Phase I, First-in-Human Study of BMS747158, a Novel ¹⁸F-Labeled Tracer for Myocardial Perfusion PET: Dosimetry, Biodistribution, Safety, and Imaging Characteristics After a Single Injection at Rest

Study Population

Healthy adults (as determined by medical history, physical examination, vital signs, electrocardiogram [ECG], electroencephalogram [EEG], neurologic examination, and clinical laboratory testing), aged 18–40 y, participated in the study. Subjects had to meet all protocol-specified inclusion criteria and none of the exclusion criteria. Subjects were to be men or women, aged 18–40 y inclusive, with a body mass index of 18–30 kg/m², with no clinically significant deviation from reference ranges in physical examination, ECG, EEG, and clinical laboratory parameters. All subjects who were women of child-bearing potential were nonpregnant and were using an adequate and medically approved method of contraception to avoid pregnancy for at least 1 mo before study enrollment through 1 mo after dosing and had a negative serum pregnancy test within 24 h before dose administration.

Subjects were excluded if they had any significant active or chronic medical illness or recent significant trauma or any condition that may have disrupted or increased permeability of the blood–brain barrier; any major surgery within 4 wk before enrollment or planned within 2 wk after completion of the study; a donation of blood or plasma to a blood bank or for a clinical study within 4 wk before enrollment; a blood transfusion within 4 wk of enrollment; any oral, transdermal, implanted, or injectable contraceptive hormones or hormone replacement therapy within 3 mo; and any radiopharmaceutical within a period equal to 10 half-lives of the isotope. Subjects were excluded if they had a positive urine screen for drugs of abuse at screening or before dosing or used any drugs (including over-the-counter medications and herbal preparations) within 2 wk before enrollment; if they had anticipated using drugs during the enrollment period through study follow-up; or for any for other sound medical, psychiatric, or social reason as determined by the principal investigator.

Study Design

This was a phase I, nonrandomized, open-label, single-dose study. Thirteen healthy adult subjects were enrolled and administered a single dose of BMS747158 at a single study center in the United States (UCLA Medical Center). Subjects were screened within 14 d before enrollment to confirm eligibility and began baseline assessments at the study center the day before study drug administration. Subjects remained at the study center until completion of the study day 2 safety assessments (24 ± 8 h after dosing). A telephone call was made to study subjects 48 ± 8 h after dosing for monitoring of adverse events (AEs). All subjects returned to the study center approximately 1 wk (5–7 d) after dosing for a follow-up safety visit and were contacted by telephone approximately 14–17 d after dosing for final monitoring of serious AEs. The major study events are displayed in Figure 2.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Study events overview. SAE = serious adverse event.

Determination of Dose and Method of Administration

The intravenous bolus injection was calculated to deliver no more than 8 mCi of ¹⁸F at the time of injection. The mean (±SD) final decay-corrected dose was 222 ± 22.2 MBq (6 ± 0.6 mCi) of ¹⁸F, with a range of 170–244 MBq (4.6–6.6 mCi). The difference between the target dose and the final dose was due to the retention of BMS747158 in the syringe. The 296-MBq (8-mCi) target dose was selected to provide adequate count statistics and was projected to be well below the maximum acceptable radiation exposure based on preclinical data. These data demonstrated that the maximum dose of BMS747158 that may be administered to a human without exceeding 50 mSv (5 rem) to the target was 742 MBq (20.0 mCi), and the injected dose that yielded an effective dose (ED) of 10 mSv (1 rem) or less was 666 MBq (18.0 mCi) (12).

On day 1, each subject received a 1- to 3-mL intravenous bolus injection of ¹⁸F-BMS747158 in a sterile solution of 5% or less ethanol containing 50 mg or less of sodium ascorbate per milliliter in water, calculated to deliver approximately the target dose of BMS747158 at the time of injection. The dose was administered in less than 10 s, followed immediately by a 3- to 5-mL saline flush. The net injected dose was calculated by subtracting the decay-corrected radioactivity in the syringe and injection tubing after injection from the assayed and decay-corrected radioactivity in the syringe before injection. The mass of BMS747158 tracer administered was 0.57 ± 0.1 μg.

PET Protocol

Whole-body PET from head to mid thigh was performed in 2-dimensional format at protocol-specified time windows outlined in Table 1 using a Siemens ECAT HR+ system and applying corrections for attenuation, randoms, and scatter. A low- to moderate-resolution spiral CT scan, from upper chest to lower abdomen, was obtained before or after the PET session to aid in organ identification and volume assessment, if needed.

Dosimetry Analyses

Estimates of radiation dosimetry for the standard organs of the adult male and female models and for the salivary glands, as well as the effective dose equivalent (13) and ED (14), were determined using the OLINDA/EXM software (15). Assessment of radiation dosimetry was based on the MIRD method, with data derived from imaging studies, using methods consistent with MIRD pamphlet no. 16 (16).

The attenuation-corrected transverse image data slice planes were combined into a single 3-dimensional image matrix for each subject and each time point using custom software. These images were then divided into 6 image sets (anterior, posterior, salivary, thyroid, source, and full) of combined coronal plane image data for each subject at each time point, grouping organs with similar anterior-to-posterior depths to optimize the region-of-interest (ROI) creation and minimize background contribution to the organs contained in each combined coronal plane image. The anterior images contained stomach wall, heart wall, and urinary bladder. The posterior images contained kidneys, lumbar spine, and spleen (when visible). The salivary images contained the salivary glands (parotid and submandibular). The thyroid images contained the thyroid. The full images combined all of the coronal image planes that contained subject image data and were used for quantification of the brain and liver. The source images contained the calibration source.

