Zilovertamab vedotin

Zilovertamab vedotin (MK-2140) is Merck's anti-ROR1 antibody-drug conjugate for diffuse large B-cell lymphoma (DLBCL), the most common aggressive blood cancer in adults [1]. Merck is running it across Phase 2 and Phase 3 programs, including a Phase 3 against R-CHOP (rituximab + cyclophosphamide + doxorubicin + vincristine + prednisone — the chemo backbone of frontline DLBCL care) in waveLINE-010 (n=1,046) [2] and a Phase 2 head-to-head against Roche's Polivy on R-CHP backbone (R-CHOP minus vincristine, dropped to leave room for the ADC's neurotoxic payload) in waveLINE-011 (n=594) [3]. The strategic bet: a more cancer-selective target than CD79b can produce a cleaner ADC for frontline DLBCL, a market Polivy already won with POLARIX [4]. DLBCL is a roughly 25,000-new-case-per-year US disease and Polivy + R-CHP is now the preferred frontline regimen in NCCN guidelines, generating north of $1.5B in 2024 — so the commercial prize is large but the comparator bar has already moved up.

Three trials matter for the regulatory story. waveLINE-010 (NCT06717347) is the Phase 3: 1,046 untreated DLBCL patients randomized to zilovertamab vedotin + R-CHP vs standard R-CHOP, PFS (progression-free survival — time until cancer progresses or patient dies) primary endpoint, recruiting [2]. waveLINE-011 (NCT06890884) is a Phase 2 head-to-head: 594 patients, zilo + R-CHP vs polatuzumab vedotin + R-CHP, complete response rate at end of treatment per Lugano criteria (the standard PET-CT-based response framework for lymphoma) [3]. waveLINE-004 (NCT05144841) is the completed Phase 2 in relapsed/refractory DLBCL, n=98 evaluable in the updated ASH 2024 readout [5,12]. ASH 2024 data: ORR 29% (95% CI 19-40), CR 13% (n=10), PR 16% (n=13), disease control rate 42% in a heavily pretreated population (median age 66, 71% with ≥3 prior lines, 58% ineligible for ASCT or CAR-T) [12]. For benchmarking, Polivy's r/r DLBCL key (Pola-BR vs BR, Sehn et al. JCO 2020) showed ORR ~45% and CR ~40% in a less heavily pretreated cohort [13] — so zilo's r/r numbers are softer in absolute terms but the populations are not apples-to-apples. waveLINE-006 (NCT05458297) in MCL/FL/CLL is active but no longer recruiting [6]. No publicly disclosed FDA breakthrough therapy, fast track, or priority review designation. Merck has not guided a specific BLA timeline; the Phase 3 PFS readout is realistically 2027 or later given recruitment scale, and waveLINE-011's CRR readout is the earlier inflection point (likely 2026-2027 based on enrollment pace). Solid tumor expansion is also in progress (Meric-Bernstam et al. 2025, Phase 2 in metastatic solid tumors) but DLBCL is the lead indication [7]. Dose schedule across the program is 1.75 mg/kg IV q3w [5,12].

ROR1 (receptor tyrosine kinase-like orphan receptor 1) is a surface protein that's busy on embryonic cells, then mostly shuts off in healthy adult tissue and reactivates on many B-cell cancers and a subset of solid tumors [1,8]. That on/off pattern is what makes ROR1 attractive for an antibody-drug conjugate: the antibody finds a flag that cancer cells carry but most normal cells don't, so the toxic payload gets delivered selectively. Zilovertamab vedotin works in three steps. The antibody binds ROR1 on the cancer cell surface. The complex gets pulled inside the cell. A linker then releases MMAE (monomethyl auristatin E), a microtubule poison, which jams the machinery cancer cells need to pull chromosomes apart during division, killing them. MMAE is membrane-permeable after release, which enables bystander killing: the freed payload can diffuse into neighboring tumor cells regardless of their ROR1 expression. That partially mitigates the heterogeneity risk discussed below — a ROR1-negative cell sitting next to a ROR1-positive cell can still take a hit — but it also widens the off-target toxicity envelope. The ADC strategy in DLBCL is already validated. Polatuzumab vedotin (Polivy, Roche) is an anti-CD79b ADC with the same MMAE payload that hit its primary endpoint in POLARIX (frontline DLBCL Phase 3) and became standard-of-care frontline [4]. So the question isn't whether ADCs work in DLBCL — they do. The question is whether ROR1 is a better target than CD79b. ROR1 is in principle more tumor-selective (less expression on normal B-cells), which could mean a cleaner safety profile or a wider therapeutic window. That's the unproven part. No anti-ROR1 drug has been approved for any indication; the only meaningfully advanced anti-ROR1 program outside Merck's is Oncternal's cirmtuzumab (anti-ROR1 mAb in CLL/MCL, in combination with ibrutinib), which has not produced a registrational dataset. Target validation in oncology therefore rests almost entirely on Merck's program. The Phase 2 head-to-head against Polivy is where this thesis gets tested [3].

