Two New Endometriosis Drugs. No US Trial

Both came out of Swedish labs. One is running across Europe, the other in China, and the American count for these six conditions went from four to zero.

Two drugs are in testing that would treat endometriosis without shutting down a woman's estrogen.

No approved treatment does that. Both came out of Swedish laboratories.

Stockholm kept one and is running it across Europe. The other was licensed out of Bayer in 2019, worldwide and exclusively, to a company in Nanjing. Its Phase 3 has no Western site.

Neither is being developed by an American company.

Every drug trial has to be registered on a public database called ClinicalTrials.gov, and each entry names who is paying for it. Pull every mid and late-stage trial opened in the past three years across six conditions, endometriosis, uterine fibroids, menopause, polycystic ovary syndrome, preeclampsia and postpartum haemorrhage, and sort them by sponsor.

45 trials started in 2024. 38 in 2025. 36 so far in 2026. The totals barely moved. The sponsors did.

Company-sponsored trials went from eight to twelve to six, and the American share of those went from one to four to none.

This year is two-thirds elapsed and trials opening in late 2026 have not all registered, so six is partial and will rise. Nothing about waiting turns a zero into a one.

Registration also varies by country, and Chinese trials file domestically on ChiCTR as well, so the international registry undercounts them rather than the reverse. The count covers interventional studies only, which leaves out observational research. Six conditions leaves out oncology, fertility and contraception.

Each of those limits makes the American count look better than it is, not worse.

One definition matters here. Sponsor means the lead sponsor named on the registry entry, the party running and paying for the trial. A company can appear on a study as a collaborator without holding that role, and collaborator fields were not audited.

The six carrying a company sponsor this year are Hope Medicine of Nanjing, Shenzhen Kangzhe Biotechnology, JW Pharmaceutical in Seoul, ResQ Medical in Israel, Insud Pharma in Spain and Zentiva in Czechia.

Pfizer was the American one in 2024, running a CDK4 inhibitor with letrozole in postmenopausal breast cancer. The four in 2025 were Tioga Pharmaceuticals in San Diego, testing a kappa agonist in vasomotor symptoms, Graminex in Michigan, Granata Bio with 659 patients across 16 sites on a menopausal gonadotropin, and Astellas Pharma Global Development, the Illinois-registered arm of a Japanese parent, which is a judgment call and counted here as one of the four.

Sort those five by condition and the zero stops looking sudden. Four of them are menopause or postmenopausal. The fifth is an assisted reproduction cycle. Across endometriosis, uterine fibroids, polycystic ovary syndrome, preeclampsia and postpartum haemorrhage, American companies sponsored nothing in 2024, nothing in 2025 and nothing this year.

Menopause is the one category in this set with an approved non-hormonal drug behind it and an endpoint the FDA has already accepted. Five trials is too few to establish why the money went there, and market size is the obvious competing explanation. But the conditions with no American sponsor share something the trial designs show plainly. Their primary endpoint is a pain score the patient reports on herself, which is harder to power a study against than a hot flash somebody counts.

When that wave finished registering, there was nothing behind it.

Every approved endometriosis drug works by shutting estrogen down. That is why the GnRH antagonists are paired with add-back therapy, and why Sumitomo Pharma is running a dedicated 1,000-patient study on what relugolix does to bone density. A drug that treats the pain without touching the hormones does not compete on price. It competes on the reason women stop taking the alternative.

Vipoglanstat is the first of the two. It is an oral inhibitor of mPGES-1, the enzyme that produces prostaglandin E2, and the research began at Karolinska Institutet. Gesynta Pharma of Stockholm is running it in the NOVA trial, 190 patients across European sites including the UK, four menstrual cycles, with dysmenorrhoea, dyspareunia and opioid rescue use among the secondary endpoints and MRI lesion change as an exploratory measure.

Christian Becker, Professor of Reproductive Sciences and co-director of the Oxford Endometriosis Care Centre, is international coordinating investigator. Half the patients were randomised as of 17 June 2026. Top-line results are expected in 2027.

In a Human Reproduction abstract presented in June 2025, the Gesynta team reported that the drug shut down PGE2 while leaving prostacyclin, which protects blood vessels, largely alone. They frame that separation as what sets it apart from coxibs and NSAIDs. It is a cardiovascular safety argument, and cardiovascular safety is what decides whether a drug can be taken for years.

