Abstract
Background
Effective contraception is critical to young women with HIV-associated tuberculosis (TB), as unintended pregnancy is associated with increased perinatal morbidity and mortality. The effects of co-administration of efavirenz and rifampicin on the pharmacokinetics of depot medroxyprogesterone acetate (DMPA) are unknown. We hypothesized that clearance of medroxyprogesterone acetate (MPA) would increase when given with rifampicin and efavirenz, thus increasing risk of ovulation.
Methods
This pharmacokinetics (PK) study assessed DMPA among HIV/TB coinfected women on an efavirenz-based antiretroviral treatment and rifampicin-based TB treatment. Plasma MPA concentrations and progesterone were measured predose (MPA only) and 2, 4, 6, 8, 10, and 12 weeks after a single DMPA 150 mg intramuscular injection. The primary outcome measure, MPA concentration (<0.1 ng/mL) at week 12, was assessed using exact 95% Clopper-Pearson confidence intervals. MPA PK parameters were calculated using noncompartmental analysis.
Results
Among 42 PK-evaluable women from 5 African countries, median age was 32 years and median CD4 was 414 cells/mm3. Five women (11.9%; 95% CI, 4.0–25.6%) had MPA <0.1 ng/mL at week 12; of these, one had MPA <0.1 ng/mL at week 10. The median clearance of MPA was 19 681 L/week compared with 12 118 L/week for historical controls. There were no adverse events related to DMPA, and progesterone concentrations were <1 ng/mL for all women for the study duration.
Conclusions
DMPA, when given with rifampicin and efavirenz, was safe. MPA clearance was higher than in women with HIV not on ART, leading to subtherapeutic concentrations of MPA in 12% of women, suggesting that more frequent dosing might be needed.
Clinical Trials Registration
Keywords: HIV, tuberculosis, ifampicin, favirenz, DMPA
Despite few data on pharmacokinetics interactions between depot medroxyprogesterone acetate (DMPA) and efavirenz/rifampicin, DMPA given with efavirenz/rifampicin was safe but 11.9% of women had medroxyprogesterone acetate concentrations <0.1 ng/mL (concentration associated with increased risk of ovulation). More data are required for definitive dosing recommendations.
Globally, tuberculosis (TB) is the leading cause of death from an infectious disease and remains the most common cause of morbidity in people living with human immunodeficiency virus (HIV) in resource-limited countries [1]. In 2017, TB alone was responsible for 300 000 deaths in persons living with HIV, with most of these deaths occurring in sub-Saharan Africa (SSA). The World Health Organization estimates the risk of developing TB in the setting of HIV is 16–27 times that of the general population [2].
Women comprise more than half of the nearly 37 million persons living with HIV and represent a disproportionate percentage of new HIV infections among individuals over 15 years old in SSA [3]. In 2017, 59% of new adult HIV infections were women in SSA, most of whom were of reproductive age [4].
Tuberculosis in pregnancy is associated with poor obstetric and perinatal outcomes, and TB and HIV coinfection in pregnancy is associated with 3 times the risk of maternal or child morbidity and mortality [5–7]. Effective contraception is therefore a critical need in this population. Management of TB during pregnancy can be complex, with concerns about adequate drug exposures and drug toxicities.
Depot medroxyprogesterone acetate (DMPA), an intermediate-acting progesterone-only injectable contraceptive, is one of the most commonly used contraceptives globally, especially in SSA because of its high efficacy and its ease and convenience of administration [8]. DMPA, given as a 150-mg intramuscular (IM) dose, produces medroxyprogesterone acetate (MPA) concentrations that stay above the therapeutic target (>0.1 ng/mL) for approximately 3 months and inhibits ovulation for up to 14 weeks [9, 10]. DMPA is administered every 3 months to ensure continuous contraception and has a perfect-use failure rate of 0.3% during the first year [9].
Rifampicin (RIF) is a critically important drug for the treatment of drug-sensitive TB, while efavirenz (EFV) is recommended for women of childbearing potential as part of a first-line combination antiretroviral therapy (cART) [11] for treatment of HIV in SSA. Rifampicin and EFV are commonly used together in SSA in patients diagnosed with TB and HIV. They are potent inducers of cytochrome (CYP) P450 enzymes, which metabolize MPA [12], and as such, interact with some hormonal contraceptives, leading to reduced contraceptive concentrations and compromised efficacy [13, 14]. Little is known about the use of DMPA in the setting of RIF-containing TB treatment and the commonly used EFV-containing cART regimens. It is anticipated that coadministration with RIF and EFV would enhance clearance of MPA, resulting in lower MPA concentrations, compared with those seen when DMPA is administered alone, potentially putting these women at risk of contraceptive failure. In this study, we explored the safety, pharmacokinetics (PK), and pharmacodynamics (PD) of DMPA over a 12-week period in women coinfected with HIV and TB receiving RIF-containing TB treatment and EFV-containing cART.
