PRL3-zumab

PRL3-zumab is a first-in-class humanized antibody from Singapore's A*STAR spin-off Intra-IMMUSG targeting PRL-3, a phosphatase that helps cancer cells move and metastasize. It is in multiple Phase 2 trials for advanced solid tumors, mainly gastric cancer and hepatocellular carcinoma (HCC), with the first basket trial (a study that tests one drug across multiple tumor types simultaneously, NCT04118114) completed at n=14 and three larger Phase 2 studies now active [1][4][5][6]. The interesting wrinkle: PRL-3 normally sits inside cells where antibodies can't reach, but on tumor cells it flips to the outer surface, which is the whole therapeutic premise [1][3].

Novel compound, no approvals anywhere. The drug cleared a Phase 1 first-in-human study in advanced refractory solid tumors and heme malignancies published in Targeted Oncology in 2023, where it was tolerated with no dose-limiting toxicity through escalation [3]. Phase 2 work is now spread across at least three active trials: NCT04452955 (n=51, active, not recruiting), NCT07541001 (n=60, active, not recruiting), and NCT07290088 (a Phase 2/3 in advanced solid tumors, n=52, recruiting), plus a recently disclosed Phase 1 in neovascular AMD (NCT07547228) that takes the same antibody into a non-oncology indication [5][6][7][8]. The original NCT04118114 basket study at National Cancer Centre Singapore is listed as COMPLETED, though whether its efficacy results have been formally published beyond the design discussion in the 2025 Cell Reports Medicine paper is not clear from public abstracts [1][4]. No FDA breakthrough, fast track, orphan, or RMAT designation has been reported. Next meaningful data is likely a pooled progression-free survival / objective response rate (ORR, the fraction of patients whose tumors shrink by a predefined amount, typically at least 30%) update from the active Phase 2s. ASCO 2026 (May 30 to June 3) has already passed at the time of writing; the next live conference catalysts are ESMO 2026 (October) and ASCO 2027. Regulatory submission is not in view; this is still signal-finding, not registration.

PRL-3 (gene PTP4A3) is a tyrosine phosphatase, an enzyme that strips phosphate tags off other proteins to flip cellular signals on or off. In normal cells it is anchored to the inner leaflet of the plasma membrane via farnesylation at its C-terminal CAAX motif (a lipid tag that pins the protein to the cytoplasmic face of the membrane), with secondary localization to the endoplasmic reticulum and late endosomes / multivesicular bodies. From there it helps push cells through the G1-to-S transition (the commit-to-divide step) and drives cell movement and invasion [UniProt O75365]. In many cancers, especially gastric, HCC, colorectal, and a swath of pediatric solid tumors, PRL-3 is dramatically overexpressed and correlates with metastasis and worse survival [2]. The unconventional part: in cancer cells, some PRL-3 flips from the inner face of the membrane to the outer face, where an antibody can actually grab it. Normal cells largely don't show this surface exposure, which is the selectivity story. PRL3-zumab binds that surface PRL-3 and recruits immune cells, mainly NK cells and macrophages, to kill the tumor through antibody-dependent cellular cytotoxicity (ADCC) [1][3]. Validation is mixed. Genetics and expression data are strong: PRL-3 is consistently elevated in metastatic disease across tumor types [2]. Animal models show tumor regression. But no other drug against PRL-3 has reached approval, so the target itself is unproven in humans. The whole premise also rests on the inner-to-outer flip being real, reproducible, and broad enough across patients to matter clinically.

The active Phase 2 program is a constellation of small, single-arm, sponsor-funded studies rather than one big registrational trial. NCT04452955 enrolls 51 patients with solid tumors with PFS as the primary endpoint [5]. NCT07541001 expands that to 60 patients, also PFS-primary [6]. NCT07290088 is labeled Phase 2/3 but enrolls only 52, again PFS-primary and currently recruiting [7]. The completed Phase 2 at National Cancer Centre Singapore (NCT04118114) ran with just 14 evaluable patients and used ORR as primary [4]. None of these trials have an active comparator arm. None publicly require PRL-3 expression for enrollment, which is a notable gap given the mechanism depends on surface-exposed target. Sample sizes in the 50 to 60 range are signal-finding, not powered for a registrational PFS hazard ratio (a statistical measure of how much faster events such as disease progression or death occur in one arm versus another; a hazard ratio of 0.7 means a 30% reduction in event rate). So the Phase 2/3 label on NCT07290088 should be read with care. The basket design described in the 2025 Cell Reports Medicine paper is a single-dose-level, multi-tumor format, which is efficient for finding which tumor types respond but does not generate the kind of head-to-head data regulators want for approval [1]. Sponsor across the newer trials is Intra-IMMUSG Pte Ltd, a small Singapore biotech, not a major pharma.

