Professional Documents
Culture Documents
PharmTech NA May2020 US WM
PharmTech NA May2020 US WM
PharmTech NA May2020 US WM
PharmTech.com
Technology Advances
Streamline Bioprocessing
Development
Solving ADC Challenges
Reformulation Strategies
Manufacturing
Microneedle Array Patches
Contamination Control Programs
Analytics
Toxicology Studies
Operations
Limits of Light Obscuration
Outsourcing
Smooth Bioprocessing Tech Transfer
Peer-Review Research
Calculating Passing Probabilities for
Content Uniformity and Dissolution Tests
EDITORIAL SALES
Editorial Director Rita Peters RPeters@mjhlifesciences.com Publisher Mike Tracey MTracey@mjhlifesciences.com
Senior Editor Agnes Shanley AShanley@mjhlifesciences.com East Coast Sales Manager Joel Kern JKern@mjhlifesciences.com
Managing Editor Susan Haigney SHaigney@mjhlifesciences.com Mid West, West Coast Sales Manager BJ Ghiglione BGhiglione@mjhlifesciences.com
European Editor Felicity Thomas FThomas@mjhlifesciences.com European Sales Manager Linda Hewitt LHewitt@mjhlifesciences.com
Manufacturing Editor Jennifer Markarian JMarkarian@mjhlifesciences.com European Senior Sales Executive Stephen Cleland SCleland@mjhlifesciences.com
Science Editor Feliza Mirasol FMirasol@mjhlifesciences.com C.A.S.T. Data and List Information Michael Kushner MKushner@mjhlifesciences.com
Assistant Editor Lauren Lavelle LLavelle@mjhlifesciences.com VP/Managing Director, Pharm/Science Group Dave Esola
Senior Art Director Marie Maresco; Graphic Designer Maria Reyes
Contributing Editors Jill Wechsler jillwechsler7@gmail.com; ADDRESS
Hallie Forcinio editorhal@cs.com; 485 Route One South, Building F, Second Floor, Iselin, NJ 08830, USA
Susan J. Schniepp sue.schniepp@mac.com; Eric Langer info@bioplanassociates.com; Tel. 732.596.0276, Fax 732.647.1235
and Cynthia A. Challener, PhD challener@vtlink.net
PharmTech.com
Correspondent Sean Milmo (Europe, smilmo@btconnect.com)
485 Route One South, Building F, Second Floor, Iselin, NJ 08830, USA Permissions Alexa Rockenstein, arockenstein@mjhlifesciences.com
Tel. 732.596.0276, Fax 732.647.1235, PharmTech.com Audience Development Research Director
EDITORIAL ADVISORY BOARD Christine Shappell cshappell@mjhlifesciences.com
Pharmaceutical Technology publishes contributed technical articles that undergo a
rigorous, double-blind peer-review process involving members of our distinguished MJH LIFE SCIENCESTM
Editorial Advisory Board. Manuscripts should be sent directly to the managing editor. Below is a partial list of the
Pharmaceutical Technology brand editorial advisory members. The full board, which includes advisory members Chairman and Founder Mike Hennessy, Sr
from Pharmaceutical Technology Europe, can be found online at PharmTech.com. Vice Chairman Jack Lepping
President and CEO Mike Hennessy, Jr
James P. Agalloco Sanjay Garg, PhD Moheb M. Nasr, PhD Chief Financial Officer Neil Glasser, CPA/CFE
President Professor and Director, Principal
Agalloco & Associates Centre for Pharmaceutical Nasr Pharma Regulatory Consulting Executive Vice President, Operations Tom Tolvé
Innovation and Development,
Larry L. Augsburger, PhD University of South Australia Garnet E. Peck, PhD Executive Vice President, Global Medical Affairs and Corporate Development
Professor Emeritus Professor Emeritus of Industrial
University of Maryland R. Gary Hollenbeck, PhD Pharmacy, Purdue University Joe Petroziello
Research Faculty Senior Vice President, Content Silas Inman
David H. Bergstrom, PhD University of Maryland Wendy Saffell-Clemmer
Senior Vice-President, School of Pharmacy Director, Research Senior Vice President, I.T. & Enterprise Systems John Moricone
Pharmaceutical Development & Baxter Healthcare
Corporate Quality Assurance Ruey-ching (Richard) Hwang, PhD
Gurvinder Singh Rekhi, PhD
Senior Vice President, Audience Generation & Product Fulfillment Joy Puzzo
Antares Pharma, Inc. Senior Director,
Pharmaceutical Sciences, Department of Pharmaceutical and Vice President, Human Resources & Administration Shari Lundenberg
Phil Borman, DSc Biomedical Sciences,
Director
Pfizer Global R&D
The University of Georgia College Vice President, Business Intelligence Chris Hennessy
Product Development & Supply Maik W. Jornitz of Pharmacy Executive Creative Director, Creative Services Jeff Brown
Medicinal Science & Technology President
Pharma R&D Susan. J. Schniepp
G-CON Manufacturing Inc. Executive Vice-President of
GlaxoSmithKline
Mansoor A. Khan, PhD Post-approval Pharmaceuticals and
Evonne Brennan Professor & Vice Dean Distinguished Fellow
International Technical Marketing Irma Lerma Rangel College of Regulatory Compliance Associates
Manager, Pharmaceutical Division, Pharmacy, Texas A&M Health David R. Schoneker
IMCD Ireland Science Center Director of Global Regulatory Affairs,
Rory Budihandojo Russell E. Madsen Colorcon
Lachman Consultants President, Aloka Srinivasan
The Williamsburg Group, LLC VP, Regulatory
Metin Çelik, PhD
President, Heidi M. Mansour, PhD Lachman Consultants
Pharmaceutical Technologies Assistant Professor Read board members’
International (PTI) College of Pharmacy biographies online at
& The BIO5 Research Institute, PharmTech.com/ © 2020 MultiMedia Pharma Sciences LLC All rights reserved. No part of this publication may be reproduced or
Zak T. Chowhan, PhD University of Arizona–Tucson transmitted in any form or by any means, electronic or mechanical including by photocopy, recording, or information
Consultant, Pharmaceutical pharmtech-editorial-
Development Jim Miller advisory-board. storage and retrieval without permission in writing from the publisher. Authorization to photocopy items for internal/
Founder and Former President, educational or personal use, or the internal/educational or personal use of specific clients is granted by MultiMedia
Suggy S. Chrai, PhD PharmSource, A Global Data Pharma Sciences LLC for libraries and other users registered with the Copyright Clearance Center, 222 Rosewood Dr.
President and CEO, Company
Chrai Associates, Inc. Danvers, MA 01923, 978-750-8400 fax 978-646-8700 or visit http://www.copyright.com online. For uses beyond
Colin Minchom, PhD those listed above, please direct your written request to Permission Dept. or email: ARockenstein@mjhlifesciences.com
Roger Dabbah, PhD Senior Director Pharmaceutical
Principal Consultant, Sciences, Shire Pharmaceuticals MultiMedia Pharma Sciences LLC provides certain customer contact data (such as customers name, addresses,
Tri-Intersect Solutions phone numbers, and e-mail addresses) to third parties who wish to promote relevant products, services, and other
R. Christian Moreton, PhD
Robert Dream Partner, Finnbrit Consulting
opportunities that may be of interest to you. If you do not want MultiMedia Pharma Sciences LLC to make your contact
Managing Director information available to third parties for marketing purposes, simply email mmhinfo@mmhgroup.com and a customer
HDR Company Fernando J. Muzzio, PhD service representative will assist you in removing your name from MultiMedia Pharma Sciences LLC lists.
Director, NSF Engineering
Tim Freeman Research Center on Structured Pharmaceutical Technology does not verify any claims or other information appearing in any of the advertisements con-
Managing Director, Organic Particulate Systems,
FreemanTechnology tained in the publication, and cannot take responsibility for any losses or other damages incurred by readers in reliance
Dept. of Chemical and Biochemical of such content. Pharmaceutical Technology welcomes unsolicited articles, manuscripts, photographs, illustrations, and
Engineering, Rutgers University
other materials but cannot be held responsible for their safekeeping or return.
Reprints: Contact Mike Tracey, mtracey@mjhlifesciences.com. Display, Web, Classified, and Recruitment Advertising:
Contact, tel. 732.346.3027. Permissions: Contact Alexa Rockenstein, tel. 732.346.3007, ARockenstein@mjhlifesciences.com
Pharmaceutical Technology’s eNewsletter Team:
To subscribe: mmhinfo@mmhgroup.com
• ePT, Editor Lauren Lavelle, ptpress@mmhgroup.com
• In the Lab, Editors Lauren Lavelle and Feliza Mirasol, ptpress@mmhgroup.com
• Equipment & Processing Report, Editor Jennifer Markarian, JMarkarian@mjhlifesciences.com
• Send news and product releases to ptpress@mmhgroup.com
STERILE.COM
For more information, visit our website at sterile.com/particlecounters VELTEK ASSOCIATES, INC.
May 2020 Volume 44 Number 5
COVER STORY
16 Technology Advances
Streamline Bioprocessing
Bioprocessing advances improve product yield, cut costs, and
streamline integration between upstream and downstream processes.
Cover Design by Maria Reyes
Images: kras99 - Stock.adobe.com
FEATURES
DEVELOPMENT
33 How a Contamination OPERATIONS
Continued on page 8
PEER-REVIEWED RESEARCH
PEER-REVIEW RESEARCH
CAPTISOL.com
Continued from page 6
Pharmaceutical Technology is selectively abstracted or indexed in: PHARMACEUTICAL TECHNOLOGY (Print ISSN: 1543-2521,
»»Biotechnology and Bioengineering Database Clarke Drive, Suite 100, Cranbury, NJ 08512. Subscription
(Cambridge Scientific Abstracts)
rates: US and possessions — 1 year (13 issues), $79.80; 2
»»Business and Management Practices (RDSI)
years (26 issues), $139.65. Canada and Mexico — 1 year,
»»Chemical Abstracts (CAS) $103.95; 2 years, $158.55. All other countries 1 year,
»»Current Packaging Abstracts $152.25; 2 years, $276.15. International price includes air-
»»Science Citation Index (Thomson) Return Undeliverable Canadian Addresses to: IMEX
Pharmaceutical Technology is proud to be a member of IPEC and PDA. Printed in the U.S.A.
How Fast
Is Too Fast?
Rita Peters
T
he rapid spread of the coronavirus joined the National Institutes of Health In April 2020, AstraZeneca and the
is a reminder of the intercon- (NIH), Foundation for the NIH, Health University of Oxford announced a devel-
nected nature of global popu- and Human Services Office of the As- opment and manufacturing partnership.
lations and economies. A few other sistant Secretary for Preparedness and Johnson & Johnson announced manu-
realities also should be obvious. First, Response, FDA, Centers for Disease facturing agreements with Catalent and
no single individual, company, or coun- Control and Prevention, and the Eu- Emergent BioSolutions for its vaccine
try can solve the pandemic crisis alone. ropean Medicines Agency in an effort candidate, as did Lonza with Moderna
Second, while getting a vaccine to pa- to advance the most promising vaccine for its mRNA vaccine. Both Johnson
tients quickly is crucial, decisions about and therapeutic candidates (1). & Johnson and Moderna have received
quality and efficacy should be based on funding from the Biomedical Advanced
science, not political or public pressure. The need for Research and Development Authority.
The competition for vital medical In late April, a Trump Administra-
resources in the early days of the pan- a coordinated tion plan—dubbed “Operation Warp
demic illustrated how uncontrolled or Speed”—committed to make 300 mil-
misguided competition and anxiety manufacturing lion doses of coronavirus vaccine avail-
can quickly overwhelm response to an
emergency. Fortunately, collaborative strategy cannot able by the end of 2020 (2). Details for
this program—including which vac-
efforts are shaping the development of
treatments and vaccines. For example,
be overlooked. cines were going to be manufactured—
were not available at press time.
the World Health Organization (WHO) While groups are collaborating on After weeks of stay-at-home orders, so-
is coordinating global efforts for vac- development and clinical trial phases; cial distancing, closed restaurants, high
cine development and clinical trials for the need for a coordinated manufactur- unemployment numbers, and bad eco-
treatment options. ing strategy cannot be overlooked. nomic news, optimism about the early
The Accelerating COVID-19 Ther- For an industry used to development arrival of a vaccine, as well as treatments
apeutic Interventions and Vaccines timelines counted in years, the COVID- such as Gilead Sciences’ remdesivir, offers
(ACTIV) partnership is a public-pri- 19 vaccine and therapy development some good news and excitement.
vate effort to develop a framework to process is moving very fast. Sponsors With anxiety for a resolution to the
prioritize vaccine and drug candidates, of the leading vaccine candidates now pandemic driving vaccine development
streamline clinical trials, coordinate in early clinical trials are expressing activity, it is crucial that science-based in-
regulatory processes, and leverage as- confidence that the product can reach formation—from the global bio/pharma
KARRAMBA PRODUCTION - STOCK.ADOBE.COM
sets to respond to the COVID-19 and patients by the end of 2020, months or experts—should drive development, ap-
future pandemics. More than a dozen years earlier than predicted by experts proval, and manufacturing decisions.
biopharmaceutical companies have just one week ago.
To achieve that goal, manufacturing References
capacity for unproven vaccines must 1. NIH, “NIH to Launch Public-Private
be created now, a risky prospect for the Partnership to Speed COVID-19 Vaccine
and Treatment Options,” Press Release,
companies developing the capacity and April 17, 2020.
Rita Peters is editorial for patients. Drug companies and con- 2. J. Jacobs and D. Armstrong, “Trump’s
director of Pharmaceutical
Technology. Send your
tract manufacturers are accepting the ‘Operation Warp Speed’ Aims to Rush
thoughts and story ideas to challenge; governments and taxpayers Coronavirus Vaccine,” Bloomberg.com,
rpeters@mjhlifesciences.com. may bear the financial risk. April 29, 2020. PT
2020 PDA
Data Integrity
Workshop
Data Governance Programs: Fit for Innovation
Learn from leading technology experts and global regulators how you can pivot your company’s data
governance program from reactive to proactive. This Workshop is designed to provide data integrity
newcomers and professionals alike with the practical knowledge needed to develop a modern, agile
data governance program.
