CLINICAL TRIALS AND TRANSLATIONAL MEDICINE
COMMENTARY
Drug Delivery Trends in Clinical Trials and Translational
Medicine: Updated Analysis of ClinicalTrials.gov Database
RODNEY J.Y. HO,1 JENNY Y. CHIEN2
1
Department of Pharmaceutics, University of Washington, and Fred Hutchinson Cancer Research Center,
Seattle, Washington
2
Eli Lilly & Company, Lilly Research Laboratories, Indianapolis, Indiana
Received 4 November 2008; accepted 6 November 2008
Published online 30 December 2008 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.21649
ABSTRACT: While the number of clinical trials has continued to grow by about 20% in
the past six months, no corresponding growth in product approval by the food and drug
administration is seen or anticipated in the near future. Late-stage clinical failures due
to lack of efficacy or toxicity continues to be a challenge. The optimization of absorption,
distribution, metabolism and elimination (ADME) has improved drug candidate selec-
tion and reduced early clinical failure. The current challenge is how to avoid late stage
clinical failures. Expanded knowledge of drug target distribution, pharmacokinetics and
validated biomarkers will allow implementation of appropriate drug delivery and clinical
trial designs to reduce drug exposure to off-target organs such as the liver and kidney
and could reduce potential untoward effects. In essence, integration of drug delivery and
targeting to reduce exposure in off-target tissues in the preclinical and clinical program
may hold the key to increasing the odds of success in drug development. In this update,
we briefly review data on clinical trials pertinent to drug delivery in the current
regulatory environment. It also provides our analysis on the emerging trends in second
generation antibody therapeutics in drug delivery and targeting. ß 2008 Wiley-Liss, Inc.
and the American Pharmacists Association J Pharm Sci 98:1928 1934, 2009
Keywords: biotechnology; DNA/oligonucleotide delivery; drug targeting; drug
transport; formulation; gene delivery; liposomes; nanoparticles; peptide delivery;
polymeric drug delivery systems
While the number of clinical trials has continued anticipated in the near future. Late stage clinical
to grow by about 20% in the past 6 months, no trial failures due to lack of efficacy or toxicity
corresponding growth in product approval by the continues to challenge pharmaceutical companies
food and drug administration (FDA) is seen or and public sponsors. The increasing number of
recalls, blackbox warnings and class action law
suits highlight the public demands for a higher
standard of safety while the FDA regulatory
Rodney JY Ho and Jenny Y Chien, Associate Editors,
Journal of Pharmaceutical Sciences. review balances risks and benefits.1 Thus, in
Correspondence to: Rodney J.Y. Ho (Telephone: 206-543-
current risk-avert climate, the FDA has re-
9434; Fax: [phone];
emphasized safety being the top priority for
E-mail: [e-mail])
review of data submitted as a part of new drug
Journal of Pharmaceutical Sciences, Vol. 98, 1928 1934 (2009)
ß 2008 Wiley-Liss, Inc. and the American Pharmacists Association applications (NDA).2
1928 JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 6, JUNE 2009
DRUG DELIVERY TRENDS IN CLINICAL TRIALS AND TRANSLATIONAL MEDICINE 1929
The optimization of absorption, distribution, netic/pharmacodynamic (PK/PD) interactions.
metabolism and elimination (ADME) has im- Even with a much more narrow definition of drug
proved drug candidate selection and greatly delivery as formulation and drug carriers, the
reduced early clinical failure in the past 10 years. U.S. market is estimated at $57 $82 billion with a
The current challenge is how to avoid late stage 6 9% annual growth rate. Regardless of one s
clinical failures. The problems that arise in the definition, novel drug delivery platforms and
late-stages of clinical trails are often not predicted strategies are and will continue to play a central
by preclinical ADME data, but rather, point to role in the global effort to improve safety, efficacy
toxicity in off-target tissues such as liver, heart, and cost-effectiveness of new or existing medicinal
and kidney. Recent research emphasis by most products. [For additional details on definition of
major pharmaceutical companies to define, early key categories, refer to Table 1 and text below].
in development, biomarkers predictive of efficacy Some of these efforts will add to the past focus on
or toxicity for each drug target entering clinical compounds that exhibit narrow therapeutic index.