The model gives this drug a 4% chance of eventually being approved. That figure starts from the historical approval rate for Phase 2 drugs in this area, which is about 13%, then adjusts based on ten specific facts about the trial and its sponsor. The biggest factors pulling the number down are the sponsor's weak approval track record, enrollment smaller than typical for this phase, and few secondary endpoints; a non-randomized trial design pushes it slightly up. The remaining facts are close to average for this stage, so they leave the final estimate near where the base rate set it.

Efficacy risk is the headline. The first Phase 2 closed at n=14, the active ones are 50 to 60 patients each, and none publicly select patients by PRL-3 expression even though the entire mechanism depends on surface-exposed PRL-3 being present on the patient's tumor cells. If surface exposure is rare or heterogeneous, the all-comer ORR will look flat and the drug will fail not because the science is wrong but because the population was wrong [1][3]. Safety risk looks lower so far. The first-in-human Phase 1 reported no dose-limiting toxicity through escalation, which is unusual and a real positive [3]. ADCC-driven antibodies typically carry infusion reactions and cytokine release, neither of which has been flagged in publications to date. Execution risk is significant. Intra-IMMUSG is a small Singapore biotech and the older trial sponsor is National Cancer Centre Singapore, so the program is academic-adjacent rather than pharma-led. There is no announced ex-Asia partner, no Phase 3 protocol, and capital for a global registrational trial is not visible. Public disclosures on Intra-IMMUSG's funding round size and runway are limited, which is itself a flag for a private biotech running multiple parallel Phase 2s. Regulatory pathway risk is asymmetric. The most plausible first approval is via Singapore's Health Sciences Authority (HSA) using local Phase 2 data, which would unlock a small commercial market but not US or EU revenue. An FDA path realistically requires either a partner-funded global Phase 3 or accelerated-approval traction through a biomarker-defined subset. Commercial risk follows from that. Even with a positive Phase 2 signal, PRL3-zumab needs a partner with oncology commercial infrastructure to reach US/EU patients, and pricing for a first-in-class antibody in second- or third-line gastric/HCC is hard to defend against generic chemotherapy unless the survival delta is large.

Worth watching, but not actionable yet. The signal that would matter is a PFS or ORR readout from NCT04452955 (n=51) or NCT07541001 (n=60) that beats historical controls in a defined indication, ideally HCC or gastric, with biomarker-enriched subset data showing the surface-PRL-3 hypothesis is real [5][6]. The competitive bar is real. In second-line HCC, PRL3-zumab would have to beat or combine with atezolizumab plus bevacizumab and lenvatinib (tremelimumab plus durvalumab is also approved first-line). In gastric cancer it competes against trastuzumab (HER2-positive subset), ramucirumab, and pembrolizumab plus chemotherapy. HCC affects roughly 900,000 new patients annually worldwide and is the third-leading cause of cancer death globally; gastric cancer hits about 1 million new patients per year, with high unmet need in second- and third-line settings. Both indications are large enough to support a first-in-class antibody if the survival delta is meaningful. A weaker but still interesting signal would be a major-pharma licensing deal, which is how this kind of academic-spinoff antibody usually gets validated commercially. The AMD expansion (NCT07547228) is plausible on mechanism rather than a random tangent. PRL-3 has been linked to VEGF signaling and pathological angiogenesis, the same pathway targeted by anti-VEGF agents already approved in neovascular AMD (ranibizumab, aflibercept). If surface-PRL-3 exposure occurs on tumor or pathological vasculature, the antibody might disrupt aberrant vessel growth, though this remains speculative and the trial is Phase 1 at n=15. AMD would be a separate regulatory pathway from the oncology program and could carry its own orphan or rare-disease optionality. Check back at ESMO 2026 (October) and ASCO 2027 for abstract drops from the Singapore group; we have not surfaced an ASCO 2026 abstract that materially changes the picture. The pediatric angle is a sleeper. The 2023 Molecular Therapy Oncolytics paper made the case for PRL-3 expression across pediatric solid tumors and laid the groundwork for first-in-child use, and orphan or rare-pediatric-disease designations could give Intra-IMMUSG a faster regulatory path and a priority review voucher; recent PRV transactions have cleared in roughly the $100M to $350M range, with rare pediatric disease vouchers tending toward the higher end [2]. Until there is either a real efficacy signal or a partner, this stays a first-in-class biology bet on an unvalidated target run by a small sponsor with limited public funding disclosure. The 11.1% PoS reflects that honestly.