Discovery and Optimization of Potent, Efficacious and Selective Inhibitors Targeting EGFR Exon20 Insertion Mutations.

Herein, we report the identification and optimization of a series of potent inhibitors of EGFR Exon20 insertions with significant selectivity over wild-type EGFR. A strategically designed HTS campaign, multiple iterations of structure-based drug design (SBDD), and tactical linker replacement led to a potent and wild-type selective series of molecules and ultimately the discovery of 36. Compound 36 is a potent and selective inhibitor of EGFR Exon20 insertions and has demonstrated encouraging efficacy in NSCLC EGFR CRISPR-engineered H2073 xenografts that carry an SVD Exon20 insertion and reduced efficacy in a H2073 wild-type EGFR xenograft model compared to CLN-081 (5), indicating that 36 may have lower EGFR wild-type associated toxicity.

Identification of Novel, Selective Ataxia-Telangiectasia Mutated Kinase Inhibitors with the Ability to Penetrate the Blood-Brain Barrier: The Discovery of AZD1390.

The inhibition of ataxia-telangiectasia mutated (ATM) has been shown to chemo- and radio-sensitize human glioma cells in vitro and therefore might provide an exciting new paradigm in the treatment of glioblastoma multiforme (GBM). The effective treatment of GBM will likely require a compound with the potential to efficiently cross the blood-brain barrier (BBB). Starting from clinical candidate AZD0156, 4, we investigated the imidazoquinolin-2-one scaffold with the goal of improving likely CNS exposure in humans. Strategies aimed at reducing hydrogen bonding, basicity, and flexibility of the molecule were explored alongside modulating lipophilicity. These studies identified compound 24 (AZD1390) as an exceptionally potent and selective inhibitor of ATM with a good preclinical pharmacokinetic profile. 24 showed an absence of human transporter efflux in MDCKII-MDR1-BCRP studies (efflux ratio <2), significant BBB penetrance in nonhuman primate PET studies (Kp,uu 0.33) and was deemed suitable for development as a clinical candidate to explore the radiosensitizing effects of ATM in intracranial malignancies.

Utilizing a Dual Human Transporter MDCKII-MDR1-BCRP Cell Line to Assess Efflux at the Blood Brain Barrier.

To facilitate the design of drugs readily able to cross the blood brain barrier (BBB), a Madin-Darby canine kidney (MDCK) cell line was established that over expresses both P-glycoprotein (Pgp) and breast cancer resistance protein (BCRP), the main human efflux transporters of the BBB. Proteomics analyses indicate BCRP is expressed at a higher level than Pgp in this cell line. This cell line shows good activity for both transporters [BCRP substrate dantrolene efflux ratio (ER) 16.3 ± 0.9, Pgp substrate quinidine ER 27.5 ± 1.2], and use of selective transporter inhibitors enables an assessment of the relative contributions to overall ERs. The MDCKII-MDR1-BCRP ER negatively correlates with rat unbound brain/unbound plasma ratio, Kpuu Highly brain penetrant compounds with rat Kpuu ≥ 0.3 show ERs ≤ 2 in the MDCKII-MDR1-BCRP assay while compounds predominantly excluded from the brain, Kpuu ≤ 0.05, demonstrate ERs ≥ 20. A subset of compounds with MDCKII-MDR1-BCRP ER < 2 and rat Kpuu < 0.3 were shown to be substrates of rat Pgp using a rat transfected cell line, MDCKII-rMdr1a. These compounds also showed ERs > 2 in the human National Institutes of Health (NIH) MDCKI-MDR1 (high Pgp expression) cell line, which suggests that they are weak human Pgp substrates. Characterization of 37 drugs targeting the central nervous system in the MDCKII-MDR1-BCRP efflux assay show 36 have ERs < 2. In drug discovery, use of the MDCKII-MDR1-BCRP in parallel with the NIH MDCKI-MDR1 cell line is useful for identification of compounds with high brain penetration. SIGNIFICANCE STATEMENT: A single cell line that includes both the major human efflux transporters of the blood brain barrier (MDCKII-MDR1-BCRP) has been established facilitating the rapid identification of efflux substrates and enabling the design of brain penetrant molecules. Efflux ratios using this cell line demonstrate a clear relationship with brain penetration as defined by rat brain Kpuu.

Brain metastatic outgrowth and osimertinib resistance are potentiated by RhoA in EGFR-mutant lung cancer.

