Journal of Veterinary Pharmacology and Therapeutics

Full Access

Interspecies allometric scaling. Part I: prediction of clearance in large animals

I. MAHMOOD

Clinical Pharmacology and Toxicology Branch (HFD-579), Office of Drug Evaluation VI, Center for Drug Evaluation and Research

The views expressed in this article are those of the authors and do not reflect the official policy of the FDA. No official support or endorsement by the FDA is intended or should be inferred.

Search for more papers by this author

M. MARTINEZ,

M. MARTINEZ

Division of Therapeutic Drugs for Food Animals (HFV-130), Office of New Animal Drug Evaluation, Center for Veterinary Medicine, Food & Drug Administration, Rockville, MD

The views expressed in this article are those of the authors and do not reflect the official policy of the FDA. No official support or endorsement by the FDA is intended or should be inferred.

Search for more papers by this author

R. P. HUNTER,

R. P. HUNTER

Elanco Animal Health, A Division of Eli Lilly and Company, Veterinary Safety/ADME, Greenfield, IN, USA

Search for more papers by this author

I. MAHMOOD,

I. MAHMOOD

Clinical Pharmacology and Toxicology Branch (HFD-579), Office of Drug Evaluation VI, Center for Drug Evaluation and Research

The views expressed in this article are those of the authors and do not reflect the official policy of the FDA. No official support or endorsement by the FDA is intended or should be inferred.

Search for more papers by this author

M. MARTINEZ,

M. MARTINEZ

Division of Therapeutic Drugs for Food Animals (HFV-130), Office of New Animal Drug Evaluation, Center for Veterinary Medicine, Food & Drug Administration, Rockville, MD

The views expressed in this article are those of the authors and do not reflect the official policy of the FDA. No official support or endorsement by the FDA is intended or should be inferred.

Search for more papers by this author

R. P. HUNTER,

R. P. HUNTER

Elanco Animal Health, A Division of Eli Lilly and Company, Veterinary Safety/ADME, Greenfield, IN, USA

Search for more papers by this author

First published: 06 September 2006

https://doi.org/10.1111/j.1365-2885.2006.00786.x

Citations: 53

Iftekhar Mahmood, Clinical Pharmacology and Toxicology Branch (HFD-579), Office of Drug Evaluation VI, Center for Drug Evaluation and Research, Food & Drug Administration, Woodmont Office Center II, Rockville, MD 20852, USA. E-mail: [email protected]

Share a link

Email
Wechat
Bluesky

Abstract

Interspecies scaling is a useful tool for the prediction of pharmacokinetic parameters from animals to humans, and it is often used for estimating a first-time in human dose. The knowledge of pharmacokinetics in veterinary species is important for dosage selection, particularly in the treatment of large zoo animal species, such as elephants, giant cats and camels, for which pharmacokinetic data are scant. Therefore, the accuracy in clearance predictions in large animal species, with and without the use of correction factors (rule of exponents), and the impact of species selection in the prediction of clearance in large animal species was examined. Based upon this analysis, it was determined that there is a much larger risk of inaccuracies in the clearance estimates in large animal species when compared with that observed for humans. Unlike in humans, for large animal species, correction factors could not be applied because there was no trend between the exponents of simple allometry and the appropriate correction factor for improving our predictions. Nevertheless, we did see an indication that the exponents of simple allometry may alert us as to when the predicted clearance in the large animal may be underestimated or overpredicted. For example, if a large animal is included in the scaling, the predicted clearance in a large animal should be considered overestimated if the exponent of simple allometry is >1.3. Despite the potential for extrapolation error, the reality is that allometric scaling is needed across many veterinary practice situations, and therefore will be used. For this reason, it is important to consider mechanisms for reducing the risk of extrapolation errors that can seriously affect target animal safety, therapeutic response, or the accuracy of withdrawal time predictions.

Introduction

Interspecies scaling is a useful tool for the prediction of pharmacokinetic parameters from animals to humans. With the advent of new approaches, it is now possible to predict human pharmacokinetic parameters with greater accuracy than was possible a decade earlier. The first-in-human dose can be selected based on pharmacokinetic data from small laboratory animals (mice, rat, rabbit, guinea pig, monkey, and dog). The principles and methods of extrapolation of pharmacokinetic parameters from laboratory animals to man have been extensively discussed (Boxenbaum & Fertig, 1984; Boxenbaum & Dilea, 1995; Lave et al., 1997; Obach et al., 1997; Mahmood, 2005). Similarly, the knowledge of pharmacokinetics in veterinary species is important for dosage selection, particularly in the treatment of large zoo animal species, such as elephants, giant cats and camels, for which pharmacokinetic data are scant. Without an understanding of the factors that can influence the accuracy of these predictions, such extrapolations can lead to ineffective doses or excessively high doses that can cause serious toxicity.

It is now well-established that clearance can be extrapolated from small laboratory animals to man with a fair degree of accuracy, provided that the rule of exponents (RE) is used (Mahmood, 2005). However, it is not known if the clearance from small animals can be extrapolated to large animals with the same degree of accuracy and whether correction factors shown to improve the accuracy of drug clearance predictions in humans can likewise be used to improve predictions in large veterinary species.

This report addresses this issue and highlights many important factors related to the prediction of clearance from small to large animals. In part II of this report, we explore some of the mathematical attributes of the regression analysis that may contribute to errors in allometric predictions in large animal species.

Methods

The general form of the allometric equation used in scaling pharmacokinetic parameters across animals is as follows:

(1)

where Y is pharmacokinetic parameter of interest; W is the body weight; a is the allometric coefficient; and b is the allometric exponent. The following three methods were used to predict drug clearance in humans as well as in large animals.

