BGB-16673

BGB-16673 (tacabrutideg / catadegbrutinib) is BeOne Medicines' oral BTK degrader, an early flagship of the protein-degrader class in hematologic oncology. Rather than blocking Bruton tyrosine kinase (BTK), the protein that keeps B-cell cancers alive, this molecule recruits the E3 ubiquitin ligase cereblon (CRBN) to drag BTK to the cell's protein-shredder for destruction. The company is running multiple Phase 3 trials in chronic lymphocytic leukemia and small lymphocytic lymphoma (CLL/SLL), including in patients who have already burned through the current standard-of-care BTK inhibitors like ibrutinib, acalabrutinib, and zanubrutinib [1][2]. The FDA granted Fast Track Designation for R/R CLL/SLL in August 2024, and the EMA granted PRIME designation in 2025 for previously treated Waldenström macroglobulinemia [15][16]. If the mechanism holds up in Phase 3, BGB-16673 becomes the first commercial validation of BTK degradation and a defensive follow-on to BeOne's own BTK inhibitor Brukinsa (zanubrutinib), which generated roughly $2.6B in 2025 and is the company's revenue engine [3]. The strategic logic is straightforward: BTK-inhibitor-experienced CLL is the fastest-growing failure mode in the class, and whoever wins this niche gets a decade of pull-through as more patients cycle off first-line therapy.

BGB-16673 is a novel compound with no prior approval anywhere. Two international nonproprietary names appear in the literature: tacabrutideg (listed as the primary name in AdisInsight and used in recent OncLive coverage) and catadegbrutinib (used by chemical suppliers and some clinical databases). Based on current WHO/USAN records, tacabrutideg appears to be the WHO proposed INN while catadegbrutinib is the more common USAN-style designation; both refer to the same compound, and BeOne has not yet settled on a single commercial brand [4][17]. The FDA granted Fast Track Designation in August 2024 for adults with R/R CLL/SLL previously treated with at least two prior lines including a BTKi and BCL2i [15]. Registrational activity spans three Phase 3 studies. NCT06970743 has completed enrollment (n=153, active-not-recruiting) in relapsed/refractory CLL/SLL previously treated with a covalent BTK inhibitor, comparing BGB-16673 to investigator's choice of bendamustine plus rituximab or high-dose methylprednisolone plus rituximab [1]. NCT06846671 is still recruiting (n=250 target) in the harder double-refractory population - patients who have failed both a covalent BTK inhibitor and a BCL2 inhibitor like venetoclax [2]. A third Phase 3 head-to-head versus pirtobrutinib began enrolling in Q4 2025 [18]. Two Phase 1 trials continue to accumulate long-term safety and expansion-cohort data, including CaDAnCe-101 (NCT05006716, n=645 target) and a China-specific study (NCT05294731) that supports NMPA filing plans [5][6]. A drug-drug interaction study with phenytoin and itraconazole was completed in healthy volunteers, standard regulatory package work [7]. Given that NCT06970743 has stopped enrolling, a first Phase 3 readout is plausible in the 2026-2027 window, with the double-refractory trial trailing by roughly a year and the head-to-head versus pirtobrutinib further out.

BTK is a signaling enzyme sitting downstream of the B-cell receptor. When a B cell sees an antigen, BTK fires and tells the cell to grow and survive. In CLL and related B-cell cancers, that signal is stuck in the 'on' position, and the malignant B cells become addicted to it [8][9]. Covalent BTK inhibitors (ibrutinib, acalabrutinib, zanubrutinib) work by parking a reactive group onto cysteine 481 of BTK, jamming the active site. The tumor's escape move is well-characterized: it mutates that cysteine to serine (C481S), and the drug can no longer bind. Non-covalent inhibitors like pirtobrutinib were engineered to work around C481S, but resistance mutations at other sites (T474, L528W) have already emerged in the clinic [10]. BGB-16673 is a chimeric degradation activating compound (CDAC), a bifunctional molecule that grabs BTK on one end and recruits the E3 ubiquitin ligase cereblon (CRBN) on the other, tagging BTK for polyubiquitination and destruction by the proteasome [19]. CRBN is the same E3 ligase co-opted by immunomodulatory drugs (IMiDs) like lenalidomide and pomalidomide, and that shared receptor has direct implications for both efficacy and safety (see risks). No BTK protein means no signal, regardless of which resistance mutation the tumor is carrying. Phase 1 data from CaDAnCe-101 presented at ASH 2024 showed objective response rates in the ~78% range in heavily pretreated CLL, including patients with C481S and other resistance mutations, and an updated ASH 2025 readout showed a 94% response rate in a heavily pretreated CLL cohort at recommended dose levels [11][20]. That is a real efficacy signal, not just target engagement.

