Keywords
gestation - hydroxyprogesterone - metabolites - pharmacokinetics - pregnancy - preterm
birth
Worldwide, preterm birth (PTB), delivery prior to 37 weeks of gestation, is the most
common cause of neonatal death and the second most common cause of death in children
under age 5 years.[1] According to the U.S. Centers for Disease Control and Prevention, the U.S. incidence
of PTB peaked in 2006 at 12.8%. This represented a 21% increase since 1990 and a 33%
increase from 1981 to 2004.[2] Since 2006, the rate has fallen slightly to a rate of 11.4% in 2013.[3]
PTB may occur spontaneously or owing to a medical indication. One significant risk
factor for spontaneous PTB is previous pregnancy history. Women with a prior spontaneous
PTB have a 2.5-fold greater risk for subsequent spontaneous PTB than those without.
In 2011, the U.S. Food and Drug Administration (FDA) approved hydroxyprogesterone
caproate (17-OHPC) injection to reduce the risk of recurrent spontaneous PTB in women
with a singleton pregnancy and history of prior singleton spontaneous PTB. At the
time of the FDA approval, limited data were available on the pharmacokinetics (PK)
of 17-OHPC in pregnant women, and no human PK studies quantitating 17-OHPC metabolites
exist. Our objective was to measure the PK of 17-OHPC and its major metabolites throughout
pregnancy.
Materials and Methods
Study Design
From July 1, 2013 to November 16, 2014, we conducted protocol HPC-PK-005 at six U.S.
academic and community centers. Each study site had institutional review board approval,
and all women gave written informed consent to participate. Thirty-one eligible women
prescribed 17-OHPC (Makena®, Hospira Inc., McPherson, KS) for recurrent PTB prevention were approached, and 30
agreed to participate. Subjects had blood drawn for 7 consecutive days at one of three
times throughout the pregnancy: cohort 1 (n = 6) after the first dose (weeks 16–20), cohort 2 (n = 8) between weeks 24 and 28, and cohort 3 (n = 16) between weeks 32 and 36 ([Table 1]). We measured serum trough level after week 1 in cohort 1 or after 2 consecutive
weekly doses in cohorts 2 and 3. In 10 subjects, we estimated 17-OHPC terminal half-life
at 28 days after their last dose.
Table 1
Study cohort blood-draw schedule for pharmacokinetic measurements
|
Cohort
|
Following first dose
|
Weeks 24–28
|
Weeks 32–36
|
|
1 (n = 6)
|
Daily for 7 days to establish C
max
|
Two consecutive trough levels
|
Two consecutive trough levels
|
|
2 (n = 8)
|
Prior to dose 2
|
Daily for 7 days to establish C
max
|
Two consecutive trough levels
|
|
3 (n = 16)
|
Prior to dose 2
|
Two consecutive trough levels
|
Daily for 7 days to establish Cmax
|
|
Elimination phase
|
Blood levels collected for 5 of the 7 days following the final dose, omitting the
weekends and within ± 1 day of days 11, 17, 24, and 28.
|
Abbreviation: C
max, maximum concentration.
After enrollment, medical and obstetric history was obtained, at least one prior PTB
was documented, and subjects underwent a brief physical examination. Maternal race/ethnicity
was self-reported. Gestational age was based upon the date of the last menstrual period
unless the subject was unsure of the date, or it was inconsistent with ultrasound
when ultrasound assessment was used. A 1-mL intramuscular injection containing 17-OHPC
250 mg was administered into the upper outer quadrant of the gluteus maximus. At weekly
visits, we documented adverse events, maternal complications, and pregnancy complications.
Subjects were followed up until delivery, and infants were followed until the later
of 28 days of life or discharge from the neonatal intensive care unit (NICU).
Blood Sample Analysis
Maternal blood was collected into 10-mL lavender-top vacutainers containing ethylenediaminetetraacetic
acid. Samples were gently inverted 5 to 10 times and centrifuged at 1,100 to 1,800 × g for 10 to 15 minutes. Plasma was transferred equally into two 5-mL cryovials and
frozen at –20°C or below.