METHODS
Study Design
This study (ClinicalTrials.gov registration no: NCT02412436) was a 12-week, phase II, open-label, single-arm, study of steady-state PK interactions among women coinfected with HIV and TB women receiving EFV-based cART and RIF-based anti-TB treatment. To be eligible to enroll, participants had to meet the following criteria: be 18–46 years old, not pregnant, have been receiving EFV plus 2 nucleoside reverse transcriptase inhibitors (NRTIs) for at least 28 days prior to study entry, and receiving RIF and isoniazid (INH) during the continuation phase of TB treatment. Participants who had received DMPA or other injectable contraceptives within 180 days or any other hormonal therapies within 30 days of study entry were excluded. Participants who were taking CYP3A4 inducers or inhibitors within 30 or 7 days, respectively, prior to study entry, were pregnant or breastfeeding, or had a contraindication to DMPA administration were also excluded. Participants provided written informed consent prior to any study evaluations.
Study Procedures
DMPA was administered as a single IM injection (150 mg) at study entry after a negative pregnancy test. Plasma MPA concentrations were measured predose and at 2, 4, 6, 8, 10, and 12 weeks postdose. As participants were considered to be at increased risk of having ovulated if progesterone concentrations were greater than 5 ng/mL on week 2 or later, progesterone concentrations were measured at weeks 2, 4, 6, 8, 10, and 12. Safety labs (blood chemistry and hematology) were collected 4 and 12 weeks after DMPA administration. Plasma HIV-1 RNA was measured at entry and week 12. Adverse events (AEs) were graded using the standardized Division of AIDS (DAIDS) Table for Grading Severity of Adult Adverse Events (http://rsc.techres.com/Document/safetyandpharmacovigilance/Table_for_Grading_Severity_of_Adult_Pediatric_Adverse_Events.pdf, clarification 2009); AEs grade 3 or higher were reportable. Adherence to HIV and TB medications was assessed for all participants at all study visits using an AIDS Clinical Trial Group (ACTG) self-report questionnaire [15].
Laboratory Assays
Laboratory specimens were batched and assayed at the end of the study at the University of Cape Town Pharmacology Specialty Laboratory. The laboratory analyzed MPA and progesterone (P4) plasma concentrations using a validated liquid-liquid extraction method and liquid chromatography–tandem mass spectrometry analysis (LC-MS/MS). MPA and P4 concentrations were determined on an AB Sciex API 5500Q mass spectrometer PA, and progesterone concentrations reported as below the lower limit of quantification (LLQ) of the assay (0.078 ng/mL for both MPA and progesterone) were assigned values of half the assay’s LLQ (ie, 0.039 ng/mL).
The historical controls were taken from Arm A of ACTG study A5093, a PK study of DMPA and selected cART regimens among women with HIV [16, 17]. Arm A was the control group and enrolled women taking either no antiretrovirals (ARVs) or NRTIs only; none were taking TB medication. The A5093 study used the same sampling methodology and similar LC-MS/MS methodology to determine MPA concentrations, but the P4 determination was performed using enzyme-linked immunosorbent assay for A5903 and LC-MS/MS for A5338. Both these studies measured plasma concentrations for MPA.
Statistical Considerations
The 2 primary outcome measures for the study were MPA less than 0.1 ng/mL and progesterone greater than 5 ng/mL at or before 12 weeks. The primary PK analysis was a per-protocol analysis. According to Mishell [10] and Smit et al [18], when MPA falls below 0.1 ng/mL, serum progesterone rises and the probability of ovulation increases once serum progesterone is above 5 ng/mL. Women who modified their TB or HIV treatment regimen, required study-prohibited medication, missed 2 consecutive visits, or missed a week 10 or week 12 visit during the study were excluded from the final analysis.
Sample size was based on the first primary outcome measure: proportion of women with subtherapeutic MPA concentrations (<0.1 ng/mL) at week 12 using a 1-sided binomial test with a significance level (ɑ) of 0.05. The null hypothesis was that the proportion of women with subtherapeutic MPA concentration would be 6% or less. If the true proportion of women with subtherapeutic MPA concentrations at week 12 is at least 21%, then a sample size of 42 women would provide 90% power to differentiate between 21% and 6%. The primary analysis compared the exact 95% Clopper-Pearson confidence interval (CI) with the prespecified 6% boundary. If the lower bound of the CI is greater than 6%, then there is sufficient evidence to conclude that the observed proportion of women with low MPA concentrations is high.
