CM336

CM336 is a BCMA x CD3 bispecific antibody from Keymed Biosciences, now running in five Phase 1/2 trials across light-chain (AL) amyloidosis, multiple myeloma with renal failure, and antibody-mediated autoimmune disease [1][2]. The lead indication tagged on this node is newly-diagnosed AL amyloidosis at low dose in a single-center Chinese study (NCT07151690, n=21) [2]. The mechanism is well-validated by teclistamab, elranatamab, and linvoseltamab in myeloma, but everything about CM336 sits one to two years behind Western peers and entirely outside Western regulatory pathways.

CM336 is a novel investigational antibody with no approvals anywhere. Five active trials, all Phase 1 or Phase 2, all in China [2][3][4][5][6]. The asset is developed and manufactured by Keymed Biosciences (HKEX: 2162), a Chengdu-based biotech building out a hematology and autoimmunity portfolio [1]. The trial captured in our node (NCT07151690) is sponsored by the Institute of Hematology & Blood Diseases Hospital in Tianjin - an investigator-initiated arrangement, not a Keymed-run key study. We've updated the structured_data to reflect this: sponsor is the Tianjin institute; Keymed is recorded separately as manufacturer/developer. No FDA designations are listed because none of these trials are running in the US, and CM336 has not entered Western regulatory pathways. The largest Keymed-sponsored AL amyloidosis trial (NCT07039578, n=90) is the closest thing to a registrational design [4]. The autoimmune cytopenia Phase 1 (NCT07175493, n=158) is unusually large for Phase 1, which signals Keymed is building a basket strategy around BCMA-driven antibody-secreting cell depletion in autoimmune disease, not just oncology [5]. No public readout timeline; no CM336 data appears to have been presented at ASH 2025 or major hematology meetings as of this writing - first interim hematologic response data is plausible in H2 2026 to H1 2027 based on response kinetics in this disease.

The biology is clean. BCMA (B-cell maturation antigen, gene TNFRSF17) is a receptor on plasma cells - the antibody-producing cells that mature from B cells [7]. CD3 is the trigger complex on T cells. A bispecific antibody is a single molecule with two grabbing arms: one arm latches onto BCMA on a plasma cell, the other arm latches onto CD3 on a T cell, dragging the T cell into contact and forcing it to kill the plasma cell. Think of it as a molecular leash that pulls killer cells next to their targets. This matters in AL amyloidosis because the disease is caused by a small clonal population of malignant plasma cells in the bone marrow producing a misfolded immunoglobulin light chain. The light chain deposits as amyloid fibrils in organs - most lethally the heart, then the kidneys - and patients die from organ failure, not from tumor burden [8]. The plasma cell clone is small, but the protein it makes is toxic. Kill the plasma cells, stop the toxic light chain at the source, organs can recover. The mechanism is validated. Teclistamab (J&J, approved 2022 for relapsed/refractory myeloma), elranatamab (Pfizer, approved 2023), and linvoseltamab (Regeneron, approved 2024) all use the same BCMA/CD3 axis with response rates of roughly 60-71% in heavily pretreated myeloma patients [9][10][12]. Daratumumab (anti-CD38) is FDA-approved in AL amyloidosis as Darzalex Faspro for newly-diagnosed disease (NDAL) since January 2021, based on ANDROMEDA [8]. Extending BCMA bispecifics into amyloidosis is biologically sensible, not novel target validation.

NCT07151690 is a single-arm, single-center, open-label Phase 2 at the Institute of Hematology & Blood Diseases Hospital in Tianjin, enrolling 21 newly-diagnosed AL amyloidosis patients [2]. Primary endpoint is hematologic VGPR (very good partial response) or better - meaning the toxic light chain is suppressed below a defined threshold (serum free light chain difference under 40 mg/L). That's the right surrogate for AL amyloidosis; organ response lags hematologic response by months. The design choices tell you what this study is and isn't. n=21, no comparator, single site - this is a proof-of-concept signal generator, not a registrational trial. The 'low-dose' framing in the trial name is the interesting tell: AL amyloidosis patients are often frail with cardiac involvement, and high-dose T-cell engager therapy carries cytokine release risk. Starting low to read safety in a vulnerable population is the right call. The broader Keymed-sponsored Phase 2 (NCT07039578, n=90) is the trial to watch for any registrational read - also single-arm, but better-powered for response rate confidence intervals [4]. Companion trials in MM with renal impairment (NCT07585760), autoimmune bullous disease (NCT06900010), and autoimmune cytopenia (NCT07175493) round out a basket strategy betting that BCMA-driven plasma cell depletion will work across antibody-mediated diseases [3][6][5].