Don’t miss out on early bird registration for this Workshop – Register before July 31 to save up to $200!
www.cordenpharma.com
www.cordenpharma.com
regulatory watch
C
oncerns about access to medi- the drug, leading to notable shortages at instructed manufacturers to send FDA
cines and diagnostics critical to some manufacturers. information on where interruptions in
containing the coronavirus pan- These concerns have spurred calls for supply affect APIs as well as drugs and
demic (COVID-19) and treating in- greater US investment in high-tech bio- extended such requirements to medical
fected patients have broadened support pharma production systems. In announc- device makers during this public health
for advanced pharmaceutical manufac- ing the COVID-19 Therapeutics Accelera- emergency. To further address these issues,
turing and test methods. FDA officials tor in early March to advance treatments Congress provided $1.5 million for the Na-
have long pressed industry to adopt for the pandemic, the Bill and Melinda tional Academies of Sciences, Medicine,
continuous manufacturing and online Gates Foundation cited the need to quickly and Engineering (NASEM) to prepare a
testing methods able to scale up quality build up manufacturing capacity to test a study on ways to strengthen the manufac-
production quickly and efficiently. The variety of drugs. Gates further empha- turing supply chain for drugs and devices
current pandemic has boosted support sized that the federal government should to avoid shortages.
for such initiatives, as global health provide suppport for building production
organizations and US policy makers facilities now, as making such investments Millions for manufacturing
recognize the importance of being able before knowing that a product will be used Added support for establishing advanced
to produce millions of doses of any is excessively risky for manufacturers. biopharma manufacturing facilities is
promising new treatments. Funds to advance efficient drug man- found in sections of the pandemic relief
While the lack of personal protective ufacturing systems to ensure access to legislation, which provides some $30 bil-
gear and respirators has dominated the vital therapies and prevent shortages due lion for federal health agencies to develop
headlines since the pandemic outbreak, to supply disruptions from China and countermeasures and vaccines, plus plat-
potential difficulties in ensuring suffi- elsewhere were included in the initial form technologies to advance US pro-
cient supplies of newly discovered thera- $8.3-billion emergency coronavirus fund- duction of new therapeutics, diagnostics,
pies and anticipated vaccines have moved ing package enacted March 5, 2020. FDA vaccines, and medical supplies. Congress
to center stage. China’s prominence in gained $61 million to support the devel- specified that a portion of $3.5 billion for
producing key APIs for antibiotics and opment of new medical countermeasures the Biomedical Advanced Research and
many common medicines, plus a recent and vaccines, of advanced manufacturing Development Authority (BARDA) should
move by the Indian government to halt for medical products, and for monitoring be used to construct or renovate US-based
the export of multiple drugs and drug in- medical product supply chains (1). The next-generation manufacturing facilities.
gredients, heightened concerns about the subsequent $2-trillion coronavirus aid and Similar directions are included in provid-
United States’ dependence on global sup- relief package approved March 27, 2020 ing added funds for the Defense Research
ply chains. Rumors that an existing ma- similarly provided an added $80 million Advanced Projects Agency (DARPA) in
laria treatment might be effective against to FDA to advance the development and the Department of Defense, building on
W.SCOTT MCGILL - STOCK.ADOBE.COM
COVID-19, for example, created a run on approval of medical countermeasures and its research programs related to advancing
vaccines (2). The legislation also directed biopharma manufacture. Among other
FDA to use some of the funds to further things, these initiatives have supported the
the adoption of advanced manufacturing development of RNA and DNA vaccines
systems for medical products and to mon- to fight infectious diseases such as Chiku-
Jill Wechsler is itor supply chains for potential threats to ngunya and Ebola and to overcome manu-
Pharmaceutical access to medicines and APIs imported facturing challenges to faster scale-up.
Technology’s
Washington editor, from abroad. To help FDA anticipate The pandemic legislation also provides
jillwechsler7@gmail.com. looming supply problems, Congress $27 billion for the Public Health & Social
14 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
Regulatory Watch
Services Emergency Fund managed by more efficient and reliable ways to produce vaccine globally (5). Such fast expansion
the Secretary of Health and Human Ser- high quality cell and gene therapies. of production capacity will be needed for
vices (HHS) to develop countermeasures These and other projects stand to assist clinical trials slated to begin this fall and
and vaccines in response to the pandemic. manufacturers on accelerated timelines then to provide emergency use access to
This includes building the Rapid Aseptic for testing promising COVID-19 therapies any promising product.
Packaging of Injectable Drugs (RAPID) and vaccines looking to establish systems
consortium with a network of up to eight for fast, reliable manufacturing scale-up References
domestic facilities to rapidly fill and finish capabilities. At a “virtual summit” in 1. H.R. 6074, 116th Congress, Public Law
millions of prefilled syringes for delivering March sponsored by the Biotechnology 116–123—Mar. 6, 2020, Coronavirus Pre-
paredness and Response Supplemental
vaccines and therapies for COVID-19 (3). Innovation Organization (BIO), industry Appropriations Act, 2020.
An innovative syringe developed by Api- leaders cited the challenge in needing to 2. Amendment H.R. 748, “An Act to amend
Ject Systems utilizes existing blow-fill-seal expand manufacturing capabilities before the Internal Revenue Code of 1986 to Re-
technology plus an interlocking needle knowing they have a viable product, and peal the Excise Tax on High Cost Employ-
hub to provide low-cost, easy-to-use inject- the fast launch of clinical trials for candi- er-Sponsored Health Coverage,” US Senate,
March 21, 2020.
ables for the Strategic National Stockpile. date vaccines aggravates those difficulties. 3. HHS, “HHS Announces New Public-Pri-
Similarly, additional funding for the On many fronts, industry is rising to vate Partnership to Develop US-Based,
National Institute of Standards and Tech- the challenge, with pharma companies High-Speed Emergency Drug Packaging
nology (NIST) supports programs to partnering with smaller biotechs and fed- Solutions,” Press Release, March 18, 2020.
accelerate production of critical materi- eral agencies that offer innovative drug 4. NIST, “NIST Funding Manufacturing In-
stitutes to Support Pandemic Response,”
als, build additional production facilities, and vaccine candidates for established Press Release, March 31, 2020.
ensure supply chains for vital ingredients, firms to test and produce. In March, 5. J&J, “Johnson & Johnson Announces a
develop and train manufacturing workers, Johnson & Johnson announced that a Lead Vaccine Candidate for COVID-19;
and return to the US the manufacture of $1-billion partnership of its Janssen unit Landmark New Partnership with US De-
critical conventional drugs (4). NIST has with BARDA planned to rapidly scale partment of Health & Human Services; and
Commitment to Supply One Billion Vac-
worked with biotech firms for several up vaccine manufacturing capacity to be cines Worldwide for Emergency Pandemic
years to address challenges in developing able to supply over one billion doses of Use,” Press Release, March 30, 2020. PT
VISIT US AT
l Small Volume Support
CPhI North America
l Critical Quality Attributes Philadelphia, PA, Sept 9-11,
Booth #2225
l GMP Synthesis & Purification
Technology Advances
Streamline Bioprocessing
Feliza Mirasol
O
ver the years, many advances ity and control of critical product bioreactor needed, directly influencing
have been made in bioprocess- quality attributes (CPQAs) via the overall footprint of operations, cap-
ing as biomanufacturers strive novel cell expression and inducer ital expenditure (CAPEX), and, subse-
to increase yield, improve product technologies with greater molec- quently, cost of goods.”
recovery, enhance product purity, and ular and cellular biology under- Furthermore, Medvedev says the use
streamline manufacturing. Innova- standing through systems biology of design of experiment (DoE) tech-
tions in technology and equipment • Accelerating the overall process niques is a promising trend in the in-
for both upstream and downstream development timeline. dustry, when these principles are cor-
processing have led to more integrated “These improvements have resulted rectly applied to process development
and efficient processes, but there still in the ability to quickly advance from- and scale-up studies. “In order to fully
remain manufacturing challenges that cell-line generation to clinical current- benefit from the use of DoE studies,
drive the need for further innovation. good manufacturing practices (cGMP) representative scale-down models are
manufacturing with bioprocesses that paramount to ensure identification of
Upstream advancements are more productive and reproducible,” high-performing small-scale process
Successful upstream bioprocessing in- observes Brian Follstad, director, up- design with seamless transfer to com-
novations have focused on key areas, stream process development, Catalent mercial manufacturing scale, reducing
KRAS99 - STOCK.ADOBE.COM
A B
C D
Visit highshearmixers.com/ultra
Or call 1-800-243-ROSS
Formulating an ADC
to be much more difficult than many
researchers had expected at the outset
of ADC research, adds Justin Sweeley,
Development Solution
senior technology manager, biologics
at Novasep. “It was initially under-
stood that finding a good target an-
tigen for an ADC-based compound
would require a target with rapid in-
Cynthia A. Challener
ternalization once it had reached the
cell,” he notes. “The result,” he says,
“was that in many cases targets that
Many antibody-drug conjugate therapies were ineffective for monoclonal anti-
are in the pipeline; however, only a handful body (mAb) therapies because of rapid
internalization became perfect targets
have been approved. What are the bottlenecks? for ADC therapies.”
This picture became much more
T
he concept of bioconjugation has Takeda); (ado-) trastuzumab emtan- complex upon the addition of payloads
been recognized for many decades, sine (Kadcyla, Genentech/Roche); to these mAb candidates, however.
as has the specific idea of anti- inotuzumab ozogamicin (Besponsa, This issue can be seen most clearly in
body-drug conjugates (ADCs). Technol- Pfizer); polatuzumab vedotin-piiq the fact that the clinical trial landscape
ogy for the production of ADCs did not (Polivy, Genentech/Roche); Fam-tras- includes well over 50 different pay-
advance sufficiently until the late 20th tuzumab deruxtecan-nxki (Enhertu, load candidates, but only seven have
century. Even then, the first ADC ap- Daiichi Sankyo and AstraZeneca); and successfully made it into commercial
proved by FDA—gemtuzumab ozogami- Enfortumab vedotin (Padcev, Astellas production, according to Sweeley.
cin (Mylotarg) from Pfizer in 2000—was Pharma/Seattle Genetics), the latter of “As a result, there has been a recent
voluntarily withdrawn in mid-2010 due which received approval in February shift in the industry to recognize that
to its association with the serious liver 2020. And, there are approximately 80 when trying to predict ADC effective-
condition called veno-occlusive disease. ADC candidates undergoing nearly 600 ness, researchers must take into ac-
Gemtuzumab ozogamicin was ulti- clinical trials (1). count the target epitope, the physical
mately re-approved in 2017 at a lower Why have more not already been ap- attributes of the payload on the mAb
recommended dose and schedule in proved? The issue of toxicity has been once conjugated, and also the effect of
VPANTEON - STOCK.ADOBE.COM
a different patient population. There a challenge, as has the complexity of the payload mechanism of action on
are now six additional FDA-approved ADCs and the consequent manufac- the specific kind of cancer being tar-
ADCs on the market: brentuximab turing challenges. geted,” Sweeley observes.
vedotin (Adcetris, Seattle Genetics/ AstraZeneca takes an empirical ap-
Cynthia A. Challener, PhD, is a Crafting the right molecule proach to target and payload selection,
contributing editor to Pharmaceutical Finding the right target, protein, linker, keeping the patient population top-of-
Technology. and payload all belong to the discovery mind, using data from clinical trial re-
22 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
CREATING TOMORROW’S SOLUTIONS
WACKER BIOTECH –
THE MICROBIAL CDMO
Your All-In-One Partner
Process
Development
Contract
Manufacturing
Wacker Biotech is your all-in-one partner for manufacturing biologics, vaccines and Live Microbial Products (LMPs). From
strain and process development through to technology transfer and manufacturing in our three EMA-/FDA-approved GMP
facilities, we are able to cover the whole process for pre-clinical, clinical and commercial supply. On top of that, our unique
microbial technologies help you to manufacture your protein extremely cost-effectively.
You can put your trust in our experience!
Wacker Biotech GmbH, Jena and Halle (Germany), Wacker Biotech B.V., Amsterdam (the Netherlands)
Phone: +49 3641 5348-0, Phone: +31 20 750 3600, info.biologics@wacker.com, www.wacker.com/biologics
Development
sults and non-clinical data to help vali- testing methods doesn’t directly result An example pointed out by Swee-
date target and payload selection. “With in better outcomes for patients. ley is the use of non-natural amino
potency being driven by the strength of acid-based click-chemistry (Ambrx
the warhead, we optimize the therapeu- Linker technology matters and Sutro). “To my knowledge, these
tic index of our molecules by selecting Finding an ADC that is better than technologies have clearly simplified
the most favorable drug-to-antibody standard therapy or that provides a the conjugation process, but have not
ratio (DAR) in designing our ADCs,” solution to an unmet need definitely shown a clear improvement in product
Coats says. “We believe,” he adds, “that represents a challenge, just as with effectiveness in the clinic,” he states.
establishing target rationale and strat- other drugs based on different ap-
egy early on are key to determining the proaches, according to Bertholjotti. Undesired immune responses
right payload.” The linker chemistry is an important ADCs are designed to be more selective
component of ADC and has a signifi- than traditional chemotherapy agents.
