3164 J. Med. Chem.

2005, 48, 3164-3170

A Bioavailability Score

Yvonne C. Martin*
Advanced Technology Division, Global Pharmaceutical Research and Development, Abbott Laboratories,
Abbott Park, IL 60064-6100

Received October 4, 2004

Responding to a demonstrated need for scientists to forecast the permeability and bioavailability
(F) properties of compounds before their purchase, synthesis, or advanced testing, we have
developed a score that assigns the probability that a compound will have F > 10% in the rat.
Neither the rule-of-five, log P, log D, nor the combination of the number of rotatable bonds
and polar surface area successfully categorized compounds. Instead, different properties govern
the bioavailability of compounds depending on their predominant charge at biological pH. The
fraction of anions with >10% F falls from 85% if the polar surface area (PSA) is e 75 Å2, to
56% if 75 < PSA < 150 Å2, to 11% if PSA is g 150 Å2. On the other hand, whereas 55% of the
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

neutral, zwitterionic, or cationic compounds that pass the rule-of-five have >10% F, only 17%
of those that fail have > 10% F. This same categorization distinguishes compounds that are
poorly permeable from those that are permeable in Caco-2 cells. Further validation is provided
with human bioavailability values from the literature.
Downloaded via BOSTON COLG on July 12, 2018 at 17:13:54 (UTC).

Introduction minimum.20-25 In some cases, utilization of a specific

Recent attention has focused on the need to assess
the potential for bioavailability problems of potential
drug candidates early in the drug discovery cycle.1 This
need is particularly acute for those drug discovery
programs that follow a genomic or crystallographic
target for which there is no cell- or animal-based assay
for potency. One approach to solve the problem has been
to use high throughput/low quantity methods such as
PAMPA chromatography2 or Caco-2 permeability mea-
surements3 to provide a prediction of whole-cell and
animal activity. However, the ideal method would be
one that would require no sample, a virtual ADME
prediction technique, and so could be used before a
compound is synthesized or purchased.
The history of the search for such virtual methods
dates to the earliest studies of the relationships between
the physical chemical and biological properties of mol-
ecules.4,5 The search continues to this day,6 with claims
of new, improved models appearing frequently.7-12
Others have suggested rules-of-thumb that improve the
chances of a compound being well absorbed.13-15
There are examples that consideration of physical
properties has been useful in the optimization of the
bioavailability or permeability within specific series of
compounds.16,17 However, the generalization of such
studies to diverse molecules has not been convincingly
demonstrated. Indeed, there is no consensus as to the
physical properties relevant for bioavailability. For
example, although an increase in lipophilicity is gener-
ally considered to be related to an increase in bioavail-
ability,18 in at least one study a decrease in cell
penetration accompanies an increase in hydrophobic-
ity.19 Others have emphasized the importance of keep-
ing hydrogen-bonding or polar surface area to a
* To whom correspondence should be addressed. R47E AP10/2, 100
Abbott Park Rd, Abbott Park IL 60064-6100. e-mail yvonne.c.martin@
abbott.com.
transporter, which has structure-activity relationships
of its own, is necessary.26 In addition to published
models, several vendors offer computer programs for
predicting permeability or bioavailability; for example,
Cerius2,10,27 Idea,28 Absolv,29-31 QikProp,32 QMPRPlus,33
VolSurf,34,35 and OraSpotter.36
This report describes our investigations that led to a
method that provides the probability that a compound
will have sufficient rat bioavailability, 10%, such that
it will not be dropped for further development because
of bioavailability concerns. We envision that the score
will be used as a complement to virtual screening, for
triaging HTS, and for selecting compounds for purchase.
We will show that it outperforms the conventional rule-
of-five13 or other simple publicly available algo-
rithms.10,15,37-40
Methods
Table 1 provides a summary of the properties of the datasets
used. The rat bioavailability data, measured by our standard
procedure,41 had been collected over a period of years. It
contains compounds that targeted a wide variety of therapeutic
targets. This set of 553 compounds contains 99 different rings
and 180 different side chains. When clustered at 0.75 similar-
ity, there are an average of 2.27 compounds per cluster. In
contrast, the Caco-2 permeability screening data was collected
during only one year with the result that this dataset is less
structurally diversesit contains an average of 4.41 compounds
per cluster. We omitted from consideration only macrolides
and a series of compounds that had been synthesized to target
the liver. In this dataset 47.74% of the compounds have rat
bioavailability less than 10%, 35.80% have rat bioavailability
greater than or equal to 10 but less than 50%, 12.66% have
bioavailability greater than or equal to 50% but less than 80%,
and 3.80% have bioavailability greater than or equal to 80%.
We used human bioavailability data curated by others to
validate our initial studies. Continuous human bioavailability
data was taken from three review articles written by experts
in interpreting such data.42-44 A compound was used in the
analysis only if the reported bioavailability is a narrow range
in all articles in which it was reported; this criterion eliminated

