array(41) {
["project_status"]=>
string(7) "ongoing"
["project_assoc_trials"]=>
array(6) {
[0]=>
object(WP_Post)#5139 (24) {
["ID"]=>
int(1268)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2014-10-20 16:17:00"
["post_date_gmt"]=>
string(19) "2014-10-20 16:17:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(298) "NCT01106677 - A Randomized, Double-Blind, Placebo and Active-Controlled, 4-Arm, Parallel Group, Multicenter Study to Evaluate the Efficacy, Safety, and Tolerability of Canagliflozin in the Treatment of Subjects With Type 2 Diabetes Mellitus With Inadequate Glycemic Control on Metformin Monotherapy"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(191) "nct01106677-a-randomized-double-blind-placebo-and-active-controlled-4-arm-parallel-group-multicenter-study-to-evaluate-the-efficacy-safety-and-tolerability-of-canagliflozin-in-the-treatment-o"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-10-24 15:50:03"
["post_modified_gmt"]=>
string(19) "2025-10-24 19:50:03"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(240) "https://dev-yoda.pantheonsite.io/clinical-trial/nct01106677-a-randomized-double-blind-placebo-and-active-controlled-4-arm-parallel-group-multicenter-study-to-evaluate-the-efficacy-safety-and-tolerability-of-canagliflozin-in-the-treatment-o/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[1]=>
object(WP_Post)#5138 (24) {
["ID"]=>
int(1271)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2014-10-20 16:18:00"
["post_date_gmt"]=>
string(19) "2014-10-20 16:18:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(302) "NCT00968812 - A Randomized, Double-Blind, 3-Arm Parallel-Group, 2-Year (104-Week), Multicenter Study to Evaluate the Efficacy, Safety, and Tolerability of JNJ-28431754 Compared With Glimepiride in the Treatment of Subjects With Type 2 Diabetes Mellitus Not Optimally Controlled on Metformin Monotherapy"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(190) "nct00968812-a-randomized-double-blind-3-arm-parallel-group-2-year-104-week-multicenter-study-to-evaluate-the-efficacy-safety-and-tolerability-of-jnj-28431754-compared-with-glimepiride-in-the"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-10-24 15:51:08"
["post_modified_gmt"]=>
string(19) "2025-10-24 19:51:08"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(239) "https://dev-yoda.pantheonsite.io/clinical-trial/nct00968812-a-randomized-double-blind-3-arm-parallel-group-2-year-104-week-multicenter-study-to-evaluate-the-efficacy-safety-and-tolerability-of-jnj-28431754-compared-with-glimepiride-in-the/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[2]=>
object(WP_Post)#5137 (24) {
["ID"]=>
int(1635)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2017-10-26 16:14:00"
["post_date_gmt"]=>
string(19) "2017-10-26 16:14:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(335) "NCT01809327 - A Randomized, Double-Blind, 5-Arm, Parallel-Group, 26-Week, Multicenter Study to Evaluate the Efficacy, Safety, and Tolerability of Canagliflozin in Combination With Metformin as Initial Combination Therapy in the Treatment of Subjects With Type 2 Diabetes Mellitus With Inadequate Glycemic Control With Diet and Exercise"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(191) "nct01809327-a-randomized-double-blind-5-arm-parallel-group-26-week-multicenter-study-to-evaluate-the-efficacy-safety-and-tolerability-of-canagliflozin-in-combination-with-metformin-as-initial"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-10-24 15:56:31"
["post_modified_gmt"]=>
string(19) "2025-10-24 19:56:31"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(240) "https://dev-yoda.pantheonsite.io/clinical-trial/nct01809327-a-randomized-double-blind-5-arm-parallel-group-26-week-multicenter-study-to-evaluate-the-efficacy-safety-and-tolerability-of-canagliflozin-in-combination-with-metformin-as-initial/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[3]=>
object(WP_Post)#5134 (24) {
["ID"]=>
int(1806)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2023-08-05 04:45:19"
["post_date_gmt"]=>
string(19) "2023-08-05 04:45:19"
["post_content"]=>
string(0) ""
["post_title"]=>
string(195) "NCT01032629 - A Randomized, Multicenter, Double-Blind, Parallel, Placebo-Controlled Study of the Effects of JNJ-28431754 on Cardiovascular Outcomes in Adult Subjects With Type 2 Diabetes Mellitus"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(189) "nct01032629-a-randomized-multicenter-double-blind-parallel-placebo-controlled-study-of-the-effects-of-jnj-28431754-on-cardiovascular-outcomes-in-adult-subjects-with-type-2-diabetes-mellitus"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-05-13 14:18:55"
["post_modified_gmt"]=>
