VOL 19, NO 6
`
`DECEMBER 1992
`
`Seminars in
`
`EDITOR-IN-CHIEF
`John W. Yarbro, MD, PhD
`
`ASSOCIATE EDITORS
`Richard S. Bornstein, MD
`Michael J. Mastrangelo, MD
`
`New Antitumor Drugs in
`Development
`
`Contributors
`
`Supplied by the British Library 29 Mar 2018, 18:43 (BST)
`
`
`
`1 of 19
`
`Celltrion, Inc. 1037
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`Seminars in Oncology
`
`Seminars in Oncology (ISSN 0093-7754) is published bimonthly by W .B. Saunders Company, Corporate
`and Editorial Offices: The Curtis Center, Independence Square West, Philadelphia, PA 19106-3399.
`Accounting and Circulation Offices: W.8. Saunders Company, 6277 Sea Harbor Dr, Orlando, FL
`32887-4800. Months of issue are February, April, June, August, October, and December. Second class
`postage paid al Orlando, FL 32887-4800, and al additional mailing offices.
`POSTMASTER: Send change of address 10 Seminars in Oncology, W.B. Saunders Company, Periodicals
`Depart men I, 6277 Sea Harbor Dr, Orlando, FL 32887-4800.
`Editorial correspondence should be addressed to John W. Yurbro, MD, PhD, Regional Cancer Center,
`Memorial Medical Ce111er, Springfield, IL 62781-000 I.
`Uusiness correspondence (subscriptions. change of address) should be addressed 10 the Publisher, W.B.
`Saunders Company, Periodicals Department, 6277 Sea I I arbor Dr, Orlando, FL 32887-4800.
`Cha11ge of address notices, including botlr the old and new addresses of the subscriber, slrould be sent at least
`01111 111ontlt in advance.
`Customer Service: 1-800-654-2452.
`Yearly subscription rates: United States and possessions: imlividuals, $98.00; insl i1u1ions, $136.00; single
`issues, $31.00. All or her countries: individuals, S 154.00; inst itutions, $154.C)(); single issues, $31.00. For all
`areas outside the United States and possessions, there is no additional charge for surface delivery. For air
`m<iil delivery, add $36.00. Students and residents: United States <ind possessions, and Canada: $59.00; all
`01her countries: $154.00. To receive student/resident rate, orders must be accompanied by name of
`affiliated institution, date of term, and the signature of program/residency coordinator on institution
`letterhead. Orders will be billed at individual rate until proof of s1arus i~ received. Current prices are in
`effect for back volumes and back issues. Back issues sold in conjunction with a subscription are on a
`prorated basis. Current and back single is.~ues exist in limited quantities and arc offered for sale subject to
`availability. 1991bo1111d1'0/11me price: $80.00: foreign customer!>, please atld $15.00 for postage. To purchase
`a 1991 bound volume, customer must be a subscriber for 1991. Cumulative Index ( 1980-1989) price: S95.00;
`foreign customers, please add $2.25 for surface delivery, or SS.00 for air mail delivery. Prices are subject to
`change without notice. Cit eeks should be made payable to W.B. Saunders Company and sent 10 Seminars in
`011cology, W.B. Saunders Company, Periodicals Department, 6277 Sea Harbor Dr, Orlando, FL
`32887-4800.
`Agents for the United Kingdom, Ireland, and Europe: Harcourt Brace Jovanovich, Ltd, 2..J-28 Oval Rd,
`London NWI 7DX, England. Agents for Australia and New Zealand: I larcourt Brace Jovanovich. Lid,
`30-52 Smidmore St (Locket.I Bag 16). Marrid.·ville, NSW 2204, Australia. Agents for Japan and Korea:
`Harcourt Brace Jovanovich. Ltd, lchibancho Central Bit.lg., 22- 1 lchihancho, Chiyoda-Ku, To!,.)'o 102,
`Japan.
