scholarly journals A Machine Learning Decision Support System (DSS) for Neuroendocrine Tumor Patients Treated with Somatostatin Analog (SSA) Therapy

Diagnostics ◽  
2021 ◽  
Vol 11 (5) ◽  
pp. 804
Author(s):  
Jasminka Hasic Telalovic ◽  
Serena Pillozzi ◽  
Rachele Fabbri ◽  
Alice Laffi ◽  
Daniele Lavacchi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Single Agent ◽  
Predictive Biomarkers ◽  
Progression Free Survival ◽  
Primary Site ◽  
Somatostatin Analog ◽  
Support Vector ◽  
Classification Models ◽  
First Line Therapy ◽  
Bayes Algorithm

The application of machine learning (ML) techniques could facilitate the identification of predictive biomarkers of somatostatin analog (SSA) efficacy in patients with neuroendocrine tumors (NETs). We collected data from 74 patients with a pancreatic or gastrointestinal NET who received SSA as first-line therapy. We developed three classification models to predict whether the patient would experience a progressive disease (PD) after 12 or 18 months based on clinic-pathological factors at the baseline. The dataset included 70 samples and 15 features. We initially developed three classification models with accuracy ranging from 55% to 70%. We then compared ten different ML algorithms. In all but one case, the performance of the Multinomial Naïve Bayes algorithm (80%) was the highest. The support vector machine classifier (SVC) had a higher performance for the recall metric of the progression-free outcome (97% vs. 94%). Overall, for the first time, we documented that the factors that mainly influenced progression-free survival (PFS) included age, the number of metastatic sites and the primary site. In addition, the following factors were also isolated as important: adverse events G3–G4, sex, Ki67, metastatic site (liver), functioning NET, the primary site and the stage. In patients with advanced NETs, ML provides a predictive model that could potentially be used to differentiate prognostic groups and to identify patients for whom SSA therapy as a single agent may not be sufficient to achieve a long-lasting PFS.

Randomized phase II study of carboplatin and paclitaxel with either linifanib or placebo for advanced nonsquamous NSCLC.

2012 ◽  
Vol 30 (15_suppl) ◽  
pp. 7512-7512
Author(s):  
Suresh S. Ramalingam ◽  
Mikhail Shtivelband ◽  
Ross A. Soo ◽  
Carlos H. Barrios ◽  
Anatoly Makhson ◽  
...  
Keyword(s):  
Single Agent ◽  
Objective Response ◽  
Predictive Biomarkers ◽  
Progression Free Survival ◽  
First Line ◽  
First Line Therapy ◽  
Line Therapy ◽  
Grade 3 ◽  
And Gender ◽  
Ecog Ps

7512 Background: Linifanib is a potent and selective inhibitor of VEGF and PDGF receptors with modest single-agent activity in NSCLC. We evaluated the combination of linifanib with carboplatin (C) and paclitaxel (P) for first-line therapy of advanced non-squamous NSCLC. Methods: Patients (pts) with stage IIIB/IV, non-squamous NSCLC, stratified by ECOG PS and gender, were randomized to receive up to six 3-wk cycles of C (AUC 6 mg/ml/min) and P (200 mg/m2) with daily placebo (Arm A), linifanib 7.5 mg (Arm B), or linifanib 12.5 mg (Arm C). The primary endpoint was progression-free survival (PFS); secondary endpoints included overall survival (OS), 12 m survival rate, and objective response rate (ORR). Safety was assessed by NCI-CTCAE v3.0. Results: 138 pts were randomized at 37 sites in 6 countries. Baseline characteristics were: median age, 61 y; men, 57%; smoker, 84%. Efficacy results are shown in the table. Thrombocytopenia was the only Grade 3/4 AE significantly higher on linifanib (Arm B: 16.7%; Arm C: 29.8%) vs. placebo (2.1%). Other adverse events (AEs) related to the dose of linifanib were diarrhea, thrombocytopenia, hypertension, weight loss, palmar-plantar erythrodysaesthesia syndrome, and hypothyroidism. Analysis of samples for predictive biomarkers including serum VEGF and placental growth factor are underway. Conclusions: The addition of linifanib to chemotherapy was tolerable at the doses tested and resulted in a significant improvement in PFS, with a modest survival improvement for Arm C in first-line therapy of advanced non-squamous NSCLC. [Table: see text]

scholarly journals Machine-Learning-Based Radiomics MRI Model for Survival Prediction of Recurrent Glioblastomas Treated with Bevacizumab

