clinical decision making. Trial registration NCT02109666 (ACTION), NCT02090556 (ASCORE).https://doi.org/10.1186/s13075-025-03484-0AbataceptMachine learningRetentionRheumatoid arthritisTreatment response |
| spellingShingle | Rieke Alten Claire Behar Pierre Merckaert Ebenezer Afari Virginie Vannier-Moreau Anael Ohayon Sean E. Connolly Aurélie Najm Pierre-Antoine Juge Gengyuan Liu Angshu Rai Yedid Elbez Karissa Lozenski Predicting abatacept retention using machine learning Arthritis Research & Therapy Abatacept Machine learning Retention Rheumatoid arthritis Treatment response |
| title | Predicting abatacept retention using machine learning |
| title_full | Predicting abatacept retention using machine learning |
| title_fullStr | Predicting abatacept retention using machine learning |
| title_full_unstemmed | Predicting abatacept retention using machine learning |
| title_short | Predicting abatacept retention using machine learning |
| title_sort | predicting abatacept retention using machine learning |
| topic | Abatacept Machine learning Retention Rheumatoid arthritis Treatment response |
| url | https://doi.org/10.1186/s13075-025-03484-0 |
| work_keys_str_mv | AT riekealten predictingabataceptretentionusingmachinelearning AT clairebehar predictingabataceptretentionusingmachinelearning AT pierremerckaert predictingabataceptretentionusingmachinelearning AT ebenezerafari predictingabataceptretentionusingmachinelearning AT virginievanniermoreau predictingabataceptretentionusingmachinelearning AT anaelohayon predictingabataceptretentionusingmachinelearning AT seaneconnolly predictingabataceptretentionusingmachinelearning AT aurelienajm predictingabataceptretentionusingmachinelearning AT pierreantoinejuge predictingabataceptretentionusingmachinelearning AT gengyuanliu predictingabataceptretentionusingmachinelearning AT angshurai predictingabataceptretentionusingmachinelearning AT yedidelbez predictingabataceptretentionusingmachinelearning AT karissalozenski predictingabataceptretentionusingmachinelearning |