Our model puts this drug's approval chances at 18%. That starts from a 34% historical baseline for Phase 2 drugs in this area, then adjusts based on ten facts about the trial and sponsor. The biggest drags are the sponsor's thin approval record, weak earlier-phase results, and smaller-than-typical enrollment; a non-randomized trial design works in its favor. The remaining factors are close to average for this stage, so they don't shift the estimate much either way.

Safety: BCMA/CD3 bispecifics carry mechanism-based cytokine release syndrome (CRS - an immune overreaction causing fever, low blood pressure, and sometimes organ failure) and infection risk. Teclistamab's label carries a black-box warning for CRS and neurotoxicity, with Grade ≥3 infections in roughly 45% of patients and hypogammaglobulinemia (depletion of normal antibodies, which raises infection risk) as plasma cells get depleted [9]. In AL amyloidosis, patients often have baseline cardiac involvement with reduced ejection fraction - CRS that would be manageable in myeloma could be lethal. The 'low-dose' framing in NCT07151690 is hedging this exact risk, and the dose level chosen will define whether the trial reads out as efficacy signal or safety signal. Efficacy: In ANDROMEDA (Kastritis NEJM 2021), the daratumumab-CyBorD arm in newly-diagnosed AL amyloidosis achieved a hematologic complete response (the primary endpoint) of 53.3% vs 18.1% for CyBorD alone, with hematologic VGPR-or-better around 78% in the dara arm at the primary analysis [8]. To matter commercially, CM336 needs to beat or complement that - not just achieve a respectable response rate against a no-treatment comparator that doesn't exist in this single-arm trial. A 21-patient study cannot answer the comparative question. Competition: BCMA is the most crowded validated target in heme malignancy. Beyond the three approved bispecifics (teclistamab, elranatamab, linvoseltamab [9][10][12]), two BCMA CAR-Ts are approved - ciltacabtagene autoleucel (Carvykti, J&J/Legend) and idecabtagene vicleucel (Abecma, BMS/2seventy) - both in relapsed/refractory myeloma, both being investigated in AL amyloidosis [13]. CAR-T differentiates on one-time administration vs repeat bispecific dosing, but requires apheresis, lymphodepletion, and a tertiary center - which may favor bispecifics for frail amyloidosis patients. Execution: single-center, n=21, no comparator, Chinese site. Even strong results will require a confirmatory multicenter trial - ideally with Western participation - for FDA or EMA submission. Keymed has not publicly disclosed a Western development partner. Commercial: AL amyloidosis is orphan-sized (US incidence ~4,500/year). Pricing power exists, but the market is consolidating around daratumumab-based regimens. A BCMA bispecific is more likely to find its niche in relapsed/refractory or dara-refractory AL amyloidosis - not front-line newly-diagnosed, which is what this trial targets.

Worth watching at low priority. The interesting story here is Keymed Biosciences, not the trial sponsor in the node (which is just the investigator-initiated host hospital). Keymed is running a basket of CM336 trials across hematology and autoimmunity that signals a real commercial program, not a one-off academic study. Financial caveat: Keymed Biosciences (HKEX: 2162) is publicly listed in Hong Kong, but the financial runway needed to execute five concurrent trials plus a future multicenter Phase 3 is not assessed in this writeup. Investors should verify current market cap, cash position, and any disclosed partnerships or fundraises directly from HKEX filings before acting on the basket-strategy thesis. Signal events to watch: (1) First hematologic response data from NCT07151690 - VGPR rates above 70% with manageable CRS would justify continuing to track. (2) Any announcement of a Western development partner or key trial design - this is the biggest gating item for the asset to matter outside China. (3) Safety specifically in the cardiac-involved subgroup. (4) Read-throughs from the broader Phase 2 (NCT07039578, n=90), which will give a more interpretable response rate confidence interval [4]. (5) First conference presentation - ASH 2026 or EHA 2026 would be the natural venues. Check back: H2 2026 at the earliest for interim data from any of the AL trials. The autoimmune indications (NCT07175493 cytopenia, NCT06900010 bullous disease) are the more strategically interesting wedge - BCMA-directed approaches in antibody-mediated autoimmune disease open larger commercial markets than amyloidosis, and that's where Keymed's bigger Phase 1 (n=158) is sitting [5].