P069 Artificial intelligence (AI)-filtered Videos for Accelerated Scoring of Colonoscopy Videos in Ulcerative Colitis Clinical Trials

Abstract Background Endoscopic assessment is a critical procedure to assess the improvement of mucosa and response to therapy, and therefore a pivotal component of clinical trial endpoints for IBD. Central scoring of endoscopic videos is challenging and time consuming. We evaluated the feasibility of using an Artificial Intelligence (AI) algorithm to automatically produce filtered videos where the non-readable portions of the video are removed, with the aim of accelerating the scoring of endoscopic videos. Methods The AI algorithm was based on a Convolutional Neural Network trained to perform a binary classification task. This task consisted of assigning the frames in a colonoscopy video to one of two classes: “readable” or “unreadable.” The algorithm was trained using annotations performed by two data scientists (BG, FA). The criteria to consider a frame “readable” were: i) the colon walls were within the field of view; ii) contrast and sharpness of the frame were sufficient to visually inspect the mucosa, and iii) no presence of artifacts completely obstructing the visibility of the mucosa. The frames were extracted randomly from 351 colonoscopy videos of the etrolizumab EUCALYPTUS (NCT01336465) Phase II ulcerative colitis clinical trial. Evaluation of the performance of the AI algorithm was performed on colonoscopy videos obtained as part of the etrolizumab HICKORY (NCT02100696) and LAUREL (NCT02165215) Phase III ulcerative colitis clinical trials. Each video was filtered using the AI algorithm, resulting in a shorter video where the sections considered unreadable by the AI algorithm were removed. Each of three annotators (EG, MM and MD) was randomly assigned an equal number of AI-filtered videos and raw videos. The gastroenterologist was tasked to score temporal segments of the video according to the Mayo Clinic Endoscopic Subscore (MCES). Annotations were performed by means of an online annotation platform (Virgo Surgical Video Solutions, Inc). Results We measured the time it took the annotators to score raw and AI-filtered videos. We observed a statistically significant reduction (Mann Whitney U test p-value=0.039) in the median time spent by the annotators scoring raw videos (10.59∓ 0.94 minutes) with respect to the time spent scoring AI-filtered videos (9.51 ∓ 0.92 minutes), with a substantial intra-rater agreement when evaluating highlight and raw videos (Cohen’s kappa 0.92 and 0.55 for experienced and junior gastroenterologists respectively). Conclusion Our analysis shows that AI can be used reliably as an assisting tool to automatically remove non-readable time segments from full colonoscopy videos. The use of our proposed algorithm can lead to reduced annotation times in the task of centrally reading colonoscopy videos.

Download Full-text

Training and deploying a deep learning model for endoscopic severity grading in ulcerative colitis using multicenter clinical trial data

Therapeutic Advances in Gastrointestinal Endoscopy ◽

10.1177/2631774521990623 ◽

2021 ◽

Vol 14 ◽

pp. 263177452199062

Author(s):

Benjamin Gutierrez Becker ◽

Filippo Arcadu ◽

Andreas Thalhammer ◽

Citlalli Gamez Serna ◽

Owen Feehan ◽

...

Keyword(s):