Method 1: simple allometry

The clearance of each drug is plotted against the body weight on a log–log scale (where logs are expressed to the base 10) and the following allometric equation is applied:

(2)

where W is the body weight and a and b are the coefficient and exponent of the allometric equation, respectively.

Method 2: product of maximum life span potential and clearance

The observed clearance values in the different animal species are multiplied by their respective maximum life-span potential (MLP), and the products are plotted as a function of body weight on a log–log scale. From the allometric equation, Cl × MLP is estimated in humans and the result is then divided by the MLP of humans (8.18 × 10⁵ h) to predict clearance in man:

(3)

and then predicted human clearance:

(4)

MLP in years is calculated from the following equation as described by Sacher (1959):

(5)

where both brain weight (BW) and body weight (W) are in kilograms.

Method 3: product of brain weight and clearance

In this approach, the clearance of animal species is multiplied by the brain weight of the species and the product is plotted as a function of body weight on a log–log scale. The allometric equation (Eqn 5) is then used to predict the clearance in humans using the human brain weight (1.53 kg):

(6)

and then predicted human clearance:

(7)

where both human brain weight (BW) and body weight (W) are expressed in kilograms.

The aforementioned three methods can only be used under specific conditions. Mahmood and Balian (1996) and Mahmood (2005) formulated a rule so that only one of these methods could be used to predict drug clearance in man from animal data. These authors termed this rule as ‘the rule of exponents’. When scaling from small animal to man, if the exponents of the simple allometry lie between 0.55 and 0.70, simple allometry will predict clearance more accurately than Cl × MLP or Cl × BW. Alternatively, the use of MLP is appropriate for drugs whose exponents lie between 0.71 and 0.99. If the exponent of simple allometry is ≥1.0, then the product of Cl × BW is the most suitable approach for predicting clearance in humans. If the exponents of the simple allometry are >1.3, it is likely that the prediction of clearance from animals to humans will be substantially higher than the observed human clearance, even if using the approach of Cl × BW. If the exponents of simple allometry are below 0.55, the predicted clearance will likely be lower than the observed human clearance.

To predict the clearance from small to large animal species, 12 drugs (Table 1) for which clearance values were available following intravenous administration in both small and large animal species were selected from the literature. For eight drugs, human data were also available. For each set of extrapolations, no less than three species were used to predict clearance in man or in large animals. The following three methods were used to predict the pharmacokinetic parameters in large animals:

Table 1. The exponents of the allometry and correlation coefficient between body weight and clearance

Drugs	Small animals without man		Small animals with man		Species used	Body weight range (kg)	References
Drugs	Exponents	R	Exponents	R	Species used	Body weight range (kg)	References
Caffeine	0.748	0.981	0.698	0.987	m, r, rab, mk	0.03–5.1	Bonati et al. (1984–85), Bachmann (1989), Peck et al. (1997), Wasfi et al. (2000)
Antipyrine	0.589	0.999	0.560	0.998	m, r, rab	0.025–2.75	Elsheikh et al. (1991), Ben-Zvi et al. (1995), Dyke et al. (1998), Balani et al. (2002)
Theophylline	0.657	0.954	0.667	0.974	m, r, rab, d	0.023–10	Ingvast-Larsson et al. (1985), Gascon et al. (1994), Elsheikh et al. (2001)
Fluconazole	0.728	0.999	0.616	0.953	m, r, gp, rab, d	0.02–13.2	Jezequel (1994), Latimer et al. (2001)
Cefoperazone	0.578	0.907	0.573	0.865	m, r, rab, mk, d	0.021–10.7	Sawada et al. (1984), Soraci et al. (1996)
Gentamicin	0.668	0.944	0.722	0.977	r, rab, d	0.23–9.2	Tanira et al. (1997), Martin-Jimenez and Riviere (2001), McNamara et al. (2001)
Propranolol	0.679	0.923	0.701	0.959	r, d, mk	0.25–14.2	Evans et al. (1973), Aramaki et al. (2000)
Furosemide	1.081	0.995	0.844	0.962	r, mk, d	0.2–12	Doyle et al. (1982), Ali et al. (1998)
Thiamphenicol	0.674	0.999	NA	NA	m, r, rab, d	0.029–16.6	Castells et al. (2001)
Sulphadimidine	1.015	0.989	NA	NA	r, d, goat	0.33–30	Riffat et al. (1982), Nouws et al. (1987), Witkamp et al. (1992)
Flunixin	0.497	0.883	NA	NA	rab, cat, d	2.7–10	Hardie et al. (1985), Lees and Taylor (1991), Landoni et al. (1995), Cheng et al. (1998), Wasfi et al. (1998), Coakley et al. (1999)
Enrofloxacin	0.863	0.987	NA	NA	m, r, rab, d	0.03–13.4	Kung et al. (1993), Gavrielli et al. (1995), Bregante et al. (1999); Haines et al. (2000)

m, mouse; r, rat; rab, rabbit; gp, guinea pig; mk, monkey; d, dog.
When man was included in the scaling a body weight of 70 kg was used.
NA, Human data not available.

1
The pharmacokinetic parameters were first predicted from small laboratory animals to humans as well as in large animals. The RE was used to predict clearance in humans but not in large animals (methods 1–3, depending on the exponents of the simple allometry).
2
The observed human data were then included with the small laboratory animal data to predict pharmacokinetic parameters in large animal species and to evaluate if the inclusion of human data in the analysis improved the prediction in large animals.
3
The clearance was predicted by including at least one large animal species in the scaling. In this method of scaling, the fitted dataset included no animal species whose body weight was less than that of the dog. Wherever possible, human data were included in the scaling. If data from two or three large animal species such as horse, camel, cow or donkey were available, the scaling was repeated sequentially, replacing one large animal by another. Prediction was always made in the species that was not included in the scaling.