Diagnostics ◽  
2021 ◽  
Vol 11 (7) ◽  
pp. 1263
Author(s):  
Samy Ammari ◽  
Raoul Sallé de Chou ◽  
Tarek Assi ◽  
Mehdi Touat ◽  
Emilie Chouzenoux ◽  
...  
Keyword(s):  
Machine Learning ◽  
Therapeutic Option ◽  
Binary Classification ◽  
Progression Free Survival ◽  
Recurrent Glioblastoma ◽  
Machine Learning Algorithms ◽  
Survival Prediction ◽  
Classification Models ◽  
Angiogenic Therapy ◽  
Recurrent Gbm

Anti-angiogenic therapy with bevacizumab is a widely used therapeutic option for recurrent glioblastoma (GBM). Nevertheless, the therapeutic response remains highly heterogeneous among GBM patients with discordant outcomes. Recent data have shown that radiomics, an advanced recent imaging analysis method, can help to predict both prognosis and therapy in a multitude of solid tumours. The objective of this study was to identify novel biomarkers, extracted from MRI and clinical data, which could predict overall survival (OS) and progression-free survival (PFS) in GBM patients treated with bevacizumab using machine-learning algorithms. In a cohort of 194 recurrent GBM patients (age range 18–80), radiomics data from pre-treatment T2 FLAIR and gadolinium-injected MRI images along with clinical features were analysed. Binary classification models for OS at 9, 12, and 15 months were evaluated. Our classification models successfully stratified the OS. The AUCs were equal to 0.78, 0.85, and 0.76 on the test sets (0.79, 0.82, and 0.87 on the training sets) for the 9-, 12-, and 15-month endpoints, respectively. Regressions yielded a C-index of 0.64 (0.74) for OS and 0.57 (0.69) for PFS. These results suggest that radiomics could assist in the elaboration of a predictive model for treatment selection in recurrent GBM patients.

scholarly journals Financial Context News Sentiment Analysis for the Lithuanian Language

2021 ◽  
Vol 11 (10) ◽  
pp. 4443
Author(s):  
Rokas Štrimaitis ◽  
Pavel Stefanovič ◽  
Simona Ramanauskaitė ◽  
Asta Slotkienė
Keyword(s):  
Machine Learning ◽  
Sentiment Analysis ◽  
Short Term Memory ◽  
Machine Learning Algorithms ◽  
Supervised Machine Learning ◽  
Experimental Investigations ◽  
Support Vector ◽  
Applied Machine Learning ◽  
Bayes Algorithm ◽  
Website Content

Financial area analysis is not limited to enterprise performance analysis. It is worth analyzing as wide an area as possible to obtain the full impression of a specific enterprise. News website content is a datum source that expresses the public’s opinion on enterprise operations, status, etc. Therefore, it is worth analyzing the news portal article text. Sentiment analysis in English texts and financial area texts exist, and are accurate, the complexity of Lithuanian language is mostly concentrated on sentiment analysis of comment texts, and does not provide high accuracy. Therefore in this paper, the supervised machine learning model was implemented to assign sentiment analysis on financial context news, gathered from Lithuanian language websites. The analysis was made using three commonly used classification algorithms in the field of sentiment analysis. The hyperparameters optimization using the grid search was performed to discover the best parameters of each classifier. All experimental investigations were made using the newly collected datasets from four Lithuanian news websites. The results of the applied machine learning algorithms show that the highest accuracy is obtained using a non-balanced dataset, via the multinomial Naive Bayes algorithm (71.1%). The other algorithm accuracies were slightly lower: a long short-term memory (71%), and a support vector machine (70.4%).

scholarly journals Breast Cancer Type Classification Using Machine Learning

2021 ◽  
Vol 11 (2) ◽  
pp. 61
Author(s):  
Jiande Wu ◽  
Chindo Hicks
Keyword(s):  
Breast Cancer ◽  
Gene Expression ◽  
Machine Learning ◽  
Triple Negative Breast Cancer ◽  
Triple Negative ◽  
Genomic Research ◽  
Support Vector ◽  
Cancer Type ◽  
Classification Models

