PharmaSUG Single Day Event
Tokyo, Japan - SAS Institute offices
October 24, 2019

We had a very successful second annual PharmaSUG Tokyo SDE- thanks once again to SAS for hosting the event, and to all who presented and attended! Please send your feedback and ideas for future events via email to This email address is being protected from spambots. You need JavaScript enabled to view it.. The slides are available for download below. Don't forget that all paid registrants will receive a $75 discount for our annual conference which will be held in San Francisco in May 2020!

Sponsored by:


Presentation (click for abstract)Presenter (click for bio)Slides
CDISC Library UpdateMike Hamidi and Anthony Chow, CDISCSlides (PDF, 2.5MB)
Electronic Data Submission and Utilization in Japan- in Preparation for the End of Transitional Period Hiromi Sugano, PMDASlides (PDF, 548KB)
Efficient Preparation of eData Submission to Both PMDA and FDACJUG ADaM TeamSlides (PDF, 4.5MB)
Central Metadata Repository for Automation in SDTM Dataset GenerationNaoko Izumi, Novartis Pharma K.K. and Ajay Sinha, Novartis Healthcare Pvt. Ltd.Slides (PDF, 393KB)
Real World Evidence using AI/ML on SAS® ViyaToru Tsunoda and Ryosuke Horiuchi, SAS Institute JapanSlides (PDF, 1.5MB)
Let's Join the SAS Global Forum: Build Your Bravery MusclesYutaka Morioka, EPSSlides (PDF, 2.9MB)
Possibility of Process Improvement by Blockchain Technology in Pharmaceutical IndustryKentaro Arai, Novartis Pharma K.K.Slides (PDF, 622KB)
Machine Learning Algorithms / Artificial Intelligence Technology in Clinical DevelopmentSatoki Fujita and Yoshitake Kitanishi, Shionogi, Ltd
Catching the Wave of Disruptive Innovations in Real World EvidenceYousuke Nishida, Chugai Pharmaceutical Co., Ltd.Slides (PDF, 530KB)
Utilization and Obstacles of Real World Data within Japan and Future PossibilitiesMasaki Nakamura, Medical Data Vision Co., Ltd.Slides (PDF, 1.4MB)
Building Real World Evidence on Cloud in PracticeNaoki Mashiko, Amazon Web Services Japan K.K.Slides (PDF, 6.1MB)


Title (click for abstract)Presenter (click for bio)Poster
Processing OMOP Data on Apache Spark Cluster by R ScriptMasafumi Okada, IQVIA Solutions, JapanPoster (PDF, 1.6MB)
Automated aCRF Generation Using PythonMitsuhiro Isozaki, Pfizer R&D JapanPoster (PDF, 1.3MB)
RPA Initiatives in ChugaiHiroshi Kosukegawa, Chugai Pharmaceutical Co.Poster (PDF, 2.1MB)
Empowering RWD Researchers with Simulated DataYuichi Koretaka, Shionogi & Co., Ltd.Poster (PDF, 1.9MB)

Conference Committee (L to R): Rob Howard, Yoshitake Kitanishi, Takashi Misawa, Yuichi Nakajima, Taku Sakaue, Yohei Takanami, Tomoki Asai, Richard Allen

Presentation Abstracts

Electronic Data Submission and Utilization in Japan- in Preparation for the End of Transitional Period
Hiromi Sugano, Principal Reviewer for Biostatistics, Office of New Drug II / Office of Advanced Evaluation with Electronic Data, Pharmaceuticals and Medical Devices Agency (PMDA)

Pharmaceuticals and Medical Devices Agency (PMDA) started to accept electronic clinical study data with New Drug Applications on October 1st, 2016. The study data have been successfully received and the new drug reviewers, mainly biostatistics reviewers and medical reviewers in PMDA, use submitted data for their new drug review. PMDA have issued several guidance and FAQs so far, and since the transitional period will be ended on March 31st, 2020, PMDA is now preparing for the next phase. In this presentation, current status and future perspective of electronic data submission will be shown. Additionally, examples of utilization of submitted data in review process and examples of reviewer-friendly style for submitted data or documents will be presented.