Is site-specific cant impact on performance. The mAb enables targeting of cancer
conjugation important? “Linker chemistry plays a critical role (or other disease) cells where the pay-
Site-specific conjugation is looked at as in in-vivo stability, but initially was as- load is delivered with high selectivity,
the second generation of ADC conjuga- sumed to have a passive role with the thus reducing the systemic toxicity in
tion techniques. “The homogeneity of exception of either being stable or la- comparison with standard chemother-
these molecules allows for much tighter bile in the acidic cytosolic cellular en- apeutic drugs.
control at both the manufacturing and vironment,” observes Sweeley. More Unfortunately, undesired immune
characterization levels,” Sweeley ex- recently, however, he notes that there responses have presented a problem
plains. Better definition for an ADC has been growing recognition that the that has been difficult to overcome.
means that its therapeutic index can linker plays an active role in conjugate These responses are largely due to
be further improved, adds Bertholjotti. hydrophobicity and therefore stability the massive number of variables
Additionally, the technology allow- of the ADC as a whole. being examined and the relatively
ing for site-specific conjugation has small number of clinical trials going
evolved dramatically from the Thi- Linker chemistry on, according to Sweeley. “Many of
omab concept piloted by Genentech to t he warheads being investigated
current techniques allowing site-spe- is an important for ADCs induce immunogenic cell
cific conjugation with any native mAb, death, which may enhance anti-tu-
such as Synaffix’s glycogen modifica- component of mor immunity,” adds Coats. He also
tion techniques or Ajinomoto’s AJI-cap observes that there are recent exam-
technology, Sweeley remarks. Many of ADC and has a ples demonstrating clinical activity
the ADCs in preclinical and clinical with a combination of an ADC with
stages are based on site-specific con- significant impact PD-1 inhibition.
jugation technology of one form or “In the process of taking any ADC
another, Bertholjotti adds. on performance. through the approval process, it is
On the other hand, Sweeley notes necessary to look at the mAb, the
that even though more site-selective T he tech nolog y, accord i ng to linker, and the payload chemistries
ADCs are expected in the future, ADCs Coats, has advanced and matured individually, the ADC as a whole,
with stochastic conjugation can still be during the past 20 years to a point and the payload and linker residues
successful. Coats agrees that site-se- where linker technology is stable and after the ADC has been internal-
lective conjugation does not seem to molecules in development demon- ized and digested within the cell.
demonstrate significant differences strate low levels of deconjugation in The end result is that for each ADC
in terms of clinical activity and safety patients. Both protease cleavable and trial, researchers must monitor all
when compared to classic non-site-se- non-cleavable linkers are in develop- of the normal undesired immune
lective conjugation. In fact, the seven ment and on the market. responses that occur during any on-
commercially approved ADCs are not Classic conjugation chemistr y cology trial, but then try to attribute
based on site-specific conjugation. approaches include maleimide and the cause of any that are observed
It is possible, according to Sweeley, N-hydroxysuccinamide-ester moi- to one of the five different potential
that the lack of a commercially avail- eties (Seattle Genetics, Roche, and sources,” Sweeley explains.
able site-specific conjugated molecule Pfizer). There are also various newer “Factoring these additional consider-
is just a result of the head start that sto- technologies for linkers, with some, ations into the normal variables of any
chastic conjugation has had, but it is Bertholjotti comments, at the proof- trial, such as the specific cancer type,
also possible that simplicity in manu- of-concept stage and others already in patient population, and level of pre-
facturing and homogeneity in analytic preclinical evaluation. treatments, etc., the picture becomes
24 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
Development
incredibly complex. Even after 20 years of clinical testing, manufacturing with the added complexity of highly potent
the issue is still not fully understood,” he concludes. drug manufacturing. “All three characteristics are complex
on their own, and combining them brings the complexity to
Lower-than-expected therapeutic window a level where it is no surprise that the vast majority of ADCs
The primary issue affecting ADC approvals today, according are manufactured by outsourcing partners,” Sweeley says.
to Sweeley, is the lower-than-expected therapeutic window For the most part, he notes that innovator companies treat
of these therapies. “If the therapeutic window of ADCs were the manufacturing of an ADC as three separate processes:
as large as people expected when ADCs first came to light, • mAb manufacturing using traditional mAb techniques
then many more approvals would already be on the mar- • Payload synthesis using traditional highly potent syn-
ket,” he asserts. This window has proven to be much more thesis techniques
complicated than originally hoped, however, and therefore • ADC conjugation using specialty contract manufactur-
the clinical impact less significant than originally imagined. ing organizations capable of both working with both
Here again, the complexity of ADCs themselves is the main biologics and highly potent payloads.
problem. For instance, Sweeley notes that if an innovator “Taken individually, all three of these steps are actually
company wants to test a new conjugation method, they might defined quite well and have been orchestrated successfully
choose a mAb and payload pair that have already shown suc- for a long time. But because ADC manufacturing requires
cess in the clinic. Similarly, a mAb company with a new anti- all three to happen together, any delays or issues in one pro-
gen target is likely to use an established conjugation approach cess will necessarily affect the other two. The reality, there-
(e.g., that for Adcetris) to lower their risk. “The best molecule, fore, is that ADC supply-chain management is one of the
however,” Sweeley explains, “might be a combination of mul- most complex processes in the pharmaceutical industry and
tiple new technologies that are too risky to investigate for a must be managed by an experienced team who can mitigate
small company with a limited budget.” risks whenever possible,” observes Sweeley.
Bertholjotti agrees. The supply chain for ADCs is com-
Complex supply chain plex and requires specialized companies to manage different
The supply chain for ADCs is highly complex as well. ADCs steps in a safe way and with the necessary quality. “If the
are unique in that their manufacture is an amalgamation of supply chain is not properly managed, delays and supply is-
classical small-molecule production and traditional mAb sues may arise. Therefore, compromises related to the supply
chain can result in critical impacts on timelines and costs
to bring a drug to market,” he asserts.
Benefits of Reformulation
that can generate outcomes, which are
difficult for prospective generic rivals to
replicate,” he states.
Reformulation strategies
Felicity Thomas There are numerous reformulation strat-
egies that are currently employed by the
bio/pharma industry, with each offering
a different way in which the patient expe-
rience can be improved from the original
Reformulation strategies can provide drug formulation. Examples of reformulation
developers with a head start to achieve approaches include the following.
Modifying the original release profile. Most
promising options that benefit the patient. reformulation strategies that are pursued
by developers of oral and parenteral dos-
A
recent study has estimated the seek initial approval for products with a age forms are based upon the modifica-
mean cost of developing a new simple lead formulation followed by the tion of the original release profile, notes
drug and bringing it to market is introduction of superior formulations Spencer. “For oral solid dosage forms,
$1.3 billion (1). Given this high invest- that can expand the patient population functional excipients can be used to refor-
ment value, the fact that drug develop- or enhance safety and efficacy.” mulate a product into dosage forms, such
ment is fraught with potential failure and as multiparticulates and matrix tablets, to
the increasing complexities of developing Getting a head start improve product performance,” he says.
difficult-to-handle novel chemical enti- “Reformulating an existing drug gives Additionally, the drug release profile
ties, reformulation strategies can provide developers a head start, rather than can be modified by using formulation
developers with promising options. developing a whole new drug from technologies, which can help improve
“Due to the high costs of drug develop- scratch,” says Jeremy Drummond, senior targetability and oral bioavailability. An
ment, as well as the high rate of clinical vice-president of Business Development, example of a strategy to improve bio-
failures, it’s vital that pharmaceutical MedPharm. “This approach cuts product availability is through the use of amor-
companies evaluate formulation oppor- development times because studies can phous solid dispersions (ASDs), adds
COLIN FEARING - STOCK.ADOBE.COM
tunities for every approved or late-stage often bridge to those of previous regu- Deanna Mudie, principal scientist, Bend
clinical product to determine how its latory submissions in particular with Research, Lonza.
commercial lifecycle can be optimized,” regard to non-clinical data.” “Enhancing the bioavailability can re-
explains Paul Spencer, head of Pharma- “One of the main benefits of reformu- duce the dose and remove the food label
ceutical Polymers and Services, Evonik lating drug products is increasing patient requirement that calls for patients to
Health Care. “To minimize regulatory adherence to medicines,” adds Henny strictly take the product with or without
risk and accelerate speed to market, it Zijlstra, director, Commercial Develop- food, which can improve patient safety,”
is common for pharma companies to ment, Lonza. Pointing to research from explains Mudie. “Food labels may pres-
28 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
ent challenges to patients with trouble colide polymers can help to reduce admin- According to Spencer, ocular drug
swallowing. Additionally, the food label istration frequency or enable localized de- delivery is currently experiencing in-
sometimes creates further challenges for livery of drugs, reveals Spencer. “Lactide/ tense focus in terms of reformulation.
patients who have to take medication glycolide polymers have attained decades “In ocular drug delivery, it is a therapeu-
without food, causing them to skip meals of literature in precisely controlling the tic priority to minimize the number of
or eat at inconvenient times.” rate of drug release from microparticles required intravitreal injections per eye,”
Other examples of reformulations of over weeks, months, or a year or more he explains. “Here, pharmaceutical
oral dosage forms that have been shown following a single administration,” he says. companies are developing drug-loaded
to improve patient adherence include “Drug developers may also reformulate implants that can safely resorb via hydro-
minitablets or orally disintegrating tab- drug products as controlled-release ver- lysis after months of drug delivery.”
lets (ODTs) that offer improved taste sions,” adds Zijlstra. “This method can Combination products. Comprised of two
sensation or swallowability for pediatric assist in reducing the dosing frequency, or more components in a single entity,
populations, confirms Spencer. thus reducing the pill burden on patients.” combination products can enable devel-
Changing the route of administration. Ad-
ministering drug products via an alter-
native route is another reformulation
approach that can overcome limitations
of the original form. “For example, con-
verting from oral to inhalation delivery
may allow developers to mitigate side ef-
fects associated with the drug interact-
ing with the gastrointestinal (GI) tract,”
says Mudie.
“For some small molecules and pep-
tides, it is possible to change the route
of administration from oral tablets to
injectable microparticles and implants,”
adds Spencer. “There are a range of suc-
cessful reformulation examples in this
regard, such as a shift from orals to ex-
tended release, injectable dosage forms to
enhance the treatment of schizophrenia,
opioid, and alcohol addition, as well as
contraception therapies.”
However, careful consideration is
required when reformulating oral dose
products as topicals, cautions Drum-
mond. “In this circumstance, the route
of delivery is significantly different,
and formulation composition can have
a major impact on efficacy as well as
chemical and physical stability,” he says.
In-vitro models that are based on rele-
vant human tissues to provide results that
are directly relatable to clinical outcomes
and can also provide answers to specific
research questions can help to optimize
reformulations, asserts Drummond.
“Using these models often leads to novel
discoveries that can extend existing pat-
ents, making reformulation an attractive
option for drug companies,” he states.
Reducing dose frequency. Injectable mi-
croparticles that use the biocompatibility
and resorption attributes of lactide/gly-
Pharmaceutical Technology MAY 2020 29
Development
opers to consistently maintain concen- into the target product profile,” he adds. tions. It may also be beneficial for treating
trations of API, minimize adverse effects, “In the case of topical reformulation, trans- schizophrenia.”
and reduce the number of dosing units forming an oral product to a topical for- Moving forward, Spencer anticipates
required, confirms Mudie. However, mulation can often mean moving from there will be an increasing number of ap-
there is the possibility that the single systemic delivery to local delivery, which plications for local delivery. “In this appli-
unit may be too large for the patient to is important to consider when assessing cation, drug concentrations in local tissues
swallow, she warns. the target patient group.” can be maximized for efficacy while min-
Liposomes. “Liposomes have also In terms of specific disease area, pros- imizing systemic drug exposure resulting
played a significant role in the success- tate cancer patients have been shown to in reduced side effects via placement of
ful reformulation of drug products,” says improve compliance after reformulation the drug product directly into target sites,
Spencer. “Decades-old cancer drugs re- of daily injections to a single injection, such as joints, the spine, the eye, infected
formulated into liposomes and pegylated extended-release, dosage form was done, areas, tumors, or the brain,” he notes.
liposomes are therapeutically efficacious Spencer continues. “Another example is “Furthermore, local delivery can min-
with improved toxicity profiles, better the reformulation of a drug from a daily imize the total amount of dose required,”
cardiac safety, and less side effects. An- oral tablet to an extended-release par- Spencer continues. “As specialized drugs
tifungal liposomes also show reduced enteral dosage form that has improved continue to become more highly potent,
toxicity along with extended-release compliance in schizophrenic and bipolar the goal of reducing injection volumes
performance, which results in longer re- patients,” he says. “In addition to elimi- and extending the period of drug release
tention times of the drug in tissues.” nating the need to remind patients to will become even more important.” As
Additionally, through reformulating take their medication, it has also helped an example, Spencer highlights a recent
pain drugs into long-acting liposome addiction patients from intentionally and approval of an ocular implant, which has
preparations, it is possible to reduce the prematurely stopping their therapy.” the capability of releasing 10 micrograms
frequency of epidural injections required, Adding to Spencer’s comments, Zi- of drug over a four-month period (3).
Spencer adds. “In all, several drugs re- jlstra explains that some relief can be Another approach that may have prom-
formulated into liposomes have been offered to oncology patients if the food ise for the future in Mudie’s opinion is the
strongly preferred to the original dosage label can be removed from the medicine creation of a dosage form that has the abil-
forms,” he states. through reformulation. “Many oncol- ity to release in the GI tract non-tradition-
ogy patients have reduced appetites as a ally. “For example, it may attach to the in-
The target patient profile side effect of chemotherapy,” she notes. testinal wall and release the drug contents
“When considering drug reformulation “However, the medicines they are re- for uptake into the bloodstream,” she says.
strategies, the target patient profile can quired to take often have food labels on “This approach could benefit patients who
play a role in developing the most effec- them, meaning that in some cases, the require frequent injections or have harsh
tive product,” asserts Zijlstra. Giving an patients must take the dose with food. GI environments.”
example, she explains that in the case of Therefore, a reformulation strategy that “Finding new molecules that have ac-
geriatric patients, who can have issues can remove this ‘food’ requirement can ceptable therapeutic profiles is not getting
swallowing larger tablets, it may be ben- improve the patient experience.” any easier,” summarizes Drummond.
eficial to reformulate a drug product into “Furthermore, formulations that com- “Fundamentally, all reformulations, cur-
a multi-particulate or sprinkle-capsule bine local delivery with extended-release rent and future, must benefit the patient,
form, thereby enabling the patient the can dramatically improve rates of brand whether by providing a new solution to
option to sprinkle the medicine onto acceptability for patients who would oth- combating an indication or improving
food or in water. erwise face a series of uncomfortable in- compliance and ease-of-use, to ensure
“Reformulation strategies can be par- travitreal injections or injections into the there is no compromise to the patient.”
ticularly beneficial to a range of chronic knee or other joints,” Spencer iterates.
diseases and patient population sub-sets, References
such as pediatrics or geriatrics,” agrees Potential path forward 1. O.J. Wouters, M. McKee, and J. Luyten,
Spencer. “Improving rates of brand ac- According to Mudie, a potential future JAMA, 323 (9) 844–853 (2020).