Background: Breast cancer is a heterogeneous disease defined by molecular types and subtypes. Advances in genomic research have enabled use of precision medicine in clinical management of breast cancer. A critical unmet medical need is distinguishing triple negative breast cancer, the most aggressive and lethal form of breast cancer, from non-triple negative breast cancer. Here we propose use of a machine learning (ML) approach for classification of triple negative breast cancer and non-triple negative breast cancer patients using gene expression data. Methods: We performed analysis of RNA-Sequence data from 110 triple negative and 992 non-triple negative breast cancer tumor samples from The Cancer Genome Atlas to select the features (genes) used in the development and validation of the classification models. We evaluated four different classification models including Support Vector Machines, K-nearest neighbor, Naïve Bayes and Decision tree using features selected at different threshold levels to train the models for classifying the two types of breast cancer. For performance evaluation and validation, the proposed methods were applied to independent gene expression datasets. Results: Among the four ML algorithms evaluated, the Support Vector Machine algorithm was able to classify breast cancer more accurately into triple negative and non-triple negative breast cancer and had less misclassification errors than the other three algorithms evaluated. Conclusions: The prediction results show that ML algorithms are efficient and can be used for classification of breast cancer into triple negative and non-triple negative breast cancer types.

Exploratory biomarker findings from cohort 2 of MODUL: An adaptable, phase 2, signal-seeking trial of fluoropyrimidine + bevacizumab ± atezolizumab maintenance therapy for BRAFwt metastatic colorectal cancer.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. 3570-3570
Author(s):  
Josep Tabernero ◽  
Axel Grothey ◽  
Dirk Arnold ◽  
Michel Ducreux ◽  
Peter J. O'Dwyer ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Metastatic Colorectal Cancer ◽  
Maintenance Treatment ◽  
Predictive Biomarkers ◽  
Experimental Treatment ◽  
Progression Free Survival ◽  
Phase 2 ◽  
First Line ◽  
Tissue Samples ◽  
First Line Therapy

3570 Background: MODUL is an adaptable, phase 2, signal-seeking trial testing novel agents as first-line therapy for predefined subgroups of patients with metastatic colorectal cancer (mCRC). Previously reported findings demonstrated that adding atezolizumab to fluoropyrimidine (FP)/bevacizumab as first-line maintenance treatment after induction with FOLFOX + bevacizumab did not improve efficacy outcomes in BRAFwt mCRC. Given these efficacy results, exploratory assessments on tumour samples were conducted to provide insights into factors that might affect efficacy of maintenance treatment and provide guidance on appropriate therapeutic strategies for BRAFwt mCRC. Methods: In patients with BRAFwt tumours (Cohort 2), experimental treatment was FP/bevacizumab + atezolizumab. Primary efficacy endpoint: progression-free survival (PFS). Overall survival (OS) was a secondary endpoint. Archival tissue samples from 104 patients were analysed by immunohistochemistry (IHC) at HistoGeneX (PD-L1; CD8/GrB/FoxP3). SP142 antibody was used for PD-L1 IHC analysis, which evaluated PD-L1low (IC 0–1) vs PD-L1high (IC > 1) in correlation with PFS and OS in the control and experimental arms. CD8/GrB/FoxP3 triplex staining was also performed to evaluate potential correlations with efficacy. Results: 445 patients with BRAFwt mCRC were randomised (2:1 ratio) to maintenance treatment in Cohort 2. Archival samples from 104 patients were analysed: FP/bevacizumab + atezolizumab (n = 82); FP/bevacizumab (n = 22). The biomarker evaluable population (BEP) for PD-L1 was n = 81 for FP/bevacizumab + atezolizumab [PD-L1low n = 35 (43%); PD-L1high n = 46 (57%)] and n = 22 for FP/bevacizumab [PD-L1low n = 16 (72%); PD-L1high n = 6 (28%)]. The BEP for CD8/GrB was n = 50 for FP/bevacizumab + atezolizumab and n = 16 for FP/bevacizumab. No difference in PFS or OS was observed in the FP/bevacizumab + atezolizumab vs FP/bevacizumab arms for PD-L1high [PFS: HR = 1.5 (95% CI 0.45−4.8), p = 0.51; OS: HR = 1.3 (95% CI 0.38−4.1), p = 0.71] or PD-L1low [PFS: HR = 0.92 (95% CI 0.47−1.8), p = 0.81; OS: HR = 0.78 (95% CI 0.4−1.5), p = 0.48]. Similar results were observed with CD8/GrBhigh [PFS: HR = 0.73 (95% CI 0.27−2.0), p = 0.55; OS: HR = 0.66 (95% CI 0.24−1.8), p = 0.44], CD8/GrBlow [PFS: HR = 1.0 (95% CI 0.42–2.5), p = 0.96; OS: HR = 0.73 (95% CI 0.3–1.8), p = 0.5], FoxP3high [PFS: HR = 0.97 (95% CI 0.37−2.5), p = 0.95; OS: HR = 0.95 (95% CI 0.36−2.5), p = 0.91] and FoxP3low [PFS: HR = 0.73 (95% CI 0.29−1.9), p = 0.53; OS: HR = 0.5 (95% CI 0.19−1.3), p = 0.18]. Conclusions: These findings suggest that PD-L1, CD8/GrB and FoxP3 might not be predictive biomarkers in BRAFwt mCRC. Further analyses are needed to further evaluate potential predictive and prognostic factors of response in this setting. Clinical trial information: NCT02291289.

