Pharmaceutical Technology eBook Solid Dosage Drug Development and Manufacturing

eBOOK SERIES

2016

SOLID DOSAGE DOSAGE DRUG DEVELOPMENT DE VELOPMENT AND MANUFACTURING

SOLID DOSAGE DRUG DEVELOPMENT AND MANUFACTUR MANUFACTURING ING 2016 2016 EXCIPIENTS

EDITORIAL Editorial Director Rita Peters [email protected] [email protected] om Senior Editor Agnes Shanley [email protected] Managing Editor Susan Haigney [email protected] Science Editor Adeline Siew, PhD [email protected] Manufacturing Editor Jennifer Markarian jmarkarian@advan [email protected] star.com Science Editor Randi Hernandez rhernandez@advanstar [email protected] .com Community Manager Caroline Hroncich chroncich@advanstar [email protected] .com Art Director Dan Ward Contributing Editors Jill Wechsl jwechsler@adva er@advanstar nstar.com .com;; Jim Mille Wechsler er jwechsl Millerr [email protected]; Hallie Forcinio [email protected]; Susan J. Schniepp [email protected]; Eric Langer [email protected]; and Cynthia A. Challener, PhD [email protected] Correspondent Sean Milmo (Europe, [email protected]) 485 Route One South, Building F, Second Floor, Iselin, NJ 08830, USA Tel. 732.596.0276, Fax 732.6 47.1235, 1235, PharmTech.com

SALES Publisher Mike Tracey [email protected]

4 Excipient Quality and Selection Irwin B. Silverstein EXCIPIENT DATABASES

16 Unifying Excipient Databases Agnes Shanley NON�GELATIN CAPSULES

20 Establishing a New Performance Standard for HPMC Capsules Agnes Shanley PROCESS SIMULATION

22 Modeling and Simulation Move Downstream Agnes Shanley

Mid-West Sales Manager Irene Onesto [email protected] East Coast Sales Manager Joel Kern [email protected] European Sales Manager Linda Hewitt [email protected] European Senior Sales Executive Stephen Cleland [email protected] Executive Assistant Barbara Sefchick [email protected]

TABLET COMPRESSION

26 Optimizing Tablet Compression Frederick J. Murray CONTINUOUS MANUFACTURING

Address 485 Route One South Building F, Second Floor Iselin, NJ 08830, USA

34 Filling the Analysis Gap in the Move to Continuous Processing Jamie Clayton

Tel. 732.596.0276, Fax 732.647.1235

PharmTech.com

ELEMENTAL IMPURITIES

40 Meeting USP Guidelines for Sr. Production Manager Karen Lenzen International Licensing Maureen Cannon [email protected], tel. 440.891.2742 or toll-free 800.225.4569 ext 2742 2742,, fax. 440.756.525 44 0.756.52555 Audience Development Manager Rochelle Ballou [email protected]

Elemental Impurity Analysis with X-ray Fluorescence Fluorescence Spectrometry Andrew Fussell GENERIC DRUGS

Pharmaceutical Technology does not verify any claims or other information appearing in any of the advertisements contained in the publication and cannot take any responsibility for any losses or other damages incurred by readers in reliance on such content. Pharmaceutical Technology welcomes unsolicited articles, manuscripts, photographs, illustrations, and other materials but cannot be held responsible for their safekeeping or return. UBM Life Sciences provides certain customer contact data (such as customers’ names, addresses, phone numbers and e-mail addresses) to third parties who wish to promote relevant products, services, and other opportunities which may be of interest to you. If you do not want UBM Life Sciences to make your contact information available to third parties for marketing purposes, simply call toll-free 866.529.2922 between the hours of 7:30 am and 5 pm CT and a customer service representative will assist you in removing your name from UBM Life Sciences’ lists. Outside the United States, please phone 218.740.6477. To subscribe: Call toll-free 888.527.7008. Outside the US, call 218.740.6477. Single issues, back issues: Call toll-free 800.598.6008. Outside the US call 218.740.6480. 218.740.6480. Reprints of all articles in this issue and past issues of this publication are available. Call 877-652-5295 ext. 121 or email bkolb@wr ightsmedia.com. Outside US, UK, direct dial: 281-41 281-419-5725. 9-5725. Ext. 121. Direct mail lists: Contact Tamara Phillips, Marketing Services, tel. 440.891.2773, [email protected]. Display, Web, Classified, and Recruitment Advertising: Contact Tod McCloskey, tel. 440.891.2739, [email protected]. Permissions: Contact Maureen Cannon, tel. 440.891.2742 or toll-free 800.225.4569 ext 2742, fax. 440.756.5255, [email protected].

46 Regulatory Considerations for Controlling Intermediates in Type-II Drug Master Files for the Manufacture of Generic Drug Substances Kandasamy Subburaj, Brian T. Connell, Srinivasa Murthy, Humcha Hariprakasha, Deborah F. Johnson, Huyi Zhang, and David J. Skanchy

54 Ad Index Issue Editor: Agnes Shanley. On the Static Cover: Science Photo Library/Getty Images; Dan Ward. On the Animated cover: Nanihta photography photography/DAJ/Tetra /DAJ/Tetra Images/Yuri_Arcurs/ Empato/E+/Image Empato/E+ /Image Source/Still Factory/Getty Images; Dan Ward. ©2016 UBM. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical including by photocopy, recording, or information storage and retrieval without permission in writing from the publisher. Authorization to photocopy items for internal/educational or personal use, or the internal/educational or personal use of specific clients is granted by UBM for libraries and other users registered with the Copyright Clearance Center, 222 Rosewood Dr. Danvers, MA 01923, 978-750-8400 fax 978-646-8700 or visit http:// www.copyright.com online. For uses beyond those listed above, please direct your written request to Permission Dept. fax 440-756-5255 or email: [email protected].

PharmTech.com

SOLID DOSAGE DRUG DEVELOPMENT AND MANUFACTUR MANUFACTURING ING 2016 2016 EXCIPIENTS

EDITORIAL Editorial Director Rita Peters [email protected] [email protected] om Senior Editor Agnes Shanley [email protected] Managing Editor Susan Haigney [email protected] Science Editor Adeline Siew, PhD [email protected] Manufacturing Editor Jennifer Markarian jmarkarian@advan [email protected] star.com Science Editor Randi Hernandez rhernandez@advanstar [email protected] .com Community Manager Caroline Hroncich chroncich@advanstar [email protected] .com Art Director Dan Ward Contributing Editors Jill Wechsl jwechsler@adva er@advanstar nstar.com .com;; Jim Mille Wechsler er jwechsl Millerr [email protected]; Hallie Forcinio [email protected]; Susan J. Schniepp [email protected]; Eric Langer [email protected]; and Cynthia A. Challener, PhD [email protected] Correspondent Sean Milmo (Europe, [email protected]) 485 Route One South, Building F, Second Floor, Iselin, NJ 08830, USA Tel. 732.596.0276, Fax 732.6 47.1235, 1235, PharmTech.com

SALES Publisher Mike Tracey [email protected]

4 Excipient Quality and Selection Irwin B. Silverstein EXCIPIENT DATABASES

16 Unifying Excipient Databases Agnes Shanley NON�GELATIN CAPSULES

20 Establishing a New Performance Standard for HPMC Capsules Agnes Shanley PROCESS SIMULATION

22 Modeling and Simulation Move Downstream Agnes Shanley

Mid-West Sales Manager Irene Onesto [email protected] East Coast Sales Manager Joel Kern [email protected] European Sales Manager Linda Hewitt [email protected] European Senior Sales Executive Stephen Cleland [email protected] Executive Assistant Barbara Sefchick [email protected]

TABLET COMPRESSION

26 Optimizing Tablet Compression Frederick J. Murray CONTINUOUS MANUFACTURING

Address 485 Route One South Building F, Second Floor Iselin, NJ 08830, USA

34 Filling the Analysis Gap in the Move to Continuous Processing Jamie Clayton

Tel. 732.596.0276, Fax 732.647.1235

PharmTech.com

ELEMENTAL IMPURITIES

40 Meeting USP Guidelines for Sr. Production Manager Karen Lenzen International Licensing Maureen Cannon [email protected], tel. 440.891.2742 or toll-free 800.225.4569 ext 2742 2742,, fax. 440.756.525 44 0.756.52555 Audience Development Manager Rochelle Ballou [email protected]

Elemental Impurity Analysis with X-ray Fluorescence Fluorescence Spectrometry Andrew Fussell GENERIC DRUGS

Pharmaceutical Technology does not verify any claims or other information appearing in any of the advertisements contained in the publication and cannot take any responsibility for any losses or other damages incurred by readers in reliance on such content. Pharmaceutical Technology welcomes unsolicited articles, manuscripts, photographs, illustrations, and other materials but cannot be held responsible for their safekeeping or return. UBM Life Sciences provides certain customer contact data (such as customers’ names, addresses, phone numbers and e-mail addresses) to third parties who wish to promote relevant products, services, and other opportunities which may be of interest to you. If you do not want UBM Life Sciences to make your contact information available to third parties for marketing purposes, simply call toll-free 866.529.2922 between the hours of 7:30 am and 5 pm CT and a customer service representative will assist you in removing your name from UBM Life Sciences’ lists. Outside the United States, please phone 218.740.6477. To subscribe: Call toll-free 888.527.7008. Outside the US, call 218.740.6477. Single issues, back issues: Call toll-free 800.598.6008. Outside the US call 218.740.6480. 218.740.6480. Reprints of all articles in this issue and past issues of this publication are available. Call 877-652-5295 ext. 121 or email bkolb@wr ightsmedia.com. Outside US, UK, direct dial: 281-41 281-419-5725. 9-5725. Ext. 121. Direct mail lists: Contact Tamara Phillips, Marketing Services, tel. 440.891.2773, [email protected]. Display, Web, Classified, and Recruitment Advertising: Contact Tod McCloskey, tel. 440.891.2739, [email protected]. Permissions: Contact Maureen Cannon, tel. 440.891.2742 or toll-free 800.225.4569 ext 2742, fax. 440.756.5255, [email protected].

46 Regulatory Considerations for Controlling Intermediates in Type-II Drug Master Files for the Manufacture of Generic Drug Substances Kandasamy Subburaj, Brian T. Connell, Srinivasa Murthy, Humcha Hariprakasha, Deborah F. Johnson, Huyi Zhang, and David J. Skanchy

54 Ad Index Issue Editor: Agnes Shanley. On the Static Cover: Science Photo Library/Getty Images; Dan Ward. On the Animated cover: Nanihta photography photography/DAJ/Tetra /DAJ/Tetra Images/Yuri_Arcurs/ Empato/E+/Image Empato/E+ /Image Source/Still Factory/Getty Images; Dan Ward. ©2016 UBM. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical including by photocopy, recording, or information storage and retrieval without permission in writing from the publisher. Authorization to photocopy items for internal/educational or personal use, or the internal/educational or personal use of specific clients is granted by UBM for libraries and other users registered with the Copyright Clearance Center, 222 Rosewood Dr. Danvers, MA 01923, 978-750-8400 fax 978-646-8700 or visit http:// www.copyright.com online. For uses beyond those listed above, please direct your written request to Permission Dept. fax 440-756-5255 or email: [email protected].

PharmTech.com

21ST CENTURY PHARMA A holistic, systems-based approach to drug development, manufacture and delivery

FORMULATION

MANUFACTURE DELIVERY

At PSE we are working with leading pharma companies to develop a holistic approach to drug development, manufacture man ufacture and delivery. We call it Systems-based Pharmaceutics (SbP). A systems-based approach considers the entire formulation-

Download the PSE, Pfizer and Eli Lilly SbP phase 1 progress report

The Advanced Process Modelling Company

to-delivery cycle as a single integrated system, relating formulation and manufacturing decisions directly to drug delivery KPIs. SbP is helping pharma companies to accelerate development, improve manufacturing, manufacturing, enhance drug product performance and increase R&D efficiency. Contact us to find out more. More online



psenterprise.com Operations in UK, USA, Switzerland, Japan, Korea, China, Taiwan, Malaysia, Thailand and UAE. UK Head Office t: +44 20 8563 0888 e: [email protected]

THE POWER TO BE CERTAIN

E��

Excipient Quality and Selection Irwin B. Silverstein

Choosing the right excipient manufacturer can help ensure the use of quality excipients.

H

ow does one define quality as applied to excipients? If

we pose the same question for APIs, the response would be to produce the ingredient under appropriate GMPs, and to the compendial monograph and the API assay.

Because the mono monograph graph provides the minimum min imum requirements, API quality is improved by reducing the presence of all materials other than the desired chemical. This is logical because, by definition, definition, the

API is “intended to furnish pharmacological activity activit y or other direct effect in i n the diagnosis, cure, mitigatio mitigation, n, treatment, or prevention prevention of disease or to affect the structure and function of the body” (1). Extraneous substances may be harmful to the patient in that they may lead to side side effects, or they are inert, iner t, thus reducing API purity and thereby compromising efficacy. Excipientt quality is described quite differently. Excipien differently. While Whi le one would again refer to compliance with the compendial monograph (if there is one) or the manufacturer’s specification, a higher assay is not always better. While this may seem counterintuitive, excipients are often complex complex mixtures mix tures that inclu i nclude de constituents arising from raw materials, catalyst, solvent, initiator residue, or side reactions. The International Pharmaceutical Excipients Council (IPEC) refers to these other unavoidable substances in the excipient as concomitant components (2). The performance of many excipients in the drug formulation may rely on the presence of such substances in the excipient. Concomitant components in the excipient may aid in solvating drug compo components, nents, improving excipient excipient functionality f unctionality,, etc. Excipient quality, therefore, is characterized as compliance to Irwin B. Silverstein, PhD, is a consultant to IPEC-Americas. 4

the monograph or specification and having a consisten consistentt concomitant composition.

Pharmaceutical Technology SOLID DOSAGE DRUG

DEVELOPMENT AND MANUFA MANUFACTURING CTURING 2016

PharmTech.com

D R A W N A D ; S E G A M I Y T T E G / Y R O T C A F L L I T S / E C R U O S E G A M I

Veltek Associates, Inc. offers two garment product lines, which are both pre-folded in our system. Comfortably styled and fitted with elastic thumb loops to reduce shifting, shifti ng, as well as tunnelized elastic wrists and ankles.