`Copyright c 1992 by W.B. Saunders Company. All rights reserved. No part of this publication may be
`reproduced or transmitted in any form or by any means, elcc1m11ic or mc.:chanical, including photocopy,
`recording, or any information storage and retrieval systc.:m, withou1 permission in writing from the
`publisher. Printed in the Uni1ed States of America.
`Correspondence regarding permission to reprint all or part of any article published in this journal should
`be addressed 10 W.B. Saunders Co .. Periodicals Department. Orlando, FL 32887. Telephone number
`1-407-345-2500.
`Other correspondence ( copyediting, production) should l>e addressed 10 W. B. Sa undcrs Co., The Curtis
`Center, Independence Square West, Philadelphia, PA 19106-3399.
`The appearance of the code al the bo11om of the fir~t page of an article in this journal indicates the
`copyright owner's consent that copies of 1hc article may be made for personal or internal use, or for the
`personal or internal use of specific clients. This consent is given on 1he condition, however, that the copier
`pay 1he stated per-copy fee through the Copyright Clearance Center, Inc. (27 Congress St, Salem. MA
`01970) for copying beyond that permitted by Sections 107 or 108 of the US Copyright Law. This consent
`docs not extend to other kinds of copying, such as copying for general distribution, for advertising or
`promotional purposes, for creating new collective works, or for resale. Absence of the code indicates that
`1he material m<iy not be processed through the Copyright Cle:1rance Center, Inc.
`Advertising representative: Cunningham Associates, PO Box 308, Park Ridge, NJ 07656, telephone
`1-201-767-4170.
`Publication of an advertisement in Seminars i11 Oncology docs not imply endorsement of its claims by the
`Editor(s) or Publisher of the journal.
`Every effort has been made to check generic and trade names, and 10 verify drug doses. The ultimate
`responsibility, however, lies with the prescribing physician. Please convey any errors to the Editors.
`
`W .B. Saunders Company
`
`Philadelphia, PA
`
`Harco11rt Bracejova11ovich1 Ju e.
`
`Supplied by the British Library 29 Mar 2016, 16:43 (BST)
`
`
`
`2 of 19
`
`Celltrion, Inc. 1037
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`The National Cancer Institute: Cancer Drug Discovery and
`Development Program
`
`Michael R. Grever, Saul A. Schepartz, and Bruce A. Chabner
`
`The discovery end development of novel therapeutic
`products for the treetment of melignancy is vitally
`important to those physicians responsible for the
`menegement of cencer petients. A description of the
`ongoing efforts et the Netionel Cancer Institute (NCI)
`is intended to provide insight into those complex
`processes necessery to eccomplish this mission. An
`update on the NCl's revised cancer screen is accompa·
`nied by a brief summary of those new agents sched·
`uled to be entered into clinical investigetion in the near
`future. The tremendous potential advantages and cha I·
`lenges associated with the use of a molecular ap·
`proach to cancer drug design are discussed. Despite
`the differences of opinion that may exist regerding the
`optimel strategies for accomplishing the mission, there
`is no disegreement regard ing the importance of the
`effort to find effective new therapies for cancer pa·
`tients.
`This is • US government work. There •re no restric(cid:173)
`tions on its use.
`
`A LTHOUGH THE DISCOVERY of effec(cid:173)
`
`tive ant icancer agents has occurred in
`academic centers and industry, the National
`Cancer Institute (NCI) has played a pivotal role
`in cancer drug discovery and development. 1•3 In
`1955, the Cancer Chemotherapy National Ser(cid:173)
`vice Center (CCNSC) at the NCI was estab·
`lished. 1 Since the creation of this national re(cid:173)
`source, the NC I has been involved with either
`the discovery o r developmental tasks that were
`essential for the approval of the majority of
`comme rcially ava ilable anticancer agents.