A secondary objective examined MPA PK parameters and compared them with historical controls. The area under the concentration-time curve (AUC) over 12 weeks (AUC0–12wk) was calculated using noncompartmental analysis (NCA). Apparent clearance (CL/F) was calculated as AUC/dose. Trough (Cmin) and maximum (Cmax) concentrations were observed values. The effect of combined EFV and RIF on the PK of MPA was examined by comparing the MPA PK parameters for women in our trial with historical controls using geometric mean ratios and 90% CIs. Fisher’s exact and Wilcoxon tests were used to compare the per-protocol population with historical controls. Correlation between PK parameters and baseline body mass index (BMI) and weight were calculated using Spearman’s correlation coefficient, and nonzero correlation was tested. The safety analysis was an intent-to-treat analysis and included all women who enrolled into the study, regardless of whether they completed the protocol’s requirements. Noncompartmental analysis and statistical analyses were performed using SAS version 9.4 and SAS/STAT version 14.1 (SAS Institute, Inc).
RESULTS
Study Population
The study enrolled 62 women between November 2015 and March 2017: Botswana (n = 7); Zimbabwe (n = 8); Kenya (n = 12); Durban, South Africa (n = 17); and Johannesburg, South Africa (n = 18). Twenty of these women were excluded from the primary PK analysis (Figure 1). Characteristics of the PK-evaluable women were similar to the overall enrolled population with respect to age, race, CD4+ counts, BMI, and HIV RNA.
Figure 1.
Consolidated standards of reporting trials diagram of participant flow. Abbreviations: PK, pharmacokinetics; TB, tuberculosis.
Of the 42 women included in the per-protocol PK analysis, 100% were African, the median age was 32 years (Q1, Q3: 27, 35) and median BMI of 20.4 kg/m2 (quartile [Q]1, Q3: 18.7, 24.4 kg/m2). The median baseline CD4 count was 414 cells/mm3 (Q1, Q3: 226, 638) with 62% of women having a CD4 count of 350 cells/mm3 or higher; 86% of women had a baseline HIV RNA of less than 400 copies/mL (Table 1) and 55% were taking a tenofovir and emtricitabine NRTI combination all were taking RIF and INH (17% were also taking ethambutol). At weeks 10 and 12, all women reported 100% adherence to their HIV and TB medications. The A5093 historical controls were similar to the A5338 per-protocol population with respect to age and BMI. The historical controls were enrolled from US sites and 79% were African-American; viral loads and CD4 counts were higher.
Table 1.
Baseline Characteristics of Study Participants and Historical Controls
| Safety Population (N = 62) | Per-protocol Population (N = 42) | Historical Controls (N = 14) | Per-protocol Population vs Historical Controls, P Value | |
|---|---|---|---|---|
| Age, years | 32 (27, 37) | 32 (27, 35) | 33 (25, 37) | .67 |
| 20–29 | 24 (39) | 17 (40) | 4 (29) | .77 |
| 30–39 | 25 (40) | 19 (45) | 8 (57) | |
| 40–49 | 13 (21) | 6 (14) | 2 (14) | |
| African race or African American | 62 (100) | 42 (100) | 11 (79) | .013 |
| Injection drug use | 0 (0) | 0 (0) | 1 (7) | .003 |
| Weight, kg | 56.0 (49.0, 63.5) | 53.8 (47.8, 61.0) | 67.9 (52.3, 71.7) | .012 |
| Body mass index, kg/m2 | 20.8 (18.9, 36.1) | 20.4 (18.7, 24.4) | 23.3 (19.6, 27.0) | .19 |
| HIV RNA <400 copies/mL | 52 (84) | 36 (86) | 2 (15) | <.001 |
| CD4, cells/mm3 | 407 (209, 644) | 414 (226, 638) | 704 (625, 963) | .001 |
| <50 | 3 (5) | 1 (2) | 0 (0) | .18 |
| 50–199 | 12 (19) | 8 (19) | 0 (0) | |
| 200–349 | 11 (18) | 7 (17) | 1 (7) | |
| ≥350 | 36 (58) | 26 (62) | 13 (93) | |
| Enrollment by country | <.001 | |||
| South Africa | 35 (56) | 20 (48) | 0 (0) | |
| Kenya | 12 (19) | 10 (24) | 0 (0) | |
| Botswana | 7 (11) | 6 (14) | 0 (0) | |
| Zimbabwe | 8 (13) | 6 (14) | 0 (0) | |
| United States | 0 (0) | 0 (0) | 14 (100) |
Data are presented as no. (%) or median (IQR) unless otherwise indicated. P values are from the Wilcoxon test (continuous variables) and Fisher’s exact test (categorical variables). Abbreviations: HIV, human immunodeficiency virus; IQR, interquartile range.