10.1021/jm0492002 CCC: $30.25 © 2005 American Chemical Society

Published on Web 04/05/2005
A Bioavailability Score Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9 3165

Table 1. Properties of the Datasets Analyzed

human bioavailability, F
human intestinal continuous category Caco-2 permeability
dataset absorption (%) (0-4) variable rat, %F category
no. of molecules 96 449 519 118 553 617
no. of monocations 36 189 233 43 171 326
no. of uncharged 28 137 151 38 150 134
no. of monoanions 20 76 83 13 134 114
no. of zwitterions 9 23 24 11 38 22
endpoint mean 77.8 54.4 2.61 - 22.31 1.11
endpoint range 1.9-100 0-100 0-4 - 0-168 0-4
pass rule-of-five 96% 94% 94% 86% 79% 72%
log P mean 1.63 2.00 2.09 2.38 3.91 4.69
log P range -3.65 to 14.36 -6.66 to 14.36 -6.66 to 14.36 -9.8 to 14.36 -2.2 to 9.5 -1.5 to 9.5
MW mean 323.9 335.5 333.3 376.8 433.6 454.4
MW range 138.1-1203 46.1-1203 46.1-1203 123-1312 160-7935 159-780
Hb donors mean 2.05 1.83 1.83 2.08 1.46 2.05
Hb donors range 0-12 0-12 0-12 0-20 0-7 0-6
Hb acceptors mean 5.03 5.56 5.45 6.08 6.19 6.23
Hb acceptors range 1-23 1-23 1-23 1-38 1-15 2-15
polar surface area mean 82.7 83.0 81.7 83.6 86.4 88.7
polar surface area range 0-204 2.91-245 2.91-245 5.34-685 11.5-261 15.8-234
clusters @0.75 similarity 83 375 244 140
compounds per cluster 1.16 1.38 2.27 4.41

118 compounds, leaving 449 with consistent values. To this Results

set we added additional compounds from a previous model9
and used their criteria to assign the human bioavailability Table 2 summarizes the results, which will be dis-
category for compounds from the review articles but not in cussed with the figures that clarify various points.
their model. Because most of the compounds reported in the We first examined properties that have traditionally
review articles have good human bioavailability, we sought been considered to be associated with permeability or
to add more compounds with low bioavailability. Recognizing bioavailability.18 Figure 1 shows that log D seems to be
that bioavailability cannot exceed intestinal absorption (HIA), somewhat associated with permeability but not with rat
we added additional compounds for which HIA is reported8 to bioavailability. (Not shown is that log D at pH values
be 0-5% and assigned them to the lowest bioavailability 4.6, 6.4, and 7.4 perform similarly. Log D at pH 7.4 is
category. Note that the majority of the compounds from the
literature for human bioavailability pass the rule-of-five.13
used for permeability because that is the pH at which
We calculated the physical properties that have traditionally
been considered to be related to permeability or bioavailabil-
ity: molecular weight, octanol-water log P,45 log D,46 polar
surface area,47 and number of rotatable bonds. For the latter
we used a daylight toolkit program to (1) count all single bonds
not in a ring, (2) subtract the number of amide bonds not in a
ring, and (3) subtract the number of symmetrical terminal
groups such as methyl, nitro, nitrile, unusbstituted- or me-
thylalkyne, trifluoromethyl, trichloromethyl, and tribromom-
ethyl. The number of hydrogen bond donors and acceptors was
calculated with a simple Daylight toolkit program.48 Except
in the case of permanently charged compounds, the properties
of the compounds are calculated for the neutral form. Com-
pounds were clustered using the Taylor algorithm.49 We
generated the predominant charge state of the molecule by a
series of empirical SMARTS targets, derived with the aid of
the Biobyte MedChem database.50 The main classes of charged
molecules are basic aliphatic nitrogens, acidic phenols, car-
boxylic acids, sulfonic acids, enolates (all tautomers), basic
amidines, phosphates, basic imidazoles and tetrazoles, N-acyl
aliphatic sulfonamides, S-aryl sulfonamides, and barbiturates.
Independent of this process we calculated the fraction in the
various ionic states with ADME Boxes.51 Where the two
methods disagreed, we again checked literature sources to
decide between them. The standard errors were calculated
as:52
Figure 1. The dependence of (a) Caco-2 permeability on log
standard error ) D at pH 7.4, the pH of measurement (red, impermeable; yellow,
slightly permeable; dark blue, somewhat permeable; light blue,

