bulletin · count of an event where statistics and the statisti-cians made significant impacts in...

Bulletin

ISSN: 2226-2393

www.icsa.orgJanuary, 2018

Volume 30, Issue 1

Invited Articles: Recent Advances in DrugDevelopmentandRegulatoryScience inChinaInvited Articles: The Court Case of US vs.Barr LabsA Conversation with Professor James C. FuHints from Hans: Writing and SupervisingPhD StudentsTerence’s Stuff: Creativity in StatisticsXL-Files: A Nobel Prize in Statistics, Finally…

2017 ICSA Awards

2017 ICSA Award Recipients with ICSA President T. Tony Cai and Executive Director Gang Li

Runze Li, Professor at Pennsylvania

State University, the recipients of the

Distinguished Achievement Award

Mei-Cheng Wang, Professor at Johns-

Hopkins University, one of the three

recipients of the Outstanding Service

Award

Zhezhen Jin, Professor at Columbia

University, one of the three recipients of

the Outstanding Service Award

Xin He, Associate Professor at University

of Maryland, one of the three recipients

of the Outstanding Service Award

ICSABulletin

Volume 30/1, January, 2018ISSN 2226-2393

Editorial StaffEditor-in-Chief

Yi [email protected]

Editorial Assistant

Xiaoyu [email protected]

Executive Committee

PresidentAiyi [email protected]

Past PresidentT. Tony [email protected]

President-ElectHeping [email protected]

Executive DirectorGang [email protected]

TreasurerHongliang [email protected]

From the Editor . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2From the 2018 President, ICSA . . . . . . . . . . . . . . . . . . . 4Results of 2017 ICSA Election . . . . . . . . . . . . . . . . . . . . 5ICSA 2018 Executives and Members of the Committees . . . . . 5Call for Nominations of Candidates for 2019 ICSA Officers . . . 7Report from Q& A’s on Career Choices and Paths: A Panel Dis-

cussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Report from the Workshop on Envisioning an Industrial Career 9Report of the ICSA Annual Banquet at JSM 2017 . . . . . . . . . 10ICSA Financial Report: July 1 Through December 31, 2017 . . . 11Recent Advances in Drug Development and Regulatory Science

in China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13The Court Case of US vs. Barr Labs . . . . . . . . . . . . . . . . . 27A Conversation with Professor James C. Fu . . . . . . . . . . . . 33Writing and Supervising PhD Students . . . . . . . . . . . . . . . 36Creativity in Statistics . . . . . . . . . . . . . . . . . . . . . . . . . 37A Nobel Prize in Statistics, Finally… . . . . . . . . . . . . . . . . 39New Fellows of IMS . . . . . . . . . . . . . . . . . . . . . . . . . . 402017 ICSA Awards . . . . . . . . . . . . . . . . . . . . . . . . . . . 412018 ICSA Applied Statistics Symposium Committees . . . . . . 422018 ICSA Applied Statistics Symposium Short Courses . . . . . 43Upcoming Events . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

Editorial ICSA Bulletin January 2018 Vol.30/1

From the EditorYi Huang

Dear ICSA Members,As the new editor-in-chief of the ICSA Bulletin, I

would like to welcome you to the new 2018 Januaryissue. Ever since I learned that I would succeedXin He as editor, I have been listening and learn-ing the engaging stories about ICSA and our Bul-letin. These reminded me constantly of the remark-able people of ICSA, particularly the former editors,executive committees, and the contributors of ICSABulletins. It is my great honor to serve as the edi-tor in 2018 - 2020. I am determined to do all that Ican to maintain the high standards set by my prede-cessors, and if I may challenge myself, make it evenmore synergistic and interesting to readers.

This issue of the Bulletin features two invitedarticles and four column articles. The first invitedarticle addresses recent exciting advances in drugdevelopment and regulatory science in China, writ-ten by Dr. Jie Chen. China is the second largestpharmaceutical market in the world. Soon afterChina joined the ICH last year, the China Food andDrug Administration (CFDA) initiated a series ofmajor changes to its policies and regulations to meetthe international standard, and to close the gapbetween innovation in science and drug/ medicalproducts’ reviewing process and make it more effi-cient. As Jie noted in his article - China’s biophar-maceutical industry is transforming from“made inChina”to“R&D in China”and now to“Innovatedin China.”Dr. Jie Chen is a prominent leader with23 years of industry experience in biopharmaceuti-cal research, development, regulation and manage-ment, particularly in the United States and China.He has served as the president of the InternationalSociety for Biopharmaceutical Statistics since 2010.Want to find out China’s exciting new changes indrug development and regulation reform? Checkout this featured article in this Bulletin, and a fullversion of this paper is published in the Therapeu-tic Innovation and Regulatory Science (2018).

The second featured article is a real regulatorystory from within US Food and Drug Administra-tion (FDA), shared by an insider, Dr. Yi Tsong, ofthe court case of US vs. Barr Labs. Even 25 years af-terwards, it is still listed as one of the top 4“US vs.Pharmaceuticals”cases nationwide. Drs. Yi Tsongand Thomas Hammerstrom were the lead scientistsrepresenting FDA, collecting, reviewing, summa-rizing, and preparing the materials for district at-

torney. So, this story is a must-read, firsthand ac-count of an event where statistics and the statisti-cians made significant impacts in regulations andlegal cases. With more than thirty years of FDA ex-perience, Dr. Tsong has been one of the top statis-tical leaders in drug safety, new drug clinical trialsas well as chemical and manufacturing control. Dr.Tsong was the 2006 President of ICSA and one ofthe most active leaders supporting ICSA. Want toknow more about the fascinating FDA story wherestatistics met policy? Want to find out more abouthow Dr. Tsong collected the evidence for the“BarrLabs”case, as well as how he used four hit songsto capture his up-and-down feelings as the case de-veloped? You are going to enjoy this fun article.

The third featured article is in the column“Conversation with a Distinguished Statistician”,

written by Dr. Naitee Ting. This article summa-rizes his conversation with Professor James C. Fuat University of Monitoba, Canada, who is an inter-nationally renowned statistician and a co-founderof the journal Statistica Sinica. Dr. Fu has maderemarkable contribution to the establishment andearly development of ICSA, served as the 2nd ICSApresident (1989-1990) and as the program chair/ co-chair for the first three ICSA International Statisti-cal Conferences. In addition to his big contribu-tions to statistical research and professional service,his vision and contribution to statistical educationis outstanding too. He founded the Department ofApplied Mathematics at Donghwa University, Tai-wan. Professor Fu shares his personal stories of theearly years of ICSA, his career path, his thoughts onthe future development of ICSA, and advice to newICSA members.

In the column “Hints from Hans”, Profes-sor Hans Rudolf Künsch offers us his personal in-sights on the relationship between Ph.D. studentsand thesis advisors, their role and expectation, andhis words of wisdom on how to make this re-lationship more synergistic and efficient. In thecolumn “Terence’s Stuff”, Professor Terry Speedpoints out the critical role of creativity and imagi-nation in statistics (the creative process, rather thana cookbook-like recipe or procedure) in determina-tion of a specific study design or data analysis plan,which influences our research to a great extent. Thisis a reprint from the author’s column “Creativityin Statistics”in IMS Bulletin, July 2014. In the col-umn“XL-Files”, Professor Xiao-Li Meng explainsthe rationale behind the launching of the Interna-

2 © www.icsa.org

ICSA Bulletin January 2018 Vol.30/1 Editorial

tional Prize for Statistics, and shares his personalpicks of six Nobel Prize worthy ideas in statistics.This is a reprint from the author’s column“A No-bel Prize in Statistics, Finally…”in IMS Bulletin, De-cember 2016.

Turning to ICSA business, this issue of the Bul-letin includes a message from the 2018 ICSA Pres-ident Aiyi Liu, results of the 2017 ICSA election;the announcement of the 2017 ICSA Awards, the2018 ICSA Executives and Members of the Com-mittees, and our new Fellows of IMS; the call forNominations of Candidates for 2019 ICSA Officers;the announcement of 2018 ICSA Applied Statis-tics Symposium Committees, and the list of 2018ICSA Applied Statistics Symposium Short Courses.This issue also contains 2017 ICSA Financial Re-port; and highlights from the ICSA Annual Ban-quet at JSM 2017, the Workshop on Envisioning anIndustrial Career (Rutgers, the State University ofNew Jersey); and the Q& A’s on Career Choices andPaths: A Panel Discussion (SUNY at Buffalo andRoswell Park Cancer Institute). Eight ICSA spon-sored or co-sponsored meetings and conferencesare announced at the end of this issue. And, don’tmiss the back cover which features the announce-

ment of the 2018 ICSA Applied Statistics Sympo-sium.

Finally, I would like to thank all the authors, con-tributors, ICSA executives and committee membersfor their strong support of and enthusiasm for theICSA Bulletin. In particular, I want to thank Drs.Xin He and Jun Yan (two former editors) for theirclear guidance and great efforts showing me andXiaoyu how to construct this first issue of ICSA Bul-letin. And, many thanks to Dr. Xiaoyu Cai for beingmy LaTeX heroine and the hard work. This issuewould not have come together so quickly withouttheir help and support. I sincerely hope you willenjoy reading this issue. Best wishes for a Prosper-ous and Healthy New Year!

Yi Huang, Ph.D.Editor-in-Chief, ICSA BulletinAssociate ProfessorDepartiment of Mathematics andStatisticsUniversity of Maryland, BaltimoreCounty

© www.icsa.org 3

From the ICSA Executives ICSA Bulletin January 2018 Vol.30/1

From the 2018 President, ICSAAiyi Liu

Dear ICSA members,I hope you all had a won-

derful Chinese New Year!As we entered the year ofdog, I wish everyone of youhealth, happiness and pros-perity throughout the newyear.