MPA Pharmacokinetics
Week-specific median MPA concentrations ranged from 0.962 (week 2) to 0.386 (week 12) ng/mL. Week-specific minima ranged from 0.283 (week 2) to 0.039 (week 12) (Figure 2 and Table 2). Five women (11.9%; 95% CI, 4.0–25.6%) had MPA concentrations less than 0.1 ng/mL at week 12; of these, one had an MPA concentration of less than 0.1 ng/mL at week 10. Because the lower bound of the CI was 4%, the proportion of women with subtherapeutic MPA concentrations at week 12 was not 6% or higher as hypothesized. Of note, only 1 of 14 (7%) historical controls had an MPA concentration less than 0.1 ng/mL at week 12 (Fisher’s exact P > .99 when comparing with 5 of 42 or 11.9%). Compared with the historical controls, Cmax was lower (geometric mean ratio [GMR], 0.76; 90% CI, 0.55–0.86) and apparent clearance was higher (GMR, 1.48; 1.24–1.78) (Table 3). The AUC for the 12 weeks after DMPA administration (AUC0–12wk) was lower than in the historical controls (GMR, 0.67; 0.56–0.81). MPA AUC0–12wk in our study was negatively correlated with baseline BMI (r = −0.36, P = .020) and baseline weight (r = −0.35, P = .025).
Figure 2.
Medroxyprogesterone acetate concentrations (ng/mL) at weeks 2, 4, 6, 8, 10, and 12 for the per-protocol population and historical controls (boxplots indicating medians, interquartile ranges, and ranges). Abbreviation: MPA, medroxyprogesterone acetate.
Table 2.
MPA Concentrations (ng/mL) at Weeks 2, 4, 6, 8, 10, and 12
| Week 2 (N = 40) | Week 4 (N = 40) | Week 6 (N = 41) | Week 8 (N = 42) | Week 10 (N = 42) | Week 12 (N = 42) | |
|---|---|---|---|---|---|---|
| Mean (SD) | 1.077 (0.587) | 0.987 (0.738) | 0.692 (0.373) | 0.628 (0.301) | 0.461 (0.233) | 0.422 (0.255) |
| Minimum, maximum | 0.283, 2.440 | 0.197, 3.420 | 0.185, 1.690 | 0.181, 1.660 | 0.083, 1.010 | 0.039, 1.110 |
| Median (Q1, Q3) | 0.962 (0.567, 1.475) | 0.673 (0.437, 1.470) | 0.600 (0.442, 0.979) | 0.603 (0.396, 0.747) | 0.444 (0.254, 0.645) | 0.386 (0.234, 0.579) |
| No. (% [95% CI]) <0.1 ng/mL | 0 (0 [0–8.8]) | 0 (0 [0–8.8]) | 0 (0 [0–8.6]) | 0 (0 [0–8.4]) | 1 (2.4 [.1–12.6]) | 5 (11.9 [4.0–25.6]) |
Abbreviations: CI, confidence interval; MPA, medroxyprogesterone acetate; Q, quartile.
Table 3.
MPA Pharmacokinetic Parameters
| Per-protocol Population (N = 42) | Historical Controls (N = 14) | Geometric Mean Ratio (90% CI) | Wilcoxon Rank-sum Test, P Value | |
|---|---|---|---|---|
| Calculated AUC(0–12wk), ng × week/mL | ||||
| Geometric mean (95% CI) | 7.44 (6.41, 8.64) | 11.04 (9.29, 13.13) | 0.67 (.56, .81) | |
| Median (Q1, Q3) | 7.63 (5.39, 11.42) | 12.38 (8.88, 13.88) | … | .004 |
| Mean (SD) | 8.25 (3.57) | 11.48 (3.10) | … | |
| Minimum, maximum | 2.39, 16.23 | 6.03, 15.64 | … | |
| Measured Cmin, ng/mL | ||||
| Geometric mean (95% CI) | 0.28 (0.22, 0.35) | 0.36 (0.23, 0.58) | 0.76 (.49, 1.16) | |
| Median (Q1, Q3) | 0.33 (0.21, 0.49) | 0.43 (0.29, 0.60) | … | .188 |
| Mean (SD) | 0.34 (0.18) | 0.45 (0.24) | … | |
| Minimum, maximum | 0.04, 0.74 | 0.04, 0.89 | … | |
| Measured Cmax, ng/mL | ||||
| Geometric mean (95% CI) | 1.07 (0.88, 1.28) | 1.55 (1.26, 1.91) | 0.69 (.55, .86) | |
| Median (Q1, Q3) | 1.04 (0.67, 1.66) | 1.74 (1.02, 2.09) | … | .023 |
| Mean (SD) | 1.26 (0.72) | 1.64 (0.56) | … | |
| Minimum, maximum | 0.31, 3.42 | 0.88, 2.58 | … | |
| Apparent clearance, L/week | ||||
| Geometric mean (95% CI) | 20 167 (17 367, 23 417) | 13 581 (11 426, 16 144) | 1.48 (1.24. 1.78) | |
| Median (Q1, Q3) | 19 681 (13 139, 27 845) | 12 117 (10 808, 16 891) | … | .004 |
| Mean (SD) | 22 705 (12 139) | 14 193 (4645) | … | |
| Minimum, maximum | 9240, 62 644 | 9588, 24 881 | … | |
Abbreviations: AUC(0–12wk), area under the concentration-time curve over 12 weeks; CI, confidence interval; Cmax, maximum concentration; Cmin, trough concentration; MPA, medroxyprogesterone acetate; SD, standard deviation; Q, quartile.