CDISC Library Update
Mike Hamidi, Head of Data Science, CDISC
Anthony Chow, Director of Data Science, CDISC

The CDISC Library update will cover various aspects, such as key functionality (e.g., elasticsearch) and future planned releases (e.g., QRS supplements, difference reports, etc.). CDISC will also highlight our new data standards browser, which includes the same content being driven by the underlying CDISC Library knowledge graph, which was previously only accessible via API. By expanding beyond just the API, CDISC Library can now be accessed by a wider stakeholder group (e.g., data managers, standards curators, etc.). Please join the CDISC data science team and learn more about these exciting updates.

Efficient Preparation of eData Submission to Both PMDA and FDA

Electronic study data submission (eData submission) to Pharmaceuticals and Medical Devices Agency began in 1st October 2016 with a 3.5-year transitional period, and will be mandatory starting in 1st April 2020. On the other hand, eData submission to Food and Drug Administration became effective as of 17th December 2016 for all studies that start after this date for New Drug Application. Although both Health Authorities (HAs) require to submit eData, there are some differences in their requirements. Under the circumstance, industries would like to file NDA to both HAs as simultaneously as possible to maximize value of its product. Thus, it is important for us to know and manage these differences.

Therefore, CDISC Japan User Group ADaM team has been creating a document to summarize differences in the requirements between both HAs and suggestion on streamlined process to achieve simultaneous submission.

In this presentation, major important differences in the requirements and the timelines, tips of streamlined process and the internal team organization to prepare eData submission will be provided.

Central Metadata Repository for Automation in SDTM Dataset Generation
Naoko Izumi, Senior Statistical Programer, Novartis Pharma K.K.
Ajay Sinha, Novartis Healthcare Pvt. Ltd.

In a world of continual improvement in processes need for automated tool has become the next tag line. In context of data submitted to Health Authorities quality and consistency of data is of prime importance. The traditional method of development needs to give way to automation to speed up the process for dataset generation so that for review of primary and secondary endpoints output enough time is subsided.

In a properly managed setup, standards and metadata can be used to drive automation. This can result in start of programming even before the actual data for that study is collected. To achieve this one simple and efficient way is to have centrally managed metadata repository that can accelerate the implementation of standards and facilitate regulatory compliance. Since most of the structure in SDTM is fixed (provided by SDTMIG) it is easier to generate SDTM datasets through metadata-driven approach.

This presentation describes metadata–driven approach that can be followed for generation of SDTM datasets. The basic advantage of this approach is that all the metadata can be managed, validated and governed centrally, while facilitating faster, more consistent dataset generation.

Real World Evidence using AI/ML on SAS® Viya
Toru Tsunoda, SAS Institute Japan
Ryosuke Horiuchi, SAS Institute Japan

The volume and diversity of structured/unstructured patient data obtained from wearable devices, insurance systems and clinical databases, which are growing enormously and are recognized as real world evidence (RWE), offer excellent opportunities for pharmaceutical sponsors to accelerate clinical development. AI/ML techniques can be applied to RWE to get better insight more efficiently. SAS® Viya is an open AI platform that provides users with Choice & Control including the state-of-the-art analytics techniques along with OSS integration in their analytics life cycle. In this presentation, its advanced features and several use cases will be shared.

Let's Join the SAS Global Forum: Build Your Bravery Muscles
Yutaka Morioka, EPS

SAS Global Forum (SASGF) is a premier worldwide event for SAS professionals. I submitted my paper to SASGF2019 held in Dallas and made a presentation on stage with my co-presenter (Jun Hasegawa of EPS Corporation), and also I was selected as a winner of the International Professional Award. This time, I want to share the process leading up to the participation in SASGF and my valuable experiences over there. What I felt deeply in Dallas is that Japanese statistical analysts need to be more actively committed to global networking and show their presence. Besides, they should be more curious and at the same time have a future-oriented vision with room in their mind. After listening to my story, I hope you feel that you can easily do it yourself.

Possibility of Process Improvement by Blockchain Technology in Pharmaceutical Industry
Kentaro Arai, Senior Statistical Programmer, Novartis Pharma K.K.

Blockchain Technology is known as distributed ledger technology and it was developed as the core technology of Bitcoin. Blockchain is a growing list of records called blocks that are linked using cryptography. The characteristic of this technology is decentralized and immutable system. In addition, it is possible to incorporate program called smart contract in Blockchain and it enables automation of pre-defined transaction. In recent years, Blockchain technology is expected to be applied to the processes in various industries. Similarly, in pharma industry, many ideas to make use of Blockchain are being considered.

