The Comparison of Machine Learning and Mechanistic In Vitro-In Vivo Extrapolation Models for the Prediction of Human Intrinsic Clearance.

Accurate prediction of human pharmacokinetics (PK) remains one of the key objectives of drug metabolism and PK (DMPK) scientists in drug discovery projects. This is typically performed by using in vitro-in vivo extrapolation (IVIVE) based on mechanistic PK models. In recent years, machine learning (ML), with its ability to harness patterns from previous outcomes to predict future events, has gained increased popularity in application to absorption, distribution, metabolism, and excretion (ADME) sciences. This study compares the performance of various ML and mechanistic models for the prediction of human IV clearance for a large (645) set of diverse compounds with literature human IV PK data, as well as measured relevant in vitro end points. ML models were built using multiple approaches for the descriptors: (1) calculated physical properties and structural descriptors based on chemical structure alone (classical QSAR/QSPR); (2) in vitro measured inputs only with no structure-based descriptors (ML IVIVE); and (3) in silico ML IVIVE using in silico model predictions for the in vitro inputs. For the mechanistic models, well-stirred and parallel-tube liver models were considered with and without the use of empirical scaling factors and with and without renal clearance. The best ML model for the prediction of in vivo human intrinsic clearance (CLint) was an in vitro ML IVIVE model using only six in vitro inputs with an average absolute fold error (AAFE) of 2.5. The best mechanistic model used the parallel-tube liver model, with empirical scaling factors resulting in an AAFE of 2.8. The corresponding mechanistic model with full in silico inputs achieved an AAFE of 3.3. These relative performances of the models were confirmed with the prediction of 16 Pfizer drug candidates that were not part of the original data set. Results show that ML IVIVE models are comparable to or superior to their best mechanistic counterparts. We also show that ML IVIVE models can be used to derive insights into factors for the improvement of mechanistic PK prediction.

Identification of Organic Anion Transporter 2 Inhibitors: Screening, Structure-Based Analysis, and Clinical Drug Interaction Risk Assessment.

Organic anion transporter 2 (OAT2 or SLC22A7) plays an important role in the hepatic uptake and renal secretion of several endogenous compounds and drugs. The goal of this work is to understand the structure activity of OAT2 inhibition and assess clinical drug interaction risk. A single-point inhibition assay using OAT2-transfected HEK293 cells was employed to screen about 150 compounds; and concentration-dependent inhibition potency (IC50) was measured for the identified "inhibitors". Acids represented about 65% of all inhibitors, and the frequency of bases-plus-zwitterions approximately doubled for "non-inhibitors". Interestingly, 9 of 10 most potent inhibitors (low IC50) are acids (pKa ∼ 3-5). Additionally, inhibitors are significantly larger and lipophilic than non-inhibitors. In silico (binary) models were developed to identify inhibitors and non-inhibitors. Finally, in vivo risk assessed via static drug-drug interaction models identified several inhibitors with potential for renal and hepatic OAT2 inhibition at clinical doses. This is the first study assessing the global pattern of OAT2-ligand interactions.

Evaluation of Prediction Accuracy for Volume of Distribution in Rat and Human Using In Vitro, In Vivo, PBPK and QSAR Methods.

Volume of distribution at steady state (Vss) is an important pharmacokinetic parameter of a drug candidate. In this study, Vss prediction accuracy was evaluated by using: (1) seven methods for rat with 56 compounds, (2) four methods for human with 1276 compounds, and (3) four in vivo methods and three Kp (partition coefficient) scalar methods from scaling of three preclinical species with 125 compounds. The results showed that the global QSAR models outperformed the PBPK methods. Tissue fraction unbound (fu,t) method with adipose and muscle also provided high Vss prediction accuracy. Overall, the high performing methods for human Vss prediction are the global QSAR models, Øie-Tozer and equivalency methods from scaling of preclinical species, as well as PBPK methods with Kp scalar from preclinical species. Certain input parameter ranges rendered PBPK models inaccurate due to mass balance issues. These were addressed using appropriate theoretical limit checks. Prediction accuracy of tissue Kp were also examined. The fu,t method predicted Kp values more accurately than the PBPK methods for adipose, heart and muscle. All the methods overpredicted brain Kp and underpredicted liver Kp due to transporter effects. Successful Vss prediction involves strategic integration of in silico, in vitro and in vivo approaches.

Transfer of multivariate calibrations between four near-infrared spectrometers using orthogonal signal correction.

The transfer of partial least squares (PLS) calibration models among four near-infrared spectrometers was investigated for the quantitative analysis of thermoset resin polymers. A comparative study of second derivatives, multiplicative scatter correction, finite impulse response filtering, slope and bias correction, model updating (MU), and orthogonal signal correction (OSC) was conducted to determine which processing methods achieved model transferability. It is shown that OSC and MU were superior to the other calibration transfer methods, leading to very robust PLS models with enhanced predictive ability. It is also shown that the transfer results obtained with OSC were not significantly different from those obtained with model updating.