The brain is a major sanctuary site for metastatic cancer cells that evade systemic therapies. Through pre-clinical pharmacological, biological, and molecular studies, we characterize the functional link between drug resistance and central nervous system (CNS) relapse in Epidermal Growth Factor Receptor- (EGFR-) mutant non-small cell lung cancer, which can progress in the brain when treated with the CNS-penetrant EGFR inhibitor osimertinib. Despite widespread osimertinib distribution in vivo, the brain microvascular tumor microenvironment (TME) is associated with the persistence of malignant cell sub-populations, which are poised to proliferate in the brain as osimertinib-resistant lesions over time. Cellular and molecular features of this poised state are regulated through a Ras homolog family member A (RhoA) and Serum Responsive Factor (SRF) gene expression program. RhoA potentiates the outgrowth of disseminated tumor cells on osimertinib treatment, preferentially in response to extracellular laminin and in the brain. Thus, we identify pre-existing and adaptive features of metastatic and drug-resistant cancer cells, which are enhanced by RhoA/SRF signaling and the brain TME during the evolution of osimertinib-resistant disease.

Unbound Brain-to-Plasma Partition Coefficient, Kp,uu,brain-a Game Changing Parameter for CNS Drug Discovery and Development.

PURPOSE: More than 15 years have passed since the first description of the unbound brain-to-plasma partition coefficient (Kp,uu,brain) by Prof. Margareta Hammarlund-Udenaes, which was enabled by advancements in experimental methodologies including cerebral microdialysis. Since then, growing knowledge and data continue to support the notion that the unbound (free) concentration of a drug at the site of action, such as the brain, is the driving force for pharmacological responses. Towards this end, Kp,uu,brain is the key parameter to obtain unbound brain concentrations from unbound plasma concentrations. METHODS: To understand the importance and impact of the Kp,uu,brain concept in contemporary drug discovery and development, a survey has been conducted amongst major pharmaceutical companies based in Europe and the USA. Here, we present the results from this survey which consisted of 47 questions addressing: 1) Background information of the companies, 2) Implementation, 3) Application areas, 4) Methodology, 5) Impact and 6) Future perspectives. RESULTS AND CONCLUSIONS: From the responses, it is clear that the majority of the companies (93%) has established a common understanding across disciplines of the concept and utility of Kp,uu,brain as compared to other parameters related to brain exposure. Adoption of the Kp,uu,brain concept has been mainly driven by individual scientists advocating its application in the various companies rather than by a top-down approach. Remarkably, 79% of all responders describe the portfolio impact of Kp,uu,brain implementation in their companies as 'game-changing'. Although most companies (74%) consider the current toolbox for Kp,uu,brain assessment and its validation satisfactory for drug discovery and early development, areas of improvement and future research to better understand human brain pharmacokinetics/pharmacodynamics translation have been identified.

Potent and Selective Inhibitors of the Epidermal Growth Factor Receptor to Overcome C797S-Mediated Resistance.

The epidermal growth factor receptor (EGFR) harboring activating mutations is a clinically validated target in non-small-cell lung cancer, and a number of inhibitors of the EGFR tyrosine kinase domain, including osimertinib, have been approved for clinical use. Resistance to these therapies has emerged due to a variety of molecular events including the C797S mutation which renders third-generation C797-targeting covalent EGFR inhibitors considerably less potent against the target due to the loss of the key covalent-bond-forming residue. We describe the medicinal chemistry optimization of a biochemically potent but modestly cell-active, reversible EGFR inhibitor starting point with sub-optimal physicochemical properties. These studies culminated in the identification of compound 12 that showed improved cell potency, oral exposure, and in vivo activity in clinically relevant EGFR-mutant-driven disease models, including an Exon19 deletion/T790M/C797S triple-mutant mouse xenograft model.

Preclinical Comparison of the Blood-brain barrier Permeability of Osimertinib with Other EGFR TKIs.