In this presentation, I will introduce the basic technology overview of Blockchain, the use case of Blockchain in pharmaceutical industry and the overview of blockchain system considered by CJUG-SDTM (CDISC Japan User Group SDTM team) Blockchain sub team.

Machine Learning Algorithms / Artificial Intelligence Technology in Clinical Development
Satoki Fujita, Data Science Group, Shionogi & Co., Ltd.
Yoshitake Kitanishi, Director, Data Science Group, Shionogi & Co., Ltd.

Data scientists are obligated to construct various models and to acquire the latest analysis methods for various kinds of data in order to make use of maximizing “Data Science”. Fashionable AI is also the same. The Artificial Intelligence (AI) that we define is a system with the series of processes of “Recognition”, “Learning” and “Action”, which assists people's activities. There are various types of data used in AI, and because of that, the models or methods used for recognition, learning, and action are different depending on the data format. However, in talking about data science, data governance is very important regardless of the data format. We have made a strategy about data governance using Python and SAS via SAS Viya, and have been maximizing data science based on effective matching such as machine learning and deep learning (CNN, RNN etc.). As one example, we introduce the "AI SAS programmer" system developed by our company which semi-automatically creates SAS programs to analyze clinical data. This system is constructed from machine learning and deep learning by selecting a programming language in Data Driven via SAS Viya, and has led to a 33% reduction in standard work time for our analysis work. We will also introduce recent topics in image analysis. A large amount of image data is required to learn the characteristics of the image, and the collection is a key point. Depending on the subject and the environment in which it is placed, it may be difficult to obtain a sufficient amount of image data. Transfer learning is one way to solve such a problem. We also report on cases where transfer learning has been implemented through SAS Viya, and its usefulness has been confirmed and verified by applying it to actual cases.

Catching the Wave of Disruptive Innovations in Real World Evidence.
Yousuke Nishida, Health Service Relations Group, Real World Data Science Dept., Chugai Pharmaceutical Co., Ltd.

The environment of real world evidence (RWE) is now changing rapidly. In particular, new players have entered the healthcare industry. How should we catch the wave of disruptive RWE innovations as a pharmaceutical company? And how can we adapt to the innovations? I would like to suggest the competency model for data scientists. I will also introduce an outline of revised Good Post-Marketing Study Practice, based on a database study which we conducted as a mock study in the Working Team 3 in Federation of Pharmaceutical Manufacturers’ Associations of JAPAN. This session will also include a discussion of Electronic Patient-Reported Outcome and digital solutions. These solutions also have issues that we have to address. Let’s think together for a bright RWE future!

Utilization and Obstacles of Real World Data Within Japan and Future Possibilities
Masaki Nakamura, Director/EBM Division Business Head, Medical Data Vision Co., Ltd.

Various databases are currently pursued with the advancement of utilization of real word data within Japan. Databases which can be explored in Japan and its characteristics, obstacles in utilizing databases, case studies output based by MDV data will be introduced within this session. Moreover, possibilities of data utilization and introduction of current situation of database study conducted post reform of the Good Post-Marketing Study Practice (GPSP) last year will be introduced.

Building Real World Evidence on Cloud in Practice
Naoki Mashiko, Amazon Web Services Japan K.K.

Globally, healthcare policies and drug payments are undergoing change. The pharmaceutical industry is responding with Real World Evidence (RWE) to capture various data types (e.g. claims, payer, EHR, mobiles/wearables, social, genomics) from clinical through post-market activities to prove drug products are efficacious, to maintain formulary preference, and to maximize reimbursement.

Cloud technologies, such as Amazon Web Services (AWS), meet the platform requirements for RWE, helping pharmaceutical organizations quickly establish a cost effective, scalable platform. In this session, we cover the benefits of cloud, how to build the RWE platform on cloud, and how to make the systems secure and compliant. More specifically, we introduce a best practice to deploy SAS Viya on the AWS cloud. By deploying the SAS platform on AWS, you get SAS analytics, data visualization, and machine learning capabilities in an AWS-validated environment. The deployment is automated by an AWS CloudFormation template and takes about one hour.