I am truly honored to serveas the 2018 ICSA presidentand looking forward to work-

ing with all of you. Established in 1987, ICSA hasbeen steadily growing, now about 1600 member-strong, and become one of the major professionalstatistical organizations. As we can all concur, ourassociation would not have gone this far withoutthe tremendous efforts and hard work from theearly pioneers and numerous volunteers, dedica-tion and services from members of various commit-tees, and constant support and loyalty from all ofus through the years. I myself would also like totake this unique opportunity to thank all past pres-idents, executive directors, members of all commit-tees, board of directors, volunteers helping variousICSA activities, and all who care about our associa-tion. In particular, I would like to express my grat-itude to members of the 2017 ICSA executive com-mittee, Professor Mei-Ling Ting Lee (University ofMaryland), 2016 ICSA president, Professor TonyCai (University of Pennsylvania), 2017 ICSA presi-dent, Dr. Gang Li (Johnson & Johnson), ICSA exec-utive director (2017-2019), and Hongliang Shi, ICSAtreasurer (2016-2018). As 2017 present-elect, I hadthe good fortune to work with and learn from themthrough the entire past year, witnessed their enthu-siasm, unselfish dedications and services to our as-sociation, and also importantly the art of decisionmaking. As I continue my service as ICSA presidentin 2018, I look forward to working with executive di-rector Dr. Gang Li (Johnson & Johnson), past presi-dent Professor Tony Cai (University of Pennsylva-nia), and president-elect Professor Heping Zhang(Yale University), and various committees.

Thanks to the program committees, local orga-nizing committees and volunteers, the IT support-ing teams, active participation from many mem-bers, the 2017 ICSA Applied Statistics Symposiumheld in Chicago, IL, and the 2017 ICSA-China con-ference held in Jilin, China, turned out to be yetanother success. This year, these two conferences

will be held respectively in June in New Brunswick,NJ and in July in Qingdao, China. Program com-mittees and local organizers have spent tremen-dous time and efforts in preparing these two meet-ings and I am confident that these two meetingswill be successful. Another flag-ship conference,the ICSA International Conference, held once everythree years, will take place in December 2019. I en-courage members to attend these three meetings topresent your research work, to enjoy meeting withother members and friends, and more importantly,of course, to show your passion about and supportto our association!

A scientific association with sizable member-ship would not be considered completely success-ful without high-impact prestigious journal(s). At-tributing to the outstanding leadership of the edito-rial board over the years, Statistica Sinica has estab-lished its reputation in both quality and impact, andhas become a leading journal in the field. Anotherjournal, Statistics in Biosciences, a sister journal ofStatistica Sinica, which targets more applied work,is on its way of rising. I encourage members to con-sider submission of their high-quality work to thesetwo journals.

In the past few years, the most discussed pro-fessional topic among statisticians is perhaps thechallenges of data sciences to the statistical profes-sion and community. I have heard mixed reactionsincluding enthusiasm, excitement, optimism, resis-tance, suspicion, and pessimism. Regardless, how-ever, the challenges and opportunities are real. Imyself truly believe that we statisticians can playan important leading role in data sciences, butonly if we are inclusive and embrace the challengesthrough involvement, learning, and understanding.

Finally I would like to emphasize that our as-sociation cannot sustain its steady growth with-out recruitment of new blood. Each year so manystudents are graduating from or entering variousstatistics programs, and they are the future of ICSA.I urge members to reach out, whenever possible, tothese young statisticians, introduce to and encour-age them to join our association. I have no doubtthat ICSA will keep growing and become more suc-cessful.

Aiyi Liu, Ph.D.2018 President, ICSASenior Investigator and Acting ChiefNational Institutes of Health

4 © www.icsa.org

ICSA Bulletin January 2018 Vol.30/1 ICSA Business

Results of 2017 ICSA Election2018 President-Elect

Heping Zhang (Yale University)

2017 Biometrics Section Chair

Lei Liu (Northwestern University School ofMedicine)

Directors of ICSA Board (2018-2020)(alphabetical order)

• Chung-Chou H. (Joyce) Chang (University ofPittsburgh)

• Haitao Chu (University of Minnesota)• Daniel Li (Juno Therapeutics)• Mark Chunming Li (Pfizer Consumer Health-

care)• Niansheng Tang (Yunnan University)

ICSA 2018 Executives and Members ofthe CommitteesEXECUTIVES

President: Aiyi Liu (2018)Past President: T. Tony Cai (2017)President-elect: Heping Zhang (2019)Executive Director: Gang Li (2017-2019)ICSA Treasurer: Hongliang Shi (2016-2018)Office of ICSA: Weiliang Qiu, [email protected],Phone: (912) 478-1277.

BOARD of DIRECTORS

Alan Chiang (2016-2018), Yanyuan Ma (2016-2018),Lu Tian (2016-2018), Bo Yang (2016-2018), HaoHelen Zhang (2016-2018), Jie Chen (2017-2019),Jianhua Hu (2017-2019), Bo Huang (2017-2019),Chiung-Yu Huang (2017-2019), Sijian Wang (2017-2019), Chung-Chou H. (Joyce) Chang (2018-2020),Haitao Chu (2018-2020), Daniel Li (2018-2020),Mark Chunming Li (2018-2020), Niansheng Tang(2018-2020).

Biometrics Section Representative

Lei Liu (2018)

STANDING COMMITTEES

Program CommitteeTony Jianguo Sun (Chair, 2018, [email protected])Guoqing Diao (2018-2020), Kai Yu (2018-2020), Li-uquan Sun (2018-2020), Yi Li (2016-2018), LanjuZhang (2016-2018), Hongmei Jiang (2016-2018),Wenguang Sun (2017-2019, USC), Dan Yang, HongTian (2017-2019); Menglin Liu (2018-2020), XiwuLin (2018-2020), Wenbin Lu (2018-2020).

Term of reference: (1) Recommend conferenceand symposium sites, including candidates for theirChairs. (2) Recommend general policy for all meet-ings, subject to approval by the Board of Directors.(3) Collaborate with symposium/conference com-mittees in organizing sessions. Chair: Attend theboard meetings at the Symposium and JSM to re-port the status.

Finance CommitteeHongliang Shi (Chair, 2016-2018)Linda Yau (2016-2018), Rochelle Fu (2019-2021).

Term of reference: (1) Manage 2 ICSA bank ac-counts (H. Shi, ICSA main account; L. Yau, ICSAJ. P. Hsu Memorial Scholarship Fund account). (2)Oversee the budget, recommend long-term finan-cial planning, and invest the Association’s assets,subject to approval by the Board of Directors. (3)

© www.icsa.org 5

mailto:[email protected]:sunj@missouri\.edumailto:sunj@missouri\.edu

ICSA Business ICSA Bulletin January 2018 Vol.30/1

Manage ICSA PayPal account for online credit cardpayment.

Nomination and Election Committee

Colin O. Wu (Chair, 2018, [email protected])Joanna Shih (2016-2018), Tsai-Hung Fan (2016-2018), Hongzhe Li (2017-2019), Niansheng Tang(2017-2019); Chin-Tsang Chiang (2018-2020); JeenLiu (2018-2020), Xin Tian (2018-2020), Ming Tan(2018-2020).

Term of reference: Nominate the candidates forPresident-elect and members of the Board of Direc-tors. The nomination should be sent by the end ofApril to the EC and board directors for approval.

Publication Committee

Yichuan Zhao (Chair, 2018, [email protected])Yi Huang (Editor of Bulletin), Mei-Cheng Wang(Co-Editor of SIB), Hongzhe Li (Co-Editor of SIB),Hsin-Cheng Huang (Co-Editor of S. Sinica), RueyTsay (Co-Editor of S. Sinica), Zhiliang Ying (Co-Editor of S. Sinica), Jiahua Chen (Editor of ICSAbook series), Din Chen (Co-Editor of ICSA book se-ries).

Term of reference: Oversee the publication pol-icy of the Association and make recommendationsto the Board of Directors.

Membership Committee

Bo Fu (Chair, 2018, [email protected])Jiajuan Liang (2016-2018), Samuel Wu (2016-2018),Pengsheng Ji (2017-2019), Yong Chen (2017-2019),Rui Feng (2017-2019), Mark Li (2018-2020), YapingWang (2018-2020), Chenguang Wang (2018-2020).

Term of reference: Recruit new members andcontact interested potential individuals and organi-zations.

Special Lecture Committee

Gang Li (Chair, 2018, [email protected])Qiman Shao (Chair, CUHK), Yazhen Wang (UW-Madison), Yin Xia (UNC/Fudan), Ming Yuan (2018-2020), Zhezhen Jin (2018-2020).

Term of reference: Nominate the speakers for1) the Peter Hall Lecture at the ICSA InternationalConference every three years, and 2) keynote speak-ers for all ICSA Conferences, including the Interna-tional Conference, Applied Statistics Symposium,

and ICSA China Conferences, and other ad hoc con-ferences. This committee is to maintain the consis-tency of quality of speakers and avoid repetitions.(Formal invitation to the candidate(s) will be sentby the Executive Director on behalf of the Executiveand Special Lecture Committees)

Awards CommitteeMei-Cheng Wang (Chair, 2018, [email protected])Ming Yen Cheng (2016-2018), Tsung-Chi Cheng(2016-2018), Yazhen Wang (2017-2019), ZongmingMa (2017-2019), Yichuan Zhao (2017-2019), Song-Xi Chen (2017-2019), Harrison Zhou (2017-2019),Frank Guanghan Liu (2018-2020), Song Yang (2018-2020).

Term of reference: Accept, evaluate, and rec-ommend nominations for ICSA Pao-Lu Hsu Awardand Distinguished Achievement Award. (Formalnotification to the winner(s) will be sent by the Exec-utive Director on behalf of the Executive and AwardCommittees)

IT CommitteeChengsheng Jiang (Chair, 2018, [email protected])Jessica Jin , Don Sun , Hongtu Zhu.

Term of reference: Maintain ICSA web-site andits functionality.

Archive CommitteeLili Yu (Chair, 2018, [email protected])Smiley Cheng, Shein-Chung Chow, Nancy Lo.

Term of reference: Plan and implement elec-tronic archive for the Association.

Lingzi Lu Award Committee (ASA/ICSA)Jichun Xie (2017-2019 , [email protected]),Hao Chen (2017-2019, UC Davis), Haonan Wang(2016-2018), Victoria Sides (2017-2019), Shelly Hur-witz (2017-2019), Eric Kolaczyk (2014-2017).