PURPOSE: Osimertinib is a potent and selective EGFR tyrosine kinase inhibitor (EGFR-TKI) of both sensitizing and T790M resistance mutations. To treat metastatic brain disease, blood-brain barrier (BBB) permeability is considered desirable for increasing clinical efficacy. EXPERIMENTAL DESIGN: We examined the level of brain penetration for 16 irreversible and reversible EGFR-TKIs using multiple in vitro and in vivo BBB preclinical models. RESULTS: In vitro osimertinib was the weakest substrate for human BBB efflux transporters (efflux ratio 3.2). In vivo rat free brain to free plasma ratios (Kpuu) show osimertinib has the most BBB penetrance (0.21), compared with the other TKIs (Kpuu ≤ 0.12). PET imaging in Cynomolgus macaques demonstrated osimertinib was the only TKI among those tested to achieve significant brain penetrance (C max %ID 1.5, brain/blood Kp 2.6). Desorption electrospray ionization mass spectroscopy images of brains from mouse PC9 macrometastases models showed osimertinib readily distributes across both healthy brain and tumor tissue. Comparison of osimertinib with the poorly BBB penetrant afatinib in a mouse PC9 model of subclinical brain metastases showed only osimertinib has a significant effect on rate of brain tumor growth. CONCLUSIONS: These preclinical studies indicate that osimertinib can achieve significant exposure in the brain compared with the other EGFR-TKIs tested and supports the ongoing clinical evaluation of osimertinib for the treatment of EGFR-mutant brain metastasis. This work also demonstrates the link between low in vitro transporter efflux ratios and increased brain penetrance in vivo supporting the use of in vitro transporter assays as an early screen in drug discovery.

Further Considerations Towards an Effective and Efficient Oncology Drug Discovery DMPK Strategy.

BACKGROUND: DMPK data and knowledge are critical in maximising the probability of developing successful drugs via the application of in silico, in vitro and in vivo approaches in drug discovery. METHODS: The evaluation, optimisation and prediction of human pharmacokinetics is now a mainstay within drug discovery. These elements are at the heart of the 'right tissue' component of AstraZeneca's '5Rs framework' which, since its adoption, has resulted in increased success of Phase III clinical trials. With the plethora of DMPK related assays and models available, there is a need to continually refine and improve the effectiveness and efficiency of approaches best to facilitate the progression of quality compounds for human clinical testing. RESULTS: This article builds on previously published strategies from our laboratories, highlighting recent discoveries and successes, that brings our AstraZeneca Oncology DMPK strategy up to date. We review the core aspects of DMPK in Oncology drug discovery and highlight data recently generated in our laboratories that have influenced our screening cascade and experimental design. We present data and our experiences of employing cassette animal PK, as well as re-evaluating in vitro assay design for metabolic stability assessments and expanding our use of freshly excised animal and human tissue to best inform first time in human dosing and dose escalation studies. CONCLUSION: Application of our updated drug-drug interaction and central nervous system drug exposure strategies are exemplified, as is the impact of physiologically based pharmacokinetic and pharmacokinetic-pharmacodynamic modelling for human predictions.

Building on the success of osimertinib: achieving CNS exposure in oncology drug discovery.

Due to the blood-brain barrier (BBB) limiting the exposure of therapeutics to the central nervous system (CNS), patients with brain malignancies are challenging to treat, typically have poor prognoses, and represent a significant unmet medical need. Preclinical data report osimertinib to have significant BBB penetration and emerging clinical data demonstrate encouraging activity against CNS malignancies. Here, we discuss the oncology drug candidates AZD3759 and AZD1390 as case examples of discovery projects designing in BBB penetrance. We demonstrate how these innovative kinase inhibitors were recognized as brain penetrant and outline our view of experimental approaches and strategies that can facilitate the discovery of new brain-penetrant therapies for the treatment of primary and secondary CNS malignancies as well as other CNS disorders.

Utilizing microphysiological systems and induced pluripotent stem cells for disease modeling: a case study for blood brain barrier research in a pharmaceutical setting.

Microphysiological systems (MPS) may be able to provide the pharmaceutical industry models that can reflect human physiological responses to improve drug discovery and translational outcomes. With lack of efficacy being the primary cause for drug attrition, developing MPS disease models would help researchers identify novel targets, study mechanisms in more physiologically-relevant depth, screen for novel biomarkers and test/optimize various therapeutics (small molecules, nanoparticles and biologics). Furthermore, with advances in inducible pluripotent stem cell technology (iPSC), pharmaceutical companies can access cells from patients to help recreate specific disease phenotypes in MPS platforms. Combining iPSC and MPS technologies will contribute to our understanding of the complexities of neurodegenerative diseases and of the blood brain barrier (BBB) leading to development of enhanced therapeutics.