Covance Inc. (West Trenton, NJ) measured 17-OHPC using a validated bioanalytical method.
17-OHPC (US Pharmacopeia) was used as a reference material and 4-Pregnen-17α-ol-3,20-dione-2,2,4,6,6,21,21,21-d8
Hexanoate as an internal standard. Analysis was performed in staggered parallel mode
using an ARIA-TX4 HPLC system with API4000 MS detection. Plasma samples were analyzed
in accordance with the laboratory's standard operating procedures. A single analysis
was performed on each sample. Each run included a duplicate calibration curve, a matrix
blank, a control zero sample (matrix blank containing internal standard), and duplicate
undiluted quality control (QC) samples at four concentrations within the calibration
range. The calibration standards were placed at the beginning and end of each analytical
run, and the QC samples were interspersed with study samples within each analytical
run. Dilution QC samples were included in any run where samples were diluted prior
to analysis.
There is no commercially available assay for 17-OHPC metabolites, thus Research Triangle
Institute (Research Triangle Park, NC) developed a metabolite assay using synthesized
standards, with purity by high performance liquid chromatography of 99.3%. For standards,
0.5 mL of blank human plasma was mixed with 0.5 mL of 2% formic acid and 25 μL of
internal standard (MPA 1 μg/mL) and 5 μL of standard spike. The samples were vortexed
and extracted on an Oasis HLB 3 cc solid phase extraction cartridge preconditioned
with methanol (2 mL) followed by water (2 mL). Cartridges were washed with 5% methanol
and then eluted with 2-mL methanol under vacuum at 10 mm Hg. The methanol extracts
were dried on a Turbovap LV under nitrogen and were resolubilized in 100 µL of 50:50
acetonitrile:water with 0.1% formic acid by vortexing (30 s); 20 μL of each was injected
on an Applied Biosystems API-5000 triple quadrupole mass spectrometer equipped with
a Waters Acquity Ultra-Performance Liquid Chromatography. Plasma samples were thawed
on ice and were briefly vortexed to ensure homogeneity. An aliquot (0.5 mL) was removed
from each plasma sample, mixed with methanol (5 μL) and 2% formic acid in water (0.5
mL), and extracted as above.
Mono-hydroxylated hydroxyprogesterone caproate (17-OHPC-OH) levels were determined
(ng/mL) based on standard curves. As no di-hydroxylated hydroxyprogesterone caproate
(17-OHPC-diOH) standard is available, we estimated levels using the 17-OHPC-OH standard
curve. Limits of quantification for the standards were 0.5 ng/mL for 17-OHPC and 0.1
ng/mL for 17-OHPC-OH, and the inter assay precision ranged from 2.5 to 6.8%.
PK Analysis
Subjects were included in PK assessments if multiple blood draws following a dose
of 17-OHPC were scheduled and at least one blood draw obtained. Subjects were included
in trough-level assessments in each interval if they did not have a PK assessment
and had blood drawn immediately prior to at least one dose of 17-OHPC. 17-OHPC and
metabolites plasma PK measurements included area under the plasma concentration-time
curve (AUC) from time = 1 to the last measurable concentration (AUClast), maximum observed concentration (C
max), time at which maximum concentration is observed (T
max), terminal elimination half-life (T
1/2), and trough concentrations (C
trough).
17-OHPC PK measurements were calculated using SAS® for Windows Version 9.4. PK mono- and di-hydroxylated metabolite measurements were
calculated using a non-compartmental analysis in Phoenix WinNonlin Version 6.3. For
PK analysis, plasma levels below limit of detection at predose to the first concentration
imputed to 0, and those that occur thereafter were treated as missing.
Statistical Methods
Data were collected using an electronic data capture system (Cmed, Timaeus Inc., Horsham,
United Kingdom) and a contract research organization (ResearchPoint Global Austin,
TX) monitored study records for data completeness and quality. We calculated mean,
median, standard deviation (SD), and range for continuous variables, and frequency
and percentage for categorical variables.