Processing OMOP data on Apache Spark Cluster by R Script
Masafumi Okada, Senior Consultant, Real World Evidence Solustions, IQVIA Solutions Japan

To handle Real World Data such as administrative claims database, the first barrier is often its large data size. Usually this kind of dataset will be stored in a relational database system, but it is sometimes troublesome for statistical programmers to handle the database. Reasons for this difficulty could be the challenge of learning complex SQL syntax or dataset-specific non-standardized column names. To resolve this, I propose a way to handle RWD easily for statistical programmers using the OMOP standardized data set and tidyverse-way of manipulating large data via the sparklyr R package.

Automated aCRF Generation Using Python
Mitsuhiro Isozaki, Pfizer R&D Japan

In order to create aCRFs, we have to classify CRF pages and consider what SDTM annotations should be put on them. But this work usually needs to be done manually. We focused instead on standardizing wording for the questions in our CRFs independent of EDC system. Program code in Python was developed to classify CRF pages using machine learning techniques and add annotations according to metadata showing the relationship between the standardized questions and SDTM variables.

RPA Initiatives in Chugai
Hiroshi Kosukegawa, Chugai Pharmaceutical Co.

We will Introduce RPA (Robotics Process Automation) initiatives at Chugai Pharmaceutical Co. First, development by RPA partners will be carried out, and we will make it possible to be developed by ourselves in the end. Therefore, we will explain the rules and systems for that.
  1. RPA policy and purpose in Chugai
  2. Established “RPA Business Selection Guidelines” and “RPA Development and Operation Guidelines
  3. Developed RPA promotion system
  4. Developed 28 business processes (50 robots) by RPA partners
  5. Rules creation and support system development for in-house production
  6. Achievements

Empowering RWD Researchers with Simulated Data
Yuichi Koretaka, Digital Intelligence Department, Shionogi & Co., Ltd.

Observational Medical Dataset Simulator Generation (OSIM) was created by OMOP. The simulated datasets include drug exposure, diagnosis, procedure and other real world information. So, these are useful for novice users to grasp what is RWD. Now, OHDSI has taken over this activity from OMOP and has created an updated simulator called SyntheaTM. In this poster, I will introduce these simulators.

Presenter Biographies

Kentaro Arai

Kentaro Arai is a Senior Statistical programmer in Novartis Pharma K.K. He has created the Blockchain sub-team in CJUG-SDTM (CDISC Japan User Group SDTM team) to discuss utilization of Blockchain. The Blockchain sub-team consists of 7 members as shown below, in alphabetical order:
  • Kentaro Arai (Novartis Pharma K.K.)
  • Yusuke Egami (Aerial Partners Inc.)
  • Ryan Hara (Covance Japan)
  • Hiroyuki Ishii (Takumi Information Technology)
  • Nobuto Matsui (FMD K&L Japan K.K.)
  • Kanji Takayama (Takumi Information Technology)
  • Tsukasa Tazawa (PerkinElmer Japan Co., Ltd.)

Anthony Chow

Anthony Chow serves as Director of Data Science at CDISC. He plays a leadership role in defining and administering processes and policies for governing the metadata that becomes CDISC standards. Anthony’s primary duties include creating, maintaining, curating, and ensuring high-quality metadata in CDISC Library, the single, trusted, authoritative source of CDISC standards metadata. He also co-leads the Controlled Terminology Relationship team to develop detailed metadata about codelist usage and rules. Previously, Anthony held essential IT roles at Allergan and Octagon Research Solutions, focused on delivering data integration and migration systems. He holds a bachelor’s degree in Computer Science from Northeastern University.


CDISC Japan User Group (CJUG) ADaM Team consists of 62 members (as of Sep 2019) from Pharma, CRO, Academia and Regulatory agencies. Their objectives are to discuss issues and provide recommendations on the ADaM standards and whole process of eData submission and to provide materials which support the creation of ADaM datasets and other ADaM-related deliverables.

Satoki Fujita

Satoki Fujita is a data scientist at Shionogi & Co., Ltd. Although a newcomer, he is actively working on the application of artificial intelligence technology.

Mike Hamidi

Mike has a Master’s of Science in Clinical Research Administration from George Washington Univ. and brings over 15 years of experience in areas of data standards consulting, program management, operational excellence, regulatory operations, and data sciences with extensive experience in managing complex global initiatives across many CROs and bio-pharmaceutical companies. Mike manages the day-to-day operations of the data science team, which includes CDISC Library, Data Exchange Standards, Real World Data, and supporting Data Science Tools. He is also part of the CDISC Submission Data Standards (SDS) leadership team and actively engaged in HL7, PhUSE, and other industry initiatives.