x
(observed proportion) × [1 - (observed proportion)] permeable; green, very permeable); and (b) rat bioavailability
.
(sample size) on log D at pH 6.4.
3166 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9 Martin

Table 2. Statistical Analysis of the Predictivity of Various Properties for Caco-2 Permeability and Rat Bioavailability
Caco-2 cell permeability rat bioavailability
number number proportion 95% number number proportion 95%
in non- non- std confidence in rat, rat, std confidence
criterion class permeable permeable error limits class F <10% F < 10% error limits
log D < 2.5 290a 174 0.600 0.029 0.058 251b 114 0.454 0.031 0.063
log D 2.5 to 5.0 216a 87 0.403 0.033 0.067 194b 101 0.521 0.036 0.072
log D > 5 75a 57 0.760 0.049 0.099 53b 28 0.528 0.069 0.137
MW < 500 411 188 0.457 0.025 0.049 390 159 0.408 0.025 0.050
MW > 500 206 165 0.801 0.028 0.056 162 105 0.648 0.038 0.075
log D < 2.5 and MW < 500 187 87 0.465 0.036 0.073 181 57 0.315 0.035 0.069
log D < 2.5 and MW > 500 103 87 0.845 0.036 0.072 69 57 0.826 0.046 0.090
log D 2.5 to 5.0 and MW < 500 168 58 0.345 0.037 0.074 124 63 0.508 0.045 0.089
log D 2.5 to 5.0 and MW > 500 48 29 0.604 0.071 0.141 70 38 0.543 0.060 0.119
log D > 5 and MW < 500 47 34 0.723 0.065 0.131 32 19 0.594 0.087 0.174
log D > 5 and MW > 500 28 23 0.821 0.072 0.145 21 9 0.429 0.108 0.216
PSA < 75 Å2 219 91 0.416 0.033 0.067 240 94 0.392 0.032 0.063
PSA 75-150 Å2 358 235 0.656 0.025 0.050 263 126 0.479 0.031 0.062
PSA > 150 Å2 40 27 0.675 0.074 0.148 50 45 0.900 0.042 0.085
pass rule-of-five13 445 215 0.483 0.024 0.047 436 191 0.438 0.024 0.048
fail rule-of-five 172 138 0.802 0.030 0.061 116 73 0.629 0.045 0.090
CLOGP e 5.88 and PSA e 131.6 Å2 10 376 179 0.476 0.026 0.052 405 181 0.447 0.025 0.049
CLOGP g 5.88 and/or PSA g 131.6 Å2 241 174 0.722 0.029 0.058 147 83 0.565 0.041 0.082
PSA < 140 Å2 and 554 312 0.563 0.021 0.042 493 220 0.446 0.022 0.045
number of rotatable bonds < 10 15
PSA g140 Å2 and/or 63 41 0.651 0.060 0.120 69 44 0.733 0.057 0.114
number of rotatable bonds g 10
ABS ) 0.85 6 1 0.167 0.152 0.304 34 5 0.147 0.061 0.121
ABS) 0.55 or 0.56 455 223 0.490 0.023 0.047 438 191 0.436 0.024 0.047
ABS < 0.20 156 129 0.827 0.030 0.061 80 68 0.850 0.040 0.080
positive charge, pass rule-of-five 368 169 0.459 0.026 0.052 344 150 0.436 0.027 0.053
positive charge, fail rule-of five 114 95 0.833 0.035 0.070 26 23 0.885 0.063 0.125
negative charge, pass rule-of-five 77 46 0.597 0.056 0.112 93 41 0.441 0.051 0.103
negative charge, fail rule-of-five 58 43 0.741 0.057 0.115 90 50 0.556 0.052 0.105
positive charge, PSA < 75 213 90 0.423 0.034 0.068 206 89 0.432 0.035 0.069
positive charge PSA 75-150 261 163 0.625 0.030 0.060 160 80 0.500 0.04 0.079
positive charge, PSA >150 8 1 0.125 0.117 0.234 4 4 1.00
negative charge, PSA < 75 6 1 0.167 0.152 0.304 34 5 0.147 0.061 0.121
negative charge, PSA 75-150 97 62 0.639 0.049 0.098 103 45 0.437 0.049 0.098
negative charge, PSA > 150 32 26 0.813 0.069 0.138 46 41 0.891 0.046 0.092
a pH 7.4. b pH 6.4.
the measurements were made; pH 6.4 is used for rat
bioavailability to account for the slightly acidic nature
of the regions of the GI tract at which most absorption
occurs.) For example, 57 of the 75 compounds, 76%, with
log D > 5 are not permeable, but only 53% of the
compounds with this log D have low rat bioavailability.
Figure 2 shows that compounds with a molecular weight
greater than 500 are in general less permeable and tend
to have lower bioavailability. However, although 80%
of the compounds with molecular weight greater than
500 are not permeable, only 65% of such compounds
have low rat bioavailability. Figure 3 and Table 2 show
that the effect of molecular weight is most pronounced
at low log D values in that the 84% of the 103
compounds with log D < 2.5 and MW > 500 are not
permeable and 83% have low bioavailability. Presum-
ably the low permeability and bioavailability of these
compounds is associated with a large number of polar
atoms.
Recently the rule-of-five has become a popular way
to characterize compounds.13 It was derived by an
analysis of the properties of orally active marketed
drugs. Figure 4 shows that 80% of the compounds that
fail the rule-of-five are not permeable, whereas 48% of
those that pass are not permeable. The contrast is less
dramatic for rat bioavailability for which the corre- Figure 2. The dependence of (a) Caco-2 permeability (legend
sponding numbers are 63% and 44%. Thus 43 of the 116 as in Figure 1) and (b) rat bioavailability on molecular weight.
A Bioavailability Score Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9 3167