Discovery of a Series of 3-Cinnoline Carboxamides as Orally Bioavailable, Highly Potent, and Selective ATM Inhibitors.

We report the discovery of a novel series of 3-cinnoline carboxamides as highly potent and selective ataxia telangiectasia mutated (ATM) kinase inhibitors. Optimization of this series focusing on potency and physicochemical properties (especially permeability) led to the identification of compound 21, a highly potent ATM inhibitor (ATM cell IC50 0.0028 µM) with excellent kinase selectivity and favorable physicochemical and pharmacokinetics properties. In vivo, 21 in combination with irinotecan showed tumor regression in the SW620 colorectal tumor xenograft model, superior inhibition to irinotecan alone. Compound 21 was selected for preclinical evaluation alongside AZD0156.

The brain-penetrant clinical ATM inhibitor AZD1390 radiosensitizes and improves survival of preclinical brain tumor models.

Poor survival rates of patients with tumors arising from or disseminating into the brain are attributed to an inability to excise all tumor tissue (if operable), a lack of blood-brain barrier (BBB) penetration of chemotherapies/targeted agents, and an intrinsic tumor radio-/chemo-resistance. Ataxia-telangiectasia mutated (ATM) protein orchestrates the cellular DNA damage response (DDR) to cytotoxic DNA double-strand breaks induced by ionizing radiation (IR). ATM genetic ablation or pharmacological inhibition results in tumor cell hypersensitivity to IR. We report the primary pharmacology of the clinical-grade, exquisitely potent (cell IC50, 0.78 nM), highly selective [>10,000-fold over kinases within the same phosphatidylinositol 3-kinase-related kinase (PIKK) family], orally bioavailable ATM inhibitor AZD1390 specifically optimized for BBB penetration confirmed in cynomolgus monkey brain positron emission tomography (PET) imaging of microdosed 11C-labeled AZD1390 (Kp,uu, 0.33). AZD1390 blocks ATM-dependent DDR pathway activity and combines with radiation to induce G2 cell cycle phase accumulation, micronuclei, and apoptosis. AZD1390 radiosensitizes glioma and lung cancer cell lines, with p53 mutant glioma cells generally being more radiosensitized than wild type. In in vivo syngeneic and patient-derived glioma as well as orthotopic lung-brain metastatic models, AZD1390 dosed in combination with daily fractions of IR (whole-brain or stereotactic radiotherapy) significantly induced tumor regressions and increased animal survival compared to IR treatment alone. We established a pharmacokinetic-pharmacodynamic-efficacy relationship by correlating free brain concentrations, tumor phospho-ATM/phospho-Rad50 inhibition, apoptotic biomarker (cleaved caspase-3) induction, tumor regression, and survival. On the basis of the data presented here, AZD1390 is now in early clinical development for use as a radiosensitizer in central nervous system malignancies.

An Investigation into the Prediction of the Plasma Concentration-Time Profile and Its Interindividual Variability for a Range of Flavin-Containing Monooxygenase Substrates Using a Physiologically Based Pharmacokinetic Modeling Approach.

Our recent paper demonstrated the ability to predict in vivo clearance of flavin-containing monooxygenase (FMO) drug substrates using in vitro human hepatocyte and human liver microsomal intrinsic clearance with standard scaling approaches. In this paper, we apply a physiologically based pharmacokinetic (PBPK) modeling and simulation approach (M&S) to predict the clearance, area under the curve (AUC), and Cmax values together with the plasma profile of a range of drugs from the original study. The human physiologic parameters for FMO, such as enzyme abundance in liver, kidney, and gut, were derived from in vitro data and clinical pharmacogenetics studies. The drugs investigated include itopride, benzydamine, tozasertib, tamoxifen, moclobemide, imipramine, clozapine, ranitidine, and olanzapine. The fraction metabolized by FMO for these drugs ranged from 21% to 96%. The developed PBPK models were verified with data from multiple clinical studies. An attempt was made to estimate the scaling factor for recombinant FMO (rFMO) using a parameter estimation approach and automated sensitivity analysis within the PBPK platform. Simulated oral clearance using in vitro hepatocyte data and associated extrahepatic FMO data predicts the observed in vivo plasma concentration profile reasonably well and predicts the AUC for all of the FMO substrates within 2-fold of the observed clinical data; seven of the nine compounds fell within 2-fold when human liver microsomal data were used. rFMO overpredicted the AUC by approximately 2.5-fold for three of the nine compounds. Applying a calculated intersystem extrapolation scalar or tissue-specific scalar for the rFMO data resulted in better prediction of clinical data. The PBPK M&S results from this study demonstrate that human hepatocytes and human liver microsomes can be used along with our standard scaling approaches to predict human in vivo pharmacokinetic parameters for FMO substrates.

