No abstract

A computational approach is described that can predict the VD(ss) of new compounds in humans, with an accuracy of within 2-fold of the actual value. A dataset of VD values for 384 drugs in humans was used to train a hybrid mixture discriminant analysis-random forest (MDA-RF) model using 31 computed descriptors. Descriptors included terms describing lipophilicity, ionization, molecular volume, and various molecular fragments. For a test set of 23 proprietary compounds not used in model construction, the geometric mean fold-error (GMFE) was 1.78-fold (+/-11.4%). The model was also tested using a leave-class out approach wherein subsets of drugs based on therapeutic class were removed from the training set of 384, the model was recast, and the VD(ss) values for each of the subsets were predicted. GMFE values ranged from 1.46 to 2.94-fold, depending on the subset. Finally, for an additional set of 74 compounds, VD(ss) predictions made using the computational model were compared to predictions made using previously described methods dependent on animal pharmacokinetic data. Computational VD(ss) predictions were, on average, 2.13-fold different from the VD(ss) predictions from animal data. The computational model described can predict human VD(ss) with an accuracy comparable to predictions requiring substantially greater effort and can be applied in place of animal experimentation.

Possible functional differences between P-glycoproteins (P-gps) encoded by the human MDR1 and mouse mdr1 and mdr3 genes with respect to drug resistance profiles and sensitivity to known modulators have been investigated. For this, the three genes were introduced and overexpressed in the same cellular background, that of Chinese hamster LR73 ovary cells, and drug-resistant clones expressing comparable amounts of the corresponding P-gps were selected under the same conditions. Analysis of the specific drug resistance profiles encoded by each P-gp for colchicine, adriamycin, vinblastine, and actinomycin D revealed overlapping but distinct patterns of drug resistance for the three isoforms. While all three P-gps conferred levels of resistance to vinblastine that did not vary by more than 2.5-fold, each isoform could be clearly distinguished by its capacity to confer resistance to colchicine and actinomycin D. Likewise, the study of structurally related and unrelated P-gp modulators indicated strong isoform-specific differences in the capacity of individual modulators to abrogate vinblastine resistance in the corresponding mdr transfectants. The study of several disubstituted piperazine analogs indicated that minor chemical modifications of the linker region of this modulator had strong effects on the sensitivity profile of each isoform to the modulator. Together, these results indicate that the three P-gp isoforms analyzed have specific and distinguishable functional characteristics with respect to interactions with drugs and modulators. These findings also suggest that P-gp positive murine transplantable tumors should be used with caution in the design and in vivo testing of novel P-gp modulators to be used to reverse multidrug resistance to tumor cells expressing human MDR1.

No abstract

No abstract