Figure 4. The predictivity of the rule-of-five for (a) Caco-2

permeability and (b) rat bioavailability. Legend as in Figures
1 and 3.

Figure 3. (a) Caco-2 permeability as a function of log D (pH

7.4) and MW. Note that a log D value could not be calculated
for 36 compounds. (b) Rat bioavailability as a function of log
D (pH 6.4) and MW (red, <10%; yellow, 10-49%; dark blue,
50-79%; green, g80%). Note that a log D value could not be
calculated for 55 compounds.
compounds that fail the rule-of-five have acceptable
bioavailability and 191 of the 436 that pass do not.
Figure 5a shows that, in accord with literature
Figure 5. (a) Caco-2 permeability (legend as in Figure 1) and
suggestions,53 in the set of compounds that have a polar
(b) rat bioavailability as a function of polar surface area in
surface area less than 75 Å2 there is a modest increase, Å2.
from 46% to 62%, in the percentage of compounds that
are permeable. However, as shown in Figure 5b, only
at very high polar surface area are some of the poorly a combination of PSA and number of rotatable bonds15
bioavailable compounds separated from those that are does separate out a set of 69 molecules with a higher
more bioavailable. Thus 90% of the 50 compounds with (73%) probability of being poorly bioavailable.
polar surface area greater than 150 Å2 are not bioavail- When one considers the predictivity of various physi-
able in the rat. At the values represented by the cal properties as a function of the net charge on the
majority of the compounds, polar surface area does not molecule, a clearer picture emerges. Figure 8 shows that
predict rat bioavailability. This suggests that the poor the rule-of-five does identify poorly bioavailable com-
bioavailability of compounds with low polar surface area pounds that are neutral (including zwitterions) or
values is due to other factors such as first pass metabo- positively charged, but not anions. Considering only the
lism or poor solubility. neutral and positively charged compounds increases the
Figure 6 shows that the recently proposed model54 proportion of poorly bioavailable compounds predicted
based on CLOGP and PSA also does not at all predict from 63% for the whole dataset to 88%. The same trend
well rat bioavailability. In contrast, Figure 7 shows that for Caco-2 permeability is seen in Figure 9.
3168 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9
Figure 6. Predictivity of rat bioavailability by the proposed
combination of polar surface area and CLOGP. Legend as in
Figure 3.
Figure 7. Predictivity of rat bioavailability by the proposed
combination of polar surface area and the number of rotatable
bonds. Legend as in Figure 3.
Figure 8. The predictivity of rat bioavailability from the rule-
of-five as a function of the charge on the predominant form of
the molecule at pH 6-7. Legend as in Figure 3.
In marked contrast, Figure 10 shows that polar
surface area predicts bioavailability of anions, but not
neutral or cationic compounds. Increasing polar surface
area is associated with increasing percentages of com-
pound that are not permeable or not bioavailable.
Based on these observations, A Bioavailability Score,
ABS, is calculated according to the rules in Figure 11.
Its performance is summarized in Figure 12. The
numerical score represents the probability that a com-
pound with the indicated properties will have rat
bioavailability greater than 10%. Contrast Figure 12
with Figure 7, and note that, at the high end of the
