Predicting Objective Response Rate (ORR) in Immune Checkpoint Inhibitor (ICI) Therapies with Machine Learning (ML) by Combining Clinical and Patient-Reported Data
ICIs are a standard of care in several malignancies; however, according to overall response rate (ORR), only a subset of eligible patients benefits from ICIs. Thus, an ability to predict ORR could enable more rational use. In this study a ML-based ORR prediction model was built, with patient-reporte...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/3/1563 |
| _version_ | 1797489079935303680 |
|---|---|
| author | Sanna Iivanainen Jussi Ekström Henri Virtanen Vesa V. Kataja Jussi P. Koivunen |
| author_facet | Sanna Iivanainen Jussi Ekström Henri Virtanen Vesa V. Kataja Jussi P. Koivunen |
| author_sort | Sanna Iivanainen |
| collection | DOAJ |
| description | ICIs are a standard of care in several malignancies; however, according to overall response rate (ORR), only a subset of eligible patients benefits from ICIs. Thus, an ability to predict ORR could enable more rational use. In this study a ML-based ORR prediction model was built, with patient-reported symptom data and other clinical data as inputs, using the extreme gradient boosting technique (XGBoost). Prediction performance for unseen samples was evaluated using leave-one-out cross-validation (LOOCV), and the performance was evaluated with accuracy, AUC (area under curve), F1 score, and MCC (Matthew’s correlation coefficient). The ORR prediction model had a promising LOOCV performance with all four metrics: accuracy (75%), AUC (0.71), F1 score (0.58), and MCC (0.4). A rather good sensitivity (0.58) and high specificity (0.82) of the model were seen in the confusion matrix for all 63 LOOCV ORR predictions. The two most important symptoms for predicting the ORR were itching and fatigue. The results show that it is possible to predict ORR for patients with multiple advanced cancers undergoing ICI therapies with a ML model combining clinical, routine laboratory, and patient-reported data even with a limited size cohort. |
| first_indexed | 2024-03-10T00:11:25Z |
| format | Article |
| id | doaj.art-b9e2d29ab44e452a97a21fc93801a593 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T00:11:25Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-b9e2d29ab44e452a97a21fc93801a5932023-11-23T15:59:41ZengMDPI AGApplied Sciences2076-34172022-01-01123156310.3390/app12031563Predicting Objective Response Rate (ORR) in Immune Checkpoint Inhibitor (ICI) Therapies with Machine Learning (ML) by Combining Clinical and Patient-Reported DataSanna Iivanainen0Jussi Ekström1Henri Virtanen2Vesa V. Kataja3Jussi P. Koivunen4Department of Oncology and Radiotherapy, Oulu University Hospital and MRC Oulu, 90220 Oulu, FinlandKaiku Health Oy, 00530 Helsinki, FinlandKaiku Health Oy, 00530 Helsinki, FinlandKaiku Health Oy, 00530 Helsinki, FinlandDepartment of Oncology and Radiotherapy, Oulu University Hospital and MRC Oulu, 90220 Oulu, FinlandICIs are a standard of care in several malignancies; however, according to overall response rate (ORR), only a subset of eligible patients benefits from ICIs. Thus, an ability to predict ORR could enable more rational use. In this study a ML-based ORR prediction model was built, with patient-reported symptom data and other clinical data as inputs, using the extreme gradient boosting technique (XGBoost). Prediction performance for unseen samples was evaluated using leave-one-out cross-validation (LOOCV), and the performance was evaluated with accuracy, AUC (area under curve), F1 score, and MCC (Matthew’s correlation coefficient). The ORR prediction model had a promising LOOCV performance with all four metrics: accuracy (75%), AUC (0.71), F1 score (0.58), and MCC (0.4). A rather good sensitivity (0.58) and high specificity (0.82) of the model were seen in the confusion matrix for all 63 LOOCV ORR predictions. The two most important symptoms for predicting the ORR were itching and fatigue. The results show that it is possible to predict ORR for patients with multiple advanced cancers undergoing ICI therapies with a ML model combining clinical, routine laboratory, and patient-reported data even with a limited size cohort.https://www.mdpi.com/2076-3417/12/3/1563ePROmachine learningprediction modelirAEimmune checkpoint inhibitorprognosis |
| spellingShingle | Sanna Iivanainen Jussi Ekström Henri Virtanen Vesa V. Kataja Jussi P. Koivunen Predicting Objective Response Rate (ORR) in Immune Checkpoint Inhibitor (ICI) Therapies with Machine Learning (ML) by Combining Clinical and Patient-Reported Data Applied Sciences ePRO machine learning prediction model irAE immune checkpoint inhibitor prognosis |
| title | Predicting Objective Response Rate (ORR) in Immune Checkpoint Inhibitor (ICI) Therapies with Machine Learning (ML) by Combining Clinical and Patient-Reported Data |
| title_full | Predicting Objective Response Rate (ORR) in Immune Checkpoint Inhibitor (ICI) Therapies with Machine Learning (ML) by Combining Clinical and Patient-Reported Data |
| title_fullStr | Predicting Objective Response Rate (ORR) in Immune Checkpoint Inhibitor (ICI) Therapies with Machine Learning (ML) by Combining Clinical and Patient-Reported Data |
| title_full_unstemmed | Predicting Objective Response Rate (ORR) in Immune Checkpoint Inhibitor (ICI) Therapies with Machine Learning (ML) by Combining Clinical and Patient-Reported Data |
| title_short | Predicting Objective Response Rate (ORR) in Immune Checkpoint Inhibitor (ICI) Therapies with Machine Learning (ML) by Combining Clinical and Patient-Reported Data |
| title_sort | predicting objective response rate orr in immune checkpoint inhibitor ici therapies with machine learning ml by combining clinical and patient reported data |
| topic | ePRO machine learning prediction model irAE immune checkpoint inhibitor prognosis |
| url | https://www.mdpi.com/2076-3417/12/3/1563 |
| work_keys_str_mv | AT sannaiivanainen predictingobjectiveresponserateorrinimmunecheckpointinhibitoricitherapieswithmachinelearningmlbycombiningclinicalandpatientreporteddata AT jussiekstrom predictingobjectiveresponserateorrinimmunecheckpointinhibitoricitherapieswithmachinelearningmlbycombiningclinicalandpatientreporteddata AT henrivirtanen predictingobjectiveresponserateorrinimmunecheckpointinhibitoricitherapieswithmachinelearningmlbycombiningclinicalandpatientreporteddata AT vesavkataja predictingobjectiveresponserateorrinimmunecheckpointinhibitoricitherapieswithmachinelearningmlbycombiningclinicalandpatientreporteddata AT jussipkoivunen predictingobjectiveresponserateorrinimmunecheckpointinhibitoricitherapieswithmachinelearningmlbycombiningclinicalandpatientreporteddata |