ICSA Representative to JSM ProgramCommitteeWenguang Sun (2018, [email protected]), Zhen Chen (2019, [email protected]).

Term of reference: Represent ICSA in the JSMProgram Committee, coordinate ICSA sponsoredand co-sponsored sessions at JSM.

6 © www.icsa.org

mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected].\edumailto:[email protected].\edumailto:[email protected]

ICSA Bulletin January 2018 Vol.30/1 ICSA Announcements

ICSA China LiaisonZhezhen Jin (2010-2020)

Term of reference: Coordinate the ICSA activi-ties in China, such as facilitate the organization ofconferences.

Call for Nominations of Candidates for2019 ICSA OfficersColin O. Wu

The ICSA 2018 Nomination Committee is seekingnominations of candidates for the 2019 officers. Thepositions include ICSA President-Elect, ICSA Bio-metrics Section Chair and ICSA Board of Directors.We need to nominate two candidates from eitheracademia or non-academia to run for the President-Elect position, two candidates to run for the Biomet-rics Section Chair position, and at least 5 candidatesto run for the Board positions. According to theICSA Bylaws, President-Elect should be from ”non-academia”, ”academia” and ”no restriction” on athree-year rotational basis - one year from ”non-academia”, followed by ”academia”, and the thirdyear open. For 2019, there will be no restrictionfor the candidates. For the Board of Directors, wehope the candidates can be balanced with respectto gender, region, professional interests, and area ofemployment (academia, industry/business, or gov-ernment).

Candidates for all positions need to be activeICSA members, and need to have strong interestsin serving the ICSA. The President-Elect serves a

three-year term - 2019 as President-Elect, 2020 asPresident, and 2021 as Past President. The Biomet-rics Section Chair serves a one-year term for 2019and sits in the 2019 Board. The Board of Directorsserve a three-year term from Jan. 2019 to Dec. 2021,and need to participate in two ICSA Board meetingseach year - one in June during the Applied Statis-tics Symposium, and the other in August during theJSM.

If you know of an active ICSA member who isappropriate for any of these positions, please con-tact Dr. Colin O. Wu ([email protected]) byApril 15, 2018.

Colin O. Wu, Ph.D.Chair, ICSA Nomination and Elec-tion Committee (2018)Mathematical StatisticianNational Heart, Lung and Blood In-stituteNational Institutes of Health

© www.icsa.org 7

mailto:[email protected]

ICSA Reports ICSA Bulletin January 2018 Vol.30/1

Report from Q& A’s on Career Choicesand Paths: A Panel DiscussionLili Tian

Dr. Lili Tian, Professor and Associate Chair & Grad-uate Director, Department of Biostatistics, SUNYat Buffalo & Professor of Biostatistics & Oncol-ogy, Department of Biostatistics & Bioinformatics,Roswell Park Cancer Institute, organized the fol-lowing panel, ”Q&A’s on Career Choices and Paths:A Panel Discussion.”

This event was jointly sponsored by the Ameri-can Statistical Association’s (ASA) Buffalo-NiagaraChapter, Committee on Career Development, Com-

mittee on Applied Statisticians, Statistical Partner-ships Among Academe, Industry, and Govern-ment (SPAIG) Committee, and International Chi-nese Statistical Association (ICSA). The Panelistswere, Ziqiang Chen, Yingdong Feng, Jia Hua,Qianya Qi, and Fan Zhang, Tingting Zhuang, KellyH. Zou. The Organizer was Lili Tian, and the Mod-erator was Yu Guan.

This panel session on careers choices and pathswas aimed at fostering the next generation of statis-ticians. As the panel organizer, Prof. Tian and themoderator discussed various topics, including (butnot limited to) the following topics. The panel dis-cussions were interactive, candid, insightful, andwell-received!

1. The differences in research between aca-demics and industries;

2. Preparations before entering the job field;

3. Ways to prepare for interviews to maximizethe employment opportunities;

4. Skill sets for statistical collaborations and con-sultations;

5. Benefits of internships;

6. Negotiation skills upon a job offer;

7. Transitions from academia as a student to anon-academic position;

8. The future directions such as data science, ar-tificial intelligence, and bioinformatics.

Lili TianProfessorDepartment of BiostatisticsUniversity of Buffalo

8 © www.icsa.org

ICSA Bulletin January 2018 Vol.30/1 ICSA Reports

Report from the Workshop onEnvisioning an Industrial Careerby the International Chinese StatisticalAssociation at Rutgers, the State Univer-sity of New Jersey

John E. Kolassa

The Graduate Program in Statistics and Biostatis-tics at Rutgers, the State University of New Jer-sey, is grateful to the International Chinese Sta-tistical Association (ICSA) for presenting a careerpanel to students in our program. The Gradu-ate Program in Statistics and Biostatistics preparesMS and PhD candidates for exciting and reward-ing careers in statistics and related fields. We pre-pare students for careers in academia, industry, andgovernment, and appreciate the contribution of theICSA in our preparation for students entering thesecond of these sectors.

The workshop was held on Busch Campus ofRutgers, the State University of New Jersey, in theseminar room and conference room of the Hill Cen-ter, home of the Department of Statistics and Bio-statistics, and the Graduate Program in Statisticsand Biostatistics, on Friday, 6 October, from 1 pmuntil quite late. The seminar room holds 50 au-dience members, and every seat was filled. Be-cause some students could only attend parts of theworkshop, the total number who participated in theevents was higher.

Gang Li of Janssen Research and Developmentdescribed the history and activities of the ICSA, thesponsor of the program. Kelly H. Zou, of PfizerInc., spoke about interesting aspects real-world dataanalysis, a new frontier in the pharmaceutical andhealth care industry. Yanzhi Hsu, of Eli Lilly andCompany, discussed the role of the statistician inthe pharmaceutical industry. Ching-Ray Yu, ofPfizer, Inc., and a Rutgers alumnus, described hiscareer path from graduate student to pharmaceuti-cal researcher, and discussed advantages and dis-advantages of an industrial career. Zhaoling Meng,of Sanofi, also offered career reflections, includingtips on making oneself attractive during the jobapplication process and during the interview, andhow to begin well at a new employer. Lei Wang, ofthe Lotus Group, offered reflections from the per-spective of a former industry statistician turned re-cruiter, valuable insights into the hiring process,and perspectives on the future of the industry andon the hiring process.

The formal portion of the event ended with apanel discussion, including answers to questionsfrom the audience. The event continued untileveryone was done talking, which lasted quite awhile.

The Rutgers students left the workshop excitedand energized and empowered to take the nextstep in their career development, and the programthanks Drs. Li, Zou, Hsu, Yu, and Meng for self-lessly sharing their time with our students. Weparticularly thank Dr. Wang, who traveled quite along distance to spend the day with us, and we alsothank her daughter who accompanied her.

John E. KolassaProfessorDepartment of Statistics and Bio-statisticsRutgers, The State University ofNew Jersey

© www.icsa.org 9


Report of the ICSA Annual Banquet atJSM 2017Chenguang Wang

It was 8:50 PM. The drivers from Baltimore Dillon’sbus service pulled their buses out from the park-ing lot, drove to the front of the Galaxy Asian Cui-sine restaurant, and waited to pick up their guests.The two buses, as if glowing in the Baltimore sum-mer night with their stylish design, almost blockedthe entire driveway. The drivers chatted away tothemselves and did not seem to be concerned. Theyknew the event in the restaurant started about anhour and half ago. They also knew it was for abunch of statisticians from a Chinese association.How long could a party of statisticians last?

Yet things did not go as they expected, at leastnot this time.

Inside the Galaxy Asian Cuisine restaurant, oneof the few Chinese restaurants in the Baltimore areathat were able to host events for over a hundredguests, the party was just reaching climax with afull house of ICSA statisticians. Not a single one ofthem showed any sign of ennui.

This was the annual ICSA banquet dinner atJSM 2017. Many of the participants were longtime ICSA members and supporters, who have wit-nessed ICSA’s growth over the years to become amore influential organization in almost all the sta-tistical fields. Even more of the participants werenew members whom the ICSA would provide op-portunities to pave and enhance their career path.

The banquet, like many before this and futureones to come, is a social event and important plat-

form for ICSA members, including renowned pro-fessors and uprising young scientists, to meet. Withthe authentic Chinese food, wine, beer, professionalfolk song performance from Dr. Aiyi Liu and thespotless Karaoke from Dr. Xiaoli Meng, the banquetoffered relaxation and entertainment at the conclu-sion of JSM 2017. If the bus drivers were there, theywould soon realize that, although numbers mightbe dull, the people working on numbers are neverboring.

As one of the organizers of the banquet, I wouldlike to take the chance to thank you all for attendingthe banquet dinner and hope you had a good time.I would also like to express a special gratitude toDr. Gang Li for his wonderful leadership and myco-organizer Dr. Weiliang Qi, who is the most idealcollaborator one can ever have. I would also like tosincerely thank our Johns Hopkins volunteer stu-dents Guanghao Qi, Gege Gui, Bingkai Wang andHaoyu Zhang for their wonderful contribution.

So long Baltimore. So long our friends. See youin Vancouver in 2018!

Chenguang Wang, Ph.D.2017 JSM Local Chair, ICSAAssistant ProfessorSidney Kimmel ComprehensiveCancer CenterJohns Hopkins University

10 © www.icsa.org

ICSA Bulletin January 2018 Vol.30/1 ICSA Reports

ICSA Financial Report: July 1 ThroughDecember 31, 2017

© www.icsa.org 11


Hongliang Shi, MSTreasurer (2016-2018), ICSADirector, BiostatisticsBlueprint Medicines

12 © www.icsa.org

ICSA Bulletin January 2018 Vol.30/1 Invited Articles

Recent Advances in Drug Developmentand Regulatory Science in Chinaby Jie Chen and Naiqing Zhao

Abstract: As the second largest pharmaceuticalmarket with a great potential for future growth,China has drawn much attention from the globalpharmaceutical community. With an increasinggovernment investment in biomedical research,the domestic biopharmaceutical (biotechnological)companies in China are turning their attention tothe development of innovative medicines and tar-geting the global market. To introduce innovativeproducts to Chinese patients sooner, to improvethe efficiency of its review and approval processes,and to harmonize its regulatory science with inter-national standards, the China Food and Drug Ad-ministration (CFDA) has initiated a series of majorchanges to its policies and regulations. This paperpresents a snapshot of China’s pharmaceutical mar-ket, and research and development status, and in-troduces technical guidelines pertaining to clinicaltrials and new drug applications. The recent waveof ground-breaking reforms in CFDA’s regulatoryscience is discussed. Examples of clinical trials andnew drug applications are provided throughout thediscussion.Keywords: China Food and Drug Administration;Clinical trials; New drug application; Statisticalguideline; Multi-regional clinical trials.

Introduction

China’s pharmaceutical market snapshot

According to the US International Trade Adminis-tration (ITA, 2017), China is the second largest phar-maceutical market in the world with total sales of$108 billion in 2015 and projected sales of $167 bil-lion by 2020, implying a compound annual growthrate (CAGR) of 9.1%. In comparison, the US phar-maceutical market, as the largest in the world, hadtotal sales of $354 billion in 2015 and forecastedsales of $497 billion by 2020, representing a CAGRof 7.0% (GlobalData, 2016). The driving factorsbehind the fast growth in China’s pharmaceuticalmarket are: (1) rapid economic growth during thepast two decades with an average of double-digitCAGR from 2005 to 2015 and a projected CAGRmaintained at 6.5%–6.8% by 2020 (IMS, 2016); (2)

a steadily growing number (percentage) of elderlypopulation aged 60 years and above that reached209 million (15.2% of total population) in 2015 andforecast to reach 358 million (25.3% ) by 2020, and492 million (36.5%) by 2050 (UN, 2015); (3) in-creased government investment in its healthcareprograms, with $348 billion in government spend-ing in 2015 (accounting for 54% of total healthcareexpenditures) (ITA, 2017); (4) the evolving regu-latory science, particularly the recent in-depth re-forms, to promote drug development and innova-tive medicines; and (5) a high demand for goodquality products to meet massive unmet medicalneeds. Figure 1 illustrates the number of prevalentcases for select diseases in China and their corre-sponding percentages of the global total number ofcases. For example, approximately 402,000 cases ofliver cancer in China in 2015 accounted for approx-imately 54% of the global total (Chen et al., 2016),and the “explosive” increase of diabetes mellitusin China increased the total number of diabetic pa-tients to 114 million or 35% of the global total of di-abetic patients (Wu, 2016).

Figure 1: Numbers of prevalent cases (in parenthe-ses) in China and their percentages of global to-tals for select diseases (Gao and Prasad, 2013; Chenet al., 2016; Wu, 2016).

With these driving factors for potential hugemarket growth, many multinational biopharma-ceutical companies have established their researchand development (R&D) facilities in China, not onlyto use local talent but also to provide easy andearly access to the market. According to the R&D-based Pharmaceutical Association Committee (RD-PAC) of China Association of Enterprises with For-eign Investment (CAEFI), at least 31 R&D centersare operated in China by RDPAC member compa-nies who (1) invested a total of $3 billion in R&Dduring the 11th five years (2006-2010); (2) intro-

© www.icsa.org 13

Invited Articles ICSA Bulletin January 2018 Vol.30/1

duced 543 new chemical entities (NCE) or biologicproducts to China and introduced 67 innovativemedicines to China’s market during the past fiveyears, which accounted for 80% of the total innova-tive medicines approved in China during the sametime period; and (3) currently manufacture and sellmore than 2000 different medicines in China’s mar-ket (http://www.rdpac.org/Index.aspx). Inaddition to government initiatives and investmentto boost biomedical research (Beier and Baeder,2017; Normile, 2017), a growing number of China-based domestic biopharmaceutical and biotechno-logical companies have increased their investmentin R&D (Ni et al., 2017; Bradshaw, 2017) to tar-get the global market (The Economist, 2017). Beierand Baeder (2017) attribute the boom of China’slife science innovation to four essential factors: (1)strong R&D infrastructure; (2) intellectual property(IP) protection; (3) adoption of international stan-dards with respect to trade, IP and drug regula-tions; and (4) cost-containment markets offering op-portunities for satisfactory reimbursement for newmedicines. As a result, China’s biopharmaceuticalindustry is transforming from “made in China” to“R&D in China” and now to “innovated in China.”

Because of the rapidly evolving biopharmaceu-tical ecosystem and increasing government invest-ment in particular, the China Food and Drug Ad-ministration (CFDA) has accelerated the reform ofits regulatory science to make the processes of prod-uct development and regulation more transpar-ent, efficient, and patient focused. Since becom-ing the agency member of the International Coun-cil on Harmonisation (ICH) in June 2017, the CFDAhas initiated several major regulation changes con-cerning clinical trials (CT) and new drug applica-tions (NDA) by encouraging translational research,global clinical trials, and innovative medicines de-velopment with priorities focusing on unmet med-ical needs. This paper briefly reviews the history ofdrug regulations, discusses technical requirementsfor clinical trials and subsequent NDA, and summa-rizes the most recent developments in China’s reg-ulatory science. Examples pertaining to the design,data analysis and interpretation, and regulatory ap-proval or disapproval of clinical trials and market-ing authorization are provided throughout the dis-cussion.

The paper is organized as follows: Section Anoverview of drug regulations briefly reviews theDrug Administration Law and Provision for DrugRegistration and recent draft revisions. An outlineof technical guidelines is also presented in this sec-tion. Guidelines for clinical trials, including sta-

tistical guideline, guideline for multiregional clin-ical trials (MRCT), and other relevant guidelines,are described in Section Guidelines for Clinical Tri-als. General considerations for the design and con-duct of clinical trials are provided in Section Gen-eral Considerations for Clinical Trial Design andConduct. Section Recent Advances in China’s Reg-ulatory Science summarizes the recent reforms inCFDA policies and regulations, and Section Con-cluding remarks provides concluding remarks.

An overview of drug regulationsChina initiated its drug regulation in 1951 whenthe Drug Administration Office was establishedwith approximately 20 staff members within theMinistry of Health. In 1978, the trial version ofDrug Administration Regulation was issued andprovided the basis for the Drug Administration Law(DAL) that was approved in 1985. The DAL pro-vides legal bases for all subsequent policies and reg-ulations, including the Provisions for Drug Reg-istration (PDR) and technical and administrativeguidelines. In 1993, the Center for Drug Evaluation(CDE), a technical division responsible for reviewand approval of clinical trials and market autho-rization of drugs, biologic products, and medicaldevices, was created with approximately 50 staffmembers; this number increased to 120 in 2002when the first version of PDR was issued. In 1998,the China (State) Food and Drug Administration (henceSFDA and later changed to CFDA) was founded,and CDE was transferred from the Ministry ofHealth to the CFDA. Between 2002 and 2010, CDEinitiated a series of efforts to draft and implementmany technical guidelines, most of which werebased on corresponding ICH or US FDA guidelineswith appropriate modifications for China’s specificneeds. In 2015, CDE was expanded to 190 employ-ees and created a position of Chief Scientist respon-sible for exploration of in-depth reform in CFDAregulatory science. In June 2017, the CFDA becamean official agency member of the ICH, a significantmilestone in the pursuit of global harmonization ofdrug development standards; see http://www.cde.org.cn/htmlPage.jsp?htmlPageId=4 fordetails on the history and major milestones of drugregulations.

The DAL, approved by the National People’sCongress, stipulates legal responsibilities for man-ufacturing, distribution, management, packaging,labeling, advertisement, and supervision of drugs.Since its issuance in 1985, the DAL has had twomajor revisions, and its third revision (draft) was

14 © www.icsa.org

http://www.rdpac.org/Index.aspxhttp://www.cde.org.cn/htmlPage.jsp?htmlPageId=4http://www.cde.org.cn/htmlPage.jsp?htmlPageId=4


made public on October 23, 2017, for soliciting com-ments and suggestions (http://www.sfda.gov.cn/WS01/CL0778/178902.html). Under theDAL, the PDR, a binding document governing thedevelopment and regulation of drugs and biologicproducts, provides operational and technical de-tails with respect to clinical trial designs and ap-plications, drug manufacturing and import, newdrug applications and approval, drug registrationand its categories, and drug registration suspen-sion. The PDR also specifies respective responsi-bilities of CFDA and provincial FDA in the processof review, inspection, approval, and post-approvalsurveillance. The second version of PDR speci-fies, in its Annex 2 “Registration Categories andApplication Information Requirements of ChemicalDrugs,” the minimum sample sizes required in dif-ferent phases of clinical development (Zhao et al.,2008). Currently, the PDR is in its third revision; the“minimum sample size requirements” will likely belifted from the new version.

In comparison with the previous (second) ver-sion, the proposed revision of PDR, released onOctober 23, 2017, for public comments (http://www.sfda.gov.cn/WS01/CL0778/178900.html), encompasses the following major changes:

(a) Clinical trial applications (CTA): (1) Focus ison target condition, clinical values, alterna-tive therapies, and trial design; (2) samplesizes are determined by study objectives andrelevant guidelines; (3) reduced sample sizesor a clinical trial waiver can be granted forrare diseases or special health conditions; (4)sponsors can test or entrust a certified thirdparty to test samples of drugs to be used inclinical trials; (5) upon CTA approval, clinicaltrial information regarding start-up, tempo-rary suspension, resumption, early termina-tion, and completion will be published at theCDE’s website; and (6) sponsors are requiredto provide a capability evaluation of clinicaltrial sites and investigators.

(b) Clinical trial conduct: (1) Any major changesin clinical trial design and conduct should beapproved by CDE; (2) serious adverse event(SAE) reporting is re-enforced; (3) annual in-spection of clinical trials will be performed byCDE and relevant provincial FDAs; (4) trialsmay be suspended or terminated if (i) SAEsare not reported to CDE, (ii) serious flaws intrial quality management exist, (iii) risks ex-ceed benefits, and (iv) any GCP violations ex-ist; (5) supplemental applications for changes

in applicants, manufacturing site and/or pro-cess, and specification of drugs during clin-ical trials are acceptable; (6) a new CTA isneeded if a suspended trial is resumed after 18months of suspension; and (7) the CFDA willmake a public announcement within 15 daysfor new drugs entering surveillance and willnot accept any NDA for the same drug with aformulation change or imported drugs of thesame kind.

(c) NDA: (1) Approval or disapproval of NDAwill be based on evidence of efficacy andsafety on Chinese patients; (2) quality as-surance documents should be available; (3)more attention will be paid to (i) clinical val-ues of innovative medicines, (ii) technical in-novation and clinical superiority of drugswith changes in formulation and/or routeof administration, (iii) consistency of genericdrugs, and (iv) biosimilarities of biologicproducts; and (4) nonapprovable NDA in-clude serious concerns in efficacy, safety, andquality of drugs, data quality and fraud, etc(13 categories).

(d) Registration categories: (1) New drugs: drugsnot marketed in any country (innovated do-mestically or overseas), and/or drugs mar-keted in other countries but not in China; and(2) modified new drugs: modified structurewith known active ingredients and clinical su-periority

The revised PDR also clarifies the responsibilitiesof provincial FDAs for review and site inspectionof generic drug applications, acceptance of supple-mental applications caused by changes in manufac-turing processes, and re-registrations of drugs thatpass the 5-year renewal period. In summary, thenew (draft) version of PDR (1) focuses on trans-parency, high quality, and predictability; (2) en-courages biomedical innovation by priority review;(3) establishes a system for risk management, sitemonitoring, and registration inspection; (4) clarifiesresponsibilities of provincial FDAs and expert con-sulting mechanism; and (5) articulates channels forsolving disputes and imposes penalties for submit-ting knowingly false information and test samples.Even though the main content of the PDR is avail-able, its Annexes are not publicly disclosed and itis unclear whether additional requirements will beposted on CTA and NDA.

In addition to the DAL and PDR, approximately180 technical guidelines cover various aspects of de-

© www.icsa.org 15

http://www.sfda.gov.cn/WS01/CL0778/178902.htmlhttp://www.sfda.gov.cn/WS01/CL0778/178902.htmlhttp://www.sfda.gov.cn/WS01/CL0778/178900.htmlhttp://www.sfda.gov.cn/WS01/CL0778/178900.htmlhttp://www.sfda.gov.cn/WS01/CL0778/178900.html


velopment and approval of drugs, biologic prod-ucts, and medical devices. These guidelines can beclassified into the following seven categories:

(a) Chemical drugs: e.g., guideline for pediatricclinical trials, guideline for data extrapolationfrom adult to pediatric populations

(b) Traditional Chinese medicines: e.g., guide-line for formulation modification of naturalherbal medicines, guideline for pharmaco-logical and toxicological research of naturalherbal medicines

(c) Biologic products: e.g., guideline for researchand evaluation of cell products, guideline forbiosimilar products

(d) Cross-subject general guidelines: e.g., biosta-tistical guideline for clinical trials, guidelinefor data management and statistical analy-sis plans and reports, guideline for electronicdata captures in clinical trials

(e) General evaluation: e.g., guideline for non-clinical safety evaluation of therapeutic/pre-ventive biologic products

(f) Technical standards: e.g., chemical drug tech-nical standards, basic requirements for sy-ringes of chemical drugs

(g) Nonclinical guidelines: e.g., guideline fornonclinical studies of potential QT intervalprolongation, guideline for toxicity studies ofrepeated dosing

See http://www.cde.org.cn/zdyz.do?method=initValue&frameStr=0 for a completelist of CFDA’s technical guidelines. However, noguidelines exist for benefit-risk or cost-effectivenessanalysis. With the CFDA joining the ICH as anagency member, it is expected that the regulatoryagency will gradually adopt ICH technical stan-dards in the development and regulation processesof medical products. Recent findings in clinicaltrial data verification have resulted in the CFDA is-suing a technical guideline for sponsors to prepareand submit a data management plan (DMP) and astatistical analysis plan (SAP) in the CTA package,and a data management report (DMR) and a statis-tical analysis report (SAR) in the NDA package; seeSection Guideline for multiregional clinical trials(MRCT) for further discussion.

Guidelines for Clinical TrialsThis section first discusses the statistical guidelinefor clinical trials and a guideline for multiregionalclinical trials. These two guidelines are critical forthe design, analysis, and reporting of clinical tri-als. Other relevant guidelines for the planning, con-duct, and quality assurance of clinical trials are thenbriefly described.

Biostatistical guideline for drug clinicaltrialsThe Guideline on Statistical Principle (GSP) for DrugClinical Trials (statistical guideline), was first draftedin 2003 based on legal requirements stipulated inthe DAL, operational and technical requirementsdescribed in the PDR, and other relevant guide-lines, such as Good Clinical Practice (GCP), ICHE9 (ICH, 1998), US FDA’s guidelines for formatand content of clinical and statistical sections ofnew drug applications (FDA, 1988), and other ICHguidelines concerning clinical trials and new drugapplications. The first version of the statisticalguideline was issued in March 2005 (Chen et al.,2008). After 10 years of implementation, the CFDArealized the limitations of the guideline and the gapbetween the GSP and ICH standards and initiatedthe first revision in 2015. The proposed revision ofthe statistical guideline was published in late 2015for public comments and was finalized in June 2016,with the following major changes (CFDA, 2016a):

(a) A new section on planning clinical investiga-tions was added to the chapter on general con-siderations of clinical trials

(b) Blinded sample size re-estimation and patientenrichment design were added to the chap-ter on clinical trial designs in which group se-quential design and subgroup analysis werealso revised

(c) Clinical trial data inspection, patient enroll-ment monitoring, and independent data andsafety monitoring committee (IDSMC) wereadded to the basic considerations of clinicaltrial conduct

(d) Statistical analysis plan and report were re-moved from the GSP to create a separateguideline

(e) Safety data analysis was enriched to cover tol-erability analysis

16 © www.icsa.org

http://www.cde.org.cn/zdyz.do?method=initValue&frameStr=0http://www.cde.org.cn/zdyz.do?method=initValue&frameStr=0


In summary, the new statistical guideline adoptsthe general principles of the ICH E9 and reflectsthe overall trend of scientific considerations in thedesign, analysis, and conduct of clinical trials. Al-though the guideline provides statistical principlesfor drug clinical trials, it also serves as a key refer-ence for statistical design, analysis, and reporting ofclinical trials for preventive biologic products andmedical devices. Following are three examples oftrial designs for new drug applications.

Example 1–Bridging design. A clinical trial wasdesigned to investigate a new treatment for soft tis-sue sarcoma, a malignant tumor that is rare in Chinawith no effective treatment. An interim analysison disease control rate (DCR) and progression-freesurvival (PFS) time from a global phase III study(without China participating) has shown promis-ing results, which prompted the sponsor to preparea CTA in China. Before finalizing the clinical trialprotocol, the sponsor held a pre-CTA meeting withthe agency, which agreed to a single-arm trial with70 Chinese patients and the overall response rate(ORR) as a primary endpoint and PFS and overallsurvival (OS) as secondary endpoints, plus a phar-macokinetics (PK) study (see Section Other relevantguidelines for clinical trials for more discussion onPK studies) to demonstrate similarity or consistencyof key PK parameters between ethnic Chinese pa-tients and non-Chinese patients. The small scale,one-arm study is neing undertaken because of thepromising results from the global study and the lowincidence rate and lack of effective treatments avail-able for the disease in China.

Example 2–Multiregional clinical trial(MRCT). An MRCT was planned to evaluate thesafety and efficacy of a new drug in patients withpancreatic cancer, a life-threatening disease with-out efficacious therapies. Pancreatic cancer is un-common but not considered a rare disease. Thesponsor proposed a two-arm, double-blind, ran-domized MRCT with PFS as the primary endpointand OS as the secondary endpoint. A total samplesize of 540 patients (including 60 planned Chinesepatients) was determined to achieve 90% powerwith a hazard ratio (HR) of 0.70 for the MRCT. Thissample size considered patient dropouts for anyreasons and censoring by the end of the follow-up,and was expected to result in approximately 390PFS events by the end of the study. The patient en-rollment window was 1.5 years for the 540 patients,with China starting enrollment 3 months before theanticipated last PSF event (because of longer timefor CTA review and approval processes), whichcould jeopardize the plan of enrolling 60 Chinese

patients. Upon discussion with, and agreementby, the agency, the sponsor extended the enroll-ment window, even after 390 PFS events had beenreached, to continue enrolling and following upChinese patients for an additional 8 months. Themain consideration for the enrollment and follow-up extension of Chinese patients was that at least60 Chinese patients were needed to satisfy the pre-defined consistency criteria of PFS results betweenChinese and non-Chinese patients.

Example 3–China-led regional study. A newdrug was approved overseas for the treatment ofgastric cancer, a prevalent malignant tumor typein China and northeast Asia. In addition to a PKstudy, the sponsor planned a two-arm, double-blind, active control, phase III clinical trial in Asiawith PFS as the primary endpoint and OS as thesecondary endpoint. The study was designed witha total of 650 patients, including 520 Chinese pa-tients (∼80% of total) and 130 non-Chinese patientsfrom two neighboring countries, to achieve an over-all 90% power to detect a treatment effect with anHR of 0.75 in all patients.

Guideline for multiregional clinical trials(MRCT)Joining global MRCTs would provide numerousbenefits to China, including bringing innovativemedicines to Chinese patients sooner, applyinginternational standards in product development(design, conduct, analysis, review, approval, andpost-market surveillance), avoiding duplicate trials,and studying heterogeneity among different ethnicpopulations. The CFDA’s MRCT guideline, issuedin June 2015, specifies that an MRCT can be a globalMRCT with at least two countries/regions includ-ing China and another country or region outsideof Asia, or a regional MRCT in which participat-ing countries are in the Asian region. For China toparticipate in an MRCT, the CDE requires that theagency must have access to all MRCT data and clin-ical trial sites, and that at least one Chinese expertjoins the IDSMC if Chinese patients exceed 20% ofthe total sample size (CFDA, 2015).