2. NICE, “Medicines Adherence: Involving
ceptance amongst patient subpopulations, reformulation strategy could be a digital Patients in Decisions About Prescribed
such as pediatrics, are also a key focus of one. “For example, developers may in- Medicines and Supporting Adherence,”
reformulation strategies.” sert a chip inside a dosage form to alert nice.org.uk, Clinical Guidance, Jan. 28, 2009.
For Drummond, the needs of the pa- either the patient, healthcare provider, or 3. Allergan, “Allergan Receives FDA Ap-
tient must be considered just as much both when the medication has success- proval for Durysta (Bimatoprost Implant)
the First and Only Intracameral Biode-
as the chemical and physical properties fully been administered,” she says. “This gradable Sustained-Release Implant to
of the drug in question, irrespective of innovation could be useful for oncology Lower Intraocular Pressure in Open-An-
whether it is a reformulation or a new patients, who may experience anxiety gle Glaucoma or Ocular Hypertension Pa-
chemical entity. “These factors all feed related to keeping track of their medica- tients,” Press Release, March 5, 2020. PT
M
icroneedle array patches (MAPs) IRV and inactivated poliovirus vaccine of production and packaging equipment
have been in development as (2). Recently, MAPs are being investi- for pharmaceutical companies, medical
an alternative to injections for gated for a vaccine to fight COVID- device companies, and other industries.
delivering vaccines and other drugs. 19 (3). The University of Pittsburgh The company is working with organiza-
Microneedle patches differ from trans- Medical Center (UPMC) announced tions and partners to develop micronee-
dermal patches that deliver medicine in an April 2, 2020 press release that dle array patch (MAP) technology, and in
through the outermost layer of the its scientists had developed a poten- January 2020, Harro Höfliger and PATH
skin (the stratum corneum), because tial vaccine against SARS-CoV-2 that hosted a workshop on MAP manufactur-
the microneedles pierce the stratum would be delivered through a MAP (3). ing attended by MAP developers and rep-
corneum and deliver the drug into The fingertip-sized patch uses 400 mi- resentatives from the World Health Or-
the epidermis or upper part of the croneedles that deliver the spike pro- ganization, UNICEF, the Bill & Melinda
dermis, but not deep enough to cause tein pieces into the skin, where the nee-
Gates Foundation, and the Gavi Vaccine
pain. Microneedle technologies include dles, which are made of sugar and the Alliance. Pharmaceutical Technology
solid microneedles coated with the drug, protein pieces, dissolve. Noting that spoke with Bernsau about some of the
hollow microneedles filled with a liquid scalability is crucial for vaccines in- considerations for MAP manufacturing.
DI STUDIO - STOCK.ADOBE.COM
drug, and dissolvable microneedles with tended for protection from pandemics,
the drug embedded in a soluble material. UPMC said that the process to make Advantages of MAPs
Microneedle patches have been stud- and purify the protein for the vaccine PharmTech: What are some of the advan-
ied in a clinical trial for delivering a flu is scalable, and that mass-producing tages of using MAPs for vaccines?
vaccine (1) and in preclinical trials for the microneedle array involves spin- Bernsau (HH): Vaccine delivery faces
delivering inactivated rotavirus vac- ning the protein-sugar mixture into a several challenges that can be addressed
cine (IRV) and co-administration of mold using a centrifuge. Advantages with MAP technology. A significant
Pharmaceutical Technology MAY 2020 31
Manufacturing
issue is that low- and middle-income One of the keys for sterile production Packaging requirements
countries often do not have a clear cold is material flow through the processing PharmTech: What are the requirements
chain, which is necessary for transport line: raw materials must be brought into for packaging of microneedle patches?
and storage of liquid, injectable vaccines. the machine; various automated steps
MAPs do not require a cold chain. An-
other concern is that a significant number
are performed; then the product must
be taken out of machine. Sterile environ-
“To a large degree
of people have a fear of regular injection
needles and the associated pain. Mi-
ments can include isolators or various
types of barrier technology, depending
the microneedles
croneedles in a patch form eliminate the
pain. In addition, MAPs can be admin-
on the cleanroom setup.
Dosing of the API, either into the hol-
are fragile.”
istered without skilled healthcare work-
ers, by trained workers or potentially by
low mold or as a coating, must be done
in a combination of high precision and
—Stefan Bernsau,
self-administration. This advantage is
beneficial for developing countries that
with high output. These are very small
doses, and different dosing technologies
Harro Höfliger
lack skilled healthcare providers. Re- are used to obtain the specific tolerances
cently, it is also being thought of as a ben- needed for a particular process. There Bernsau (HH): To a large degree the mi-
efit for all countries for use in pandemics, are some new developments in the mar- croneedles are fragile; they are typically
where patients could potentially have the ket for dosing. VAXXAS, for example, packaged in an applicator, with a rigid
vaccine delivered for self-administration, has developed their own high-speed and container to protect the needles. The
thus avoiding the need for people to go to high-precision dosing technology that is patch must also be packaged in a sterile
doctors’ offices or hospitals. similar to ink-jet printing technology. As environment. After primary packaging,
another example, scraper technology can it must also be packaged for transpor-
Microneedle array be a solution for dissolvable micronee-
dles, depending on the viscosity and the
tation. An advantage of MAPs is that
temperature is not as much of an issue,
patches are being physical characteristics of the liquid or
the vaccine.
compared to traditional liquid vaccines,
because the MAPs vaccine is dry.
investigated for For the automation of the lines, we use
dedicated servo-driven units because the Outlook
a vaccine to fight parts being handled are so small—the
whole array may be about the size of a
PharmTech: Do you foresee in-country,
for-country production with MAP
COVID-19. penny, for example. Robotic arms can
cause turbulence in the airflow, which
technology?
Bernsau (HH): It is far too early to know
can be a problem regarding the aseptic whether production in developing coun-
Manufacturing considerations requirements of such small, precise units. tries is feasible, because the initial invest-
PharmTech: What are the biggest chal- Another proven technology is a so-called ment in these lines is high and a lot of
lenges in microneedle patch manufac- ‘walking beam,’ which uses pick-and- technical knowledge and skill are needed
turing and what are some best practices place handling for small and delicate to run these lines. However, some organi-
for addressing these challenges? parts inside an isolator. zations are looking at this possibility, and
Bernsau (HH): Companies developing PharmTech: What are some best prac- the current pandemic may cause more
products at the laboratory scale are tices for inspection of MAPs after governments globally to think about
looking at issues such as dosing (either manufacturing? in-country production.
by coating or filling) and drying. At Bernsau (HH): Several quality attributes
the commercial scale, however, auto- need to be inspected, particularly that References
mation is crucial for obtaining output the required amount of drug is there (ei- 1. NIH, “Researchers Develop Micronee-
at an appropriate cost, with a reason- ther coated or filled). Inspection of mi- dle Patch for Flu Vaccination,” Press
Release, June 27, 2017.
able total cost of ownership. As a ma- croneedles is more difficult than syringe 2. Micron Biomedical, “Micron Receives
chine manufacturer, we want to join inspection because of the small size of the Additional Funding from the CDC for
in the development process as early as microneedles. Cameras are used to view Collaboration on the Development of a
possible so that we can provide input the top and side. Other critical pieces are Microneedle Patch for IRV-IPV Co-ad-
to developers for how to optimize the the light source, appropriate software, ministration,” Press Release, Aug. 27,
process for commercial automation. and hardware for fast computing speed. 2018.
3. UPMC, “COVID-19 Vaccine Candidate
One of the constraints for vaccine A best practice is 100% inspection of each Shows Promise,” Press Release, April 2,
manufacturing is that most cannot be part, but a challenge is the speed and an 2020.
terminally sterilized, therefore they must eventual reflectance of image acquisition 4. J. Markarian, Pharm. Tech. 43 (11) 48–
be produced in a sterile environment. to accomplish this. 49,54 (2019). PT
The number
How a Contamination one source of
contamination
Control Program Impacts in the cleanroom
Product Sterility environment is
people.
Aaron Mertens and Joe McCall
Elements of a
contamination control program
A well-defined contamination control program To meet the quality system requirement
detailed in ICH Q10 for a CCP, the fol-
is essential to maintain quality through aseptic lowing elements must be considered:
manufacture of parenteral drug products. • Environmental monitoring (EM)
• Gowning
A
robust contamination control International Council for Harmoniza- • Cleaning and disinfection
program (CCP) does more than tion (ICH) Q10 includes the following • Materials and equipment airlocks
simply remove microbial con- systems, each of which is integrally • Sterilization
tamination from cleanroom surfaces related to CCP (1): • Parts and equipment use at filling
and equipment; a CCP ensures aseptic • Provision of facilities, utilities, line.
processes result in a sterile finished and equipment Best practices within these elements
product. CCPs are critical to prod- • Production, including packaging ensure the risk of a non-sterile product is
uct and patient safety and must be and labeling minimized. Documentation within work
multifaceted to consider all aspects • Quality control and assurance. instructions and/or standard operating
of contamination to be effective. For Assembling a multidisciplinary procedures (SOPs) provides direction to
pharmaceutical, biotechnology, and team to evaluate microbiological risk trained staff, ensuring consistent execu-
medical device manufacturing, CCPs to the product identifies key areas of tion of the best practices.
are one of the foundational elements focus in the CCP including elements Environmental monitoring. Good EM
of a facility’s pharmaceutical quality of the science-based assessment, poli- practices demonstrate microbial control
system (PQS), as described in FDA cy-based management, and risk com- and identify adverse trends within the
SEVENTYFOUR - STOCK.ADOBE.COM
Guidance for Industry, Q10 Pharma- munication (2). This exercise provides cleanroom environment. Data collected
ceutical Quality System. The scope of direction and prioritization within the using an environmental particle counter
elements of the CCP. over time validates the cleaning and dis-
Aaron Mertens is a Microbiologist ICH Q9, Quality Risk Management infection process and gives evidence that
and Technical Services Manager at provides a framework for the risk the contamination control program is
STERIS Life Sciences, and Joe McCall,
previously with STERIS, is the Associate management process within a phar- effective. Figure 1 shows an overview of
Director of QA Technical Service at ADMA maceutical organization. The goal is to points to consider when starting new or
Biologics. proactively identify and manage prod- evaluating existing EM practices.
Pharmaceutical Technology MAY 2020 33
Manufacturing
Gowning. The number one source of
Figure 1. Overview for evaluating environmental monitoring practices.
contamination in the cleanroom environ-
ment is people. Operators working in the
cleanroom must adhere to strict gowning
practices to ensure particulate and micro-
Identification organism contamination stemming from
• Formal program identifies and banks relevant
microbial isolates. the operator is kept to a minimum. Con-
tinued personnel training and monitoring
Response to Results
is an essential part of the CCP. All opera-
• Defined response plan to alert, action, and tors should be included in a training pro-
trends based on risk, regulatory guidance and
requirements, and historical facility gram and undergo formal qualification
performance capacity.
(initial and periodic assessment of skill).
Trending Training should be conducted by subject
• Formal documented trend analysis program,
periodic, and trend analyses are reported to matter expert(s) (SMEs), and employees
and signed-off by the Plant.
should be restricted from cleanroom ac-
Sampling Plan
cess until fully qualified.
• Formal documented risk analysis. Aseptic operators should be monitored
• Sample site selection, frequency, test methods,
and excursion response plan. at appropriate sites and frequencies (based
on risk assessment). There should be an
SOP-defined program for addressing poor
Table I. Cleanroom environmental contamination control program. EPA is US Environmental Protection Agency. DE is
disinfectant efficacy. USP is United States Pharmacopeia. SOP is standard operation procedure. ISO is International
Organization for Standardization. RTU is ready to use.
Formal program for qualification, use, and disposal of disinfectants. Disinfectants in the United States are
Selection of agents Environmental Protection Agency (EPA) registered, with defined criteria for selection and suitability of agents
(i.e., substrate compatibility, label claims, etc.)
More than one type; use of a broad-spectrum disinfectant with periodic use of sporicidal (frequency driven by
environmental monitoring data), use of sterile 70% isopropyl alcohol (IPA) for aseptically gowned operators in ISO
Rotation
5/ISO 7 zones (e.g., frequently on gloves, surfaces after interventions). Disinfectants have detergent capacity, or
formal program exists for cleaning of surfaces prior to disinfection.
Frequency and method for rinsing is based on usage of disinfectant(s). Procedure is defined in standard
operating procedures (SOPs); use of high purity water (United States Pharmacopeia purified water or water
Residue removal
for injection) or 70% IPA for surface rinsing and residue removal. Formulated detergents used as needed,
based on residue type and build-up.
Application Well defined in SOP (e.g., specify two or three bucket systems, stipulate mop type(s), hands-on operator
method(s) training, periodic audit of program by QA to verify application technique and contact time.
Formulated cleaner used to help remove debris and soil. SOP-defined plan for shutdown recovery
Recovery from loss (e.g., 3X Clean & Disinfection with final round of sporicidal). In-situ study performed (1x) to demonstrate
of control (planned ability to recover from loss of aseptic control. EM program used to generate in-situ data to demonstrate
or unplanned) effectiveness, i.e., area mapping via contact plates after worst-case event; mapping done pre- and post-
recovery to document ability to reestablish cleanroom conditions.
Disinfectant efficacy Formal DE study performed using facility-specific substrates (surfaces) and microbial isolates. Substrates include
ALL FIGURES COURTESY OF THE AUTHORS.
Wet contact time Contact time is established in the DE study, and reflected in actual practice of disinfectant application in the facility.
Sterile vs. non- Selection is based on risk. If used in ISO 5 zone, solutions must be sterile—either purchased as sterile or
sterile introduced into ISO 5 zone via validated process (e.g., sterile filtration).
If not using RTU, then use-dilution is supported by DE study, with SOP preparation instructions including
Use-dilution
specified quality of water (e.g., USP Purified, WFI), method for accurate measurement, shelf-life.