scale, ABS identifies a set of 34 compounds with high
average bioavailability, and at the low end of the scale,
poorly absorbed compound are also more clearly identi-
fied.
Figure 13 shows that ABS is also related to human
bioavailability. However, the fraction of compounds that
have low bioavailability is proportionally lower in all
Martin
Figure 9. The predictivity of Caco-2 permeability from the
rule-of-five as a function of the charge on the predominant form
of the molecule at pH 6-7. Not shown are an additional eight
compounds with a net charge -2 that violate the rule-of-five
and have %F < 10%. Legend as in Figure 1.
Figure 10. The influence of net charge on the predictivity of
polar surface area for rat bioavailability. Legend as in Figure
3.
categories, presumably because some poorly bioavailable
compounds do not make it to market.
Discussion
The results presented here were a surprise to us. They
were generated after commercial ADME-prediction
programs could not predict these data. To investigate
further their lack of predictivity, we investigated whether
the programs fail for certain classes of compounds, ionic
state being one of the classifications considered. A
chance observation led to the relationships shown in
Figure 8.
The model by Yoshida and Topliss provides one
precedent for the observation that different physical
properties govern the bioavailability of anions rather
than neutral compounds and cations.9 Their model
contains a term, log D6.5 - log D7.4 that differentiates
anionic compounds (positive ∆ log D) from basic com-
pounds (slightly negative ∆ log D) and neutral com-
pounds (zero ∆ log D). Hence, their equation suggests
different behavior for anions than for neutral and
cationic compounds.
We know of no particular explanation for these
observations. Indeed, there is a poor correlation between
Caco-2 permeability and human bioavailability category,
Figure 14. Clearly, however, one can believe that the
charge state of the molecule has a key effect on how it
is perceived by biomolecules including membranes,
enzymes, and transporters.
A Bioavailability Score
Figure 11. A pictorial representation of ABS. Legends as in Figures 1 and 3.
Figure 12. A summary of the performance of ABS for rat F.
Legend as in Figure 3.
Figure 13. The predictivity of ABS for the human bioavail-
ability of 519 compounds. Legend: red, <20% F; yellow, 20-
49% F; dark blue, 50-79% F; green, g80% F.
Figure 14. The relationship between Caco-2 permeability and
human bioavailability. Legend as in Figure 13.
Two components of the rule-of-five are the sum of the
number of nitrogen and oxygen atoms and the number
of hydrogen-bond donating atoms in the molecule.
Although these properties are related to polar surface
area, our investigation suggests that polar surface area
alone, not combined with molecular weight or CLOGP,
is related to bioavailability of anions.
Although in a drug discovery program the ultimate
aim is to produce a compound that is bioavailable in
humans, rat bioavailability is frequently used as a
surrogate for this property, and compounds that have
poor rat bioavailability may have difficulty gaining
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9 3169
management approval for development. We anticipate
that ABS will be used in very early drug discovery, for
example in selecting compounds to purchase or deciding
which of several series of hits will be most attractive to
pursue with synthesis. ABS is calculated for every
compound registered into the Abbott structure database
