Periconceptional exposure to lopinavir, but not darunavir, impairs decidualization: a potential mechanism leading to poor birth outcomes in HIV-positive pregnancies

STUDY QUESTION: Does HIV protease inhibitor (PI)-based combination antiretroviral therapy (cART) initiated at periconception affect key events in early pregnancy, i.e. decidualization and spiral artery remodeling? SUMMARY ANSWER: Two PIs, lopinavir and darunavir, currently offered as cART options in HIV-positive pregnancies were evaluated, and we found that lopinavir-based cART, but not darunavir-based cART, impaired uterine decidualization and spiral artery remodeling in both human ex vivo and mouse in vivo experimental models. WHAT IS KNOWN ALREADY: Early initiation of cART is recommended for pregnant women living with HIV. However, poor birth outcomes are frequently observed in HIV-positive pregnancies exposed to PI-based cART, especially when it is initiated prior to conception. The correlation between early initiation of PI-cART and adverse birth outcomes is poorly understood, due to lack of data on the specific effects of PI-cART on the early stages of pregnancy involving uterine decidualization and spiral artery remodeling. STUDY DESIGN, SIZE, DURATION: Lopinavir and darunavir were evaluated in clinically relevant combinations using an ex vivo human first-trimester placenta-decidua explant model, an in vitro human primary decidual cell culture system, and an in vivo mouse pregnancy model. The first-trimester (gestational age, 6–8 weeks) human placenta-decidua tissue was obtained from 11 to 15 healthy women undergoing elective termination of pregnancy. C57Bl/6 female mice (four/treatment group) were administered either lopinavir-cART, darunavir-cART or water by oral gavage once daily starting on the day of plug detection until sacrifice. PARTICIPANTS/MATERIALS, SETTING, METHODS: Human: Spiral artery remodeling was assessed by immunohistochemical analysis of first-trimester placenta-decidua explant co-culture system. Trophoblast migration was measured using a placental explant culture. A primary decidual cell culture was used to evaluate the viability of immune cell populations by flow cytometry. Soluble factors, including biomarkers of decidualization and angiogenesis, were quantified by ELISA and Luminex assay using decidua-conditioned media. Mouse: In the mouse pregnancy model, gestational day 6.5 or 9.5 implantation sites were used to assess decidualization, spiral artery remodeling and uterine natural killer (uNK) cell numbers by immunohistochemistry. Transcription factor STAT3 was assayed by immunohistochemistry in both human decidua and mouse implantation sites. MAIN RESULTS AND THE ROLE OF CHANCE: Lopinavir-cART, but not darunavir-cART, impaired uterine decidualization and spiral artery remodeling in both experimental models. Lopinavir-cART treatment was also associated with selective depletion of uNK cells, reduced trophoblast migration and defective placentation. The lopinavir-associated decidualization defects were attributed to a decrease in expression of transcription factor STAT3, known to regulate decidualization. Our results suggest that periconceptional initiation of lopinavir-cART, but not darunavir-cART, causes defective maturation of the uterine endometrium, leading to impairments in spiral artery remodeling and placentation, thus contributing to the poor birth outcomes. LARGE SCALE DATA: N/A LIMITATIONS, REASONS FOR CAUTION: The human first-trimester placenta/decidua samples could only be obtained from healthy females undergoing elective termination of pregnancy. As biopsy is the only way to obtain first-trimester decidua from pregnant women living with HIV on PI-cART, ethics approval and participant consent are difficult to obtain. Furthermore, our animal model is limited to the study of cART and does not include HIV. HIV infection is also associated with immune dysregulation, inflammation, alterations in angiogenic factors and complement activation, all of which could influence decidual and placental vascular remodeling and modify any cART effects. WIDER IMPLICATIONS OF THE FINDINGS: Our findings provide mechanistic insight with direct clinical implications, rationalizing why the highest adverse birth outcomes are reported in HIV-positive pregnancies exposed to lopinavir-cART from conception. We demonstrate that dysregulation of decidualization is the mechanism through which lopinavir-cART, but not darunavir-cART, use in early pregnancy leads to poor birth outcomes. Although lopinavir is no longer a first-line regimen in pregnancy, it remains an alternate regimen and is often the only PI available in low resource settings. Our results highlight the need for reconsidering current guidelines recommending lopinavir use in pregnancy and indicate that lopinavir should be avoided especially in the first trimester, whereas darunavir is safe to use and should be the preferred PI in pregnancy. Further, in current times of the COVID-19 pandemic, lopinavir is among the top drug candidates which are being repurposed for inclusion in clinical trials world-over, to assess their therapeutic potential against the dangerous respiratory disease. Current trials are also testing the efficacy of lopinavir given prophylactically to protect health care workers and people with potential exposures. Given the current extraordinary numbers, these might include women with early pregnancies, who may or may not be cognizant of their gestational status. This is a matter of concern as it could mean that women with early pregnancies might be exposed to this drug, which can cause decidualization defects. Our findings provide evidence of safety concerns surrounding lopinavir use in pregnancy, that women of reproductive age considering participation in such trials should be made aware of, so they can make a fully informed decision. STUDY FUNDING/COMPETING INTEREST(S): This work was supported by funding from the Canadian Institutes of Health Research (CIHR) (PJT-148684 and MOP-130398 to L.S.). C.D. received support from CIHR Foundation (FDN143262 to Stephen Lye). S.K. received a TGHRI postdoctoral fellowship. The authors declare that there are no conflicts of interest. L.S. reports personal fees from ViiV Healthcare for participation in a Women and Transgender Think Tank.