1600 Garments Breathable Comfortable High bacterial efficiency

1700 Garments High filtration efficiency Low particulate and shedding performance Excellent water repelency

Face Masks Breathable Reduces goggle fogging due to absorption efficiency Soft and comfortable

E�� Specified limits for excipients

Povidone is more likely than hydrazine to impact

As required by clause 8.2.4.6 of the ANSI excipi- performance of the excipient in some drug formulaent GMP standard, excipient manufacturers are

tions. A test method suitable for detecting low levels

expected to identify concomitant components

of this monomer was developed, and the compen-

present in the excipient whenever possible and to

dium was updated accordingly.

specify limits for those components that have been

An additional substance, 2-pyrollidone, was sub-

shown to be either important to excipient perfor- sequently found in Povidone. This substance is mance or known to have an adverse impact to the

formed during the polymerization reaction when

patient (3). Impurities known to be present in the

some vinyl pyrrolidone decomposes rather than

excipient are also required to have specified upper polymerizes. While it is unlikely that the presence limits based upon safety considerations, regulatory

of hydrazine or vinyl pyrrolidone beneficially im-

requirements, customer requirements, and, if ap- pacts the performance of Povidone in the drug forplicable, the compendium.

mulation, the same conclusion cannot be drawn for

Povidone and its monograph illustrate these

2-pyrrolidone. 2-Pyrrolidone is often used as a sol-

points. Povidone is the homo-polymerized mono- vent, and therefore, its presence in the excipient may mer vinyl pyrrolidone. It is sold in various mo- play a beneficial role in certain drug formulations lecular weights. In the late 1980s, GAF Chemicals, by helping to solvate the API. While it is possible to a manufacturer of Povidone, was made aware of

remove this substance through further processing, it

the presence of hydrazine, a toxic substance, in

is not feasible for the manufacturer to asses the im-

Povidone. The company identified the mechanism

pact on performance for all drug formulations that

of hydrazine formation as a by-product of the po- use Povidone. Therefore, it is important to control, lymerization reaction. Through modification of

but not limit, the quantity of 2-pyrrolidone so that

the process, the level of hydrazine was reduced

the performance of each lot of Povidone is consis-

to what was deemed acceptable for safe use of Po- tent in the various drugs that use this excipient. vidone in pharmaceuticals. Because Hydrazine is not expected to be beneficial in Povidone, it

Non-homogeneity

is thus considered an undesirable component. A These examples with one excipient illustrate how test method was developed, appropriate specified

control of all the components in the material are

limits were established, and the compendium was

needed in order to assure consistent quality. An-

updated accordingly.

other aspect that needs equivalent control is the

In the early 1990s, vinyl pyrrolidone was iden- degree of homogeneity of solid excipients, partified as a suspect carcinogen. As a consequence, ticularly those supplied in powder form. However, manufacturing methods were again modified

many excipients are also manufactured in much

to reduce the level of residual vinyl pyrrolidone

larger volumes for other markets where a larger

monomer to a toxicologically safe level. Vinyl pyr- degree of variation is tolerable. rolidone has solvating properties and is a reactive

To illustrate a common cause of non-homoge-

molecule. Therefore, residual vinyl pyrrolidone in

neity, consider that excipient manufacture often

6


DEVELOPMENT AND MANUFACTURING 2016

PharmTech.com

Advancing Drug Delivery Through Superior Polymer Design Safe and effective delivery of today’s complex drug molecules is increasingly a function of collaboration between polymer experts and pharmaceutical product development teams. Acting on enabling scie nce, producing superior polymer excipient technology, and creating systems for optimum bioavailability, Ashland works alongside formulators to advance drug-speci�c delivery platforms. Explore how you can solve intricate drug formulation and development challenges with polymer expertise from Ashland. Visit us at Ashland.com/pharmaceutical

Ashland [email protected] www.ashland.com/pharmaceutical ®Registered trademark, Ashland or its subsidiaries, registered in various countries ™Trademark, Ashland or its subsidiaries, registered in various countries ©2016, Ashland AD-13475

E�� involves drying the substance. The ability to

Common impurities in excipients, which are not

dry material to a consistent residual moisture

needed for excipient performance, may include

throughout the lot is inherently difficult due to

residual process aids, additives, by-products, and

the many operating variables. Spray drying is a

material that sheds from filter media. In addition,

case in point. Operating parameters for the spray

contaminants, which are to be avoided, can occur

dryer include the temperature and dew point of

from environmental factors such as personnel hy-

inlet air, burner temperature, concentration of

giene, equipment failure, contact with packaging,

the excipient in the aqueous solution, spray pat- etc. and include rust, oil, grease, insect fragments, tern of the excipient solution, rate of drying, and outlet air temperature. During a 24-hour cycle,

extractable and leachable materials, etc. Excipient quality is, therefore, best ex pressed as conformance to GMPs

The stability of the excipient can pose a risk if the material is likely to degrade during storage or shipment when temperature and/or humidity are not controlled within acceptable limits.

as well as to compendia or a specification and consistent composition, lot to lot. Consistent composition within each lot is also an expression of excipient quality, but oftentimes

the ambient air temperature and humidity may

such consistency is difficult to achieve without

differ considerably from day to night. Also the ex- a blending step. cipient concentration may vary due to prior man-

Generally, it is expected that a more consis-

ufacturing steps. Achieving a consistent moisture

tent excipient composition will result in a more

level requires frequent sampling of dried material

predictable performance in the final drug for-

and adjustment of spray-dryer operating param- mulation. In the selection of an excipient for a eters. As drying conditions become more severe

drug formulation, consideration should be to

in order to maintain constant residual moisture, include an excipient whose composition profile however, it is possible to cause some degradation

has known a nd tolerable variation with mini-

manifest as charring of the excipient. This is typi- mal number of concomitant components and cally manifested as burnt particles (4). Consistent

impurities.

moisture content in the excipient, therefore, may be a tradeoff with the quantity of burnt particles

Selection of excipient suppliers

in the product.

The European Union Directive Guidelines on the Formalized Risk Assessment for Ascertaining the

Excipient impurities

Appropriate Good Manufacturing Practice for Ex-

Excipient impurities are specific entities that

cipients of Medicinal Products of Human Use (5)

should not be present and/or need to be con- provides the following characteristics for assesstrolled for safety, toxicological, or other reasons. ing the manufacture and supply of excipients: 8



PharmTech.com

E�� Potential presence of transmissible spongiform

or molecular weight, therefore, can be considered

encephalopathy (TSE)

dedicated. Also equipment used to produce vari-

t

Potential for viral contamination

ous grades of an excipient that are then sold in dif-

t

Potential for microbiological or endotoxin con- ferent markets (e.g., food, cosmetic, or industrial

t

tamination

applications), but produced using the same chem-

t

Potential for the presence of impurities

istry and raw materials, should also be considered

t

Supply chain complexity and security

dedicated.

Excipient stability

Using dedicated equipment reduces the risk

t

Tamper-evident packaging.

that the excipient will be contaminated by the

t

Each of these characteristics can be related to

presence of other substances (e.g., other raw ma-

excipient “quality” and used to assess excipient

terial, intermediate, or finished product residue in

suppliers. Note that each of these considerations

the production equipment). Using multi-purpose

is in addition to manufacturing the excipient in

equipment relies heavily on verifying cleaning ef-

conformance to excipient GMP.

fectiveness and the ability to detect potential re-

In addition, the ANSI excipient GMP standard (3) sidual contaminants to assure the minimization highlights the following criteria to assess for risk to

of potential cross-contaminants in the excipient.

protect an excipient from contamination:

Where multi-use equipment is used, it is advisable

t

Hygienic practices: excipient contamination due

to review the excipient manufacturer’s cleaning

to personnel hygiene, illness, attire, unauthor- validation report. ized access, food, medication, tobacco, etc. t

When possible, it is preferable to source the ex-

Infrastructure, building: excipient contamina- cipient from a supplier that does not use animaltion, cross-contamination, mix-ups

t

derived raw materials at risk for bovine spongiform

Infrastructure, equipment: excipient contamina- encephalopathy (BSE)/TSE in the manufacture of tion due to material of construction, utilities, the excipient. Otherwise, the excipient user will water, process materials, and work environment

have to ensure the excipient presents minimal risk

(air handling, cleaning/sanitation, pest control

from TSE contaminants. A risk assessment should

and drainage).

include confirmation the animals used in the manufacture of the animal-derived raw material come

Minimizing contamination risk

from a country designated as negligible TSE risk.

Using an excipient manufacturer that produces Alternatively, the excipient manufacturer should the excipient in dedicated equipment is a lower

demonstrate that the animal-derived raw material

risk to excipient quality as a result of reduced risk

was processed under conditions that have been de-

of cross-contamination. Equipment can be con- fined to inactivate the TSE risk materials if present. sidered dedicated when it is used to manufacture

TSE risks are also present when the excipient is

products utilizing the same chemistry and raw

manufactured in multi-purpose equipment where

materials. Equipment used to manufacture an ex- the other products are animal derived. If there is a cipient in various particle size, density, viscosity, risk of TSE material residue on equipment, the ex10



PharmTech.com

Vcaps ® Plus Capsules

Not all capsules are created equal. Download Whitepaper

The new standard in HPMC capsules: Immediate release in-vivo performance bioequivalent to gelatin and consistent dissolution across changing pH and ionic strength. Vcaps® Plus capsules can optimize capsule disintegration for pharmaceutical products with the ability to release contents independent of pH and ionic strength of the test media. And a new in-vivo study demonstrates that Vcaps Plus capsules are equivalent to gelatin in terms of human pharmacokinetics profile. Finally, a superior specialty polymer capsule with proven gelatin-like performance without cross-linking potential, Vcaps Plus is the new capsule standard with benefits of narrow weight variability, impressive resistance to heat and humidity, and validated manufacturing experience. By reducing variability, Vcaps Plus capsules are quickly becoming a powerful new tool to reduce timelines in drug product development.

Contact us at 800-845-6973 for our full technical package and samples for your development review.

Proven Bioequivalence to Gelatin 8000 ) L m / g n ( n o i t a r t n e c n o C n e h p o n i m a t e c A a m s a l P n a e M

Molecule = Acetaminophen Human cross-over study in 24 subjects

7000 6000

Gelatin Capsules

5000

Vcaps Plus Capsules

4000 3000 2000 1000

0

2

Read the full report at VcapsPlus.com. © 2016 Capsugel Belgium NV All rights reserved.

8

10

ominal Hours (hrs) Post Dose

Mean in-vivo acetaminophen plasma concentration profiles over 12 hours.

True pH and Ionic Media Independence 100 90 80 d e v l o s s i d e n i e f f a C %

70 60

pH 1.2 USP pH 6.8 USP pH 6.8 JP2 Simulated milk fluid pH 1.2 – 2 g KCI/L pH 1.2 – 9 g KCI/L

50 40 30 20 10 0 0

New in-vivo results comparing Vcaps Plus HPMC Capsules and gelatin capsules demonstrate bioequivalency of HPMC in multiple dissolution profiles.

4

3

6 9 12 15 18 21 24 27 30 35 40 45 50 55 60 75 Time in minutes

In-vitro dissolution

of caffeine in Vcaps Plus capsules

12

E�� cipient manufacturer should demonstrate cleaning

borne particulate to reduce the risk of microbial

procedures that show residual TSE risk material is

contamination only requires use of a HEPA filter

either reduced to an acceptable level on the equip- if the excipient is purported to be sterile. ment surface or is inactivated.

Contamination of the excipient with undersir-

The risk of viral, microbiological, or endotoxin

able components can arise from such sources as

contamination arises from raw materials, water, nearby manufacturing operations, processing and the environment. Where the manufacture of

equipment (e.g., filters and traps), and utilities.

the excipient uses viral agents or there is a risk of

Filters pose a risk from shedding their material of

contamination with viral agents, adequate mea- construction and from traps that are improperly sures of sanitation or sterilization by the supplier

maintained, allowing trapped impurities through.

are expected.

Utilities such as nitrogen, compressed air, and steam may contami-

The complexity of the supply chain from excipient manufacturer to pharmaceutical facility is a consideration in selecting a supplier.

nate the excipient with impurities such as compressor oil and boiler additives. It is common for the

Excipient manufacturers should use at least pota- excipient to be produced at a site where many ble water where water is used in the process after the

other products are also manufactured. Some of

starting point for GMP or when water is a potential

these other products may be toxic (e.g., herbicides

source of microbial contamination in the finished

or pesticides) or they may use toxic ingredients

excipient. Water that is used for temperature control

in their manufacture. Where toxic substances are

that does not contact excipient during manufacture volatile enough to become airborne contaminants, poses minimal risk under normal operating condi- manufacturers should take appropriate measures tions and therefore need not be potable. For excipi- to minimize the risk of contamination. It is iments that are intended for drug products where the

portant for the user to assess the risk of airborne

presence of endotoxin poses a risk to patient safety

contamination and the measures taken to protect

and water comes into direct contact with the ex- the excipient during an onsite audit. cipient during processing, higher purity water such as United States Pharmacopieal Convention (USP)

Supply chain considerations

water for injection may be expected to be used.

