Daratumumab SC

This is an NCI-sponsored Phase 2 trial (NCT05907759) testing subcutaneous daratumumab - Johnson & Johnson's anti-CD38 antibody, sold as Darzalex Faspro - in three rare and aggressive lymphoproliferative cancers: primary effusion lymphoma (PEL), plasmablastic lymphoma (PBL), and multicentric Castleman disease (MCD) [1]. These diseases sit at the plasma-cell/B-cell boundary, are frequently HIV-associated and driven by Kaposi's sarcoma-associated herpesvirus (HHV-8), and have few good options after frontline chemo (typically EPOCH-R for PEL/PBL) fails. The bet is mechanistic: these tumors express CD38, the same target daratumumab obliterates in multiple myeloma, where the IV and SC formulations together generated roughly $11.7B for J&J in 2024 [2]. This is investigator-driven label-expansion science, not a J&J-led registration push - the molecule is owned and made by J&J (originated at Genmab), but the sponsor here is the NCI [1]. A positive ORR readout would build the case for daratumumab in CD38-positive non-myeloma plasma-cell-like cancers, an area where the only approved anti-CD38 antibodies (daratumumab, isatuximab) have minimal labeled use. The commercial upside for J&J is small; the scientific upside - extending CD38 biology into HHV-8-driven disease - is the real story.

Daratumumab itself is not investigational. The IV form was first approved by FDA in November 2015 for relapsed/refractory multiple myeloma, and the SC formulation (daratumumab plus recombinant human hyaluronidase, branded Darzalex Faspro) was approved in May 2020 [3]. The SC version now dominates J&J's anti-CD38 franchise because it shortens administration from a multi-hour IV infusion to roughly a five-minute injection and cuts infusion-reaction rates substantially. What is investigational here is the indication. PEL, PBL, and MCD have no anti-CD38 antibody approved for them. For frontline context: PEL and PBL are typically treated first-line with dose-adjusted EPOCH-R (etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin, rituximab), and HHV-8-positive MCD is typically treated with rituximab-based regimens - patients eligible for this trial have failed those backbones. This Phase 2 has no FDA breakthrough, fast-track, or orphan designation tied to it - it's a trial run through NCI's academic trial infrastructure (the Cancer Therapy Evaluation Program, CTEP) [1]. Recruitment is open at n=28 total across all three arms combined - which works out to as few as ~9 patients per disease cohort if accrual splits evenly, and tells you this is a signal-finding study, not a registrational one [1]. Readout timing is not publicly disclosed; given the rarity of these tumors (PEL alone has fewer than 1,000 US cases annually) and the small enrollment cap, expect partial data in 2027 at the earliest. A positive signal would more plausibly trigger a follow-on Janssen-sponsored or multi-center academic consortium trial than direct label expansion.

CD38 is a protein that sits on the outside of cells and acts as both an enzyme (it chops up the NAD+ molecule and helps cells signal calcium) and a marker [4]. Healthy plasma cells (the antibody factories of the immune system) display a lot of CD38 on their surface. So do multiple myeloma cells, which are malignant plasma cells. So do the cells in PEL, PBL, and MCD - these are all cancers of cells that have differentiated toward the plasma cell end of the B-cell lineage, which is why they keep CD38 up high. Daratumumab is a human IgG1κ antibody that latches onto CD38 and then kills the cell four ways: it recruits immune killer cells to chew it up (antibody-dependent cellular cytotoxicity, ADCC, and antibody-dependent cellular phagocytosis, ADCP), it triggers the complement protein cascade (CDC), and it directly nudges the cell toward apoptosis [5]. The mechanism is as well-validated as oncology gets: thousands of myeloma patients respond, and emerging Phase 2 data in CD38-positive non-myeloma settings (e.g., daratumumab plus GDP in CD38+ relapsed/refractory peripheral T-cell lymphoma) supports activity beyond plasma-cell malignancies [11]. The open question is whether PEL/PBL/MCD biology - driven by HHV-8 and often HIV, with plasmablastic rather than mature plasma-cell phenotype - preserves enough CD38 dependence to make killing CD38-positive cells therapeutically meaningful. Open Targets gives CD38-myeloma an evidence score of 0.62; the lymphoma adjacencies are weaker but biologically reasonable [6].