scholarly journals The role of abemaciclib in treatment of advanced breast cancer

2018 ◽  
Vol 10 ◽  
pp. 175883591877692 ◽  
Author(s):  
Amelia McCartney ◽  
Erica Moretti ◽  
Giuseppina Sanna ◽  
Marta Pestrin ◽  
Emanuela Risi ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Advanced Breast Cancer ◽  
Single Agent ◽  
Endocrine Resistance ◽  
Predictive Biomarkers ◽  
Progression Free Survival ◽  
Advanced Breast ◽  
Significant Activity ◽  
Epidermal Growth

Until recently, the mainstay of treatment in the majority of hormone receptor (HR)-positive, human epidermal growth factor 2 receptor (HER2)-negative advanced breast cancer (ABC) has consisted of single-agent endocrine therapy (ET). However, as understanding of endocrine resistance has grown, newer targeted agents have come to the fore. Inhibition of cyclin-dependent kinase complexes 4 and 6 (CDK4/6) combined with ET has shown significant activity in HR+ HER2− ABC, with impressive results in terms of progression-free survival (PFS) when compared with ET alone. This review summarizes the seminal findings pertaining to CDK4/6 inhibition in this population, specifically focusing on abemaciclib, contrasted with palbociclib and ribociclib. Potential directions for future studies are discussed, as a way of addressing outstanding issues such as establishing optimal treatment sequencing and agent combinations, appropriate patient selection to derive maximal benefits, predictive biomarkers and the employment of CDK4/6 inhibition beyond the ABC setting.

Classification study of solvation free energies of organic molecules using machine learning techniques

RSC Advances ◽  
2014 ◽  
Vol 4 (106) ◽  
pp. 61624-61630 ◽  
Author(s):  
N. S. Hari Narayana Moorthy ◽  
Silvia A. Martins ◽  
Sergio F. Sousa ◽  
Maria J. Ramos ◽  
Pedro A. Fernandes
Keyword(s):  
Machine Learning ◽  
Support Vector Machine ◽  
Random Forest ◽  
Organic Molecules ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Classification Models ◽  
Free Energies ◽  
Learning Techniques ◽  
Solvation Free Energies

Classification models to predict the solvation free energies of organic molecules were developed using decision tree, random forest and support vector machine approaches and with MACCS fingerprints, MOE and PaDEL descriptors.

EVERSUN: A phase II trial of everolimus alternating with sunitinib as first-line therapy for advanced renal cell carcinoma (RCC) (ANZUP trial 0901).