One discrepancy is worth recording. ClinicalTrials.gov lists the NOVA start date as 15 October 2025. Gesynta announced first patient dosed on 11 March 2026. The registry field is study start rather than first dosing, which explains it, but anyone counting trials by start date is counting a study that had not yet dosed a patient.

HMI-115 is the second. It is a human monoclonal antibody that blocks the prolactin receptor, discovered at BioInvent International in Lund using the company's n-CoDeR phage display platform under a 2008 collaboration with Bayer. Bayer carried it as BAY 1158061. In April 2019, Bayer licensed it worldwide and exclusively to Hope Medicine.

The Phase 2 ran 142 women across the United States, Poland and China. In the 240mg arm, mean dysmenorrhoea pain fell 42% from baseline. Mean non-menstrual pelvic pain fell 50%. Both were significant against placebo. Bone mineral density did not move. Neither did estradiol, LH, FSH or progesterone. The results were published in November 2025 in The Lancet Obstetrics, Gynaecology & Women's Health by a team from Peking University, Peking Union Medical College Hospital and Shandong University.

Jeffrey Jensen, Leon Speroff Professor of Obstetrics and Gynecology at Oregon Health & Science University, commented on the result. He said the antibody works without interrupting the menstrual cycle or dropping hormone levels, which is what spares patients the hot flashes and bone thinning that come with GnRH blockers. He also noted that it does not act as a contraceptive, leaving it usable by women who are trying to get pregnant.

The FDA granted HMI-115 Fast Track Designation on 4 December 2025. China's Center for Drug Evaluation had already listed it as a Breakthrough Therapy.

The Phase 3 opened on 29 January 2026 with 540 patients across five sites. Peking Union Medical College Hospital in Beijing. Nanjing Women and Children's Healthcare Hospital. The Second Affiliated Hospital of Soochow University. The International Peace Maternity and Child Health Hospital in Shanghai. Tianjin Medical University General Hospital.

The Phase 2 had sites in the United States and Poland. The Phase 3, carrying American regulatory support, has none.

BioInvent booked a €1 million milestone on 26 March 2026 when the trial opened. That is what the discovering company earns from a Phase 3 of its own antibody.

The licence Bayer granted covers every market, not only China. If HMI-115 is approved in the United States, it will be approved as a Hope Medicine drug.

Chinese sponsors have not run a programme like this before. Narrow the same registry to Phase 3 trials in endometriosis and uterine fibroids started between 2019 and 2023, and 34 studies come back. Qilu Pharmaceutical ran elagolix. Nanjing Chia-tai Tianqing ran elagolix, the same compound AbbVie was running in an 800-patient Phase 3 across 179 sites. Bio Genuine of Shanghai ran BG2109 and Jiangsu Hengrui ran SHR7280, both GnRH antagonists in the same class as relugolix and linzagolix. Qilu returned in January 2024 with a Phase 3 of relugolix itself. Across five years, every Chinese Phase 3 in those two conditions was the same molecule as a Western drug or the same class.

The Korean trial in this year's set is that older pattern. JW Pharmaceutical's KLH-2109 is linzagolix, discovered by Kissei in Japan.

Kissei licensed Korean rights to JW in June 2024, having already licensed the compound to ObsEva for markets outside Asia and to Theramex for Europe, where it launched in Germany as Yselty. Kissei terminated its Chinese licence in September 2024 and started its own 288-patient Phase 3 in endometriosis in June 2025. JW is running a national registration trial inside a Japanese programme.

So six sponsors is not six programmes. It is one first-in-class Phase 3 on a licensed European antibody, one national registration trial, and four smaller studies in postpartum haemorrhage, fibroids and post-menopausal hypertension.

The other thirty trials this year, 83% of the set, are run by universities, hospitals and government bodies. That is not the weak part. The London School of Hygiene and Tropical Medicine is running 30,000 patients on tranexamic acid routes for postpartum haemorrhage. Vall d'Hebron is running 15,160 on aspirin timing in preeclampsia. The University of Calgary is running 8,805 on postpartum thromboembolism. George Washington University is running 4,120 on blood pressure targets in hypertensive pregnancy.

The University of Pennsylvania is running 1,150 on aspirin dose. Those trials change practice at a scale no company trial reaches.

What they produce is guidelines, and guidelines have no acquirer.

Vipoglanstat reads out next year. HMI-115 finishes in March 2028.

Neither one is an American drug.

The Luteal covers the business, science and policy of women's health. Nothing here is medical advice.