Artificial Intelligence ◽

Ulcerative Colitis ◽

Clinical Trials ◽

Deep Learning ◽

Mayo Clinic ◽

Automated System ◽

Phase Iii ◽

Plain Language ◽

Data Set ◽

Endoscopic Videos

Introduction: The Mayo Clinic Endoscopic Subscore is a commonly used grading system to assess the severity of ulcerative colitis. Correctly grading colonoscopies using the Mayo Clinic Endoscopic Subscore is a challenging task, with suboptimal rates of interrater and intrarater variability observed even among experienced and sufficiently trained experts. In recent years, several machine learning algorithms have been proposed in an effort to improve the standardization and reproducibility of Mayo Clinic Endoscopic Subscore grading. Methods: Here we propose an end-to-end fully automated system based on deep learning to predict a binary version of the Mayo Clinic Endoscopic Subscore directly from raw colonoscopy videos. Differently from previous studies, the proposed method mimics the assessment done in practice by a gastroenterologist, that is, traversing the whole colonoscopy video, identifying visually informative regions and computing an overall Mayo Clinic Endoscopic Subscore. The proposed deep learning–based system has been trained and deployed on raw colonoscopies using Mayo Clinic Endoscopic Subscore ground truth provided only at the colon section level, without manually selecting frames driving the severity scoring of ulcerative colitis. Results and Conclusion: Our evaluation on 1672 endoscopic videos obtained from a multisite data set obtained from the etrolizumab Phase II Eucalyptus and Phase III Hickory and Laurel clinical trials, show that our proposed methodology can grade endoscopic videos with a high degree of accuracy and robustness (Area Under the Receiver Operating Characteristic Curve = 0.84 for Mayo Clinic Endoscopic Subscore ⩾ 1, 0.85 for Mayo Clinic Endoscopic Subscore ⩾ 2 and 0.85 for Mayo Clinic Endoscopic Subscore ⩾ 3) and reduced amounts of manual annotation. Plain language summary Patient, caregiver and provider thoughts on educational materials about prescribing and medication safety Artificial intelligence can be used to automatically assess full endoscopic videos and estimate the severity of ulcerative colitis. In this work, we present an artificial intelligence algorithm for the automatic grading of ulcerative colitis in full endoscopic videos. Our artificial intelligence models were trained and evaluated on a large and diverse set of colonoscopy videos obtained from concluded clinical trials. We demonstrate not only that artificial intelligence is able to accurately grade full endoscopic videos, but also that using diverse data sets obtained from multiple sites is critical to train robust AI models that could potentially be deployed on real-world data.

Download Full-text

The case for introducing pre-registered confirmatory pharmacological pre-clinical studies

Journal of Cerebral Blood Flow & Metabolism ◽

10.1177/0271678x18760109 ◽

2018 ◽

Vol 38 (5) ◽

pp. 749-754 ◽

Cited By ~ 1

Author(s):

Olivia Kiwanuka ◽

Bo-Michael Bellander ◽

Anders Hånell

Keyword(s):

Clinical Trial ◽

Traumatic Brain Injury ◽

Clinical Trials ◽

Brain Injury ◽

Network Analysis ◽

Clinical Studies ◽

Phase Iii ◽

Phase Iii Clinical Trial ◽

Phase Iii Clinical Trials ◽

Experimental Drugs

When evaluating the design of pre-clinical studies in the field of traumatic brain injury, we found substantial differences compared to phase III clinical trials, which in part may explain the difficulties in translating promising experimental drugs into approved treatments. By using network analysis, we also found cases where a large proportion of the studies evaluating a pre-clinical treatment was performed by inter-related researchers, which is potentially problematic. Subjecting all pre-clinical trials to the rigor of a phase III clinical trial is, however, likely not practically achievable. Instead, we repeat the call for a distinction to be made between exploratory and confirmatory pre-clinical studies.

Download Full-text

A Review of Phase III Clinical Trials of Prostate Cancer Chemoprevention

Annals of The Royal College of Surgeons of England ◽

10.1308/003588407x179125 ◽

2007 ◽

Vol 89 (3) ◽

pp. 207-211 ◽

Cited By ~ 18

Author(s):

JF Thorpe ◽

S Jain ◽

TH Marczylo ◽

AJ Gescher ◽

WP Steward ◽

...