Manufacturing
was ordered. Investigational sampling in
Figure 3. Root cause analysis.
the unclassified storage area revealed the
presence of genetically identical profile
of mold and Bacillus species, to that re-
covered from the bulk material and post-
SIP tank interior. No proactive measures
were taken to protect the integrity of the
tank while it was held in an unclassified
environment. Upon reintroduction to the
classified area, exterior surfaces of the tank
were disinfected and treated with a spori-
cide, and CIP and SIP cycles of the tank
were completed. A concurrent review of
the tank’s SIP validation package showed
that although it passed validation criteria,
the biological indicator (BI) located at the
rupture disc connection was a worst-case
location for steam penetration. The in-
vestigation logically concluded that the
rupture disc zone of the tank had been
exposed to the contaminating microor-
ganisms while it was stored in the unclassi-
fied area, and that the ensuing CIP and SIP
cycles were ineffective at removing surface
Figure 4. Bulk tank. contaminants from this specific area of the
tank’s interior.
The resulting preventative action in-
cluded a new policy to provide physical
barrier protection of critical equipment
and surfaces when moved to or stored in
areas of lower classification. Figure 4 shows
the location of the rupture disk on the
tank). Figure 5 shows the good manufactur-
ing practice (GMP) equipment cover being
applied to protect the tank from potential
environmental contamination.
GMP equipment covers manufactured
using nonwoven spunbonded polyolefin
material can protect equipment from
particulate and microbial contamination
during storage and when out of use. Re-
gardless of how robust the CCP, environ-
mental monitoring demonstrates that mi-
while out of the controlled environment the tank had been stored revealed the croorganisms are present at varying levels
for several weeks. Upon reintroduction presence of genetically identical profile in any manned cleanroom environment.
to the classified area, exterior surfaces of mold and Bacillus species to that re- Additionally, aging facilities contribute
of the tank were disinfected and treated covered from the tank interior. to potential microbial risk to equipment
with a sporicide. Clean-in-place (CIP) A review of the bulk tank’s history in the cleanroom environment. It is im-
and steam-in-place (SIP) of the tank was through the equipment logbook showed portant that barrier protection of critical
performed; however, review of the tank’s the only significant event prior to the onset surfaces be part of the CCP, especially
SIP validation package showed that the of the contamination was a burst rupture during equipment staging and storage.
interior area at the rupture disc connec- disc. Interviews with the manufacturing Minimizing risk of microbial contamina-
tion proved to be a worst-case location technician revealed that the tank had been tion, through prevention and protection of
for steam penetration. Investigational relocated for several weeks to a non-classi- the product contact surfaces, should be the
sampling in the unclassified area where fied area while a replacement rupture disc highest priority element in the CCP.
36 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
barrier protection increases sterility
Figure 5. GMP equipment cover.
assurance of critical surfaces and fur-
ther enhances the CCP.
The case study exemplifies how a sim-
ple solution like utilizing barrier protec-
tion over a bulk product tank would have
prevented contamination of the bulk
product. A proactive microbiological risk
assessment in addition to the contamina-
tion control program will identify oppor-
tunities like this, where simple solutions
minimize risk to the product. A holistic
microbial risk assessment of all contami-
nation control program elements ensures
production of safe medicines and products.
References
1. ICH, Q10 Pharmaceutical Quality System
(FDA, April 2009).
2. Quality Assurance and Food Safety, “Mi-
crobiological Risk Assessment,” Global
Manager, Food Safety Services Innova-
Summary airlock and sterilization procedures, tion, June 2014.
As part of a facility PQS, a robust con- equipment use at the filling line) to 3. ICH, Q9 Quality Risk Management (FDA,
tamination control program includes minimize the microbial risk to the November 2005).
a diverse set of elements (e.g., EM, finished product, as well as to meet 4. J. McCall, Case Study, unpublished man-
gowning, cleaning and disinfection, requirements in ICH Q10. Additional uscript, 2008. PT
Visit www.renishaw.com/raman
Renishaw, Inc. 1001 Wesemann Drive, West Dundee, Illinois, 60118, United States
T +1 847 286 9953 F +1 847 286 9974 E raman@renishaw.com
www.renishaw.com
A
statistical sampling plan is not included within United
The CuDAL-Excel program, based on Microsoft
States Pharmacopeia (USP) <905> Uniformity of Dos-
(MS) Excel, has been developed to calculate age Units and USP <711> Dissolution, which are only
the United States Pharmacopeia (USP) passing intended to determine conformance of a sample taken
probability of content uniformity and dissolution from routine batches or the pharmaceutical market. Passing
tests for both sampling plan 1 and sampling plan USP <905> tests once does not provide statistical assurance that
2 scenarios and for both immediate release and a batch will meet the target quality (1), thus testing against USP
acceptance criteria is insufficient for batch release or process val-
extended release requirements. The users can
idation purposes. Statistical acceptance criteria with statistically
obtain the passing probability by simply entering valid sampling plans can better assess the quality of a process.
the input variables, with wide applications For in-process sampling of dosage units, FDA’s current rec-
for process validation/verification and batch ommendation is to use nested sampling plans—to test replicate
release. As a user-friendly program, CuDAL‑Excel samples (more than one, typically three, which should be sam-
should bring more benefits to the industry pled at closest positions) from each location of the stratified plan
(2). In this way, the data can be subjected to variance component
practitioners than other existing programs/tools.
analysis, which divides the total variance into “between location”
and “within-location” components (2), increasing the evaluation
confidence for batch quality. The “between-location” component
is the variability across the sampling locations in a blender, or
during compression, encapsulation, or filling process, while the
“within-location” component is the variability between samples
within a given sampling location.
Bergum published a method to calculate a lower bound on the
probability of meeting acceptance criteria of multiple stage tests
(3), such as the USP content uniformity and dissolution tests, and
gave examples in a book (4). Later, Bergum wrote a SAS program
(CuDAL version 1), which implements his calculation method (5).
ELENA ABRAZHEVICH - STOCK.ADOBE.COM
Tergus Pharma is an industry leading CDMO for all Topical Drug Product Development needs f rom
the concept stage to Clinical Trial Materials (CTM) and cGMP Commercial Manufacturing.
proposition to our clients by bringing 505 (b)(1) monitoring and responding to the challenges
and 505(b)(2) molecules to clinical phase and created by Corona Virus, protecting our
complex generics to match RLD requirements colleagues, while ensuring we continue our
The Equation is Simple: The Art + Science of by producing clinical trial materials for various
Formulation Analytical IVRT Skin Penetration Commercial cGMP CMC Reg Clinical Trial
Development R&D (Franz Diffusion) (IVPT Studies) Manufacturing Consulting Materials
www.TergusPharma.com
Peer-Review Research
Table I. CuDAL-Excel program sheets.
Calculate standard deviation (SD) for Calculate the mean, SD, maximum, and
1 NA 1
sampling plan (SP) 1 minimum
Calculates the mean, within-location SD
Calculate standard deviation (SD)
2 (denoted as SE), between-location SD NA 2
for SP2
(denoted as SM), maximum and minimum
Calculate the passing probability against
3 CUSP1 NA 1
USP <905> Content Uniformity test
Calculate the passing probability against
4 CUSP2 NA 2
USP <905> Content Uniformity test
Calculate the passing probability
5 DissSP1 Immediate-release 1
against USP <711> Dissolution test
Calculate the passing probability
6 DissSP2 Immediate-release 2
against USP <711> Dissolution test
Calculate the passing probability
7 ExtDissSP1 Extended-release 1
against USP <711> Dissolution test
Calculate the passing probability
8 ExtDissSP2 Extended-release 2
against USP <711> Dissolution test
*SD, standard deviation; NA, not applicable; SP, sampling plan; SE, within-location SD; SM, between-location SD; USP, United States Pharmacopeia;
CUSP1, content uniformity sampling plan 1; CUSP2, content uniformity sampling plan 2; DissSP1, dissolution sampling plan 1; DissSP2, dissolution
sampling plan 2; ExtDissSP1, extended release dissolution sampling plan 1; ExtDissSP2, extended release dissolution sampling plan 2
based on readily accessible software would be more acceptable. quires estimation of statistical lower and/or upper bounds for the
R language programs, transformed from SAS programs, were mean and standard deviation (SD) of the samples of appropriate
written by www.pharmstat.com. Pramote Cholayudth, coauthor size. The calculation method assumes that the test results can be
of this paper, first developed (8) and later revised (9) a Microsoft approximated by a normal distribution. For detailed calculation
(MS) Excel program to compute the probability of passing a USP methodology, interested readers can unhide the hidden MS Excel
content uniformity test; however, it is no longer consistent with columns and rows to understand the calculation algorithm. The
Bergum’s current CuDAL method. The International Society for open-source CuDAL-Excel file is available from the author.
Pharmaceutical Engineering (ISPE) has also published five com-
mon acceptance limit tables for sampling plans 1 and 2 based CuDAL-Excel inputs
on ASTM E2709/E2810 and an Excel Workbook (10). However, Overall mean. The calculation of overall mean for both sam-
due to the limited number of combinations of acceptable proba- pling plan (SP) 1 and SP2 are the same. Overall mean = AV-
bility, number of samples and sample locations, and confidence ERAGE(D2:D101) or = AVERAGE(F3:Y72), where D2:D101 or
levels, the tables are not capable of satisfying diverse company F3:Y72 are the individual results of all samples. All measurements
requirements. of dosage units and criteria values are in percentage label claims
The present work reports the CuDAL-Excel, a set of MS Excel (%LC).
programs transformed and extended from Bergum’s CuDAL ver- Calculation of SD for SP1. SD = STDEV(D2:D101), where D2:D101
sion 2 SAS program. The CuDAL-Excel is used to evaluate data are the individual results of samples from all the locations.
against the USP 29–43 <905> Content Uniformity and <711> Dis- Calculation of SD for SP2. Within-location SD (denoted as SE) =
solution for both immediate-release and extended-release dosage SQRT(AVERAGE(D3:D72)), where D3 = VAR(F3:Y3), with F3:Y3
forms (the test for extended-release dissolution is an additional as the individual results of samples from the first location, and
function to CuDAL). It is designed for industry practitioners for D72 = VAR(F72:Y72), with F72:Y72 as the individual results of
batch release and process validation. samples from the last location.
Between-location SD (denoted as SM) = STDEV(C3:C72),
CuDAL-Excel program where C3 = AVERAGE(F3:Y3), with F3:Y3 as the individual
The CuDAL-Excel is an MS Excel file that contains eight MS results of samples from the first location, and C72 = AVER-
Excel sheets with functions listed in Table I, and it calculates the AGE(F72:Y72), with F72:Y72 as the individual results of samples
passing probability of result data against USP multiple stage tests, from the last location.
<905> Uniformity of Dosage Units and <711> Dissolution. The Sample size. The sample sizes are different for SP1 and SP2.
CuDAL-Excel is based on Bergum’s methodology and is trans- • For SP1, one (n=1) dosage unit is sampled from each of the L
formed and extended from CuDAL SAS programs as well as from locations, with a total sample size of L. The locations should
R programs written by www.pharmstat.com. The calculation re- be equally spaced throughout the batch.
40 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
HOW ARE YOU PROGRESSING
DRUG DEVELOPMENT DURING
THE PANDEMIC?
More informa�on at
absorp�on.com/bcs
For quotes, email contact@absorp�on.com
©2020 Absorption Systems, LP. All rights reserved.
Peer-Review Research
• For SP2, more than one (n>1) dosage units are sampled from Table II. United States Pharmacopeia (USP) 29–43
each of the L locations, with a total sample size of n×L. The <905> Uniformity of Dosage Units test procedure and
locations should be equally spaced throughout the batch, acceptance criteria.
while the n samples within a location should be sampled as
close as possible. Stage (S) Number tested Pass stage if:
Confidence level. Confidence levels as well as probability values ∙ + 2.4 s ⩽
AV = |M – X|
(P-values) are typically 50%, 90%, or 95%. A Parenteral Drug S1 10 15.0, where M is defined
Association (PDA) Technical Report suggests the use of a 90% below.
confidence level to provide 95% coverage (6). In the regulatory ∙ + 2.0 s ⩽
(1) AV = |M – X|
arena, a confidence level of 95% is often used (1). 15.0, using all 30 results
(S1 + S2).
Probability lower bound (passing probability). The probability (or S2 20 (2) No dosage unit is
coverage) lower bound, denoted as “passing probability,” is the outside the maximum
lowest probability of passing the USP acceptance criteria, and 90%, allowed range of 0.75 ×
M to 1.25 × M.
95%, and 99% are commonly used, with 95% as the usual value
employed in practice (1). M is defined as follows:
∙ 101.5,
Probability of passing If 98.5 ⩽ X⩽ ∙ (AV = ks)
M =X
then
USP 29–43 <905> Uniformity of Dosage Units test
M = 98.5% (AV = 98.5 –
The USP <905> Uniformity of Dosage Units test procedure and If T ⩽ 101.5 %LC ∙ < 98.5, then
If X ∙ + ks)
X
acceptance criteria are listed in Table II (11). The two-stage test
∙–
and its acceptance criteria have not changed from USP 29 to the ∙ > 101.5
If X
M = 101.5 (AV = X
101.5 + ks)
current USP 43.
Sheet CUSP1 (CU sampling plan 1). The Excel sheet CUSP1 needs ∙ T, then
If 98.5 ⩽ X⩽ AV = ks
users to input cells B1–B5 with the target content, confidence level, If T > 101.5 %LC ∙ < 98.5, then
If X ∙ + ks
AV = 98.5 – X
number of samples tested, average, and SD of all sample results. ∙ > T, then ∙ – T + ks
If X AV = X
Then cell B6 will display the calculated passing probability. The
merged cell range A7–B11 gives the narrative conclusion. Note: At each stage calculate the sample average, X, ∙ and the sample
The calculation formulas are demonstrated in the CuDAL-Ex- standard deviation, s.
∙ sample
S, stage; AV, acceptance value; s, sample standard deviation; X,
cel file. With given inputs (cells B1:B5), the statistical lower bound, average; T, target content; M, reference value; k, acceptability constant.
LLU (B23), and upper bound, ULU (B43), of the mean, can be The information is derived from USP General Chapter <905> “Uniformity of
calculated. For LLU, the passing probability of stage 1 is calcu- Dosage Units” (11).
lated in B30 and the passing probability of stage 2 is in B41. B28 probability increases with decreasing SD and with a mean closer
and B29 are the probability integration results for stage 1, while to center target, and that a higher SD can be tolerated when the
B35 and B36 are the probability integration results for stage 2. mean is close to the center target.