string(19) "2025-05-13 18:18:55"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(238) "https://dev-yoda.pantheonsite.io/clinical-trial/nct01032629-a-randomized-multicenter-double-blind-parallel-placebo-controlled-study-of-the-effects-of-jnj-28431754-on-cardiovascular-outcomes-in-adult-subjects-with-type-2-diabetes-mellitus/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[4]=>
object(WP_Post)#5136 (24) {
["ID"]=>
int(1808)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2019-08-12 15:10:00"
["post_date_gmt"]=>
string(19) "2019-08-12 15:10:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(188) "NCT01989754 - A Randomized, Multicenter, Double-Blind, Parallel, Placebo-Controlled Study of the Effects of Canagliflozin on Renal Endpoints in Adult Subjects With Type 2 Diabetes Mellitus"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(182) "nct01989754-a-randomized-multicenter-double-blind-parallel-placebo-controlled-study-of-the-effects-of-canagliflozin-on-renal-endpoints-in-adult-subjects-with-type-2-diabetes-mellitus"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-10-02 10:04:00"
["post_modified_gmt"]=>
string(19) "2025-10-02 14:04:00"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(231) "https://dev-yoda.pantheonsite.io/clinical-trial/nct01989754-a-randomized-multicenter-double-blind-parallel-placebo-controlled-study-of-the-effects-of-canagliflozin-on-renal-endpoints-in-adult-subjects-with-type-2-diabetes-mellitus/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[5]=>
object(WP_Post)#5135 (24) {
["ID"]=>
int(1902)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2023-08-05 04:45:19"
["post_date_gmt"]=>
string(19) "2023-08-05 04:45:19"
["post_content"]=>
string(0) ""
["post_title"]=>
string(229) "NCT02065791 - A Randomized, Double-blind, Event-driven, Placebo-controlled, Multicenter Study of the Effects of Canagliflozin on Renal and Cardiovascular Outcomes in Subjects With Type 2 Diabetes Mellitus and Diabetic Nephropathy"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(194) "nct02065791-a-randomized-double-blind-event-driven-placebo-controlled-multicenter-study-of-the-effects-of-canagliflozin-on-renal-and-cardiovascular-outcomes-in-subjects-with-type-2-diabetes-mell"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-10-28 13:13:55"
["post_modified_gmt"]=>
string(19) "2025-10-28 17:13:55"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(243) "https://dev-yoda.pantheonsite.io/clinical-trial/nct02065791-a-randomized-double-blind-event-driven-placebo-controlled-multicenter-study-of-the-effects-of-canagliflozin-on-renal-and-cardiovascular-outcomes-in-subjects-with-type-2-diabetes-mell/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
}
["project_title"]=>
string(142) "Bayesian machine learning for the identification of benefiting subgroups and treatment effect heterogeneity for canagliflozin in T2DM patients"
["project_narrative_summary"]=>
string(633) "This study will develop statistical methodology for identifying individuals that respond positively to a treatment using state-of-the-art Bayesian machine learning and decision theoretic methods, with a focus on identifying subgroups of individuals that respond either positively or negatively to canagliflozin as a treatment for renal or cardiovascular diseases. The methods developed will be compared to other approaches for assessing treatment effect heterogeneity and subgroup identification using machine learning. Insights from this study may help improve the power and robustness of subgroup identification in clinical trials."
["project_learn_source"]=>
string(10) "web_search"
["principal_investigator"]=>
array(7) {
["first_name"]=>
string(7) "Antonio"
["last_name"]=>
string(6) "Linero"
["degree"]=>
string(15) "PhD, Statistics"
["primary_affiliation"]=>
string(29) "University of Texas at Austin"
["email"]=>
string(32) "antonio.linero@austin.utexas.edu"
["state_or_province"]=>
string(2) "TX"
["country"]=>
string(13) "United States"
}
["project_key_personnel"]=>
array(2) {
[0]=>
array(6) {
["p_pers_f_name"]=>
string(7) "Entejar"
["p_pers_l_name"]=>
string(4) "Alam"
["p_pers_degree"]=>
string(7) "Masters"
["p_pers_pr_affil"]=>
string(29) "University of Texas at Austin"
["p_pers_scop_id"]=>
string(0) ""
["requires_data_access"]=>
string(3) "yes"
}
[1]=>
array(6) {
["p_pers_f_name"]=>
string(8) "Poorbita"
["p_pers_l_name"]=>
string(5) "Kundu"
["p_pers_degree"]=>
string(7) "Masters"
["p_pers_pr_affil"]=>
string(31) "University of California, Davis"
["p_pers_scop_id"]=>
string(0) ""
["requires_data_access"]=>
string(3) "yes"
}
}
["project_ext_grants"]=>
array(2) {
["value"]=>
string(3) "yes"
["label"]=>
string(65) "External grants or funds are being used to support this research."
