The impact of the molecular classification of glioblastoma on the interpretation of therapeutic clinical trial results

Abstract Background Eligibility criteria are the primary strategy for screening the target participants of a clinical trial. Automated classification of clinical trial eligibility criteria text by using machine learning methods improves recruitment efficiency to reduce the cost of clinical research. However, existing methods suffer from poor classification performance due to the complexity and imbalance of eligibility criteria text data. Methods An ensemble learning-based model with metric learning is proposed for eligibility criteria classification. The model integrates a set of pre-trained models including Bidirectional Encoder Representations from Transformers (BERT), A Robustly Optimized BERT Pretraining Approach (RoBERTa), XLNet, Pre-training Text Encoders as Discriminators Rather Than Generators (ELECTRA), and Enhanced Representation through Knowledge Integration (ERNIE). Focal Loss is used as a loss function to address the data imbalance problem. Metric learning is employed to train the embedding of each base model for feature distinguish. Soft Voting is applied to achieve final classification of the ensemble model. The dataset is from the standard evaluation task 3 of 5th China Health Information Processing Conference containing 38,341 eligibility criteria text in 44 categories. Results Our ensemble method had an accuracy of 0.8497, a precision of 0.8229, and a recall of 0.8216 on the dataset. The macro F1-score was 0.8169, outperforming state-of-the-art baseline methods by 0.84% improvement on average. In addition, the performance improvement had a p-value of 2.152e-07 with a standard t-test, indicating that our model achieved a significant improvement. Conclusions A model for classifying eligibility criteria text of clinical trials based on multi-model ensemble learning and metric learning was proposed. The experiments demonstrated that the classification performance was improved by our ensemble model significantly. In addition, metric learning was able to improve word embedding representation and the focal loss reduced the impact of data imbalance to model performance.

Download Full-text

Patient Income Level and Cancer Clinical Trial Participation

Journal of Clinical Oncology ◽

10.1200/jco.2012.45.4553 ◽

2013 ◽

Vol 31 (5) ◽

pp. 536-542 ◽

Cited By ~ 114

Author(s):

Joseph M. Unger ◽

Dawn L. Hershman ◽

Kathy S. Albain ◽

Carol M. Moinpour ◽

Judith A. Petersen ◽

...

Keyword(s):

Clinical Trial ◽

Clinical Trials ◽

Universal Access ◽

Participation Rate ◽

Treatment Decision ◽

Trial Participation ◽

Lower Income ◽

Clinical Trial Participation ◽

Clinical Trial Results ◽

The Impact

Purpose Studies have shown an association between socioeconomic status (SES) and quality of oncology care, but less is known about the impact of patient SES on clinical trial participation. Patients and Methods We assessed clinical trial participation patterns according to important SES (income, education) and demographic factors in a large sample of patients surveyed via an Internet-based treatment decision tool. Logistic regression, conditioning on type of cancer, was used. Attitudes toward clinical trials were assessed using prespecified items about treatment, treatment tolerability, convenience, and cost. Results From 2007 to 2011, 5,499 patients were successfully surveyed. Forty percent discussed clinical trials with their physician, 45% of discussions led to physician offers of clinical trial participation, and 51% of offers led to clinical trial participation. The overall clinical trial participation rate was 9%. In univariate models, older patients (P = .002) and patients with lower income (P = .001) and education (P = .02) were less likely to participate in clinical trials. In a multivariable model, income remained a statistically significant predictor of clinical trial participation (odds ratio, 0.73; 95% CI, 0.57 to 0.94; P = .01). Even in patients age ≥ 65 years, who have universal access to Medicare, lower income predicted lower trial participation. Cost concerns were much more evident among lower-income patients (P < .001). Conclusion Lower-income patients were less likely to participate in clinical trials, even when considering age group. A better understanding of why income is a barrier may help identify ways to make clinical trials better available to all patients and would increase the generalizability of clinical trial results across all income levels.

Download Full-text

The basal subtype to predict clinical benefit from neoadjuvant chemotherapy: Final results from a phase II clinical trial of DDMVAC plus bevacizumab.

Journal of Clinical Oncology ◽

10.1200/jco.2015.33.7_suppl.291 ◽

2015 ◽

Vol 33 (7_suppl) ◽

pp. 291-291 ◽

Cited By ~ 1

Author(s):

Arlene O. Siefker-Radtke ◽

Woonyoung Choi ◽

Sima P. Porten ◽

Yu Shen ◽

Ashish M. Kamat ◽

...

