Artificial Intelligence for Quantitative Modeling in Drug Discovery and Development: An Innovation and Quality Consortium Perspective on Use Cases and Best Practices.

Recent breakthroughs in artificial intelligence (AI) and machine learning (ML) have ushered in a new era of possibilities across various scientific domains. One area where these advancements hold significant promise is model-informed drug discovery and development (MID3). To foster a wider adoption and acceptance of these advanced algorithms, the Innovation and Quality (IQ) Consortium initiated the AI/ML working group in 2021 with the aim of promoting their acceptance among the broader scientific community as well as by regulatory agencies. By drawing insights from workshops organized by the working group and attended by key stakeholders across the biopharma industry, academia, and regulatory agencies, this white paper provides a perspective from the IQ Consortium. The range of applications covered in this white paper encompass the following thematic topics: (i) AI/ML-enabled Analytics for Pharmacometrics and Quantitative Systems Pharmacology (QSP) Workflows; (ii) Explainable Artificial Intelligence and its Applications in Disease Progression Modeling; (iii) Natural Language Processing (NLP) in Quantitative Pharmacology Modeling; and (iv) AI/ML Utilization in Drug Discovery. Additionally, the paper offers a set of best practices to ensure an effective and responsible use of AI, including considering the context of use, explainability and generalizability of models, and having human-in-the-loop. We believe that embracing the transformative power of AI in quantitative modeling while adopting a set of good practices can unlock new opportunities for innovation, increase efficiency, and ultimately bring benefits to patients.

Meta-Analysis of Noncompartmental Pharmacokinetic Parameters of Ertugliflozin to Evaluate Dose Proportionality and UGT1A9 Polymorphism Effect on Exposure.

Ertugliflozin, a sodium-glucose cotransporter 2 inhibitor, is primarily metabolized via glucuronidation by the uridine 5'-diphospho-glucuronosyltransferase (UGT) isoform UGT1A9. This noncompartmental meta-analysis of ertugliflozin pharmacokinetics evaluated the relationship between ertugliflozin exposure and dose, and the effect of UGT1A9 genotype on ertugliflozin exposure. Pharmacokinetic data from 25 phase 1 studies were pooled. Structural models for dose proportionality described the relationship between ertugliflozin area under the plasma concentration-time curve (AUC) or maximum observed plasma concentration (Cmax ) and dose. A structural model for the UGT1A9 genotype described the relationship between ertugliflozin AUC and dose, with genotype information on 3 UGT1A9 polymorphisms (UGT1A9-2152, UGT1A9*3, UGT1A9*1b) evaluated as covariates from the full model. Ertugliflozin AUC and Cmax increased in a dose-proportional manner over the dose range of 0.5-300 mg, and population-predicted AUC and Cmax values for the 5- and 15-mg ertugliflozin tablets administered in the fasted state demonstrated good agreement with the observed data. The largest change in ertugliflozin AUC was in subjects carrying the UGT1A9*3 heterozygous variant, with population-predicted AUC (90% confidence interval) values of 485 ng·h/mL (458 to 510 ng·h/mL) and 1560 ng·h/mL (1480 to 1630 ng·h/mL) for ertugliflozin 5 and 15 mg, respectively, compared with 436 ng·h/mL (418 to 455 ng·h/mL) and 1410 ng·h/mL (1350 to 1480 ng·h/mL), respectively, in wild-type subjects. Overall, the mean effects of the selected UGT1A9 variants on ertugliflozin AUC were within ±10% of the wild type. UGT1A9 genotype did not have any clinically meaningful effects on ertugliflozin exposure in healthy subjects. No ertugliflozin dose adjustment would be required in patients with the UGT1A9 variants assessed in this study.

PharmTeX: a LaTeX-Based Open-Source Platform for Automated Reporting Workflow.