The complexity of the supply chain from excipi-

There is also the potential for airborne microor- ent manufacturer to pharmaceutical facility is also ganisms to contaminate the excipient. Generally, a consideration in selecting a supplier. Although airborne microbes that can contaminate the ex- delivery from an excipient manufacturing site cipient can be controlled by filtering the air, such

directly to the pharmaceutical manufacturing

as when the excipient is exposed to the air during

facility provides the least opportunity for the ex-

packaging, to remove particles. Removal of air- cipient to become contaminated or tampered with 12



PharmTech.com

en route; generally direct delivery is uncommon

should verify through the paper trail that the ship-

and only applies to full truckloads of the excipi- ment of an excipient lot has come from the excipient. More often, less-than-truckload quantities are shipped by common carrier. Oftentimes, the ship-

ent manufacturer. The stability of the excipient can pose a risk if

ment goes from the manufacturer to a warehouse

the material is likely to degrade during storage or

of the transport carrier. There, the shipment may

shipment when temperature and/or humidity are

be cross-docked to a truck heading to the desired

not controlled within acceptable limits. Generally,

destination or another intermediate destination. excipients such as inorganic salts, minerals, modiWhile tampering with the excipient at the trans- fied food ingredients, and synthetic substances are port warehouse is unlikely, there is the possibil- stable materials. Also, many excipients have been ity for the packaged excipient to be exposed to

in commerce for an extended number of years and,

extremes of weather (temperature, humidity, and precipitation), for the packaging to be damaged through mishandling, or for the tamper-evident seal

Where delivery is not direct from the excipient manufacturer, the pharmaceutical company should periodically establish the pedigree of the excipient.

to be accidentally broken. Excipients are often sold through a distributor. therefore, their stability has been well established Distributors can sell the excipient in the unopened

and characterized. Stability issues occur more

excipient manufacturer’s package or the distributor

frequently from exposure to moisture or oxygen

may repackage the excipient into smaller packages. rather than temperature extremes. However, unExcipients may also be shipped in bulk to a manu- less studies have shown the excipient to be affected facturer’s terminal or a distributor where the excip- by extremes of temperature, humidity, or exposure ient is either stored in bulk tanks or packaged from

to oxygen, there is little cause for concern regard-

the tank truck or rail car into discrete containers. ing excipient storage. Any time the excipient is handled other than in

For moisture- and/or oxygen-sensitive excipients,

the original container is an opportunity for the

the excipient packaging should be considered and

excipient to become contaminated, adulterated, or

assessed when selecting a supplier. The excipient

otherwise compromised. Therefore, the fewer such

supplier should provide evidence for the suitability

activities in the supply chain, the lower the risk.

of the packaging used to protect the excipient from

Where delivery is not direct from the excipi- moisture and oxygen. ent manufacturer, the pharmaceutical company

Finally, tamper-resistant packaging is an im-

should periodically establish the pedigree of the

portant consideration in the selection of a sup-

excipient. As discussed in the IPEC-Americas

plier. Though packages can be sealed with tam-

and IPEC-Europe Excipient Pedigree position

per-evident closures, the package materials can be

paper of 2008 (6), the pharmaceutical company

susceptible to tampering via a puncture. However,



13

E�� impervious packaging such as steel drums may not

been made and the risk is kept to an accept-

be appropriate for some excipients. High-density

able maximum.

polyethylene (HDPE) drums may be compatible

t

No viral risk material is used in excipient man-

with the excipient and are more resistant to prod-

ufacture unless an adequate assessment has

uct tampering than fiber drums or bags. Though

been made and the risk is kept to an acceptable

bags and supersacks are a challenge to make tam-

maximum.

per resistant due to their material of construction,

t

oftentimes they are the only packaging available.

tamination risk from microbes or endotoxin un-

Package openings should be protected with tam-

less an adequate assessment has been made and

per-evident seals unique to the excipient manufacturer. Such seals are characterized by their having

the risk is kept to an acceptable maximum. t

to be opened to gain access to the excipient, cannot be reapplied if broken or otherwise removed,

t

Transport of the excipient from the manufacturer is directly to the user.

t

tic. While numbered seals are desirable, they are impractical for excipients where the number of

The risk of contamination is mitigated through the implementation of GMP controls.

and are unique in that they have the manufacturer’s name, logo, or inherent design characteris-

Adequate measures are taken to control the con-

The excipient is stable under the conditions of storage and shipment.

t

Each excipient package is tamper resistant and, if

containers in a lot often exceeds 100 and can be

feasible, closed with a tamper-evident seal.

in excesses of 1000. Using a tamper-evident seal,

These considerations will help to ensure the use

however, provides no benefit if incoming inspec- of quality excipients in the manufacture of phartion by the pharmaceutical firm fails to match

maceuticals.

the appearance of the seal to an authentic seal (or photo) provided by the excipient manufacturer.

Acknowledgement The author would like to thank those colleagues

Conclusion

at IPEC who took the time to provide comments

Excipient quality is best characterized as confor- to this article. mance to GMPs and the compendial monograph or specification with a consistent composition pro- References 1. ICH, Q7, Good Manufacturing Practice Guide for Active Phar-

file lot to lot and within lot. In the selection of an excipient supplier, the following characteristics minimize the risk to excipi The excipient is manufactured using dedicated equipment or mixed-use equipment with suffiNo TSE risk material is used in excipient manufacture unless an adequate assessment has 14

Pharmaceutical Technology SOLID

3. NSF/IPEC/ANSI Standa rd for Pharmaceutica l Excipients , Good 363-2014. 4. IPEC-Americas, Technically Unavoidable Particle Profile Guide (IPEC, 2013). 5. EUR-Lex, Official Journal of the European Union, 2015/C95/02,

cient cleaning validation. t

2. IPEC, Excipient Composition Guide (IPEC, 2009). Manufacturing Practices (GMP) for Pharmaceutical E xcipients,

ent quality: t

maceutical Ingredient (ICH, November 2000).

21.3.2015, C95/10-C95/13. Website: http://bit.ly/EurLexTC15 6. IPEC, IPEC-Europe Excipient Pedigree position paper, ww w. ipecamericas.org/ipec-store PT

DOSAGE DRUG DEVELOPMENT AND MANUFACTURING 2016

PharmTech.com

An obsession with quality de finitely works to your advantage. That s why Mikart is the Contract Development and Manufacturing Organization (CDMO) that delivers the attention to detail you need, plus the speed and responsiveness you want. ʼ

A Textbook Case For (Quality Control) OCD. We re quick to respond with personalized solutions that get your projects completed faster and meet your individual requirements. All while remaining fixated on quality. ʼ

In addition to Formulation Development, Mikart also provides Clinical Trial Supplies, Regulatory Filing Support, plus Manufacturing and Packaging solutions. You ll find Mikart has everything you re looking for in a pharma partner: innovative technology and equipment, highly skilled people and 38 years of solid CDMO experience. ʼ

ʼ

To experience just how responsive we can be, contact us at 1-888-4MIKART or send us an email to [email protected]

Ready. Responsive.

Mikart, Inc. | 1750 Chattahoochee Avenue | Atlanta GA 30318 www.mikart.com | 1-888-4MIKART

Right for You.

E�� D��

Unifying Excipient Databases Properties data critical, says NIPTE’s Hoag Agnes Shanley

More databases currently provide information on pharmaceutical excipients, but are they providing enough information of the right kind, and can they help users address variability?

O

nce taken for granted as inactive fillers, excipients play a

vital role in any pharmaceutical formulation. The wrong excipient can reduce a drug’s efficacy or interact with

APIs or other ingredients, resulting in side effects or worse.

Excipients can also be major sources of variability. Improved analytics have shown significant differences in measured physical properties such as moisture content, particle size, or bulk density, between the same excipient sourced from different suppliers, and even between different lots sold by the same vendor. Differences that may seem small can have an impact on final drug product quality.

Variability means risk “We have variability from the excipients, the API, the manufacturing, the processing conditions, and all of these … result in variability of the product,” Stephen Hoag, a professor at the University of Maryland at Baltimore’s school of pharmacy, commented before an FDA panel (1). The panel had asked for recommendations for future generic-drug research funding through the Generic Drug User Fee Amendments of 2012 (GDUFA). “In terms of risk analysis, that’s where harm to the patient can come in,” Hoag said. As recognition increases of the pivotal role that excipients play in drug safety and quality, and the increasing quality and supplychain risks involved, a growing number of databases are being developed to provide more information on these materials. One of the first such projects began in 2009 when FDA funded t he creation of a database by the National Institute for Pharmaceutical Technology and Education (NIPTE), a consortium of universities and industry, to catalog physical properties and other information 16



PharmTech.com

S E G A M I Y T T E G / + E / O T A P M E

E�� D�� on excipients. Professor Hoag has been leading

introduced. IPEC has continued to provide the

these efforts.

agency with feedback on the effort’s progress (4, 5).

The goal of the NIPTE–FDA database, the Ex-

A separate excipient data-gathering effort is the

cipients Knowledge Base (2), is to offer detailed

STEP (Safety and Toxicity of Excipients for Pediat-

information on the physical material properties

rics) database (6), designed to help pharmaceutical

of commercial excipients. This is exactly the type

formulation scientists screen excipients for use in

of data that engineers routinely use in other in- children’s medications. dustries, including chemical, automotive, and

Knowledge gaps

aerospace, when developing new products.

Despite the increased focus on gathering informa-

“There’s a real lack of knowledge of how to go from material to clinical properties.” —Stephen Hoag, University of Maryland and NIPTE

tion on excipients, significant gaps in knowledge remain. At the FDA GDUFA hearing, Professor Hoag said, “There’s a real lack of knowledge of how to go from material to clinical properties.” Most of the industry’s knowledge of these relationships, he said, is gained in a hit-or-miss fashion, based on empirical observation. More fundamen-

The database, which is housed on Pharmahub, tal knowledge, and development of models, would a data sharing community developed at Purdue

help with change control, he said, and would allow

University, is designed to be a shared resource for

the industry to deal with unexpected factors.

the global industry with community features that,

Hoag has suggested that separate databases be

for example, allow users to input their experiences. merged to provide increased functionality and So far, it has between 500 and 1000 regular users.

more information on material properties. He discussed formulation issues and plans for the NIPTE

FDA’s Inactive Ingredients Database

database with Pharmaceutical Technology .

At around the same time that NIPTE started to work on their database, FDA began to catalog the inactive ingredients used in products (in final dos- Future plans for the database age form) that it had approved. A working group

PharmTech: What progress has been made with the

was established in 2011, involving the International

NIPTE database, and what are the plans for its fu-

Pharmaceutical Excipients Council (IPEC), the Ge- ture, both short and long term? neric Pharmaceuticals Association (GPhA), and a

Hoag: FDA funding has come to an end, so we

cross-disciplinary team from FDA’s Office of Ge- are looking for funding that would help us to add neric Drugs, to develop FDA’s Inactive Ingredients

more data to improve it. So far, we have had dis-

Database (IID) (3).

cussions on this topic with both GPhA and IPEC.

In 2015, the database’s IT underpinnings were

We are seeing that more companies are starting to

improved, and a more user-friendly interface was

use the database. Some major upgrades have been

18



PharmTech.com

made to the underlying Pharma Hub software, in-

PharmTech: Are companies concerned, as they

cluding new enhancements to improve security, data

were in the early days of the process analytical

visualization, and calculations based on the data.

technology and quality by design implementations,

IPEC is interested in our work and some mem- that sharing best practices would amount to giving ber companies have been using it, and adding data, away competitive advantage? but in a password-protected way. In March 2016,

Hoag: Some suppliers worry about how posting

NIPTE will be exploring the idea of a new center

data will affect their competitive positions. In the

for pharmaceutial technology and drug-develop- end, some people recognize the need for this datament research with FDA.

base, while others worry about it. We have set it up as a community tool and a place where people can

Need for greater depth and functionality PharmTech: You’ve noted a ‘proliferation’ of excipi-

share individual experiences and best practices. There is still a lot of ‘reinventing the wheel,’ and

ent databases today. Do they provide sufficient

redundant testing and evaluation going on within

information to help industry deal with problems

individual pharmaceutical companies. I’ve seen

such as variability and complexity?

cases within a company where one lab in the US

Hoag: IPEC is primarily interested in FDA’s IID, does a characterization study, then another of the

and we’ve talked about how the two databases

company’s labs in Europe runs the same study

might be merged.

without realizing that it had already been done.

PharmTech: How does the philosophy behind

NIPTE’s database differ from that behind the IID?

The NIPTE database could help prevent wasted efforts like that.

Hoag: IID is not really a database but more of

Our goal is to make people aware of this tool,

a spreadsheet with useful information in it. Our

and to fund research and work that would improve

database is focused on properties and on allowing

it. At this point, the project needs more support.

users to mine data. PharmTech: You have repeatedly mentioned the

need for better modeling in pharmaceutical de velopment and formulation, based on an understanding of material properties. Is its reliance on batch processes the reason why the pharmaceutical industry has not developed more innovative approaches to modeling? Hoag: Batch focus is not the main reason, but

growing interest in continuous manufacturing will drive home the need for some of this work. Currently, the biggest holdup appears to be inertia and investment in the status quo and established ways of doing things.

References 1. Transcript , Public Hearing, FDA Generic Drug User Fee Amendments of 2012 Regulatory Science Initiative, Request for Public Input for Fiscal Year 2015, Generic Drug Research, Part 15, Public Hearing, Friday, June 5, 2015, www.fda.gov/downloads/ForIndustry/UserFees/GenericDrugUserFees/ UCM455846.pdf. 2. The NIPTE-FDA Excipients Knowledge Base, www.pharmahub. org/excipientsexplore. 3. The FDA Inactive Ingredients Database, www.accessdata.fda. gov/scripts/cder/iig/index.cfm. 4. R. Iser, “Inactive Ingred ient Database-FDA Update,” GPhA Fall Technical Conference, October 30, 2013, www.fda.gov/downloads/AboutFDA/CentersOffices/OfficeofMedicalProductsandTobacco/CDER/UCM375889.pdf. 5. “IPEC-Americas suggests improvements to FDA’s Inactive Ingredients Database,” Pharmaceutical Technology Sourcing Management , October 30, 2015, ww w.pharmtech.com/ipec-americasuggests-improvements-fda-s-inactive-ingredients-database 6. S. Salunke et al., International Journal of Pharmaceutics, 457, (1) (November 20, 2013), pp. 310-322. PT



19

N��-G�� C��

Establishing a New Performance Standard for HPMC Capsules Agnes Shanley

A new generation of cellulose-derived materials addresses the variability in disintegration and product dissolution that were seen in

F

or nearly a hundred years, two-piece hard shell capsules made from gelatin have been the preferred medium for encapsulating solid oral-dosage forms. The market, however, began to recognize the need for alternatives decades ago.

Gelatin, derived from pork and beef byproducts, posed a concern

the first generation of

for patients with religious or dietary restrictions, but it could also

gelatin alternative HPMC

pose formulation problems when encapsulating hygroscopic and

capsules, while offering

moisture-sensitive ingredients.

in-vitro and in-vivo

The plant-based material, hydroxypropyl methylcellulose (HPMC),

performance comparable to

was developed as an alternative to gelatin for two-piece hard shell

that of gelatin capsules.

capsules. The first manufacture of the material in capsule form, however, required the use of secondary gelling agents, which resulted in variability in both disintegration and product dissolution. Approximately eight years ago, a second generation of HPMC-polymer capsules was introduced. Matt Richardson, PhD, manager, Pharmaceutical Business Development at Capsugel, and Michael Morgen, PhD, senior principal scientist, Bend Research, a division of Capsugel Dosage Form Solutions, discussed these next-generation materials with Pharmaceutical Technology.