Results
Study Population
[Table 2] shows subject characteristics. Two-thirds were Caucasian, 10% reported smoking during
pregnancy, and mean ± SD pre-pregnancy body mass index (BMI) was 31.9 ± 10.3 kg/m2. The median number of prior preterm deliveries was one, and 11 (37%) subjects had
more than one prior PTB.
Table 2
Study cohort characteristics and previous pregnancy history
|
All subjects
(N = 30)
|
Cohort 1
(n = 6)
|
Cohort 2
(n = 8)
|
Cohort 3
(n = 16)
|
|
Age (years)
|
|
Mean ± SD
|
29.4 ± 5.5
|
29.4 ± 6.6
|
29.4 ± 6.6
|
29.4 ± 6.6
|
|
Ethnicity, n (%)
|
|
Hispanic or Latino
|
2 (6.7)
|
0 (0.0)
|
1 (12.5)
|
1 (6.3)
|
|
Non-Hispanic or non-Latino
|
28 (93.3)
|
6 (100.0)
|
7 (87.5)
|
15 (93.8)
|
|
Race, n (%)
|
|
White
|
20 (66.7)
|
4 (66.7)
|
6 (75.0)
|
10 (62.5)
|
|
Black, African-American,
or African heritage
|
10 (33.3)
|
2 (33.3)
|
2 (25.0)
|
6 (37.5)
|
|
Pre-pregnancy weight (kg)
|
|
Mean ± SD
|
83.8 ± 27.58
|
67.5 ± 15.6
|
87.2 ± 33.5
|
88.2 ± 27.1
|
|
Pre-pregnancy BMI (kg/m2)
|
|
Mean ± SD
|
31.9 ± 10.3
|
27.0 ± 7.0
|
33.1 ± 12.1
|
33.1 ± 10.4
|
|
Substance abuse, n (%)
|
|
Smoking
|
3 (10.0)
|
1 (16.7)
|
0
|
2 (12.5)
|
|
Alcohol
|
0
|
0
|
0
|
0
|
|
Drugs
|
2 (6.7)
|
1 (16.7)
|
0
|
1 (6.3)
|
|
Previous preterm births
|
|
Mean ± SD
|
1.6 ± 1.1
|
2.2 ± 2.0
|
1.8 ± 0.7
|
1.3 ± 0.6
|
|
>1 Previous preterm birth, n (%)
|
11 (36.7)
|
2 (33.3)
|
5 (62.5)
|
4 (25.0)
|
Abbreviations: BMI, body mass index; SD, standard deviation.
Data shown as column number (%) or as continuous values with SD.
Twenty (67%) of 30 subjects received 17-OHPC through gestational week 36. Six subjects
discontinued early due to a premature delivery, one declined the week-36 dose, one
was incarcerated and thus withdrawn at the time of incarceration, one was found to
be ineligible and was withdrawn after one dose, and one did not receive the week-36
dose due to scheduling difficulties. All subjects in cohorts 1 and 2 had blood drawn
for PK measurements, as did 11 of 16 subjects in cohort 3; the five subjects did not
have blood drawn because of delivery or were lost to follow-up. The median number
of injections per subject was 19 (range 1–22), and all subjects received ≥80% of injections
within a 7-day interval.
PK Results
[Table 3] summarizes PK results, including AUC 1–7, C
max, T
max, and mean trough levels. 17-OHPC and 17-OHPC-OH levels were adequate for analysis,
while the levels of 17-OHPC-diOH were below detectable levels to allow for detailed
analysis.