Every year, the pharmaceutical industry generates a large number of scientific reports related to drug research, development, and regulatory submissions. Many of these reports are created using text processing tools such as Microsoft Word. Given the large number of figures, tables, references, and other elements, this is often a tedious task involving hours of copying and pasting and substantial efforts in quality control (QC). In the present article, we present the LaTeX-based open-source reporting platform, PharmTeX, a community-based effort to make reporting simple, reproducible, and user-friendly. The PharmTeX creators put a substantial effort into simplifying the sometimes complex elements of LaTeX into user-friendly functions that rely on advanced LaTeX and Perl code running in the background. Using this setup makes LaTeX much more accessible for users with no prior LaTeX experience. A software collection was compiled for users not wanting to manually install the required software components. The PharmTeX templates allow for inclusion of tables directly from mathematical software output as well and figures from several formats. Code listings can be included directly from source. No previous experience and only a few hours of training are required to start writing reports using PharmTeX. PharmTeX significantly reduces the time required for creating a scientific report fully compliant with regulatory and industry expectations. QC is made much simpler, since there is a direct link between analysis output and report input. PharmTeX makes available to report authors the strengths of LaTeX document processing without the need for extensive training. Graphical Abstract ᅟ.

Systematic review and meta-analysis of serious infections with tofacitinib and biologic disease-modifying antirheumatic drug treatment in rheumatoid arthritis clinical trials.

BACKGROUND: Tofacitinib is an oral Janus kinase inhibitor for the treatment of rheumatoid arthritis (RA). Tofacitinib modulates the signaling of cytokines that are integral to lymphocyte activation, proliferation, and function. Thus, tofacitinib therapy may result in suppression of multiple elements of the immune response. Serious infections have been reported in tofacitinib RA trials. However, limited head-to-head comparator data were available within the tofacitinib RA development program to directly compare rates of serious infections with tofacitinib relative to biologic agents, and specifically adalimumab (employed as an active control agent in two randomized controlled trials of tofacitinib). METHODS: A systematic literature search of data from interventional randomized controlled trials and long-term extension studies with biologics in RA was carried out. Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) consensus was followed for reporting results of the review and meta-analysis. Incidence rates (unique patients with events/100 patient-years) for each therapy were estimated based on data from randomized controlled trials and long-term extension studies using a random-effects model. Relative and absolute risk comparisons versus placebo used Mantel-Haenszel methods. RESULTS: The search produced 657 hits. In total, 66 randomized controlled trials and 22 long-term extension studies met the selection criteria. Estimated incidence rates (95% confidence intervals [CIs]) for abatacept, rituximab, tocilizumab, and tumor necrosis factor inhibitors were 3.04 (2.49, 3.72), 3.72 (2.99, 4.62), 5.45 (4.26, 6.96), and 4.90 (4.41, 5.44), respectively. Incidence rates (95% CIs) for tofacitinib 5 and 10 mg twice daily (BID) in phase 3 trials were 3.02 (2.25, 4.05) and 3.00 (2.24, 4.02), respectively. Corresponding incidence rates in long-term extension studies were 2.50 (2.05, 3.04) and 3.19 (2.74, 3.72). The risk ratios (95% CIs) versus placebo for tofacitinib 5 and 10 mg BID were 2.21 (0.60, 8.14) and 2.02 (0.56, 7.28), respectively. Risk differences (95% CIs) versus placebo for tofacitinib 5 and 10 mg BID were 0.38% (-0.24%, 0.99%) and 0.40% (-0.22%, 1.02%), respectively. CONCLUSIONS: In interventional studies, the risk of serious infections with tofacitinib is comparable to published rates for biologic disease-modifying antirheumatic drugs in patients with moderate to severely active RA.

Population pharmacokinetics of azithromycin and chloroquine in healthy adults and paediatric malaria subjects following oral administration of fixed-dose azithromycin and chloroquine combination tablets.

