Using machine learning for mortality prediction and risk stratification in atezolizumab‐treated cancer patients: Integrative analysis of eight clinical trials
Abstract Background Few models exist to predict mortality in cancer patients receiving immunotherapy. Our aim was to build a machine learning‐based risk stratification model for predicting mortality in atezolizumab‐treated cancer patients. Methods Data from 2538 patients in eight atezolizumab‐treate...
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-02-01
|
| Series: | Cancer Medicine |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/cam4.5060 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846157611285610496 |
|---|---|
| author | Yougen Wu Wenyu Zhu Jing Wang Lvwen Liu Wei Zhang Yang Wang Jindong Shi Ju Xia Yuting Gu Qingqing Qian Yang Hong |
| author_facet | Yougen Wu Wenyu Zhu Jing Wang Lvwen Liu Wei Zhang Yang Wang Jindong Shi Ju Xia Yuting Gu Qingqing Qian Yang Hong |
| author_sort | Yougen Wu |
| collection | DOAJ |
| description | Abstract Background Few models exist to predict mortality in cancer patients receiving immunotherapy. Our aim was to build a machine learning‐based risk stratification model for predicting mortality in atezolizumab‐treated cancer patients. Methods Data from 2538 patients in eight atezolizumab‐treated cancer clinical trials across three cancer types (non‐small‐cell lung cancer, bladder transitional cell carcinoma, and renal cell carcinoma) were included. The whole cohort was randomly split into development and validation cohorts in a 7:3 ratio. Machine‐learning algorithms (extreme gradient boosting, random forest, logistic regression with lasso regularization, support vector machine, and K‐nearest neighbor) were applied to develop prediction models. Model performance was mainly assessed by area under the receiver operating characteristic curve (AUC) value, calibration plot, and decision curve analysis. The probability of death risk was then stratified. Results One thousand and three hundred and seventy‐nine (54.33%) patients died. The random forest (RF) model was overall the best in terms of predictive performance, with the AUC of 0.844 (95% confidence interval [CI]: 0.826–0.862) in the development cohort and 0.786 (95% CI: 0.754–0.818) in the validation cohort for predicting mortality. Twelve baseline variables contributing to mortality prediction in the RF model were C‐reactive protein, PD‐L1 level, cancer type, prior liver metastasis, derived neutrophil‐to‐lymphocyte ratio, alkaline phosphatase, albumin, hemoglobin, white blood cell count, number of metastatic sites, pulse rate, and Eastern Cooperative Oncology Group (ECOG) performance status. A total of 1782 (70.2%) patients were separated into the high‐risk and 756 (29.8%) low‐risk groups. Patients in the high‐risk group were significantly more likely to die, experience disease progression, discontinue study, and discontinue treatment than patients in the low‐risk group (all p values < 0.001). Risk groups were not associated with immune‐related adverse events and grades 3–5 treatment‐related adverse events (all p values > 0.05). Conclusion RF model has good performance in mortality prediction and risk stratification for cancer patients receiving atezolizumab monotherapy. |
| format | Article |
| id | doaj-art-a3aa5f50e78e416f93774c6be39e27e2 |
| institution | Kabale University |
| issn | 2045-7634 |
| language | English |
| publishDate | 2023-02-01 |
| publisher | Wiley |
| record_format | Article |
| series | Cancer Medicine |
| spelling | doaj-art-a3aa5f50e78e416f93774c6be39e27e22024-11-25T07:56:32ZengWileyCancer Medicine2045-76342023-02-011233744375710.1002/cam4.5060Using machine learning for mortality prediction and risk stratification in atezolizumab‐treated cancer patients: Integrative analysis of eight clinical trialsYougen Wu0Wenyu Zhu1Jing Wang2Lvwen Liu3Wei Zhang4Yang Wang5Jindong Shi6Ju Xia7Yuting Gu8Qingqing Qian9Yang Hong10National Institute of Clinical Research, The Fifth People's Hospital of Shanghai Fudan University Shanghai ChinaShanghai Long For Health Data Technology Co.ltd Shanghai ChinaShanghai Long For Health Data Technology Co.ltd Shanghai ChinaShanghai Long For Health Data Technology Co.ltd Shanghai ChinaDepartment of Biostatistics Fudan University School of Public Health Shanghai ChinaDepartment of Urology The Fifth People's Hospital of Shanghai, Fudan University Shanghai ChinaDepartment of Respiratory Medicine The Fifth People's Hospital of Shanghai, Fudan University Shanghai ChinaNational Institute of Clinical Research, The Fifth People's Hospital of Shanghai Fudan University Shanghai ChinaNational Institute of Clinical Research, The Fifth People's Hospital of Shanghai Fudan University Shanghai ChinaNational Institute of Clinical Research, The Fifth People's Hospital of Shanghai Fudan University Shanghai ChinaNational Institute of Clinical Research, The Fifth People's Hospital of Shanghai Fudan University Shanghai ChinaAbstract Background Few models exist to predict mortality in cancer patients receiving immunotherapy. Our aim was to build a machine learning‐based risk stratification model for predicting mortality in atezolizumab‐treated cancer patients. Methods Data from 2538 patients in eight atezolizumab‐treated cancer clinical trials