}
["project_funding_source"]=>
string(21) "NSF Award DMS 2144933"
["project_date_type"]=>
string(18) "full_crs_supp_docs"
["property_scientific_abstract"]=>
string(1541) "Background: Canagliflozin has been shown to reduce the risk of cardiovascular and renal events in patients with T2DM. However, the effect of canagliflozin may vary across patients, and identifying subgroups of patients who benefit most from treatment is of great clinical interest. A common challenge is the need to use data-driven methods both to identify subgroups and to quantify uncertainty about the effect within subgroups, as this leads to the problem of post-selection inference bias.
Objective: To identify subgroups of T2DM patients with expected high expected cardiovascular and renal benefits of canagliflozin using Bayesian machine learning to simultaneously account for subgroup estimation and treatment effect estimation uncertainty.
Study Design: A posthoc observation study of patients in the the CANVAS and CANVAS-R trials.
Participants: Patients enrolled in the CANVAS NCT01989754, NCT02065791, NCT0103262, NCT01809327, NCT00968812, NCT01106677.
Primary and Secondary Outcome Measures: HbA1c change from baseline and the composite endpoint MACE for cardiovascular events.
Statistical Analysis: Bayesian causal forests will be used obtain estimates of the posterior distribution of the individual-level treatment effect, and Bayesian decision theory will be used to post-process these results to obtain subgroups of patients with high expected benefits. Within each subgroup, treatment efficacy is quantified via a utility function.
"
["project_brief_bg"]=>
string(1866) "There is growing interest in developing data-adaptive subgroup estimation techniques in both the statistical and clinical literature. Ideally such techniques should leverage both advances in statistical machine learning and causal inference to provide highly-accurate estimates of individual treatment effects, as well as deal in a principled fashion with the post-selection inference problem that arises when the subgroups themselves are estimated from the same data that we wish to estimate their treatment effects on.
The landmark CANVAS and CANVAS-R clinical trials provided encouraging evidence that a new class of diabetes drugs called sodium-glucose cotransporter 2 (SGLT2) inhibitors can help mitigate these risks. Patients taking canagliflozin experienced significantly fewer major cardiovascular events compared to placebo. The primary composite endpoint measured cardiovascular death, non-fatal heart attack, and non-fatal stroke - and canagliflozin demonstrated a clear reduction in these events. We aim to explore how the benefits of canagliflozin differ with respect to patient phenotypes and clinical variables. While some T2DM patients sharing specific demographic and medical characteristics may benefit highly from the drug, others may benefit mildly or poorly from the same. Also efficacy quantification may be driven by very precise criteria/cutoffs depending on the arena of application. In this project, we will identify patient subgroups using data-driven and objective-tailored statistical approaches, so as to identify reasonably large, intuitively interpretable, and arguably homogeneous benefiting subgroups.
In addition to insights about the effect of canagliflozin on T2DM patients, the statistical methodology developed will be broadly useful in subgroup identification for generic clinical trials.
"
["project_specific_aims"]=>
string(1172) "The objective of this project is to develop and assess the performance of Bayesian nonparametric methods and Bayesian machine learning methods in both (i) accurately estimating the individual-level treatment effect of canagliflozin on health endpoints, (ii) identifying subgroups of patients with high expected benefits, and (iii) adequately addressing post-selection inference problems. We assess a class of carefully constructed Bayesian causal forests as an estimation procedure and virtual twins as a subgroup estimation procedure. Using results from multiple trials, we will assess the relative merits of estimating subgroup effects using post-selection inference techniques versus using data from follow-up trials.
Specifically for the CANVAS and CANVAS-R trials, we aim to identify a highest-benefitting phenogroup of T2DM patients, with proper uncertainty quantification. We will consider:
- the largest effect on the a subgroup level;
- maximizing within-group homogeneity and between-group heterogeneity of treatment effects;
- minimizing the false discovery rate for a desired efficacy level of the treatment.
"
["project_study_design"]=>
array(2) {
["value"]=>
string(7) "meta_an"
["label"]=>
string(52) "Meta-analysis (analysis of multiple trials together)"
}
["project_purposes"]=>
array(4) {
[0]=>
array(2) {
["value"]=>
string(56) "new_research_question_to_examine_treatment_effectiveness"
["label"]=>
string(114) "New research question to examine treatment effectiveness on secondary endpoints and/or within subgroup populations"
}
[1]=>
array(2) {
["value"]=>
string(37) "develop_or_refine_statistical_methods"
["label"]=>
string(37) "Develop or refine statistical methods"
}
[2]=>
array(2) {
["value"]=>
string(34) "research_on_clinical_trial_methods"
["label"]=>
string(34) "Research on clinical trial methods"
}
[3]=>
array(2) {
["value"]=>
string(50) "research_on_clinical_prediction_or_risk_prediction"
["label"]=>
string(50) "Research on clinical prediction or risk prediction"
}
}
["project_research_methods"]=>
string(122) "This study will use all patients included in the original CANVAS and CANVAS-R studies. There are no systematic exclusions."