HPMC vs. conventional materials PharmTech: What makes HPMC capsules different, in

terms of performance properties, from conventional gelatin technology? Richardson (Capsugel): HPMC capsules, specifically those made by

thermo-gelation, differ from conventional gelatin capsules in several ways. Gelatin, derived from animal collagen, is composed of amino acid chains, while HPMC is cellulose-based. The respective poly 20



PharmTech.com

L E G U S P A C F O Y S E T R U O C S I E G A M I

mers impart different properties to each capsule. second-generation HPMC capsules offer disinteCompared to gelatin, HPMC has lower moisture

gration/product dissolution profiles that have been

absorption, more flexibility, higher thermal stabil- shown to provide equivalent in-vivo performance ity, and better resistance to chemical crosslinking. to gelatin capsules for a range of drug molecules. The manufacture of second-generation HPMC

PharmTech: What are their limitations?

capsules by thermo-gelation is quite different than

Morgen (Bend Research): HPMC capsules have

that of first-generation HPMC capsules and gelatin

higher oxygen permeability than gelatin capsules,

capsules. Both first-generation HPMC and gelatin

so they may not be ideal for oxidation-sensitive

capsules are made by dipping stainless-steel pins into

formulations, or those that have an odor.

a hot polymer solution. The polymer is deposited on the stainless-steel pin and then dried to achieve a tar- Recent in-vivo and dissolution test results geted water content. In the thermo-gelation process

PharmTech: Please discuss the significance of recent

for second-generation HPMC capsules, heated stain- in-vivo and dissolution test results. less-steel pins are dipped into a solution of HPMC to form capsules without use of gelling agents.

Morgen (Bend Research): The most recent study

focused on investigating the clinical performance

PharmTech: What were the specific performance

of second-generation HPMC capsules relative to

limitations with first-generation HPMC capsules?

that of gelatin capsules. It aimed to demonstrate

Morgen (Bend Research): First-generation HPMC

that, in-vivo, the slight disintegration time delay for

capsules avoided gelatin’s potential for crosslink- second-generation HPMC capsules does not lead ing. They were formulated with gelling systems, to differences in pharmacokinetics for relatively however, and variability in disintegration and

soluble, well-absorbed drugs, but that the perfor-

product dissolution rates was often observed, de- mance would be equivalent to that of gelatin cappending on pH and/or ionic strength of the test

sules for these applications. This equivalence was

media. Second-generation HPMC capsules were

shown, using three different compounds. Formula-

designed to reduce the performance variability of

tion scientists now have more options for optimiz-

first-generation HPMC capsules.

ing capsule disintegration and product release.

PharmTech: What other specific advantages does

the second-generation material offer?

PharmTech: What other innovations are being

investigated for second-generation capsules?

Richardson (Capsugel): HPMC capsules have a lower

Morgen: For some applications, the capsules

average water content (6% compared with roughly

may provide superior bioavailability, particularly

14% for gelatin capsules), making them an excellent

for low solubility drugs, by sustaining supersatu-

choice for low moisture or hygroscopic formula- rated concentrations of dissolved drug in the gastions. They also demonstrate high levels of thermal

trointestinal (GI) tract through crystallization in-

stability and are quite f lexible, compared with gela- hibition. This is a potential advantage,specifically tin capsules, which can occasionally become brittle, for high-energy salt and amorphous drug forms, especially with low-moisture APIs and excipients. as well as for weakly basic drugs that are capable Perhaps their most important attribute is that

of supersaturating in the GI tract. PT



21

P�� S��

Modeling and Simulation Move Downstream Agnes Shanley

Computer-based tools promise better quality products, improved process control, and increased R&D efficiencies, but will require different workflows.

M

any industries use computer-based simulation and modeling routinely to troubleshoot and improve their processes. Pharmaceutical manufacturers have not

yet embraced the technology, at least not for down-

stream applications. Some companies may use non-computer based simulation to optimize conditions, for example, compaction simula-

tors for tablet compression. Generally, though, the industry’s most advanced computer-based systems are reserved for pharmacokinetics and early research. For chemists, engineers, and technicians who work in development and manufacturing, part of the challenge to using models is that they can be more difficult to develop for batch processes, which can involve more variability in materials, process conditions, and other factors. In addition, the physical and material properties data required for this work, which are freely available to engineers in petrochemical processing, for instance, via references such as Perry’s

Chemical Engineering Handbook, are not readily available to professionals in pharmaceutical manufacturing and development. Most of the data on materials depend on the specific run, batch, equipment, and facility. Increased interest in continuous processing, however, is convincing more professionals in the industry to study and apply in-silico technologies and models to gain process knowledge and improve manufacturing. Modeling could also improve drug development, enabling development of higher quality, more robust products, and increasing R&D efficiency. Indicative of the trend is a new four-year initiative that was launched in the United Kingdom in January 2016 to help streamline drug development and manufacturing by leveraging better 22


DEVELOPMENT AND MANUFACTURING 2016 P h a r m T e c h . c o m

S E G A M I Y T T E G / S R U C R A I_ R U Y

Award-Winning On-Dose Authentication & Brand Protection Solutions • Covert, Inert and Easy to Apply • Machine-Readable Microtags • Custom Codes by Product

                  Click on the video link below to learn how on-dose    

Y PROD UC T

T F E

A S

A

E

H

Contact us now: www.trutags.com/awardwinning

F

T

& H

T L

O

H E

AWARDS2013

Y

E

A

R

P�� S�� computer-based process modeling and simula- facturing. Its software is being used by Big Pharma tion. The $29-million program, called Advanced

companies, but as of yet, only by 2–20 people within

Digital Design of Pharmaceutical Therapeutics

each company. “Our biggest challenge,” notes Berm-

(ADDoPT), aims to develop the tools required for

ingham, “is integrating modeling tools into heavily

an “efficient, knowledge-based, quality by design- experiment-focused workflows.” It is important that oriented [pharmaceutical] supply chain” (1). Astra- modeling be used before numerous experiments are Zeneca, Bristol Myers Squibb, GlaxoSmithKline, performed that, individually, yield little information and Pfizer are participating in the program, as are

content, he says.

the universities of Cambridge, Leeds, and Strath-

Academic programs, including the Center for

clyde, all of which are doing advanced research in

Structured Organic Particulate Systems (C-SOPS),

materials science and continuous manufacturing. involving Rutgers and Purdue, use the software for Cambridge’s Crystallographic Data Center and the

advancing science and transferring knowledge to

Hartree Center, a high-performance computing fa- industry, Bermingham says. For manufacturing, cility, are also partners.

process modeling software allows users to ana-

Participating solutions providers include Process

lyze reasons why a process may not be inherently

Systems Enterprise (PSE), a specialist in process

robust. It also permits the cost-effective develop-

modeling and optimization; the process-control

ment of process-control systems. This approach

company Perceptive Engineering; and Britest, a

can help make R&D more efficient.

nonprofit organization that focuses on using quali-

“It’s inefficient to have process characterization

tative methods to improve process understanding

and scale-up depend overwhelmingly on experi-

and value.

ments, and, in some cases, statistical modeling.

One of ADDoPT’s goals will be to diversify the

The days when armies of chemists could run these

modeling techniques available to pharmaceutical

experiments each day are over,”Bermingham says.

manufacturers, says Sean Bermingham, principal

He points to mechanistic model-based scaleup

consultant and head of PSE’s Life Sciences division, of a spray-drying process as an example. Instead of and the technical lead of ADDoPT.

doing 20 different spray-drying experiments, one

PSE is a spinoff of Imperial College UK and was

would be able to start with vapor sorption experi-

set up in 1997 to sell process modeling and optimi- ments, estimate isotherm parameters, then work zation software for the chemicals and oil and gas

on drying kinetics using single-droplet studies,

sectors. In 2010, it set up a pharmaceutical and

then move on to spray drying using a handful of

life-sciences division.

experiments to characterize fluid and particle flow

Process optimization and modeling is widely

at the various scales of interest.

used in petrochemicals, and, with companies such

Another factor that is driving interest in simula-

as Aspen Technologies and Scale-up Systems, it has

tion is continuous processing. Most of the petro-

moved into some facets of API development and

leum industry’s processes are run continuously. As

manufacturing. PSE has focused on expanding use

more pharmaceutical manufacturers explore the

of modeling technologies into drug-product manu- potential benefits of 24/7 operation, modeling has 24



PharmTech.com

become more important. Among the pharmaceu-

Another computer-based simulation and mod-

tical companies that use mechanistic modeling

eling platform for downstream pharmaceutical

tools, Bermingham says, 50–60% of their use is to

manufacturing application is F-CAD, developed

support continuous manufacturing efforts.

by the Center for Innovation in Computer-Aided

In contrast to mechanistic modeling, statisti- Pharmaceutics (CINCAP). Utilizing artificial incal models, such as JMP’s, create black-box type

telligence, and the concept of “cellular automata”

models to identify critical process parameters and

(2) to model complex systems, the tool looks like

critical quality attributes, all essential for quality- any CAD platform used to model new car and airby-design (QbD) work. Mechanistic modeling cuts

plane prototypes (3).

through the noise and multivariate data by using

A number of companies are evaluating the

first-principles equations that represent the process

system. However, as one of its developers notes,

and can be used not only for QbD work but also to

using this type of technology will require that

control and improve the process and to adjust to

pharmaceutical manufacturers adopt workf lows

unanticipated changes.

more often seen in the automotive and aviation

However, using this approach requires a differ- industries (3). Specifically, he notes, samples used ent mindset and new workflows. It also requires

in Phase I and II work will have to be prepared at

new types of data, less process data (i.e., less use

conditions used in Phases III and IV (i.e., using a

of APIs), but more materials property data, a point

mechanical simulator of high-speed rotary presses).

that thought leaders in the industry continue to

This would require engineers and pharmaceutical

emphasize (see Feature, p. 16). High-performance

scientists to collaborate more closely. It would also

computing will also be needed to sort through

require closer connection between the IT and doc-

massive amounts of material property data. One

umentation systems used by different functional

characteristic that would help develop solid dos- groups in development and manufacturing. age-form manufacturing processes more efficiently is “compressability.” At this point, that variable can vary significantly when the blend of API particles and excipients is changed. “You may run tests on a compactor simulator, but when you change the blend composition or the particle size distribution

References 1. A. Shanley, “Can Better Modeling Reduce Pharmaceutical De velopment and Manufacturing Costs?” Pharmtech.com, March 1, 2016, www.pharmtech.com/can-better-modeling-reducepharmaceutical-development-and-manufacturing-costs 2. M. Puchkov et al., “3-D Cellular Automata in Computer Aided Design of Pharmaceuticals,” in Formulation Tools for Pharma-

of the API and/or excipients, the blend’s behavior

ceutical Development , J. Aguilar Editor, Woodhead Series in

will change. One of the research aims in ADDoPT

Biomedicine, 2013, Elsevier, p. 155. Chapter available via

is to generate correlations for compressability and

QBAJ&pg=PA155&lpg=PA155&dq=CINCAP+switzerland+pha

other material properties,” says Bermingham. AD-

rma&source=bl&ots=5H8WWo3EEW&sig=uT5pn7rJDr79wN

DoPT’s pharma partners will provide real-world measurements from their labs of different compressability values for different blends, in order to develop a compressibility correlation. Pharmaceutical Technology SOLID

Google Books, https://books.google.com/books?id=gGlEAgAA

OqaAQ9kXzEaQw&hl=en&sa=X&ved=0ahUKEwi7jsunn6fLA hXDnYMKHW6uC8QQ6AEIIjAB#v=onepage&q=CIN CAP%20switzerland%20pharma&f=false 3. H. Leuenberger, European Journal of Pharmaceutical Science, Vol. 16 (February 2016), www.ncbi.nlm.nih.gov/ pubmed/26876764.

PT


25

T�� C��

Optimizing Tablet Compression Frederick J. Murray

A process optimization template provides a practical approach for maximizing output and product quality of an existing tablet compression process.

A

ll aspects of pharmaceutical manufacturing face increased pressure to improve production efficiency and uptime, while maintaining the highest standards for product quality. For tablet compression equipment,

many companies seek to increase productivity of the existing equipment platform prior to considering additional equipment. A structured optimization template provides a practical approach to maximizing both output and product quality of an existing tablet compression process.

Tablet weight control The key process parameter for any tablet compression operation is tablet weight control. Assuming uniform distribution of the blend, the ability to hold precise tablet weight is an absolute requirement to delivering the prescribed dosage of active material. Tablet weight control is influenced by a number of factors including press speed, flow properties of the granulation, filling depth, feeder paddle configuration, and working lengths of the upper and lower punches. Press speed. The press speed and pitch circle diameter of the tablet

press die table will determine the tangential velocity of the press tools. The lower press tools pass under the feed frame to fill the dies, and based on the length of the feeder opening and tangential velocity of the tools, the feeder dwell time may be determined. It is obvious that a longer feeder dwell-time will permit more time to fill the dies and that there may be a critical speed limit where it is impossible to achieve uniform die filling because the feeder dwell time is too short. Frederick J. Murray is President of KORSCH America Inc., tel: 508.238.9080, fred. [email protected]. 26

Flow properties of the granulation. A product with robust flow proper-

ties will fill the dies uniformly at high speeds. Products with marginal or poor flow properties need long feeder dwell times and often



PharmTech.com

S E G A M I Y T T E G / X R O W R E T T U H S

T�� C�� require a modified feeder paddle configuration to

in the tablet weight. Precision tablet weight control

fill the dies uniformly.

thus mandates excellent press-tool tolerances and

Filling depth. All tablet presses use a volumetric

a corresponding press-tool maintenance program.

fill to obtain the desired tablet weight. The fill

Understanding the issues that impact tablet weight

depth (i.e., position of the lower punch in the die

control and can cause tablet weight variability is cru-

to achieve the desired tablet weight) is determined

cial to a successful process optimization effort.

by the shape of the tablet and the bulk density of the material. A small tablet diameter with a deep

Tablet hardness

filling depth is obviously more difficult to fill than

In addition to tablet weight, thickness and hard-

a larger tablet diameter with a small filling depth. ness are key quality parameters for tablet producThe ability to fill dies with deep filling depths can

tion, and these attributes can be measured in real

certainly be rate limiting.