Progesterone Concentrations
All progesterone concentrations remained below 1 ng/mL throughout the study.
Safety
Of the 62 women enrolled in this study, there were 21 AEs that were grade 3 or higher reported in 8 (13%) women. The majority of the AEs were grade 3 (n = 16 of 21); there were 4 grade 4 AEs. None of the AEs experienced were related to DMPA.
DISCUSSION
This is the first study to evaluate the effects of RIF- and EFV-containing treatment on the PK and PD of MPA in women with HIV-associated TB. This study enrolled an African population from 5 diverse resource-limited countries where there is high usage of DPMA and looked carefully at hormonal levels. Women in our study had reduced AUC, lower Cmax, and higher clearance of MPA than historical controls—women with HIV not receiving EFV or RIF. There were no grade 3 or higher AEs attributed to DMPA for the duration of the study.
Pharmacological interactions between hormonal contraceptives and RIF were first described in the 1970s when unintended pregnancies were reported in women with TB [19]. There are no prior studies documenting the interaction between DMPA and RIF. Pharmacokinetic parameters for DMPA from earlier studies showed that concentrations of MPA typically reach 1 ng/mL and stay above the therapeutic target of 0.1 ng/mL for about 3 months postdose and then decline. Reduced AUC, lower Cmax, and higher clearance in this study as compared with historical data and historical controls used for this study are hypothesized to be indicative of induction of metabolizing enzymes or transporters, leading to lower MPA concentrations and a shorter time to subtherapeutic concentrations. In addition to A5093, a study by Nanda et al [20] also demonstrated higher AUC (10.4 ng × week/mL) in their study population of women with HIV receiving cART (zidovudine, lamivudine, and EFV) than those who were not receiving ART. These 2 studies are the only PK studies that looked at interactions between DMPA and cART, specifically EFV. Most of the data on interactions between ART and DMPA were based on observational or retrospective analyses. Nanda et al’s study did not find significant differences in MPA concentrations between the cART and their control group. The A5093 study also did not find a difference in MPA AUC, Cmax, and clearance between study groups. Although 3 women in the A5093 study had MPA concentrations of less than 0.1 ng/mL (1 in the control arm, 1 in the nelfinavir arm, and 1 in the EFV arm), the authors concluded that no clinically significant changes in MPA concentrations occurred in the A5093 study.
While more rapid clearance of MPA in the presence of RIF-based TB treatment may be of concern, it is also important to consider markers of increased risk of ovulation. Progesterone concentrations above 1 ng/mL were not observed in our study, suggesting that no women ovulated. Earlier studies showed that time required to regain cyclic ovarian function after a single DMPA injection largely depended on the duration of effective circulating MPA concentrations rather than the inability of the hypothalamus, pituitary, or ovary to recover from effects of the drug after its elimination [21]. Thus, follicle maturation resumes at MPA concentrations of approximately 0.5 to 0.25 ng/mL, but ovulation apparently fails to occur as long as serum MPA concentrations exceed 0.1 ng/mL [21]. Based on these prior studies, ovulation may occur when MPA concentrations fall below 0.1 ng/mL [10, 21]. In our study, 5 women had low concentrations of MPA (4 were <0.078 ng/mL and 1 was 0.094 ng/mL) at week 12. One of the women with an MPA concentration less than 0.078 ng/mL at week 12 also had a low MPA concentration at week 10 (0.083 ng/mL). In our historical control population, 1 woman out of 14 had MPA concentrations less than 0.1 ng/mL at week 12, with none having MPA concentrations less than 0.1 ng/mL before week 10.
Participants who had received DMPA in the past 180 days were excluded from participating in this study, and our results were measured after a single injection. Earlier work done on DMPA did not show that repeated injections affect some of the PD measures such as return to fertility, and it is thought that the time to undetectable concentrations of MPA after the last of several injections should be approximately the same as that after a single injection [22]. This assumption is based on the fact that MPA is cleared rapidly from the bloodstream and the prolonged presence in the blood is related to the slow release from the injection site (eg, a 2-phase release). This hypothesis agrees with the clinical finding that the time of resumption of fertility is unrelated to the number of injections the woman has received [23]. There is therefore no reason to suspect that the repeated injections will affect time to concentration of 0.1 ng/mL.
The drug–drug interactions (DDIs) between EFV, RIF, and hormonal contraceptives pose a significant challenge to women desiring effective contraceptives in SSA where choices are already limited. We have shown in our study that the clearance of MPA is swifter in patients receiving RIF and EFV, suggesting that DMPA should be dosed more frequently than every 12 weeks. This finding contributes to a growing body of literature suggesting that DDIs limit the effectiveness of several forms of hormonal contraception: both oral contraceptives and subdermal implants have been shown to be less effective when used concurrently with EFV [13, 24]. There are no data on the use of implants with RIF, but this is likely to result in less effectiveness of implants due to the potency of RIF. Thus, women in SSA living with HIV and being treated for TB have few available and effective contraceptive choices.