and reported back to the person who submitted it. These
values are also presented on the HTS screening report
and are accessible through ISIS and Spotfire. Addition-
ally, ABS calculations on compounds proposed for
synthesis or purchase can be performed on an internal
web page.
Conclusions
The physical properties that govern the bioavailability
and permeability of compounds that are negatively
charged at pH 6-7 differ from those that govern these
properties of compounds that are uncharged or posi-
tively charged at this pH. For anions the important
property is polar surface area (PSA) whereas for the
other compounds the rule-of-five has predictive ability.
A Bioavailability Score, ABS, is formulated as the
probability that a compound will have >10% bioavail-
ability in rat or measurable Caco-2 permeability. ABS
is 0.11 for anions for which PSA is >150 Å2, 0.56 if PSA
is between 75 and 150 Å2, and 0.85 if PSA is <75 Å2.
For the remaining compounds ABS is 0.55 if it passes
the rule-of-five and 0.17 if it fails. ABS also identifies
poorly- and well-absorbed compounds tested in humans.
Acknowledgment. The author thanks Elizabeth
Everitt in the drug metabolism/PK department and
Mark Schurdak in the high throughput screening
department for the Caco-2 data. Kennan Marsh in the
drug metabolism/PK department and Jonathan Trum-
bull in the automation engineering department provided
the animal pharmacokinetics data. Additionally, thanks
to Mark Bures and Isabella Lico Haight, who prepared
the original database of human ADME properties, and
John Topliss who supplied the data for the compounds
from his report.
References
(1) Van De Waterbeemd, H.; Gifford, E. Admet in Silico Modelling:
Towards Prediction Paradise? Nature Rev. Drug Discovery 2003,
2, 192-204.
(2) Kansy, M.; Senner, F.; Gubernator, K. Physicochemical High
Throughput Screening - Parallel Artificial Membrane Perme-
ation Assay in the Description of Passive Absorption Processes.
J. Med. Chem. 1998, 41, 1007-1010.
3170 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9
(3) Artursson, P.; Karlsson, J. Correlation between Oral Drug
Absorption in Humans and Apparent Drug Permeability Coef-
ficients in Human Intestinal Epithelial (Caco-2) Cells. Biochem.
Biophys. Res. Commun. 1991, 175, 880-885.
(4) Overton, E. Osmotic Properties of Cells in the Bearing on
Toxicology and Pharmacy. Z. Phys. Chem. 1897, 22, 189-209.
(5) Meyer, H. Zur Theorie Der Alkolnarkose I. Welche Eigenschaft
Der Anaesthetica Bedingt Ihre Narkotische Wirkung? Arch. Exp.
Pathol. Pharmakol. 1899, 42, 109-118.
(6) Clark, D. E. Prediction of Intestinal Absorption and Blood-Brain
Barrier Penetration by Computational Methods. Comb. Chem.
High Throughput Screening 2001, 4, 477-496.
(7) Hirono, S.; Nakagome, I.; Hirano, H.; Matsushita, Y.; Yoshii, F.
et al. Non-Congeneric Structure-Pharmacokinetic Property Cor-
relation Studies Using Fuzzy Adaptive Least-Squares: Oral
Bioavailability. Biol. Pharm. Bull. 1994, 17, 306-309.
(8) Wessel, M. D.; Jurs, P. C.; Tolan, J. W.; Muskal, S. M. Prediction
of Human Intestinal Absorption of Drug Compounds from
Molecular Structure. J. Chem. Inform. Comput. Sci. 1998, 38,
726-735.
(9) Yoshida, F.; Topliss, J. G. QSAR Model for Drug Human Oral
Bioavailability. J. Med. Chem. 2000, 43, 2575-2585.
(10) Egan, W. J.; Merz, K. M.; Baldwin, J. J. Prediction of Drug
Absorption Using Multivariate Statistics. J. Med. Chem. 2000,
43, 3867-3877.
(11) Wolohan, P. R. N.; Clark, R. D. Predicting Drug Pharmacokinetic