Table 3
PK results
|
Time period
|
Dose 1
|
Weeks 24–28
|
Weeks 32–36
|
Elimination
|
|
Number of subjects for PK assessment
|
6
|
8
|
11[a]
|
10
|
|
Number of trough levels taken
|
19
|
19
|
12
|
|
|
17-OHPC (mean ± SD)
|
|
AUC (ng × h/mL)[b]
|
571.4 ± 195.2
|
1,269.6 ± 285.0
|
1,268.0 ± 511.6
|
|
|
C
max (ng/mL)
|
5.0 ± 1.5
|
12.5 ± 3.9
|
12.3 ± 4.9
|
|
|
T
max (h)
|
119.8 ± 50.1
|
28.6 ± 9.0
|
43.7 ± 20.2
|
|
|
Half-life (days)
|
|
|
|
16.3 ± 3.6
|
|
Clearance (λ
z) (/h)
|
|
|
|
0.002 ± 0.0004
|
|
C
trough (ng/mL)
|
7.5 ± 7.1
|
10.0 ± 4.8
|
8.9 ± 2.9
|
|
|
17-OHPC-OH (mean ± SD)
|
|
AUC (ng × h/mL)[b]
|
208.5 ± 92.4
|
157.1 ± 64.6
|
211.2 ± 113.1
|
1,205.2 ± 995.2
|
|
C
max (ng/mL)
|
1.9 ± 0.7
|
1.5 ± 0.7
|
1.8 ± 1.0
|
3.2 ± 3.0
|
|
T
max (h)
|
107.7 ± 44.9
|
66.4 ± 53.3
|
91.5 ± 43.8
|
299.3 ± 213.4
|
|
Half-life (days)
|
|
|
|
19.7 ± 6.2
|
|
C
trough (ng/mL)
|
0.6 ± 0.3
|
1.0 ± 0.7
|
0.8 ± 0.8
|
|
|
17-OHPC-diOH (mean ± SD)
|
|
C
max (ng/mL)
|
<0.1
|
0.28 ± 0.12
|
0.48 ± 0.36
|
|
|
T
max (h)
|
n/a
|
81.0 ± 51.2
|
104.7 ± 55.0
|
|
|
C
trough (ng/mL)
|
<0.1
|
0.28 ± 0.14
|
0.42 ± 0.22
|
|
Abbreviations: AUC, area under the time concentration curve; C
max, maximum concentration; C
trough, trough concentration; 17-OHPC, hydroxyprogesterone caproate; 17-OHPC-diOH, di-hydroxylated
hydroxyprogesterone caproate; 17-OHPC-OH, mono-hydroxylated hydroxyprogesterone caproate;
n/a, not applicable; PK, pharmacokinetic; SD, standard deviation; T
max, time to maximum concentration.
a Five of 16 patients did not have PK blood drawn due to delivery or lost to follow-up.
b AUC1–7 for treatment intervals and AUC0–28 for elimination phase.
Individual plots of 17-OHPC levels by gestational age and subject are shown in [Fig. 1]. 17-OHPC PK, as measured by C
max and AUC1–7, were moderately variable in each cohort with % coefficient of variation (%CV) values
ranging from 30 to 40%. After the first dose, the mean ± SD C
max level of 5.0 ± 1.5 ng/mL was achieved by 119.8 ± 50.1 hours. The exposure to 17-OHPC,
as assessed by AUC1–7 after the first injection, was 571.4 ± 195.2 ng × h/mL. By 24 to 28 weeks and 32
to 36 weeks, mean ± SD AUC1–7 was 1,269.6 ± 285.0 ng × h/mL and 1,268.0 ± 511.6 ng × h/mL, and mean ± SD C
max values were 12.5 ± 3.9 ng/mL and 12.3 ± 4.9 ng/mL, respectively. Mean ± SD T
max values were 28.6 ± 9.0 hours at weeks 24 to 28 and 43.7 ± 20.2 hours at weeks 32
to 36. There was slow accumulation of the drug after the first administration, but
by week 24, an approximate steady state was reached.
Fig. 1 17-OHPC levels (ng/mL) by subject and time in gestation: (A) Postdose 1; (B) Weeks 24–28; (C) Weeks 32–36, and (D) Elimination phase. C
max and AUC1–7 % coefficient of variation ranged from 30 to 40%.