The error in the prediction of a pharmacokinetic parameter for an individual drug was defined as:

when Error = 1, the prediction was without error.

Results

Prediction of clearance from laboratory animals to large animal species with and without human data

Table 1 summarizes the allometric exponents and the corresponding correlation coefficients (R) of the regression of log clearance vs. log body weight. A good correlation between body weight and clearance was noted across all drugs with or without the inclusion of human data. The exponent of the simple allometry (without human data) for clearance ranged from 0.497 to 1.081 and from 0.560 to 0.844 with human data. For some drugs, inclusion of human data with the small animal data altered the exponent of the allometry almost by ≥0.1. This difference had a substantial impact on the prediction of clearance in large animal species.

The prediction error associated with the allometric scaling of clearance for these compounds is provided in Table 2. The clearances of all eight drugs in humans (for which human data were available) were predicted with great accuracy when the RE was applied. However, despite the very high correlation coefficient, the predicted clearances in large animals were grossly in error. No particular pattern of error was identified and depending upon the drug, predicted values could be either substantially lower or higher than the observed values. The error in the predicted clearances for some drugs was several-fold. For several compounds, the inclusion of human data with small laboratory animal data improved the prediction of clearance. To evaluate the accuracy in smaller animal species, clearance in cat, goat and sheep was also predicted for some compounds. Unlike that seen with the large animal species, but similar to that seen for humans, the predicted clearance values in cat, goat, and sheep were reasonably accurate when derived from equations based solely on laboratory animals and without the RE.

Table 2. Predicted and observed clearance (mL/min) in large animals from laboratory animals

Drugs	Body weight (kg)	Observed	Predicted	Predicted with man	Error	Error with man
Caffeine
Man	70	98	71	NA	0.72	NA
Horse	400	340	566	393	1.66	1.16
Donkey	270	221	422	299	1.91	1.35
Camel	365	82	528	369	6.44	4.50
Antipyrine
Man	70	43	55	NA	1.28	NA
Horse	450	2790	164	133	0.06	0.05
Camel	375	1613	148	120	0.09	0.07
Sheep	25	20	30	26	1.50	1.30
Goat	15	22	23	20	1.05	1.10
Theophylline
Man	70	45	42	NA	0.93	NA
Horse	450	387	138	149	0.36	0.39
Camel	375	2288	123	132	0.05	0.06
Goat	30	267	23	24	0.09	0.09
Fluconazole
Man	70	22	29	NA	1.32	NA
Horse	500	167	119	43	0.71	0.26
Cat	2.4	1.5	2.4	1.6	1.60	1.07
Cefoperazone
Man	70	74	79	NA	1.07	NA
Horse	415	4981	220	215	0.04	0.04
Calf	80	653	85	83	0.13	0.13
Sheep	40	107	57	56	0.53	0.52
Gentamicin
Man	70	100	72	NA	0.72	NA
Horse	318	567	230	318	0.41	0.56
Sheep	94	87	75	94	0.86	1.08
Goat	28	32	24	28	0.75	0.88
Cat	11	7	11	11	1.57	1.57
Propranolol
Man	70	850	728	NA	0.86	NA
Horse	533	12 181	2891	3303	0.24	0.27
Furosemide
Man	70	154	296	NA	1.92	NA
Horse	500	6450	5878	1384	0.91	0.21
Camel	247	1334	2744	737	2.06	0.55
Thiamphenicol
Calf	75	327	244	NA	0.75	NA
Sheep	66	449	224	NA	0.50	NA
Pig	26.6	356	121	NA	0.34	NA
Sulphadimidine
Horse	275	253	159	NA	0.63	NA
Cattle	603	498	352	NA	0.71	NA
Camel	200	136	115	NA	0.85	NA
Sheep	25	19	14	NA	0.74	NA
Flunixin
Horse	237	492	69	NA	0.14	NA
Donkey	447	427	50	NA	0.12	NA
Mule	467	654	70	NA	0.11	NA
Camel	475	705	71	NA	0.10	NA
Calf	119	397	35	NA	0.09	NA
Sheep	45	29	22	NA	0.76	NA
Enrofloxacin
Mare	540	3331	6388	NA	1.92	NA
Camel	453	2088	5919	NA	2.83	NA
Cow	710	5775	8723	NA	1.51	NA
Sheep	45	209	807	NA	3.86	NA

NA, not applicable.

Inability of RE to improve allometric predictions in large animal species

In humans, the prediction of clearance was improved by the incorporation of MLP or brain weight. The application of these correction factors depends on the exponents of the simple allometry as described previously. To determine whether or not the RE can be applied to allometric scaling in large animal species, the influence of MLP and brain weight was examined using two drugs whose clearance values have been estimated across a wide variety of animal species: enrofloxacin and furosemide (Table 3).

Table 3. Comparison of the predicted Cl (mL/min) based upon simple allometry (SA), maximum life-span potential (MLP) and brain weight (BRW)

Species	Body weight (kg)	Observed Cl	Predicted Cl (SA)	Predicted Cl (MLP)
Enrofloxacin
Sheep*	45	209	863	903
Cow*	700	5775	9216	16 992
Horse*	550	3392	7484	10 188
Camel*	453	2088	6330	8142
Horse^†	550	3392	3635	3155
Camel^†	453	2088	3121	2384
Furosemide				Predicted Cl (BRW)
Horse^‡	500	6450	2730	24 057
Camel^‡	247	1334	1349	4576

*Allometric relationships obtained with SA or MLP × Cl using small animals (mouse, rat, rabbit, dog) was as follows: SA: Cl = 32.3W^0.863, r = 0.991; MLP × Cl = 21.2W^1.169, r = 0.989.
^†Allometric equation is based upon data from dog, sheep and cow (SA: Cl = 25.5W³, r = 0.89; MLP × Cl = 35.3W^0.864, r = 0.915).
^‡Allometric relationships obtained with SA or brain weight × Cl using small animals (rat, rhesus monkey, and dog) was as follows: SA: Cl = 5.46W^1.0, r = 0.979; Cl × BRW = 0.051W^2.0, r = 0.999.