Properties Using Molecular Interaction Fields and Simca. J.
Comput.-Aided Mol. Des. 2003, 17, 65-76.
(12) Oprea, T. I.; Zamora, I.; Ungell, A. L. Pharmacokinetically Based
Mapping Device for Chemical Space Navigation. J. Comb. Chem.
2002, 4, 258-266.
(13) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J.
Experimental and Computational Approaches to Estimate Solu-
bility and Permeability in Drug Discovery and Development
Settings. Adv. Drug Delivery Rev. 1997, 23, 3-25.
(14) Navia, M. A.; Chaturvedi, P. R. Design Principles for Orally
Bioavailable Drugs. Drug Discovery Today 1996, 1, 179-189.
(15) Veber, D. F.; Johnson, S. R.; Cheng, H. Y.; Smith, B. R.; Ward,
K. W. et al. Molecular Properties That Influence the Oral
Bioavailability of Drug Candidates. J. Med. Chem.. 2002, 45,
2615-2623.
(16) Hill, A. P.; Hyde, R. M.; Robertson, A. D.; Wollard, P. M.; Glen,
R. et al. Oral Delivery of 5-HT1D Receptor Agonists: Towards
the Discovery of 311c90, a Novel Anti-Migraine Agent. Headache
1994, 34, 308.
(17) Rosenberg, S. H.; Spina, K. P.; Condon, S. L.; Polakowski, J.;
Yao, Z. et al. Studies Directed toward the Design of Orally Active
Renin Inhibitors. 2. Development of the Efficacious, Bioavailable
Renin Inhibitor (2s)-2- Benzyl-3- [[(1-Methylpiperazin-4-Yl)-
Sulfonyl]Propionyl]-3-Thiazol-4-Yl- L-Alanine Amide of (2s,3r,-
4s)-2-Amino-1-Cyclohexyl-3,4-Dihydroxy-6- Methylheptane (a-
72517). J. Med. Chem. 1993, 36, 460-467.
(18) Camenisch, G.; Folkers, G.; Vandewaterbeemd, H. Shapes of
Membrane Permeability-Lipophilicity Curves - Extension of
Theoretical Models with an Aqueous Pore Pathway. Eur. J.
Pharm. Sci. 1998, 6, 321-329.
(19) Sawada, G. A.; Barsuhn, C. L.; Lutzke, B. S.; Houghton, M. E.;
Padbury, G. E. et al. Increased Lipophilicity and Subsequent
Cell Partitioning Decrease Passive Transcellular Diffusion of
Novel, Highly Lipophilic Antioxidants. J. Pharmacol. Exp. Ther.
1999, 288, 1317-1326.
(20) Paterson, D. A.; Conradi, R. A.; Hilgers, A. R.; Vidmar, T. J.;
Burton, P. S. A Non-Aqueous Partitioning System for Predicting
the Oral Absorption Potential of Peptides. Quant. Structure-
Act. Relat. 1994, 13, 4-10.
(21) Burton, P. S.; Conradi, R. A.; Ho, N. F. H.; Hilgers, A. R.;
Borchardt, R. T. How Structural Features Influence the Biomem-
brane Permeability of Peptides. J. Pharm. Sci. 1996, 85, 1336-
1340.
(22) Palm, K.; Stenberg, P.; Luthman, K.; Artursson, P. Polar
Molecular Surface Properties Predict the Intestinal Absorption
of Drugs in Humans. Pharm. Res. 1997, 14, 568-571.
(23) Norinder, U.; Osterberg, T.; Artursson, P. Theoretical Calculation
and Prediction of Caco-2 Cell Permeability Using Molsurf
Parametrization and Pls Statistics. Pharm. Res. 1997, 14, 1786-
1791.
(24) Stenberg, P.; Luthman, K.; Artursson, P. Prediction of Membrane
Permeability to Peptides from Calculated Dynamic Molecular
Surface Properties. Pharm. Res. 1999, 16, 205-212.
(25) Norinder, U.; Osterberg, T.; Artursson, P. Theoretical Calculation
and Prediction of Intestinal Absorption of Drugs in Humans
Using Molsurf Parametrization and Pls Statistics. Eur. J.
Pharm. Sci. 1999, 8, 49-56.
Martin
(26) Tamai, I.; Nakanishi, T.; Nakahara, H.; Sai, Y.; Ganapathy, V.
et al. Improvement of L-Dopa Absorption by Dipeptidyl Deriva-
tion, Utilizing Peptide Transporter Pept1. J. Pharm. Sci. 1998,
87, 1542-1546.
(27) Cerius 2; Accelrys: San Diego, CA.
(28) iDEA pkExpress; Lion Bioscience AG: Heidelberg, Germany.
(29) Absolv; Sirius Analytical Instruments Ltd: East Sussex UK.