Keyword(s):

Gene Expression ◽

Clinical Trial ◽

Neoadjuvant Chemotherapy ◽

Bone Metastases ◽

Gene Expression Profiling ◽

Expression Profiling ◽

Urothelial Cancer ◽

Basal Subtype ◽

Clinical Trial Results ◽

The Impact

291 Background: Gene expression profiling (GEP) suggests 3 main subtypes of urothelial cancer: basal, which historically has the worst prognosis with high proliferation and HIF-1 expression; p53-like, with decreased proliferation and increased markers of extracellular matrix (ECM); and luminal which has increased proliferation compared to p53-like tumors. We hypothesized that GEP of transurethral resections (TUR) and cystectomy specimens from patients on a neoadjuvant trial would predict benefit from chemotherapy. Methods: Sixty patients enrolled on a neoadjuvant trial of DDMVAC+B. TUR and cystectomy specimens were available for gene expression profiling in 39 and 33 patients, respectively, with matched specimens in 23 patients. The validation set consisted of 49 patients treated with perioperative MVAC on a previously published clinical trial. Results: Chemotherapy was quite active with pT0N0 and ≤ pT1N0 down-staging rates of 38% and 53%, respectively. Basal tumors had improved survival compared to luminal and p53-like (5-year OS 91%, 73% and 36%, p=0.015). A validation cohort of patients treated with perioperative MVAC confirmed this survival benefit (5-year OS basal, luminal, and p53-like 77%, 57%, and 57%, respectively, p =0.027). The use of bevacizumab in basal tumors did not confirm evidence of significant benefit in these small numbers of patients (5-year OS bevacizumab: 91% vs MVAC: 77%, p=0.68) Bone metastases within 2 years associated exclusively with the p53-like subtype (p53-like: 100%, luminal: 0%, basal 0%, p≤0.001). The p53-like subtype was enriched at cystectomy (basal to p53-like in 3/5 (60%), luminal to p53-like in 5/7 (71%), suggesting chemo-resistance in p53-like tumors. Conclusions: In contrast to historical expectations, the basal subtype was predictive of clinical outcomes from neoadjuvant chemotherapy, reflecting the impact of chemotherapy on highly proliferative tumors. Bone metastases were associated with the p53-like subtype which is enriched for ECM. We can no longer think of urothelial cancer as one disease; subtyping should be considered for all tumors, and may have implications on selecting therapy. Clinical trial information: NCT00506155.

Download Full-text

FDA analysis of patient enrollment by region in clinical trials for approved oncological indications.

Journal of Clinical Oncology ◽

10.1200/jco.2017.35.15_suppl.2539 ◽

2017 ◽

Vol 35 (15_suppl) ◽

pp. 2539-2539 ◽

Cited By ~ 2

Author(s):

Bindu Kanapuru ◽

Harpreet Singh ◽

Lola A. Fashoyin-Aje ◽

Adrian Myers ◽

Geoffrey Kim ◽

...

Keyword(s):

Clinical Trial ◽

Clinical Trials ◽

North America ◽

Global Scale ◽

Malignant Solid Tumor ◽

Trends Over Time ◽

Clinical Trial Results ◽

The Baltic ◽

The Impact ◽

Patient Enrollment

2539 Background: Clinical trials are increasingly conducted on a global scale in an effort to accelerate accrual. This analysis attempts to quantify and characterize participants in trials submitted to support approval of drugs for oncology indications by the region of enrollment. Methods: Demographic information was extracted for patients enrolled in clinical trials submitted to the FDA from 2005-2015. Only trials submitted to support approval for malignant solid tumor or hematology indications were included. Countries were grouped into regions for further analysis. A total of 178,024 patients with information regarding age and country were included in this analysis. Results: Forty five percent (80,460) of clinical trial participants were enrolled from Europe, 36% (63,958) from North America (includes U.S.A and Canada) and 8.4% (14,975) from Asia. Countries in Latin America, Middle East/Africa and the Baltic States/Russia enrolled the remainder 10.5% of the patients. Among 99,556 participants < 65 years of age; 38.7% (38,538) were enrolled from North America, 40.5% (40,362) from Europe, 9.7 % (9674) from Asia and 11% from the rest of the regions. Europe enrolled the highest number of cancer patients aged 65 years or older; 51.1% (40,098) compared to 32.4% (25,420) from North America and 6.8 % (5301) from Asia. Conclusions: Majority of patients enrolled into clinical trials submitted for oncology drug approvals were from regions other than North America, with highest number enrolled from Europe particularly in the older age group. While it is interesting to speculate, the reasons for differential enrollment of patients between Europe and North America and the impact of these findings on interpretation of clinical trial results need additional exploration. Analysis of trends over time may be useful to address this issue. [Table: see text]