In large animal species, the RE did not improve prediction of clearance (Table 3). Furthermore, unlike that seen with humans, there was no trend in the exponents of the simple allometry that could guide in the selection of a particular correction factor. Nevertheless, for these same compounds, RE did substantially improve our estimates of clearance in humans. For example, the human clearance of furosemide using simple allometry (based upon rat, rhesus monkey and dog) was 381 mL/min. This is substantially higher than the observed human Cl value of 154 mL/min. However, when brain weight was used as a correction factor (allometric exponent in this case = 1.0), the prediction improved substantially (163 mL/min). On the other hand, the application of brain weight introduced substantial error in the predicted clearance of camel and horse.

In the case of enrofloxacin (for which there is no human data), our clearance prediction (based upon data in mice, rat, rabbit and dog) went from 1263 mL/min when using simple allometry to 370 mL/min when using MLP (allometric coefficient was 0.863). Application of the MLP introduced substantial error in the predicted clearance of cattle, horses, and camels. The analysis from these two drugs (and several others not shown here) indicates that although the RE substantially helps in improving the predicted human clearance, it is not applicable to the large animals. At this time, the reasons for this are not known.

Impact of species selection in the prediction of clearance in large animal species

In this set of evaluations, the smallest animal included in the data set for the prediction of clearance was the dog. Human data were included in the scaling when available. At least one large animal species (horse, camel, donkey, mule, or cattle) was included in the scaling. The results of this analysis are shown in Table 4. The errors associated with the predicted clearance based upon the inclusion of larger animal species (termed ‘current’) were compared with the errors in predictions when the scaling was based upon data solely from the small laboratory animal species (as reported in Table 2). These ‘Table 2’ error values represent those error estimates obtained when human data were excluded from the regression analysis. Treating each scaling event for each animal species as a single ‘observation’, the magnitude of error observed across compounds when using either the species included in Table 2 vs. those in Table 4 (termed ‘current’), was compared to assess if there are differences in the overall accuracy characteristics of these allometric estimates (Fig. 1). Note that when error = 1, the clearance value predicted by scaling equals the observed clearance value.

Table 4. Predicted and observed clearance (mL/min) in large animals with and without inclusion of a large animal based on SA

Species used	Exponents	R	Species predicted	Observed Cl	Predicted Cl	Error
Species used	Exponents	R	Species predicted	Observed Cl	Predicted Cl	Current	Table 2
Caffeine
Dog, man, donkey	0.727	0.998	Horse	312	340	0.92	1.66
Dog, man, donkey	0.727	0.998	Camel	292	82	3.56	6.44
Dog, man, horse	0.762	0.999	Donkey	221	256	1.16	1.91
Dog, man, horse	0.762	0.999	Camel	82	323	3.94	6.44
Antipyrine
Dog, man, camel	0.827	0.765	Horse	2790	860	0.31	0.06
Dog, man, horse	0.951	0.795	Camel	1613	1038	0.64	0.09
Theophylline
Dog, goat, horse	0.688	0.793	Camel	2288	478	0.21	0.05
Dog, goat, camel	1.261	0.957	Horse	387	3657	9.45	0.36
Enrofloxacin
Dog, sheep, cow	0.784	0.897	Mare	3331	3552	1.07	1.92
Dog, sheep, cow	0.784	0.897	Camel	2088	3095	1.47	2.83
Dog, sheep, mare	0.676	0.969	Cow	5775	3063	0.53	1.51
Dog, sheep, mare	0.676	0.969	Camel	2088	2261	1.08	2.83
Dog, mare, camel	0.554	0.987	Cow	2088	3230	1.55	1.51
Dog, mare, camel	0.554	0.987	Sheep	209	703	3.36	3.86
Cefoperazone
Dog, man, horse	1.189	0.875	Calf	653	348	0.53	0.13
Dog, man, horse	1.189	0.875	Sheep	107	153	1.43	0.53
Dog, man, calf	0.703	0.504	Horse	4981	758	0.15	0.04
Dog, man, calf	0.703	0.504	Sheep	107	146	1.36	0.53
Furosemide
Dog, man, horse	1.149	0.934	Camel	1334	1740	1.30	2.06
Dog, man, camel	0.864	0.923	Horse	4981	1632	0.33	0.91
Sulphadimidine
Dog, goat, horse	1.248	0.990	Cattle	498	737	1.48	0.71
			Camel	136	186	1.37	0.85
			Sheep	19	14	0.74	0.74
Dog, goat, cattle	1.161	0.993	Horse	253	217	0.86	0.63
			Camel	136	150	1.10	0.71
			Sheep	19	13	0.68	0.74
Dog, goat, camel	1.175	0.988	Cattle	498	555	1.11	0.71
			Horse	253	220	0.87	0.63
			Sheep	19	13	0.68	0.74
Flunixin
Sheep, calf, donkey	1.690	0.920	Horse	492	1868	3.80	0.14
			Mule	654	2012	3.08	0.11
			Camel	705	2070	2.94	0.10
Sheep, mule, donkey	1.386	0.987	Horse	492	754	1.53	0.14
			Calf	397	120	0.30	0.09
			Camel	705	820	1.16	0.10
Dog, sheep, donkey	1.173	0.973	Horse	492	707	1.44	0.14
			Calf	397	150	0.38	0.09
			Camel	29	759	1.08	0.10
			Mule	654	744	1.14	0.11
Dog, sheep, camel	1.109	0.986	Horse	492	565	1.15	0.14
			Calf	397	130	0.33	0.10
			Sheep	29	44	1.52	0.76
			Mule	654	593	0.91	0.11

Details are in the caption following the image — **Figure 1**
Open in figure viewer PowerPoint

Comparison of error seen from large animal predictions based on small species vs. larger animal species.