(30) Zhao, Y. H.; Le, J.; Abraham, M. H.; Hersey, A.; Eddershaw, P.
J. et al. Evaluation of Human Intestinal Absorption Data and
Subsequent Derivation of a Quantitative Structure-Activity
Relationship (QSAR) with Abraham Descriptors (Vol 90, Pg 749,
2001). J. Pharm. Sci. 2002, 91, 605.
(31) Zhao, Y. H.; Le, J.; Abraham, M. H.; Hersey, A.; Eddershaw, P.
J. et al. Evaluation of Human Intestinal Absorption Data and
Subsequent Derivation of a Quantitative Structure-Activity
Relationship (QSAR) with the Abraham Descriptors. J. Pharm.
Sci. 2001, 90, 749-784.
(32) QikProp; Schrödinger, Inc.
(33) QMPRPLUS; Simulations plus, inc.: Lancaster, CA 93534-2902.
(34) Volsurf; Tripos, Inc.: St. Louis, MO.
(35) Cruciani, G.; Pastor, M.; Guba, W. Volsurf: A New Tool for the
Pharmacokinetic Optimization of Lead Compounds. Eur. J.
Pharm. Sci. 2000, 11, S29-S39.
(36) Oraspotter; zyxbio: Cleveland, OH.
(37) Clark, D. E. Rapid Calculation of Polar Molecular Surface Area
and Its Application to the Prediction of Transport Phenomena.
1. Prediction of Intestinal Absorption. J. Pharm. Sci. 1999, 88,
807-814.
(38) Clark, D. E. Rapid Calculation of Polar Molecular Surface Area
and Its Application to the Prediction of Transport Phenomena.
2. Prediction of Blood-Brain Barrier Penetration. J. Pharm. Sci.
1999, 88, 815-821.
(39) Norinder, U.; Haeberlein, M. Computational Approaches to the
Prediction of the Blood-Brain Distribution. Adv. Drug Delivery
Rev. 2002, 54, 291-313.
(40) Feher, M.; Sourial, E.; Schmidt, J. M. A Simple Model for the
Prediction of Blood-Brain Partitioning. Intl. J. Pharm. 2000, 201,
239-247.
(41) Kempf, D. J.; Marsh, K. C.; Paul, D. A.; Knigge, M. F.; Norbeck,
D. W. et al. Antiviral and Pharmacokinetic Properties of C2
Symmetric Inhibitors of the Human Immunodeficiency Virus
Type 1 Protease. Antimicrob. Agents Chemother. 1991, 35, 2209-
2214.
(42) Benet, L. Z.; Øie, S.; Schwartz, J. B. Design and Optimization
of Dosage Regimens: Pharmacokinetic Data. Goodman & Gil-
man’s the Pharmacological Basis of Therapeutics, 9th ed.;
McGraw-Hill: New York, 1996; pp 1707-1792.
(43) Bertz, R. J.; Granneman, G. R. Use of in Vitro and in Vivo Data
to Estimate the Likelihood of Metabolic Pharmacokinetic Inter-
actions. Clin. Pharmacokinet. 1997, 32, 210-258.
(44) Sietsema, W. K. The Absolute Oral Bioavailabilty of Selected
Drugs. Intl. J. Clin. Pharmacol., Ther. Toxicol. 1989, 27, 179-
211.
(45) Hoekman, D.; Leo, A. J. CLOGP; 4.3 ed.; Biobyte: Claremont,
CA.
(46) ADME Batch; Pharma Algorithms: Toronto, Canada.
(47) Pearlman, R. S. Savol3; College of Pharmacy, The University
of Texas at Austin, Austin, TX. Pearlman@vax.phr.utexas.edu.:
Austin, TX.
(48) Daylight Chemical Information Systems. Manual to Version 4.61;
4.62 ed.; Daylight Chemical Information Systems: Mission Viejo,
CA 92691.
(49) Martin, Y. C.; Kofron, J. L.; Traphagen, L. M. Do Structurally
Similar Molecules Have Similar Biological Activity? J. Med.
Chem. 2002, 45, 4350-4358.
(50) Leo, A. Medchem 2002 Database; Biobyte: Claremont, CA.
(51) ADME Boxes; 1.0 ed.; Pharma Algorithms: Toronto, Canada.
(52) Jaffe, A. J.; Spirer, H. F. Misused Statistics: Straight Talk for
Twisted Numbers; Marcel Dekker: New York, 1987; pp 146-
149.
(53) Palm, K.; Luthman, K.; Ungell, A. L.; Strandlund, G.; Artursson,
P. Correlation of Drug Absorption with Molecular-Surface
Properties. J. Pharm. Sci. 1996, 85, 32-39.
(54) Egan, W. J.; Lauri, G. Prediction of Intestinal Permeability. Adv.
Drug Delivery Rev. 2002, 54, 273-289.
JM0492002