`The primary responsibility of the NCl's pre(cid:173)
`clinical drug evaluation program was intended
`to focus on the treatment of malignancy. How·
`ever, the public health emergency that emerged
`over the past decade from the human immuno (cid:173)
`deficiency virus (HIV) necessitated the involve(cid:173)
`ment of the NCI in the discovery and develop(cid:173)
`ment of effective therapeutic products to treat
`
`From tire Oevt!lopmental V1erapeutics Program, and tire
`Office of tire Director, D111isio11 of Cancer Treatment. National
`Cancer lnstitllle, Btthesda, MO.
`Address repn111 requests to Michael R. Grever, MD. National
`Canctr Jnstit11te. Executive Pla:a North, Room 843, Bethesda.
`MD20892.
`This is a US go1'emment work Tlrere are no restrictions on
`its rise.
`0093-7754/9211906-0003$0.00/0
`
`patients with acquired immunodeficiency syn(cid:173)
`drome (AIDS). Although substantial changes in
`the organizational structure occurred, many of
`the preclinical functions (ie, pharmacology, tox(cid:173)
`icology, analytical chemistry, formulation re(cid:173)
`search, etc.) are essentia lly ide ntical for both
`programs. Thus, resources already in existence
`for cancer drug discovery and development
`were rapidly mobilized in response to the criti·
`cal public health issues associated with HIV(cid:173)
`induced illness. The National Institute of AJ(cid:173)
`lergy and Infectious Diseases (NIAID) a lso
`maintains a preclinical drug evaluation program
`that works closely with the NCI to address this
`major crisis.
`The NCI will remain committed to the ardu(cid:173)
`ous tasks of drug discovery and development
`because a meaningful extension of high-quality
`life for patients with either cancer or AIDS
`hinges on continued success in these areas. T he
`propensity for malignancies to develop in pa(cid:173)
`tients receiving effective treatment for A IDS
`underscores the necessity for the NCI to be
`integrally involved in the search for novel thera(cid:173)
`peutic agents for both fatal diseases. The Devel(cid:173)
`opmenta l Therapeutics Program (DTP) of the
`Division o f Cancer Treatment (OCT) at the
`NC I is responsible for those preclinical activi(cid:173)
`ties necessary for both cancer and AIDS drug
`discovery and development. It is important to
`emphasize that promising drugs may be submit(cid:173)
`ted to the program for consideration at virtually
`any stage in development in order to maximally
`use the preclinical and clinical resou rces of the
`NCI. This willingness to cooperate with industry
`is reflected in the growing number of Investiga(cid:173)
`tio nal New D rug Appl ications (INDs) being
`fi led in recent years, many of these representing
`cooperative drug development efforts with indus(cid:173)
`try. The purpose of this article is to describe the
`current structure and functional operation of
`this important preclinical program.
`
`OVERVIEW OF NCI CANCER DRUG DISCOVERY
`AND DEVELOPMENTAL EFFORTS
`
`The actual preclinical responsibilities and
`tasks of the DTP are accomplished through the
`
`622
`
`Seminars in Oncology, Vol 19, No 6 !0ecember), 1992: pp 622-638
`
`
`
`3 of 19
`
`Celltrion, Inc. 1037
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`CANCER DRUG DISCOVERY AND DEVELOPMENT AT NCI
`
`623
`
`use of intramural research efforts and extramu(cid:173)
`ral resources involving diverse mechanisms (eg,
`contracts, grants, cooperative agreements).
`There are five intramural laboratories and nine
`extramural branches within the DTP that bring
`agents from the point of discovery to clinical
`trial (Table I). Although many of the research
`functions are conducted within a specific labora(cid:173)
`tory or branch, the majority of the projects
`require an interdisciplinary collaboration be(cid:173)
`tween these well-defined administrative units.
`The basic preclinical drug discovery and devel(cid:173)
`opmental tasks are summarized in Table 2.
`Over the past 2 years, an emphasis has been
`placed on the use of agent-specific working
`groups to usher each agent over the many
`preclinical hurdles in the drug evaluation
`" pipeline. ·· This general approach to project
`management directed at a specific therapeutic
`product has been successfully used by the phar(cid:173)
`maceutical industry. In the past at the NCI ,
`specific agents were handed on to individual
`administrative units as each defined task was
`completed. This change in product manage(cid:173)
`ment will expedite the complex processes en(cid:173)
`countered with drug development.