17-OHPC-OH PK following the first dose and at 24 to 28 weeks varied, as exhibited
by %CV estimates of 37 to 46%. 17-OHPC-OH PK was more variable at 32 to 36 weeks,
with %CV estimates of 54 to 56%. After dose 1, the maximum 17-OHPC-OH level was 1.9 ± 0.7
ng/mL, achieved by ∼113.6 hours (median). Exposure to 17-OHPC-OH, as assessed by AUC1–7, was 208.5 ± 92.4 ng × h/mL. Mean ± SD C
max estimates at 24 to 28 weeks and 32 to 36 weeks were 1.5 ± 0.7 ng/mL and 1.8 ± 1.0
ng/mL, respectively. Mean ± SD AUC1–7 estimates were 157 ± 64.6 ng × h/mL and 211 ± 113.1 ng × h/mL, respectively, at the
same time points. Exposure to the 17-OHPC-OH was approximately 6 to 10 times lower
than that of 17-OHPC. 17-OHPC C
max and AUC increased following repeat injections. Summaries of 17-OHPC and 17-OHPC-OH
levels are presented in [Figs. 2] and [3], respectively.
Fig. 2 Average daily 17-OHPC plasma concentration (ng/mL) throughout pregnancy by time in
gestation and blood-draw schedule.
Fig. 3 Average daily 17-OHPC-OH plasma concentration (ng/mL) throughout pregnancy by time
in gestation and blood-draw schedule.
17-OHPC-diOH levels were ∼40 times lower than 17-OHPC and four to five times lower
than 17-OHPC-OH. 17-OHPC and 17-OHPC-OH trough levels were consistent at all measured
time points and were, in general, also consistent with those for subjects sampled
for 7 days in the same intervals.
The terminal half-lives of 17-OHPC and 17-OHPC-OH were determined. Although 10 subjects
had blood draws after week 36, serum levels did not decline in a log-linear manner
for six subjects; therefore, only 4 of the 10 subjects contributed data to the determination.
The half-life was 16.3 ± 3.6 days and 19.7 ± 6.2 days for 17-OHPC and 17-OHPC-OH,
respectively. At the end of the sampling interval, measurable levels were still evident
for 17-OHPC (2.9–14.8 ng/mL) and 17-OHPC-OH (1.4–2.0 ng/mL). There was no difference
in mean 17-OHPC trough levels measured between weeks 24 and 28 for women with a pre-pregnancy
BMI <30 kg/m2 versus women with a BMI ≥30 kg/m2 (10.3 ± 4.9 versus 8.4 ± 2.9, respectively, p = 0.22). Four (29%) of 14 subjects with a trough level less than the median at 24
to 28 weeks delivered preterm, compared with two (14%) of 14 with a value above the
median.
Maternal Adverse Events, Pregnancy Complications, and Infant Outcomes
Twenty-four (80.0%) of 30 subjects experienced adverse events. The most common maternal
adverse events were gastrointestinal. No subject discontinued 17-OHPC early except
for preterm delivery. Twelve (40%) of 30 women experienced pregnancy complications.
Six (21%) infants were born preterm (at <37 weeks), three spontaneous and three medically
indicated. One subject had a miscarriage at 18 weeks 4 days, which was deemed unlikely
to be drug related. Infant outcomes were available for 28 of 30 subjects. Twenty-seven
infants were live-born. Mean gestational age was 36.3 weeks and mean birth weight
was 2,965.5 g. Seven (25.9%) infants were admitted to the NICU for a median of 10
days (range 1–99).
Discussion
Our results define the PK parameters of 17-OHPC and its metabolites in 17-OHPC-treated
pregnant women. At 24 weeks, 17-OHPC trough plasma levels were 8.9 to 10 ng/mL, and
corresponding C
max values were ∼12.5 ng/mL, which is lower than reported by Caritis et al.[4]
[5] The primary clinical trial of 17-OHPC for PTB prevention used 250-mg weekly injections,
with no PK data to inform that decision.[6] The dose selected was the most commonly reported dose and was close to the drug
solubility limit in 1 mL of the oil-based formulation. Previous studies used doses
ranging from 250[7]
[8] to 1,000 mg/week.[9]
Dose response may be an important factor for 17-OHPC efficacy for PTB prevention.