2012 ◽  
Vol 30 (15_suppl) ◽  
pp. TPS4681-TPS4681 ◽  
Author(s):  
Ian D. Davis ◽  
Val Gebski ◽  
Mark D. Chatfield ◽  
Peter S. Grimison ◽  
George Kannourakis ◽  
...  
Keyword(s):  
Disease Progression ◽  
Phase Ii ◽  
Prognostic Score ◽  
Single Agent ◽  
Locally Advanced ◽  
Progression Free Survival ◽  
First Line ◽  
Free Survival ◽  
First Line Therapy ◽  
Line Therapy

TPS4681 Background: Treatment of RCC has improved due to better understanding of its biology. New targeted therapies have improved time to progression and overall survival but the optimal sequencing of agents is unknown. Currently drugs are given sequentially, usually starting with sunitinib and often followed by an mTOR inhibitor or another VEGFR-targeted therapy, but resistance to both drugs eventually occurs probably due to host adaptive responses. We hypothesize that resistance might be delayed by planned alternation of treatments. Methods: EVERSUN is a single-arm, two-stage, multicenter, phase II clinical trial aiming to determine the activity and safety of an alternating regimen of two therapies with different targets (sunitinib and everolimus) in patients with advanced RCC. Key eligibility criteria: RCC with a clear cell component; metastatic or locally advanced disease not suitable for resection; ECOG performance status 0-1; low or intermediate MSKCC prognostic score. The primary endpoint is the status of being alive and progression-free (RECIST 1.1) 6 months after registration. Target accrual of 55 subjects gives 95% power and 95% confidence to distinguish between 6-month progression free survival rates of 64% or lower vs 84% or higher using a Simon 2-stage minimax design. The criteria for further evaluation come from the pivotal trial of single agent sunitinib as first line therapy for RCC, in which the 6-month progression free survival rate was 74%. Trial treatment is administered in 12-week (wk) cycles consisting of 4 wks of sunitinib (50 mg daily) followed by 2 wks rest, followed by 5 wks of everolimus (10 mg daily) followed by 1 wk rest. Disease progression is interpreted as failure of the most recent drug taken. Participants who stop one drug because of toxicity or disease progression, on or before the 6 month assessment, will continue the other drug until subsequent progression or prohibitive toxicity on the second drug. EVERSUN is an ANZUP Cancer Trials Group Ltd. trial coordinated by the NHMRC Clinical Trials Centre. Accrual commenced in September 2010 with 38/55 participants recruited as of the 31-Jan-12 from 17 Australian sites (ACTRN12609000643279).

Phase II trial of volasertib (BI 6727) versus chemotherapy (CT) in platinum-resistant/refractory ovarian cancer (OC).

2013 ◽  
Vol 31 (15_suppl) ◽  
pp. 5504-5504 ◽  
Author(s):  
Eric Pujade-Lauraine ◽  
Beatrice E. Weber ◽  
Isabelle Ray-Coquard ◽  
Ignace Vergote ◽  
Frédéric Selle ◽  
...  
Keyword(s):  
Kinase Inhibitor ◽  
Demographic Data ◽  
Single Agent ◽  
Pegylated Liposomal Doxorubicin ◽  
Predictive Biomarkers ◽  
Progression Free Survival ◽  
Overall Response ◽  
Platinum Resistant ◽  
Measurable Disease ◽  
Platinum Refractory

