certepetide

Certepetide (LSTA1, formerly CEND-1) is a cyclic peptide developed by Lisata Therapeutics (Nasdaq: LSTA) to enhance chemotherapy delivery into stroma-dense solid tumors, principally pancreatic ductal adenocarcinoma (PDAC, the dominant histology of pancreatic cancer) [1]. Mechanism: certepetide binds neuropilin-1 (NRP1) and triggers the CendR transcytosis pathway - active vesicle-mediated ferrying of molecules across endothelial cell walls - briefly opening leaky vessels and fibrotic stroma so any co-infused drug penetrates several-fold deeper [1]. The AGITG-led randomized Phase 2b ASCEND trial (NCT05042128) read out positively at ESMO GI Barcelona in July 2025: objective response rate (ORR, fraction of patients whose tumors shrink ≥30%) of 45.2% versus 19% placebo, and median progression-free survival (PFS) of 7.5 vs 4.7 months (HR 0.60, 95% CI 0.34-1.07, P=0.08), added to gemcitabine plus nab-paclitaxel in first-line metastatic PDAC [2]. Everything around it has decayed: the Lisata-sponsored BOLSTER Phase 2 in cholangiocarcinoma (bile duct cancer) was wound down for capital allocation reasons in 2025 [3][5]; the Qilu Pharmaceutical China license (NCT06261359) was mutually terminated on January 27, 2026 [10]; and on March 6, 2026 Lisata signed a definitive merger to be acquired by Kuva Labs at $5.00/share cash plus two $1.00 CVRs tied to China rights reversion and any future certepetide NDA filing [11]. The equity story is essentially closed; the live question is whether Kuva - which originally licensed certepetide as a targeting/delivery agent for its NanoMark MR imaging platform - funds a confirmatory PDAC Phase 3 or repurposes the asset for imaging.

Novel compound. Never approved anywhere. The program is in transition between sponsors. Strongest data point: ASCEND (NCT05042128), the 158-patient AGITG-led randomized double-blind Phase 2b in first-line metastatic PDAC. Cohort B results presented at ESMO GI Barcelona July 2025 showed ORR 45.2% (certepetide arm) vs 19% (placebo), 6-month PFS 60.8% vs 25%, and median PFS 7.5 vs 4.7 months (HR 0.60, 95% CI 0.34-1.07, P=0.08) [2]. The Phase 1b precursor (NCT03517176) had shown ORR of 61.5% at the 3.2 mg/kg dose, which is what built the original bull case [9]. Discontinued or terminated: BOLSTER (NCT05712356), the Lisata-sponsored Phase 2 in cholangiocarcinoma, was wound down with enrollment capped at approximately 20 patients in the second-line cohort. Lisata's disclosed rationale was capital allocation and accelerated data readout - a financial/strategic decision, not a futility analysis or safety signal [3][5]. That distinction matters: investors should not treat the BOLSTER stop as scientific repudiation. The Qilu Pharmaceutical partnership (NCT06261359, Phase 2 in first-line PDAC in China, n=120) was mutually terminated on January 27, 2026, returning Greater China rights to Lisata [10]. On March 6, 2026, Lisata entered a definitive merger agreement with Kuva Labs: $5.00/share cash plus two non-tradeable CVRs of $1.00 each, payable on (i) China rights reversion to Lisata and (ii) NDA filing for certepetide in any indication/jurisdiction; expected to close Q2 2026 [11]. Surviving under investigator-initiated or academic sponsorship: NCT05121038, a 50-patient Phase 1 of CEND-1 with neoadjuvant FOLFIRINOX (a four-agent chemotherapy regimen) ± panitumumab in resectable PDAC [6] - note the panitumumab arm is unusual since panitumumab is an anti-EGFR antibody and ~90% of PDAC carries KRAS mutations that render EGFR inhibition largely ineffective; the trial does not publicly disclose KRAS wild-type selection, which is a design concern worth flagging. And LSTA1-GBM-2A, an Estonia-based randomized Phase 2a in newly diagnosed glioblastoma added to temozolomide [1]. No FDA breakthrough, fast-track, RMAT, or orphan designations have been publicly disclosed for the program.

This is the most interesting part of the story. Solid tumors are notoriously bad targets for drug delivery: they build dense fibrotic stroma, their blood vessels are leaky in some places and clogged in others, and the resulting high interstitial pressure inside the tumor pushes drugs back out. Pancreatic and biliary cancers are the worst offenders. Certepetide is designed to override that. It's a cyclic peptide modeled on iRGD, a sequence that binds tumor-expressed integrins. After binding, tumor-surface proteases cleave the peptide and expose a C-terminal arginine. That exposed end docks into neuropilin-1 (NRP1, a co-receptor on endothelial and tumor cells) and triggers the CendR pathway - short for C-end Rule - which briefly opens a transcytosis route (active vesicle-mediated transport of molecules through cells rather than between them) across vessel walls and into stroma [1]. In preclinical models, any chemo or antibody co-infused with certepetide gets delivered several-fold deeper into tumor than it would alone. Certepetide doesn't kill cancer cells itself; it just opens the door for whatever drug rides along. The glioblastoma extension is mechanistically coherent rather than scope creep: NRP1 is overexpressed on GBM tumor vasculature, and preclinical work has shown the same CendR pathway enhances drug penetration across the abnormal blood-brain barrier in GBM models - that's the explicit rationale for adding certepetide to temozolomide in LSTA1-GBM-2A [1]. Mechanistic validation is real but incomplete. The CendR pathway is published and reproduced in academic labs. ASCEND now provides the first randomized clinical evidence that the combination beats chemo alone in PDAC on ORR and PFS [2]. What's still missing: confirmatory Phase 3, and a biomarker to identify which patients benefit most.