Orally Bioavailable and Blood-Brain Barrier-Penetrating ATM Inhibitor (AZ32) Radiosensitizes Intracranial Gliomas in Mice.

Inhibition of ataxia-telangiectasia mutated (ATM) during radiotherapy of glioblastoma multiforme (GBM) may improve tumor control by short-circuiting the response to radiation-induced DNA damage. A major impediment for clinical implementation is that current inhibitors have limited central nervous system (CNS) bioavailability; thus, the goal was to identify ATM inhibitors (ATMi) with improved CNS penetration. Drug screens and refinement of lead compounds identified AZ31 and AZ32. The compounds were then tested in vivo for efficacy and impact on tumor and healthy brain. Both AZ31 and AZ32 blocked the DNA damage response and radiosensitized GBM cells in vitro AZ32, with enhanced blood-brain barrier (BBB) penetration, was highly efficient in vivo as radiosensitizer in syngeneic and human, orthotopic mouse glioma model compared with AZ31. Furthermore, human glioma cell lines expressing mutant p53 or having checkpoint-defective mutations were particularly sensitive to ATMi radiosensitization. The mechanism for this p53 effect involves a propensity to undergo mitotic catastrophe relative to cells with wild-type p53. In vivo, apoptosis was >6-fold higher in tumor relative to healthy brain after exposure to AZ32 and low-dose radiation. AZ32 is the first ATMi with oral bioavailability shown to radiosensitize glioma and improve survival in orthotopic mouse models. These findings support the development of a clinical-grade, BBB-penetrating ATMi for the treatment of GBM. Importantly, because many GBMs have defective p53 signaling, the use of an ATMi concurrent with standard radiotherapy is expected to be cancer-specific, increase the therapeutic ratio, and maintain full therapeutic effect at lower radiation doses. Mol Cancer Ther; 17(8); 1637-47. ©2018 AACR.

The Identification of Potent, Selective, and Orally Available Inhibitors of Ataxia Telangiectasia Mutated (ATM) Kinase: The Discovery of AZD0156 (8-{6-[3-(Dimethylamino)propoxy]pyridin-3-yl}-3-methyl-1-(tetrahydro-2 H-pyran-4-yl)-1,3-dihydro-2 H-imidazo[4,5- c]quinolin-2-one).

ATM inhibitors, such as 7, have demonstrated the antitumor potential of ATM inhibition when combined with DNA double-strand break-inducing agents in mouse xenograft models. However, the properties of 7 result in a relatively high predicted clinically efficacious dose. In an attempt to minimize attrition during clinical development, we sought to identify ATM inhibitors with a low predicted clinical dose (<50 mg) and focused on strategies to increase both ATM potency and predicted human pharmacokinetic half-life (predominantly through the increase of volume of distribution). These efforts resulted in the discovery of 64 (AZD0156), an exceptionally potent and selective inhibitor of ATM based on an imidazo[4,5- c]quinolin-2-one core. 64 has good preclinical phamacokinetics, a low predicted clinical dose, and a high maximum absorbable dose. 64 has been shown to potentiate the efficacy of the approved drugs irinotecan and olaparib in disease relevant mouse models and is currently undergoing clinical evaluation with these agents.

Expanding the Armory: Predicting and Tuning Covalent Warhead Reactivity.

Targeted covalent inhibition is an established approach for increasing the potency and selectivity of potential drug candidates, as well as identifying potent and selective tool compounds for target validation studies. It is evident that identification of reversible recognition elements is essential for selective covalent inhibition, but this must also be achieved with the appropriate level of inherent reactivity of the reactive functionality (or "warhead"). Structural changes that increase or decrease warhead reactivity, guided by methods to predict the effect of those changes, have the potential to tune warhead reactivity and negate issues related to potency and/or toxicity. The half-life to adduct formation with glutathione (GSH t1/2) is a useful assay for measuring the reactivity of cysteine-targeting covalent warheads but is limited to synthesized molecules. In this manuscript we assess the ability of several experimental and computational approaches to predict GSH t1/2 for a range of cysteine targeting warheads, including a novel method based on pKa. Furthermore, matched molecular pairs analysis has been performed against our internal compound collection, revealing structure-activity relationships between a selection of different covalent warheads. These observations and methods of prediction will be valuable in the design of new covalent inhibitors with desired levels of reactivity.

