About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Another Use for Xenon | Main | Ezetimibe In the Marketplace »

March 21, 2014

Dosing by Body Surface Area

Email This Entry

Posted by Derek

We were talking about allometry around here the other day, which prompts me to mention this paper. It used the reports of resveratrol dosing in animals, crudely extrapolated to humans, to argue that the body surface area normalization (BSA) method was a superior technique for dose estimation across species.

Over the years, though, the BSA method has taken some flak in the literature. It's most widely used in oncology, especially with cytotoxics, but there have been calls to move away from the practice, calling it a relic with little scientific foundation. (The rise of a very obese patient population has also led to controversy about whether body weight or surface area is a more appropriate dose-estimation method in those situations). At the same time, it's proven useful in some other situations, so it can't be completely ignored.

But it seems that the FASEB paper referenced in the first paragraph, which has been cited hundreds of times since 2008, may be overstating its conclusions. For example, it says that "BSA normalization of doses must be used to determine safe starting doses of new drugs because initial studies conducted in humans, by definition, lack formal allometric comparison of the pharmacokinetics of absorption, distribution, and elimination parameters", and cites its reference 13 for support. But when you go to that reference, you find that paper's authors concluding with things like this:

The customary use of BSA in dose calculations may contribute to the omission of these factors, give a false sense of accuracy and introduce error. It is questionable whether all current cancer treatment strategies are near optimal, or even ethical. BSA should be used for allometric scaling purposes in phase I clinical trials, as the scaling of toxicity data from animals is important for selecting starting doses in man, but the gradual discontinuation of BSA-based dosing of cytotoxic drugs in clinical practice is seemingly justified.

Citing a paper for support that flatly disagrees with your conclusions gets some points for bravado, but otherwise seems a bit odd. And there are others - that reference that I linked to in the second paragraph above, under "taken some flak", is cited in the FASEB paper as its reference 17, as something to do with choosing between various BSA equations. And it does address that, to be sure, but in the context of wondering whether the whole BSA technique has any clinical validity at all.

This is currently being argued out over at PubPeer, and it should be interesting to see what comes of it. I'll be glad to hear from pharmacokinetics and clinical research folks to see what they make of the whole situation.

Comments (17) + TrackBacks (0) | Category: Pharmacokinetics | The Scientific Literature


1. Fat Old Man on March 21, 2014 9:20 AM writes...

Dosing by BSA is fine for IV formulations like most cytotoxics where fine control of the dose for AE reasons is practical. Patients like to take 1 tablet once a day so dosing by BSA is not so practical there, too many dose units to develop.

Permalink to Comment

2. newnickname on March 21, 2014 9:31 AM writes...

"Citing a paper for support that flatly disagrees with your conclusions gets some points for bravado"

Hah! I know of some papers that selectively IGNORE work from their own group that contradicts or undermines the new results and then, as you point out, selectively cite other sources as "supporting" that do NOT, in fact, support their claims.

If you get that truck-driving gig, please take me along for the ride. I can spell you on the night shift.

Permalink to Comment

3. RTW on March 21, 2014 10:04 AM writes...

Interesting to note that Oncology has started to look at instead of dosing Pt containing cytotoxic drugs via BSA - to actually do dose escalation and measure blood levels, to give a more correct therapeutic dosage. They want to measure each patients PK/PD responses to these to tailor them. Some people tolerate for example Oxaliplatin better as well as much worse than similarly set BSA levels. Likely because of individual PK/PD response.

Permalink to Comment

4. simpl on March 21, 2014 11:52 AM writes...

Here is a model from the olden days:

- If at dose X the efficacy is bad, double the dose.
- If the side effects are bad, half the dose.
- If you can't get efficacy without bad side-effects, the drug is bad.

As example, here is a link to guanfacine, with tables on efficacy and side-effects from 0.5mg to 5mg.
It's a pretty clear case, I'd say.

5mg of guanfacine was the launch dose for cv, but the standard dose dropped to 1mg pretty fast.

Permalink to Comment

5. Anonymous on March 21, 2014 1:55 PM writes...

dosing based on Lean Body Mass has been proposed as a better way to go.

Permalink to Comment

6. David E. Young, MD on March 21, 2014 3:11 PM writes...

All of this discussion is appropriate. For many cytotoxics, body surface area is just one part of safe and efficacious dosing. Take Carboplatin, of instance, where dosing is now handled by determinng the "area under the curve" based upon BSA and renal function. As RTW mentions above, a sister drug, Oxaliplatin may be best dosed by that method. As an oncologist, I generally round off my doses, but base my rounding on organ function and prior experience with that patient. If my patient getting adjuvant Fluorouracil for a resected colon cancer, and there are no side effects, I will then increase the dose for subsequent treatment.