The pivotal study, NCT06970743, is a randomized open-label Phase 3 in relapsed/refractory CLL/SLL after prior covalent BTK inhibitor exposure. Primary endpoint is progression-free survival by independent review committee. Enrollment is 153, which is small for a Phase 3 and implies a large expected effect size versus the control arm [1]. The comparator, investigator's choice of bendamustine plus rituximab (BR) or high-dose methylprednisolone plus rituximab, is defensible in a post-BTKi setting where options genuinely narrow, but it is a soft comparator. BR typically delivers median PFS in the 12-18 month range in this population, and HDMP+R is worse. That helps BGB-16673 clear the bar, but regulators may push for cross-trial context against pirtobrutinib (Jaypirca, Eli Lilly), which is already approved for CLL/SLL after at least two prior lines including a BTKi and BCL2i, with reported median PFS around 19-20 months in the BRUIN CLL registrational data [10]. BeOne has already launched a dedicated Phase 3 head-to-head against pirtobrutinib (Q4 2025) to answer that question directly [18]. NCT06846671 targets a more brutal population (n=250, recruiting), double-refractory after both BTK and BCL2 inhibitors, where no standard of care exists and any credible PFS or ORR signal has commercial teeth [2]. Randomizing against investigator's choice again is pragmatic, though it invites the same questions about comparator quality. Open-label design is standard for oral targeted agents against chemo but adds bias risk on PFS assessment, partially mitigated by IRC review.

Our model estimates a 22% 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 its light or open-label blinding and more secondary endpoints than usual; it is held back by the sponsor's thin or weak approval record and 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 sits mostly in the comparator gap. If BR performs at the high end of historical range and BGB-16673 lands in the middle of its Phase 1 distribution, the PFS delta could compress. The 153-patient sample is thin for a Phase 3 registrational study and any subgroup imbalance amplifies. Safety risk is where the interesting exposure lives. BTK is expressed in myeloid cells and platelets, and CRBN-mediated degradation is systemic, not just tumor-restricted. Phase 1 has shown neutropenia, thrombocytopenia, and infections consistent with on-target biology, but at rates that so far look manageable [11]. The CRBN-specific risk is the known IMiD-class hazard: cereblon binders can drive neo-substrate degradation of IKZF1 and IKZF3 (Ikaros and Aiolos), producing immunomodulatory effects that are pharmacologically real (this is how lenalidomide works) and can also cause off-target cytopenias, fatigue, and rash. Whether CaDAnCe-101 has systematically screened for IKZF1/3 protein levels in peripheral blood has not been publicly disclosed, and that is exactly the readout investors should demand at future ASH updates. There is also a drug-drug interaction consideration: patients on lenalidomide-containing regimens (common in prior lines for lymphoma) may exhibit competitive CRBN binding that reduces BGB-16673 efficacy. Nurix's NX-2127 was placed on partial clinical hold in 2024 over CMC issues, a reminder that the manufacturing side of protein degraders is not fully solved [13]. Commercial risk is real on two fronts: (1) pirtobrutinib (Jaypirca, Eli Lilly) is already on-market for post-BTKi CLL with an established safety profile, and payers will want head-to-head or clean cross-trial data - BeOne's own Q4 2025 head-to-head trial acknowledges this; (2) Nurix's bexobrutideg (NX-5948), partnered with Roche, is the primary BTK-degrader commercial competitor and is running its own pivotal DAYBreak CLL-201 study, with additional early-line and autoimmune expansion - Roche's global commercial footprint is the actual competitive threat, not Nurix alone [21][22]. Nurix's other asset, NX-2127, remains a separate earlier-stage program.

Worth watching, and specifically for three data points. First, updated CaDAnCe-101 durability at future ASH meetings: the ASH 2025 94% ORR readout was strong, but median duration of response holding past 24 months in the covalent-BTKi-experienced cohort is what converts this to a serious commercial asset regardless of Phase 3 comparator debates [20]. Second, any signal of unexpected off-target or CRBN-neo-substrate degradation in longer follow-up. The Nurix/Roche cohort (bexobrutideg / NX-5948, plus NX-2127) is running parallel programs against overlapping patient sets, and the class will rise or fall together on tolerability. Third, China: BeOne is a dual-listed company with foundational operations in China, Brukinsa's China revenue is a meaningful share of its ~$2.6B 2025 global total, and NCT05294731 is positioned to support an NMPA filing that would let BeOne defend its domestic BTKi installed base as it transitions to degrader chemistry. BeOne's strategic position is strong here: Brukinsa is the fastest-growing BTK inhibitor globally and BGB-16673 is a defensive line item against the eventual erosion of covalent BTKi franchises [3]. Losing the degrader race to Nurix/Roche would not kill the company but would concede the next decade of BTK-class economics. The first Phase 3 readout in the 2026-2027 window is the moment. Monitor BeOne's 10-Q filings for any language change on trial timelines, which is usually the first public signal that a readout is slipping or accelerating [14].