5504 Background: Volasertib (V) is a potent and selective cell cycle kinase inhibitor that induces mitotic arrest and apoptosis by targeting Polo-like kinases. This study investigated V vs CT as 3rd- or 4th-line therapy in patients (pts) with platinum-refractory or resistant OC. Methods: Pts were randomized to V 300 mg IV Q3W or investigator’s choice single-agent CT (pegylated liposomal doxorubicin, topotecan, paclitaxel, gemcitabine) until progression or intolerance. Primary endpoint was 24-wk disease control rate (DCR; % of pts with complete/partial response [PR] or stable disease [SD]). Secondary endpoints included safety, progression-free survival (PFS), best overall response (RECIST 1.1) and explorative biomarkers. Results: 109 pts received V (n=54) or CT (n = 55) for a median (range) of 95 (22–716) and 114 (7–351) days, respectively. Demographic data were balanced between the treatment arms. Overall, median age was 62.0 yr; ECOG PS 0–1: 103 pts; 2 prior CTs: 51 pts; ≥3 prior CTs: 57 pts; platinum-resistant: 78 pts; platinum-refractory: 31 pts; measurable disease: 89 pts. 24-wk DCR (95% CI) for V vs CT was 31% (18–43) vs 43% (30–57), and median PFS was 13.1 vs 20.6 wks (HR = 1.01; 95% CI: 0.66–1.53). Six V pts vs 0 CT pts are ongoing for PFS 1 yr after randomization. Best overall response in pts with measurable disease (V/CT) was: PR, 7/8 pts; SD, 24/24 pts. Adverse events (AEs) led to discontinuation in 20 pts (V, n = 5; CT, n = 15); no V pts and 8 CT pts discontinued due to treatment-related AEs (including neuropathy in 3 CT pts). Most frequent all grade AEs (% of pts) regardless of relatedness were neutropenia (61%), anemia (54%), thrombocytopenia (46%), nausea (37%) and asthenia (33%) with V, and asthenia/nausea (47% each), abdominal pain (38%), anemia (36%) and neutropenia/vomiting (31% each) with CT. There were 3 fatal AEs per arm. Conclusions: Single-agent V showed antitumor activity in OC in a range similar to CT. AEs with V were mainly hematologic and manageable, with fewer non-hematologic AEs than CT. Exploration of potential predictive biomarkers for V activity is ongoing. Clinical trial information: NCT01121406.

Updated meta-analysis (MA) of salvage therapy for metastatic urothelial cancer (mUC): Comparing outcomes of immunotherapy (IT) versus single agent and doublet chemotherapy (CT).

2017 ◽  
Vol 35 (6_suppl) ◽  
pp. 341-341 ◽  
Author(s):  
Andrea Necchi ◽  
Daniele Raggi ◽  
Guru Sonpavde ◽  
Patrizia Giannatempo ◽  
Luigi Mariani ◽  
...  
Keyword(s):  
Salvage Therapy ◽  
Meta Analysis ◽  
Single Agent ◽  
Predictive Biomarkers ◽  
Progression Free Survival ◽  
Random Effects Models ◽  
Level Data ◽  
Metastatic Urothelial Cancer ◽  
Programmed Cell Death 1 ◽  
Ecog Ps

341 Background: Doublet CT demonstrated improved overall response rate (ORR) and progression-free survival (PFS) compared to single agent CT in a MA of patients (pts) receiving II line CT in trials of mUC ( Raggi et al, Ann Oncol 2016). We aimed to update the analyses adding trials of salvage IT. Methods: We searched for arms of phase 2 or 3 studies of salvage single agent anti-programmed cell death-1/Ligand-1 (PD-1/PD-L1) agents pembrolizumab, nivolumab, atezolizumab, durvalumab, avelumab, single agent CT and doublet CT. Random-effects models were used to pool trial level data according to treatment arm, including ORR, median PFS (mPFS), median overall survival (mOS). Univariable (UVA) and multivariable (MVA) analyses were performed, adjusting for ECOG-PS 2 and liver metastases. Results: 7 IT trials were analyzed (n=1,041), 22 arms received single agent CT (n=1,202), and 24 doublet CT (n=708). The pooled ORR was: 21.2% (95%CI: 14.9-29.2) with IT, 14.2% (95%CI: 11.1-17.9) with single agent CT and 31.9% (95%CI: 27.3-36.9) with doublet CT. Pooled mPFS was 1.8, 2.69 and 4.05 months, respectively. Pooled mOS was 8.27, 6.98 and 8.50 months. Pooled median ORR and mOS of IT for PD-L1+ pts were: 30.7% (95%CI: 23.2-39.2) and 11.60 months. Results of UVA and MVA are shown in the table. Only UVA was possible for PD-L1+ pts. Conclusions: Among trials of salvage therapy for mUC, IT was associated with significantly higher ORR and mOS in PD-L1+ pts compared with single agent or doublet CT, while significant differences were not seen in unselected pts. These results are hypothesis-generating and suggest the importance of developing companion predictive biomarkers. [Table: see text]

Sign in / Sign up

Export Citation Format

Share Document