ASCEND (NCT05042128) is now the central data set, not a forward-looking study: 158-patient AGITG-led randomized double-blind Phase 2b, gemcitabine/nab-paclitaxel ± certepetide in first-line metastatic PDAC, with PFS as the Cohort B primary endpoint [2]. PFS HR 0.60 with P=0.08 is directionally positive but does not formally clear statistical significance at the 0.05 threshold - for a Phase 2b, this is treated as supportive but not confirmatory. ORR 45.2% vs 19% is the cleaner signal. BOLSTER (NCT05712356) in cholangiocarcinoma was the other near-term randomized readout under Lisata sponsorship; it was wound down for capital reasons before generating a definitive efficacy answer [3][5]. The Qilu Phase 2 in first-line PDAC (NCT06261359, n=120, ORR primary) was mutually terminated when the license was returned in January 2026 [10]. ORR alone in PDAC is a known weak surrogate for overall survival (OS), so its loss does not significantly degrade the asset's evidence base. Surviving investigator-initiated work: NCT05121038 (FOLFIRINOX ± panitumumab in resectable PDAC, n=50, active not recruiting) [6] - the panitumumab arm requires explanation given PDAC's KRAS-mutant biology, and none is publicly disclosed. LSTA1-GBM-2A in newly diagnosed glioblastoma is signal-finding, not registrational [1]. There is currently no announced confirmatory Phase 3 path; whether Kuva funds one post-close is the next material question.

Our model estimates a 1% chance this drug is eventually approved. That starting point comes from the historical approval rate for Phase 2 drugs in this area, which is about 13%. The estimate is then adjusted using ten facts about the trial and sponsor - in this case, heavier-than-usual blinding, a thin or weak sponsor approval record, few secondary endpoints, and weak or limited earlier-phase results all pull the number down. The remaining facts are close to average for this stage, so they leave the estimate roughly where it landed after those penalties.

Efficacy. ASCEND is positive but not definitive. PFS p=0.08 misses the conventional 0.05 threshold, and Phase 2b PDAC readouts have a poor track record of replicating in Phase 3. The PEGPH20/HALO 301 failure in 2019 is the closest cautionary precedent - different mechanism (hyaluronidase enzymatic degradation vs CendR transcytosis), same broad thesis, did not move OS in Phase 3 [12]. Certepetide has no biomarker to select responders; the bet is on a population-level penetration effect, which is harder to confirm at Phase 3 scale. Safety. Toxicity in early trials reads as a chemo profile consistent with a co-infused peptide [8]. Long-term implications of repeated CendR pathway activation, and off-target NRP1 effects on vascular biology and neuronal guidance, remain incompletely characterized. Sponsor and execution - now the dominant risk. Lisata's wholly-owned program shrank to investigator-initiated trials. The Qilu partnership was mutually terminated [10]. As of Q3 2025 disclosures, Lisata had roughly $20M cash and runway into Q3 2026, consistent with the forced-exit dynamic of the Kuva merger [13]. Lisata entered a definitive merger agreement to be acquired by Kuva Labs in March 2026 [11]. Kuva originally licensed certepetide as a targeting agent for its NanoMark MR imaging platform - there is no public commitment to a confirmatory PDAC Phase 3 post-close. The two milestone CVRs ($1.00 China rights reversion, $1.00 any-indication NDA filing) signal that Kuva is buying optionality, not registrational commitment. Commercial. Even with a positive Phase 3, PDAC has tough payer dynamics and lukewarm physician adoption for added-on agents that don't move OS materially. Certepetide would need clear OS benefit and predictable safety to drive uptake on top of gem/nab-paclitaxel or FOLFIRINOX.

Watch list, low conviction, transitioning sponsor. The mechanism is genuinely interesting and ASCEND is the cleanest randomized readout the platform has produced - ORR 45.2% vs 19% and PFS HR 0.60 (P=0.08) in first-line metastatic PDAC is meaningful, even if it does not formally clear statistical significance [2]. But the equity vehicle is essentially closed: Lisata signed a definitive merger with Kuva Labs in March 2026 at $5.00/share cash plus two $1.00 CVRs tied to (i) Greater China rights reversion and (ii) certepetide NDA filing in any indication/jurisdiction [11]. Cash position was running thin (Q3 2025 roughly $20M, runway into Q3 2026) [13], consistent with a forced-exit dynamic. The CVR structure is itself the most honest valuation signal in this story. Kuva is paying $5.00 cash for certain assets and pricing the certepetide registrational upside at $2.00 contingent - that is the market's implied probability of NDA filing, discounted for time. If you believe the asset more than that implies, LSTA CVRs at deal close are a positive-EV bet. If you do not, the deal is fairly priced. What would flip the story positive: Kuva publicly commits to a confirmatory PDAC Phase 3 with realistic timelines; an out-license to a large oncology partner with execution muscle; biomarker work that retroactively segments ASCEND responders; or repurposing as an imaging delivery agent (NanoMark) producing a near-term regulatory path. None has been announced. The GBM Phase 2a is too small and too early to move anything material [1]. Check back when Kuva reports its post-close development plan for certepetide, when CVR milestone updates are filed, or when the investigator-initiated NCT05121038 reports its KRAS-status data - that last point would resolve the panitumumab design question.