Lead Investigator: Takashi Sozu, Tokyo University of Science
Title of Proposal Research: Matrix Decomposition in Meta-Analysis for Extraction of Adverse Event Pattern and Patient-Level Safety Profile
Vivli Data Request: 5894
Funding Source: None
Potential Conflicts of Interest: None
Summary of the Proposed Research:
The purpose of assessing adverse events (AEs) in clinical studies is to evaluate what AE patterns are likely to occur during treatment. In contrast, it is difficult to specify which of these patterns occurs in each patient. To tackle this challenging issue, we construct a new statistical model including nonnegative matrix factorization (NMF) by incorporating prior knowledge of AE-specific structures such as severity, combination therapy, and patient information. In our model, the large patient-AE occurrence matrix is decomposed into two small rank matrices. A matrix decomposition in which the elements of these matrices are restricted to positive values is called NMF. Because insufficient information is derived from a single clinical study, we have to extend our model to analyze data from multiple clinical studies simultaneously, as in a meta-analysis. To this end, we assume that a treatment-dependent term can be separated from a study-dependent term. The extracted typical treatment-specific AE patterns coincided with background knowledge. We demonstrated the extraction of AE patterns and patient-level safety profiles using the data in the Project Data Sphere (submitted).
Next, using the data sets in Vivli, we would like to incorporate temporal order of AEs, patient information including age, weight and blood test values into our statistical model. We believe that our research will allow us to extract treatment-specific AE patterns and understand what patient information makes a certain pattern more likely to occur, and then will enable the proposal of drugs that are likely to reduce AEs for each patient.
Requested Studies:
A Randomized, Double-Blind, Placebo-Controlled Phase 3 Study of Capecitabine and Cisplatin With or Without Ramucirumab as First-line Therapy in Patients With Metastatic Gastric or Gastroesophageal Junction Adenocarcinoma (RAINFALL)
Sponsor: Eli Lilly & Co.
Study ID: NCT02314117
Sponsor ID: 15372
A Randomized, Double-blind, Multicenter Phase 3 Study of Irinotecan, Folinic Acid, and 5-Fluorouracil (FOLFIRI) Plus Ramucirumab or Placebo in Patients With Metastatic Colorectal Carcinoma Progressive During or Following First-Line Combination Therapy With Bevacizumab, Oxaliplatin, and a Fluoropyrimidine
Sponsor: Eli Lilly & Co.
Study ID: NCT01183780
Sponsor ID: 13856
A Randomized, Multicenter, Double-Blind, Placebo-Controlled Phase 3 Study of Weekly Paclitaxel With or Without Ramucirumab (IMC-1121B) Drug Product in Patients With Metastatic Gastric Adenocarcinoma, Refractory to or Progressive After First-Line Therapy With Platinum and Fluoropyrimidine (RAINBOW)
Sponsor: Eli Lilly & Co.
Study ID: NCT01170663
Sponsor ID: 13894
A Randomized, Double-Blind, Phase 3 Study of Docetaxel and Ramucirumab Versus Docetaxel and Placebo in the Treatment of Stage IV Non-Small Cell Lung Cancer Following Disease Progression After One Prior Platinum-Based Therapy
Sponsor: Eli Lilly & Co.
Study ID: NCT01168973
Sponsor ID: 13852
A Phase 3, Randomized, Double-Blinded Study of IMC-1121B and Best Supportive Care (BSC) Versus Placebo and BSC in the Treatment of Metastatic Gastric or Gastroesophageal Junction Adenocarcinoma Following Disease Progression on First-Line Platinum- or Fluoropyrimidine-Containing Combination Therapy
Sponsor: Eli Lilly & Co.
Study ID: NCT00917384
Sponsor ID: 13893
A Multicenter, Multinational, Randomized, Double-Blind, Phase III Study of IMC-1121B Plus Docetaxel Versus Placebo Plus Docetaxel in Previously Untreated Patients With HER2-Negative, Unresectable, Locally-Recurrent or Metastatic Breast Cancer
Sponsor: Eli Lilly & Co.
Study ID: NCT00703326
Sponsor ID: 13892
The primary purpose of evaluating adverse events (AEs) in clinical studies is to understand the patterns of AEs caused by treatment at the population level. A secondary purpose is to know which AE patterns are likely to occur for each patient to support patient-level treatment strategies. To achieve these objectives simultaneously, an extended model of nonnegative matrix factorization (NMF) that can integrate data from multiple clinical studies has been proposed. An important challenge that has not yet been addressed is the estimation of AE patterns for each drug in combination therapy. In this report, we constructed a new statistical model to achieve this challenge. We illustrated the estimation of the AE patterns by applying our model to the data from two actual Phase III clinical studies of docetaxel and ramucirumab combination therapy.