["project_main_outcome_measure"]=>
string(419) "Primary outcome includes the change in HbA1c from Baseline to the 26th week, 3 point MACE or progression of albuminuria, cardiovascular mortality, non-fatal myocardial infarction, and non-fatal stroke. Secondary endpoints included BMI, lipid profiles, urinary albumin. Lipid profiles consisted of triglyceride (TG), low-density lipoprotein cholesterol, (LDL- C) and high-density lipoprotein cholesterol (HDL-C).
"
["project_main_predictor_indep"]=>
string(274) "In addition to the assigned treatment regime, we will use as predictors the patient characteristics reported in Table 1 of the original CANVAS trial publication (Neal et al., 2017), which includes age, sex, race, smoking status and history of diabetes and vascular diseases."
["project_other_variables_interest"]=>
string(266) "In addition to baseline demographic variables and medical history, we will consider models that make use of biological measures such as blood pressure, cholesterol levels, and body mass index, as reported in the original CANVAS trial publication (Neal et al., 2017)."
["project_stat_analysis_plan"]=>
string(1697) "For each trial (or subset of trials) that we analyze we will do the following:
1. We will divide the individual-level data into training and testing sets.
2. Using the training set in step 1, we will fit a Bayesian causal forest, with appropriately tuned treatment effect priors that encourage treatment effect homogeneity, to estimate the individual level treatments effects, i.e., the CATE (conditional average treatment effect) and obtain uncertainty quantification for the CATE estimates.
3. Using the results from step 2, we will use customized classification and regression tree (CART) software to identify subgroups of patients with high expected benefits. We will use the virtual twins approach to build these subgroups as well as use new decision-theoretic criteria that we will develop.
4. After identifying subgroups, we will produce point estimates and unceratinty quantification for the subgroup treatment effects.
5. Finally, we will benchmark the results of both the individual-level treatment effect estimates and the subgroup treatment effect estimates using the heldout testing set.
The following modeling techniques will be used throughout:
- Bayesian causal forests [Hahn et al., 2020]
- Virtual twins [Forest et al., 2011]
- Bayesian decision theoretic subgroup estimation [Sivaganesan et al., 2017]
- We will compare our methodological developement with alternate techniques from the double machine learning literature [Chernozhukov et al., 2018], which are based on sample splitting and machine learning techinques.
"
["project_software_used"]=>
array(1) {
[0]=>
array(2) {
["value"]=>
string(7) "rstudio"
["label"]=>
string(7) "RStudio"
}
}
["project_timeline"]=>
string(154) "Project start date: July 1, 2024
Analysis completion: November 1, 2024
Manuscript draft completion: December 1, 2024
"
["project_dissemination_plan"]=>
string(174) "Potentially suitable journals include Biostatistics, Statistics in Medicine, Biometrics, Statistical Methods in Medical Research, and Journal of Biopharmaceutical Statistics."
["project_bibliography"]=>
string(1831) "
- Foster, J. C., Taylor, J. M., & Ruberg, S. J. (2011). Subgroup identification from randomized clinical trial data. Statistics in medicine, 30(24), 2867-2880.
- Neal, B., Perkovic, V., Mahaffey, K. W., De Zeeuw, D., Fulcher, G., Erondu, N., … & Matthews, D. R. (2017). “Canagliflozin and cardiovascular and renal events in type 2 diabetes”. New England Journal of Medicine, 377(7), 644-657.
- Hahn, P. R., Murray, J. S., & Carvalho, C. M. (2020). “Bayesian regression tree models for causal inference: Regularization, confounding, and heterogeneous effects (with discussion)”. Bayesian Analysis, 15(3), 965-1056.
- Sivaganesan, S., Müller, P., & Huang, B. (2017). “Subgroup finding via Bayesian additive regression trees”. Statistics in medicine, 36(15), 2391-2403.
- Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C., Newey, W., & Robins, J. (2018). “Double/debiased machine learning for treatment and structural parameters”. The Econometrics Journal, 21(1), C1-C68.