waveLINE-010 (NCT06717347) is the program-defining Phase 3: 1,046 patients with previously untreated DLBCL, randomized to zilovertamab vedotin + R-CHP vs R-CHOP, PFS primary [2]. Comparator choice is the strategic question. POLARIX established Polivy + R-CHP as the new frontline benchmark on PFS, so running zilo + R-CHP vs R-CHOP rather than vs Polivy + R-CHP means a Phase 3 win doesn't directly dislodge Polivy — it just shows zilo beats the old standard. Merck has to win twice: vs R-CHOP in waveLINE-010 to get approval, and vs Polivy in the market. That's why waveLINE-011 (NCT06890884) matters more for the commercial story. It's a Phase 2 head-to-head, 594 patients, zilo + R-CHP vs Polivy + R-CHP, primary endpoint complete response rate at end of treatment per Lugano [3]. This is the trial that says whether zilovertamab is actually differentiated. The choice of CRR at EOT (not PFS) means readout is faster but the endpoint is softer — response rates tell you about depth, not durability. waveLINE-004 (NCT05144841) was the Phase 2 in r/r DLBCL, n=98 evaluable, completed [5,12]. The ASH 2024 update showed sustained antitumor activity in a heavily pretreated, ASCT/CAR-T-ineligible-or-refractory population (see status section for ORR/CR numbers). Those numbers set the efficacy floor for the frontline program, and they are below Polivy's r/r benchmark in absolute terms, though the populations are not directly comparable [13]. waveLINE-003 (NCT05139017) is a Phase 2 dose-finding combo study still recruiting [9]. Enrollment in waveLINE-010 (1,046) is larger than POLARIX (879), which stretches the timeline.

Our model estimates a 16% chance this drug is eventually approved. It starts from the historical base rate for Phase 2 drugs in this area (about 21%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by larger-than-typical enrollment for this phase, its light or open-label blinding, and more secondary endpoints than usual; it is held back by weak or limited earlier-phase results. The other facts land near average for this stage, so they leave the estimate roughly where the base rate put it.

Efficacy risk first, and it's the central one. ROR1 expression in DLBCL is genuinely heterogeneous, and the spread by detection method is wide: roughly 15% of unselected DLBCL samples show ROR1 membrane positivity by IHC (the clinically relevant surface target for an ADC), about 60% are positive by in situ hybridization at the transcript level, and an earlier IHC study using a different antibody clone reported 83% cytoplasmic positivity [14]. The ADC needs surface ROR1 to engage, so the 15% figure is the load-bearing number. waveLINE-010 does not pre-select for high expressors, which means most of the unselected trial population may have tumors with low membrane target density. MMAE bystander killing softens this somewhat (released payload can hit neighboring cells regardless of ROR1 status), but it doesn't eliminate the dilution problem — the trial could miss its endpoint not because the drug fails biologically but because too many patients in the active arm have tumors with sub-threshold target density. Safety risk is MMAE-driven. The payload is the same microtubule poison used in Polivy and Adcetris, and it carries the same dose-limiting toxicity profile: peripheral neuropathy, neutropenia, infusion reactions. Hard to differentiate on safety against Polivy when the payload is identical. Any worse neuropathy signal in the head-to-head and the differentiation story dies. Commercial risk is the structural problem. POLARIX made Polivy + R-CHP the standard frontline DLBCL care [4]. waveLINE-010 vs R-CHOP doesn't beat Polivy on its own terms, even with a clean win. Payers and KOLs will demand head-to-head evidence. That's why waveLINE-011 is where the program's commercial future actually lives. Execution risk: a 1,046-patient Phase 3 in frontline DLBCL takes years to enroll and read out. Merck has the infrastructure but timeline drag adds competitive pressure — CD20xCD3 bispecifics like epcoritamab and glofitamab are eating r/r DLBCL share and may move earlier in line, changing the comparator math during waveLINE-010's recruitment window.

Worth watching, but not the headline event for the BioCosm DLBCL map yet. The single data point that matters for thesis-testing is waveLINE-011's complete response rate readout — Phase 2 head-to-head vs Polivy + R-CHP, likely 2026-2027 on enrollment pace. A clean win there (CRR materially above Polivy) and the program becomes a real frontline DLBCL contender. A tie or loss and the Phase 3 design problem becomes acute, because beating R-CHOP doesn't beat the actual current standard of care anymore. The waveLINE-004 r/r data is now in hand at the ASH 2024 cut: ORR 29%, CR 13% in a heavily pretreated population [12]. That's a real efficacy floor, but it's a soft one relative to Polivy's r/r numbers [13] and to what frontline expectations need to look like. The GCB (germinal center B-cell-like, one of two main DLBCL molecular subtypes) vs ABC subtype breakdown in waveLINE-004 has not been published in detail; that distribution matters for predicting frontline performance because ROR1 expression and biology may differ by subtype. Commercial context: DLBCL is roughly a 25,000-new-US-case-per-year disease, and Polivy + R-CHP frontline is now NCCN-preferred. Polivy generated more than $1.5B in 2024 globally and continues to grow. So the commercial prize is real, but Merck would be entering against an entrenched standard, not an open field. Merck doesn't need this drug to hit. Keytruda generated about $29.5B in 2024 [11] and continues to expand indication by indication. Zilovertamab is a portfolio shot, not a P&L bet, which means Merck will let the data drive the decision rather than push through ambiguous signals to protect a fragile program. That makes the trial readouts cleaner reads on the science, not on corporate pressure to keep the program alive. Check back at the next waveLINE-004 update (durability tail), or when waveLINE-011 reports its first interim look on CRR.