BACKGROUND: Population pharmacokinetics (PK) of azithromycin (AZ) and chloroquine (CQ) following administration of fixed-dose combination tablet formulations of AZ and CQ (AZCQ) was evaluated using data from two studies: 1) in children with symptomatic uncomplicated falciparum malaria in sub-Saharan Africa; and 2) in healthy adults in the United States. METHODS: Study 1 included paediatric subjects randomized to either AZCQ or artemether-lumefantrine treatment in Cohort 1 (age 5-12 years) and Cohort 2 (age 6-59 months). Dosing of AZCQ was approximately 30 mg/kg AZ and 10 mg/kg CQ once daily for 3 days (for ≥20 kg weight: AZ/CQ 300/100 mg per tablet; 5 to <20 kg weight: AZ/CQ 150/50 mg per tablet). Study 2 included adults randomized to receive either two AZCQ tablets (AZ/CQ 250/155 mg per tablet) or individual commercial tablets of AZ 500 mg and CQ 300 mg. Serum AZ and plasma CQ concentrations from both studies were pooled. Population PK models were constructed using standard approaches to evaluate the concentration-time data for AZ and CQ and to identify any covariates predictive of PK behaviour. RESULTS: A three-compartment PK model with linear clearance and absorption adequately described AZ data, while a two-compartment model with linear clearance and absorption and an absorption lag adequately described CQ data. No overall bias or substantial model misspecification was evident using diagnostic plots and visual predictive checks. Body weight as an allometric function was the only covariate in the final AZ and CQ PK models. There were significantly lower AZ (0.488 vs 0.745 [mg•h/L]/[mg/kg], p < 0.00001) and CQ (0.836 vs 1.27 [mg•h/L]/[mg/kg], p < 0.00001) exposures (AUCinf) normalized by dose (mg/kg) in children compared with the adults. CONCLUSIONS: The PK of AZ and CQ following administration of AZCQ was well described using a three- and two-compartment model, respectively. AZ and CQ exhibited linear absorption and clearance; the model for CQ included an absorption lag. Weight was predictive of exposure for both AZ and CQ. Assuming equivalent dosing (mg/kg), AZ and CQ exposure in children would be expected to be lower than that in adults, suggesting that children may require a higher dose (mg/kg) than adults to achieve the same AZ and CQ exposure.

Ziprasidone and the corrected QT interval: a comprehensive summary of clinical data.

BACKGROUND: Prolongation of the corrected QT interval (QTc) is understood to be a predictor of risk for ventricular arrhythmia; consequently, data on QTc effects of drugs are used by regulatory bodies to evaluate potential safety risks. Clinical pharmacology studies in adults receiving oral ziprasidone demonstrated a dose-dependent mean increase (4.5-19.5 milliseconds [ms]) in QTc over the range of 40-160 mg/d with a small incremental increase (22.5 ms) at 320 mg/d. In a comparative study of ziprasidone versus five antipsychotics, the mean QTc increase at steady state maximum concentration (C(max)) for ziprasidone was 15.9 ms. Accordingly, the effects of ziprasidone on QTc were studied in phase II-IV randomized controlled trials (RCTs). OBJECTIVE: The objective of this study was to provide clinicians and clinical researchers with a comprehensive analysis of QTc changes associated with ziprasidone based on data from Pfizer-sponsored phase II-IV RCTs in schizophrenia or bipolar disorder patients, safety reports and post-marketing surveillance. METHODS: The following analyses of data were conducted to obtain a comprehensive summary of QTc data on ziprasidone: (i) post hoc analyses (using primarily descriptive statistics) of pooled QTc data (Fridericia correction) from more than 40 phase II-IV adult ziprasidone RCTs organized according to the following subgroups: all monotherapy studies in schizophrenia and bipolar disorder, all intramuscular (IM) studies, adjunctive studies in bipolar disorder and fixed-dose oral studies; (ii) post hoc analyses from 36 phase II-IV adult ziprasidone RCTs exploring the relationship between QTc change from baseline and baseline QTc in adults; (iii) post hoc analyses from phase II-IV adult ziprasidone RCTs modelling QTc change as a function of ziprasidone concentration in both adult (17 studies) and paediatric (5 studies) subjects; (iv) cardiac adverse event (AE) reports from all phase II-IV adult ziprasidone RCTs in schizophrenia; (v) a large simple trial entitled Ziprasidone Observational Study of Cardiac Outcomes (ZODIAC) in 18 154 subjects with schizophrenia (the only previously reported results included here); and (vi) cardiac-related AEs presented in a ziprasidone post-marketing surveillance report created in 2007. RESULTS: A total of 4306 adults received ziprasidone in placebo- and active-comparator phase II-IV RCTs and had evaluable QTc data. One subject reached a QTc ≥480 ms; 33 (0.8%) had a QTc ≥450 ms. QTc prolongation ≥30 ms was observed in 389 subjects (9.0%); ≥60 ms in 30 (0.7%); and ≥75 ms in 12 (0.3%). In the placebo-controlled studies, mean change in QTc from baseline to end of study was 3.6 (± 20.8) ms in the ziprasidone group; the corresponding QTc change in the pooled placebo group was -0.3 (± 20.6) ms. Data from IM studies, and bipolar studies in which ziprasidone was used adjunctively with lithium, valproate or lamotrigine, demonstrated similar QTc effects. A scatter-plot of QTc prolongation against baseline QTc showed QTc prolongation ≥60 ms exclusively in adult subjects with a baseline QTc ≤400 ms. The final concentration-response analysis model, comprising 2966 data points from 1040 subjects, estimates an increase in QTc of 6 ms for each 100 ng/mL increase in ziprasidone concentration. The large simple trial (ZODIAC) failed to show that ziprasidone is associated with an elevated risk of non-suicidal mortality relative to olanzapine in real-world use. Post-marketing data over a 5-year period did not show a signal of increased cardiac AEs. CONCLUSIONS: These analyses provide the first comprehensive summary of QTc changes associated with ziprasidone based on Pfizer-sponsored phase II-IV RCTs, safety reports and post-marketing surveillance. The results of the analyses of pooled data from phase II-IV RCTs in adults demonstrate a modest mean increase in QTc, infrequent QTc prolongation ≥60 ms (<1.0%) and rare observation of QTc ≥480 ms. These data are consistent with results from ziprasidone clinical pharmacology studies, safety reports and post-marketing surveillance. Taken together, they provide the most comprehensive evidence published to date that ziprasidone appears to be safe when used as indicated in patients with schizophrenia or bipolar disorder.

Effects of high-dose ziprasidone and haloperidol on the QTc interval after intramuscular administration: a randomized, single-blind, parallel-group study in patients with schizophrenia or schizoaffective disorder.

BACKGROUND: Antipsychotic agents have been associated with a prolonged QT interval. Data on the effects of ziprasidone and haloperidol on the QTc interval are lacking. OBJECTIVE: This study aimed to characterize the effects of 2 high-dose intramuscular injections of ziprasidone and haloperidol on the QTc interval at T(max). METHODS: This randomized, single-blind study enrolled patients with schizophrenia or schizoaffective disorder in whom long-term antipsychotic therapy was indicated. Patients were randomized to receive 2 high-dose intramuscular injections of ziprasidone (20 and 30 mg) or haloperidol (7.5 and 10 mg) separated by 4 hours. The primary outcome measure was the mean change from baseline in QTc at the T(max) of each injection. Each dose administration was followed by serial ECG and blood sampling for pharmacokinetic determinations. Twelve-lead ECG data were obtained immediately before and at predetermined times after injections. ECG tracings were read by a blinded central reader. Blood samples were obtained immediately before and after injections. Point estimates and 95% CIs for mean QTc and changes from baseline in QTc were estimated. No between-group hypothesis tests were conducted. For the assessments of tolerability and safety profile, patients underwent physical examination, including measurement of vital signs, clinical laboratory evaluation, and monitoring for adverse events (AEs) using spontaneous reporting. RESULTS: A total of 59 patients were assigned to treatment, and 58 received study medication (ziprasidone, 31 patients; haloperidol, 27; age range, 21-72 years; 79% male). After the first injection, mean (95% CI) changes from baseline were 4.6 msec (0.4-8.9) with ziprasidone (n = 25) and 6.0 msec (1.4-10.5) with haloperidol (n = 24). After the second injection, these values were 12.8 msec (6.7-18.8) and 14.7 msec (10.2-19.2), respectively. Mild and transient changes in heart rate and blood pressure were observed with both treatments. None of the patients had a QTc interval >480 msec. Two patients in the ziprasidone group experienced QTc prolongation >450 msec (457 and 454 msec) and QTc changes that exceeded 60 msec (62 and 76 msec) relative to the time-matched baseline values. With haloperidol, QTc interval values were <450 msec with no changes >60 msec. Treatment-emergent AEs were reported in 29 of 31 patients (93.5%) in the ziprasidone group and 25 of 27 patients (92.6%) in the haloperidol group; most events were of mild or moderate severity. Frequently reported AEs were somnolence (90.3% and 81.5%, respectively), dizziness (22.6% and 7.4%), anxiety (16.1% and 7.4%), extrapyramidal symptoms (6.5% and 33.3%), agitation (6.5% and 18.5%), and insomnia (0% and 14.8%). CONCLUSIONS: In this study of the effects of high-dose ziprasidone and haloperidol in patients with schizophrenic disorder, none of the patients had a QTc interval >480 msec, and changes from baseline QTc interval were clinically modest with both drugs. Both drugs were generally well tolerated.

Effects of Oral Ziprasidone and Oral Haloperidol on QTc interval in patients with Schizophrenia or Schizoaffective disorder.

STUDY OBJECTIVE: To characterize the effect of oral ziprasidone and haloperidol on the corrected QT (QTc) interval under steady-state conditions. Design. Prospective, randomized, open-label, parallel-group study. SETTING: Inpatient clinical research facility. Patients Fifty-nine adults (age range 25-59 yrs) with schizophrenia or schizoaffective disorder who had no clinically significant abnormality on electrocardiogram (ECG) at screening. Intervention. During period 1 (days -10 to -4), antipsychotic and anticholinergic drugs were tapered. On the first day (day -3) of period 2, the drugs were discontinued, and placebo was given for the next 3 days (days -2 to 0). On the last day (day 0) of period 2, serial baseline ECGs were collected. During period 3 (days 1-16), patients received escalating oral doses of ziprasidone and haloperidol to reach steady state. Period 4 (days 17-19) allowed for study drug washout and initiation of outpatient antipsychotic therapy; safety assessments were also performed during this period. MEASUREMENTS AND RESULTS: At each steady-state dose level, three ECGs and a serum or plasma sample were collected at the predicted time of peak exposure to the administered drug. Point estimates and 95% confidence intervals (CIs) were determined for the mean QTc interval at baseline and for the mean change from baseline in QTc at each steady-state dose level. Mean changes from baseline in the QTc interval (msec) for ziprasidone were 4.5 (95% CI 1.9-7.1), 19.5 (95% CI 15.5-23.4), and 22.5 (95% CI 15.7- 29.4) for steady-state doses of 40, 160, and 320 mg/day, respectively; for haloperidol, -1.2 (95% CI -4.1-1.7), 6.6 (95% CI 1.6-11.7), and 7.2 (95% CI 1.4-13.1) for steady-state doses of 2.5, 15, and 30 mg/day. Although no patient in either treatment group experienced a QTc interval of 450 msec or greater, the QTc interval increased 30 msec or more in 11 and 17 ziprasidone-treated patients at 160 and 320 mg/day, respectively, and in 3 and 5 haloperidol-treated patients at 15 and 30 mg/day, respectively. Most treatment-emergent adverse drug reactions were mild in intensity, and none were severe. CONCLUSION: The QTc interval in ziprasidone- and haloperidol-treated patients increased with dose. Treatment with high doses of ziprasidone or haloperidol did not result in any patient experiencing a QTc interval of 450 msec or greater.

Disease progression meta-analysis model in Alzheimer's disease.

BACKGROUND: Various authors have evaluated disease progression in Alzheimer's disease (AD), using patient data from individual clinical studies or pooled data across various trials. We conducted a systematic review of public data sources from 1990 to 2008 for all available AChE inhibitor studies, as well as clinical studies that evaluated the rate of deterioration in AD patients. Unique to this analysis, we developed a model based on literature data to describe the longitudinal response in the Alzheimer's Disease Assessment Scale-Cognitive (ADAS-cog) (change from baseline) in mild to moderate severity AD patients. The model was used to estimate disease progression for both placebo-treated patients and acetylcholinesterase (AChE)-inhibitor treated patients, and factors that affected disease progression. METHODS: We collected 576 mean ADAS-cog changes from baseline data points of 52 trials, representing data from approximately 19,972 patients and more than 84,000 individual observations. The model described the rate of disease progression, the evident placebo effect, and the symptomatic effect of AChE-inhibitors. Baseline ADAS-cog, Mini-Mental State Examination score, age, and year of publication were tested as covariates. RESULTS: The disease progression in mild to moderate AD patients across all available and relevant literature sources was estimated as 5.5 points per year. An Emax-type model best described the symptomatic drug effect of AChE inhibitors. The rate of disease progression (underlying disease progression) was no different between placebo and AChE-inhibitors groups. Baseline ADAS-cog is a significant covariate in disease progression. Baseline age was also tested as a covariate in the rate of disease progression, but the model was unable to describe any effects of age, likely because of the narrow distribution of mean age (literature-level analysis). There was no significant impact of publication year in the model. CONCLUSIONS: Baseline ADAS-cog is a significant covariate affecting the rate of disease progression, and it describes or at least explains the different rates of deterioration evident in early or late stages of the disease. There was no significant impact of publication year in the model, suggesting that disease progression has not slowed in more recent trials.

Population pharmacokinetic analysis of varenicline in adult smokers.

AIMS: To characterize the population pharmacokinetics of varenicline and identify factors leading to its exposure variability in adult smokers. METHODS: Data were pooled from nine clinical studies consisting of 1878 subjects. Models were developed to describe concentration-time profiles across individuals. Covariates were assessed using a full model approach; parameters and bootstrap 95% confidence intervals (CI) were estimated using nonlinear mixed effects modelling. RESULTS: A two-compartment model with first-order absorption and elimination best described varenicline pharmacokinetics. The final population parameter estimates (95% CI) were: CL/F, 10.4 l h(-1) (10.2, 10.6); V(2)/F, 337 l (309, 364); V(3)/F, 78.1 l (61.9, 98.9); Q/F, 2.08 l h(-1) (1.39, 3.79); K(a), 1.69 h(-1) (1.27, 2.00); and A(lag), 0.43 h (0.37, 0.46). Random interindividual variances were estimated for K(a)[70% coefficient of variation (CV)], CL/F (25% CV), and V(2)/F (50% CV) using a block covariance matrix. Fixed effect parameters were precisely estimated [most with % relative standard error < 10 and all with % relative standard error < 25], and a visual predictive check indicated adequate model performance. CL/F decreased from 10.4 l h(-1) for a typical subject with normal renal function (CLcr = 100 ml min(-1)) to 4.4 l h(-1) for a typical subject with severe renal impairment (CLcr = 20 ml min(-1)), which corresponds to a 2.4-fold increase in daily steady-state exposure. Bodyweight was the primary predictor of variability in volume of distribution. After accounting for renal function, there was no apparent effect of age, gender or race on varenicline pharmacokinetics. CONCLUSIONS: Renal function is the clinically important factor leading to interindividual variability in varenicline exposure. A dose reduction to 1 mg day(-1), which is half the recommended dose, is indicated for subjects with severe renal impairment.

Single-dose pharmacokinetics and safety of ziprasidone in children and adolescents.

OBJECTIVE: The purpose of this study was to provide single-dose pharmacokinetic, safety, and tolerability data for ziprasidone in youths with tic disorder, for comparison to adult studies to discern whether ziprasidone pediatric dosing could be modeled from adult data. METHOD: A single-dose, open-label study of ziprasidone was conducted in youths (ages 7-16 years) with Tourette's disorder or chronic tic disorder. Dosing of ziprasidone oral suspension (40 mg/mL) was weight adjusted: >60 kg, 20 mg (group 1, n = 8); 31 to 60 kg, 10 mg (group 2, n = 8); and 16 to 30 kg, 5 mg (group 3, n = 8). Patients were assessed for serum ziprasidone concentration, safety, tolerability, and electrocardiogram pre- and postdose. RESULTS: Twenty-four patients were evaluated for safety and tolerability, and 23 were evaluated for pharmacokinetics. Regression analysis of AUC(0-infinity) and Cmax values versus weight-normalized dose showed linear, dose-related changes in ziprasidone exposure. Ziprasidone was well tolerated with frequent, although transient, somnolence. No clinically significant change from baseline was observed in Bazett's or Fridericia's corrected QT(c) interval, and change in QT(c) interval was not related to serum ziprasidone concentration. CONCLUSIONS: Oral ziprasidone exhibited linear pharmacokinetics and dose-related exposure in youths with Tourette's disorder or chronic tic disorder, which are comparable to adult data. A single dose of ziprasidone was well tolerated without clinically significant effects on electrocardiograms collected around the time of maximum serum concentration.

Pharmacokinetics, safety, and tolerability of intramuscular ziprasidone in healthy volunteers.

Little has been published regarding the pharmacokinetics of the intramuscular (IM) formulation of Ziprasidone. The authors report results from 2 early phase I studies in healthy volunteers: a trial of single 5-, 10-, or 20-mg IM doses of ziprasidone in 24 subjects and an open-label 3-way crossover trial of 5-mg intravenous (IV), 5-mg IM, and 20-mg oral ziprasidone in 12 subjects. Absorption of IM ziprasidone was rapid (Tmax < 1 hour). The IM pharmacokinetic profile was consistent between studies and linear, with dose-related increases in exposure observed. The mean IM elimination t(1/2) was short and approximately 2.5 hours. The mean bioavailability for the 5-mg IM ziprasidone dose was approximately 100%. Adverse events were generally mild to moderate, and no subjects were discontinued from the study. No significant effects on renal function or other laboratory values were noted. These results support the use of IM ziprasidone in treating acutely agitated patients with schizophrenia.