Divalproex to Divalproex Extended Release Conversion

Sandeep Dutta; Ronald C. Reed


Clin Drug Invest. 2004;24(9) 

In This Article

Variability in AUC, C

The variability in Cmin values was relatively greater than the variability in AUC and Cmax values ( Table II ). From first principles, a higher variability can be expected for Cmax and Cmin than for AUC, since the Cmax and Cmin values are single-point determinations, whereas AUC is a time-averaged value. Therefore, the lower variability of Cmax compared with Cmin is surprising.

The distributions in divalproex ER/divalproex AUC, Cmax and Cmin ratios (from studies 4 and 5), clinically important parameters for maintenance of efficacy and avoidance of toxicity following conversion from divalproex to divalproex ER, were assessed using cumulative frequency distribution plots. Figure 2 illustrates that divalproex ER/divalproex Cmin ratio has higher variability than AUC and Cmax ratios because the slope of the distribution curve is shallower for the divalproex ER/divalproex Cmin ratio. As AUC is generally considered to be a more relevant factor for safety and efficacy of VPA, the lower variability associated with AUC while converting from divalproex to divalproex ER provides comfort regarding the clinical robustness of the conversion to an 8-20% higher divalproex ER dose.

Cumulative frequency distributions of divalproex extended release (ER)/divalproex area under the plasma concentration-time curve (AUC), maximum plasma concentration (Cmax) and minimum plasma concentration (Cmin) ratios from studies 4 and 5, which compared divalproex daily doses with an 8-20% higher divalproex ER dose.

As the results of the statistical analysis for AUC central values and Cmin means indicate, the cumulative distributions show that the median AUC and Cmin ratios are 1.0 or a little higher following conversion to 8-20% higher divalproex ER doses. Some individual ratios are rather low (e.g. <0.50), but a comparable number are high (e.g. >1.5), with a Cmin ratio being more likely than an AUC ratio to differ substantially from 1.0. The lower or higher observed Cmin ratios might be largely due to intrasubject variability. It should be noted that the number of individuals (n = 99) assessed in this composite analysis across two studies is large compared with the typical single pharmacokinetic study. Therefore, a larger range, but not larger variability (i.e. percentage coefficient of variation), in the Cmin ratios would be expected when compared with studies with fewer participants.

In summary, the variability in Cmin values is relatively greater than the variability in AUC and Cmax values. Plasma VPA Cmin values for divalproex ER on average are equivalent to those for divalproex when the divalproex ER dose is 8-20% higher than the total daily divalproex dose, but may vary across patients after conversion. The AUC ratios are equivalent with 8-20% higher doses of divalproex ER and few (<5%) volunteers/patients have AUC ratios below 0.5 or above 1.5. These studies demonstrated that, upon immediate conversion (i.e. within 6-12 hours) from a divalproex regimen (q6h, q8h or q12h) to an 8-20% higher divalproex ER once-daily regimen, bioavailability was not a concern in a significant proportion of volunteers/patients. Therefore, if satisfactory clinical response has not been achieved, plasma VPA concentrations should be measured to determine whether they are in the recommended therapeutic range (50-100 mg/L).[2]