From these results, we see that for any of the large animal species, the predicted clearances are highly dependent upon the species included in the regression analysis. For some species and compounds, there were large discrepancies between the predicted clearance values, depending upon the large animal species included in the allometric assessment. With regard to comparing the error from this ‘current’ assessment to the error values reported in Table 2, for most (but not all) compounds, the magnitude of the error was substantially reduced when at least one large animal species was included in data set used for the predictions. As seen in Fig. 1, there was a much greater proportion of error falling within the range of 0.5–1.5 (i.e., more accurate predictions) when large animal species were included in the data set when compared with when only laboratory animals were included in the analysis. This is consistent with the importance of including large animal species in the allometric analysis, even though there was a slightly greater risk of overestimating the true clearance values when large animal species were included in the regression analysis.

In general, when the equation is based solely upon small laboratory species, there is a great propensity for substantially underestimating clearance and, unlike that observed for humans, the drug clearance in large animal species is generally poorly predicted based on data obtained in laboratory animals.

Discussion

Despite the potential for extrapolation error, the reality is that allometric scaling is needed across many veterinary practice situations and therefore will be used. For this reason, it is important to consider mechanisms for reducing the risk of extrapolation errors that can seriously affect target animal safety, therapeutic response, or the accuracy of withdrawal time predictions. Based upon this analysis, it appears that inclusion of at least one large animal in the scaling (with or without human data) can improve predictions. As discussed in part II, there may be several reasons for this finding.

Another important consideration is the exponent of allometry. As mentioned previously, if the exponents of the simple allometry are <0.55 then the predicted clearance in humans is likely to be lower than the observed values. The magnitude of this underestimation cannot be determined by the value of the exponent. If the exponents of the simple allometry are >1.3, the predicted clearance in humans will be overestimated by several orders of magnitude, even when using brain weight as a correction factor. In large animals, correction factors could not be applied because there was no trend seen between the exponents of the simple allometry and the appropriate correction factor for improving our predictions. Nevertheless, we did see an indication that the exponents of simple allometry may alert us to when the predicted clearance in the large animal may be underestimated or overpredicted. An example of this was seen with flunixin. The exponent of allometry based upon small laboratory animals was 0.497. The predicted clearances of flunixin in six large animal species were substantially underestimated (Table 2). When the large animal species were included in the scaling, the exponents of allometry increased (Table 4). Depending upon the species included in the regression analysis, the allometric exponents ranged from 1.11 to 1.69. The predicted clearances of flunixin were near the observed clearances for large animal species when the exponent of the allometry was either 1.173 or 1.109. However, clearance values were grossly overestimated when the allometric exponent was 1.69. Therefore, it is suggested that if a large animal species is included in the scaling, the predicted clearance in the other large animal species should be considered overestimated if the exponent of the simple allometry is >1.3.

The use of allometric scaling to predicted clearance across species always has the risk of ignoring potential differences in clearance mechanisms and metabolic pathways. For example, most species either convert ketoprofen from the R to the S enantiomer (rodents, dogs, rabbits, monkey, horse, cat) or do not convert at all (cattle and llamas). Asian elephants are the only studied species that appears to convert S to R. Elephants also appear to produce a glycine conjugate not reported in other species (Hunter et al., 2003). In addition, the vast majority of large animals for which pharmacokinetic data are available are herbivores. Conversely, the smaller animal species are either omnivores (mouse, rat, monkey and human) or carnivores (dogs and cats). This difference in feeding behavior itself can influence drug metabolism and renal elimination. A detailed discussion of the differences in drug metabolism that may influence extrapolation error when small animals are used for human preclinical drug development has been reviewed elsewhere (Martinez, 2006). While large carnivore data (lion, tiger, polar bear, etc) would be helpful, the obvious dangers hinder this area of research. Some marine mammal species may be worthy of consideration, but due to body fat composition of species such as killer whales (Orcinus orca), the results would be of limited use.

Another point of consideration is the interspecies differences in the contribution of various organ systems and differences in body composition as it pertains to the body weight estimates. For example, when using ruminants in an allometric evaluation, we also need to consider the percentage of the estimated value of weight that is associated with rumen fluids. In an adult cow (e.g., 1300 lb), the rumen can contain more than 80 gallons of grain and fluid (Ensminger et al., 1990). When we consider sheep, 13% of their body weight is attributable to rumen contents and almost 12% of their live weight is associated with fleece and pelt (MacGregor, 1980). On the other hand, the wet weight of digestive contents in 10 African wild elephants in Kruger National Park was estimated to be approximately 10% of the total body weight (A. Craigmill, personal communication; Isaza & Hunter, 2004). When considering this value on a log scale, the latter translates into W = 2.7 for digestive contents and 3.7 for the body weight (where an adult African elephant is estimated to weigh between 4100 and 5000 kg) (Schmidt, 1986). From these values, we can see how these certain species-specific anatomical traits can greatly influence the accuracy of the predicted clearance.

Even if these normal physiological differences could be ignored, there remains the error attributable simply to extrapolation. For large animal species, when generating an equation using data derived from mice, rats and dogs, there is almost a 600-fold range of weights. When using that equation to extrapolate to man, there is a five- to sixfold difference in weight between actual (dog) data vs. extrapolated (human) data. On the other hand, if an allometric equation is established solely on the basis of data obtained in large animal species such as cattle, horses, and donkeys, there is less than a twofold range in weights. Therefore, using only large animal species for generating the allometric equation will compromise our ability to adequately describe the slope and intercept as a function of body weight. These potential sources of extrapolation error are further examined in part II.