The integration calculation results, int2 (H304, sum of H4:H303), Sheet CUSP2 (CU sampling plan 2). Sheet CUSP2, compared with
int3 (N304, sum of N4:N303), iint2 (T304, sum of T4:T303), iint3 sheet CUSP1, has more intermediate parameters (B17–B31 in
(Z304, sum of Z4:Z303), and the detailed calculation steps are sheet CUSP2) than SP1 (B13–B15 in sheet CUSP1), due to the
demonstrated in the CuDAL-Excel file. The maximum one of the more complicated calculation algorithm of SIGMA. These cal-
passing probabilities of stage 1 and stage 2 (B30 and B41, respec- culations involve the number of samples per location, number of
tively) is the overall passing probability for LLU (B42). The cal- locations, within-location SD, and between-location SD. However,
culation of overall passing probability for ULU (B62) is the same. all the other calculation steps are almost the same as sheet CUSP1.
The minimum one of overall passing probabilities for LLU (B42) The detailed calculation steps are shown in CuDAL-Excel file
and ULU (B62) is the final passing probability and shown in B6. sheet CUSP2.
For example, if target content = 100, confidence level = 95%, For example, if the target content = 100, confidence level = 95%,
number of samples tested = 30, mean of tested results = 99.9% number of samples tested per location = 3, number of locations =
label claim, SD of the tested results = 3.996, the calculated proba- 20, mean of tested results = 99.9% label claim, within-location SD
bility of passing USP Content Uniformity test for a future sample of the tested results = 4, between-location SD = 3.3, the calculated
taken from the batch is approximately 97.78% (0.977799163). The probability of passing USP Content Uniformity test for a future
obtained probability is larger than 95%, thus the batch is well sample taken from the batch is about 86.09% (0.860912446). The
within the release requirement, or the process is valid to produce obtained probability 86.09% is too small to accept, thus the batch
products meeting USP Content Uniformity test requirements. is not released or the process needs to be improved.
For another example (as shown in Figure 1), if the given target For another example (as shown in Figure 2), if the given target =
= 100, confidence level = 90%, and number of samples = 18, the 100, mean = 100, confidence level = 90%, number of samples per
passing probability is calculated by CuDAL-Excel with varying location = 3, and number of locations = 20, the passing probability
mean and SD. The obtained contour chart shows that passing is calculated by CuDAL-Excel with varying within-location SD
42 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
Figure 1: The contour chart of passing probability Figure 2: The contour chart of passing probability against
against United States Pharmacopeia content United States Pharmacopeia content uniformity test
uniformity test calculated with CuDAL-Excel (sheet calculated with CuDAL-Excel (sheet CUSP2) from varying
CUSP1 of the MS Excel file) from varying standard within-location SD (SE) and between-location SD (SM),
deviation (SD), based on target = 100, confidence level based on target = 100, mean = 100, confidence level =
= 90%, and number of samples = 18. 90%, number of samples per location = 3, and number of
locations = 20. SD is standard deviation.
Between-location SD (SM)
7
> 0.95
105
6
Mean
5
100
95
3
2
90
85
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 9 10
SD Within-location SD (SE)
and between-location SD. The contour chart shows that passing Extended-release SP1 (sheet ExtDissSP1). For example, if the acceptable
probability increases with decreasing within-location and be- dissolution limit for this non-final time point is a range, dissolu-
tween-location SD. To meet a passing probability, the allowable tion lower limit = 48%, dissolution upper limit = 72%, confidence
within-location SD is larger than between-location SD. level = 95%, number of samples tested =18, mean of tested results
= 56%, SD of the tested results = 5, the calculated probability of
Probability of passing USP 43 <711> Dissolution test passing USP dissolution test (extended-release) for a future sam-
The USP Dissolution test procedures and acceptance criteria are ple taken from the batch is approximately 94.37% (0.943742397).
summarized in Table III and Table IV (12). The obtained probability 94.37% is acceptable (if there is a pre-
Immediate-release dosage form. For immediate-release dosage defined acceptance limit of 90%), thus the batch is meeting the
forms, examples for calculating passing probability of sampling release requirement, or the process is valid to produce products
plan 1 and sampling plan 2 are respectively demonstrated below. meeting USP Dissolution test requirements.
Immediate-release SP1 (sheet DissSP1). For example, if the Q limit = 80, Extended-release SP2 (sheet ExtDissSP2). For example, if the acceptable
confidence level = 95%, number of samples tested =30, mean of dissolution limit for this non-final time point is a range, dissolu-
tested results = 84% label claim, SD of the tested results = 4.704%, tion lower limit = 48%, dissolution upper limit = 72%, confidence
the calculated probability of passing USP Content Uniformity level = 95%, number of samples tested per location = 3, number
test for a future sample taken from the batch is about 91.16% of locations = 20, mean of tested results = 56%, within-location
(0.911628045). The obtained probability 91.16% is acceptable (if SD of the tested results = 5.2, between-location SD of the tested
there is a predefined acceptance limit of 90%), thus the batch is results = 4.5, the calculated probability of passing USP Disso-
meeting the release requirement, or the process is valid to produce lution test (extended-release) for a future sample taken from
products meeting USP Dissolution test requirements. the batch is approximately 93.38% (0.933772184). The obtained
Immediate-release SP2 (sheet DissSP2). For example, if the Q limit = 80, probability 93.38% is not acceptable (if there is a predefined ac-
confidence level = 95%, number of samples tested per location = ceptance limit of 95%), thus the batch is not meeting the release
3, number of locations = 20, mean of tested results = 88% label requirement, or the process needs to be improved.
FIGURES ARE COURTESY OF THE AUTHORS
claim, within-location SD of the tested results = 7.5, between-lo- In another example where an extended-release tablet dissolu-
cation SD = 5, the calculated probability of passing USP Content tion is tested at three time points, the dissolution specification
Uniformity test for a future sample taken from the batch is about is when the dissolution percentage at the first time-point is not
94.31% (0.94309498). The obtained probability 94.31% is accept- more than 40%, the dissolution percentage at the second time-
able (if there is a predefined acceptance limit of 90%), thus the point is between 50% and 75%, and the dissolution percentage
batch is meeting the release requirement, or the process is valid at the last time point is not less than 90%. When the given con-
to produce products meeting USP Dissolution test requirements. fidence level = 95%, number of samples tested per location = 3,
Extended-release dosage form. For extended-release dosage forms, and number of locations = 20, the passing probability of data
examples for calculating passing probability of sampling plan 1 obtained at the first time point is calculated by ExtDissSP2 to be
and sampling plan 2 are respectively demonstrated below. 0.993062574 (inputs are the data from the final time point?=No,
Pharmaceutical Technology MAY 2020 43
Peer-Review Research
Table III. United States Pharmacopeia (USP) 43 <711> Table IV. United States Pharmacopeia (USP) 43 <711>
Dissolution test procedure and acceptance criteria, Dissolution test procedure and acceptance criteria,
immediate-release dosage forms. extended-release dosage forms.
Stage Number tested Acceptance criteria Level Number tested Acceptance criteria
Stage Each of the 6 individual units should be Each of the individual values should be within
6 Level
1 equal or larger than Q +5%. each of the specified ranges, and each of the
1 6
individual values at the final test time point
The mean value of all the 12 units from should be ⩾ the specified amount.
Stage both Stage 1 and Stage 2 should not
6 The mean value of all the 12 units from both
2 be less than Q, and none of the 12 units
should be less than Q − 15%. Level 1 and Level 2 should be within each
of the specified range and the mean value
The mean value of all the 24 units from of the final test time point should be ⩾ the
all three stages (Stage 1, Stage 2, and Level specified amount; each of the individual
2 6
Stage Stage 3) should not be less than Q, the values should be within 10% of label con-
12
3 number of units that are less than Q – 15% tent outside each of the specified ranges;
should not be more than 2, and each of the and each of the individual values of the final
individual 24 units should be ⩾ Q – 25%. test time point should be ⩾10% of the la-
beled content below the specified amount.
Q, quantity, expressed as a percentage of the labeled content. The mean value of all the 24 units from all
The information is derived from USP General Chapter <711> “Dissolution” (12). the three levels (Level 1, Level 2, and Level
3) should be within each of the specified
dissolution lower limit = 0, dissolution upper limit = 40, mean ranges and the mean value of the final
test time point should be ⩾ the speci-
= 35, within-location SD = 3.4, and between-location SD = 2.9), fied amount; the number of units that are
that of the second time point is 0.995913738 (inputs are the >10% of labeled content outside each of
data from the final time point?=No, dissolution lower limit = Level the specified ranges should not be more
3 12 than 2; the number of units of the final test
50, dissolution upper limit = 75, mean = 63, within-location SD time point that are >10% of labeled content
= 4.3, and between-location SD = 4.7), and that of the last time below the specified range should not be
point is 0.962366544(inputs are the data from the final time more than 2; and each of the 24 individual
units should be within 20% of labeled con-
point?=Yes, dissolution lower limit = 90, mean = 95, within-lo- tent outside each of the specified ranges
cation SD = 2.9, and between-location SD = 3.7). The overall while for the final test time point all values
passing probability is calculated, by manually multiplying the should be equal or larger than 20% of la-
beled content below the specified amount.
above three probability values, to be 0.951784997 (= 0.993062574
× 0.995913738 × 0.962366544). That means the passing proba- The information is derived from USP General Chapter <711> “Dissolution” (12).
bility of the data tested against product specification as per USP
procedures is 95.18%. Therefore, the CuDAL-Excel can be used as an alternative of
CuDAL SAS programs. It is essential that users validate the
Validation of CuDAL-Excel spreadsheet before use in a regulated environment. The Cu-
The CuDAL-Excel was validated by comparison of results DAL-Excel file and its user manual can be freely downloaded
obtained from the CuDAL-Excel with those from Bergum’s from CuDal_Excel@163.com. With this file, the users can obtain
CuDAL SAS programs. The results, as listed in Table V, show that the passing probability by simply entering the input values. The
the above two programs generate the same passing probability CuDAL-Excel can be used for process validation/verification
values. The calculated intermediate parameters (not shown) are and batch release. As a user-friendly Excel-based program, Cu-
also the same for the two programs. The above results demon- DAL-Excel should bring more benefits to practitioners than the
strate that the two programs are equivalent. other tools.
As Bergum’s CuDAL programs do not cover probability de- Access to tool. The tool can be downloaded from
termination for dissolution of extended-release dosage forms, it CuDal_Excel@163.com.
was not possible to viably compare against the original SAS pro- Editor’s Note: The email address is provided by the authors and
grams. The authors have done their best to check the calculation Pharmaceutical Technology does not assume any liability for the
steps/algorithm of sheets ExtDissSP1 and ExtDissSP2 under the contents of the linked file.
principle of CuDAL statistical methodology. Users may decide
to perform additional validation. Acknowledgement
This work fully respects and benefits from James Bergum’s great
Conclusion statistical methodology and original SAS programs. The authors’
The developed CuDAL-Excel is used to calculate the passing intention is to make the great methodology performable on a
probability of collected data against USP multi-stage Content software MS Excel which is widely accessible to industry prac-
Uniformity and Dissolution tests. The programs are based on titioners. The authors are grateful to the reviewers of this paper
Bergum’s methodology, and the functions have been validated. for their valuable comments.
44 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
Table V. Comparison of results from CuDAL-Excel and Bergum’s CuDAL SAS programs. LC is label claim.
Passing probability
Target Confidence Number of
Mean (%LC) SD
(%LC) level (%) samples
CuDAL-Excel Bergum’s CuDAL SAS
Between- Passing
Confidence Number of Number of Within-location probability
Target MEAN location
level (50.0– samples per locations standard
(%LC) (%LC) standard Bergum’s CuDAL
99.0%) location (NN) (LOC) deviation (SE) CuDAL-Excel
deviation (SM) SAS
100 95 3 20 99.9 4.5 3.1 0.845161545 0.8451615445
104 90 3 20 98 5.1 2.7 0.657672831 0.6576728309
References
8. P. Cholayudth, Pharm. Tech., 28 (9) 86–98 (2004).
1. ASTM Standard Number E2810-11(R2017), “Standard Practice
9. P. Cholayudth, J. Valid. Technol., 14 (2) 62–72 (2008).
for Demonstrating Capability to Comply with the Test for Unifor-
10. ISPE, “Common Acceptance Limit Tables for Sampling Plans 1 & 2
mity of Dosage Units” (ASTM International, West Conshohocken,
Based on ASTM E2709/E2810,” ispe.org [Accessed April 18, 2020].
PA, 2017).
11. USP, General Chapter <905>, “Uniformity of Dosage Units” USP (US
2. J.S. Bergum, et al., Pharm. Eng., 34 (2) 28–39 (2014).
Pharmacopeial Convention, Rockville, MD, 2014).
3. J.S. Bergum, Drug Dev. Ind. Pharm., 16 (14) 2153–2166 (1990).
12. USP, General Chapter <711>, “Dissolution” USP (US Pharmacopeial
4. J.S. Bergum and M.L. Utter, “Process Validation,” in Encyclopedia of
Convention, Rockville, MD, 2011). PT
Biopharmaceutical Statistics, S-C Chow, Ed. (Marcel Dekker, New
York, 2000), pp. 422–439.
5. J.S. Bergum, SAS Programs, “Content Uniformity and Dissolution Lei Lei* is senior R&D director of Cutia Therapeutics, Shanghai,
Acceptance Limits (CUDAL), Version: 1.0,” validation completed Feb. China, and he was principal scientist of Shanghai Johnson
11, 2002. & Johnson Pharmaceutical Company, leilei-cn@163.com.
6. J.S. Bergum and H. Li, Pharm. Tech., 31 (10) 90–100 (2007). Pramote Cholayudth is principal and general manager of PM
7. J.S. Bergum, SAS Programs, “Content Uniformity and Dissolution Consult, Bangkok, Thailand, cpramote2000@yahoo.com.
Acceptance Limits (CUDAL), Version: 2.0,” validation completed
*To whom all correpondence should be addressed.
Nov. 1, 2007.
to Genotoxins
NDMA had also been found in metformin,
the fourth most widely prescribed drug
in the United States, and one of the most
widely prescribed diabetes treatments in
the world. In March 2020, after testing 38
Agnes Shanley
lots of metformin from 22 suppliers, Vali-
sure filed a citizen’s petition with FDA (6).