NCT05907759 is a Phase 2, single-arm, open-label basket trial run by the NCI in patients with relapsed/refractory PEL, PBL, or MCD [1]. Enrollment target is 28 patients total across the three disease cohorts - meaning as few as ~9 patients per cohort if accrual splits evenly, which is the floor for any kind of per-cohort signal interpretation. Primary endpoint is overall response rate (ORR), with secondary endpoints covering duration of response, progression-free survival, and safety [1]. Patients receive subcutaneous daratumumab on a standard weekly-then-tapered schedule. There is no comparator arm, no randomization, and no biomarker enrichment beyond clinicopathologic diagnosis - investigators are not pre-screening for CD38 expression level, which is a missed opportunity given that response in heme malignancies tracks with CD38 density. The design is appropriate for hypothesis generation in ultra-rare cancers where a controlled trial is logistically impossible, but it limits what you can conclude. A 30% ORR in PBL would be interesting; a 30% ORR in MCD would be much less impressive, particularly because the MCD cohort likely mixes HHV-8-positive disease (the form most relevant to this CD38/plasmablast hypothesis) with HHV-8-negative idiopathic MCD (iMCD), where siltuximab (anti-IL-6) is already FDA-approved and produces durable responses. A separate Italian Phase 2 (NCT04915248, n=28, RECRUITING) is testing daratumumab plus bortezomib/dexamethasone specifically in PBL [7]. Read those two trials together - if both produce signal in PBL, the case for a registration-grade study tightens considerably.

Our model gives this drug an 8% chance of eventually being approved. That figure starts from the historical approval rate for Phase 2 drugs in this area - about 21% - then adjusts based on ten facts about the trial and its sponsor. The estimate goes up because of a non-randomized design and open-label blinding, but goes down because the sponsor has a thin approval record and earlier-phase results were weak. Most other factors land near average, so they don't shift the number much.

Efficacy risk is the dominant concern. PEL, PBL, and MCD are biologically related but not identical - PBL is most plasma-cell-like and likely the highest-CD38 cohort; PEL is HHV-8-driven and phenotypically more variable; MCD is itself split into HHV-8-positive multicentric disease (where the CD38/plasmablast rationale is strongest) and HHV-8-negative idiopathic MCD (where siltuximab already produces durable responses and the bar for 'responding' is high). Without per-cohort CD38 expression data, you cannot tell whether a non-response means the drug failed or the patient was poorly selected. Safety risk is minimal - daratumumab's profile is among the cleanest in heme-onc, and SC administration cut grade 3+ infusion reactions to under 1% [9]. The main on-target effect is suppression of normal plasma cells and modestly higher infection risk, which matters more in HIV-positive populations common in PEL/PBL. Execution risk is real: 28 patients across three ultra-rare cancers means enrollment will take years, and accrual in PEL specifically has historically stalled multi-center academic consortium trials. Commercial risk is structural and time-bound - this is an NCI trial in indications too small to justify a dedicated J&J registration effort, and daratumumab's core composition-of-matter patents are estimated to expire circa 2029-2031 (varies by geography). Biosimilar entry would compress pricing and weaken any commercial case for chasing rare-indication label additions. The SC formulation patents (covering the recombinant hyaluronidase co-formulation) extend exclusivity further but are more vulnerable to challenge than the antibody composition claims. Even with positive data, payer coverage for off-label Darzalex Faspro use in PEL/PBL/MCD will be inconsistent without an NCCN listing or an FDA label addition.

Worth watching, but not a J&J commercial signal. Darzalex's growth thesis is driven by frontline multiple myeloma combinations (the PERSEUS, MAIA, and GRIFFIN trials - Phase 3 studies establishing daratumumab-based quadruplet regimens as standard of care in newly diagnosed myeloma) and now smoldering myeloma, not by rare-lymphoma label expansions [2]. The reason to track NCT05907759 is biological, not financial: a clean ORR signal in CD38-positive HHV-8-driven cancers would open up a small but real adjacent market for anti-CD38 antibodies (daratumumab and isatuximab) and could prompt Janssen-sponsored confirmatory work. The specific data point to watch is per-cohort ORR with CD38 expression correlatives - if investigators publish those, the signal-to-noise jumps. The Italian PBL trial (NCT04915248) is the parallel readout to track [7]; if both produce daratumumab response signals in PBL specifically, that's the most likely path to a formal label expansion. Check back when either trial reports interim data - realistically mid-2027. The broader commercial story to watch instead has two layers. First, CD38 biology has two commercialized antibodies (daratumumab and isatuximab), with felzartamab advancing in late-stage trials (Phase 3 in IgA nephropathy, Phase 2 history in heme-onc as MOR202) as a potential third entrant; rare-lymphoma extension is upside, not thesis. Second, AbbVie's etentamig + daratumumab combo in frontline myeloma (NCT07095452) is a more material competitive event for J&J's franchise - etentamig is a bispecific antibody being combined with daratumumab rather than competing against it, but the combo could shift share of the frontline regimen mix [10]. Looming over all of this: biosimilar daratumumab in the early 2030s reshapes the entire commercial calculus for rare-indication expansion.