In a secondary analysis of the Meis et al trial,[6] Heyborne et al[10] concluded that 17-OHPC was effective in preventing repeat PTB only in women with
a pre-pregnancy BMI <30 kg/m2. However, Meis et al[11] previously reported that a high pregravid BMI in the placebo group was associated
with a negative (protective) association with preterm delivery, while BMI did not
affect the rate of preterm delivery in the 17-OHPC group. We did not find differences
in PTB or 17-OHPC trough levels between obese and non-obese women. In a secondary
analysis of 17-OHPC and omega-3 to prevent PTB,[12] Caritis et al[13] found that subjects with 17-OHPC plasma levels in the lowest quartile had a higher
risk of PTB and delivery at earlier gestational ages than women in the upper quartiles
had. Our data did not demonstrate differences in PTB by 17-OHPC level quartile. Our
sample size was not powered to detect differences in PTB by maternal BMI or 17-OHPC
levels. Additional studies may be needed in certain populations, such as women with
a higher BMI.
At the time of the FDA approval of 17-OHPC for PTB prevention, there was little understanding
of 17-OHPC PKs. A study of women with endometrial cancer on long-term 17-OHPC (1,000 mg
daily for 5 days followed by 1,000 mg every 2 weeks) found that 17-OHPC levels peaked
2 weeks following drug initiation, and that women treated with weekly injections had
higher serum levels than those treated every 2 weeks had.[14] Subsequently, two studies reported the PK of 17-OHPC. In 61 women with singleton
pregnancies receiving 17-OHPC, Caritis et al[4] reported that the half-life of 17-OHPC was 16.2 days, and levels were higher at
31 to 35 weeks than 20 to 25 weeks. In women pregnant with twins or triplets, the
half-life of 17-OHPC was 10 days.[5]
The half-life of 16.3 days we report is an apparent half-life, the net result of drug
being eliminated from the plasma while simultaneously being replaced by drug absorbing
from the oil depot in the muscle. The half-life of 17-OHPC itself in the blood may
be much shorter. In a PK study of a novel oral formulation of 17-OHPC, dosing twice
a day was needed to maintain blood levels.[15] In a study of a cremophor:ethanol-based formulation in rats, the half-life was 10.6
hours when given intravenously but was not reached after 120 hours when given intramuscularly.[16]
We provide information that has not been previously published, including PK after
the first dose and quantification of 17-OHPC-OH and 17-OHPC-diOH. A limitation of
our analysis is that we did not collect trough samples during intensive PK studies
at 24 to 28 and 32 to 36 weeks, thus we can only speculate on trough levels as pregnancy
advances. Further research is needed to determine whether these metabolites are active.
Our findings confirm previous observations that 17-OHPC has a long terminal half-life
and accumulates with multiple dosing until a steady state is reached. However, our
reported levels are lower than those reported by Caritis et al.[4] Given the similar methodology, the reason for the difference is unclear, although
there were differences in sampling times relative to dosing. Our results were produced
using a validated assay where validation and sample analysis were commensurate with
international standards of Good Laboratory Practice.
Concentrations of 17-OHPC and 17-OHPC-OH reach steady state by weeks 24 to 28, but
there is considerable variability among subjects. The reasons for this variability
remain unclear. Possibly there are absorption or metabolism differences, and these
differences could impact drug levels. Other factors that can lead to variability of
plasma levels include age, hormonal status, conditions at the absorption site, circulation
to the absorption site, and area of the absorbing surface (determined largely by the
route of administration).[5]
[17]
[18] Our objective was to measure the PKs of 17-OHPC and its metabolites throughout pregnancy.
Future studies are needed to determine the reasons for inter-woman variability and
the potential impact of efficacy for recurrent PTB prevention.
