P-503 Immunological therapies in women undergoing assisted conception: a systematic review and meta-analysis

Abstract

Abstract Study question What is the effectiveness and safety of immunological therapies in women undergoing assisted conception? Summary answer The evidence on the overall effectiveness and safety of immunomodulators was scarce and uncertain, although some drugs may lead to improved outcomes in selected women. What is known already Embryo implantation is a pre-requisite for pregnancy success. It requires a state of immune tolerance within the maternal endometrium permitting the attachment and subsequent invasion into the decidua of the semi-allogenic blastocyst. Mechanistic data suggesting that heightened immune responses may impair implantation have led to the widespread use of immunomodulatory therapies in clinical practice. These include drugs such as aspirin, heparin, corticosteroids, intralipid and intravenous immunoglobulin (IVIG). However, despite the common use of immunomodulatory drugs in women having in vitro fertilisation (IVF), there is no consensus on their effectiveness and safety, owing to a paucity of well-designed interventional studies. Study design, size, duration We searched MEDLINE, EMBASE, PubMed and CENTRAL until September 2022. We selected randomised controlled trials (RCTs) investigating immunological therapies in women undergoing IVF, including aspirin, heparin, corticosteroids, granulocyte-colony stimulating factor (G-CSF), intralipid, intravenous immunoglobulin (IVIG), tumour necrosis factor-alpha inhibitors and peripheral blood mononuclear cells (PBMCs), evaluated alone or in combination, versus no intervention, placebo or any other immunomodulator(s). The primary outcomes were the rates of live birth or ongoing pregnancy (LBR/OPR) and miscarriage per participant. Participants/materials, setting, methods Two reviewers independently selected studies and extracted data. We conducted pairwise meta-analyses according to participants’ phenotypes as reported by the trialists, including: good prognosis, previous implantation failure, autoimmunity, high inflammation, thin endometrium, and low ovarian reserve. Using the Cochrane-RoB-1 tool, we restricted our primary analyses to RCTs at low risk of selection and other biases. We presented effect estimates as risk ratio (RR) with 95% confidence interval (CI) and considered I2>50% as representing substantial heterogeneity. Main results and the role of chance Our searches identified 14,893 records. We included a total of 84 studies in the qualitative synthesis, of which 43 contributed to the meta-analyses, evaluating in total 7,301 participants and 12 treatment comparisons. The primary analysis showed that aspirin resulted in little to no difference in LBR (RR 0.97, 95% CI 0.94-1.23; n = 688 women; moderate-certainty evidence) and MR (RR 1.20, 95% CI 0.73-1.99; n = 988 women; low-certainty evidence) compared with placebo. The data did not identify a difference in LBR/OPR in women using subcutaneous heparin (RR 1.30, 95% CI 0.80-2.12; n = 150 women; very low-certainty evidence) or IVIG (RR 1.28, 95% CI 0.32-5.16; n = 51 women; very low-certainty evidence) versus placebo. For the remaining interventions, the primary analysis was uncertain of an effect or insufficient for quantitative synthesis. However, the sensitivity analysis including all studies, irrespective of risk of bias, showed that corticosteroids may improve LBR in women with thyroid autoimmunity (RR 2.33, 95% CI 1.04-5.25; n = 60 women; very low-certainty evidence); intrauterine G-CSF may improve LBR in women with a thin endometrium (RR 2.57, 95% CI 1.24-5.29; n = 304 women; very low-certainty evidence); and intrauterine PBMCs may improve LBR in women with previous implantation failure (RR 2.03, 95% CI 1.33-3.10; n = 312 women; very low-certainty evidence). Limitations, reasons for caution For most of the interventions under study, the evidence was of low or very-low certainty, owing to limitations in study design, imprecision, indirectness and inconsistency. Further, very few studies reported on adverse events, highlighting a worrying lack of data on the safety of immunological therapies for women undergoing assisted conception. Wider implications of the findings A scarcity of well-designed RCTs limits the evidence on immunomodulators in women undergoing IVF. However, we identified positive signals for some therapies in specific conditions (e.g. corticosteroids for thyroid autoimmunity, intrauterine G-CSF for thin endometrium, and PBMCs in women with previous implantation failure). This merits further investigation in high-quality trials. Trial registration number PROSPERO registration number CRD42021294031