References

Ali, B.H., Wong, Y.C., Alhadrami, G.A., Charles, B.G. & Bashir, A.K. (1998) Plasma pharmacokinetics of intravenous and intramuscular furosemide in the camel (Camelus dromedarius). Research in Veterinary Science, 64, 69–72.
10.1016/S0034-5288(98)90118-7
CAS PubMed Web of Science® Google Scholar
Aramaki, S., Mori, M., Nakata, M., Shinohara, A. & Koizumi, T. (2000) Pharmacokinetics of propranolol and its metabolites in horses after intravenous or oral administration. Biological and Pharmaceutical Bulletin, 23, 1333–1340.
10.1248/bpb.23.1333
CAS PubMed Web of Science® Google Scholar
Bachmann, K. (1989) Predicting toxicokinetic parameters in humans from toxicokinetic data acquired from three small mammalian species. Journal of Applied Toxicology, 9, 331–338.
10.1002/jat.2550090509
CAS PubMed Web of Science® Google Scholar
Balani, S.K., Zhu, T., Yang, T.J., Liu, Z., He, B. & Lee, F.W. (2002) Effective dosing regimen of 1-aminobenzotriazole for inhibition of antipyrine clearance in rats, dogs, and monkeys. Drug Metabolism and Disposition, 30, 1059–1062.
10.1124/dmd.30.10.1059
CAS PubMed Web of Science® Google Scholar
Ben-Zvi, Z., Rubin, M., Van Creveld, C. & Yagil, R. (1995) Antipyrine disposition in the dehydrated camel. Journal of Veterinary Pharmacology and Therapeutics, 18, 137–140.
10.1111/j.1365-2885.1995.tb00567.x
CAS PubMed Web of Science® Google Scholar
Bonati, M., Latini, R., Tognoni, G., Young, J.F. & Garattini, S. (1984) Interspecies comparison of in vivo caffeine pharmacokinetics in man, monkey, rabbit, rat, and mouse. Drug Metabolism Review, 15, 1355–1383.
10.3109/03602538409029964
CAS PubMed Web of Science® Google Scholar
Boxenbaum, H. & Dilea, C. (1995) First-time-in-human dose selection: allometric thoughts and perspectives. Journal of Clinical Pharmacolology, 35, 957–966.
10.1002/j.1552-4604.1995.tb04011.x
CAS PubMed Web of Science® Google Scholar
Boxenbaum, H. & Fertig, J.B. (1984) Scaling of antipyrine intrinsic clearance of unbound drug on 15 mammalian species. European Journal of Drug Metabolism and Pharmacokinetics, 9, 177–183.
10.1007/BF03189622
CAS PubMed Web of Science® Google Scholar
Bregante, M.A., Saez, P., Aramayona, J.J., Fraile, L., Garcia, M.A. & Solans, C. (1999) Comparative pharmacokinetics of enrofloxacin in mice, rats, rabbits, sheep, and cows. American Journal of Veterinary Research, 60, 1111–1116.
CAS PubMed Web of Science® Google Scholar
Castells, G., Capece, B.P., Perez, F., Marti, G., Arboix, M. & Cristofol, C. (2001) Allometric analysis of thiamphenicol disposition among seven mammalian species. Journal of Veterinary Pharmacology and Therapeutics, 24, 193–197.
10.1046/j.1365-2885.2001.00331.x
CAS PubMed Web of Science® Google Scholar
Cheng, Z., McKeller, Q. & Nolan, A. (1998) Pharmacokinetic studies of flunixin meglumine and phenylbutazone in plasma, exudate and transudate in sheep. Journal of Veterinary Pharmacology and Therapeutics, 21, 315–321.
10.1046/j.1365-2885.1998.00144.x
CAS PubMed Web of Science® Google Scholar
Coakley, M., Peck, K.E., Taylor, T.S., Matthews, N.S. & Mealey, K.L. (1999) Pharmacokinetics of flunixin meglumine in donkeys, mules, and horses. American Journal of Veterinary Research, 60, 1441–1444.
CAS PubMed Web of Science® Google Scholar
Doyle, E., Chasseaud, L.F. & Miller, J.N. (1982) Comparative pharmacokinetics of frusemide in female rhesus monkeys, cynomolgus monkeys and baboons. Comparative Biochemistry Physiology C, 71C, 89–93.
10.1016/0306-4492(82)90015-6
CAS PubMed Web of Science® Google Scholar
Dyke, T.M., Sams, R.A. & Hinchcliff, K.W. (1998) Antipyrine pharmacokinetics and urinary excretion in female horses. American Journal of Veterinary Research, 59, 280–285.
CAS PubMed Web of Science® Google Scholar
Elsheikh, H.A., Ali, B.H., Homeida, A.M., Hassan, T. & Hapke, H.J. (1991) Pharmacokinetics of antipyrine and sulphadimidine (sulfamethazine) in camels, sheep and goats. Journal of Veterinary Pharmacology and Therapeutics, 14, 269–275.
10.1111/j.1365-2885.1991.tb00837.x
CAS PubMed Web of Science® Google Scholar
Elsheikh, H.A., Ali, B.H., Zahurin, M., Mustafa, A.M., Alhadrami, G. & Bashir, A.K. (2001) Comparative pharmacokinetics of theophylline in camels (Camelus dromedarius) and goats (Caprus hircus). Journal of Veterinary Medical A Physiology and Pathology Clinical Medicine, 48, 581–586.
10.1046/j.1439-0442.2001.00401.x
CAS PubMed Web of Science® Google Scholar
Ensminger, M.E., Oldfield, J.E. & Heinemann, W.W. (1990) Feeds and Nutrition, p. 53. Ensminger Publishing Company, Clovis, CA.
Google Scholar
Evans, G.H., Nies, A.S. & Shand, D.G. (1973) The disposition of propranolol III. Decreased half-life and volume of distribution as a result of plasma binding in man, monkey, dog, and rat. Journal of Pharmacology and Experimental Therapeutics, 186, 114–122.