Our study has several limitations. We enrolled women from SSA only, whereas our historical controls were enrolled in the United States. Since the populations are not the same, we cannot exclude the presence of other factors that could affect the differences between the 2 populations. However, US patients tend to have a higher BMI, which is associated with lower MPA concentrations, and so one might expect that if there was a bias, it would be toward the null. Different laboratories were used for the analysis of MPA. Although they both used LC-MS/MS, and one can compare the validated concentration ranges, precision, and reproducibility, there could still be important differences. Another limitation was the small sample size. Although 42 PK-evaluable women were sufficient for the assumed proportion of 21%, this number was too small for the observed proportion of 11.9%, leading to a wide CI and, therefore, inconclusive dosing recommendations. It is hoped that more robust results will be obtained from a planned nonlinear mixed-effects PK-modeling study, which will include data from other similar studies of ARVs and DMPA, and may provide more definitive guidance. In addition, because all participants were taking both RIF and EFV, the relative contributions of each to changes in PK parameters could not be ascertained. However, since we expect that EFV would add only modestly to the induction effects of RIF, we anticipate that our study results could be applicable to patients with TB who are taking RIF alone without EFV [25].
In conclusion, in women with HIV-associated TB receiving EFV and RIF, MPA clearance appeared to be more rapid than in historical controls living with HIV but not receiving these medications. The use of DMPA with RIF-based TB treatment and EFV-containing HIV treatment was safe.
While none of the women had an increase in progesterone concentrations, suggesting that ovulation most likely did not occur, 11.9% (5 of 42) of women had MPA concentrations less than 0.1 ng/mL, with one of these women having this low concentration as early as week 10. These low concentrations are concerning and suggest the possibility of contraceptive failure with the current dosing schedule in this population. Given the risks associated with unintended pregnancy in women with TB disease and recognizing the limited alternatives in most settings where TB is common, it seems prudent to consider shortening the DMPA dosing interval, most likely to every 8–10 weeks.
Supplementary Material
Appendix
Study Team Members
| Member First Name | Member Surname | Member Institution | Member Study Role | Member E-mail Address |
|---|---|---|---|---|
| Jhoanna | Roa | ACTG Network Coordinating Center | Clinical Trials Specialist | jroa@s-3.com |
| Laura | Moran | ACTG Network Coordinating Center | Clinical Trials Specialist | lmoran@s-3.com |
| Flavia | Miiro | Joint Clinical Research Center (JCRC)/Kampala CRS | CSS Representative | flatmiiro@gmail.com |
| Catherine | Godfrey | HIV Research Branch, Therapeutics Research Program, DAIDS, NIAID, NIH | DAIDS Medical Officer | adriana.andrade@nih.gov |
| Kristine | Coughlin | ACTG Data Management Center | Data Manager | coughlin@fstrf.org |
| Ian | Mugisa | Joint Clinical Research Center (JCRC)/Kampala CRS | Field Representative | mugisaian@yahoo.co.uk |
| Zephney | Van Der Spuy | Department of Obstetrics and Gynecology, University of Cape Town | Investigator | zephne.vanderspuy@uct.ac.za |
| Cynthia | Firnhaber | University of the Witwatersrand Helen Joseph (WITS HJH) CRS | Investigator | csfirnhaber@gmail.com |
| Kelly | Dooley | Johns Hopkins University CRS | Investigator | kdooley1@jhmi.edu |
| Mark | Lojacono | ACTG Data Management Center | Laboratory Data Manager (LDM) | lojacono@fstrf.org |
| Vandana | Kulkarni | Byramjee Jeejeebhoy Government Medical College CRS | Laboratory Technologist | vandana@jhumitpune.com |
| David | Haas | Vanderbilt Therapeutics (VT) CRS | Pharmacogenomics | david.haas@vanderbilt.edu |
| Helen | McIlleron | Division of Clinical Pharmacology, University of Cape Town | Pharmacologist | helen.mcilleron@uct.ac.za |
| Michelle | Kendall | Statistical and Data Analysis Center | Statistician | kendall@sdac.harvard.edu |
| Yoninah | Cramer | Statistical and Data Analysis Center | Statistician | cramer@sdac.harvard.edu |
| Susan E. | Rosenkranz | Statistical and Data Analysis Center | Statistician | sue@sdac.harvard.edu |
| Xingye (Shirley) | Wu | Statistical and Data Analysis Center | Statistician | xwu@sdac.harvard.edu |
| Rosie | Mngqibisa | Durban International CRS, Enhancing Care Foundation | Study Chair | mngqibisa@ecarefoundation.com |
| Jennifer | Robinson | Johns Hopkins University CRS | Study Vice-Chair | jrobin87@jhmi.edu |
| Susan E. | Cohn | Northwestern University CRS | Study Vice-Chair | susan-cohn@northwestern.edu |
Abbreviations: ACTG, AIDS Clinical Trial Group; CRS, Clinical Research Site; DAIDS, Division of AIDS; NIAID, National Institute of Allergy and Infectious Diseases; NIH, National Institutes of Health.