`Although various approaches could be used
`to describe the NCl's drug discovery and devel(cid:173)
`opmental efforts, the schematic diagram de(cid:173)
`picted in Fig 1 outlines the current functional
`components of the DTP. The procurement of
`defined chemical entities and crude natural
`products for testing in either the NCl's cancer
`screen or the antiviral screen is coordinated
`
`Table 1. Developmental Therapeutics Program of the
`National Cancer Institute Administrative Units
`
`Intramural Laboratories
`Biological chemistry
`Molecular pharmacology
`Medicinal chemistry
`Drug discovery research and development
`Pharmaceutical chemistry
`Extramural Branches
`Drug synthesis and chemistry
`Biological testing
`Natural products
`Antiviral evaluation
`Pharmacology
`Toxicology
`Pharmaceutical resources
`Information technology
`Grants and contracts operations
`
`Table 2. National Cancer Institute Preclinical Program
`
`Drug Discovery:
`functions
`
`Drug Development:
`Functions
`
`Acquisition
`Defined chemical enti·
`ties
`Crude natural products
`
`Chemical modification:
`Lead optimization
`Retest modified struc·
`tu res
`Examine structure·ac·
`tivity relationship
`
`In vivo testing
`Evaluate therapeutic
`index
`Optimize dose/route/
`schedule
`
`Scale· up bulk drug pro·
`duction
`
`Formulation research and
`production
`
`Pharmacological studies
`
`Toxicological evaluation
`
`File investigational new
`drug application (IND)"
`
`"This task performed by CTEP, OCT. and NCI.
`
`through two extramural branches (ie, Drug
`Synthesis and Chemistry Branch and the Natu(cid:173)
`ral Products Branch). Vigorous programs of
`acquisition are supported to search for novel
`chemical structures or natural products to be
`tested in the in vitro screens. Substantial effort
`has been made to seek submissions from aca(cid:173)
`demic institutions and industrial sources. The
`advent of the in vitro screens has made the
`requirement for large quantities of material for
`the initial submission less important. In the
`past, the cancer screen used in vivo models for
`the initial antitumor evaluation and thus re(cid:173)
`quired submission of gram quantities of mate(cid:173)
`rial. In contrast, the current screen can provide
`an initial evaluation using an amount in the
`range of 25 to 50 mg. This microacquisition
`approach has enabled individual chemists work(cid:173)
`ing with limited resources to contribute novel
`chemical structures for evaluation and has al(cid:173)
`tered the profile of the suppliers to the program
`as shown in Fig 2. Whereas past acquisition of
`agents for evaluation was somewhat limited to
`those suppliers with resources for making large
`quantities of new chemicals, the current ap(cid:173)
`proach should optimize the chemical diversity
`of compounds submitted to the program.
`
`Acquisition
`The chemical acquisition program is aided by
`an extramural contractor with an annual target
`to procure approximately 10,000 new chemical
`structures that have not been previously exam-
`
`
`
`4 of 19
`
`Celltrion, Inc. 1037
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`'Pharmaceutlcal
`Academic
`l ndlvlclual
`Intramural
`
`- - - - - - . . Decision Network
`Committee
`
`L
`
`.......
`
`I ~
`
`Tesllng
`
`Operating Committees - - - - - - (cid:173)
`(Wort<ing Groups)
`
`LPC
`(lnlfamural lab)
`
`~ t
`
`OTP
`Exiramural
`Branches
`
`Fig 1. Preclinical cancer drug
`discovery and development at
`the NCI. This schematic represen·
`tation of the major components
`of the preclinical program out(cid:173)
`lines the overall flow that agents
`traverse at the NCI during the
`processes involved with drug dis(cid:173)
`covery and development. Suppli·
`ers may enter agents Into this
`system at any stage in their de·
`velopment in order to use the
`preclinical and clinical resources
`of the NCI.