time as the tablet press is producing tablets. The

Feeder paddle configuration. Most modern tablet

tablet hardness will determine the dissolution rate,

presses use variable-speed power feeders with ro- which is crucial to ensure effective drug delivery. tating feeder paddles that assist the transfer of ma- The tablet hardness is a function of the volume of terial into the die. Feeder paddles generally come

material in the die, and the magnitude and rate of

in a rectangular profile; however, many tablet

compression force applied to the tablet. In most

presses are available with alternate feeder paddle

cases, increasing the press force applied to the

designs, including round profiles and beveled pro- tablet will increase the corresponding tablet hardfiles. There is no real handbook to define the best

ness. Some products, however, have a maximum

feeder paddle for any given product, and empirical

hardness threshold, and higher press forces actu-

testing, such as described in this article, is required

ally cause the tablet hardness to be reduced. Un-

to identify the optimal paddle configuration.

derstanding the relationship between press force

Upper- and lower-punch working lengths. The role of

press-tool working lengths on tablet weight control

and tablet hardness is a crucial component to any process optimization study.

is often overlooked. Variability in the lower-punch

The rate that press force is applied to the tablet

working length will directly impact the volume of

is a function of the tangential velocity of the press

material in each die and can, in itself, cause tablet

tools, the diameter of the compression rollers, and

weight variation. Variability in the upper-punch

the geometry of the press tool, and it is generally

working length will not directly impact the volu- referred to as compression dwell time. In simple metric fill; however, because most modern tablet

terms, the compression process imparts energy

presses are using press-force control systems as the

into the tablet by applying a force over a period of

basis for tablet weight control and because upper

time. If the time is reduced (when the press speed

punch working length variability will alter the tab- is increased), then the force must be increased let thickness (and therefore, the press force), the

to impart the same amount of energy. The press

force control system is now reacting to tolerance is- speed is thus another dimension to the press force– sues in the upper punches and not actual variability tablet hardness relationship. 28



PharmTech.com

Process optimization template With this fundamental understanding of the key process parameters and those factors that can impact quality parameters, one can use the process optimization template, which consists

Figure 1: Tablet weight variation shown as relative standard deviation (Srel%) of individual tablet weight as a function of press speed. 6% 5% 4%

% l 3% e r S t h 2% g i e W 1% 0% 20

30

of the following four sets

40

50

60

70

80

90

Press speed (RPM)

of empirical testing. Press speed/tablet quality. This testing consists

of making the product to

Figure 2: Tablet thickness variation shown as relative standard deviation (Srel%) of individual tablet thickness as a function of press speed.

specification (i.e., target

6%

weight, thickness, and

5%

hardness) and then measuring the standard deviation of individual tablet weight, thickness, and

4%

% l e r S 3% s s e 2% n k c i h 1% T

hardness values across

0% 20

a defined speed range.

30

40

50

60

70

80

90

Press speed (RPM)

Graphs of standard de viation of individual tablet weight, thickness, and hardness variation versus

Figure 3: Tablet hardness variation shown as relative standard deviation (Srel%) of individual tablet hardness as a function of press speed.

press speed will clearly

40%

define the press speed

35%

range in which quality

30%

tablets can be produced. Figure 1 plots the rela-

tive standard deviation (Srel %) of individual tablet weight as a function of

% l e r S s s e n d r a H

25% 20% 15% 10% 5% 0% 20

30

press speed. In this ex-

40

50

60

70

80

90

Press speed (RPM)

ample, there is a clear inPharmaceutical Technology SOLID DOSAGE DRUG


29

T�� C�� crease in the tablet weight variability at press speeds higher than 60 RPM, thus 60 RPM is the maximum press speed given the flow properties of the material tested. Better flow properties would likely permit higher press speeds.

Figure 4: Tablet hardness vs. compression force at different press speeds (i.e., compression dwell times). 12 10

) P8 K ( s s6 e n d r4 a H 2 0

Figure 2 plots the rela-

2

30 RPM

4

45 RPM

6 60 RPM

75 RPM

8

90 RPM

10

Main compression force (kN)

tive standard deviation (Srel %) of individual tablet thickness as a function of press speed and shows

(i.e., compression dwell time) and will confirm

consistent control across the press speed range.

the press force value that corresponds to the de-

Figure 3 plots the standard deviation (Srel %) of

individual tablet hardness as a function of press

sired tablet hardness. The representative graph shown in Figure 4 indi-

speed. These data mirror the tablet weight data, cates that the compression dwell time does impact with consistent control up to the 60 RPM press

the tablet hardness, especially at the higher press

speed level.

force levels. At 90 RPM, the tablet hardness is, on

The process capability (Cp) may be calculated

average, 80% lower than the same tablet produced

for tablet weight, thickness, and hardness at each

at the 30 RPM press speed level. Based on required

press speed using Equation 1.

tablet hardness range, the expected press force range can be easily determined.

Cp = (USL – LSL) / 6 * ∑

[Eq. 1]

Feeder speed/feeder paddle optimization. This test

consists of running the press at different press Where Cp is process capability index, USL is

speeds and adjusting the feeder speed across the

upper specification limit, LSL is lower specifica- range, while recording the relative standard detion limit, and Σ is standard deviation.

viation of individual tablet weights. This measure-

Nominal process capability values of 1.33 or 1.50, ment is performed at each press speed and with or higher, are generally indicative of a process that

different feeder paddles designs (standard, round,

is under control.

beveled) to determine the optimal feeder speed

Tablet hardness/compression force. This testing

and feeder-paddle configuration.

consists of running the press at different com-

If the results show that the standard deviation of

pression forces and measuring tablet hardness at

tablet weight is not impacted significantly by the

the different press-speed levels. The force versus

feeder speed, then it makes sense to run the feeder

hardness plot will confirm the ability to achieve

speed at the lowest value to avoid overmixing or

the desired tablet hardness at each press speed

shearing the granulation in the feeder.

30



PharmTech.com

M i c r o e n S p e c a c i p s a l u i a s t l i s t o n

Does your current CDMO leave you with blurred vision?         in the door then seem to go in multiple directions. There is no clear direction; and the invoices are piling up. At Coating Place our experienced staff believes the best results come with a clear understanding of your ideas and goals. Trust the knowledge, quality and             

     

                

   

T�� C�� Figure 5: Weight variation shown as relative standard deviation (Srel%) vs. feeder speed at a series of press speeds (30– 90 RPM indicated as different colors) using a rectangular feeder paddle.

lock in the optimal feeder speed at each press speed. Figure 7 shows the im-

6%

pact of the feeder-paddle

5%

configuration on the rela-

4%

tive standard deviation of

% 3% l e r S t 2% h g i e 1% W

individual tablet weight at a variety of different press speeds and feeder speeds.

0% 10

20

30 RPM

30

45 RPM

40

60 RPM

50

60

rable performance of the

90 RPM

75 RPM

The results show compatwo paddles at the 30 RPM

Feeder speed (RPM), rectangular paddles

press speed (see Figure 7a). Figure 6: Weight variation shown as relative standard deviation (Srel%) vs. feeder speed at a series of press speeds (30– 90 RPM indicated as different colors) using a round feeder paddle.

At the 60 RPM press speed, the rectangular paddle clearly performs better

6%

(see Figure 7b). Results are

5%

mixed at the 90 RPM press

4%

speed (see Figure 7c).

% 3% l e r S 2% t h g i e 1% W

Fill cam optimization.

Most presses offer a variety of fill cams to cover

0% 10

20

30 RPM

30

45 RPM

40

60 RPM

75 RPM

50

60

90 RPM

Feeder speed (RPM), round paddles

a range of filling depths. The purpose of the fill cam is to overfill each die and then push some material back into the

The following example shows the impact of feeder feeder to ensure optimal die filling as the lower speed on tablet weight variation, using a rectangular

punch moves through the dosing cam. For exam-

feeder paddle (see Figure 5) and a round feeder paddle

ple, a standard EURO or TSM B turret will offer

(see Figure 6), at different press speeds. This infor- a fill depth range of 0–18 mm. This filling depth mation can be used to establish the optimal feeder

is achieved by a range of filling cams, as follows:

speed at each press speed, to be stored in the product recipe. In this example, it can be concluded that the

6 mm Fill Cam

Fill Cam Range 0–6 mm

higher feeder speeds (> 40 RPM) and the rectangular

10 mm Fill Cam


feeder paddle generally produce better results (i.e.,

14 mm Fill Cam


less weight variation), and the data can be used to

18 mm Fill Cam

Fill Cam Range 8–18 mm.

32



PharmTech.com

Because there is overlap in the range of different fill cams, the optimal fill cam must be selected empirically. For a product that requires a final filling depth of 8 mm, the press can operate with either the 10-mm fill cam or the 14-mm fill cam. Although it may seem logical that a deeper fill cam will be better, that is not always the case. In summary,

Figure 7: Weight variation shown as relative standard deviation (Srel%) of tablet weight vs. feeder speed and feeder paddle configuration for rectangular paddles (blue lines) and round paddles (red lines) at (a) 30 RPM, (b) 60 RPM, and (c) 90 RPM press speeds.

(a) 30 RPM press speed 6% 5% 4%

% l e r 3% S t 2% h g i e 1% W 0% 10

lap between multiple fill cams can be optimized only by testing each fill cam across the desired speed range.

30

40

50

60

(b) 60 RPM press speed 6% 5% 4%

% l e r 3% S t 2% h g i e 1% W 0% 10

products that have filling depths in the over-

20

20

30

40

50

60

(c) 90 RPM press speed

6% 5%

% 4% l e r S 3% t h 2% g i e 1% W 0% 10

20

30

40

50

60

Figure 8 plots the relaRectangle Paddles

tive standard devia-

Round Paddles

Feeder speed (RPM)

tion of tablet weight at different press speeds using two dif ferent fill cams (10 mm and 14

Figure 8: Weight variation shown as relative standard deviation (Srel %) of tablet weight at different press speeds using two different fill cams (10 mm, blue line and 14 mm, red line) at the same dosing setting and tablet weight.

mm) at the same dos-

6%

ing setting and tab-

5%

let weight. The results show comparable results through 60 RPM, but the deeper fill cam

4%

% l e r 3% S t h 2% g i e W 1% 0%

(14 mm) clearly extends the process control (i.e.,

30

45

75

60

10 mm Fill Cam

90

14 mm Fill Cam

Press speed (RPM)

Contin. on page 45



33

C�� M��

Filling the Analysis Gap in the Move to Continuous Processing Jamie Clayton

Effective analysis is key for the successful continuous manufacturing of soliddosage pharmaceuticals.

T

he benefits of continuous manufacturing justify significant investment, as evidenced by collaborations such as the MIT/Novartis Center for Continuous Manufacturing

(1). Batch production dominates within the pharmaceuti-

cal industry but many expect continuous processing to contribute a substantial share of manufacturing capacity. The move to continuous manufacturing requires changes in analytical practices in order to support new manufacturing models. Dynamic powder testing can contribute to the development of efficient continuous processes.

The benefits of continuous manufacturing Although innovative in many areas, the pharmaceutical industry has historically focussed less on processing than other research and development areas. Patent protection previously ensured that R&D costs could be properly recouped, but as costs rise and time to market increases, profitability cannot be guaranteed. When patents expire, profitability relies on efficient production. Furthermore, a regulatory focus on risk suggests a need for greater understanding of processes and improved quality control. A shift from batch processes, which are heavily dependent on manual intervention, to automated, continuous operation is highly attractive. In batch production, sequential steps are undertaken, with analysis Jamie Clayton is operations director at Freeman Technology. 34

performed in between. Batch-to-batch variability and products out of specification (OOS) are common problems. The necessary rework and waste levels are unacceptably high.



PharmTech.com

S E G A M I Y T T E G / J A D

THINI UPGRADE Optimize your capability Taking your processing capabilities to the next level is a strategy for success; one that often demands upgrading equipment to meet new production goals and future requirements. For immediate access to the best technologies at the right price while maximizing returns from surplus idle equipment, Federal Equipment Company is ready to help you optimize operations while recovering equipment value through accurate appraisals, strategic liquidations and full logistical support.

When you think equipment, think Federal Equipment 1.877.536.1538

>

www.fedequip.com

Download

Download Article Article

C�� M�� Continuous processing is widely used in chemi- of a problem. With continuous manufacture, there cal and food industries and offers the following

is a question of how to define a ‘batch’. A batch be-

important advantages:

comes associated with an operating period, which

t

Reduce costs, labor, and waste

begins when start-up completes and ends at a de-

t

Optimize asset utilization

fined point. Any OOS products are therefore as-

t

Simplify scale-up

sociated with a time period rather than a discrete

t

Improve containment.

batch number, which can make problems difficult

These are environmental and economic im- to isolate. provements, but there are also technical benefits.

Optimized processing relies on understanding

A batch step has a beginning and an end; be- and controlling the materials and process variables tween these points the product continuously

that define clinical efficacy. The pharmaceutical

changes (e.g., a blending process starts with the

industry has traditionally focused on developing

unmixed constituents and proceeds to a homo- and adhering to repeatable processes. This apgeneous state). A well-controlled continuous pro- proach relies on consistent feed and provides little cess should operate at steady state for the major- flexibility to respond to variation. This is a critical ity of the time. A steady-state operation requires

limitation, as feed variability is a major source of

effective monitoring, as exemplified by widely

failure. The transition from batch processing to-

used techniques such as in-line particle size and

ward knowledge-based continuous manufacture,

near-infrared (NIR) analysis. Steady-state opera- however, has its challenges. tion means continuous processing is associated with consistent output, which equates to consis- A toolkit for more efficient manufacture tent product quality. Scale-up is also simplified as

Operations such as milling, roller compaction,

smaller units can be run for longer, avoiding the

and tableting can be considered semi-continuous,

implications of changes in geometry and volume. as they are constantly fed during a batch camBatch production does, however, have benefits. paign. The challenge involves engineering the One benefit is flexibility; a suite of batch equip- equipment for reliable, prolonged operation and ment can easily be reconfigured for different prod- successfully integrating the necessary compoucts. Batch production also simplifies containment

nents into an optimized continuous process. Au-

Center for Structured Organic Particulate Systems (C-SOPS) For 10 years, researchers at the Center for Structured Organic Particulate Systems (C-SOPS) have worked to transform pharmaceutical manufacturing into a sciencedriven discipline, in the areas of materials formulation and characterization, design and scale up of material structuring, structural characterization and modeling, and integrated systems science. C-SOPS also operates three test beds to develop new continuous manufacturing processes. A National Science Foundation Engineering Research Center, C-SOPS is a consortium of academic institutions and more than 40 industry partners from both bio/pharmaceutical companies and industry equipment suppliers.