Keyword(s):

Prostate Cancer ◽

Clinical Trial ◽

Clinical Trials ◽

Cancer Prevention ◽

Age Of Onset ◽

Cancer Chemoprevention ◽

Phase Iii ◽

Phase Iii Clinical Trials ◽

Prostate Cancer Prevention ◽

Phase Iii Trials

INTRODUCTION Prostate cancer is an excellent target for chemoprevention strategies; given its late age of onset, any delay in carcinogenesis would lead to a reduction in its incidence. This article reviews all the completed and on-going phase III trials in prostate cancer chemoprevention. PATIENTS AND METHODS All phase III trials of prostate cancer chemoprevention were identified within a Medline search using the keywords ‘clinical trial, prostate cancer, chemoprevention’. RESULTS In 2003, the Prostate Cancer Prevention Trial (PCPT) became the first phase III clinical trial of prostate cancer prevention. This landmark study was terminated early due to the 24.8% reduction of prostate cancer prevalence over a 7-year period in those men taking the 5α-reductase inhibitor, finasteride. This article reviews the PCPT and the interpretation of the excess high-grade prostate cancer (HGPC) cases in the finasteride group. The lack of relationship between cumulative dose and the HGPC cases, and the possible sampling error of biopsies due to gland volume reduction in the finasteride group refutes the suggestion that this is a genuine increase in HGPC cases. The other on-going phase III clinical trials of prostate cancer chemoprevention – the REDUCE study using dutasteride, and the SELECT study using vitamin E and selenium – are also reviewed. CONCLUSIONS At present, finasteride remains the only intervention shown in long-term prospective phase III clinical trials to reduce the incidence of prostate cancer. Until we have the results of trials using alternative agents including the on-going REDUCE and SELECT trials, the advice given to men interested in prostate cancer prevention must include discussion of the results of the PCPT. The increased rate of HGPC in the finasteride group continues to generate debate; however, finasteride may still be suitable for prostate cancer prevention, particularly in men with lower urinary tract symptoms.

Download Full-text

ACTR-47. PHASE-III TRIAL OF STANDARD CHEMOTHERAPY VS. CHEMOTHERAPY GUIDED BY CANCER STEM CELL TEST IN RECURRENT GLIOBLASTOMA (NCT03632135)

Neuro-Oncology ◽

10.1093/neuonc/noz175.089 ◽

2019 ◽

Vol 21 (Supplement_6) ◽

pp. vi24-vi24

Author(s):

Tulika Ranjan ◽

Dawit Aregawi ◽

Christine Lu-Emerson ◽

Jason Schroeder ◽

Mark Anderson ◽

...

Keyword(s):

Clinical Trial ◽

Overall Survival ◽

Median Overall Survival ◽

Tumor Response ◽

Chemotherapeutic Agents ◽

Response To Therapy ◽

Phase Iii ◽

Recurrence Time ◽

Cell Kill ◽

Recurrent Gbm

Abstract Stupp treatment protocol for patients with glioblastoma (GBM) has improved the median overall survival to 14.6 months. However, no investigations have defined effective strategies against recurrence and the prognosis of recurrent GBM patients remains poor. Personalized medicine with assay-guided treatment targeting chemotherapy resistant cancer stem cells (CSCs) alongside the bulk tumor cells is a new paradigm in cancer treatment that may result in improved patient’s outcome. We are using ChemoID, a CLIA and CAP certified CSC cytotoxicity assay for predicting response to chemotherapeutic agents. Our prospective analysis of 61 GBM patients demonstrated that ChemoID-guided treatment significantly improved tumor response. For every 5% increase in cell kill of CSCs by assay-guided chemotherapy, 12-month patient response (non-recurrence of cancer) increased 2.5-fold, OR=2.3 (p=0.01). We also found that median recurrence time was 20-months versus 3-months for patients with a positive (>40% cell kill) CSC test versus negative, whereas median recurrence time was 13-months versus 4-months for patients with a positive (>55% cell kill) bulk test versus negative. We are conducting a multi-institutional phase-III clinical trial (NCT03632135) to determine the clinical validity of the ChemoID assay as a predictor of clinical response in recurrent GBM. The study has been designed as a parallel group controlled clinical trial and the participants are randomized to either standard of care chemotherapy chosen by the physician or ChemoID-guided therapy. Response to therapy will be measured by MRI imaging using RANO criteria. Primary endpoint of median overall survival (OS) and secondary endpoints of OS at 6, 9, and 12 months, median progression free survival (PFS), PFS at 4, 6, 9, and 12 months, objective tumor response, time to recurrence, and quality of life will be measured. Trial is open and currently 22 subjects have been enrolled. Interim analysis of the trial will be conducted in approximately 12 months.