`
`624
`
`GREVER, SCHEPARTZ. AND CHABNER
`
`Agudsttion lnosn Committee
`
`Extramural
`Biological Tesllng
`and
`~ Natural Product Branches
`. .
`,...--,--..,,.-----,~ ~~~
`Primary Cancer
`Screen
`(In Vitro)
`
`LO ORO
`(ln1ramural Lab)
`
`. .
`
`Drug Syn1hesls
`and Chemtslry
`Branch
`
`// ~~~
`I ~ !~~~.!:~~-=----~
`..
`
`Phase I Cl!o)ca! lnyesllgAJign
`
`ined by the NCI. A major stipulation for submis(cid:173)
`sion of an agent is disclosure of the chemical
`structure to permit the NCI staff to verify that
`the compound has not been previously submit-
`
`1980 - 1903
`
`1084 - 1987
`Year or Assignment
`
`1988 - 1992
`
`Fig 2. Agents acquired for testing by the NCI: Industry
`versus university. Over the past decade, there has been a
`change in the relative contribution of various suppliers of
`chemical agents for testing in the NCI cancer screen. The
`previous cancer drug screens used In vivo testing procedures
`that required the submission of gram quantities of prospective
`egents for Initial testing. Since the NCl's development of the
`current in vitro screening procedures permits en initiel analy(cid:173)
`sis of antitumor activity to be made with much smaller
`amounts of material. the diversity of chemicels acquired for
`testing can now be increased by extending the opportunity for
`compound submission to many suppliers with limited re·
`sources for lerge-scale production. • . Industry; and El, univer(cid:173)
`sity.
`
`ted by another supplier. After testing, the NCI
`will provide a full report of the results from the
`in vitro assay(s) for anticancer and/or anti-HIV
`activity. In general, suppliers are encouraged to
`permit compound testing in both NCI screens
`(ie, the anticancer screen and the anti-HIV
`screen).4·s This testing is performed at no cost to
`the supplier, and the results are maintained in
`strict confidence.
`The NCI has recognized that natural prod(cid:173)
`ucts represent a tremendous potential resource
`for cancer drug discovery. Many of the currently
`used anticancer agents are natural products or
`are derived from leads discovered from natural
`sources. 6 The complex chemical structures that
`are found within nature far exceed the imagina(cid:173)
`tion and synthetic capabilities of medicinal
`chemists. The chemical entities that are discov(cid:173)
`ered to have potential medicinal value may
`serve either as a drug candidate or as a critically
`important lead for structural modification. Thus,
`careful assessment of the chemical repository
`existing within nature provides substantial op(cid:173)
`portunity to secure novel drug candidates.
`Collection contractors for the NCI have made
`a concerted effort to use native expertise to
`
`
`
`5 of 19
`
`Celltrion, Inc. 1037
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`CANCER DRUG DISCOVERY AND DEVELOPMENT AT NCI
`
`625
`
`identify those materials (eg, tropical plants)
`reported in folklore to have medicinal value.
`This effort to use existing knowledge is termed
`ethnopharmacology. A high yield of positive
`leads may indeed be forthcoming as a result of
`interaction of the collectors and the native
`experts.
`Therefore, the natural product acqu1s1t1on
`program is interested in the procurement of
`crude material of diverse origin ( eg, plant,
`marine, or microbial organisms). The natural
`products are initially classified extensively by
`either botanists or microbiological experts with
`respect to their identity and location of collec(cid:173)
`tion. Furthermore, additional information relat(cid:173)
`ing to the conditions of collection are retained
`by the Natural Products Branch in its Natural
`Product Repository located in Frederick, MD.
`The actual field collections for the raw material
`include diverse geographical areas spanning
`from the Far East, to Central and South Amer(cid:173)
`ica, and finally into the tropical areas of Africa.
`The information relative to the actual condi(cid:173)
`tions of collection and processing are critically
`important to insure that the same chemical
`compound will be isolated in a reasonable yield
`from subsequent recollections.