trials or development program will likely provide Since our last report, there was an approximately
some relief. Another important and complemen- 22% increase in documented drug-delivery related
tary solution for preventing late stage clinical clinical trials in which drug delivery system/
failure will likely come from scientists ability to technology category contributing 27% (Table 2).
define tissues and cells localization of the ther- However, these numbers reflect predominantly
apeutic targets and distribution of drug candi- drug delivery technologies that advanced to
dates to off-target tissues. Knowledge of drug clinical evaluation. Therefore, in our commen-
target distribution, pharmacokinetics and vali- taries, we will summarize and update the readers
dated biomarkers will allow implementation of with drug delivery technologies currently in
appropriate drug delivery and clinical trial development and discuss emerging trends in
designs to reduce drug exposure to off-target the translation of drug delivery concepts into
organs such as the liver and kidney and could pharmaceutical products.
reduce potential untoward effects. In essence, According to the FDA Modernization Act of
integration of drug delivery and targeting to 1997 and the 2007 Food and Drug Amendments
reduce exposure in off-target tissues in the Act, all drug, biologic and device clinical trials
preclinical and clinical program may hold the now require registration. Consequently, all the
key to improving the odds of success in drug mandatory clinical trial information filed with
development. ClinicalTrials.gov is now a centralized resource.
In our last commentary,3 drug delivery was At the time of this writing, there were 63,362
defined broadly as physical dosage form, mole- trials listed in this database, of which 55,339 are
cular design or other physical approaches built on evaluating treatment effects in 158 countries
the fundamentals of metabolism, pharmacoki- around the world, including the U.S. With the
Table 1. Categorical Organization of Descriptors Found in ClinicalTrials.gov
(I) Drug Delivery Technology, (II) Biological Molecule (III) Drug Metabolism
System and Device Platform/Technology and PK/PD Interactions
Device Antibody Drug metabolism inhibitor
Drug delivery system Biologics and vaccines Drug transport modulator
Dosage form Peptide Drug interactions
Formulation Recombinant proteins Drug metabolic induction
Formulation comparison Anitbody conjugates Active metabolite
Transdermal Antisense
Aerosol OR Inhalation Oligonucleotide
Route siRNA
Sustained release Aptamer
Lipid formulation
Liposome
Nanoparticles
Microparticles OR microcarriers
Prodrugs
Colloid
DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 6, JUNE 2009
1930 HO AND CHIEN
Table 2. The Increase in Number of Drug Delivery Related Clinical Trials Between
March and October 2008
March October Net
Categories 2008 2008 Change (%)
(I) Drug delivery technology and system 9807 12,492 2685 (27.4)
(II) Biological molecular platform 5333 5979 646 (12.1)
(III) PK/PD interactions 2445 3031 586 (24.0)
Subtotal 17,585 21,502 3917 (22.3)
up-to-date clinical trial data as a starting point, to dose recommendations and dose adjustment
we develop our commentaries to inform scientists in patients to achieve targeted exposure to the
in the pharmaceutical science community about parent drug or its active metabolites as new
the translational trends in the state-of-the-art drug entities. The safety issue with terfenadine
drug delivery technologies and innovative for- (Seldane) and its replacement with metabolite
mulation strategies. The commentaries as well as fexofenadine (Allegra) is a good example.4 In
data summarized in the tables and graphs are also addition, under the   Study Type  search field,
available online at the Journal of Pharmaceutical specifying trial type as   Interventional trials 
Sciences website; http://www3.interscience.wiley. searches for studies on evaluation of therapies.
com/journal/117935713/grouphome/home.html. Thus, search results can be further refined by trial
We have developed a set of descriptors that are status (  open  or   closed  for recruitment), by
grouped into three categories for analysis of therapy type, sponsor, study specific information,
trends in clinical trials. The three categories as geography, patient/subject demographics, or
shown in Table 1 are (1) Drug Delivery Technol- phase of development.