36


TestBed1focuses onthe simultaneousdevelopmentof formulations,continuous manufacturing, and analytical control methodologies for solid oral products. Test Bed 2 is designed to create an integrated continuous manufacturing platform to produce film-based drug products with controlled-release properties. Test Bed 3 is based on drop-on-demand manufacturing, and uses liquid-phase processing to avoid challenges normally associated with conventional powder-based processes. Information about consortium members, research projects, and test bed programs can be found at http://ercforsops.org. —The Editors of Pharmaceutical Technology


PharmTech.com

tomation is important but so are analy tical tools that provide the knowledge required to optimize

Figure 1: Measuring flow energy to quantifying changes in flow properties following consolidation.

multi-component systems. The Engineering Research Center for Structured Organic Particulate Systems (C-SOPS) is a group at the forefront of

Y T I S N E D / Y G R E N E W O L F

research in this area. The group applies modeling, in-line analysis, and techniques such as powder rheology to integrate sequential blending, dry

Flowability Change >1000%

Density Change ≈40% (max)

granulation, lubrication, and tableting. A key focus is to develop solutions to avoid three com-

INCREASED TAPPING

mon tableting issues: segregation, agglomeration, and compaction (2).

of powders; however, the need for accurate, process-

Whether improving batch processes, or design- relevant data exposes limitations and highlights the ing, monitoring, and controlling a continuous

merits of techniques such as dynamic testing.

process, analytical tools are needed that deliver

In dynamic testing, axial and rotational forces

relevant data and expand understanding of how

acting on a blade are measured as it rotates through

processes work, reinforcing FDA’s process analyti- a powder sample to determine values of flow encal technology (PAT) initiative. PAT is defined as

ergy that quantify how a powder flows under

“a system for designing, analyzing, and controlling

conditions that reflect processing environments.

manufacturing through timely measurement of

Powders can be characterized in consolidated, con-

critical quality and performance attributes of raw

ditioned, aerated, or even fluidized states to mea-

and in-process materials and processes, with the

sure the response to stress and air content. The

goal of ensuring final product quality” (3). Real- impact of moisture, flow additives, compaction, time analysis is therefore important but so are tech- attrition, and segregation can be evaluated. niques that, for example, provide robust analysis of

Figure 1 contrasts the change in bulk density in-

feeds prior to introduction to the plant. Identifying

duced by tapping with the corresponding change

techniques that provide the information required to

in f low energy. Flow energy increases by an order

achieve process efficiency is fundamental.

of magnitude greater than density suggesting that flow energy measurements are significantly more

. R O H T U A E H T F O Y S E T R U O C E R A S E R U G I F

Focus on powder testing

sensitive in quantifying the impact of the change.

Tablets are the most common drug-delivery vehicle, Furthermore, this indicates how density changes and most drugs are handled in solid form at some

could be misleading when quantifying how con-

point, demonstrating the need for suitable powder

solidation impacts a process.

testing tools. Numerous methods for characteriz-

This experiment emphasizes the importance

ing powders exist, including angle of repose, f low of selecting a suitable analytical technique for a through an orifice, and tapped density. These sim- given application. It is increasingly acknowledged ple techniques provide some insight into the nature

that no single powder test suits every application



37

C�� M�� Figure 2: Dynamic testing can provide differentiation of materials that shear test s classify as identical.

2 2

5

1800

vanillin ethylvanillin

1600 1400

4

J m , y g r e n e l a t o T

a P k . s s e r 3 t s r a e h S 2

1200 1000 800 600 400 2

1

200 0 0

1

2

3

4

5

6

7

8

9

Applied normal stress, kPa

2

0 0 Test number Tip speed, mm/s

vanillin ethylvanillin

2 -1 00

- 10 0

- 10 0

4

6

8

-1 00

-1 00 -100 -100

-100

10 -70

-40

-10

and that tests should represent the conditions to

process design and enable effective monitoring

which the material is exposed.

and control. It is essential to consider what in-

Shear testing, for example, is a well-estab- formation is required and how to obtain it. Aplished technique developed to support hopper

plying this approach to powder characterization

design protocols advanced by Jenike (4, 5). It is

highlights limitations with traditional techniques.

still widely applied, with modern instrumenta- Innovative techniques such as dynamic testing tion delivering improved reproducibility, and is

and, in particular, instruments that combine dy-

a valuable tool for characterizing a powder in a

namic testing with methods such as shear and

static, consolidated state, but there are limitations. bulk property analysis, present an efficient and Figure 2 shows shear and flow energy data for versatile choice for those demanding new levels

two excipients: vanillin and ethylvanillin. Shear

of efficiency.

testing suggests these materials are identical, while dynamic testing identifies clear differ- References ences. In this case, the flow energy correlated with in-process behavior, illustrating how materials classified as identical by shear testing may actually process differently. This highlights the importance of employing methods that simulate process conditions.

Looking ahead The rise of continuous manufacture, and the drive for greater efficiency, increases the need for analytical techniques that support intelligent 38


1. Novarti s-MIT Center for Continuous Manufact uring website, https://novartis-mit.mit.edu/, accessed March 1, 2016. 2. Center for Structu red Organic Partic ulate Systems, Test Bed 1, Continuous Powder Manufacturing, http://csops.rutgers.edu/ research/test-beds/test-bed-1, accessed March 1, 2016. 3. FDA, Progress Report on Process Analytic al Technolog y, www.fda.gov/drugs/developmentapprovalprocess/manufacturing/questionsandanswersoncurrentgoodmanufacturingpracticescgmpfordrugs/ucm072006.htm, accessed March 1, 2016. 4. A.W. Roberts, Basic Principles of Bulk Solids Storage, Flow and Handling (The Institute for Bulk Materials Handling Research, Callaghan, NSW, Australia, 1993). 5. The Institut ion of Chemical Engineers/Europea n Federation of Chemical Engineering, Standard Shear Testing Technique for Par ticulate Solids Using the Jenik e She ar Cell , A Report of the EFCE Working Party on the Mechanics of Particulate Solids (1989). PT


PharmTech.com

Download

Case Study

“I need a CDMO that’s a true PARTNER .” With Rottendorf Pharma as your CDMO, you can be certain you are working with a �nancially stable company committed to you in a long-term relationship. We will know your goals and bring proven ideas to the table on how to achieve them. Need a true partner for your solid oral doses? Rottendorf owns it. www.Rottendorf.com/partner

MANUFACTURING

•

PACKAGING

•

F ORMULATION DEVELOPMENT

•

A NALYTICAL METHOD S

E�� I��

Meeting USP Guidelines for Elemental Impurity Analysis with X-ray Fluorescence Spectrometry Andrew Fussell

This article discusses the USP guidelines for monitoring elemental impurities in drug products.

T

here are strict limits and clear regulatory guidelines for the monitoring of elemental impurities in drug products. An important source of these impurities is the catalysts, which are necessary during drug production. These catalysts in-

clude palladium, platinum, rhodium, iridium, and ruthenium—all elements that, at high concentrations, can be harmful to patients.

Previously, the United States Pharmacopeia (USP ) chapter <231> (1) set the standard for the control of elemental impurities; however, the qualitative approach lacked selectivity and sensitivity and could fail to detect certain elements, such as mercury, at toxicologically relevant levels. For some time, it has been widely recognized within the pharmaceutical industry that the methodology set out in USP <231> is outdated, compared to more modern quantitative techniques. In response, the US Pharmacopeial Convention (USP) has developed new regulations, chapters <232> and <233> (2, 3), through a process of consultation with the industry, chemists, and toxicologists. USP <232> and <233> contain changes to the concentration limits for elemental impurities, as well as introducing flexibility in the choice of testing methods. In an effort to harmonize the regulations, the International Council for Harmonization (ICH) worked Dr. Andrew Fussell is the pharmaceutical segment manager at PANalytical. 40

together with representatives from the European Pharmacopoeia (Ph. Eur.), the Japanese Pharmacopeia, and USP to create the ICH Q3D guideline for elemental impurities (4). ICH Q3D has been ad-



PharmTech.com

S E G A M I Y T T E G / S E G A M I A R T E T

opted verbatim by the European Medicines Agency (EMA). The ICH Q3D guideline does not stipu-

Table I: Twenty elements spiked into pure cellulose, lactose, or calcium carbonate matrix. International Council for Harmonization (ICH) oral delivery risk assessment elements shown in bold.

late analytical procedures but instead states that

PAN standards

pharmacopeial procedures or suitable alternative

Concentration range (μg/g)

PAN standards

Concentration range (μg/g)

As

0-100

Se

0-200

Cd

0-100

TI

0-200

Pb

0-100

Cu

0-1000

released onto the market and legacy products, which

Hg

0-100

Zn

0-1000

will need to be retested.

Co

0-200

Mo

0-200

V

0-20 0

Ru

0-200

facturers choice in the analytical method used for

Ni

0-1000

Rh

0-200

the evaluation of levels of elemental impurities so

Cr

0-500

Pd

0-200

long as the technique has been validated in line

Mn

0-1000

Ir

0-200

Fe

0-1000

Pt

0-200

procedures should be used. These regulations (USP and ICH Q3D) will apply both to products newly

In a significant move, USP <233> allows manu-

with the requirements. One option emerging as

a favored alternative is X-ray fluorescence spec- ing. It can be used for the elemental and chemical trometry (XRF).

analysis of solid, powdered, and liquid samples, making it particularly valuable to pharmaceuti-

X-ray analysis in pharmaceuticals

cals. One of the main advantages of XRF is that it

Reliable, accurate analytical processes are the basis

requires minimal if any sample preparation prior

for most activities in the pharmaceutical industry

to analysis. In addition, the instrumentation is well

and X-ray-based analytical techniques underpin

adapted to automation.

many of these procedures. X-ray diffraction (XRD)

This is in sharp contrast to inductively coupled

is an established technique in the industry and is

plasma atomic emission spectroscopy (ICP–AES)

widely used for qualitative and quantitative analy- and inductively coupled plasma mass spectrometry sis of solid phases (5). XRD has long been the ac- (ICP–MS), which feature in the two sample methcepted technique for establishing the crystalline

ods outlined in USP <233>. ICP–AES uses ICP to

drug “fingerprint” needed for drug approval, pat- excite the atoms in a sample to emit electromagent descriptions, and for the identification of dif- netic radiation—the wavelength indicates the presferent drug batches (6). The use of XRD has high- ence of an element while the intensity indicates the lighted the general benefits of X-ray analysis to the

concentration. In ICP–MS, ICP is used to ionize

pharmaceutical industry, including the following:

a sample, while mass spectrometry separates and

t

Speed of analysis

quantifies the elements present. Both of these tech-

t

Simple (or no) sample preparation

niques are already used in parts of the industry;

t

Non-destructive measurement.

however, the sample preparation is intensive, and

In parallel, XRF is considered a proven tech- the dilutions required can lead to errors in analysis. nique for material analysis in a broad range of in-

XRF analysis can be divided into wavelength

dustries and applications from measuring sulfur in

dispersive (WD) and energy dispersive (ED) tech-

oil to analyzing coating thickness in metal finish- niques. The difference between these two techPharmaceutical Technology SOLID DOSAGE DRUG


41

E�� I��

Table II: International Council for Harmonization (ICH) and United States Pharmacopeia (USP) limits and validation sample set up. PDE is permissible daily exposure. Element

ICH PDE (μg/g)

ICH threshold (30%)

USP conc. limit for oral drug products (μg/g)

Calc. USP max. daily dose (g/ day)

Lower spike conc. (μg/g)

Upper spike conc. (μg/g)

As

15

4.5

3

5

1

9

Cd

5

1.5

2

2.5

1

9

Pb

5

1.5

2

2.5

1

9

Co

50

15

10

5

5

20

V

100

30

10

10

5

20

Ni

200

60

20

10

5

60

Cu

3000

900

300

10

15

150

Cr

11000

3300

1100

10

5

75

Mo

3000

900

300

10

5

120

Ru

100

30

10

10

5

15

Rh

100

30

10

10

5

15

Pd

100

30

10

10

5

15

Ir

100

30

10

10

5

15

Pt

100

30

10

10

5

15

inner-shell electrons of the elements in the sample.

Table III: Limits of detection—Epsilon 3X; PDE is permissible daily exposure; LLD is lower limit of detection. Element As

The interaction of the high energy X-rays with the

ICH PDE (μg/g)

ICH threshold (30%)

LLD (μg/g) Measurement (3m of blank) time (min)

electrons results in the emission of X-rays of char-

15

4.5

0.1

acteristic energies for the elements in the sample.

30 Pb

5

1.5

0.1

Cd

5

1.5

0.4

Co

50

15

0.1

Cr

11000

3300

0.2

Ni

200

60

0.1

Cu

3000

900

0.2

Mo

3000

900

0.4

V

100

30

0.2

Ru

100

30

0.4

Rh

100

30

0.9

Pd

100

30

0.9

Ir

100

30

0.3

Pt

100

30

0.3

The WD technique uses an analyzing crystal to

30

separate the characteristic X-rays for the different

30

elements, which are then detected by an X-ray detector. The ED techniques do not require an analyzing crystal, but instead, the X-ray detector itself

15

determines both the energy and the intensity of the characteristic X-rays (7).

15

In this study, EDXRF systems were used. EDXRF spectrometers discriminate each specific X-radia-

30

tion line based on the energy of the produced photon. Compared to other techniques, EDXRF spec-

5

trometers offer a number of advantages in that they

niques is due to the different detection methods. tend to be smaller, simpler in design, faster, have Both techniques typically use an X-ray tube to

fewer engineered parts, and are typically cheaper.

generate high energy X-rays that interact with the

In addition, the advances in XRF technology in

42



PharmTech.com

recent years make this method particularly attractive to the pharmaceutical industry, particularly as it is now a pharmacopeial method with USP

Table IV: Limits of detection—Epsilon 5; PDE is permissible daily exposure; LLD is lower limit of detection Element

ICH PDE (μg/g)

ICH threshold (30%)

LLD (μg/g) Measurement (3m of blank) time (min)

<735> X-ray fluorescence spectrometry and Ph.

As

15

4.5

0.8

10

Eur. chapter 2.2.37.