- Oikonomou EK, Suchard MA, McGuire DK, Khera R. “Phenomapping-Derived Tool to Individualize the Effect of Canagliflozin on Cardiovascular Risk in Type 2 Diabetes. Diabetes Care.” 2022 Apr 1;45(4):965-974. doi: 10.2337/dc21-1765. PMID: 35120199; PMCID: PMC9016734.
- Nugent, Ciara, Wentian Guo, Peter Müller, and Yuan Ji. “Bayesian approaches to subgroup analysis and related adaptive clinical trial designs.” JCO precision oncology 3 (2019): 1-9.
- Schnell, Patrick M., Qi Tang, Walter W. Offen, and Bradley P. Carlin. “A Bayesian credible subgroups approach to identifying patient subgroups with positive treatment effects.” Biometrics 72(4) (2016): 1026-1036.
"
["project_suppl_material"]=>
bool(false)
["project_coi"]=>
array(3) {
[0]=>
array(1) {
["file_coi"]=>
array(21) {
["ID"]=>
int(14967)
["id"]=>
int(14967)
["title"]=>
string(40) "SV_57KskaKADT3U9Aq-R_6LAMxSL8h152Bcx.pdf"
["filename"]=>
string(40) "SV_57KskaKADT3U9Aq-R_6LAMxSL8h152Bcx.pdf"
["filesize"]=>
int(20174)
["url"]=>
string(89) "https://yoda.yale.edu/wp-content/uploads/2024/05/SV_57KskaKADT3U9Aq-R_6LAMxSL8h152Bcx.pdf"
["link"]=>
string(86) "https://yoda.yale.edu/data-request/2024-0600/sv_57kskakadt3u9aq-r_6lamxsl8h152bcx-pdf/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "1800"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(40) "sv_57kskakadt3u9aq-r_6lamxsl8h152bcx-pdf"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(14965)
["date"]=>
string(19) "2024-05-29 22:03:00"
["modified"]=>
string(19) "2024-05-29 22:03:05"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
}
[1]=>
array(1) {
["file_coi"]=>
array(21) {
["ID"]=>
int(14968)
["id"]=>
int(14968)
["title"]=>
string(37) "Entejar-Alam_conflict-of-interest.pdf"
["filename"]=>
string(37) "Entejar-Alam_conflict-of-interest.pdf"
["filesize"]=>
int(20563)
["url"]=>
string(86) "https://yoda.yale.edu/wp-content/uploads/2024/05/Entejar-Alam_conflict-of-interest.pdf"
["link"]=>
string(83) "https://yoda.yale.edu/data-request/2024-0600/entejar-alam_conflict-of-interest-pdf/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "1800"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(37) "entejar-alam_conflict-of-interest-pdf"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(14965)
["date"]=>
string(19) "2024-05-29 22:03:02"
["modified"]=>
string(19) "2024-05-29 22:03:05"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
}
[2]=>
array(1) {
["file_coi"]=>
array(21) {
["ID"]=>
int(14969)
["id"]=>
int(14969)
["title"]=>
string(27) "Conflict-of-Interest_PK.pdf"
["filename"]=>
string(27) "Conflict-of-Interest_PK.pdf"
["filesize"]=>
int(20396)
["url"]=>
string(76) "https://yoda.yale.edu/wp-content/uploads/2024/05/Conflict-of-Interest_PK.pdf"
["link"]=>
string(73) "https://yoda.yale.edu/data-request/2024-0600/conflict-of-interest_pk-pdf/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "1800"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(27) "conflict-of-interest_pk-pdf"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(14965)
["date"]=>
string(19) "2024-05-29 22:03:03"
["modified"]=>
string(19) "2024-05-29 22:03:05"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
}
}
["data_use_agreement_training"]=>
bool(true)
["certification"]=>
bool(true)
["search_order"]=>
string(1) "0"
["project_send_email_updates"]=>
bool(false)
["project_publ_available"]=>
bool(true)
["project_year_access"]=>
string(4) "2024"
["project_rep_publ"]=>
bool(false)
["project_assoc_data"]=>
array(0) {
}
["project_due_dil_assessment"]=>
array(21) {
["ID"]=>
int(15901)
["id"]=>
int(15901)
["title"]=>
string(47) "YODA Project Due Diligence Assessment 2024-0600"
["filename"]=>
string(51) "YODA-Project-Due-Diligence-Assessment-2024-0600.pdf"
["filesize"]=>
int(112352)
["url"]=>
string(100) "https://yoda.yale.edu/wp-content/uploads/2024/05/YODA-Project-Due-Diligence-Assessment-2024-0600.pdf"
["link"]=>
string(93) "https://yoda.yale.edu/data-request/2024-0600/yoda-project-due-diligence-assessment-2024-0600/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "1885"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(47) "yoda-project-due-diligence-assessment-2024-0600"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(14965)
["date"]=>
string(19) "2024-11-05 18:22:04"
["modified"]=>
string(19) "2024-11-05 18:22:04"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
["project_title_link"]=>
array(21) {
["ID"]=>
int(16216)
["id"]=>
int(16216)
["title"]=>
string(46) "YODA Project Protocol - 2024-0608 - 2024-12-11"
["filename"]=>