Download Full-text

Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI Extension

BMJ ◽

10.1136/bmj.m3164 ◽

2020 ◽

pp. m3164 ◽

Cited By ~ 6

Author(s):

Xiaoxuan Liu ◽

Samantha Cruz Rivera ◽

David Moher ◽

Melanie J Calvert ◽

Alastair K Denniston

Keyword(s):

Artificial Intelligence ◽

Clinical Trial ◽

Clinical Trials ◽

Clinical Trial Design ◽

Trial Protocols ◽

General Readership ◽

Peer Reviewers ◽

Multi Stakeholder ◽

Analysis Of Error

Abstract The CONSORT 2010 (Consolidated Standards of Reporting Trials) statement provides minimum guidelines for reporting randomised trials. Its widespread use has been instrumental in ensuring transparency when evaluating new interventions. More recently, there has been a growing recognition that interventions involving artificial intelligence (AI) need to undergo rigorous, prospective evaluation to demonstrate impact on health outcomes. The CONSORT-AI extension is a new reporting guideline for clinical trials evaluating interventions with an AI component. It was developed in parallel with its companion statement for clinical trial protocols: SPIRIT-AI. Both guidelines were developed through a staged consensus process, involving a literature review and expert consultation to generate 29 candidate items, which were assessed by an international multi-stakeholder group in a two-stage Delphi survey (103 stakeholders), agreed on in a two-day consensus meeting (31 stakeholders) and refined through a checklist pilot (34 participants). The CONSORT-AI extension includes 14 new items, which were considered sufficiently important for AI interventions, that they should be routinely reported in addition to the core CONSORT 2010 items. CONSORT-AI recommends that investigators provide clear descriptions of the AI intervention, including instructions and skills required for use, the setting in which the AI intervention is integrated, the handling of inputs and outputs of the AI intervention, the human-AI interaction and providing analysis of error cases. CONSORT-AI will help promote transparency and completeness in reporting clinical trials for AI interventions. It will assist editors and peer-reviewers, as well as the general readership, to understand, interpret and critically appraise the quality of clinical trial design and risk of bias in the reported outcomes.

Download Full-text

Obtaining Valid Laboratory Data in Clinical Trials Conducted in Resource Diverse Settings: Lessons Learned from a Microbicide Phase III Clinical Trial

PLoS ONE ◽

10.1371/journal.pone.0013592 ◽

2010 ◽

Vol 5 (10) ◽

pp. e13592 ◽

Cited By ~ 9

Author(s):

Tania Crucitti ◽

Katrien Fransen ◽

Rashika Maharaj ◽

Tom Tenywa ◽

Marguerite Massinga Loembé ◽

...

Keyword(s):

Clinical Trial ◽

Clinical Trials ◽

Laboratory Data ◽

Lessons Learned ◽

Phase Iii ◽

Phase Iii Clinical Trial

Download Full-text

Outcomes in white (W) patients (pts) and those of African (A) descent receiving adjuvant therapy for colon cancer (CC)

Journal of Clinical Oncology ◽

10.1200/jco.2007.25.18_suppl.4017 ◽

2007 ◽

Vol 25 (18_suppl) ◽

pp. 4017-4017 ◽

Cited By ~ 1

Author(s):

G. Yothers ◽

W. Blackstock ◽

N. Wolmark ◽

R. M. Goldberg ◽

M. J. O’Connell ◽

...

