Machine learning model to predict oncologic outcomes for drugs in randomized clinical trials

Schperberg, Alexander; Boichard, Amélie; Tsigelny, Igor F.; Richard, Stéphane; Kurzrock, Razelle

doi:10.1002/ijc.33240

Cited by 16 publications

(18 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…PushpanjaliGupta et al(26) developed a machine learning model based on the random forest algorithm to predict ve-year disease-free survival in patients with colon cancer with an accuracy of 84% and an AUC value of 0.82 ± 0.10. There are many more studies on machine learning models based on random forests (27)(29) The AUC values of ve predictive models in our study are 0.8931 for XGBoost, 0.8906 for SVM, 0.8512 for NB, and 0.9088 for RF, 0.8319 for LR.…”

mentioning

confidence: 64%

Development and interpretation of a clinicopathological-based model for the identification of microsatellite instability in Colorectal Cancer

Jiang

Cao

Yan

et al. 2022

Preprint

View full text Add to dashboard Cite

Background:Chemotherapy is not recommended for patients with deficient mismatch repair(dMMR)in colorectal cancer(CRC), so assessing the status of MMR is crucial for the selection of subsequent treatment. Therefore, this study aims to build predictive models to accurately and rapidly identify dMMR.Methods:A retrospective analysis based on the clinicopathological data of patients with colorectal cancer from May 2017 to December 2019 at Wuhan Union Hospital was performed. The variables were subjected to co-linearity analysis, least absolute shrinkage and selection operator (LASSO) regression and random forest (RF) feature screening. Four sets of machine learning models (i.e., extreme gradient boosting (XGBoost), support vector machine (SVM), naive bayes (NB), and random forest and a conventional logistic regression (LR) model were built for training and testing. Receiver operating characteristic (ROC) curves are plotted to evaluate the predictive performance.Findings:A total of 2279 patients were included in the analysis and were randomly divided into a training group and a test group.The final 12 clinicopathological features were incorporated into the predictive models. The Area Under the curve (AUC) values of the five predictive models were 0.8931 in XGBoost, 0.8906 in SVM, 0.8512 in NB, 0.9088 in RF and 0.8319 in LR. The results showed that the random forest exhibited the best recognition ability and outperformed the conventional logistic regression method in identifing dMMR and proficient mismatch repair (pMMR).Conclusion:Our predictive models built on routine clinicopathological data can significantly improve the diagnostic performance of dMMR and pMMR. Meanwhile, four machine learning models outperformed conventional logistic regression model.

show abstract

mentioning

confidence: 64%

Development and interpretation of a clinicopathological-based model for the identification of microsatellite instability in Colorectal Cancer

Jiang

Cao

Yan

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Schperberg et al combined clinical trials, drug-related biomarkers, and molecular profile information to construct an RF model to predict drug oncologic outcomes in randomized clinical trials. The Spearman correlation ( ) between their predicted model's and actual outcomes was statistically significant (progression-free survival (PFS): = 0.879, overall survival (OS): = 0.878, P < .0001) 79 . Nasief et al collected CT images of patients during chemotherapy and compared the changes in radiomics features therein.…”

Section: Ai In Prognosismentioning

confidence: 98%

“…These facts make it challenging to predict the prognosis of PC. Due to its excellent computational power, AI was used to analyze PC prognoses, including survival time 204 - 221 , recurrence risk 78 , 221 - 224 , metastasis 225 - 230 , therapy response 79 - 81 , 231 - 240 , etc.…”

Section: Ai In Prognosismentioning

confidence: 99%

See 1 more Smart Citation

Artificial intelligence in pancreatic cancer

Huang¹,

Huang²,

Zhang³

et al. 2022

Theranostics

View full text Add to dashboard Cite

Pancreatic cancer is the deadliest disease, with a five-year overall survival rate of just 11%. The pancreatic cancer patients diagnosed with early screening have a median overall survival of nearly ten years, compared with 1.5 years for those not diagnosed with early screening. Therefore, early diagnosis and early treatment of pancreatic cancer are particularly critical. However, as a rare disease, the general screening cost of pancreatic cancer is high, the accuracy of existing tumor markers is not enough, and the efficacy of treatment methods is not exact. In terms of early diagnosis, artificial intelligence technology can quickly locate high-risk groups through medical images, pathological examination, biomarkers, and other aspects, then screening pancreatic cancer lesions early. At the same time, the artificial intelligence algorithm can also be used to predict the survival time, recurrence risk, metastasis, and therapy response which could affect the prognosis. In addition, artificial intelligence is widely used in pancreatic cancer health records, estimating medical imaging parameters, developing computer-aided diagnosis systems, etc. Advances in AI applications for pancreatic cancer will require a concerted effort among clinicians, basic scientists, statisticians, and engineers. Although it has some limitations, it will play an essential role in overcoming pancreatic cancer in the foreseeable future due to its mighty computing power.

show abstract

“…In drug discovery (DD), a plethora of techniques that are based on ML are being published every month. These are applied to many of the stages of the DD pipeline, from binding site characterization and prediction of ligand affinities to calculations of ADMET properties [1,2] and development and analysis of clinical trials [3,4]. Still, the ML application that has produced the largest impact on DD is possibly the application to protein structure prediction, such as RosettaFold [5] and AlphaFold2 (AF2) [6].…”

Section: Introductionmentioning

confidence: 99%

Are Deep Learning Structural Models Sufficiently Accurate for Virtual Screening? Application of Docking Algorithms to AlphaFold2 Predicted Structures

Díaz-Rovira

Martín²,

Beuming³

et al. 2022

Preprint

View full text Add to dashboard Cite

Machine learning protein structure prediction, such as RosettaFold and AlphaFold2, have impacted the structural biology field, raising a fair amount of discussion around its potential role in drug discovery. While we find some preliminary studies addressing the usage of these models in virtual screening, none of them focus on the prospect of hit-finding in a real-world virtual screen with a target with low sequence identity. In order to address this, we have developed an AlphaFiold2 version where we exclude all structural templates with more than 30% sequence identity. In a previous study, we used those models in conjunction with state of the art free energy perturbation methods. In this work we focus on using them in rigid receptor ligand docking. Our results indicate that using out-of-the-box Alphafold2 models is not an ideal scenario; one might think in including some post processing modeling to drive the binding site into a more realistic holo target model.

show abstract

Machine learning model to predict oncologic outcomes for drugs in randomized clinical trials

Cited by 16 publications

References 58 publications

Development and interpretation of a clinicopathological-based model for the identification of microsatellite instability in Colorectal Cancer

Development and interpretation of a clinicopathological-based model for the identification of microsatellite instability in Colorectal Cancer

Artificial intelligence in pancreatic cancer

Are Deep Learning Structural Models Sufficiently Accurate for Virtual Screening? Application of Docking Algorithms to AlphaFold2 Predicted Structures

Contact Info

Product

Resources

About