string(46) "YODA-Project-Protocol-2024-0608-2024-12-11.pdf"
["filesize"]=>
int(216967)
["url"]=>
string(95) "https://yoda.yale.edu/wp-content/uploads/2024/05/YODA-Project-Protocol-2024-0608-2024-12-11.pdf"
["link"]=>
string(88) "https://yoda.yale.edu/data-request/2024-0600/yoda-project-protocol-2024-0608-2024-12-11/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "1885"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(42) "yoda-project-protocol-2024-0608-2024-12-11"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(14965)
["date"]=>
string(19) "2024-12-11 18:07:25"
["modified"]=>
string(19) "2024-12-11 18:07:25"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
["project_review_link"]=>
array(21) {
["ID"]=>
int(15903)
["id"]=>
int(15903)
["title"]=>
string(36) "YODA Project Review - 2024-0600_site"
["filename"]=>
string(38) "YODA-Project-Review-2024-0600_site.pdf"
["filesize"]=>
int(1315587)
["url"]=>
string(87) "https://yoda.yale.edu/wp-content/uploads/2024/05/YODA-Project-Review-2024-0600_site.pdf"
["link"]=>
string(80) "https://yoda.yale.edu/data-request/2024-0600/yoda-project-review-2024-0600_site/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "1885"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(34) "yoda-project-review-2024-0600_site"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(14965)
["date"]=>
string(19) "2024-11-05 18:22:48"
["modified"]=>
string(19) "2024-11-05 18:22:48"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
["project_highlight_button"]=>
string(0) ""
["request_data_partner"]=>
string(15) "johnson-johnson"
["human_research_protection_training"]=>
bool(false)
["request_overridden_res"]=>
string(1) "3"
}
data partner
array(1) {
[0]=>
string(15) "johnson-johnson"
}
pi country
array(0) {
}
pi affil
array(0) {
}
products
array(1) {
[0]=>
string(8) "invokana"
}
num of trials
array(1) {
[0]=>
string(1) "6"
}
res
array(1) {
[0]=>
string(1) "3"
}
General Information
How did you learn about the YODA Project?:
Internet Search
Conflict of Interest
Request Clinical Trials
Associated Trial(s):
- NCT01106677 - A Randomized, Double-Blind, Placebo and Active-Controlled, 4-Arm, Parallel Group, Multicenter Study to Evaluate the Efficacy, Safety, and Tolerability of Canagliflozin in the Treatment of Subjects With Type 2 Diabetes Mellitus With Inadequate Glycemic Control on Metformin Monotherapy
- NCT00968812 - A Randomized, Double-Blind, 3-Arm Parallel-Group, 2-Year (104-Week), Multicenter Study to Evaluate the Efficacy, Safety, and Tolerability of JNJ-28431754 Compared With Glimepiride in the Treatment of Subjects With Type 2 Diabetes Mellitus Not Optimally Controlled on Metformin Monotherapy
- NCT01809327 - A Randomized, Double-Blind, 5-Arm, Parallel-Group, 26-Week, Multicenter Study to Evaluate the Efficacy, Safety, and Tolerability of Canagliflozin in Combination With Metformin as Initial Combination Therapy in the Treatment of Subjects With Type 2 Diabetes Mellitus With Inadequate Glycemic Control With Diet and Exercise
- NCT01032629 - A Randomized, Multicenter, Double-Blind, Parallel, Placebo-Controlled Study of the Effects of JNJ-28431754 on Cardiovascular Outcomes in Adult Subjects With Type 2 Diabetes Mellitus
- NCT01989754 - A Randomized, Multicenter, Double-Blind, Parallel, Placebo-Controlled Study of the Effects of Canagliflozin on Renal Endpoints in Adult Subjects With Type 2 Diabetes Mellitus
- NCT02065791 - A Randomized, Double-blind, Event-driven, Placebo-controlled, Multicenter Study of the Effects of Canagliflozin on Renal and Cardiovascular Outcomes in Subjects With Type 2 Diabetes Mellitus and Diabetic Nephropathy
What type of data are you looking for?:
Individual Participant-Level Data, which includes Full CSR and all supporting documentation
Request Clinical Trials
Data Request Status
Status:
Ongoing
Research Proposal
Project Title:
Bayesian machine learning for the identification of benefiting subgroups and treatment effect heterogeneity for canagliflozin in T2DM patients
Scientific Abstract:
Background: Canagliflozin has been shown to reduce the risk of cardiovascular and renal events in patients with T2DM. However, the effect of canagliflozin may vary across patients, and identifying subgroups of patients who benefit most from treatment is of great clinical interest. A common challenge is the need to use data-driven methods both to identify subgroups and to quantify uncertainty about the effect within subgroups, as this leads to the problem of post-selection inference bias.