Notes
Acknowledgments. The authors acknowledge and thank all the participants in ACTG A5338 and all ACTG personnel who made this study possible. The authors gratefully acknowledge the support of the ACTG network and the scientific committees for their contribution during protocol development and study conduct. The authors especially want to acknowledge and thank Susan L. Rosenkranz, Yoninah Cramer, and Kristine Coughlin for their contributions in study design, data monitoring, and interim analysis. The authors also wish to acknowledge the following study team members for their contributions: Gary Maartens, Zephne Van Der Spuy, David Haas, Akbar Shahkolahi, Jhoanna C. Roa, Laura Moran, Michelle Wildman, Ian S. Mugisa, Asuman Ssentamu, Vandana Kulkarni, and Flavia Nakayima Miiro. A sincere thank you to Jennifer Norman and her team at the University of Cape Town Speciality Laboratory for the analysis of MPA and progesterone samples. The authors acknowledge the following sites, site staff, and investigators for performing all the work at their various sites: Sr P. Madlala and M. Chikowore, Durban International Clinical Research Site 11201, UM1 AI069432-08; Drs M. Rassool and T. Mwelase, University of the Witwatersrand Helen Joseph CRS 11101 UM1 AI069463 (Clinical Trials Unit grant number) and AI068636 (ACTG network grant number); and Drs F. A. Mbata, E. Ouma Kisumu, and K. Cain, Kisumu CRS 31460, 5UM1AI068636. Dr U. Chakalisa and M. S. Raesi Gaborone CRS 12701, UM1 AI069456-08; and Prof J. G. Hakim and Dr P. Mukwekwerere, Parirenyatwa CRS 30313 UM1AI069436.
Disclaimer. The findings and conclusions in this manuscript are those of the authors and do not necessarily represent the official position of the funding agencies. The contributions made by C. Godfrey in the writing of this paper were made in her capacity as an National Institutes of Health (NIH) employee, but the views expressed in this paper do not necessarily represent those of the NIH.
Financial support. The research reported in this publication was supported by the National Institute of Allergy and Infectious Diseases (NIAID) of the National Institutes of Health (NIH) under award numbers UM1 AI068634 (Statistical and Data Management Center of the AIDS Clinical Trials Group), UM1 AI068636, UM1 AI69471 and UM1 AI106701. This work was supported in part by the Emory–Centers for Disease Control and Prevention (CDC) HIV/AIDS Clinical Trials Unit award number UM1AI069418 from the NIH (NIAID) and the US CDC (Division of HIV/AIDS Prevention).
Potential conflicts of interest. S. E. C. is a scientific advisor on enrollment of women into trials for Merck, Inc.. The other authors report no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
Presented in part: 26th Conference on Retroviruses and Opportunistic Infections (CROI 2019); Seattle, Washington; 4–7 March 2019. Abstract 78.
References
- 1. World Health Organization. Global tuberculosis report. Secondary global tuberculosis report. 2018. Available at: http://apps.who.int/iris/bitstream/handle/10665/274453/9789241565646-eng.pdf?ua=1. Accessed 21 December 2018. [Google Scholar]
- 2. UNAIDS data 2018. Secondary UNAIDS data 2018. 2018. Available at: http://www.unaids.org/sites/default/files/media_asset/unaids-data-2018_en.pdf. Accessed 21 December 2018. [Google Scholar]
- 3. UN wom en: facts and figures: HIV and AIDS. Secondary UN women: facts and figures: HIV and AIDS July 2018. Available at: http://www.unwomen.org/en/what-we-do/hiv-and-aids/facts-and-figures.12/21/2018. Accessed 21 December 2018. [Google Scholar]
- 4. UNAIDS. Fact sheet, World AIDS Day 2018. Secondary fact sheet, World AIDS Day 2018. Available at: http://www.unaids.org/sites/default/files/media_asset/UNAIDS_FactSheet_en.pdf. Accessed 21 December 2018. [Google Scholar]
- 5. Sobhy S, Babiker Z, Zamora J, Khan KS, Kunst H. Maternal and perinatal mortality and morbidity associated with tuberculosis during pregnancy and the postpartum period: a systematic review and meta-analysis. BJOG 2017; 124:727–33. [DOI] [PubMed] [Google Scholar]
- 6. World Health Organization. TB and women fact sheet. Secondary TB and women fact sheet 2015. Available at: https://www.who.int/tb/publications/tb_women_factsheet_251013.pdf. Accessed 25 January 2019. [Google Scholar]
- 7. Getahun H, Sculier D, Sismanidis C, Grzemska M, Raviglione M. Prevention, diagnosis, and treatment of tuberculosis in children and mothers: evidence for action for maternal, neonatal, and child health services. J Infect Dis 2012; 205(Suppl 2):S216–27. [DOI] [PubMed] [Google Scholar]