ROIs were drawn around all organs that showed uptake above background using custom software developed and validated for this purpose. Absolute radioactivity was determined by normalizing ROI sums by a calibration factor derived from the calibration source. Region counts were also adjusted for activity containing underlying and overlying tissue that was not part of the organ or tissue being quantified by use of background ROIs. Total body region counts were also corrected for off-body background counts. Appropriate normalization of region sizes for organ and adjacent regions were made. Unobstructed regions of organs with significant overlap from other activity-containing organs were also used where necessary. To estimate the activity in the lower legs (which were not imaged), an ROI on the upper thigh was used. Activities were also normalized when necessary to account for 100% of the injected activity and to ensure conservative (slight overestimates) determination of absorbed dose. Where urinary excretion data were available beyond the end of the imaging regimen, these data were used to determine whole-body retention.

Urinary Excretion Assay

Urinary excretion was collected for up to 8 h. Samples were marked for time of collection and volume of the entire void determined, and two 1-mL samples were assayed for radioactive (¹⁸F) content. A known amount of ¹⁸F was prepared and used as a 1-mL standard. Background counts were also determined. Activity in each urine void was then determined using Equation 1.AVoid = V[CTSS1 − CTSBC + CTSS2 − CTSBC2]Eq. 1where V is volume in milliliters of the total void, A_Void is activity in the total void, CTS_S1 is counts recorded in sample 1, CTS_S2 is counts recorded in sample 2, CTS_B is background counts, and C is the calibration factor (counts per unit activity based on counting standard).

Kinetic data for the brain, heart wall, kidneys, liver, red marrow (lumbar spine regions were used), salivary glands, spleen, stomach wall, thyroid, and urinary bladder for the subjects in the study were determined using the image quantification methodology. Absolute activity was converted to fractional dose by dividing by the total activity administered. Organ and tissue data were fit using nonlinear least-squares regression with sums of exponentials of the form shown in Equation 2, where f and λ are the model parameters that are determined in the fitting process, F_ij(t) is the fraction of the total injected activity, t is the time after injection, i is the i^th ROI, j is the j^th subject, and k is the k^th exponential term. Between 1 and 4 exponential terms were used, as appropriate.Fij(t)=∑kfijke−λijktj.Eq. 2

The regression was performed using custom software that determines initial parameter values based on the temporal variation of the kinetic data and the use of pretabulated estimates for various time–activity scenarios as selected by the user. Once these data were fit, residence times were determined by integration of these empirically determined functions (sums of exponentials) from time zero to infinity, taking into account physical decay. Remainder-of-body residence times were determined by subtraction of appropriate organ residence times from whole-body residence times. Urinary bladder residence times were determined using the parameters determined by fitting the whole-body activity data with a urinary bladder model as implemented in the OLINDA/EXM software with a 3.5-h bladder-voiding interval. Red marrow residence time was determined using an ROI drawn on a portion of the lumbar spine. The lumbar spine was assumed to contain 16.1% (17) of the total red marrow.

Organ and Tissue Dosimetry Estimates

Absorbed dose estimates for all target organs were determined using the OLINDA/EXM software with the adult male model. The resulting absorbed dose estimates were scaled using the total body mass of the individual subjects relative to that of the radiation transport phantom. Salivary gland dosimetry was determined using a conservative estimate of the S value for salivary glands based on the reference man total mass of the parotid and submaxillary salivary glands (18) and assuming a spheric shape. S values for spheres were produced by the OLINDA/EXM software and were linearly scaled on the basis of the relative total body mass of reference man to that of subject. These S values were then multiplied by the residence times to produce final salivary gland dose estimates.

Safety and Tolerability Evaluation

Subjects were closely monitored on site for 24 ± 8 h after drug administration, and safety monitoring continued for approximately 2 wk (14–17 d) after dosing. Several safety measures were included in this first-in-human study specifically to monitor for potential safety signals in the heart and brain. Multiple 12-lead ECGs were obtained at baseline and during anticipated peak BMS747158 levels and at 24 ± 8 h after dosing. Troponin-T levels were measured before and after the dose at 1, 8, and 24 h. In addition, a follow-up safety visit at approximately 1 wk (5–7 d) after dosing was conducted. Monitoring for neurophysiologic effects was conducted additionally through EEG assessments and neurologic examinations at scheduled time points before and after the dose.

An independent data-monitoring committee was established to monitor the safety of subjects participating in this clinical study. For each of the first 6 subjects enrolled, at least 1 data-monitoring committee member reviewed and assessed safety information through day 2 before the study drug was administered to the next subject. The remaining 7 subjects were enrolled with a minimum 48-h window between successive drug administrations.

The safety and tolerability endpoints were number and percentage of AEs and change from baseline in vital signs, laboratory values, ECG, EEG, neurologic examination, and physical examination that were of clinical significance as reported by the principal investigator.

Statistical Analyses

All statistical analyses and all summary tables and listings were prepared using SAS (release 9.1.3; SAS Institute, Inc.). Standard descriptive summaries included the N, mean, median, SD, or coefficient of variation (%CV); minimum and maximum for continuous variables; and the number and percentage for categoric variables.