ogy, System and Device, (2) Biological Molecule In this update, we will briefly review data on
Platform or Technology, and (3) Drug Metabolism clinical trials pertinent to drug delivery technol-
and PK/PD Interactions. The first category covers ogies or issues associated with specific drug
all the known drug delivery devices and systems, delivery systems (including strategies related to
including biopolymers, drug carriers as well as metabolism, transport and drug drug interac-
pro-drug platforms. Each specific terminology is tions) in the current regulatory environment. This
entered in the   search term  field. The rationale will be followed by our analysis on the emerging
for the choice of these broad and overlapping trends in second generation antibody therapeutics
descriptors/search terms is to maximize search in drug delivery and targeting.
results for subsequent analyses. (Please logon to
the accompanying online version of these tables
will provide embedded links for readers to explore
additional details for specific trials.) Therefore, OVERVIEW OF DRUG DELIVERY
these categories are by no means comprehensive. TECHNOLOGY IN CLINICAL DEVELOPMENT
The second category highlights biological and
molecular approaches, including recombinant While industry spending on discovery of novel
proteins, antibody derivatives, peptides and drug delivery technology has stabilized in the past
oligonucleotide platforms including siRNA and 5 years, much focus has now turned to the evalua-
aptamer technologies. Each of these technology tion and application of existing mature or most
platforms has unique development issues. For promising technologies. As shown in Table 3, the
instance, development strategies and issues for number of trials evaluating dosage forms or
conjugated antibody are different from antibody formulations appears to be comparable, but they
due to its molecular modification. The category of are lower than those that evaluate devices.
drug metabolism and PK/PD interaction strate- Therefore, the general trend in clinical trials
gies pertains to metabolic drug drug interaction may reflect marketing driven factors or tailor-
mediated enhancement or reduction in overall ing to clinical needs (safety and tolerability) in
drug exposure. Although novel therapies based on patient subpopulations. Surprisingly, few trials
the knowledge of drug metabolism and PK/PD are focused on improving delivery based on drug s
and are not traditionally considered as drug ADME properties with dosage forms, drug deliv-
delivery approaches, such strategies often lead ery systems or devices.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 6, JUNE 2009 DOI 10.1002/jps
DRUG DELIVERY TRENDS IN CLINICAL TRIALS AND TRANSLATIONAL MEDICINE 1931
Table 3. Summary of the Number of Clinical Trials Listed in ClinicalTrials.gov Organized as the Three Major Drug
Delivery Categories (Updated on October 16, 2008)
Sum of
All Intervention Phase I Phase II Phase III Phase IV Phases
(I) Drug delivery technology and system
Device 4691 3951 443 665 703 781 2592
Dosage form 3297 2885 636 958 716 331 2641
Drug delivery system 1464 1395 301 476 358 175 1310
Formulation 3057 2651 545 855 677 317 2394
Liposome 302 298 91 67 48 14 220
Transdermal 287 274 30 63 107 53 253
Formulation comparison 338 311 73 72 81 47 273
Route 289 275 77 80 60 48 265
Sustained release 153 146 18 38 71 19 146
Lipid formulation 106 96 18 28 18 15 79
Nanoparticles 68 61 19 44 3 1 67
Aerosol and Inhalation 64 63 10 12 20 12 54
Prodrugs 49 47 21 18 8 1 48
Colloid 40 39 0 5 9 4 18
Drug delivery tech and system Subtotal 12,492
(II) Biological molecule platform/technologies
Antibody 3215 3037 905 1504 680 146 3235
Biologics and vaccines 1509 1483 623 655 468 147 1893
Peptide 874 787 272 322 119 82 795
Recombinant proteins 343 327 146 122 54 16 338
Antibody conjugates 182 176 29 54 81 17 181
Antisense 70 69 38 38 11 0 87
Oligonucleotide 81 73 33 38 11 2 84
siRNA 23 18 7 6 6 1 20
Aptamer 9 9 3 4 4 1 12
Biological molecule platform Subtotal 5979
(III) PK/PD interactions
Metabolic inhibitor 1111 1060 110 210 368 273 961
Drug transport modulator 833 816 93 184 292 198 767
Drug interactions 978 789 283 156 73 100 612
Metabolic induction 300 257 24 63 38 47 172
Active metabolite 110 109 51 27 7 15 100
PK/PD interactions Subtotal 3031
All the clinical trial data filed with ClinicalTrials.gov database were sorted according to the delivery platform listed under the
three major categories. All clinical trials include observation and intervention studies. As not all intervention studies are Phase I IV
clinical studies, the Sum of Phases will be the same as that of the number all Intervention studies.
