W. P. Martins, C. O. Nastri
Ovarian hyperstimulation syndrome (OHSS) is a much feared and potentially devastating iatrogenic complication of assisted reproductive technology (ART)1. Despite the widespread adoption of preventive strategies against OHSS (most notably the gonadotropin-releasing hormone agonist trigger and the ‘freeze-all’ policy), severe OHSS persists as a life-threatening condition that has not been fully eradicated2. In a nationwide French cohort study, severe OHSS was responsible for nearly 5000 hospital admissions over a 6-year period, accounting for approximately 20 000 days of inpatient care3. While the relative incidence of severe OHSS has decreased, the absolute burden remains significant and the consequences for the affected individuals are profound. Fatalities continue to occur in young (< 30 years of age), otherwise healthy women, including oocyte donors, who succumb to acute massive pulmonary edema or cardiac arrest driven by hemodynamic disruption associated with severe OHSS4. Current management of established severe OHSS is passive, relying on risk reduction or supportive care (i.e. intravenous fluids, anticoagulation, paracentesis) while waiting for the syndrome to resolve spontaneously1, 2. A specific treatment capable of actively terminating the pathophysiology once it has begun is lacking. OHSS is driven fundamentally by the supraphysiological secretion of vascular endothelial growth factor from multiple hyperstimulated corpora lutea (CLs). We propose that the therapeutic strategy should focus not on managing the downstream symptoms (such as ascites and thrombosis), but on eliminating the upstream source of these complications. Herein, we present the induction of rapid, pharmacological luteolysis using prostaglandin F2α (PGF2α) analogs as a potential treatment for OHSS. While this approach is novel in human in-vitro fertilization, it is the standard of care in veterinary medicine, in which PGF2α is used routinely to regress the CL. We propose two potential pharmacological candidates for clinical investigation: cloprostenol and carboprost (Table 1). Cloprostenol is a potent synthetic PGF2α analog used extensively to synchronize estrus in cattle, horses and other mammals (Table S1). It is characterized by high luteolytic potency and a favorable safety profile. Its high specificity for the CL suggests it could be an excellent drug for treating OHSS; indeed, a study in heifers demonstrated that a single low dose of cloprostenol triggers a dramatic decline (> 90%) in serum progesterone concentrations within just 24 h, confirming rapid and complete functional luteolysis5. However, as cloprostenol is not currently approved for human use, its introduction into clinical practice would require completion of the full investigational new drug (IND) process, including Phase I safety trials. 250 μg (1 mL of standard 250 μg/mL solution) by IM injection. Subsequent doses of 250 μg might be administered at 1.5 to 3.5 h intervals. Dose may be increased to 500 μg (2 mL) if uterine contractility is judged to be inadequate after several doses of 250 μg. Optional test dose of 100 μg (0.4 mL) may be administered initially. Conversely, the PGF2α analog carboprost tromethamine (15-methyl PGF2α) has been approved for human use by the United States Food and Drug Administration (FDA) since 1979, as well as by the European Medicines Agency (EMA), for the treatment of postpartum hemorrhage. As such, its pharmacokinetics and safety profile in humans are well established. The standard 250-μg intramuscular dose of carboprost is effective for inducing tetanic uterine contractions but it is associated with significant smooth muscle side effects, including severe nausea, vomiting and diarrhea6. While these side effects are often deemed acceptable in the context of a life-threatening hemorrhage, the use of carboprost in less emergent settings has been discouraged historically. However, recent trials have demonstrated that these adverse side effects can be mitigated significantly using sedatives, such as propofol, remimazolam or sufentanil (Table S2). A key insight driving this proposal is the large disparity in dosing requirements between uterine contraction and luteolysis. The standard 250-μg dose of carboprost used in obstetrics is designed to induce powerful, sustained uterine contractions (approximately 4 μg/kg for a 60-kg woman). However, comparative animal data suggest that the threshold for inducing biochemical luteolysis is a fraction of that. Crucial evidence for a low-dose strategy comes from a study that investigated cloprostenol in donkeys (body weight, 100–140 kg), in which doses between 37.5 μg and 125 μg were administered7. They found that a low dose of 37.5 μg (approximately 0.3 μg/kg) was sufficient to induce luteolysis, significantly reducing the estrous cycle length and CL volume compared with controls. Importantly, adverse side effects such as cramping and sweating were dose-dependent; they were prevalent at higher doses but absent or negligible at the lowest effective luteolytic doses. Therefore, we propose initiating human trials with a 25-μg dose of either carboprost or cloprostenol (equivalent to 0.1 mL of the standard 250 μg/mL solution), administered preferentially via subcutaneous injection. This dose corresponds to approximately 0.4 μg/kg for a 60-kg woman, representing only 10% of the standard obstetric dose of carboprost. For the treatment of severe OHSS, we suggest injection should be repeated every 6–12 h until improvement, using the same dose or escalating dosage by 12.5 μg (not exceeding 100 μg) depending on the side effects observed with the previous dose. This approach allows for safe dose escalation while minimizing adverse side effects. Since PGF2α side effects are strictly dose-dependent, this strategy would be significantly better tolerated while maintaining sufficient potency to trigger regression of the CL. Even if dose escalation is required, we estimate the maximum dose for efficacy would likely be ≤ 100 μg (approximately 1.5 μg/kg for a 60-kg woman), which would still represent only 40% of the current dose of carboprost used routinely in labor wards. For a patient with severe OHSS at risk of thrombosis, renal failure or prolonged hospitalization, it can be considered that transient nausea or cramping caused by a PGF2α analog are acceptable side effects. However, it is imperative to state that PGF2α induces luteolysis and will therefore result in the termination of an established pregnancy. Therefore, this therapy is contraindicated strictly in cycles involving fresh embryo transfer and suggested solely for: ‘freeze-all’ cycles, in which no embryo transfer has occurred and the priority is the rapid resolution of ovarian stimulation; or life-threatening OHSS, which occurs in rare cases of established pregnancy in which the severity of the maternal condition (e.g. progressive organ failure) necessitates termination of the pregnancy to save the life of the mother. Severe OHSS is currently being managed with passive supportive care, while a definitive treatment might exist. The biological rationale for PGF2α-induced luteolysis is sound, and the precedent in animal models is overwhelming. We call for dose-finding trials to determine the minimum effective luteolytic dose of carboprost or cloprostenol in non-pregnant women, as targeted luteolysis represents the logical next step in the evolution of ART safety. Data sharing not applicable to this article as no datasets were generated or analysed during the current study. Table S1 Studies reporting dosage of cloprostenol used to induce luteolysis in different mammals. Table S2 Efficacy of pharmacological interventions in mitigating adverse effects of 250 μg of intramuscular carboprost during Cesarean delivery. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.