Discovery of nitrosamines in three of the world’s Far from being isolated instances, ni-
trosamine contamination is a systemic
most widely prescribed drugs is driving efforts to issue that manufacturers of all types and
sizes need to take seriously. Although
better detect, control, and prevent their generation more is being learned about how it oc-
in APIs and finished drug products. curs, there are still gaps in that knowledge,
says Ron Najafi, CEO of Emery Pharma.
S
ince traces of the compounds were quate manufacturing and quality testing Emery is testing a number of pharmaceu-
first found in batches of APIs practices. In the case of the first contami- ticals in an attempt to clarify root causes
used to make valsartan and other nated API, for valsartan, one supplier had of such contamination. The company has
sartan angiotensin receptor blockers tweaked its synthesis process to improve already developed and validated new an-
(ARBs) in 2018, drug manufacturers yield and efficiency. Not only the man- alytical methods for such tests.
have been paying much closer attention ufacturer but the regulators who okayed Later in 2020, The United States Phar-
to nitrosamines —in particular NDMA that change failed to realize that it would macopeial Convention (USP) will release
(N-Nitrosodimethylamine)—and how result in contamination (1). new guidance to help manufacturers bet-
they enter the pharmaceutical supply But ARBs were only the first sign of ter face, address, and prevent nitrosamine
chain. “Nitrosamines are extremely trouble, and, for one drug (so far), the contamination of pharmaceuticals and
carcinogenic, as academic researchers problems go much deeper than manu- APIs. Included will be an informational
have known for decades. The fact that facturing and quality control. In 2019, chapter, as well as a series of six validated
NDMA is used to this day as a control extensive testing found that ranitidine, reference standards that can be used as
to induce cancer in rats just underscores one of the most widely prescribed controls and in method validation, says
MOTORTION - STOCK.ADOBE.COM
its immense carcinogenic potential,” treatment for ulcers and acid indi- Jaap Venema, chief scientific officer with
says David Light, CEO of Valisure, gestion (whether as the trade-named USP. Venema believes that decisive action
an analytical pharmacy that filed the Zantac or as one of many generic and is required now, since the contaminant
first citizen’s petition on nitrosamine over-the-counter brands), was also con- has been found in three very different
contamination in ranitidine with FDA. taminated with NDMA. Valisure and types of products from very different
Some cases of contamination have Emery Pharma filed citizen’s petitions sources. Longer term, he says, more work
been traced to solvents and to inade- with FDA in 2019 and 2020 detailing is needed in assessing the risk, not only
46 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
A SPECIALTY CDMO TRANSFORMING
ORAL FORMULATION CHALLENGES
INTO PRODUCT SOLUTIONS
R I G H T D R U G , R I G H T P AT I E N T, R I G H T F O R M
As a speciality CDMO, Adare provides extensive expertise and advanced knowledge to guide your project
from product development through commercial manufacturing. We specialize in ODTs, taste masking, and
customized drug release for the Pharmaceutical, Animal Health, and OTC markets. With over 30 years of
proven legacy, Adare has successfully developed and manufactured more than 40 products sold by partners
in more than 100 countries globally.
LEARN HOW ADARE PHARMACEUTICALS CAN MAKE YOUR PRODUCTS MORE PATIENT-CENTRIC
ADAREPHARMA.COM | BUSDEV@ADAREPHARMA.COM
Event Overview
Formulators must consider many variables in the formulation Presenters
of oral solid dosage form (OSDF) drugs. Digitalization can
help formulators of OSDF address certain challenges they Brett Burns
typically face in their formulation journey. Global Strategic
Marketing Manager
In this webcast, experts will review some of these challenges
and the science behind ZoomLabTM, an online formulation BASF Pharma Solutions
optimization and prediction tool available to all formulators,
to help reduce significant amounts of trial and error during Dr. Ferdinand Brandl
the formulation process. A demonstration of the tool’s Head of Laboratory
ability to predict optimal formulation for direct compression, BASF SE, Nutrition and
including aspects of processability, tabletability, and content Health, Development
uniformity—using excipients from a range of companies— Pharma Solutions
will be featured.
Moderator
Key Learning Objectives
• Learn about the variables that effect the optimization of
Rita Peters
an OSDF formulation Editorial Director
• Learn how a digital tool accounts for these variables to
Pharmaceutical Technology
optimize an OSDF formulation
• Learn how a digital tool can make formulating an OSDF
more efficient Sponsored by
Presented by
P
articulate matter by light obscu- be inaccurate (such as an instrument enough that they are not mitigated
ration (LO) testing quantifies undercounting clear particles due to prior to generating new issues. In such
particles in suspension by their clarity issues or under-sizing aspheri- cases, check to see if a different sen-
projected shadow onto a photo sensor. cal particles due to shape issues). This sor type may be utilized (likely still in
This is the preferred method of sub-vis- article discusses both cases, with a conjunction with dilution). Manufac-
ible particle quantification (generally focus on those where inaccurate results turers typically have multiple types of
< 100 µm), as it has a high testing are generated because the added risk sensors available, each with their own
throughput and robustness compared of the inaccuracy was not discovered. specialty, that may be of aid. However,
to other means, such as membrane or if dilution cannot be made sufficient,
flow imaging microscopy. However, Usable results particles may need to be quantified by
not all preparations are suitable for unable to be generated another technique.
testing by LO. In some cases, the issues The most apparent symptom of a vis-
may prevent usable results from being cosity issue is the instrument’s inability Inaccurate results generated—
generated (such as an instrument being to pull solution at the predefined flow variability by instrument
unable to maintain a flow rate due to rate. This almost always prevents results Contrast limitations affect preparations
GREEN WIND - STOCK.ADOBE.COM
viscosity issues or unable to measure from being generated, though in bor- exponentially as the refractive indices
particles in an opaque solution), and in derline situations the solution may still of particles and solution approach each
other cases the results generated may flow through the instrument and simply other, which can lead to otherwise iden-
generate errors at a higher-than-normal tical instruments generating noticeable
Brent Denton is scientist, John Bak
is principal scientist, and Jonathon frequency. Errors should be reviewed differences in their results due to slight
Salsbury is associate director; all at PPD carefully to determine if they are related variation in their construction/assembly.
Laboratories GMP Lab. to an increased load on the motor (or Such an unexpected variation should be
52 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
Practical Considerations
for Bringing Drug
Candidates to Phase I
Clinical Trial
L IVE WE BCAST Register for this free webcast at:
www.pharmtech.com/pt_p/practical_considerations
Wednesday, May 27, 2020
11am EDT | 8am PDT | All attendees will receive a FREE executive
4pm BST | 5pm CEST summary of the webcast!
Presenters
Event Overview
New drug innovators face key challenges to advance
their compound to clinical trials. An investigational
new drug (IND) submission is a key milestone for every
Philip Jones, PhD drug candidate. This webcast will discuss the best
Vice President, Therapeutics Discovery
MD Anderson Cancer Center
practices to advance small-molecule drug candidates
quickly from discovery to IND filing, including safety
assessment and chemistry, manufacturing, and
controls (CMC) programs with real-world case studies.
Figure 1. Size distributions of a single standard preparation across three different light obscuration sensors of the same
manufacturer and model. Each curve represents an (n=3) mean. All replicates were sampled in staggered order to reduce bias.
evaluated to determine if it can be at- for particles in advance. Preliminary counter A, though otherwise identical
tributed to an instrument error, or if it testing can be performed by slightly al- to counter B and counter C, must have
may be a contrast issue. The contrast of tering the solution refractive index, and a different sensitivity to contrast issues.
a particle in suspension is defined by the thus particle contrast, between analyses
absolute value of the difference in refrac- and checking for an amplified differ- Inaccurate results generated—
tive indices (Δn) between the particle and ence in particle counts. variability by flow rate
surrounding fluid. Particle counts begin To illustrate how instrument vari- An aspherica l par ticle ca n f low
to be noticeably affected when Δn is less ability can be attributed to contrast is- through an instrument sensor either
than approximately 0.10, with more sig- sues in even a relatively mild case, see in line with the surrounding laminar
nificant effects when Δn is less than ap- Figure 1. Three instruments equipped flow and appear smaller than actual-
proximately 0.05 (3,4,5). with the same model of sensor that ity, or be rotating, or “tumbling,” and
If the chemical composition of a would typically be considered identi- appear larger than actuality (6). If re-
particle population is known, the bulk cal are compared, with one producing sults are sensitive to flow rate, it may
refractive indices of each may be used noticeably different results. A 30-µm be a symptom of aspherical particles,
as rough approximations to determine standard preparation (Thermo 9030 as increased flow rate can increase the
if contrast issues are likely. While the glass standard: 29.5 µm diameter, 2.44 likelihood of tumbling. It is also likely
refractive index of a fluid often can g/cm 3 density, 1.52 refractive index) that different instrument types affect
be easily determined or well-approxi- was suspended in a water–glycerol the incidence of tumbling. Sensitivity
mated, a particle’s refractive index can solution (Δn of approximately 0.11) to flow rate (as a symptom of aspher-
ALL FIGURES COURTESY OF THE AUTHORS.
be more challenging and laborious to and analyzed on each LO counter ical particles) may appear subtly as a
obtain. Additionally, a single refractive (HIAC Royco Model 9703 or 9703+ broadened size distribution, but may
index is rarely representative: even if a Liquid Particle Counting System with a indicate that further evaluation is
particle population is both isotropic and United States Pharmacopeia [USP] <788> needed. While certain other types of
chemically homogeneous, the effective calibrated HRLD 400 sensor). Com- particles, such as proteins, are also
refractive index may be a function of pared to counters B and C, the results sensitive to flow rate, evaluation for
particle size due to edge effects. Because obtained from counter A were subtly, aspherical particles should always
of this, it is advised to reactively test for though noticeably, different: both the be performed if they are at all a con-
any suspected contrast issue rather than mean and standard deviation of the sideration. Depending on the level of
determining refractive index value(s) particle distribution increased. Thus, accuracy required, an additional cal-
54 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
Figure 2. Example light obscuration sensor calibration curve, split into two sections, in mV (electric response of photo-
sensor) vs. µm (diameter [d] of the standard used). Figure 2a shows the low range response, which best fits to an
exponential. Figure 2b shows the high range response, which best fits as linear. Sensor geometry dictates a thin cross-
sectional slice of fluid be analyzed—approximately just 10 µm in some cases. While a sensor can be calibrated to much
larger particle sizes, the curve itself changes form as particles (and resulting displaced volume, proportional to shadow
intensity) shift from being fully visible in a single view (shadow intensity as a function of d cubed), to only a cross-section
being visible at a given moment (shadow intensity as a function of d squared). Due to photo-sensor output voltage being
proportional to the square root of shadow intensity, for standards at and below 15 µm, the electrical response is a direct
exponential of approximate order 1.5, and for standards at and above 25 µm, the response is linear order to reduce bias.
ibration curve or other means of cor- To illustrate how the apparent size of cause any additional issues. If results
relation may be warranted. However, an aspherical particle is impacted by its generated appear inaccurate, further
any LO result generated with an ad- orientation, see Figure 2. It is likely that evaluation should be performed spe-
ditional calibration curve should only only a portion of a particle is within the cific to the situation. When diagnos-
be considered a worst-case scenario, as sensor’s field of view at any given time, as ing any symptom, it is important to be
counts and sizes may be biased high LO instruments are configured to reduce aware of both the chemical and physi-
for particles less aspherical than those particle coincidence by only viewing a cal characteristics of any expected par-
represented by the additional calibra- thin slice of solution (increasing the max- ticles, as they can give critical insight.
tion curve. imum analyzable particle concentration). Symptoms that generate inaccurate
The simplest and easiest way to As the field of view begins to approximate results should be evaluated especially
check for aspherical particles is micros- a cross-section for larger particle sizes, carefully due to the risk associated
copy, though any technique that incor- the apparent size of an aspherical par- with their potential to be overlooked.
porates visual analysis may be appli- ticle will become orientation dependent. In most cases, measures can be taken
cable. Observed particles do not need An aspherical particle that stays in line to ensure testing per LO is suitable,
to be highly aspherical to indicate an with the surrounding laminar flow will though in certain situations other
impact on results, as symptoms may be appear relatively small, while a tumbling techniques should be considered.
noticeable for particles with an aspect aspherical particle may pass through the
ratio (the ratio of the minor to major sensor sideways and appear much larger. References
axis) up to approximately 0.5, with 1. L.O. Narhi, et al., J Pharm Sci 104, 1899–
more significant symptoms appearing Conclusion 1908 (2015).
2. T. Werk, D.B. Volkin, and H.C. Mahler,
if aspect ratios are approximately 0.25 In summary, a variety of issues can Eur J Pharm Sci 53, 95–108 (2014).
or less. However, aspherical particles arise while performing LO testing, 3. Z. Hu and D.C. Ripple, J Res Natl Inst
typically only noticeably impact results causing symptoms that either make re- Stand Technol 119, 674–682 (2014).
when they are a large portion of the sults unable to be generated, or gener- 4. P. Oma, D.K. Sharma, and D. King, USP
particle population. Therefore, if they ate inaccurate results. If results cannot Pharmacop Forum 36, 311–320 (2010).
5. D.C. Ripple and Z. Hu, Pharm Res 33,
are not observed upon visual analysis it be generated, dilution is typically an 653–672 (2016).
is highly unlikely they are a root cause appropriate next step, though it must 6. R.E. Cavicchi, et al., J Pharm Sci 104, 971–
of any symptom. be done with care to ensure it does not 987 (2015). PT
Transfer of Bioprocesses to
in manufacturability too early, and then
the process performance changes in the
development stage or when moving a
Outsourcing Partners
program from one site to another. Estab-
lishing expectations for process perfor-
mance with limited data can, Hastings
says, often over-constrain the develop-
ment team, leading them to necessarily
Cynthia A. Challener
forego improvements in order to achieve
a manufacturability specification.