The total number of clinical trials and inter- reflecting the trend of maturation and probability
vention trials associated with drug delivery of clinical success.
systems or devices or select molecular technology As one might expect, the number of clinical
or platforms descriptors are summarized accord- trials using antibody and antibody conjugate
ing to respective phase of development (Table 3). platform continues to dominate and is signifi-
Interestingly, 3037 clinical trials (51% in the cantly higher than the number of trials evaluating
biological and molecular platform) are associated recombinant proteins. What is not expected,
with antibody technology. The majority of the however, is the higher number of ongoing trials
trials are listed in early (i.e., Phase 2) stage of with peptides as a molecular delivery platform
development (Fig. 1) with only small percentage than with recombinant proteins. Perhaps
(6%) of the trials conducted in Phase IV. Overtime, advances in molecular design, allowing for the
the numbers may increase in Phases III and IV, development of small peptides that resemble
DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 6, JUNE 2009
1932 HO AND CHIEN
ciated with delivery systems, accounting for 26%
of the trials in the Disease category. Cancer (24%)
and infections (13%) are the second and third most
common disease categories. This is followed by
diabetes (12%), and central nervous system and
behavioral/mental diseases (10% each). Approxi-
mately 5% of trials are associated with HIV/AIDS.
However, this high-level summarization most
likely reflects global concern associated with
mortality and morbidity in those disease states
rather than the value of alternative delivery
technology to those organ systems.
Figure 1. Percent distribution of clinical trials asso-
EMERGING TRENDS IN ANTIBODY-BASED
ciated with biological/molecular technology by phase of
THERAPEUTICS
clinical trial being conducted.
Antibody-based therapeutics is the fastest grow-
active-site motifs, has paved the way to avoid
ing platform for developing new products for
using large recombinant proteins. It is also
cancer diagnosis and treatments to autoimmune
noteworthy that siRNA and aptamer platforms,
and cardiovascular diseases. With 22 antibody
while low in number, have begun to fill the space
therapeutics that produce over 18 billion dollars
in the various phases of drug development. A
sales in 2006, there are at least 350 antibody
surprising finding is the low number of trials
molecules in clinical development that span 3215
evaluating prodrugs; a more in-depth analysis of
clinical studies. The early work on immunoglo-
this subject area is needed and will be addressed
bulin biology, biotransformation and pharmaco-
in a commentary dedicated to prodrugs in the
kinetics provided solid foundations on how
future. Clearly, further mining of the database is
immunoglobulin G, IgG, antibodies distribute
needed in the specific drug delivery areas of
and are cleared in vivo. It produces a predictable
interest to the readers. This database summar- and sustained circulating plasma half-life around
ization provides an excellent starting point.