An Investigation into the Prediction of in Vivo Clearance for a Range of Flavin-containing Monooxygenase Substrates.

Flavin-containing monooxygenases (FMO) are metabolic enzymes mediating the oxygenation of nucleophilic atoms such as nitrogen, sulfur, phosphorus, and selenium. These enzymes share similar properties to the cytochrome P450 system but can be differentiated through heat inactivation and selective substrate inhibition by methimazole. This study investigated 10 compounds with varying degrees of FMO involvement to determine the nature of the correlation between human in vitro and in vivo unbound intrinsic clearance. To confirm and quantify the extent of FMO involvement six of the compounds were investigated in human liver microsomal (HLM) in vitro assays using heat inactivation and methimazole substrate inhibition. Under these conditions FMO contribution varied from 21% (imipramine) to 96% (itopride). Human hepatocyte and HLM intrinsic clearance (CLint) data were scaled using standard methods to determine the predicted unbound intrinsic clearance (predicted CLint u) for each compound. This was compared with observed unbound intrinsic clearance (observed CLint u) values back calculated from human pharmacokinetic studies. A good correlation was observed between the predicted and observed CLint u using hepatocytes (R2 = 0.69), with 8 of the 10 compounds investigated within or close to a factor of 2. For HLM the in vitro-in vivo correlation was maintained (R2 = 0.84) but the accuracy was reduced with only 3 out of 10 compounds falling within, or close to, twofold. This study demonstrates that human hepatocytes and HLM can be used with standard scaling approaches to predict the human in vivo clearance for FMO substrates.

Utilization of Structure-Based Design to Identify Novel, Irreversible Inhibitors of EGFR Harboring the T790M Mutation.

A novel series of covalent inhibitors of EGFR (epidermal growth factor receptor) kinase was discovered through a combination of subset screening and structure-based design. These compounds preferentially inhibit mutant forms of EGFR (activating mutant and T790M mutant) over wild-type EGFR in cellular assays measuring EGFR autophosphorylation and proliferation, suggesting an improved therapeutic index in non-small cell lung cancer patients would be achievable relative to established EGFR inhibitors. We describe our design approaches, resulting in the identification of the lead compound 5, and our efforts to develop an understanding of the structure-activity relationships within this series. In addition, strategies to overcome challenges around metabolic stability and aqueous solubility are discussed. Despite limitations in its physical properties, 5 is orally bioavailable in mice and demonstrates pronounced antitumor activity in in vivo models of mutant EGFR-driven cancers.

Interpreting physicochemical experimental data sets.

With the wealth of experimental physicochemical data available to chemoinformaticians from the literature, commercial, and company databases an increasing challenge is the interpretation of such datasets. Subtle differences in experimental methodology used to generate these datasets can give rise to variations in physicochemical property values. Such methodology nuances will be apparent to an expert experimentalist but not necessarily to the data analyst and modeller. This paper describes the differences between common methodologies for measuring the four most important physicochemical properties namely aqueous solubility, octan-1-ol/water distribution coefficient, pK(a) and plasma protein binding highlighting key factors that can lead to systematic differences. Insight is given into how to identify datasets suitable for combining.

Using quantitative structure activity relationship models to predict an appropriate solvent system from a common solvent system family for countercurrent chromatography separation.

Countercurrent chromatography (CCC) is a form of liquid-liquid chromatography. It works by running one immiscible solvent (mobile phase) over another solvent (stationary phase) being held in a CCC column using centrifugal force. The concentration of compound in each phase is characterised by the partition coefficient (Kd), which is the concentration in the stationary phase divided by the concentration in the mobile phase. When Kd is between approximately 0.2 and 2, it is most likely that optimal separation will be achieved. Having the Kd in this range allows the compound enough time in the column to be separated without resulting in a broad peak and long run time. In this paper we report the development of quantitative structure activity relationship (QSAR) models to predict logKd. The QSAR models use only the molecule's 2D structure to predict the molecular property logKd.