Experience, communication, collaboration, Change management during process
development, meanwhile, can challenge
transparency, planning, and prioritization both sponsors and CDMOs as they work
contribute to success. to expedite processes. “In this day and age
of accelerated processes and ever-chang-
T
echnology transfer of bioprocesses,
tiple challenges that can include meeting ing team members, capturing rationale for
while common in the biopharma- the material demands of their clinical change becomes increasingly important
ceutical industry, can be complex
or commercial strategy and achieving to guard against knowledge loss as team
and present numerous challenges. Proj-aggressive timelines. These constraints members leave a company and to help
must be met while accommodating the
ects with accelerated development time- guide development in an unemotional
lines that also involve multiple contract
facility fit and scale-up needs determined manner,” Hastings asserts.
service providers have the potential to
by both the established process and the While commercial processes typically
magnify the difficulties. Both contract
CDMO’s platform, equipment, and ca- have robust quality and change-con-
manufacturing organizations (CMOs)/ pabilities, according to Frank V. Ritacco, trol systems, earlier-phase projects can
contract development and manufactur- director of scientific and technical affairs often change quickly and with limited
ing organizations (CDMOs) and sponsor for pharma services at Patheon, part of traceability. “We recommend taking a
companies can facilitate the process Thermo Fisher Scientific. balanced approach in order to capture
using pragmatic approaches that mit- Abel Hastings, director of process pertinent information while not hinder-
igate risks and ensure cooperation be-sciences at Fujifilm Diosynth Biotech- ing development,” comments Hastings.
METAMORWORKS - STOCK.ADOBE.COM
tween all parties involved. nologies, adds there are some themes Some sponsors use iterative risk assess-
that often slow the progression of ments or failure mode and effects anal-
Material, specification, and sponsor projects, including materials ysis tools to quickly capture changes
change-management issues challenges, specification ambiguity, and build catalogs of development and
When transferring a bioprocess to a and challenges with historical change lifecycle changes. “This approach al-
CDMO, manufacturers often face mul- management. “The implications of an lows the user to both convey historical
Cynthia A. Challener, PhD, is a ill-selected material can linger and details and to project forward-looking
contributing editor to Pharmaceutical hinder manufacturability. Once se- risks in order to steer development in a
Technology. lected, materials dogma can be diffi- planned way,” Hastings observes.
56 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
Phase-related challenges risks,” she notes, “can be minimized by
to Ritacco. “Fast-tracking the timeline
The overall challenge in tech transfer is and schedule without compromising
having a good depth of knowledge of the
to ensure that the sponsor’s process will product quality and patient safety is
facility’s specific equipment and well-es-
be reproduced in a similar, robust, and tablished scaling models.” an additional challenge,” Bulpin says.
compliant way. Manufacturing product Variations in equipment and material While some product development
in the quality and quantity in the defined steps can be reduced, Bulpin insists that
handling at facilities can complicate the
timeframe are indicators of a successful tech transfer process, agrees Bulpin. “If
risk analysis and decision justifications
transfer of the manufacturing process to the target facility uses different equipment,
should not be underestimated. “Sufficient
the CMO/CDMO, according to Andrew the raw materials must be compatible understanding of the process is necessary
Bulpin, head of process solutions at Mil- to ensure control of operating parameters
with that equipment,” he explains. In ad-
liporeSigma. “This overall challenge,” he to consistently deliver safe and efficacious
dition, to ensure a successful scale up, it is
says, “is mainly linked to the process itself, critical that raw materials are available at
product to patients, and achieving this on
but it also relates to the capacity to trans- an accelerated timeline requires a skilled
large scale and equipment capacity is eval-
fer the process knowledge built for several team leveraging well-established pro-
uated. “The scale-up strategy of the CMO/
years to the CMO/CDMO that will be inte- CDMO, as well as its expertise in various
cesses and methods, as well as past ex-
grated to ensure a successful tech transfer.” process types, molecules, and innovative
perience to streamline development and
It often also varies depending on the de- process methods—including perfusion, avoid pitfalls,” adds Ritacco.
velopment phase of the bioprocess being It is also extremely important for spon-
precipitation, and single-pass tangential
transferred. Sponsor companies’ ear- sor companies to work with a partner
flow filtration—is also key,” asserts Bulpin.
ly-phase programs are typically challenged who has extensive experience with tech
For cell-culture processes in particu-
by transferring processes that do not fit transfer and scale up into commercial
lar, scale up can be challenging because
into their external partners’ platform launch, according to Schirmer. “A typical
scaling to larger volume is not linear, and
operations, according to Emily Schirmer, program with an external partner is de-
multiple engineering parameters must be
senior director of process development at signed to have a process-acquisition phase,
optimized to match process performance
Catalent Biologics. “Those partners that developed at the bench scale and con-which includes a paper transfer followed
can offer a high level of flexibility and unit by establishing the process at a small scale.
firmed at pilot to manufacturing scales
operation expertise are more successful Then the scale up occurs within the fa-
of 2000 L and beyond, according to Ri-
in accommodating unique or non-tradi- cility infrastructure. These scale-up ac-
tacco. Harvest and downstream purifica-
tional processes,” she remarks. tivities can, however, be minimized by
tion processes also need to be scaled up to
In addition, as programs progress performing the process acquisition at the
match product quality and impurity pro-
through clinical stages, process transfer manufacturing scale to meet accelerated
files, and sometimes adapted to address
becomes more difficult because there timelines,” she says.
differences in product titer and impurity
is less inherent flexibility allowed and profiles observed at larger scale. Fujifilm often observes three partic-
process changes or adaptations become ular challenges with accelerated pro-
Even if projects don’t involve scale-up,
more challenging to justify, Schirmer grams: prioritization, collaboration and
generally they still require a thorough
adds. Bulpin agrees that the process pa- communication, and planning. Most
investigation of which parameters need
rameters in Phase I or II can be slightly sponsors with accelerated designations
adjustment to fit best within the CMO/
adapted or adjusted to ensure a successful CDMO facility and equipment con- greatly reduce timelines, Hastings says,
tech transfer, while in Phase III the pro- by focusing on refining their processes
straints without impacting specifications,
cess is more locked with defined critical to reliably achieve CQAs, often at a cost
Hastings observes. “In our view, scale
process parameters, critical raw materials, to the yield. “The most successful proj-
changes and facility-fit adjustments are
and designated critical quality attributes ects have a cross-functional team that is
not complications but rather just part of
(CQAs). Thus, any changes in Phase III what we do,” he says. clear on their priorities and what success
should be assessed, justified and, if nec- looks like. Ensuring all members of the
essary, validated. Accelerating complications team work together to settle on the right
Many new biologic drug candidates balance of yield and CQA reliability will
Scale-up, equipment, are awarded some form of accelerated ensure teams aren’t derailed by conflict-
and parameter adjustments approval designation. While acceler- ing priorities,” observes Hastings.
Adding in a scale-up element when trans- ated approval designations may allow One problem Hastings has observed,
ferring a process increases the overall risk a faster path to the clinic or to com- however, is that as teams turn attention
for project success and comparability. “It mercial approval for breakthrough to accelerating development timelines,
is difficult to discern the differences ob- therapies, the time it takes to develop a they have a tendency to unintentionally
served from facility to facility and any bioprocess that will be robust, scalable, reduce the circle of communication. “The
disparities that can be observed on scale and reproducibly deliver target product most successful teams, however, enlarge
up independently,” Schirmer says. “These quality remains significant, according their circle to engage all functional areas,
Pharmaceutical Technology MAY 2020 57
Outsourcing
allowing everyone to be part of the pro- Hastings also notes that some pro- with or without scale up—as an entire
cess of building the best balance of speed, cesses include historical elements that project with dedicated subject matter
reliability, and risk,” he asserts. are difficult to translate into and out of experts, communication f low, clear
Planning for an accelerated designation batch records. “In some cases, sponsors responsibilities, and planning. Dedi-
project, meanwhile, requires a balance of deal with this issue by maintaining some cating one individual to the manage-
creativity, dedication, and pragmatism, of the same staff throughout transfers to ment of one CMO/CDMO group helps
according to Hastings. “Teams need to be minimize the chances that so-called tribal with this approach, except for quality
able to challenge their platform expecta- knowledge is lost,” he observes. assurance and regulatory topics, which
tions to strike a balance to improve time- Transparency between all team Bulpin says should be managed glob-
lines. This planning cannot be done once, members, adds Schirmer, can actually ally to ensure process compliance.
but rather needs to evolve as the scientific be the catalyst to a successful devel- The preparation and review of thor-
understanding of the process grows and opment, scale up, and tech transfer of ough process descriptions, facility fit,
teams need to be aware that any new data a drug with an accelerated approval and tech transfer documentation, with
might require an adjustment to the project designation. “Whether the process appropriate quality review and over-
plan,” he adds. technology transfer occurs between sight, go a long way toward achieving
the sponsor company and the exter- a successful tech transfer, adds Ritacco.
Engaging with multiple nal partner, or between two external The experience and expertise of the
outsourcing partners partners, the importance of transpar- teams involved is also critical. “A highly
For many projects, it is necessary to ency is the same. Communication is skilled and experienced team will antic-
transfer a bioprocess from one CMO/ also critical. If all parties are aligned ipate potential challenges and leverage
CDMO to another. Regardless of the rea- and understand the path forward, it is their deep understanding of bioprocess
son, it is essential that everyone involved much more likely that the transfer will development, tech transfer, and manu-
does not overlook the complexities as- be a success,” she comments. facturing to achieve the desired process
sociated with leveraging previous data The general constraints are the same performance and product quality in a
and maintaining technical continuity, for any tech transfer process, including new facility, even when the timeline
according to Hastings. those between outsourcing providers, may be challenging,” he asserts.
To proactively address this issue, Fu- although the confidentiality becomes It is also important, according to
jifilm identifies new, unique, or difficult more marked and the training process Hastings, for everyone involved in a
(NUD) elements, such as process-specific can be slightly different. “The aim re- tech transfer project to recognize up-
factors that depend on equipment idio- mains the same,” Bulpin says. “In ad- front that priorities may change during
syncrasies that may not be fully evident dition, any process, analytical, equip- the development cycle for biologics
to a sponsor while the process is running ment, raw material, and/or regulatory with accelerated approval designa-
at a CDMO. “When a process is moved gaps should be listed and mitigated to tions. Risk-based tools can be used to
from one CDMO to another, the receiving meet specifications and ensure a suc- segment process characterization into
site rarely has access to complete equip- cessful tech transfer.” two or more phases and to design the
ment-specific details, which can make in- process to prioritize CQAs over yield.
terpretation of previously generated data Minimize risk for optimum success “Using this approach, it is possible for
challenging and incur risk,” Hastings says. Technology transfer introduces additional sponsors to start their process perfor-
“At Fujifilm, we work hard to identify risk to the development and commercial- mance qualification (PPQ) in parallel
NUDs as early as possible and then cate- ization of biopharmaceuticals, so mini- with some process characterization
gorize them as equipment-specific versus mizing risk is fundamental to successful work, thereby shortening the timeline
equipment-agnostic. This approach allows technology transfer of bioprocesses. from clinical manufacturing to PPQ,”
us to suggest studies that may improve fa- A clear project plan with milestones Hastings observes.
cility fit,” he adds. that are mutually agreed upon sets On the other hand, it can result in
Differences in equipment throughout the stage for a successful program, post-approval yield fluctuation, which
the process, as well as differences in scale, Schirmer observes. “Risks should be makes business-planning difficult. Fu-
require optimization of multiple engineer- identified at the onset of the project jifilm see this issue as an opportunity to
ing parameters to match process perfor- and continuously re-evaluated to de- move through a natural transition point.
mance in cell culture, harvest, and down- termine effectiveness of mitigations “Purposefully pivoting from biologic li-
stream purification processes between and any additional risks,” she says. In cense application-enabling activities
facilities, Ritacco agrees. He also points addition, risk assessments should be to business-enabling process improve-
out that analytical methods have to be backed by scientific reasoning and ments at the optimum time will allow
transferred, established, qualified, and ul- process data. programs to benefit from both accel-
timately validated to assure consistency in The main strategy, according to erated PPQ and long-term commercial
analytical results between manufacturers. Bulpin, is to consider tech transfer— pay-back,” he concludes. PT
58 Pharmaceutical Technology MAY 2020 P h a r mTe c h . c o m
pharma capsules
I N T E R N A T I O N A L C E N T R E F O R D I F F R A C T I O N D A T A
PDF-4/Organics
Standardized Data
Solve Difficult
More Coverage Problems,
Get Better Results
and Characterization
Room temperature crystal structure publication for Atazanavir, a
potential drug for treating COVID-19, is available from International
Centre for Diffraction Data, ICDD ® Targeted Molecular
Collection for Graphics and
Commercial & Digitized
Read the Powder Diffraction Journal paper for Free Now
Regulatory Patterns
via Cambridge CoreShare: Fields
https://doi.org/10.1017/S0885715620000135
Mixing/Blending/Drying
9150.pgs 01.10.2018
magenta
yellow 01:49
cyan
black ADVANSTAR_PDF/X-1a
PT-HouseAd-2020-onefourth.indd 1 4/29/2020 4:18:04 PM
Ad Index
COMPANY PAGE COMPANY PAGE COMPANY PAGE
Adare Pharmaceuticals................................. 47 FG Clean Wipes.............................................. 29 STERIS Corp. Life Sciences Marketing.... 25, 26
BASF................................................................ 49 International Centre for Diffraction Data..... 60 Syntegon Pharma Technology, Inc............... 27
through temporary voluntary engagements, placements, or experts. Pharmaceutical companies probably look more for
positions is what helps improve the chances for long-term or experts in a particular subject, whereas service providers,
permanent employment. And don’t forget to network through such as consultancies or contract research organizations, may
portals such as LinkedIn, Bing, or similar sites that have a good have a need for experts with more varied backgrounds.
reputation with industry and job agencies. Also, it’s important We may not always find the job we want, but we can always
to write a succinct and well-thought-out curriculum vitae, and learn from what we do, and it will always be a beneficial
there is a lot of great advice available for free online on how to personal and job experience. PT
Attend the Conference virtually and access all the high-quality content and presentations
you’ve come to expect from your home or office.
NOW
VIRTUAL
JULY 20-22
EXHIBITION: JULY 20-22
TRAINING COURSES: JULY 23-24
#PDAAnnual
supplying clinical trials of the future.
ready today, innovating for tomorrow.
introducing onebio® suit e
Small Molecule & Biologics | Forecasting & Planning | Manufacturing | Packaging & Labeling | Logistics & Cold Chain | Comparator Sourcing