CAS PubMed Web of Science® Google Scholar
Gascon, A.R., Calvo, B., Hernandez, R.M., Dominguez-Gil, A. & Pedraz, J.L. (1994) Interspecies scaling of cimetidine-theophylline pharmacokinetic interaction: interspecies scaling in pharmacokinetic interactions. Pharmaceutical Research, 11, 945–950.
10.1023/A:1018914816137
CAS PubMed Web of Science® Google Scholar
Gavrielli, R., Yagil, R., Ziv, G., Creveld, C.V. & Glickman, A. (1995) Effect of water deprivation on the disposition kinetics of enrofloxacin in camels. Journal of Veterinary Pharmacology and Therapeutics, 18, 333–339.
10.1111/j.1365-2885.1995.tb00600.x
PubMed Web of Science® Google Scholar
Haines, G.R., Brown, M.P., Gronwall, R.R. & Merritt, K.A. (2000) Serum concentrations and pharmacokinetics of enrofloxacin after intravenous and intragastric administration to mares. Canadian Journal of Veterinary Research, 64, 171–177.
CAS PubMed Web of Science® Google Scholar
Hardie, E.M., Hardee, G.E. & Rawlings, C.A. (1985) Pharmacokinetics of flunixin meglumine in dogs. American Journal of Veterinary Research, 46, 235–237.
CAS PubMed Web of Science® Google Scholar
Hunter, R.P., Isaza, R. & Koch, D.E. (2003) Oral bioavailability and pharmacokinetic characteristics of ketoprofen enantiomers after oral and intravenous administration in Asian elephants (Elephas maximus). American Journal of Veterinary Research, 64, 109–114.
10.2460/ajvr.2003.64.109
CAS PubMed Web of Science® Google Scholar
Ingvast-Larsson, C., Paalzow, G., Paalzow, L., Ottosson, T., Lindholm, A. & Appelgren, L.E. (1985) Pharmacokinetic studies of theophylline in horses. Journal of Veterinary Pharmacology and Therapeutics, 8, 76–81.
10.1111/j.1365-2885.1985.tb00927.x
CAS PubMed Web of Science® Google Scholar
Isaza, R. & Hunter, R.P. (2004) Drug delivery to captive Asian Elephants – treating goliath. Current Drug Delivery, 1, 291–298.
10.2174/1567201043334641
CAS PubMed Google Scholar
Jezequel, S.G. (1994) Fluconazole: interspecies scaling and allometric relationships of pharmacokinetic properties. Journal of Pharmacy and Pharmacology, 46, 196–199.
10.1111/j.2042-7158.1994.tb03777.x
CAS PubMed Web of Science® Google Scholar
Kung, K., Riond, J.L. & Wanner, M. (1993) Pharmacokinetics of enrofloxacin and its metabolite ciprofloxacin after intravenous and oral administration of enrofloxacin in dogs. Journal of Veterinary Pharmacology and Therapeutics, 16, 462–468.
10.1111/j.1365-2885.1993.tb00212.x
PubMed Web of Science® Google Scholar
Landoni, M.F., Cunningham, F.M. & Lees, P. (1995) Determination of pharmacokinetics and pharmacodynamics of flunixin in calves by use of pharmacokinetic/pharmacodynamic modeling. American Journal of Veterinary Research, 56, 786–794.
CAS PubMed Web of Science® Google Scholar
Latimer, F.G., Colitz, C.M., Campbell, N.B. & Papich, M.G. (2001) Pharmacokinetics of fluconazole following intravenous and oral administration and body fluid concentrations of fluconazole following repeated oral dosing in horses. American Journal of Veterinary Research, 62, 1606–1611.
10.2460/ajvr.2001.62.1606
CAS PubMed Web of Science® Google Scholar
Lave, T., Dupin, S., Schmitt, C., Chou, R.C., Jaeck, D. & Coassolo, P. (1997) Integration of in vitro data into allometric scaling to predict hepatic metabolic clearance in man: application to 10 extensively metabolized drugs. Journal of Pharmaceutical Sciences, 86, 584–590.
10.1021/js960440h
CAS PubMed Web of Science® Google Scholar
Lees, P. & Taylor, P.M. (1991) Pharmacodynamics and pharmacokinetics of flunixin in the cat. British Veterinary Journal, 147, 298–305.
10.1016/0007-1935(91)90002-5
CAS PubMed Web of Science® Google Scholar
MacGregor, R. (1980) The Structure of the Meat Animals. A Guide to their Anatomy and Physiology. 225 pp. Technical Press, Oxford.
Google Scholar
Mahmood, I. (2005) Interspecies Pharmacokinetic Scaling: Principles and Application of Allometric Scaling. 393 pp. Pine House Publishers, Rockville, MD.
Google Scholar
Mahmood, I. & Balian, J.D. (1996) Interspecies scaling: predicting clearance of drugs in humans. Three different approaches. Xenobiotica, 26, 887–895.
10.3109/00498259609052491
CAS PubMed Web of Science® Google Scholar
Martin-Jimenez, T. & Riviere, J.E. (2001) Mixed effects modeling of the disposition of gentamicin across domestic animal species. Journal of Veterinary Pharmacology and Therapeutics, 24, 321–332.
10.1046/j.1365-2885.2001.00346.x
CAS PubMed Web of Science® Google Scholar
Martinez, M.N. (2006) Interspecies differences in physiology and pharmacology: extrapolating preclinical data to human populations. In: Preclinical Drug Development Eds M. Rogge & D. Taft pp. 11–66. Marcel Dekker, New York (in press).
Google Scholar