- 8. Darroch JE, Singh S. Trends in contraceptive need and use in developing countries in 2003, 2008, and 2012: an analysis of national surveys. Lancet 2013; 381:1756–62. [DOI] [PubMed] [Google Scholar]
- 9. Pfizer. Depo-Provera (depot medroxyprogesterone acetate) package label. New York, NY: Pharmacia & Upjohn Co, Division of Pfizer Inc, 2014. [Google Scholar]
- 10. Mishell DR., Jr Pharmacokinetics of depot medroxyprogesterone acetate contraception. J Reprod Med 1996; 41:381–90. [PubMed] [Google Scholar]
- 11. World Health Organization. Updated recommendations on first-line and second-line antiretroviral regimens and post-exposure prophylaxis and recommendations on early infant diagnosis of HIV: interim guidance. Geneva: World Health Organization, 2018. [Google Scholar]
- 12. Kobayashi K, Mimura N, Fujii H, et al. Role of human cytochrome P450 3A4 in metabolism of medroxyprogesterone acetate. Clin Cancer Res 2000; 6:3297–303. [PubMed] [Google Scholar]
- 13. Scarsi KK, Darin KM, Nakalema S, et al. Unintended pregnancies observed with combined use of the levonorgestrel contraceptive implant and efavirenz-based antiretroviral therapy: a three-arm pharmacokinetic evaluation over 48 weeks. Clin Infect Dis 2016; 62:675–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Patel RC, Onono M, Gandhi M, et al. Pregnancy rates in HIV-positive women using contraceptives and efavirenz-based or nevirapine-based antiretroviral therapy in Kenya: a retrospective cohort study. Lancet HIV 2015; 2:e474–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Chesney MA, Ickovics JR, Chambers DB, et al. Patient Care Committee and Adherence Working Group of the Outcomes Committee of the Adult AIDS Clinical Trials Group (AACTG). Self-reported adherence to antiretroviral medications among participants in HIV clinical trials: the AACTG adherence instruments.. AIDS Care 2000; 12:255–66. [DOI] [PubMed] [Google Scholar]
- 16. Watts DH, Park JG, Cohn SE, et al. Safety and tolerability of depot medroxyprogesterone acetate among HIV-infected women on antiretroviral therapy: ACTG A5093. Contraception 2008; 77:84–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Cohn SE, Park JG, Watts DH, et al. ; ACTG A5093 Protocol Team Depo-medroxyprogesterone in women on antiretroviral therapy: effective contraception and lack of clinically significant interactions. Clin Pharmacol Ther 2007; 81:222–7. [DOI] [PubMed] [Google Scholar]
- 18. Smit J, Botha J, McFadyen L, Beksinska M. Serum medroxyprogesterone acetate levels in new and repeat users of depot medroxyprogesterone acetate at the end of the dosing interval. Contraception 2004; 69:3–7. [DOI] [PubMed] [Google Scholar]
- 19. Long-acting pr ogestogen-only contraception. Drug Therapeutics Bulletin 1996; 34:93–6. [DOI] [PubMed] [Google Scholar]
- 20. Nanda K, Amaral E, Hays M, Viscola MA, Mehta N, Bahamondes L. Pharmacokinetic interactions between depot medroxyprogesterone acetate and combination antiretroviral therapy. Fertil Steril 2008; 90:965–71. [DOI] [PubMed] [Google Scholar]
- 21. Ortiz A, Hirol M, Stanczyk FZ, Goebelsmann U, Mishell DR. Serum medroxyprogesterone acetate (MPA) concentrations and ovarian function following intramuscular injection of depo-MPA. J Clin Endocrinol Metab 1977; 44:32–8. [DOI] [PubMed] [Google Scholar]
- 22. Cornette JC, Kirton KT, Duncan GW. Measurement of medroxyprogesterone acetate (Provera) by radioimmunoassay. J Clin Endocrinol Metab 1971; 33:459–66. [DOI] [PubMed] [Google Scholar]
- 23. Schwallie PC, Assenzo JR. The effect of depo-medroxyprogesterone acetate on pituitary and ovarian function, and the return of fertility following its discontinuation: a review. Contraception 1974; 10:181–202. [DOI] [PubMed] [Google Scholar]
- 24. Patel RC, Morroni C, Scarsi KK, Sripipatana T, Kiarie J, Cohen CR. Concomitant contraceptive implant and efavirenz use in women living with HIV: perspectives on current evidence and policy implications for family planning and HIV treatment guidelines. J Int AIDS Soc 2017; 20:21396. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Habtewold A, Aklillu E, Makonnen E, et al. Long-term effect of rifampicin-based anti-TB regimen coadministration on the pharmacokinetic parameters of efavirenz and 8-hydroxy-efavirenz in Ethiopian patients. J Clin Pharmacol 2016; 56:1538–49. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.