`The NCI's Natural Product Repository repre(cid:173)
`sents an extremely valuable national resource
`for new drug discovery. The diversity of organ(cid:173)
`isms that have been stored in this facility is
`depicted in Table 3. The Natural Products
`Branch coordinates the systematic extraction
`(both aqueous and organic) of this raw material.
`At present, these crude extracts are tested at a
`single concentration in the cancer screen. Over
`the past 6 months, more than 25,000 different
`extracts have been tested. When a crude extract
`is found to have antitumor activity, further
`fractionation of the positive lead extract is
`necessary to isolate the chemical constituents
`responsible fo r the observed effect. The ulti(cid:173)
`mate goal of this process is the careful character(cid:173)
`ization of the chemical entities within the natu-
`
`Table 3. Natural Products Repository of the
`National Cancer Institute
`
`Major Category
`
`Organisms Received
`
`Extracts
`
`Terrestrial plants
`Marine plants
`Marine invertebrates
`Cyanobacteria and fungi
`
`30.521
`1,973
`4,504
`5,767
`
`39,572
`2.197
`8,434
`17,960
`
`ral products with potential antitumor effects.
`The Laboratory of Drug Discovery Research
`and Development (LDDRD) is the intramural
`DTP laboratory with expertise in natural prod(cid:173)
`uct chemistry that directs the search for those
`intriguing chemical entities within the extracts
`that are responsible for the antitumor activity.
`The Natural Product Branch supports their
`efforts to purify the interesting extracts and
`coordinates the re-collection of raw material to
`provide sufficient biomass for isolation of those
`chemical entities thought to have potential
`medicinal value. A tremendous effort is made to
`determine whether these interesting leads from
`nature will in fact represent valid drug candi(cid:173)
`dates.
`
`SELECTION OF CHEMICALS AND NATURAL
`PRODUCTS FOR EVALUATION
`The process of selecting defined chemical
`structures for initial evaluation in either the
`cancer or AIDS screens is made by the DTP's
`Acquisition Input Committee in collaboration
`with medicinal chemists in the Drug Synthesis
`and Chemistry Branch. The tremendous testing
`capacity requires an organized approach to
`establish some basis for prioritizing the large
`number of agents submitted to the program.
`This group selects agents based on their unique(cid:173)
`ness of chemical structures, and computerized
`programs have been developed that assist in this
`process.7 Obviously, new chemical classes of
`agents that have not had extensive clinical
`evaluation are prime targets. Furthermore, new
`agents with either a potential novel mechanism
`of action or selectivity for a key target within
`neoplastic cells will be assigned a high priority
`for testing.
`Screening efforts are intended to find interest(cid:173)
`ing lead compounds a nd thus require high
`volume testing. The processes for selection and
`prioritization of testing become critically impor(cid:173)
`tant. Therefore, the Acquisition Input Commit(cid:173)
`tee is responsible for interacting with the suppli(cid:173)
`ers, selecting the queue for testing of the
`numerous agents and extracts in the respective
`screens, and reporting the in vitro data to the
`suppliers in a reasonable time frame (eg, recent
`turnaround time is averaging 3 months).
`The NCI Chemical Repository currently holds
`over 400,000 chemical structures within its data
`base. The program is currently converting this
`
`
`
`6 of 19
`
`Celltrion, Inc. 1037
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`626
`
`GREVER, SCHEPARTZ, AND CHABNER
`
`two-dimensional chemical str.uctural data base
`into a three-dimensional representation. The
`creation of this new data base will markedly
`enhance the assessment of the uniqueness of
`chemical structures submitted to the program.
`Furthermore, this modification of the chemical
`data will provide an opportunity for improved
`evaluation of structure-activity relationships and
`will introduce the potential capability for molec(cid:173)
`ular modeling in the future design of novel
`therapeutic products.