Download Full-text

Positive Outcomes Influence the Rate and Time to Publication, but Not the Impact Factor of Publications of Clinical Trial Results

PLoS ONE ◽

10.1371/journal.pone.0054583 ◽

2013 ◽

Vol 8 (1) ◽

pp. e54583 ◽

Cited By ~ 777

Author(s):

Pilar Suñé ◽

Josep Maria Suñé ◽

J. Bruno Montoro

Keyword(s):

Clinical Trial ◽

Impact Factor ◽

Positive Outcomes ◽

Clinical Trial Results ◽

The Impact

Download Full-text

The significance of NNT parameter for the evaluation of therapy effectiveness

Aktualności Neurologiczne ◽

10.15557/an.2016.0022 ◽

2016 ◽

Vol 16 (4) ◽

pp. 175-179

Author(s):

Andrzej Potemkowski ◽

Keyword(s):

Multiple Sclerosis ◽

Health Care ◽

Clinical Trial ◽

Magnetic Resonance ◽

Magnetic Resonance Images ◽

Disability Progression ◽

Advantages And Disadvantages ◽

Number Of Patients ◽

Clinical Trial Results ◽

The Impact

Clinical trials provide a practicing clinician with an abundance if not excess of data to draw on when comparing treatment methods and deciding the potentially optimal therapeutic option for a given patient. NNT, or number needed to treat, has been identified as one of the parameters useful for assessing the effectiveness of therapy. It represents the number of patients who must undergo a given health-care intervention instead of another one to see a difference in the effectiveness of obtaining a desired outcome within a set timeframe. NNT is a derivative of absolute risk reduction (ARR) or initial risk and its relative reduction. It represents the relative superiority of a given treatment. NNT is primarily used for comparing the advantages and disadvantages of alternative health-care interventions, and its assessment is important for estimating the clinical value of statistically significant clinical trial results. Utilization of NNT allows to predict therapy outcome both in terms of its effectiveness and tolerance. Also, clinical trial results presented in the form of NNT may be easily shared with patients, their families, and the institutions deciding the availability of a given drug. In multiple sclerosis, clinical trial results have been concerned with the impact of therapeutics on decrease of annualized relapse rate (ARR) and reduction of lesions visible in magnetic resonance images as well as slowing of disability progression. Analyses of first-line multiple sclerosis treatments reveal their NNT referred to prevention of relapses, disability progression and lesions in magnetic resonance image to vary significantly. Similar differences exist across NNT values established for second-line treatments or the oral therapies being currently introduced. The data clearly show that when evaluating clinical trial results, it is not enough to consider only given parameters, as they must all be critically and constructively analysed. NNT’s importance is also stressed as a clear parameter to be used for the evaluation of economic outcomes in healthcare.

Download Full-text

Genetics and genomics of pulmonary arterial hypertension

European Respiratory Journal ◽

10.1183/13993003.01899-2018 ◽

2019 ◽

Vol 53 (1) ◽

pp. 1801899 ◽

Cited By ~ 72

Author(s):

Nicholas W. Morrell ◽

Micheala A. Aldred ◽

Wendy K. Chung ◽

C. Gregory Elliott ◽

William C. Nichols ◽

...

Keyword(s):

Pulmonary Arterial Hypertension ◽

Arterial Hypertension ◽

Genetic Diagnosis ◽

Molecular Classification ◽

Incomplete Penetrance ◽

Therapeutic Approaches ◽

Genomic Studies ◽

Pulmonary Arterial ◽

The Impact

Since 2000 there have been major advances in our understanding of the genetic and genomics of pulmonary arterial hypertension (PAH), although there remains much to discover. Based on existing knowledge, around 25–30% of patients diagnosed with idiopathic PAH have an underlying Mendelian genetic cause for their condition and should be classified as heritable PAH (HPAH). Here, we summarise the known genetic and genomic drivers of PAH, the insights these provide into pathobiology, and the opportunities afforded for development of novel therapeutic approaches. In addition, factors determining the incomplete penetrance observed in HPAH are discussed. The currently available approaches to genetic testing and counselling, and the impact of a genetic diagnosis on clinical management of the patient with PAH, are presented. Advances in DNA sequencing technology are rapidly expanding our ability to undertake genomic studies at scale in large cohorts. In the future, such studies will provide a more complete picture of the genetic contribution to PAH and, potentially, a molecular classification of this disease.

Download Full-text