To the Editor:
The article by Griggs et al1
gave guidelines for appropriate dosing for obese adult patients that use body-surface area (BSA) to determine dose. The oncology community is in general agreement that patients should be given an optimal dose that results in clinical benefit and that a reduction in “dose-intensity may compromise disease-free survival (DFS) and overall survival (OS) in the curative setting.”1(p1553)
Historically, dosing of obese patients has been problematic, and to date, there has been no standardized approach. Although Griggs et al propose the use of BSA to determine dose, it seems that this recommendation lacks true level-one evidence. The new recommendation aims to achieve the same maximum dose-intensity in the obese, arguing for full weight–based calculation of BSA to determine doses as opposed to using a reduced weight metric or dose capping. However, there is growing evidence that there are significant limitations with respect to BSA-based dosing.
BSA dosing was originally derived, in 1916 using eight patients, by DuBois and DuBois2
to adjust for basal metabolic rates in estimating the human starting dose from animal doses. This formula was used by Freireich in the 1960s to achieve uniformity in dosing patients who were being treated with phase I cytotoxics.3
However, there is no scientific basis for such use of BSA, and there is growing evidence that this approach is, in fact, invalid.4-10
BSA dosing is associated with high pharmacokinetic variability and is a poor indicator of optimal drug exposure. On this point, Baker et al4
reviewed 33 investigational agents and found that BSA-based dosing reduced interpatient variability for only five (15%). Moreover, the reduction in clearance variability was between 15% and 35%, which indicates that only up to one third of the variability was explained by BSA. Felici et al5
reported the variability in clearance of the most commonly used cytotoxics to be between 25% and 70%, with most drugs showing variability above 35%. They concluded: “BSA failed to individualize the effects of the majority of agents explored.”5(p1677)
More recently, the exposure effect relationship has been well-characterized for many chemotherapeutics. For carboplatin, fluorouracil, docetaxel, and paclitaxel, the levels of exposure beyond which unacceptable toxicity occurs have been identified.11-17
These maximum tolerated exposures (MTEs) have generally been derived from biologic effects correlated to exposure levels after BSA-based dosing. The exposure for fluorouracil varies 30- to 100-fold and for docetaxel up to seven-fold. This variability reflects differences in the way individual patients clear drugs, which underscores the need for an alternative to BSA-based dosing. It would be more appropriate to titrate an individual patient's dose to an MTE by means of therapeutic drug monitoring. In this regard, the most widely used measure of exposure is area under the concentration curve.
Some may argue that there is insufficient evidence proving the benefit of therapeutic drug monitoring. Yet, have such rigorously controlled randomized trials been performed to prove the benefit of BSA-based dosing? In fact, the literature shows that BSA-based dosing does not achieve its goal of reducing pharmacokinetic variability and that it is merely used as a matter of convenience. It gives the false impression that we are practicing personalized medicine by using a patient-specific metric. If anything, the calculations involved in determining the dose—no matter how simple—are likely to result in increased dosing errors. To provide our patients with the best care, we need to understand the issues and take the appropriate steps to ensure that the MTE is, in fact, achieved.
Presently, analytic technologies are routinely available in clinical chemistry laboratories, and we are able to pharmacokinetically guide dosing in the clinic. Such a guided approach is already routine with transplantation agents, antiepileptics, and antibiotics. In addition, this approach would deal with many of the issues acknowledged by Griggs et al.1
The extrapolation of the rather sparse published data from the obese subpopulations of trials to the extremes of morbid obesity and pharmacogenetic or other covariate effects on drug clearance would be of no concern if dose management were to be guided by pharmacokinetics.
Moving away from BSA-based dosing presents challenges, but we should realize the potential gains and initiate clinical studies to support the better alternative. Oncology has made great strides during the past 50 years. Let us hope that, in another 50 years, we are not still recommending BSA-based dosing and that we will have moved to dosing a patient with the right drug at whatever dose required to achieve the right exposure.