Leukine

Emory University is running a Phase 2 test of sargramostim, an already-approved bone marrow stimulant, as a way to regrow blood vessels in legs choked by peripheral artery disease (PAD). NCT03304821 uses the same molecule sold as Leukine for restarting white blood cell production after chemotherapy, but here the goal is different: mobilize progenitor cells that might build collateral arteries around blocked segments and let patients walk further before their calves cramp. The trial enrolled 47 patients against placebo and completed in 2022, with no peer-reviewed publication as of mid-2026. This is an academic repurposing effort against a therapeutic-area backdrop littered with failed angiogenesis programs, and the four-year publication silence is itself a data point.

Sargramostim (brand: Leukine) is a recombinant human GM-CSF that has been FDA-approved since 1991 for neutrophil recovery after chemotherapy, autologous and allogeneic bone marrow transplant, and induction chemotherapy in acute myeloid leukemia [1]. Partner Therapeutics acquired the drug from Sanofi in 2018 and remains the current commercial owner [2]. Partner Therapeutics is not passive with the asset: in 2020 the company received FDA Emergency Use Authorization for sargramostim in hospitalized COVID-19 patients with hypoxemia, and has publicly pursued an indication-expansion strategy across non-oncology settings [7]. That said, in PAD specifically, the compound has no FDA designations, no breakthrough or fast track status, and no orphan tag. NCT03304821 is investigator-initiated, sponsored by Emory University in Atlanta rather than a commercial developer [3]. Per the current registry state, the trial has moved into completed status (primary completion 2022), but no peer-reviewed manuscript, preprint, or major conference abstract has surfaced in the four years since. Any path to labeling for peripheral vascular disease would require confirmatory Phase 3 work, and no such program has been publicly disclosed by Partner Therapeutics. Reading the tea leaves: this is a small mechanistic proof-of-concept whose long publication silence, more than anything else, suggests the results were not compelling enough to pursue aggressively.

GM-CSF is a cytokine that tells bone marrow to produce more granulocytes and macrophages, the shock troops of the immune system. Think of it as a factory manager who ramps up production of white blood cells on demand, which is why it works clinically after chemotherapy wipes out those cells. In PAD, the logic is different. Leg arteries clogged with atherosclerotic plaque starve calf muscle of oxygen, which is why patients feel cramping pain when they walk. The therapeutic bet is that GM-CSF also mobilizes endothelial progenitor cells (EPCs), a rare population of bone marrow cells that can travel to oxygen-starved tissue and help build new small vessels (collaterals) around the blockage [4]. Preclinical work and small human studies have shown that GM-CSF raises circulating EPC counts. Whether those cells actually build durable, functional collateral arteries in older PAD patients is a much harder question. The mechanism is biologically plausible, but the field has been burned repeatedly: VEGF, FGF, and HGF gene therapies for the same indication all showed early angiogenic activity and then missed primary endpoints in Phase 3 [5]. GM-CSF is entering the same graveyard with the same hopeful cell-mobilization story.

NCT03304821 is a Phase 2 randomized placebo-controlled study run out of Emory University, enrolling 47 patients with intermittent claudication and testing subcutaneous sargramostim against placebo [3]. The primary endpoint is peak walking time on a standardized graded treadmill protocol, the field-standard functional readout for PAD trials, with 6-minute walk distance and quality-of-life scales as common secondaries. These endpoints are notoriously placebo-sensitive: unblinded exercise counseling alone can move walking distance by 20 to 30 percent, so any real signal has to clear a high noise floor. An n of 47 is powered to detect large effect sizes only, and even a clean positive at that sample size would need a much larger confirmatory trial to support a label. The study reached its primary completion date in 2022. That the results have not been published in the four years since is a soft negative signal on its own, because positive PAD trials in a small academic setting typically publish within 12 to 24 months of completion. The design is fit for hypothesis testing, not for a label, and no commercial sponsor has committed to fund a confirmatory Phase 3.

Our model estimates a 18% chance this drug is eventually approved. It starts from the historical base rate for Phase 3 drugs in this area (about 57%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by more secondary endpoints than usual; it is held back by the sponsor's thin or weak approval record, weak or limited earlier-phase results, and a randomized design. The other facts land near average for this stage, so they leave the estimate roughly where the base rate put it.

Efficacy risk is the whole ballgame. Every serious therapeutic angiogenesis program in PAD has failed at Phase 3, whether the mechanism was growth factor delivery, gene therapy, or autologous cell transplant [5]. Walking-distance endpoints reward placebo response and supervised exercise, both of which are unavoidable confounders. Safety risk is real but bounded: at cell-mobilizing doses, GM-CSF can trigger neutrophilia (too many circulating white blood cells, which can drive inflammation), capillary leak (fluid escaping from small vessels into tissue, causing swelling and low blood pressure), bone pain, and fever, and there are theoretical concerns about promoting inflammation in already-atherosclerotic vessels in an elderly cardiovascular population. The safety profile is at least well characterized after decades of oncology use, unlike a novel biologic. Execution risk is high because this is a single-site academic study without the trial machinery a commercial sponsor would bring, and the four-year publication delay after primary completion suggests execution or interpretive problems the team has not resolved. Commercial risk is the biggest kill switch: Partner Therapeutics has pursued non-oncology indications for Leukine (notably a COVID-19 EUA in 2020) [7], but has made no public statement committing to a PAD program. Leukine's oncology pricing would face payer pushback for a chronic vascular indication where cilostazol (a generic, inexpensive pill that improves blood flow and walking distance in PAD, and is the guideline-endorsed first-line drug) and supervised exercise are already available and cheap.

Downgrade from watchlist to effectively dormant. NCT03304821 reached primary completion in 2022. Four years on, no peer-reviewed publication, preprint, or major conference abstract has appeared, and no follow-on trial with an industry sponsor has been posted. Small positive PAD trials typically publish within 12 to 24 months of completion, so this silence reads as a soft negative signal on its own. The therapeutic-area context makes the read harsher: multiple better-funded programs with more direct angiogenic mechanisms have already failed at Phase 3, and current-generation PAD cell/gene programs (for example CD34+ autologous cell approaches in the Caladrius/Losordo lineage) have themselves struggled to differentiate from placebo. Specific data that would flip this back to signal: a peer-reviewed readout showing peak walking time improvement of more than 60 seconds versus placebo with a p-value that survives correction, plus a public statement from Partner Therapeutics indicating interest in developing the indication commercially. Absent both, sargramostim in PAD is a stalled academic curiosity rather than a commercial story. If a manuscript or industry-sponsored follow-on trial appears, revisit; otherwise no further review needed.