21 30 days in the absence of binding to specific
We also evaluated queries of interest to thera- targets that remove the IgG. While the versatile
peutic area or disease experts, using the terminol- therapeutic potential and advantage of long half-
ogies for organ system or disease category in the
life were recognized early, a number of technical
ClinicalTrials.gov database, searched under drug
challenges, in producing human monoclonal or
delivery systems (Table 4). As summarized in
recombinant antibodies, limited antibody applica-
Figure 2, the cardiovascular is the most common
tion in the 1970s to acute or sub-acute human
organ system evaluated in clinical trials asso- diseases.5 Recent technological advances fueled
Table 4. Summary of the Number of Clinical Trials Related to Device, Drug Delivery System, Formulation or
Dosage Form that are Listed According to the Disease Target or Organ Systems
Disease All Intervention Phase I Phase II Phase III Phase IV Sum of Phases
Cardiovascular 1899 1704 192 342 447 413 1394
Cancer and neoplasm 1721 1584 471 753 256 62 1542
Central nervous system 811 686 132 189 166 85 572
Immune system 1055 962 236 384 190 95 905
Diabetes, endocrine and metabolism 915 799 127 202 169 165 663
Digestive system 699 636 117 225 129 70 541
Infection 933 839 197 258 184 110 749
Behavioral and mental disorders 790 689 80 187 183 117 567
AIDS 330 299 96 74 53 30 253
Other 908 615
Subtotal 8813
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 6, JUNE 2009 DOI 10.1002/jps
DRUG DELIVERY TRENDS IN CLINICAL TRIALS AND TRANSLATIONAL MEDICINE 1933
rationale for Fc sequence modification of IgG
antibody is to reduce receptor mediated cytotoxi-
city or inflammatory responses related to Fc
domain. Decoupling binding and neutralization
of target with that of biological functions of IgG
will be particularly useful for removing circulat-
ing targets such as tumor-necrosis-factor (TNF) in
autoimmune conditions and amyloid precursor
protein in Alzheimer s disease. While there is
little doubt that binding affinity of antibody will
provide efficacy, the antibody binding also tends
to induce inflammatory responses. Antibodies
capable of neutralizing these antigens without
eliciting potential Fc-mediated inflammatory
Figure 2. Percent distribution of clinical trials asso-
cytokine response will likely improve clinical
ciated with delivery systems, dosage form, formulation,
safety. The clinical clue to the possibility of
or devices by diseases. [Color figure can be seen in the
decoupling binding and Fc-mediated function
online version of this article, available on the website,
was discovered and demonstrated in patients
www.interscience.wiley.com.]
receiving Rituxan (rituximab). Rituxan antitumor
activity in Non-Hodgkin s lymphoma (NHL)
significant progress in transition from mouse to requires rituximab antibody binding to Fc recep-
human monoclonal antibodies to reduce immune tors in the patients to produce anti-cancer
response and provide antibodies for chronic response. The Fc receptor (i.e., FcgR-IIIa) genetic
human use. Currently, we are in transitions from polymorphism studies point to the Fc-receptor
mouse to chimera, humanized and eventually gene isoform called FCGR3A V/V and F/F that
human antibodies. Some believe that compared produce functional and ablated genotypes.7 NHL s
to mouse antibody, switching to chimeric and patients on Rituxin with FCGR3A at 158 amino-
humanized antibody provides about 25% higher acid position being V/V, compared to V/F or F/F
clinical success rate. There are several technolo- exhibit significantly higher [90 vs. 51 ( p ź 0.03)] 1-
gies including phage-array human antibody- year response rate. The discovery of the link
binding library, Xenomouse, and rational protein between the function of Fc-receptor and clinical
design using human IgG sequences that greatly response in NHL s patients prompted the race to
improve efficiency in producing a well-defined strategically modify the Fc sequences or recogni-
human IgG that recognizes a given   drug target  tion motif that related to Fc-receptor (FcgR-IIIa)
or therapeutic target. Some of these high- binding. Alteration of the composition of glycosyl
throughput technologies have pushed the envelop residues of Fc regions by producing the recombi-
on binding affinity of a typical therapeutic anti- nant antibody in the absence of fucose in the
body by about a thousand fold (from nM to pM antibody s glycosylation site was shown to over-
range).6 Some of the   binding optimized or affinity come disparity in Fc binding affinity to Fc-
matured  antibodies are now in clinical trials. receptor.8 These data led to the identification
Numax (motavizumab), for example, is a second and modification of amino acid sequences on IgG s
generation   affinity matured  antibody to respira- Fc domain that regulates decoupling of binding to
tory syncytia virus (RSV). Information submitted Fc-mediated function.9 Such antibodies will likely
in the biologic license application (BLA) to FDA in improve safety in diseases such as Alzheimer s
early 2008 suggests that the second generation where antibody mediated inflammatory responses
antibody motavizumab produced significantly are a major concern.10 In clinical trials, vaccinat-
higher antiviral potency against RSV than the ing Alzheimer s patients with an immunogen
existing product Synagis (palivizumab). Thus, we   AN-1792,  designed to elicit endogenous anti-
are likely to see more affinity optimized antibodies body against the beta-amyloid (Ab) plaques
entering clinical testing. had led to removal of plaques and improved
Another antibody innovation that fuels clinical quality of life; however, 6% of subjects developed
testing is IgG antibody that lacks Fc-mediated aseptic meningitis.11 Thus   AN-1792  trial was
biologic function but exhibits similar binding terminated.12 Also, Élan and Wyeth codeveloped
capabilities and circulating plasma half-life. The bapineuzumab an antibody to the Ab plaques
DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 6, JUNE 2009
1934 HO AND CHIEN
for Alzheimer s disease neuropathology, which 5. Ho RJ, Gibaldi M. 2003. Antibodies and Deriva-
tives. In: Ho RJ, Gibaldi M, editors. Biotechnology
recently completed phase II (n ź 234) study. In a
and Biopharmaceuticals: Transforming Proteins
July 29, 2008, press release, the bapineuzumab, a
and Genes into Drugs. Hoboken, NJ: John Wiley
first generation humanized monoclonal antibody,
and Sons. pp. 271 311.