McNamara, D.R., Nafziger, A.N., Menhinick, A.M. & Bertino, J.S., Jr (2001) A dose-ranging study of gentamicin pharmacokinetics: implications for extended interval aminoglycoside therapy. Journal of Clinical Pharmacology, 41, 374–377.
10.1177/00912700122010221
CAS PubMed Web of Science® Google Scholar
Nouws, J.F., Firth, E.C., Vree, T.B. & Baakman, M. (1987) Pharmacokinetics and renal clearance of sulfamethazine, sulfamerazine, and sulfadiazine and their N4-acetyl and hydroxy metabolites in horses. American Journal of Veterinary Research, 48, 392–402.
CAS PubMed Web of Science® Google Scholar
Obach, R.S., Baxter, J.G., Liston, T.E., Silber, B.M., Jones, B.C., Macintyre, F., Rance, D.J. & Wastall, P. (1997) The prediction of human pharmacokinetic parameters from preclinical and in vitro metabolism. Journal of Pharmacology and Experimental Therapeutics, 283, 46–58.
CAS PubMed Web of Science® Google Scholar
Peck, K., Mealey, K.L., Matthews, N.S. & Taylor, T.S. (1997) Comparative pharmacokinetics of caffeine and three metabolites in clinically normal horses and donkeys. American Journal of Veterinary Research, 58, 881–884.
CAS PubMed Web of Science® Google Scholar
Riffat, S., Nawaz, M. & Zia-Ur-Rehman (1982) Pharmacokinetics of sulphadimidine in normal and febrile dogs. Journal of Veterinary Pharmacology and Therapeutics, 5, 131–135.
10.1111/j.1365-2885.1982.tb00508.x
CAS PubMed Google Scholar
Sacher, G.A. (1959) Relation of Lifespan to Brain Weight and Body Weight in Mammals. 115 pp. Ciba Foundation Colloquium, Ageing, Basilea.
Google Scholar
Sawada, Y., Hanano, M., Sugiyama, Y. & Iga, T. (1984) Prediction of the disposition of beta-lactam antibiotics in humans from pharmacokinetic parameters in animals. Journal of Pharmacokinetics and Biopharmaceutics, 12, 241–261.
10.1007/BF01061720
CAS PubMed Web of Science® Google Scholar
Schmidt, M. (1986) Elephants (Proboscidea). In Zoo & Wild Animal Medicine, 2nd edn. Ed. M.E. Fowler. 883 pp. Sauders, Philadelphia, PA.
Google Scholar
Soraci, A.L., Mestorino, O.N. & Errecalde, J.O. (1996) Pharmacokinetics of cefoperazone in horses. Journal of Veterinary Pharmacology and Therapeutics, 19, 39–43.
10.1111/j.1365-2885.1996.tb00006.x
CAS PubMed Web of Science® Google Scholar
Tanira, M.O., Ali, B.H. & Bashir, A.K. (1997) Effect of endotoxin on gentamicin pharmacokinetics in old and young adult rats. Life Sciences, 60, 413–424.
10.1016/S0024-3205(96)00666-2
CAS PubMed Web of Science® Google Scholar
Wasfi, I.A., Boni, N.S., Abdel Hadi, A., Elghazali, M., Zorob, O., Alkatheeri, N.A. & Barezaiq, I.M. (1998) Pharmacokinetics, metabolism and urinary detection time of flunixin after intravenous administration in camels. Journal of Veterinary Pharmacology and Therapeutics, 21, 203–208.
10.1046/j.1365-2885.1998.00122.x
CAS PubMed Web of Science® Google Scholar
Wasfi, I.A., Boni, N.S., Elghazali, M., Abdel Hadi, A.A., Almuhrami, A.M., Barezaig, I.M. & Alkatheeri, N.A. (2000) The pharmacokinetics, metabolism and urinary detection time of caffeine in camels. Research in Veterinary Science, 69, 69–74.
10.1053/rvsc.2000.0389
CAS PubMed Web of Science® Google Scholar
Witkamp, R.F., Yun, H.I., Van't Klooster, G.A., Van Mosel, J.F., Van Mosel, M., Ensink, J.M., Noordhoek, J. & Van Miert, A.S. (1992) Comparative aspects and sex differentiation of plasma sulfamethazine elimination and metabolite formation in rats, rabbits, dwarf goats, and cattle. American Journal of Veterinary Research, 53, 1830–1835.
CAS PubMed Web of Science® Google Scholar

Citing Literature

Volume29, Issue5

October 2006

Pages 415-423

Interspecies allometric scaling. Part I: prediction of clearance in large animals

Abstract

Introduction

Methods

Method 1: simple allometry

Method 2: product of maximum life span potential and clearance

Method 3: product of brain weight and clearance

Results

Prediction of clearance from laboratory animals to large animal species with and without human data

Inability of RE to improve allometric predictions in large animal species

Impact of species selection in the prediction of clearance in large animal species

Discussion

References

Citing Literature

Figures

References

Information

About Wiley Online Library

Help & Support

Opportunities

Connect with Wiley

Interspecies allometric scaling. Part I: prediction of clearance in large animals

Abstract

Introduction

Methods

Method 1: simple allometry

Method 2: product of maximum life span potential and clearance

Method 3: product of brain weight and clearance

Results

Prediction of clearance from laboratory animals to large animal species with and without human data

Inability of RE to improve allometric predictions in large animal species

Impact of species selection in the prediction of clearance in large animal species

Discussion

References

Citing Literature

Figures

References

Related

Information