Development of design and analysis for multistage-randomized trials for selecting optimal treatment regimes

• Project/Area Number: 16K16015
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Statistical science
• Research Institution: The University of Tokyo
• Principal Investigator: Shinozaki Tomohiro 東京大学, 大学院医学系研究科(医学部), 助教 (60644482)
• Project Period (FY): 2016-04-01 – 2019-03-31
• Project Status: Completed (Fiscal Year 2018)
• Budget Amount: ¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000); Fiscal Year 2018: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000); Fiscal Year 2017: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000); Fiscal Year 2016: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
• Keywords: 統計的因果推論 / 臨床試験 / 個別化医療 / 動的治療レジメ / セミパラメトリック推測 / 動的レジメ

Outline of Final Research Achievements

As an approach to "personalized medicine" for delivering better treatments to each patient, this research project focused on "dynamic treatment regimes," which use biomarker information varying with treatment sequence and/or response to previous treatment to adaptively select subsequent treatment strategy. Among the candidate dynamic regimes, we should select the "optimal treatment regime" that maximizes the efficacy on targeted long-term clinical outcomes through validated statistical methods applied to adequately designed clinical studies including multistage-randomized clinical trials. I have made some methodological contribution for tackling challenges in estimating optimal treatment regimes from a statistical causal inference perspective.

Academic Significance and Societal Importance of the Research Achievements

適切な治療法の確立には、基礎研究だけでなく現実の患者を対象とした臨床データとその適切な統計解析が必要である。特に、長期にわたって治療法を計画するためには、最初に計画した治療法を遵守するだけではなく、臨床データの経過・変化に応じて、適応的に以降の治療を変更するような治療方針を取るべきである。このような一連の治療の流れを「治療レジメ」と呼び、中でも長期的なアウトカムを最も改善するものを「最適な治療レジメ」と呼ぶが、現実の医療データからこのような最適レジメを選ぶことは至難の業である。本研究課題では、「統計的因果推論」のアプローチを通して、統計的観点からこの目的を達成するための方法論的研究を行った。

Research Products

• [Journal Article] Misuse of Regression Adjustment for Additional Confounders Following Insufficient Propensity-Score Balancing (2019)
  • Author(s): Tomohiro Shinozaki, Masanori Nojima
  • Journal Title: Epidemiology Volume: In press
  • Peer Reviewed
• [Journal Article] Hazard Ratio Estimators after Terminating Observation within Matched Pairs in Sibling and Propensity Score Matched Designs (2019)
  • Author(s): Tomohiro Shinozaki, Mohammad Ali Mansournia
  • Journal Title: International Journal of Biostatistics Volume: Ahead of print
  • DOI: 10.1515/ijb-2017-0103
  • Peer Reviewed / Open Access / Int'l Joint Research
• [Journal Article] On hazard ratio estimators by proportional hazards models in matched-pair cohort studies (2017)
  • Author(s): Tomohiro Shinozaki, Mohammad A. Mansournia, Yutaka Matsuyama
  • Journal Title: Emerging Themes in Epidemiology Volume: 14 Pages: 6-6
  • DOI: 10.1186/s12982-017-0060-8
  • Peer Reviewed / Open Access / Int'l Joint Research
• [Journal Article] Re: Biases in randomized trials: a conversation between trialists and epidemiologists (Letter) (2017)
  • Author(s): Tomohiro Shinozaki, Yasuhiro Hagiwara, Yutaka Matsuyama
  • Journal Title: Epidemiology Volume: 印刷中
  • DOI: 10.1097/ede.0000000000000663
  • Peer Reviewed / Open Access / Acknowledgement Compliant
• [Presentation] G-estimation of structural nested restricted mean time lost models to estimate causal effect of treatment on survival outcome (2018)
  • Author(s): Yasuhiro Hagiwara, Tomohiro Shinozaki, Yutaka Matsuyama
  • Organizer: The XXIXth International Biometric Conference
  • Int'l Joint Research
• [Presentation] 治療方針の因果効果と臨床試験 (2018)
  • Author(s): 篠崎智大
  • Organizer: 日本行動計量学会 第46回大会
  • Invited
• [Presentation] 構造ネストモデルによる繰り返し治療の総合効果と直接効果の推定 (2018)
  • Author(s): 篠崎智大
  • Organizer: 2018年度統計関連学会連合大会
  • Invited
• [Presentation] IPTW推定量と二重ロバスト推定量 (2018)
  • Author(s): 篠崎智大
  • Organizer: 第24回日本薬剤疫学会学術総会
  • Invited
• [Presentation] 傾向スコアと層別・回帰を組み合わせた交絡調整の誤用について (2018)
  • Author(s): 篠崎 智大，野島 正寛
  • Organizer: 2018年度日本計量生物学会年会
• [Presentation] 時間依存性交絡の調整 (2017)
  • Author(s): 篠崎 智大
  • Organizer: 2017年度日本計量生物学会年会 チュートリアル
  • Place of Presentation: 東京
  • Year and Date: 2017-03-16
  • Invited
• [Presentation] Comparison of several approaches to adjust overall survival (OS) for extensive crossovers in placebo-controlled randomized phase 3 trials (2017)
  • Author(s): Shogo Nomura, Tomohiro Shinozaki, Chikuma Hamada
  • Organizer: The 2017 ENAR Spring Meeting
  • Int'l Joint Research
• [Presentation] 時間依存性交絡の調整 (2017)
  • Author(s): 篠崎 智大
  • Organizer: 2017年度日本計量生物学会年会
  • Invited
• [Presentation] 周辺構造モデルによるパネルデータの解析 (2017)
  • Author(s): 篠崎 智大
  • Organizer: 日本臨床疫学会 第1回年次学術大会
  • Invited
• [Presentation] ペアワイズ層別Coxモデルによるoverall Cの推測 (2016)
  • Author(s): 篠崎 智大，松山 裕
  • Organizer: 2016年度統計関連学会連合大会
  • Place of Presentation: 金沢
  • Year and Date: 2016-09-04
• [Presentation] Estimation of modified effectiveness in randomized trials with subsequent treatments using dynamic marginal structural models (2016)
  • Author(s): Tomohiro Shinozaki, Yasuhiro Hagiwara, Yutaka Matsuyama
  • Organizer: UK Causal Inference Meeting 4th Annual Meeting
  • Place of Presentation: London
  • Year and Date: 2016-04-13
  • Int'l Joint Research
• [Book] Chapter "Data Analysis on Real World Data" in "Epidemiologic Research on Real World Medical Data in Japan" (Editor: Osamu Sudo) (2019)
  • Author(s): Tomohiro Shinozaki, Yutaka Matsuyama
  • Publisher: Springer Japan