Objective: To identify subgroups of T2DM patients with expected high expected cardiovascular and renal benefits of canagliflozin using Bayesian machine learning to simultaneously account for subgroup estimation and treatment effect estimation uncertainty.
Study Design: A posthoc observation study of patients in the the CANVAS and CANVAS-R trials.
Participants: Patients enrolled in the CANVAS NCT01989754, NCT02065791, NCT0103262, NCT01809327, NCT00968812, NCT01106677.
Primary and Secondary Outcome Measures: HbA1c change from baseline and the composite endpoint MACE for cardiovascular events.
Statistical Analysis: Bayesian causal forests will be used obtain estimates of the posterior distribution of the individual-level treatment effect, and Bayesian decision theory will be used to post-process these results to obtain subgroups of patients with high expected benefits. Within each subgroup, treatment efficacy is quantified via a utility function.
Brief Project Background and Statement of Project Significance:
There is growing interest in developing data-adaptive subgroup estimation techniques in both the statistical and clinical literature. Ideally such techniques should leverage both advances in statistical machine learning and causal inference to provide highly-accurate estimates of individual treatment effects, as well as deal in a principled fashion with the post-selection inference problem that arises when the subgroups themselves are estimated from the same data that we wish to estimate their treatment effects on.
The landmark CANVAS and CANVAS-R clinical trials provided encouraging evidence that a new class of diabetes drugs called sodium-glucose cotransporter 2 (SGLT2) inhibitors can help mitigate these risks. Patients taking canagliflozin experienced significantly fewer major cardiovascular events compared to placebo. The primary composite endpoint measured cardiovascular death, non-fatal heart attack, and non-fatal stroke - and canagliflozin demonstrated a clear reduction in these events. We aim to explore how the benefits of canagliflozin differ with respect to patient phenotypes and clinical variables. While some T2DM patients sharing specific demographic and medical characteristics may benefit highly from the drug, others may benefit mildly or poorly from the same. Also efficacy quantification may be driven by very precise criteria/cutoffs depending on the arena of application. In this project, we will identify patient subgroups using data-driven and objective-tailored statistical approaches, so as to identify reasonably large, intuitively interpretable, and arguably homogeneous benefiting subgroups.
In addition to insights about the effect of canagliflozin on T2DM patients, the statistical methodology developed will be broadly useful in subgroup identification for generic clinical trials.
Specific Aims of the Project:
The objective of this project is to develop and assess the performance of Bayesian nonparametric methods and Bayesian machine learning methods in both (i) accurately estimating the individual-level treatment effect of canagliflozin on health endpoints, (ii) identifying subgroups of patients with high expected benefits, and (iii) adequately addressing post-selection inference problems. We assess a class of carefully constructed Bayesian causal forests as an estimation procedure and virtual twins as a subgroup estimation procedure. Using results from multiple trials, we will assess the relative merits of estimating subgroup effects using post-selection inference techniques versus using data from follow-up trials.
Specifically for the CANVAS and CANVAS-R trials, we aim to identify a highest-benefitting phenogroup of T2DM patients, with proper uncertainty quantification. We will consider:
- the largest effect on the a subgroup level;
- maximizing within-group homogeneity and between-group heterogeneity of treatment effects;
- minimizing the false discovery rate for a desired efficacy level of the treatment.
Study Design:
Meta-analysis (analysis of multiple trials together)
What is the purpose of the analysis being proposed? Please select all that apply.:
New research question to examine treatment effectiveness on secondary endpoints and/or within subgroup populations
Develop or refine statistical methods
Research on clinical trial methods
Research on clinical prediction or risk prediction
Software Used:
RStudio
Data Source and Inclusion/Exclusion Criteria to be used to define the patient sample for your study:
This study will use all patients included in the original CANVAS and CANVAS-R studies. There are no systematic exclusions.
Primary and Secondary Outcome Measure(s) and how they will be categorized/defined for your study:
Primary outcome includes the change in HbA1c from Baseline to the 26th week, 3 point MACE or progression of albuminuria, cardiovascular mortality, non-fatal myocardial infarction, and non-fatal stroke. Secondary endpoints included BMI, lipid profiles, urinary albumin. Lipid profiles consisted of triglyceride (TG), low-density lipoprotein cholesterol, (LDL- C) and high-density lipoprotein cholesterol (HDL-C).