across three cancer types (non‐small‐cell lung cancer, bladder transitional cell carcinoma, and renal cell carcinoma) were included. The whole cohort was randomly split into development and validation cohorts in a 7:3 ratio. Machine‐learning algorithms (extreme gradient boosting, random forest, logistic regression with lasso regularization, support vector machine, and K‐nearest neighbor) were applied to develop prediction models. Model performance was mainly assessed by area under the receiver operating characteristic curve (AUC) value, calibration plot, and decision curve analysis. The probability of death risk was then stratified. Results One thousand and three hundred and seventy‐nine (54.33%) patients died. The random forest (RF) model was overall the best in terms of predictive performance, with the AUC of 0.844 (95% confidence interval [CI]: 0.826–0.862) in the development cohort and 0.786 (95% CI: 0.754–0.818) in the validation cohort for predicting mortality. Twelve baseline variables contributing to mortality prediction in the RF model were C‐reactive protein, PD‐L1 level, cancer type, prior liver metastasis, derived neutrophil‐to‐lymphocyte ratio, alkaline phosphatase, albumin, hemoglobin, white blood cell count, number of metastatic sites, pulse rate, and Eastern Cooperative Oncology Group (ECOG) performance status. A total of 1782 (70.2%) patients were separated into the high‐risk and 756 (29.8%) low‐risk groups. Patients in the high‐risk group were significantly more likely to die, experience disease progression, discontinue study, and discontinue treatment than patients in the low‐risk group (all p values < 0.001). Risk groups were not associated with immune‐related adverse events and grades 3–5 treatment‐related adverse events (all p values > 0.05). Conclusion RF model has good performance in mortality prediction and risk stratification for cancer patients receiving atezolizumab monotherapy.https://doi.org/10.1002/cam4.5060cancer immunotherapymachine learningmortality predictionrisk stratification |
| spellingShingle | Yougen Wu Wenyu Zhu Jing Wang Lvwen Liu Wei Zhang Yang Wang Jindong Shi Ju Xia Yuting Gu Qingqing Qian Yang Hong Using machine learning for mortality prediction and risk stratification in atezolizumab‐treated cancer patients: Integrative analysis of eight clinical trials Cancer Medicine cancer immunotherapy machine learning mortality prediction risk stratification |
| title | Using machine learning for mortality prediction and risk stratification in atezolizumab‐treated cancer patients: Integrative analysis of eight clinical trials |
| title_full | Using machine learning for mortality prediction and risk stratification in atezolizumab‐treated cancer patients: Integrative analysis of eight clinical trials |
| title_fullStr | Using machine learning for mortality prediction and risk stratification in atezolizumab‐treated cancer patients: Integrative analysis of eight clinical trials |
| title_full_unstemmed | Using machine learning for mortality prediction and risk stratification in atezolizumab‐treated cancer patients: Integrative analysis of eight clinical trials |
| title_short | Using machine learning for mortality prediction and risk stratification in atezolizumab‐treated cancer patients: Integrative analysis of eight clinical trials |
| title_sort | using machine learning for mortality prediction and risk stratification in atezolizumab treated cancer patients integrative analysis of eight clinical trials |
| topic | cancer immunotherapy machine learning mortality prediction risk stratification |
| url | https://doi.org/10.1002/cam4.5060 |
| work_keys_str_mv | AT yougenwu usingmachinelearningformortalitypredictionandriskstratificationinatezolizumabtreatedcancerpatientsintegrativeanalysisofeightclinicaltrials AT wenyuzhu usingmachinelearningformortalitypredictionandriskstratificationinatezolizumabtreatedcancerpatientsintegrativeanalysisofeightclinicaltrials AT jingwang usingmachinelearningformortalitypredictionandriskstratificationinatezolizumabtreatedcancerpatientsintegrativeanalysisofeightclinicaltrials AT lvwenliu usingmachinelearningformortalitypredictionandriskstratificationinatezolizumabtreatedcancerpatientsintegrativeanalysisofeightclinicaltrials AT weizhang usingmachinelearningformortalitypredictionandriskstratificationinatezolizumabtreatedcancerpatientsintegrativeanalysisofeightclinicaltrials AT yangwang usingmachinelearningformortalitypredictionandriskstratificationinatezolizumabtreatedcancerpatientsintegrativeanalysisofeightclinicaltrials AT jindongshi usingmachinelearningformortalitypredictionandriskstratificationinatezolizumabtreatedcancerpatientsintegrativeanalysisofeightclinicaltrials AT juxia usingmachinelearningformortalitypredictionandriskstratificationinatezolizumabtreatedcancerpatientsintegrativeanalysisofeightclinicaltrials AT yutinggu usingmachinelearningformortalitypredictionandriskstratificationinatezolizumabtreatedcancerpatientsintegrativeanalysisofeightclinicaltrials AT qingqingqian usingmachinelearningformortalitypredictionandriskstratificationinatezolizumabtreatedcancerpatientsintegrativeanalysisofeightclinicaltrials AT yanghong usingmachinelearningformortalitypredictionandriskstratificationinatezolizumabtreatedcancerpatientsintegrativeanalysisofeightclinicaltrials |