Keyword(s):

Clinical Trials ◽

Socioeconomic Factors ◽

Response To Therapy ◽

Phase Iii ◽

Clinical Factors ◽

Cox Models ◽

Time To Death ◽

Time To Recurrence ◽

Kaplan Meier ◽

Treatment Type

4017 Background: Published reports suggest that CC pts of A descent have inferior survival compared to W pts. Whether these differences are explained by clinical factors at diagnosis, socioeconomic factors impacting access to care, or intrinsic differences in the biology of the tumors or the response to therapy is unclear. Pts in clinical trials have data collected for important baseline clinical factors and should receive comparable oncologic care regardless of socioeconomic factors. Methods: We analyzed data from 13,435 individual pts on 11 phase III adjuvant CC trials accrued from 1977 to 2002. Analysis was restricted to stage II and III pts, with race reported as black or white. Endpoints were overall survival (OS - time to death), recurrence-free survival (RFS - time to recurrence or death), and recurrence-free interval (RFI - time to recurrence censoring for death). Cox models stratified by study controlled for gender, stage, age, and treatment type (rx) to determine the effect of race. Kaplan-Meier estimates (KM) were adjusted (adj) by the Xie-Liu method for study, gender, stage, age, and rx. Results: A pts (n=1134, 8.4%) were younger than W (median 58 vs 61, p<0.001) and more likely female (55 vs 45%, p<0.001). A pts had poorer OS than W pts ( table ). OS results were consistent in subsets defined by gender, stage, and age. RFS results were attenuated compared to OS, but still favored improved RFS in W pts ( table ). RFI results were further attenuated and not significantly different by race ( table ). Conclusions: Even with identical rx for CC in controlled clinical trials, A pts have poorer OS and RFS than W pts. The OS deficit was consistent across subgroups, and neither deficit was explained by differences in gender, stage, age, or rx. RFI was similar for both races, suggesting that the OS and RFS differences may be largely due to deaths unrelated to CC. [Table: see text] [Table: see text]

Download Full-text

Paucity of economic analyses of phase III clinical trials conducted by the cooperative oncology groups: A review of the evidence

Journal of Clinical Oncology ◽

10.1200/jco.2007.25.18_suppl.17052 ◽

2007 ◽

Vol 25 (18_suppl) ◽

pp. 17052-17052

Author(s):

K. Fitzner ◽

J. McKoy ◽

C. L. Bennett

Keyword(s):

Clinical Trial ◽

Clinical Trials ◽

Cost Effectiveness ◽

Cancer Care ◽

Phase Iii ◽

Economic Data ◽

Direct Medical Costs ◽

Phase Iii Clinical Trials ◽

Need To Evaluate ◽

Economic Analyses

17052 Background: Cancer care is expensive, accounting for $72 billion in direct medical costs. New oncology drugs are frequently costly, and can be > $100,000 per patient. Hence, assessments of the costs and cost-effectiveness of cancer pharmaceuticals alongside phase III clinical trials conducted by the NCI-sponsored cooperative oncology groups represents an important opportunity to generate relevant economic data. Methods: Review of published cost and cost-effectiveness analyses for cancer drugs conducted alongside phase III clinical trials conducted by the NCI-sponsored cooperative clinical trial groups. Results: See Table . Conclusions: Despite increasing concerns over the high costs of cancer pharmaceuticals and the need to evaluate the costs and cost-effectiveness of these agents, NCI sponsored phase III clinical trials rarely include economic assessments. Future phase III clinical trials with expensive new cancer agents conducted by cooperative clinical trials groups should include prospective economic assessments. [Table: see text] No significant financial relationships to disclose.

Download Full-text

Reconsenting patients with cancer on clinical trials: Does added risk influence continued participation?