Main Predictor/Independent Variable and how it will be categorized/defined for your study:
In addition to the assigned treatment regime, we will use as predictors the patient characteristics reported in Table 1 of the original CANVAS trial publication (Neal et al., 2017), which includes age, sex, race, smoking status and history of diabetes and vascular diseases.
Other Variables of Interest that will be used in your analysis and how they will be categorized/defined for your study:
In addition to baseline demographic variables and medical history, we will consider models that make use of biological measures such as blood pressure, cholesterol levels, and body mass index, as reported in the original CANVAS trial publication (Neal et al., 2017).
Statistical Analysis Plan:
For each trial (or subset of trials) that we analyze we will do the following:
1. We will divide the individual-level data into training and testing sets.
2. Using the training set in step 1, we will fit a Bayesian causal forest, with appropriately tuned treatment effect priors that encourage treatment effect homogeneity, to estimate the individual level treatments effects, i.e., the CATE (conditional average treatment effect) and obtain uncertainty quantification for the CATE estimates.
3. Using the results from step 2, we will use customized classification and regression tree (CART) software to identify subgroups of patients with high expected benefits. We will use the virtual twins approach to build these subgroups as well as use new decision-theoretic criteria that we will develop.
4. After identifying subgroups, we will produce point estimates and unceratinty quantification for the subgroup treatment effects.
5. Finally, we will benchmark the results of both the individual-level treatment effect estimates and the subgroup treatment effect estimates using the heldout testing set.
The following modeling techniques will be used throughout:
- Bayesian causal forests [Hahn et al., 2020]
- Virtual twins [Forest et al., 2011]
- Bayesian decision theoretic subgroup estimation [Sivaganesan et al., 2017]
- We will compare our methodological developement with alternate techniques from the double machine learning literature [Chernozhukov et al., 2018], which are based on sample splitting and machine learning techinques.
Narrative Summary:
This study will develop statistical methodology for identifying individuals that respond positively to a treatment using state-of-the-art Bayesian machine learning and decision theoretic methods, with a focus on identifying subgroups of individuals that respond either positively or negatively to canagliflozin as a treatment for renal or cardiovascular diseases. The methods developed will be compared to other approaches for assessing treatment effect heterogeneity and subgroup identification using machine learning. Insights from this study may help improve the power and robustness of subgroup identification in clinical trials.
Project Timeline:
Project start date: July 1, 2024
Analysis completion: November 1, 2024
Manuscript draft completion: December 1, 2024
Dissemination Plan:
Potentially suitable journals include Biostatistics, Statistics in Medicine, Biometrics, Statistical Methods in Medical Research, and Journal of Biopharmaceutical Statistics.
Bibliography:
- Foster, J. C., Taylor, J. M., & Ruberg, S. J. (2011). Subgroup identification from randomized clinical trial data. Statistics in medicine, 30(24), 2867-2880.
- Neal, B., Perkovic, V., Mahaffey, K. W., De Zeeuw, D., Fulcher, G., Erondu, N., … & Matthews, D. R. (2017). “Canagliflozin and cardiovascular and renal events in type 2 diabetes”. New England Journal of Medicine, 377(7), 644-657.
- Hahn, P. R., Murray, J. S., & Carvalho, C. M. (2020). “Bayesian regression tree models for causal inference: Regularization, confounding, and heterogeneous effects (with discussion)”. Bayesian Analysis, 15(3), 965-1056.
- Sivaganesan, S., Müller, P., & Huang, B. (2017). “Subgroup finding via Bayesian additive regression trees”. Statistics in medicine, 36(15), 2391-2403.
- Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C., Newey, W., & Robins, J. (2018). “Double/debiased machine learning for treatment and structural parameters”. The Econometrics Journal, 21(1), C1-C68.
- Oikonomou EK, Suchard MA, McGuire DK, Khera R. “Phenomapping-Derived Tool to Individualize the Effect of Canagliflozin on Cardiovascular Risk in Type 2 Diabetes. Diabetes Care.” 2022 Apr 1;45(4):965-974. doi: 10.2337/dc21-1765. PMID: 35120199; PMCID: PMC9016734.
- Nugent, Ciara, Wentian Guo, Peter Müller, and Yuan Ji. “Bayesian approaches to subgroup analysis and related adaptive clinical trial designs.” JCO precision oncology 3 (2019): 1-9.
- Schnell, Patrick M., Qi Tang, Walter W. Offen, and Bradley P. Carlin. “A Bayesian credible subgroups approach to identifying patient subgroups with positive treatment effects.” Biometrics 72(4) (2016): 1026-1036.