In addition to demonstrating the similarity ofPK parameters between Chinese subjects and non-Chinese subjects, an MRCT is also concerned withthe heterogeneity of intrinsic and extrinsic factorsthat may impact the consistency eveluation of pri-mary (and key secondary) endpoints and the over-all sample size and its allocation to individual coun-tries. The guideline categorizes these factors as(1) absorption, distribution, metabolism, excretion

© www.icsa.org 17


(ADME); (2) disease etiology and prevalence; (3)prognostic and risk factors; (4) medical practice (di-agnosis, treatment, standard of care, etc.); (5) cul-tural differences; and (6) other ethnic differences.See Section General Considerations for ClinicalTrial Design and Conduct for further discussion ofthese factors in designing an MRCT. The guidelinealso discusses the possibility of different dosages,availability of active controls and different efficacyendpoints for Chinese patients.

The MRCT guideline provides only a general de-scription regarding the overall sample size for anMRCT and its allocation to individual countries,stating that the total sample size should satisfy thestudy objective and that the sample size allocationratio should meet the requirements of subgroupanalysis with considerations of disease epidemiol-ogy and sample representation of individual coun-tries/regions. Per the ICH E17 (ICH, 2017), theoverall sample size should be determined by test-ing the primary hypothesis based on data from allcountries/regions, and the sample size allocationto individual regions should be decided “such thatclinically meaningful differences in treatment ef-fects among regions can be described without sub-stantially increasing the sample size on the primaryhypothesis.” While pre-CTA consultation meetingswith CDE are strongly recommended to discusstrial design including subpopulation sample sizes,sponsors should consider the following when spec-ifying sample size allocation ratio: (1) minimumsample size requirement by regulation (if any); (2)proportionality to disease incidence or prevalence;(3) expected differences in treatment effect; (4) alter-native treatment options; (5) statistical methods forconsistency evaluation of treatment effect; and (6)safety concerns. In general, the following three sta-tistical methods can be used to determine the min-imum sample size of Chinese patients for the eval-uation of consistency between Chinese populationand overall population.

Directional treatment effects (DTE) approach.The DTE approach is used to determine the mini-mum sample size for Chinese patients whose treat-ment effect is expected to be directionally similar tothat of the overall population as estimated from theMRCT. Let ΔCN, ΔNCN and Δ denote, respectively,the standardized treatment effects for Chinese popu-lation, non-Chinese population and overall popula-tion. Then, conditional on a statistically significantresult of the MRCT, the DTE approach can be ex-pressed as

𝑃{ΔCN > 0 | Δ ΔCN, ΔNCN > 𝑍1−𝛼/2} > 1 − 𝛽CN,(1)

where ΔCN, ΔNCN and Δ are the estimates ofΔCN, ΔNCN and Δ, respectively, 𝑍𝑝 is the 𝑝% per-centile of the standard normal distribution, 𝛼 is thesignificance level for the MRCT, and 1 − 𝛽CN is theprobability of observing a DTE among Chinese pa-tients. This DTE approach, often called Methods 2as it is the second method proposed in the guidingdocument of the Pharmaceutical and Medical De-vice Association (PMDA) of Japan (MHLW, 2012),is generally suitable to clinical trials for uncom-mon to moderately prevalent diseases with similardisease etiology across countries/regions, such asmelanoma and multiple myeloma.

Retentive treatment effect (RTE) approach.The RTE approach looks for a minimum samplesize that can demonstrate that the treatment effectamong Chinese patients retains at least a certainproportion, say 𝜋 (0 < 𝜋 ≤ 1), of the overall treat-ment effect of the MRCT. Given a significant resultof the MRCT, the sample size for Chinese patientsto show RTE can be expressed as

𝑃{ΔCN > 𝜋Δ ΔCN, ΔNCN | Δ ΔCN, ΔNCN> 𝑍1−𝛼/2} > 1 − 𝛽CN. (2)

The RTE approach, often called Method 1 asit appeared first in the PMDA guideline (MHLW,2007), is usually appropriate for highly prevalentdiseases with similar or slightly different etiol-ogy such as chronic obstructive pulmonary disease(COPD) and breast cancer.

Equivalent treatment effect (ETE) approach.The ETE approach is used to establish equivalencein treatment effect between Chinese patients andthe overall patient population represented in theMRCT. Equivalence can be concluded if the differ-ence (or ratio) of treatment effects between the sub-population and overall population falls within therange of pre-defined upper and lower margins. Thesample size for demonstrating ETE is conditional ona significant result of the MRCT and can be writtenas

𝑃{𝛾𝐿 <ΔCN

𝜋Δ ΔCN, ΔNCN< 𝛾𝑈 | Δ ΔCN, ΔNCN

> 𝑍1−𝛼/2} > 1 − 𝛽CN, (3)

18 © www.icsa.org


where 𝛾𝐿 and 𝛾𝑈 are pre-specified lower and up-per margins for the ratio of estimated treatment ef-fects between Chinese patients and all patients (theformula for a difference in treatment effects can bederived similarly). This approach is suitable forclinical trials demonstrating similarity or equiva-lence in treatment effects (either too high or too lowratio is unacceptable).

If a significant effect in Chinese patients is re-quired for the new drug registration, a commonpractice is to conduct an MRCT with an extensionenrollment of additional Chinese patients (and/orextended follow-up time for time-to-event end-points), or an MRCT plus a China-only trial. In ei-ther case, additional Chinese patients are accrued(and/or followed up) and included, together withthose in the MRCT, in the final analysis to ensurea sufficient sample size (or events) of Chinese pa-tients to demonstrate a significant treatment effect.A large sample size of Chinese patients is requiredto show a statistically significant treatment effectin Chinese patients, and is generally required forhighly prevalent diseases such as diabetic mellitus,liver cancer and esophageal cancer.

Figure 2: Approximate distributions of hazard ra-tios (HRs) for global MRCT and Chinese patientsand probability of observing Chinese HR> 1 (globalHR=0.78, Chinese sample size=800).

The MRCT guideline prescribes that, in additionto the overall analysis of the MRCT data, the sta-tistical analysis report (SAR) should focus on theconsistency among Chinese patients, non-ChineseAsian patients and all patients in the MRCT in termsof primary and key secondary efficacy endpointsas well as major safety endpoints. Specifically, theSAR should include (1) analyses of all patients andthe subgroup of Chinese patients in terms of effi-cacy and safety; (2) consistency evaluation betweenChinese patients and all patients in terms of pri-

mary and key secondary endpoints; and (3) furtherexplorations among participating subpopulations ifinconsistency is observed. The example below dis-cusses inconsistency in the primary efficacy end-point between Chinese patients and all patients inan MRCT.

Example 4–Consistency in primary endpoint.A two-arm MRCT was designed to investigate thetreatment effect of a new drug over an active controlwith a total of 18,000 patients (including 800 Chi-nese patients) and 1:1 randomization to detect anHR of 0.8 at a one-sided significant level 𝛼 = 0.025with 90% power. At the end of the study, the overallHR of the MRCT was estimated as 0.78 while the HRfor Chinese patients was estimated at 1.07 (whichmight have been expected because of small sam-ple size). These findings put the regulatory agencyin a difficult position to either approve (based onthe MRCT results) or disapprove (no evidence oftreatment effect among Chinese patients). In thiscase, the sponsor explored ethnic heterogeneity interms of the primary endpoint that included: (1) ev-idence of heterogeneity between Chinese patientsand non-Chinese patients from pharmacokineticsand clinical and epidemiological perspectives; (2)treatment interactions (using 0.20 as a cut-off forthe 𝑝-value); (3) confounding factors (not stratifiedduring randomization) that could be used to obtainan adjusted HR; (4) propensity score with multipleconfounding factors to obtain adjusted treatmenteffects; (5) forest plot of HRs among participatingcountries or regions; and (6) stochastic simulation(e.g., bootstrap) to estimate the probability of ob-serving an HR = 1.07 given a sample size 800 andassumed HR = 0.8. For example, Figure 2 showsthat given the global HR of 0.78 and Chinese sam-ple size of 800 patients, the probability of observingan HR of 1.07 or greater in Chinese patients is ap-proximately 0.2339, suggesting that an HR > 1.07 inChinese patients could be due to small sample size.

Other relevant guidelines for clinical tri-alsIn July 2016, the CFDA issued a technical guide-line for the data management plan (DMP), statis-tical analysis plan (SAP), data management report(DMR), and statistical analysis report (SAR) for clin-ical trials. CFDA requires the DMP and SAP arerequired to be submitted in the CTA package andthe DMR and SAR to be submitted in the final NDApackages. These requirements, particularly the in-clusion of DMP and SAP in the CTA package, seemto have been prompted by the findings from a re-

© www.icsa.org 19


cent initiative on drug clinical trial data verificationconducted by CDE starting in July 2015. The verifi-cation initially targeted 2033 applications for regis-trations of new drugs, generic drugs and importeddrugs. After completing the first-round self verifi-cation requested by the agency, 1316 (64.7%) appli-cations were withdrawn voluntarily by applicants.Of the 313 (15.8%) applications for which clinicaltrial data were verified by the CFDA, 38 were sus-pected of data fraud (16 for NDA, 17 for genericdrugs, and 5 for imported drugs). The agency re-jected 30 of the 38 applications and started furtherinvestigation against 11 clinical trial organizationsor contract research organizations (CROs) for possi-ble data fraud (CFDA, 2017b). The inclusion of theDMP and SAP in the CTA packages is intended toassure the CFDA that the sponsor has the plan andcapability for high-quality data management andstatistical analysis of clinical trials. The contentsof the DMP and SAP are based primarily on ICHE3, E6 and E9 (http://www.ich.org/products/guidelines/efficacy/article/efficacy-guidelines.html). Specifically, the DMPshould describe the sponsor’s detailed plan on (1)data flow and data management flow; (2) data col-lection and management system; (3) data manage-ment steps consisting of the case report form (CRF)and database design, data entry, data audit and en-quiry, medical coding, external data management,blinded auditing, database lock and unlock, dataoutput, and data management documentation. Inaddition, the sponsor should provide in the finalNDA package the DMR and SAR, which usuallyhave the same structures as the DMP and SAP andpresent in the respective documents the results, de-viations, and any other observations during datamanagement and statistical analysis; see CFDA(2016b) for details.