`The procurement of sufficient material for
`secondary in vivo testing can present a formida(cid:173)
`ble challenge for the entire program consider(cid:173)
`ing the diversity and complexity of chemical
`structures involved. The substantial time, effort,
`and cost are incurred either with the re(cid:173)
`collection and isolation of natural products or
`the resynthesis of chemical entities. However,
`these tasks are extremely important considering
`that an in vitro antitumor effect does not ensure
`that the lead will translate into a useful antican(cid:173)
`cer drug. The responsibilities for obtaining an
`amount of material sufficient to permit these
`secondary evaluations for the targeted agents
`are shared by three branches (Drug Synthesis
`and Chemistry, Pharmaceutical Resources, and
`Natural Products).
`Secondary drug evaluation efforts require
`variable quantities of material depending on the
`potency of the agent under consideration (ie, a
`range from a milligram to several gram quanti(cid:173)
`ties). The difficulties encountered with scaling
`up resources for the procurement of several
`grams of either an interesting natural product
`or a complex synthetic chemical entity can
`involve a delay of more than 6 months in testing
`and can cost thousands of dollars. The procure(cid:173)
`ment of smaller quantities for secondary in vivo
`evaluation may suffice in the case of a highly
`potent natural product. The challenges of an
`inadequate drug supply can adversely affect
`both the preclinical and the subsequent clinical
`evaluation of an agent (eg, taxol). Despite the
`difficulties of dealing with the complex and
`chemically unique structures that are character(cid:173)
`istic of the natural products, this group presents
`truly exciting opportunities for introducing new
`mechanistic classes of drugs for clinical investi(cid:173)
`gation.
`
`THE NCI CANCER SCREEN
`
`The two basic approaches to cancer drug
`discovery involve either empirical methods or
`attempts at rational drug design. The majority
`of the current antineoplastic agents were discov(cid:173)
`ered either through empirical screening efforts
`or represent chemical modifications derived
`from lead compounds discovered in a cancer
`drug screen.~.v Thus, antitumor screens have
`been extremely useful in identifying "leads" (ie,
`the first chemical entity in a series discovered to
`be active), which may be subsequently modified
`to enhance the therapeutic index.2
`Historically, the NCI cancer screens have
`relied heavily on murine leukemias (L1210 and
`P388) as the index tumors in an in vivo screen(cid:173)
`ing modet. 1.m.r 1 This approach identified drug
`candidates that obviously had the capability of
`crossing cell membranes and surviving an in vivo
`route of administration. However, the drugs
`that resulted from the earlier cancer screens
`were shown to be primarily effective in the
`treatment of human leukemias and lymphomas.
`The success in the treatment of these malignan(cid:173)
`cies may have reflected the index tumors tha~
`had been used in the primary screen.
`Subsequently, changes in the NCI's approach
`to screening between 1975 to 1985 included the
`addition of animal solid tumors and human
`tumor xenografts to a secondary in vivo tumor
`panel in an effort to enhance the identification
`of agents that would be effective in the treat(cid:173)
`ment of human solid tumors. HI However, the
`initial stage of the NCI cancer screen still used a
`murine leukemia/lymphoma cell line and re(cid:173)
`quired demonstration of activity in this line in
`order to be referred to the secondary tumor
`panel. Thus, the addition of the solid tumor
`panel did not result in a major change in the
`general propensity for the cancer screen to
`identify agents with primary activity in the
`hematologic malignancies.
`In 1985, a major modification of the cancer
`screening strategy was undertaken after exten(cid:173)
`sive consultation and discussion with advisory
`groups. The NCI made the decision to use
`human tumor cell lines in an in vitro assay as the
`primary cancer screen. 10 A total of sixty human
`tumor cell lines, derived from seven cancer
`types (lung, colon, melanoma, renal, ovarian,
`brain, and leukemia), formed the basis of this
`
`
`
`7 of 19
`
`Celltrion, Inc. 1037
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`CANCER DRUG DISCOVERY AND DEVELOPMENT AT NCI
`
`627
`
`novel approach. The initial concept proposed
`that leads showing disease-panel specificity
`would be identified and could be subsequently
`examined further by in vivo models derived
`from the most sensitive in vitro index tumor
`lines. This new screen was unique in several
`respects. In particular, it changed the general
`approach from a compound-oriented drug dis(cid:173)
`covery effort to a disease-panel oriented exer(cid:173)
`cise. Additional reasons for changing from the
`previous approach to the current NCI screen
`included (1) emphasis on human tumor cells
`derived from solid tumors; (2) development of a
`high-volume screening device that could adapt
`to either numerous chemical agents or extracts
`(in support of the renewed orientation to natu(cid:173)
`ral products as a source for cancer drug discov(cid:173)
`ery); and (3) establishment of an in vitro screen
`that would avoid the use of mice, save on the
`amount of material required for the initial
`screening evaluation of a compound, and lend
`itself to automation.