Ryosuke Horiuchi

Ryosuke Horiuchi is a Customer Advisory from SAS Institute Japan. After working as a Healthcare Information Technologist at several hospitals in Niigata prefecture, he moved to a pharmaceutical industry and started his career as a Data Analyst who applies AI/ML techniques to clinical trials databases and/or real-world data (RWD). Data analytics is his delight.

Mitsuhiro Isozaki

Mitsuhiro Isozaki is a manager in the statistical programming group at Pfizer R&D Japan. He has 15 years of experience in statistical programming languages such as SAS and R.

Naoko Izumi

Naoko Izumi is a Statistical Programmer for Data Sciences & Scientific Operations (DSSO) in Novartis Pharma K.K. (NPKK). She works on creating analysis reports of clinical trials, and supports electronic data submissions. She also belongs to Japan eCTD society, and participates on the eSub Data subcommittee. One of the missions of the subcommittee is to prepare and submit electronic data more smoothly with good collaboration from each department involved.

Yoshitake Kitanishi

Yoshitake Kitanishi is Head of Data Science with Shionogi Co., Ltd. and has been using SAS/R for 15+ years.

Yuichi Koretaka

Yuichi Koretaka works in the Big Data Strategies group at SHIONOGI & Co., Ltd. He is in charge of RWD analysis and consultation. He also works on digital-related things.

Hiroshi Kosukegawa

Bio coming soon.

Naoki Mashiko

Naoki Mashiko is a Senior Solution Architect for enterprise healthcare and life sciences (HCLS) customers in Amazon Web Services (AWS) Japan, technically helping customers to build new businesses and services on AWS. He also leads the internal technical community for HCLS in Japan

Yutaka Morioka

Yutaka Morioka works for EPS Corporation. He works as a clinical data scientist, actively engaging in disseminating various SAS programming techniques from introductory level to advanced. He has presented many times at seminars and SAS users meetings.

Masaki Nakamura

Masaki Nakamura currently is head of managing the large scale data utilization business of MDV.

Yousuke Nishida

Yousuke Nishida works at Chugai Pharmaceutical Co., Ltd. His mission is to gather information and network in order to explore and implement a strategy of improving the Chugai brand through behavioral economics, digital solutions and other disciplines. He founded the external digital health team “Digital Health Insight” this year.

Masafumi Okada

Masafumi has over 15 years of experience in epidemiology, biostatistics, and medical informatics. He has worked on several clinical study projects that obtained data from EMR data. He also has over 5 year experience with CDISC standards. As a biostatistician, he is also an experienced R programmer.

Ajay Sinha

Ajay Sinha has over 14 years of experience in Life Science industry. He is an Advanced SAS Certified Professional and holds various other professional certifications. He has anchored the setup of large Biostatistics/SAS Programming unit both at India and US locations. He has served in various leadership positions in the past and has helped start-ups to set up and stabilize vital functional units, mid-sized CROs to accelerate productivity with enhanced quality, and IT to deploy new software into the system successfully (including computer validation inspections (IQ/OQ/PQ) in compliance with 21 Code of Federal Regulations (21 CFR, Part 11)). Ajay has worked in various therapeutic areas under the statistical programming umbrella. He is currently working at Novartis as Associate Director (Group Head) in the Statistical Programming Groups, helping streamline various activities in the SP function. He has served as Guest Faculty and Expert Speaker on SAS Programming for various institutions, and has also presented papers at national and international conferences. He is influential, results-oriented and a consummate professional leader with excellent people management skills.

Hiromi Sugano

Hiromi Sugano is a Principal Reviewer for Biostatistics for Pharmaceuticals and Medical Devices Agency (PMDA), Japan. She is in charge of the biostatistics review and consultation in Office of New Drug II. She has mainly reviewed cardiovascular disease related drug so far. Additionally, she works for Office of Advanced Evaluation with Electronic Data, and she is in charge of supporting utilization of submitted and accumulated electronic data in PMDA through offering training and practical assistance for usage of analysis software to reviewers.

Toru Tsunoda

Toru Tsunoda is a Customer Advisor from SAS Institute Japan. After working for several management consulting firms, he joined SAS Institute Japan. He is in charge of the health and life science industries.