Cd

5

1.5

0.3

10

Pb

5

1.5

0.3

10

Co

50

15

1.7

Ni

200

60

0.4

Cu

3000

900

7.2

stances is from intentionally added metals, namely

V

100

30

0.4

metal catalysts. To combat this risk, the industry

Cr

11000

3300

0.3

has adopted a risk-based approach to assessing the

Mo

3000

900

0.2

potential presence of elemental impurities in drug

Ru

100

30

0.2

products. During the manufacture of drugs, risk

Rh

100

30

0.1

assessments have to be conducted that are in ac-

Pd

100

30

0.4

Ir

100

30

0.3

Pt

100

30

0.4

In practice It is recognized that the greatest risks in drug sub-

cordance with USP <232> or ICH Q3D, which lists the permissible daily exposure (PDE) for potential impurity elements in pharmaceutical materials. To demonstrate how XRF fulfills the guideline

5

5 5 5 5 5

Table V: Calibration results for the Epsilon 5. RMS is root mean squared. Element

Conc. range (μg/g)

R value

RMS error (μg/g)

criteria, pharmaceutical calibration standards for

As

0-100

0.9998

0.6

elemental impurity analysis were developed by

Cd

0-100

0.9999

0.3

spiking 20 elements into pure cellulose, lactose,

Pb

0-100

0.9989

1.6

or calcium carbonate matrix, as shown in Table I.

Co

0-200

0.9986

3.7

The calibration samples were prepared in triplicate

V

0-200

0.9999

1.0

using 1 g of loose powder, using a 6 µm polypro-

Ni

0-1000

0.9998

6.7

Cu

0-1000

0.9998

7.0

Cr

0-500

0.9999

2.9

Mn

0-1000

0.9996

10.8

Mo

0-200

0.9994

2.5

triplicate by mixing pure paracetamol (acetamino-

Ru

0-200

1.0000

0.6

phen) with known concentrations of test elements

Rh

0-200

1.0000

0.5

from the calibration materials. Each test element

Pd

0-200

0.9997

1.7

was validated at two different concentrations as

Ir

0-200

0.9999

1.1

required by USP <233>. In most cases, the vali-

Pt

0-200

0.9998

1.5

pylene foil. The results presented in this work are for the cellulose matrix. The validation samples were also prepared in

dation samples were above and below the target concentration. The powders were mixed for one

was prepared and filled with 1 g of sample. Next,

minute in plastic vials with ZrO2 milling balls. The

the sample was inserted into the instrument for

total sample preparation took less than five min- measurement. Table II shows the sample set up used. utes. For analysis, a loose powder sample holder The first system used was the Epsilon 3X, which Pharmaceutical Technology SOLID DOSAGE DRUG


43

E�� I�� Table VI: Validation spike results for Epsilon 5. USP is United States Pharmacopeia . Expected spike conc. (μg/g)

Element As

Cd

Pb

Co

V

Ni

Cu

Pd

Measurement conc. (μg/g)

Percentage recovery

Within USP <233> limit (70-150%)

Measurement time (min)

1

1.19 ± 0.03

119

YES

5

9

8.87 ± 0.22

99

YES

10

1

0.77 ± 0.14

77

YES

5

9

8.30 ± 0.12

92

YES

10

3

2.78 ± 0.21

93

YES

5

9

10.33 ± 0.15

115

YES

10

5

6.90 ± 1.32

138

YES

5

20

16.29 ± 0.01

81

YES

3

5

6.52 ± 0.13

130

YES

5

20

23.03 ± 0.31

115

YES

2.5

5

5.42 ± 1.29

108

YES

5

60

57.23 ± 2.34

95

YES

3

15

16.30 ± 1.58

109

YES

5

150

144.36 ± 6.87

96

YES

3

5

4.98 ± 0.11

100

YES

3

15

14.48 ± 0.04

97

YES

3

Figure 1: Calibration lines for As (lef t), Cd (middle), and Pd (right) show good linearity. 0.10

0.10

0.04

0.03

0.05 0.02

0.05

0.01

0.00

0.00

0.00 0

50

100

Concentration (μg/g)

0

50

100


0

50

100

150

200


is a benchtop EDXRF spectrometer with a 10-posi-

The samples were also analyzed with the Epsi-

tion sample changer. The analysis was performed in

lon 5, a floor standing EDXRF spectrometer with

air atmosphere not requiring any external gasses. A

a 133-position sample changer. For this instrument,

total one off calibration time was 15 hours for all 20

the measurements were performed in helium at-

elements. The measurement time used to obtain the

mosphere. The total one off calibration time was

detection limits (Table III) was between 5–30 minutes

significantly reduced when compared with the Ep-

per element, while the validation spike sample mea- silon 3X at 6 hours for all 20 elements. The limit surement was between 2.5–30 minutes per element. of detection measurement time and the validation Measurement time can be shortened depending on

spiked sample measurement time were also re-

the number of elements, sample size, and accuracy

duced to 5–10 minutes and 2.5–10 minutes per ele-

and precision requirements needed.

ment, respectively. Again, this time can be reduced

44



PharmTech.com

depending on the number of elements, sample size, as reduced potential for error through minimized and the accuracy and precision requirements. The

sample preparation, the non-destructive nature

limits of detection for the Epsilon 3X and Epsilon

of the analysis, reduced chemical waste, and the

5 are shown in Tables III and IV.

lower total cost of ownership and operation. Many

Good linearity was achieved for all the elements

companies are picking up the opportunity to use

tested (as noted in Table V), Examples of the cali- alternative techniques and taking advantage of the bration curves for As, Cd, and Pd are shown in

methods now available. It is important for these

Figure 1. Repeatability was ascertained by analyz- techniques to be fully explored, as it will allow

ing the standards measured 20 times consecutively the pharmaceutical industry to find the technique as unknowns. Further, the validation spike results

and methods that best suit their requirements.

shown in Table VI show a reasonable recovery percentage for a number of elements mixed as pow- References ders with paracetamol.

Conclusion By allowing the use of alternative techniques, such as XRF, USP is now supporting pharmaceutical manufacturers who want to make their own analytical choices. EDXRF was shown to be a powerful technique capable of reaching the required USP <232> and ICH Q3D PDE limits for oral drugs. XRF also provides a number of advantages over the USP <233> wet chemical techniques, such

1. 2. 3. 4.

USP, Chapter <231>, USP 35–NF 30, (USP, 2012). USP, Chapter <232>, USP 35–NF 30, (USP, 2012). USP, Chapter <233>, USP 35–NF 30, (USP, 2012). ICH, Q3D Guideline for Elemental Impurities (ICH, 2014), www.ich.org/fileadmin/Public_Web_Site/ICH_Products/ Guidelines/Quality/Q3D/Q3D_Step_4.pdf, accessed Feb. 22 2016. 5. Vyas K., X-ray Diff raction in Pharma Industry (2013), ww w. icdd.com/ppxrd/12/abstracts/P26.pdf, acce ssed Feb. 22, 2016. 6. Randa ll C., Rocco W., Ricou P. , XRD in Pharmaceutical Analysis: A Versatile Tool for Problem-Solving (2010), www.americanpharmaceuticalreview.com/Featured-Articles/115052-XRD-inPharmaceutical-Analysis-A-Versatile-Tool-for-Problem-Solving/, accessed Feb. 22, 2016. 7. Brouwer P., Theory of XRF–Get ting acquainted with the principles (book let) (2013), w ww.panalytical.com/Technologybackground/Theory-of-XRF-getting-acquainted-with-theprinciples-booklet.htm, accessed Feb. 22, 2016). PT

T�� C�� — contin. from page 33

maintains low weight variation) through the 75

necessary to expand the process optimization tem-

RPM press speed.

plate to incorporate representative granulations for nominal and more difficult batches.

Conclusion

Most modern tablet presses incorporate a prod-

Applying a methodical approach to tablet-com- uct recipe capability that will permit optimized pression process optimization and evaluating the

press settings to be stored and automatically re-

process control under a range of different condi- trieved to eliminate redundant optimization efforts tions will permit a full understanding of the pro- and expensive losses at start-up. In many cases, cess and will facilitate the opportunity to achieve

there are opportunities to implement process opti-

optimal tablet quality and production output. This

mization measures on existing products and obtain

template assumes limited batch-to-batch variabil- benefits from the resulting improvements in tablet ity. If batch-to-batch variability is high, it may be

quality and production efficiencies. PT



45

G�� D��

Regulatory Considerations for Controlling Intermediates in Type-II Drug Master Files for the Manufacture of Generic Drug Substances Kandasamy Subburaj, Brian T. Connell, Srinivasa Murthy, Humcha Hariprakasha, Deborah F. Johnson, Huyi Zhang, and David J. Skanchy

T

he Generic Drug User Fee Amendments of 2012 (Public The authors have undertaken an analysis of Law 112–144, Title III), commonly referred to as GDUFA, the current state of control was signed into law on July 9, 2012 and is designed to of intermediate identity speed the delivery of safe and effective generic drugs to and quality, based on the public. Since the law came into effect, Type II drug master file information submitted in a pseudorandom sample of 120 (DMF) and abbreviated new drug application (ANDA) submissions drug master files that have have reached unprecedented levels. been submitted to FDA. Quality by design (QbD) is an essential part of the modern approach to pharmaceutical quality (1). With the introduction and implementation of the International Council on Harmonization David J. Skanchy is a senior supervisory regulatory review officer and Kandasamy Subburaj, Bria n T. Connell, Srinivasa Murthy, Humcha Hariprakasha, Deborah F. Johnson, and Huyi Zhang, are chemists, all at the Office of New Drug Products under the US Food and Drug Administration’s Office of Pharmaceutical Quality within the Center for Drug Evaluation and Research. Disclaimer:

This article represents the views of the authors and not of FDA. 46

(ICH) guidelines (2–5) and FDA initiatives (6) in the past few years, pharmaceutical manufacturers have gradually and continuously adopted QbD principles and elements in the field. The ICH Q11 guideline further clarifies the principles and concepts described in ICH Q8 Guidelines on Pharmaceutical Development , ICH Q9 Quality Risk Management , and ICH Q10 Pharmaceutical Quality System as they pertain to the development and manufacture of drug substances. ICH Q11 also describes approaches to developing and understanding the manufacturing process of the drug substance. That guidance recognizes that critical quality attributes (CQAs) of a drug substance can be “(1) included on the specification and confirmed through testing the final drug substance, or (2) included on the specification and



PharmTech.com

S E G A M I Y T T E G / Y H P A R G O T O H P A T H I N A N

confirmed through upstream controls, or (3) not

undertaken an analysis of the current state of con-

included on the specification but ensured through

trol of intermediate identity and quality, based on

upstream controls.” One of the upstream controls

information submitted in a pseudorandom sample

is isolation of intermediate(s), where applicable, of 120 DMFs that have been submitted to FDA. with appropriate characterization and controls in While the results of this analysis are not intended place for impurities, which include starting ma- to be extrapolated to all other Type II DMFs curterials, previous intermediates, reagents, solvents, rently available for reference, they do provide some catalysts, and reaction by-products. ICH Q7 (7) insight into the current state of intermediate condefines an intermediate as “a material produced

trol. The results of this analysis are presented in

during steps of the manufacturing of an API that

the appropriate section throughout the remainder

undergoes further molecular change or purifica- of this article. tion before it becomes an API. Intermediates may

Identification and addresses the intermediates that can be isolated, so characterization of intermediates

or may not be isolated.” This article, however, only that intermediate specifications can be established

Regulation 21 Code of Federal Regulations (CFR)

and documented to ensure that quality attributes

314.50(d)(1)(i) describes the regulatory require-

are met.

ments of drug substance information that should

The isolation of intermediates, with correspond- be provided in an application and one of the key ing specifications, provides significant opportu- components is the proof of identity (The regulation nities to purge potential impurities that may be

requires “a full description of the drug substance

carried from the starting material(s) or earlier

including its physical and chemical characteristics

intermediate(s). This purging is especially impor- and stability; the name and address of its manutant for regulatory starting materials that may be

facturer; the method of synthesis (or isolation) and

outsourced from multiple suppliers. Isolation and

purification of the drug substance; the process con-

control of intermediates are ways to help mitigate

trols used during manufacture and packaging; and

the risk to drug substance quality when new start- the specifications necessary to ensure the identity, ing material suppliers have been changed/added. strength, quality, and purity of the drug substance Additionally, intermediate controls can serve as

and the bioavailability of the drug products made

the control point(s) for some CQA(s) of the drug

from the substance, including, for example, tests,

substance. Because intermediates are considered

analytical procedures, and acceptance criteria re-

to be critical materials in API production, the

lating to stability, sterility, particle size, and crys-

specifications should be designed to control the

talline form”) (8). The ICH common technical doc-

critical material attributes (CMAs) that may have

ument (CTD) format dedicates section 3.2.S.3.1 to

an impact on the CQAs of a drug substance. The

Elucidation of Structure and other Characteristics

CMAs of an intermediate are its chemical identity

of the drug substance (9, 10). Comprehensive struc-

and quality, which equate to its characterization

tural elucidation information is typically provided,

and level of impurities present. The authors have

including but not limited to infrared ultraviolet/



47

G�� D�� niques should be used to

Figure 1: Routine assay and regioisomer control.

distinguish these isomers, No 13% API 44%

Yes 87%

Intermediate 17% Unknown 39%

accompanied by a precise interpretation of this supporting information. For example, to establish the E/Z configura-

Routine Intermediate Assay/Purity Test

Regioisomer Control Point

tion of an alkene that is introduced by a Wittig reaction in an intermedi-

visible spectroscopy, mass spectrometry, elemental

ate stage, both 1D and 2D NMR spectra can be re-

analysis, one- and two-dimensional 1H and 13C nu- corded and interpreted. Proton coupling constants, clear magnetic resonance spectroscopy, and X-ray

nuclear Overhauser effect (NOE) enhancements,

crystallography in some cases. For many synthetic

and 1H-13C long-range multiplet bond correlation

small-molecular drug substances, the synthetic

(HMBC) are methods that may provide solid evi-

route, including the choice of starting materials

dence of the configuration of a particular alkene

and reagents, and the process design and controls, isomer. Only 87% of the intermediates examined also serves as supplementary confirmation of the

included one assay or purity test (Figure 1). In cases

drug substance structure. Therefore, the character- where regioisomeric impurities were possible, analization of intermediates is often considered to be a

ysis revealed that the control point was not explic-

verification of structure and can be done by an IR itly discussed in 39% of the sampled intermediates, spectra and a chromatographic comparison to an

while 17% had a control for these impurities in the

established reference standard. Although it would

intermediate and the regioisomeric impurity, or

be rare that the manufacturer has synthesized a

downstream regioisomeric API analogs, was con-

completely different molecule, or even the wrong trolled in the final APIs for the remaining 44% of enantiomer or diastereomer, there are cases where the identity of the wrong compound was taken for

the intermediates studied. It is important for the applicant to sufficiently

granted, such as the one described recently (11). address the characterization issue of the intermeIn fact, 96% of the intermediates examined in the

diate especially when certain structural features

authors’ study included at least one identity test. would not be unambiguously determined by simHowever, often times an enantiomer, diastereo- ple techniques. It is equally important for the applimers, or regioisomers are possible at a given step, cant to provide the interpretation of the supporting or a semi-synthetic process isolates/extracts a com- information, describing how the identity and/or plex structure as an intermediate. In such cases

the challenging structural features are assured by

where the manufacturer relies on the identity of

the data provided. Data can also be collected on

the intermediate to infer a structural feature of the

the undesired isomer(s) to highlight the contrast-

final drug substance, appropriate analytical tech- ing spectral features. If an undesired isomer is not 48



PharmTech.com

. S R O H T U A E H T F O Y S E T R U O C E R A S E R U G I F L L A

Figure 2: Manufacturing process of the drug substance.