`In the current NCI anticancer screen, each
`chemical agent is tested over a broad concentra(cid:173)
`tion range against every cell line in the panel. Hl
`The tumor cells are inoculated over a series of
`standard 96-well microtiter plates on day 0.
`These cells are then preincubated on the micro(cid:173)
`titer plate for 24 hours. The test drugs ( 400/wk)
`are added to the wells in five 10-fold dilutions
`starting from the highest soluble concentration.
`Because the NCI appreciated the tremen(cid:173)
`dous potential for novel drug discovery within
`the area of natural products, extensive efforts
`were made to adapt the new cancer screen to
`test large numbers of crude natural product
`extracts. Although the initial approach to screen(cid:173)
`ing the extracts did involve serial dilutions, the
`capacity for screening of the materials from the
`Natural Product Repository has been markedly
`enhanced by modifying the approach to the
`initial screening of the crude extracts. Over the
`past 6 months, the crude extracts have been
`tested at single concentrations, thus increasing
`the capacity to test over 1,000 extracts per week.
`Follow-up studies or examination of specific
`fractions of the initial extracts are then coordi(cid:173)
`nated by the Natural Products Branch in collab(cid:173)
`oration with the natural product chemists within
`LDDRD.
`The actual assay involves an incubation of
`
`either the chemical agents or extracts for 48
`hours with the tumor cell lines.4 At the termina(cid:173)
`tion of the assay, the cells are fixed in situ,
`washed, and dried. Sulforhodamine B (SRB), a
`pink anionic dye that binds to the basic amino
`acids of trichloroacetic acid (TCA )-fixed cells, is
`added. The cells are washed again, and the
`remaining dye is a function of the adherent cell
`mass. The bound stain is solubilized and mea(cid:173)
`sured spectrophotometrically. The data are rou(cid:173)
`tinely entered from automatic reading devices
`into the computer for subsequent storage and
`analysis.
`Exhaustive measures were undertaken to in(cid:173)
`sure highly reproducible assays for identifying
`those chemicals or natural product extracts
`capable of producing either tumor cell cytotoxic(cid:173)
`ity or growth inhibition. II> Multiple parameters
`(eg, clear definition of assay end points, assay
`conditions, including specific media require(cid:173)
`ments, reporting devices for results, and meth(cid:173)
`ods for data analysis) had to be optimized
`before implementation of the new cancer screen.
`The NCI cancer screen has been fully opera(cid:173)
`tional since April 1990. Since that time, over
`27,000 defined chemical entities and thousands
`of crude natural product extracts have been
`screened. The agents that have been shown to
`have antitumor activity represent a diverse ar(cid:173)
`ray of chemical structures, and many represent
`chemical classes that have not had extensive
`prior clinical evaluation. Approximately 4% of
`the agents screened have been referred for
`further testing based on the in vitro screening
`data. These agents (about l,000) have been
`prioritized and are currently undergoing second(cid:173)
`ary in vitro and in vivo evaluation.
`
`PRESENTATION AND ANALYSIS OF THE
`CANCER SCREENING DATA
`The reproducibility of the biological data
`from the cancer screen has been quite encourag(cid:173)
`ing. A standard dose-response curve for the
`individual agents is provided for each of the 60
`cell lines. In ad