In many NDA applications, a PK study of Chi-nese subjects (or patients) is required to investi-gate the ADME of a drug and demonstrate equiv-alence of PK parameters between Chinese and non-Chinese populations. A commonly used design(per the CFDA’s guideline for PK studies) is AABBand BBAA design (especially for drugs with highvariability in ADME). The PK parameters shouldinclude at least the maximum concentration (𝐶max),time to 𝐶max (𝑇max) and area under the curve forthe first 24 hours (AUC1−24) or from time 0 to apre-defined time 𝑡 (AUC0−𝑡) (The choice of AUC1−24,AUC1−𝑡 or both depends on the half-life of thedrug). Bioequivalence can be concluded if the 90%confidence intervals on the ratios of geometric (orarithmetic) means for all estimated PK parameters

between Chinese and non-Chinese subjects fall be-tween 0.8 and 1.25.

In March 2016, the agency issued a guideline forpediatric clinical trials stating that data from adultclinical trials should be fully utilized, and that thealgorithm of pediatric drug development shouldfollow the three disease categories: (1) For pediatricspecific diseases, safety and PK data should first beobtained from adult subjects to guide pediatric clin-ical trials; (2) for diseases shared by both adult andpediatric populations with no efficacious therapiesavailable, a phase II proof of concept (POC) studyon adult patients should be conducted first to guidepediatric clinical trials; and (3) for diseases sharedby both adult and pediatric populations with effec-tive treatments available, data in adult clinical trialscan be used to guide pediatric clinical development.A complete list of CFDA guidelines for drug clini-cal trials is available at http://www.cde.org.cn/zdyz.do?method=initValue&frameStr=0.

A wide array of guidelines for the research anddevelopment of therapeutic and preventive biologicproducts, including cell products, preventive DNAproducts, HIV vaccines and combination preven-tive biologic products, are available. These guide-lines cover pre-clinical safety evaluation, develop-ment and validation of biologic assays, and efficacyand safety studies in clinical development. Detailsare available at http://www.cde.org.cn/zdyz.do?method=initValue&frameStr=0#top. InFebruary 2015, the CFDA issued its first guidelinefor the development and evaluation of biosimilarproducts in terms of efficacy, safety, and productquality. The biosimilar guideline specifies stepwiseapproaches to the comparative structural and func-tional characterization (that is, structural similar-ity using the state-of-the-art analytical methodolo-gies such as fingerprint-like analysis algorithm andfunctional similarity in terms of toxicity, immuno-genicity, and PK/PD) and totality of evidence ap-proach, with the latter focusing on clinical similar-ity even with minor structural and/or functionaldifferences.

General Considerations for Clini-cal Trial Design and ConductPlanning and conduct of clinical trials in Chinashould consider not only specific regulatory re-quirements, but also unmet medical needs. Ac-cording to the China Statistical Yearbook (Sheng,Laiyun, 2016), the top 10 leading causes of deathsin China are (1) malignant neoplasm; (2) heart dis-

20 © www.icsa.org

http://www.ich.org/products/guidelines/efficacy/article/efficacy-guidelines.htmlhttp://www.ich.org/products/guidelines/efficacy/article/efficacy-guidelines.htmlhttp://www.ich.org/products/guidelines/efficacy/article/efficacy-guidelines.htmlhttp://www.cde.org.cn/zdyz.do?method=initValue&frameStr=0http://www.cde.org.cn/zdyz.do?method=initValue&frameStr=0http://www.cde.org.cn/zdyz.do?method=initValue&frameStr=0#tophttp://www.cde.org.cn/zdyz.do?method=initValue&frameStr=0#top


eases; (3) cerebrovascular diseases; (4) chronic res-piratory diseases; (5) endocrine, nutritional andmetabolic diseases; (6) neurological disorders; (7)digestive system disease; (8) transport injuries; (9)self-harm and violence injuries; and (10) infectiousdiseases. Cancer patients with lung, hepatocellu-lar, esophageal, gastric, and pancreatic carcinomas,which are among the most malignant canceroustypes, have the greatest unmet medical needs be-cause of high prevalence rates or limited treatmentoptions. The government prioritizes medical re-search and drug development activities for diseasesthat are aging-related or of public heath impor-tance, such as COPD, HIV/AIDS control and injuryprevention. Clinical trials for innovative medicinestargeting diseases with unmet medical needs arelikely to receive priority review; see Section RecentAdvances in China’s Regulatory Science for addi-tional discussion.

Different designs of clinical trials may result indifferent strengths of evidence for therapeutic ef-fects (both efficacy and safety) of a new drug in Chi-nese patients. Statistically powered China-led re-gional or China only studies provide the strongestevidence for treatment effects in the Chinese popu-lation with less heterogeneity concerns and can beused to investigate a new drug for China-specificprevalent diseases with a large patient base. Thisfacilitates patient enrollment, which shortens studyduration. Conversely, an MRCT offers opportuni-ties to evaluate treatment effects in a broader popu-lation and hence, heterogeneity of treatment effects(if any) among subpopulations, providing earlieraccess to more effective medicines globally. As dis-cussed in Section Guideline for multiregional clin-ical trials (MRCT), certain important intrinsic andextrinsic factors must be considered in the designand analysis of an MRCT for cross-ethnicity consis-tency evaluation. Specifically, sponsors should con-sider the following factors:

(a) Patient composition of subpopulations: Thecomposition of the Chinese subpopulationcould differ from that of other subpopulationswith respect to disease stage, gender, age, ed-ucation, and economic status. For example,Chinese patients are more likely to be in thelate stage of head and neck cancer when en-rolled.

(b) Medical practice and concurrent conditions:(1) Traditional Chinese medicines may beused concurrently in Chinese patients; and(2) disease profiles and concurrent conditionsmay differ between Chinese and non-Chinese

patients if these factors are not specified in theinclusion and exclusion criteria.

(c) Health care policy: (1) Medical insurance pro-gram coverage may impact the lines of treat-ments; and (2) treatment options may dif-fer from country to country, and drugs ap-proved in other countries may not be availablein China.

(d) Genomic factors: The genomic profiles of theChinese population may differ from those ofthe non-Chinese population. For example,EGFR mutation is present in approximately40% of the Chinese population versus only15% in the European population (Zhang et al.,2016).

(e) Differences in lifestyle, diet, or environmen-tal factors: (1) According to WHO, about 60%of lifestyle-related factors contribute to indi-vidual health and quality of life (WHO, 2004);(2) factors such as diet and body mass index(BMI), exercise, sleep, sexual behaviors, sub-stance use, medication abuse, etc., may im-pact or modify treatment effect; and (3) air,soil and water pollution may play a role in dis-ease etiology and lead to differentiated treat-ment effects.

(f) Cultural differences: The following culture-related factors may play important roles in theevaluation of treatment effects: (1) culture-bound syndrome when interacting with tra-ditional medicines; (2) patient reported out-comes; (3) cultural belief and mental illness;(4) interaction with healthcare providers; and(5) toxicity tolerability or perception, (6) ad-herence to prescribed treatment regimens.

Sponsors must also consider the following in theconduct of an MRCT:

(a) Imbalanced patient assignment betweentreatment and control: Extremely imbalancedpatient assignment because of a small propor-tion of allocation or centralized randomiza-tion may cause unreliable estimates of treat-ment effects.

(b) Difference in investigator’s perception/diag-nosis: A patient may be diagnosed by theinvestigator as “progression” with treatmentswitch when the patient has in fact no pro-gressive disease.

© www.icsa.org 21


(c) Interim analysis: Can a drug be approved inChina based on a positive global interim re-sult? This will depend on the disease area,safety profile of the drug, interim results andavailable treatment options.

(d) Differentiated drop-off rates: In Chinese pa-tients, a high tolerability to toxicity couldcause a low drop-out rate, and lack of efficacycould result in a high drop-out rate.

(e) Compliance with study protocol: Patientsmay take extra pills when they skip or missthe previous dose.

In general, factors that could modify treatment ef-fect but potentially be treated differently amongsubpopulations during the course of clinical trialsshould be aligned at the protocol level to ensureconsistency of data quality and balanced patientcharacteristics as prescribed at the design stage.Two examples are provided below to illustrate somedesign considerations of clinical trials for new drugapplications.

Example 5–One-arm trial. Ranibizumab injec-tion, a monoclonal antibody fragment, was firstapproved by the US FDA in 2006 and then byEuropean Medicines Agency (EMA) in 2010 fortreatment of age-related wet macular degeneration.These approvals, together with the rare disease sta-tus of age-related wet macular degeneration and thelack of effective treatment, led the agency in 2012 toapprove market authorization of Lucentis in Chinabased on a one-arm study with 160 Chinese pa-tients.

Example 6–Inappropriate selection of doses forreference drug. A clinical trial was designed toshow non-inferiority of a new drug over an activecontrol drug that was approved and marketed withtwo doses (high and low) in China. The treatmentguideline for the target disease recommends thehigh dose for treatment use. The sponsor, how-ever, chose the low dose as a reference dose in thenon-inferiority trial. Even though the clinical trialreached positive results and concluded that the newdrug was non-inferior to the (low dose of) activecontrol, the regulatory agency rejected the market-ing application of the new drug because the choiceof the low dose as a reference dose was inconsistentwith the treatment guideline.

Recent Advances in China’s Regu-latory ScienceSince early 2016, the CFDA has initiated several ma-jor changes in its regulations for drugs and med-ical devices and these changes have created theheadline news in pharmaceutical business-focusedmedia, e.g., Wang and Davidson (2017), Gregory(2017), Bloomberg News (2017), Cyranoski (2017),Balzano et al. (2017) and McCallister (2017). Thechanges reflect the agency’s ambitious frameworkto modernize its regulatory science by reducingthe time for the review and approval processes,encouraging innovative medical product develop-ment, and promoting international standards. Thissection summarizes these remarkable regulation re-forms in the order as they were announced.