But what you say reminds me of the folly of training programs, where the pharmacist would be asked to prepare 750 mg per meter squared of Cyclophosphamide and deliver a bag of drug with a label of "Cyclophosphamide, 1,087.6 mg". I am sure that the drug sponsers for the clnical study were happy that we paid that much attention to detail even though that level of accuracy was meaningless.

Permalink to Comment

7. Anonymous on March 21, 2014 4:05 PM writes...

This is interesting. Given a roughly constant shape and density of the human body, surface area A should vary with mass M as: A = k M^(2/3)

Also, if a drug's dose correlates more with body surface area than with body mass, this indicates that it depends more on the surface areas of cell *membranes* than on cell or tissue *volumes*. In other words the dose is driven by kinetics (PK) rather than by mass equilibria!

Permalink to Comment

8. Erebus on March 22, 2014 7:45 AM writes...

For dose conversions from rats to humans, I think that those BSA calculations are 'industry standard', these days. The FDA even gives guidelines on how to use 'em to estimate starting doses for Phase-I clinical trials. (They discuss other methods, but those BSA equations feature most prominently. The document is called "Guidance for Industry - Estimating the Maximum Safe Starting Dose in Initial Clinical Trials for Therapeutics in Adult Healthy Volunteers" and is a free download.)

...There are other options, and I wholeheartedly agree that these (X*3/37=HED) equations are very crude -- but, really, all we need for dose conversions is a reasonable starting point. Measurements can be taken from there, and adjustments can be made. In any case, it seems that those BSA calculations generally provide an adequate dose for Phase I trials. More often than not, anyhow...

Where clinical use is concerned, with cytotoxics and other narrow-therapeutic-window agents, I'm not a huge fan of the BSA method. It ignores too many important variables.

Permalink to Comment

9. Hamstav on March 22, 2014 1:53 PM writes...

When you go back to Dubois & Dubois original paper on BSA, you find all kind of funny things that can help you measure the changes in the last hundred years. First, you can read the study participant's name and comments on their morphology. Then, you realize that there are almost no children in his experiment. The study wanted to find the formula that best predicts an estimation of the BSA based on cylinders theory.

In the end, what you get is an equation that estimates the estimation of BSA. Most studies afterward tries to find simpler formulas to correlate to the Dubois & Dubois equations. You then get an estimation of the estimation of the estimation...

The problem when drugs are marketed with BSA dosing, is that BSA is not easily calculated and is based on 2 measures plus a calculation that will differ from one clinician to the other. In the end, it's not exceptional to have dosing that differ form up to 15% depending on the method of determination.

Since the calculations depends on relatively complex formulas or tools, there are some errors in prescriptions and this is what bothers me the most as a pharmacist. If there are no proven clinical benefits and it promotes error, we should stop doing this. BSA should be used only if proven to be a superior method of dosing compared to body weight, and I doubt this would happen a lot in clinical practice.

Permalink to Comment

10. Carl Lumma on March 22, 2014 6:55 PM writes...

A large body of allometric research suggests M^(3/4) where M is body mass. Advanced on an empirical basis by Max Kleiber, it was given theoretical footing by West, Brown & Enquist in the late '90s. Subsequent debate has been lively and has only seemed to strengthen WBE's case.

Permalink to Comment

11. Anonymous on March 23, 2014 9:35 AM writes...

'Derek Lowe to fill in for Jerry Remy on NESN'
....... For a moment , I tht........

Permalink to Comment

12. Anonymous on March 23, 2014 9:35 AM writes...

'Derek Lowe to fill in for Jerry Remy on NESN'
....... For a moment , I tht........

Permalink to Comment

13. Anonymous BMS Researcher on March 23, 2014 9:30 PM writes...

Picking the starting dose for a first-in-human trial is always a challenging exercise. Not only is there the question "which allometric scaling equation to use?" but also there is the question "to which model organism should we apply that allometric equation?" In my experience, all too frequently when we apply one of the standard equations to preclinical data from several species, we get rather different projected human doses depending which species is used. Of course we have no way to tell which species would be the best predictor.

Permalink to Comment

14. ex-FDA reviewer on March 24, 2014 5:09 AM writes...

#13, you use the more sensitive species

Permalink to Comment

15. JS on March 24, 2014 12:13 PM writes...

The authors should have realized there's a special dose-conversion algorithm for resveratrol: HED = X^0 - 1.

Permalink to Comment

16. emjeff on March 24, 2014 1:13 PM writes...

The reason BSA (sort of) works is that it is dependent on weight and height - which are two pretty good ways of measuring body size. As pointed out above, the clearance of most drugs scale to weight to the 3/4 power. So, why would you use a crude surrogate for body size (BSA) when a very accurate one (weight) is available? The answer is, you wouldn't, unless you were fighting 100 years of medicinal folklore...

Permalink to Comment

17. Anonymous on March 25, 2014 12:39 PM writes...

@16: Question is: Does it vary with body shape at fixed weight? If so that would suggest BSA has some effect that is independent of weight.

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry