Accelerating drug development at Bristol Myers Squibb through innovation.

This paper focuses on the use of novel technologies and innovative trial designs to accelerate evidence generation and increase pharmaceutical Research and Development (R&D) productivity, at Bristol Myers Squibb. We summarize learnings with case examples, on how we prepared and continuously evolved to address the increasing cost, complexities, and external pressures in drug development, to bring innovative medicines to patients much faster. These learnings were based on review of internal efforts toward accelerating R&D focusing on four key areas: adopting innovative trial designs, optimizing trial designs, leveraging external control data, and implementing novel methods using artificial intelligence and machine learning.

A randomized, controlled, repeat-dose study of batefenterol/fluticasone furoate compared with placebo in the treatment of COPD.

BACKGROUND: Batefenterol (BAT) is a bi-functional molecule with both muscarinic antagonist and ß2-adrenoceptor agonist pharmacology. This Phase II, randomized, placebo-controlled, double-blind study evaluated the safety and tolerability of BAT 300 µg with fluticasone furoate (FF) 100 µg administered via the ELLIPTA inhaler (BAT/FF 300/100). METHODS: Subjects with stable chronic obstructive pulmonary disease were randomized 2:1 to receive BAT/FF 300/100 or placebo once daily for 6 weeks. The primary endpoint was change from baseline in 0-4-h weighted mean (WM) heart rate (HR, measured by electrocardiogram [ECG]) on Day 42. Other endpoints included WM and maximum 0-4-h corrected QT interval (ECG on Days 1, 28, and 42), HR measured by Holter monitoring (Day 42), and standard safety assessments. Study protocol was approved by an Investigational Review Board. RESULTS: Sixty-two patients were randomized and received ≥1 dose of study medication (BAT/FF 300/100 n = 42; placebo n = 20). Mean age was 62.5 years (standard deviation [SD] 8.17). Study completion rates were 83% (BAT/FF 300/100) and 100% (placebo). Screening mean (SD) post-bronchodilator percentage-predicted forced expiratory volume in 1 s was 57.57 (11.42) in the BAT/FF 300/100 group and 55.68 (14.03) in the placebo group. BAT/FF 300/100 was non-inferior to placebo for the primary endpoint, treatment difference: - 2.2 beats per minute (bpm), 95% confidence interval [CI]: - 6.2, 1.7). There were no clinically relevant differences between treatment groups in WM or maximum 0-4-h corrected QT interval, or mean HR based on Holter monitoring on Day 42 (BAT/FF 300/100: 76.3 bpm [SD 11.38]; placebo: 84.8 bpm [SD 9.87]). Adverse events (AEs) occurred in 38% (BAT/FF 300/100) and 35% (placebo) of patients. AEs in ≥2 subjects with BAT/FF 300/100 were dysgeusia (10%), diarrhea (7%), nasopharyngitis (7%), and cough (5%). AEs leading to discontinuation occurred in two subjects who received BAT/FF 300/100: post-treatment severe pneumonia (serious AE) and non-serious AEs of moderate vomiting and severe gastroenteritis; both were not considered drug-related. No deaths occurred. CONCLUSIONS: Six weeks of BAT/FF 300/100 treatment was non-inferior to placebo for change from baseline in HR, with no new clinically relevant general or cardiovascular safety signals. TRIAL REGISTRATION: Clinicaltrials.gov: NCT02573870 (submitted October 12, 2015).

Randomized dose-finding study of batefenterol via dry powder inhaler in patients with COPD.

BACKGROUND: Batefenterol is a novel bifunctional muscarinic antagonist ß2-agonist in development for COPD. The primary objective of this randomized, double-blind, placebo-controlled, active comparator, Phase IIb study was to model the dose-response of batefenterol and select a dose for Phase III development. PATIENTS AND METHODS: Patients aged ≥40 years with COPD and FEV1 ≥30% and ≤70% predicted normal were randomized equally to batefenterol 37.5, 75, 150, 300, or 600 µg, placebo, or umeclidinium/vilanterol (UMEC/VI) 62.5/25 µg once daily. The primary and secondary endpoints were weighted-mean FEV1 over 0-6 hours post-dose and trough FEV1, analyzed by Bayesian and maximum likelihood estimation Emax of dose-response modeling, respectively, on day 42. RESULTS: In the intent-to-treat population (N=323), all batefenterol doses demonstrated statistically and clinically significant improvements from baseline vs placebo in the primary and secondary endpoints (191.1-292.8 and 182.2-244.8 mL, respectively), with a relatively flat dose-response. In the subgroup reversible to salbutamol, there were greater differences between batefenterol doses. Lung function improvements with batefenterol ≥150 µg were comparable with those with UMEC/VI. Batefenterol was well tolerated and no new safety signals were observed. CONCLUSION: Batefenterol 300 µg may represent the optimal dose for Phase III studies.

Treatment of spontaneous preterm labour with retosiban: a phase II pilot dose-ranging study.

AIMS: The aims of the present study were to investigate the maternal, fetal and neonatal safety and tolerability, pharmacodynamics and pharmacokinetics of intravenous (IV) retosiban in pregnant women with spontaneous preterm labour (PTL) between 340/7 and 356/7  weeks' gestation. METHODS: In parts A and B of a three-part, double-blind, placebo-controlled, multicentre study, women were randomized 3:1 (Part A) or 2:1 (Part B) to either 12-h IV retosiban followed by a single dose of oral placebo (R-P) or 12-h IV placebo followed by single-dose oral retosiban (P-R). RESULTS: A total of 29 women were randomized; 20 to R-P and nine to P-R. An integrated analysis found that adverse events were infrequent in mothers/newborns and consistent with events expected in the population under study or associated with confounding factors. Retosiban was rapidly absorbed after oral administration, with an observed half-life of 1.45 h. Efficacy analyses included 19 women. While not statistically significant, those receiving R-P more frequently achieved uterine quiescence in 6 h (R-P, 63%; 95% credible interval [CrI]: 38, 84; P-R, 43%; 95% CrI: 12, 78) and more achieved a reduction of ≥50% in uterine contractions in 6 h (R-P, 63%; 95% CrI: 38, 84; P-R, 29%; 95% CrI: 4, 64). The number of days to delivery was increased in women receiving R-P (median 26 days for R-P vs. 13 days for P-R). CONCLUSIONS: Intravenous retosiban has a favourable safety and tolerability profile and might prolong pregnancies in women with PTL. The study provides the rationale and dosing strategy for further evaluation of the efficacy of retosiban in the treatment of PTL.

Platelet aggregation unchanged by lipoprotein-associated phospholipase A2 inhibition: results from an in vitro study and two randomized phase I trials.

BACKGROUND: We explored the theorized upregulation of platelet-activating factor (PAF)- mediated biologic responses following lipoprotein-associated phospholipase A2 (Lp-PLA2) inhibition using human platelet aggregation studies in an in vitro experiment and in 2 clinical trials. METHODS AND RESULTS: Full platelet aggregation concentration response curves were generated in vitro to several platelet agonists in human plasma samples pretreated with rilapladib (selective Lp-PLA2 inhibitor) or vehicle. This was followed by a randomized, double-blind crossover study in healthy adult men (n = 26) employing a single-agonist dose assay of platelet aggregation, after treatment of subjects with 250 mg oral rilapladib or placebo once daily for 14 days. This study was followed by a second randomized, double-blind parallel-group trial in healthy adult men (n = 58) also treated with 250 mg oral rilapladib or placebo once daily for 14 days using a full range of 10 collagen concentrations (0-10 µg/ml) for characterizing EC50 values for platelet aggregation for each subject. Both clinical studies were conducted at the GlaxoSmithKline Medicines Research Unit in the Prince of Wales Hospital, Sydney, Australia. EC50 values derived from multiple agonist concentrations were compared and no pro-aggregant signals were observed during exposure to rilapladib in any of these platelet studies, despite Lp-PLA2 inhibition exceeding 90%. An increase in collagen-mediated aggregation was observed 3 weeks post drug termination in the crossover study (15.4% vs baseline; 95% confidence interval [CI], 3.9-27.0), which was not observed during the treatment phase and was not observed in the parallel-group study employing a more robust EC50 examination. CONCLUSIONS: Lp-PLA2 inhibition does not enhance platelet aggregation. TRIAL REGISTRATION: 1) Study 1: ClinicalTrials.gov NCT01745458 2) Study 2: ClinicalTrials.gov NCT00387257.

Inhibition of p38 mitogen-activated protein kinase reduces inflammation after coronary vascular injury in humans.

OBJECTIVE: To evaluate whether a p38α/ß mitogen-activated protein kinase inhibitor, SB-681323, would limit the elevation of an inflammatory marker, high-sensitivity C-reactive protein (hsCRP), after a percutaneous coronary intervention (PCI). METHODS AND RESULTS: Coronary artery stents provide benefit by maintaining lumen patency but may incur vascular trauma and inflammation, leading to myocardial damage. A key mediator for such stress signaling is p38 mitogen-activated protein kinase. Patients with angiographically documented coronary artery disease receiving stable statin therapy and about to undergo PCI were randomly selected to receive SB-681323, 7.5 mg (n=46), or placebo (n=46) daily for 28 days, starting 3 days before PCI. On day 3, before PCI, hsCRP was decreased in the SB-681323 group relative to the placebo group (29% lower; P=0.02). After PCI, there was a statistically significant attenuation in the increase in hsCRP in the SB-681323 group relative to the placebo group (37% lower on day 5 [P=0.04]; and 40% lower on day 28 [P=0.003]). There were no adverse safety signals after 28 days of treatment with SB-681323. CONCLUSIONS: In the setting of statin therapy, SB-681323 significantly attenuated the post-PCI inflammatory response, as measured by hsCRP. This inflammatory dampening implicates p38 mitogen-activated protein kinase in the poststent response, potentially defining an avenue to limit poststent restenosis.

Operating characteristics of a partial-block randomized crossover bioequivalence study for dutasteride, a drug with a long half-life: investigation through simulation and comparison with final results.

Studies to establish bioequivalence (BE) of a drug are important elements in support of drug applications. A typical BE study is conducted as a single dose, randomized, 2-period crossover design. For drugs with long half lives (≥ 48 hours) and evaluation of multiple BE objectives in 1 trial, this design may not be adequate. A parallel design may then be a more appropriate choice. However, parallel designs require increased sample size, which can become substantial. One option that is a compromise between the complete randomized block design and the parallel design is a partial-block crossover design. This approach came about during the development of a combination of dutasteride and tamsulosin. Previous experience with performing single-dose dutasteride studies suggested that 28 days of washout is needed between treatments because of its half-life of 7-9 days. Simulations were performed to assess the operating characteristics of this design using a previously developed PK model. Four scenarios were developed, and each scenario was simulated 500 times. The results showed that this design demonstrated acceptable consumer and producer risk. Partial-block crossover designs should be considered for studies when the half-life of the drug is long and there are more than 2 periods.

Ethanol does not alter the pharmacokinetic profile of the controlled-release formulation of carvedilol.

The concomitant ingestion of alcohol may alter the release of a drug from a modified-release dosage form, posing a potential risk to patients. In a randomized, open-label, 4-period cross-over study, the pharmacokinetic profiles of R(+) and S(-) carvedilol were compared after a single oral dose of carvedilol controlled-release formulation (administered following a standard meal) was given alone or concomitantly with ethanol. Thirty-nine healthy subjects participated in this study. Following coadministration of carvedilol controlled-release 40 mg with ethanol (approximately 38 g), ethanol ingestion 2 hours before or 2 hours after carvedilol controlled-release administration, area under the curve for the R(+) and S(-) carvedilol enantiomers was similar compared with carvedilol controlled-release given alone. Carvedilol exposure was not affected by the concomitant administration of ethanol and carvedilol controlled-release. Maximum plasma concentrations for the R(+) and S(-) carvedilol enantiomers were similar except when ethanol was ingested 2 hours after carvedilol controlled-release administration, where there was a modest decrease in maximum plasma concentration for R(+) and S(-) carvedilol (16% and 17%, respectively). Carvedilol controlled-release given alone or concomitantly with ethanol ingestion was generally well tolerated, and no serious or severe adverse events were reported. Ethanol did not alter the pharmacokinetic profile of carvedilol controlled-release.

Pharmacokinetic properties of a new controlled-release formulation of carvedilol.

This review summarizes the pharmacokinetics (PK) of carvedilol after administration of a new once-daily controlled-release (CR) formulation. The plasma concentration-time profiles for both R(+)- and S(-)-carvedilol indicate that carvedilol CR will provide coverage over a 24-hour period similar to the current immediate-release (IR) twice-daily formulation. Exposures for both enantiomers, based on area under the curve (AUC), maximum plasma concentrations (C(max)), and trough concentrations, are equivalent for carvedilol CR compared with carvedilol IR. C(max) and AUC of the enantiomers of carvedilol increase in an approximate dose-proportional manner after administration of carvedilol CR over the dose range of 10-80 mg, indicating that the formulation provides consistent PK performance across the dose strengths proposed for marketing. The intrasubject and intersubject variability of carvedilol CR was comparable to carvedilol IR. For carvedilol CR, mean AUC and C(max) were increased <20% after a high-fat meal compared with a standard meal. The CR and IR formulations of carvedilol exhibited equivalent steady-state PK characteristics in the target hypertension and heart failure populations. The availability of once-daily dosing is expected to improve treatment adherence and thereby enhance the effectiveness of carvedilol in routine clinical use.

Development of a pharmacokinetic/pharmacodynamic model for carvedilol to predict beta1-blockade in patients with congestive heart failure.

To determine whether the controlled-release (CR) formulation of carvedilol given once daily provides 24-hour beta1-receptor blockade similar to the currently marketed immediate-release (IR) formulation given twice daily, changes in exercise-induced heart rate after bicycle ergometry were measured. The pharmacokinetic (PK)/pharmacodynamic (PD) relation between S(-)-carvedilol concentration-the enantiomer with beta-blocking activity-and change in exercise-induced heart rate was defined in healthy subjects and was best described using a direct effect inhibitory E(max) model (with E(max) being the maximum effect). The population estimates for E(max) and concentration at 50% of the maximum effect (EC50) were 19.2 beats per minute (an approximately 13% maximum decrease in exercise-induced heart rate) and 7.7 ng/mL, respectively. The PK/PD model was used to predict PD effects in patients with mild-to-severe heart failure and in patients after myocardial infarction with left ventricular dysfunction who had received both the IR and CR formulations of carvedilol. In these patients, carvedilol CR had equivalent predicted overall PD (area under the effect curve) and trough (PD(min)) effects compared with carvedilol IR, indicating 24-hour beta-blocking coverage for the new CR formulation of carvedilol given once daily.

Pharmacokinetic and pharmacodynamic comparison of controlled-release carvedilol and immediate-release carvedilol at steady state in patients with hypertension.

Carvedilol is indicated for the treatment of essential hypertension and mild-to-severe chronic heart failure, as well as the reduction of cardiovascular mortality in clinically stable post-myocardial infarction patients with left ventricular dysfunction. Carvedilol is a racemic mixture of R(+) and S(-) enantiomers that combines beta(1)-, beta(2)-, and alpha(1)-adrenoceptor blockade. For all indications, the immediate-release (IR) formulation of carvedilol is taken twice daily. A controlled-release (CR) formulation of carvedilol that allows once-daily dosing has recently been developed. In this double-blind, parallel-group, crossover study, 122 patients with essential hypertension were randomly allocated to receive low and high doses of carvedilol or placebo. Patients received either a constant low dose (CR 20 mg once daily or IR 6.25 mg twice daily) or were titrated to a high dose (CR 80 mg once daily or IR 25 mg twice daily) before being crossed over to an equivalent dose of the alternative formulation. The pharmacokinetic (PK) and pharmacodynamic (PD) profiles were compared between patients receiving carvedilol CR and carvedilol IR. The PK profiles for R(+)- and S(-)-carvedilol for the 2 formulations were equivalent (based on area under the curve, maximum plasma concentration [C(max)], and trough drug concentration). Consistent with an extended-release formulation, carvedilol CR delayed C(max) by 3.5 hours compared with carvedilol IR. For both carvedilol CR and IR, the attenuation of exercise-induced heart rate in patients with hypertension was maintained over the entire 24-hour period, and the 2 formulations demonstrated equivalent beta(1)-blocking effects at trough (end of the dosing interval [PD(min)]), suggesting that the rate of absorption does not interfere with the PD effect. In this first direct comparison of carvedilol CR and IR in subjects with hypertension, fewer adverse events were reported while subjects were receiving carvedilol CR (59.1% overall) compared with carvedilol IR (77.5% overall). This was true regardless of dose received. Headache was the most commonly reported adverse event for subjects receiving either formulation of carvedilol and placebo. Importantly, dizziness and headache were reported less often when subjects received carvedilol CR. This is the first study to show that both formulations had comparable beta(1)-adrenergic blockade in patients with essential hypertension under steady-state conditions. Notably, carvedilol CR provides consistent beta(1)-adrenergic blockade over 24 hours with a once-daily dose.

Pharmacokinetic profile of controlled-release carvedilol in patients with left ventricular dysfunction associated with chronic heart failure or after myocardial infarction.

We compared the pharmacokinetic (PK) profiles of repeated dosing with the currently available immediate-release (IR) carvedilol (given twice daily) and a newly developed controlled-release (CR) formulation (given once daily) in patients with left ventricular dysfunction (LVD). We enrolled 188 patients with stable mild, moderate, or severe heart failure as well as survivors of a recent acute myocardial infarction (MI) with asymptomatic LVD (left ventricular ejection fraction /=2 weeks. Patients were then switched to the corresponding dose of carvedilol CR (10, 20, 40, or 80 mg once daily), and PK variables were reassessed after an additional 2 weeks. The primary measures included trough plasma concentration (C(tau)), maximum plasma concentration (C(max)), and area under the concentration-time curve (AUC([0-t])) for both enantiomers of carvedilol of each formulation of the drug. The AUC((0-t)) and the trough and maximum carvedilol concentrations for both the R(+) and S(-) enantiomers were similar when carvedilol IR was compared with carvedilol CR as follows: 3.125 mg twice daily versus 10 mg once daily, 6.25 mg twice daily versus 20 mg once daily, 12.5 mg twice daily versus 40 mg once daily, and 25 mg twice daily versus 80 mg once daily, respectively. Based on a pooled analysis, the AUC((0-t)), C(max), and C(tau) for both R(+) and S(-) were equivalent for the CR and IR formulations, with point estimates and 90% confidence intervals within the bioequivalence limits of 80%-125%. The fluctuation index (the CR-IR ratio for [C(max) - C(min)]/C(ss) where C(min) is the minimum observed concentration over the dosing interval and C(ss) is the concentration at steady state) for both R(+)- and S(-)-carvedilol was approximately 1, indicating that the peak-to-trough fluctuation in plasma concentration for carvedilol after use of carvedilol CR once daily was similar to that for carvedilol IR given twice daily. The median time to maximum observed plasma concentration (t(max)) was approximately 3 hours longer for both enantiomers after administration of carvedilol CR as compared with carvedilol IR. These data demonstrate that the new carvedilol CR formulation given once daily is equivalent to the currently available carvedilol IR formulation given twice daily in patients with heart failure and asymptomatic post-MI LVD.

Pharmacodynamics and pharmacokinetics of oral contraceptives co-administered with alosetron (Lotronex).

The 5-HT3 receptor antagonist alosetron (Lotronex) is indicated for use in women with severe, chronic, diarrhea-predominant irritable bowel syndrome (IBS) who have failed conventional therapy. Oral contraceptives (OCs) and alosetron are potential co-medications in women of childbearing age. This study assessed the effect of alosetron co-administration on pharmacodynamic markers of contraceptive efficacy, on the pharmacokinetics of estrogen and progesterone OC components, and on the activity of biochemical markers for the risk of thrombosis. This was an open label, nonrandomized two-way crossover study in 18 healthy women stabilized for 3 months on a low-dose OC containing ethinyl estradiol (EE) and levonorgestrel (LN). Alosetron had no effect on serum concentrations of luteinizing hormone (LH) or follicle-stimulating hormone. Ovarian activity grades (assessing follicle size, progesterone, and 17beta-estradiol concentrations) were similar during OC use with and without alosetron. Steady-state (Day 21) AUC24, Cmax, and tmax of both LN and EE were similar during coadministration of alosetron with an OC. Concentrations and activity of biochemical markers of thrombosis risk were not different in the presence of alosetron. These results indicate that alosetron does not alter the pharmacokinetics or pharmacodynamic markers of efficacy for a low-dose combination OC. The results also suggest that thromboembolic risk is not increased when alosetron is co-administered with an OC.

Fluticasone propionate aqueous nasal spray provided significantly greater improvement in daytime and nighttime nasal symptoms of seasonal allergic rhinitis compared with montelukast.

BACKGROUND: The safety and efficacy of intranasal corticosteroids for the treatment of allergic rhinitis is well documented in the literature. Additionally, an expert panel has concluded that intranasal corticosteroids are the first line of therapy when obstruction is a major component of rhinitis. Montelukast is a leukotriene receptor antagonist recently approved for the treatment of seasonal allergic rhinitis (SAR). OBJECTIVE: This randomized, double-blind, double-dummy, parallel-group study was conducted to compare the effectiveness of a 15-day course of intranasal fluticasone propionate 200 microg, once daily (FP200QD), to oral montelukast 10 mg, once daily (MON10QD), in relieving daytime and nighttime nasal symptoms associated with SAR. METHODS: The intent-to-treat (ITT) analysis population consisted of 705 eligible males and females (> or = 15 years) with SAR randomized to either FP200QD (N = 353) or MON10QD (N = 352). The primary efficacy endpoint was the mean change from baseline in subject-rated daytime total nasal symptom scores (the sum of four individual scores: nasal congestion, itching, rhinorrhea, and sneezing), evaluated via visual analog scales, and averaged over weeks 1 to 2. Secondary endpoints included the four daytime individual nasal symptom scores, the nighttime total, and individual nasal symptom scores (each evaluated on a four-point scale from 0 to 3). RESULTS: Statistically significant differences favoring FP200QD over MON10QD were observed for the mean change from baseline in daytime total nasal symptom scores (P < 0.001), daytime individual nasal symptom scores (P < 0.001), nighttime total (P < 0.001), and all individual nasal symptom scores (P < or = 0.002) over the 15-day treatment period. FP200QD and MON10QD were both well tolerated. CONCLUSIONS: The results of this well controlled study demonstrated that FP200QD was consistently superior to MON10QD with regard to every efficacy endpoint evaluated, including daytime and nighttime nasal congestion, in subjects with SAR.