Prioritization of backlog applications for re-view. The time from submission to review and ap-proval of CTA or NDA is usually quite lengthy, pri-marily because of a large volume of applicationsand an insufficient number of CDE staff handlingthese applications (see Section An overview of drugregulations for a brief review of CDE staffing his-tory). The immediate consequence of the largebacklog of applications is the resulting delay inmarketing of new drugs in China to meet Chinesepatients’ needs. On February 26, 2016, the CFDAannounced ia priority review initiative covering thefollowing:

(a) Drug categories: (1) innovative drugs notmarketed anywhere; (2) imported innovativedrugs (marketed in other countries but not inChina); (3) therapies with preparations usingadvanced technologies, or innovative treat-ment methods, showing clear clinical bene-fits; (4) CTAs for drugs with 3-year patentexpiration, and manufacturing applicationsfor drugs with 1-year patent expiration; (5)CTAs with clinical trials approved by FDAand EMA, and NDAs for drugs approved byFDA and EMA; (6) NDAs for TCMs that showclear advantages in treatment and preventionof diseases of public health importance; and(7) NDAs in national critical research pro-grams.

(b) Disease areas: (1) AIDS; (2) tuberculosis; (3)viral hepatitis; (4) rare diseases; (5) malignanttumors; (6) pediatric drugs; and (7) drugs forprevalent diseases in the elderly population.

Visit http://www.sda.gov.cn/WS01/CL0844/145260.html for more information on this

22 © www.icsa.org

http://www.sda.gov.cn/WS01/CL0844/145260.htmlhttp://www.sda.gov.cn/WS01/CL0844/145260.html


initiative.Establishment of advisory committees. To im-

prove transparency, scientific accuracy, and the reg-ulatory review system, the CFDA announced onMarch 19, 2017, the creation of an advisory bodyfor the following 10 committees: (1) reform and de-velopment policy; (2) safety and risk management;(3) theory and formulation of TCM; (4) GMP andquality assurance for TCM; (5) pharmacology forchemical drugs and biologic products; (6) pharma-cotoxicology; (7) pharmaceutical and clinical phar-macology; (8) biostatistics; (9) pediatric drugs; and(10) legal, economic and social. Ten disease-specificcommittees for TCM and 16 disease-specific com-mittees for chemical drugs were also established.These advisory committees will assist CDE in thereview and evaluation processes of applications;visit http://www.sda.gov.cn/WS01/CL0087/170640.html for a detailed announcement.

Adjustment of approval authority. On march17, 2017, the CFDA announced transferring the ap-proval authorization from the CFDA to CDE for(1) drug CTA (both domestic and imported drugs);(2) drug supplemental applications (both domes-tic and imported drugs); and (3) imported drug re-registration. This change is expected to improvereview and approval efficiency and decrease thewaiting time (http://www.sda.gov.cn/WS01/CL0053/171383.html).

Expedited review of innovative products. OnMarch 11, 2017, the CFDA announced an initiativeto implement the expedited review and approvalprocesses for innovative drugs and medical devices:(1) Conditional approval can be given based onearly or mid-stage trial results for life-threateningdiseases for which no efficacious treatments areavailable; (2) for rare diseases, clinical trial waiver orconditional approval can be granted based on lim-ited data or overseas approval; (3) establishing re-view teams comprising medical physicians, phar-macologists, toxicologists and statisticians for newdrug review process (clinical and safety); (4) estab-lishing review teams comprising experts in medicalscience, diagnosis, electronics, material science, andbioengineering for review of medical products frombasic science perspective; (5) encouraging clinicalapplications of innovative drugs; and (6) encourag-ing development of TCM (http://www.sda.gov.cn/WS01/CL0087/172567.html).

In-dept reform on review and approval pro-cesses for innovative products. On October 8,2017, the General Office of the Communist Partyof China Central Committee and the State Counciljointly announced an in-depth reform on the review

and approval processes to encourage innovations ofdrugs and medical devices (http://www.sda.gov.cn/WS01/CL1747/178283.html). Specifi-cally, the reform covers

(a) Clinical trials: (1) Clinical trial site: clini-cal/medical institutions with clinical trial ca-pability may register via the CFDA web siteas clinical trial sites; (2) enhanced ethics com-mittee system: no re-approval is needed if aclinical trial is approved by the ethics com-mittee of the primary investigator’s institu-tion; (3) CTA: the CFDA MUST issue a recom-mendation letter for a CTA within a specifiedtime period; (4) overseas clinical trial data: theCFDA accepts overseas clinical trial data tosupport NDA in China; (5) expanded use ofclinical trial drugs: for life threatening dis-eases with no efficacious therapies, the newtreatment can be used in non-clinical trial pa-tients if (i) it shows benefits; (ii) it is approvedby EC; (iii) patients sign the informed consentform (ICF); (iv) no major safety concerns ex-ist; and (6) legal responsibility for clinical tri-als and clinical trial data: any type of fraudwill be prosecuted.

(b) Regulatory approval: (1) Products for life-threatening diseases or public health urgencywith no efficacious therapies: (i) early- or mid-stage results with predictable clinical bene-fits may support conditional marketing au-thorization; manufacturers need to stipulaterisk management plan, (ii) priority review canbe given to scientifically impactful innovativeproducts; and (2) products for rare diseases:clinical trial waiver and/or conditional ap-proval can be granted for products approvedin other countries if the sponsor provides arisk management plan (and possibly commitsto follow-up studies).

Adjustment for imported drug registration.On October 10, 2017, the CFDA announced theadjustment of imported drug registration, whichstates that (1) simultaneous phase I clinical trials canbe conducted (except for preventive biologic prod-ucts); (2) sponsors may follow the PDR to submitan NDA upon completion of MRCTs; (3) no over-seas approval in the sponsor’s country is requiredfor CTA and NDA of imported drug and therapeu-tic biologic products; and (4) clinical trial waiversbased on MRCT data can be granted per relevantclauses in the PDR (http://www.sda.gov.cn/WS01/CL0050/178363.html).

© www.icsa.org 23

http://www.sda.gov.cn/WS01/CL0087/170640.htmlhttp://www.sda.gov.cn/WS01/CL0087/170640.htmlhttp://www.sda.gov.cn/WS01/CL0053/171383.htmlhttp://www.sda.gov.cn/WS01/CL0053/171383.htmlhttp://www.sda.gov.cn/WS01/CL0087/172567.htmlhttp://www.sda.gov.cn/WS01/CL0087/172567.htmlhttp://www.sda.gov.cn/WS01/CL1747/178283.htmlhttp://www.sda.gov.cn/WS01/CL1747/178283.htmlhttp://www.sda.gov.cn/WS01/CL0050/178363.htmlhttp://www.sda.gov.cn/WS01/CL0050/178363.html


Requirements on acceptance of overseas clini-cal trial data. To implement the “in-dept reform onreview and approval processes for innovative prod-ucts” announced by the central government, CDEposted on October 20, 2017, the technical require-ments for the acceptance of overseas clinical trialdata (http://www.cde.org.cn/news.do?method=largeInfo&id=314075). The requirementscan be summarized as follows:

(a) General principles: Submitted MRCT datamust be authentic, complete, accurate andtraceable.

(b) Completeness: (1) Overseas clinical trials tosupport NDA (not marketed anywhere): sub-mit ALL (not selective) clinical trial data; (2)overseas marketed products to support NDA:submit ALL overseas clinical trial data ANDperiodic safety update report (PSUR) and/ordevelopment safety update report (DSUR); (3)MRCT to support NDA: submit ALL domes-tic and overseas clinical trial data and results;and (4) overseas early clinical trial data to sup-port domestic CTA: submit all analysis resultsand data from overseas early clinical trials.

(c) Basic technical requirements for overseas clin-ical trial data to support NDA: The NDApackage must include (1) biopharmaceutics:bioavailability and bioequivalence; (2) clinicalpharmacology: pharmacokinetics and phar-macodynamics, regional / ethic differences tosupport efficacy and safety; (3) efficacy: re-sults from global pivotal trials and clinical tri-als in China, consistency in efficacy betweenoverall population and Chinese population;(4) safety: ALL safety data and analysis, con-sistency of overall population and Chinesepopulation; and (5) under PDR requirementsand based on ICH E5, sponsor must evaluateconsistency of Chinese population and over-all population in terms of efficacy and safety,to support extrapolation of results from non-Chinese population to Chinese population.

(d) Acceptability of overseas clinical trial data:

(i) Fully acceptable – (1) Authentic, compli-ant with GCP and clinical trial inspec-tion and data verification requirements;(2) support target indications with re-spect to efficacy and safety; and (3) noethnic sensitivity factors that impact ef-ficacy and safety,

(ii) Partially acceptable – (1) Authentic, com-pliant with GCP and CT inspection re-quirements; (2) support target indica-tions in terms of efficacy and safety; (3)include some ethnic sensitivity factorsthat impact efficacy and safety leadingto uncertainty of extrapolation from non-Chinese to Chinese populations; and (4)sponsor should analyze the influentialfactors and (following discussion withagency) conduct additional (domestic)clinical trials,

(iii) Unacceptable – (1) Serious issues exist inclinical trial data with respect to authen-ticity, completeness, accuracy, and trace-ability; (2) overseas clinical trial data donot support target indications in termsof efficacy and safety; (3) ethnic fac-tors exist that impact efficacy and safety;and (4) sponsor should follow innovativemedicine development guidelines andconduct a series of domestic clinical tri-als to support efficacy and safety claimsfor the Chinese population.

(e) Life-threatening conditions, orphan diseases,pediatric products: Applications that belongto “partially acceptable” can be granted “con-ditional” approval with sponsor’s commit-ments to collecting and submitting more effi-cacy and safety data in Chinese patients aftermarketing authorization in China.

Concluding remarksThis paper provides an overview China’s biophar-maceutical market, its potential for future growth,and the country’s drug regulatory system. It alsodescribes guidelines pertaining to clinical trials andnew drug applications. A high-level summary ofthe ground-breaking policy and regulation changes(resulting from recent reforms) that will directly im-pact the drug development programs in China isalso provided. With the CFDA joining the ICH asan agency m

bulletin · count of an event where statistics and the statisti-cians made significant impacts in...

Documents