did not induce aseptic meningitis, but treated
6. Carter PJ. 2006. Potent antibody therapeutics by
patients are twice as likely to exhibit vasogenic
design. Nat Rev 6:343 357.
edema. It is not known whether edema is related
7. Cartron G, Dacheux L, Salles G, Solal-Celigny P,
to antibody-mediated cytotoxic or inflammatory
Bardos P, Colombat P, Watier H. 2002. Therapeutic
functions. Whether second generation huma-
activity of humanized anti-CD20 monoclonal anti-
nized-antibodies to Ab plaques, such as   PF-
body and polymorphism in IgG Fc receptor Fcgam-
04360365  which lack Fc mediated functions will
maRIIIa gene. Blood 99:754 758.
improve safety without compromising efficacy is 8. Okazaki A, Shoji-Hosaka E, Nakamura K,
Wakitani M, Uchida K, Kakita S, Tsumoto K,
remained to be determined.
Kumagai I, Shitara K. 2004. Fucose depletion from
human IgG1 oligosaccharide enhances binding
enthalpy and association rate between IgG1 and
REFERENCES
Fc[gamma]RIIIa. J Mol Biol 336:1239 1249.
9. Wilcock DM, Alamed J, Gottschall PE, Grimm
1. Csellar M, Perrow K. 2007. New Survey Finds J, Rosenthal A, Pons J, Ronan V, Symmonds K,
Strong Public Interest in Prescription Drug Safety Gordon MN, Morgan D. 2006. Deglycosylated anti-
Issues: Americans Trust FDA, but express conflict- amyloid-beta antibodies eliminate cognitive deficits
ing views about how to manage safety risks. ed.: and reduce parenchymal amyloid with minimal
American University s Center for Congressional vascular consequences in aged amyloid precursor
and Presidential Studies. [cited 2008 October 20] protein transgenic mice. J Neurosci 26:5340
Available from: http://www.prnews-wire.com/mnr/ 5346.
dialogue/29804/. 10. Jack CR Jr, Marjanska M, Wengenack TM, Reyes
2. Woodcock J, Woosley R. 2008. The FDA critical path DA, Curran GL, Lin J, Preboske GM, Poduslo JF,
initiative and its influence on new drug develop- Garwood M. 2007. Magnetic resonance imaging of
ment. Annu Rev Med 59:1 12. Alzheimer s pathology in the brains of living trans-
3. Chien JY, Ho RJ. 2008. Drug delivery trends in genic mice: A new tool in Alzheimer s disease
clinical trials and translational medicine. J Pharm research. Neuroscientist 13:38 48.
Sci 97:2543 2547. 11. Robinson SR, Bishop GM, Lee HG, Munch G. 2004.
4. Ho RJ, Gibaldi M. 2003. Not all protein drugs and Lessons from the AN 1792 Alzheimer vaccine: Lest
vaccines of the same name are identical. In: Ho RJ, we forget. Neurobiol Aging 25:609 615.
Gibaldi M, editors. Biotechnology and Biopharma- 12. Brody DL, Holtzman DM. 2008. Active and passive
ceuticals: Transforming Proteins and Genes into immunotherapy for neurodegenerative disorders.
Drugs. Hoboken, NJ: John Wiley and Sons. pp. 2 9. Annu Rev Neurosci 31:175 193.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 6, JUNE 2009 DOI 10.1002/jps