Journal of Clinical Oncology ◽

10.1200/jco.2009.27.15_suppl.e15610 ◽

2009 ◽

Vol 27 (15_suppl) ◽

pp. e15610-e15610

Author(s):

A. Elegbede ◽

A. Andrei ◽

K. D. Holen

Keyword(s):

Clinical Trial ◽

Clinical Trials ◽

Adverse Event ◽

Side Effects ◽

Adverse Events ◽

Study Participant ◽

P Value ◽

Study Participation ◽

Grade 5 ◽

Study Results

e15610 Background: The general policy endorsed by multiple professional societies and cooperative groups regarding patients on cancer clinical trials states that subjects should be informed of new adverse events or significant developments during study participation and re-consented to continue on study. However, no information is known as to the effect of re-consenting on a patients’ decision to continue study participation. Our research question addresses how the severity of reported risk to other study participants will impact the subjects’ decision to continue participation in a clinical trial. Methods: We surveyed 34 patients with gastrointestinal (GI) tumors all of whom were currently enrolled in a clinical trial. The survey portrayed hypothetical adverse reactions affecting another study participant ranging from Grade 1 to Grade 5 according to the National Cancer Institutes Common Terminology Criteria for Adverse Effects v. 3.0. The survey asked about subjects’ opinions of the theoretical adverse event categorized as “would not be concerned,” “would be concerned, but would continue the study,” and “would discontinue the study.” Results: Patients willingness to continue the study was highest at Grade 1 with 97% of all participants. However, willingness to continue participation progressively declined as the severity of adverse events increased such that only 44% of participants would continue participation with a reported Grade 5 adverse event. Conclusions: Among surveyed GI cancer patients, willingness to continue participation in a clinical trial declined significantly as the severity of adverse events increased from Grade 1 to Grade 3 - 5 (p-value < 0.001. This could be due to multiple factors, including the terminal nature of the patients’ cancer, the side effects of study therapy and the patients’ response to study treatment. This data could produce a reasonable adverse event grade cut-off for re-consenting patients regarding new side effects. No significant financial relationships to disclose.

Download Full-text

A review of the patterns of docetaxel use for hormone refractory prostate cancer (HRPC) at the Princess Margaret Hospital

Journal of Clinical Oncology ◽

10.1200/jco.2009.27.15_suppl.e16161 ◽

2009 ◽

Vol 27 (15_suppl) ◽

pp. e16161-e16161

Author(s):

S. N. Chin ◽

L. Wang ◽

A. Lau ◽

M. Moore ◽

S. S. Sridhar

Keyword(s):

Clinical Trial ◽

Clinical Trials ◽

Clinical Practice ◽

Median Duration ◽

Response Rates ◽

Routine Clinical Practice ◽

Phase Iii ◽

Line Treatment ◽

Second Line ◽

First Line

e16161 Background: Docetaxel is standard of care for the treatment of HRPC, based on two large randomized clinical trials. The aim of this study was to determine if docetaxel use and effectiveness in routine clinical practice was similar to that seen in the TAX 327 randomized phase III clinical trial. Methods: A retrospective chart review was undertaken to assess patterns of docetaxel use for HRPC at our institution for the 2-year period since its approval for the first-line treatment of HRPC in 2005. Results: Eighty-eight patients, median age 71 and baseline PSA 107, received docetaxel in the first line setting. Main reasons for initiating docetaxel were rising PSA (90%) and progressive symptoms (71%). Eighteen percent of patients received docetaxel for rising PSA alone. A median of 7 cycles was administered. PSA response rates were 61%, time to response 1.5 months, and response duration 6.8 months. Disease progression was the most common reason for treatment discontinuation (36%). Main toxicities were fatigue (32%) and neuropathy (22%). Kaplan Meier survival analysis showed median duration of survival was 15.9 months (95% CI 12.4–20.5) from first drug use. 1-year survival was 0.63 (95% CI 0.52–0.72). Post-docetaxel, 36 patients received second-line treatment, mostly with mitoxantrone (89%). Second-line response rates were 22%, and median duration of response was 4 months. Conclusions: In routine clinical practice, docetaxel is a well-tolerated regimen for the treatment of HRPC. Response rates and toxicity profiles were comparable to the randomized trials. However, compared with the TAX 327 clinical trial, survival was slightly shorter than expected (15.9 vs. 18.9 months), possibly due to inclusion of patients with poorer performance status and comorbidities, who may be excluded from clinical trials. Second-line response rates were also comparable with previous reports. No significant financial relationships to disclose.

Download Full-text