A

Step 1

B

Step 2

Step 3 C (RSM)

D

Step 4

E

Step 5

F (Crude drug substance)

Step 6

G (Final drug substance) Puri�cation

available, the applicant should consider describing

turer can consider implementing single or multiple

the data that would be expected if any undesired

points of control for a specific CQA, particularly

isomers were formed, so that it is clear that the

impurities, depending on the risk associated with

proposed specifications are specific to the desired

the CQA and the ability of individual controls to

intermediate and not to its closely related isomers. detect potential problems. There are many options This notion also applies to the full characterization

available to the manufacturer for controlling im-

of final drug substance.

purities in a drug substance. The decision about where to control each particular impurity is up

Impurity control in intermediates

to the applicant. Some of the more common op-

Impurities are unwanted materials present in the

tions include controlling all impurities in the final

API that have no therapeutic value, may be harm- drug substance at their respective ICH limits (the ful, and therefore, need to be controlled. Potential

traditional approach); at an intermediate stage in

impurities at a given intermediate stage include

the process at their respective drug substance ICH

residual starting materials and previous interme- limits; or at an intermediate stage in the process at diates, impurities carried over from the starting

higher than their respective drug substance ICH

materials, and previous intermediates, reagents, limits when coupled with demonstration that the solvents, catalysts, and reaction by-products. The

impurities are absent in multiple batches of the

identification of reaction by-products requires

drug substance and thus do not require routine

some chemistry knowledge, previous experience, drug substance testing. and literature knowledge. Investigation of batch

Impurities may be divided into three categories:

analyses for observed impurities at ICH Q3A (12)

organic impurities, residual solvents, and inor-

levels is the key step to determine what impurities

ganic impurities. Each category is discussed sepa-

need to be controlled. It is essential for applicants

rately in the following sections.

to demonstrate process understanding by examin-

Organic impurities. In the fictional example pro-

ing the origins and fates of impurities to support vided in Figure 2, intermediate C is the proposed the proposed controls at various stages.

regulatory starting material (RSM). The crude

The control of impurities should always be a

drug substance F, an intermediate, was manu-

component of a drug substance overall control

factured in three steps (steps 3, 4, and 5) from

strategy. Under the QbD paradigm, the manu- the RSM C via intermediates D and E. RSM C is facturing process of the drug substance and the

considered to be a critical material in the overall

impurity profile must be understood step-by-step. manufacturing process and cGMP guidance apWhen developing a control strategy, a manufac- plies from this step of the manufacturing process. Pharmaceutical Technology SOLID DOSAGE DRUG


49

G�� D�� cess based on multiple

Figure 3: Routine impurity control.

Between 3%-10% 27%

Greater than 10% 8% Content less than 2% 65%

Between 2%-5% 25%

Greater than 5% 7% Content less than 2% 68%

batch analyses. Inclusion of a study demonstrating how the amount of a particular organic impurity decreases throughout the manufacturing process

Content of each individual impurity

Total impurity content

from the proposed limit to an acceptable level in

CQAs of RSM C include assay, organic impu- the final drug substance would provide stronger rities, residual solvents, and residual metal from justification than the submission of end testing catalysts carried over from Steps 1 and 2. These

data alone. As long as RSM C is a process-related

attributes should be controlled with appropriate

impurity, and not also a degradant that can be

limits in RSM C, thereby reducing the risk of fur- (re)generated in later steps, further control in the ther carryover. ICH Q11 clearly states that the ana- drug substance may not be warranted. Multi-point lytical methods for the regulatory starting material

control of impurities can also be implemented for

should be capable of detecting impurities in the

intermediates up to and including the drug sub-

starting material, and the fate and purge of these

stance. A relatively high limit of residual reagent X

impurities and their derivatives in subsequent pro- in intermediate D can be controlled at a lower limit cessing steps should be understood.

in intermediate E, and then shown to be absent in

Often the proposed specification of the RSM C

the drug substance. The impurity, therefore, will

is too wide or does not list all possible impurities. be well controlled and the chance of carr y over to When the assay of the starting material is set to a

the drug substance will be slim.

wide range (e.g., Not less than 80%), the applicant

It is often beneficial for the sponsor to propose

should discuss what makes up the remaining (20%)

a limit for impurity Y in the RSM C specification

of the mass balance and how this will affect, via

with an acceptance criteria at or below the accept-

the downstream reaction steps (Steps 3 thru 6), the

able drug substance limit (based on ICH Q3A and

ultimate purity of the drug substance.

MDD of the drug substance), so that monitoring

When the limit for a particular impurity (e.g., re- of Y is optional in the drug substance release specisidual RSM C in intermediate D) is set higher than

fication (unless the impurity is also a degradant).

the ICH limit (based ICH Q3A and maximum

Alternatively, the limit for impurity Y in RSM C

daily dose of the drug substance), then the appli- can be proposed at a higher limit than the acceptcant needs to demonstrate that residual RSM C

able drug substance limit. In this case, the spon-

(and any by-products formed due to its continued

sor must demonstrate by a spike and purge study

reaction) will be purged out to the acceptable ICH

that impurity Y is purged out during the successive

level in the drug substance during the successive

steps of the manufacturing process when present at

steps (Steps 4, 5, and 6) of the manufacturing pro- the maximum acceptable limit in the RSM C. The 50



PharmTech.com

same procedure can be used when setting a limit for an organic process impurity in intermediates

Figure 4: Control of potentially genotoxic impurities (when discussed).

D, E, or F. It is important to note that spike/purge studies should be performed in a manner represen-

Inadequate 37%

tative of the commercial process in order for the results to be predictive.

Partially adequate 44%

19% Fully adequate

It is often the case that the manufacturing of RSM C or the intermediates involves the use of potentially genotoxic material (e.g., substituted amino- or nitro-aromatics, alkyl halides, sulfonyl

chlorides, or other potentially genotoxic reagents). based the process data provided in the DMFs. In the case that impurity Z in RSM C possesses

Overall, 79% of the proposed individual impu-

a genotoxic alerting functionality, it can be con- rity limits were adequately justified. Additionally, trolled in RSM C at a permitted daily exposure, total impurity content was controlled at less than based on ICH M7 Assessment and Control of DNA 2% in 68% of the intermediates examined, at beReactive (Mutagenic) Impurities in Pharmaceuti- tween 2%–5% for 25% of the intermediates, and cals to Limit Potential Carcinogenic Risk (13) and

at greater than 5% for 7% of the intermediates.

the MDD of the drug substance. Such control

Overall, 92% of the proposed total impurity lim-

is sufficient to negate the need for any further

its were adequately justified based on the process

downstream control of impurity Z. Alternatively, data provided in the DMFs. a higher limit can be set in RSM C, only if it has

However, only 66% of the sampled DMFs that

been established by a spike and purge study that

use potentially genotoxic impurities (PGIs) as raw

this impurity Z is purged out during the successive

materials or intermediates provided any discussion

steps of the manufacturing process when present

of their control. Of those that did, 63% provided at

at the maximum acceptable limit in the RSM. As

least some appropriate discussion and justification

another option, the sponsor can provide pharma- of their control strategy. The remaining 37% were cology/toxicology data to justify a higher proposed judged to be inadequate based on the available inlimit. The same procedures can be used when set- formation (Figure 4). ting a limit for a potentially genotoxic process impurity in intermediates D, E, or F.

In approximately 25% of the sample of DMFs, the undesired enantiomeric impurity is a possible

In the authors’ analysis, 65% of intermediates

process impurity. More than 97% of the time, the

controlled individual impurities at not more than

resulting enantiomeric impurity is controlled in

2% each, 27% of intermediates controlled indi- the API rather than in an intermediate. vidual impurities between 3%–10%, and the re-

As can be seen in these examples, it is necessary

mainder (8%) had at least one impurity controlled

to have a thorough understanding of the manu-

above 10% (Figure 3). However, 21% of the proposed

facturing process of the drug substance, including

intermediate impurity limits were not justified

the steps leading to the RSM, so that the potential



51

G�� D�� Figure 5: Potential unwanted side reactions.

HOMe + CH3SO2C1

CH3SO2OMe (requires genotoxic impurity control)

HOEt + CH3SO2H

CH3SO2OEt (requires genotoxic impurity control)

for carryover of impurities to the drug substance

higher than the USP <467> limit if it can be demon-

can be minimized.

strated that the process is capable of removing that

Residual solvents. The control of residual solvents

solvent to the appropriate level in the final drug sub-

should always be a component of a drug substance

stance. A decision not to test for a particular solvent

overall control strategy. Residual solvents may be

needs to be justified through adequate supportive

controlled through final testing on a drug substance

data. Inclusion of a study demonstrating how the

or during the manufacturing process as in-process

amount of a particular residual solvent decreases

controls or components of an intermediate specifi- throughout the manufacturing process from the cation. The decision about where to control a par- proposed limit to an acceptable level in the final ticular residual solvent is up to the applicant.

drug substance would provide stronger justification

The first option simply requires that all solvents used in the process be specified and tested for in

than the submission of end testing data alone. There are times when controlling a residual sol-

the final drug substance with acceptance criteria vent at an intermediate stage may be especially imless than or equal to the United States Pharma- portant. Tightly controlling residual amounts of copeia (USP ) chapter <467> limit for that solvent. lower chain alcohols in intermediates, for example, Class III solvents can be controlled through in- may be important when sulfonic acids (e.g., methdividual testing or by a loss on drying test with a

ane sulfonic, benzene sulfonic, or p-toluene sul-

limit of not more than 0.5%.

fonic acids) or their corresponding acid chlorides

It may be more efficient to control solvents that

are used in the next step(s) of the manufacturing

are used early in the drug substance manufactur- process (Figure 5). Such tight control, at levels far ing process through in-process tests or as part of

below the USP <467> limits, will mitigate the risk

the specification of a RSM or intermediate, as op- of potentially genotoxic alkyl sulfonic ester impuposed to including them in the drug-substance

rity formation and carryover to the drug substance.

release specification. This is especially true if a

Conversely, controlling residual amounts of

solvent is used only during the early steps of the

alkyl sulfonic acid or acid chlorides may be im-

manufacturing process and not at any later point. portant when alcohols are used in the following Residual solvents that can be shown to be at or

steps of the process. There may also be cases in

below their USP <467> limit in an intermediate

which residual alcohols or other solvents have the

do not warrant testing in the final drug substance, potential to react with the drug substance to form if they are not used in the later part of the manu- undesired side products. facturing process.

When these types of materials are part of a man-

A Class II or III residual solvent may be controlled

ufacturing process, it is important that a thorough

upstream in the manufacturing process at a level

discussion is presented about the possibility of

52



PharmTech.com

sulfonic acid esters being present and the control

Figure 6: Potentially genotoxic impurities byproduct control.

strategies put in place to ensure that the risk of carryover to the final drug substance is minimized. This should be a preemptive discussion on part of

Not controlled 31%

Controlled 69%

the applicant, not the result of a question coming from a regulatory agency. In this analysis, only 69% of the intermediates, where this type of PGI was possible, were adequately controlled (Figure 6). The remainder required additional justification. Class I solvents are rarely used, but if a process

controlled in the drug substance through a residue

does use this type of solvent, it must be tested, ei- on ignition test. Transition metals and some main ther at an intermediate stage or in the final drug

group metals, due to their higher level of toxicity,

substance. Non-testing of a Class I solvent is not

require specific controls. The US Pharmacopeial

acceptable. Most Class I solvents, such as benzene, Convention is in the process of introducing Genhowever, can be introduced into a process as im- eral Chapter <232> Elemental Impurities, which purities of another solvent. If large quantities of

will update the types of analytical methods used

solvents such as toluene, hexane, acetone, or meth- to monitor residual metals and set safe limits for anol are used near the end of a manufacturing pro- patient exposure. The ICH Q3D guideline will process, then residual benzene needs to be controlled. vide “a global policy for limiting metal impurities Benzene can be controlled as part of an intermedi- qualitatively and quantitatively in drug products ate’s specifications or in the final drug substance. A and ingredients” (14). discussion and supporting data should be provided

Many transition metals, such as palladium

to justify the proposed benzene limit in the process

or platinum, are introduced into a process as

solvent in order to ensure that the benzene level in

catalysts absorbed onto carbon. The removal of

the drug substance is less than 2ppm.

these metals is usually accomplished by filtra-

Whatever control strategy is chosen, it should

tion through some type of filter aid. If the f iltra-

be accompanied by a well thought-out discussion

tion process is not thorough, residual amounts

of the facts and not justified by simply submitting

of these metals may carry through and would

the analytical results of multiple batches of drug

most likely not be removed by other purification

substance showing the absence of the solvent.

methods such as extraction or recrystalliz ation.

Inorganic impurities. The control of metal impuri- Other metals may be removed from the process

ties in a drug substance is equally as important as

by precipitation of an insoluble salt or extrac-

the control of organic and residual solvent impuri- tion using a chelating agent. Whatever process ties. Residual metals are typically introduced into

is used for the removal of the metal, a control

a drug substance as part of a reagent or catalyst.

strategy for residual amounts of that metal must

Some metals may be completely innocuous, in- be in place. This control may be done by adding cluding sodium and potassium, and may simply be Pharmaceutical Technology SOLID

a test for the metal to an intermediate specifica-


53

Pharmaceutical Technology eBook Solid Dosage Drug Development and Manufacturing

Recommend Documents