Emavusertib (CA-4948)

NCT07271667 (TakeAim CLL, protocol CA-4948-203) is a Phase 2 trial from Curis testing emavusertib, an oral IRAK4 inhibitor, in combination with zanubrutinib in chronic lymphocytic leukemia and other B-cell malignancies, with undetectable measurable residual disease (uMRD) by ClonoSEQ NGS as the primary endpoint [1]. The mechanistic bet is that hitting IRAK4 deepens BTK-driven responses by closing off a parallel survival signal that drives residual disease and resistance. For Curis (CRIS), a micro-cap that ended Q1 2026 with $15.0M cash against a $9.0M quarterly burn and a stated going-concern doubt, emavusertib is functionally the entire company [6][7][12].

Emavusertib is a novel small molecule, never approved anywhere. It is an oral inhibitor of IRAK4 with secondary activity against FLT3. NCT07271667 (CA-4948-203, TakeAim CLL) is Phase 2, recruiting, with eleven active clinical sites as of June 26, 2026; Curis has reaffirmed guidance that the initial five patients will be dosed by end of July 2026, with initial data expected in Q4 2026 (December 2026) [1][7][12]. The full multi-cohort trial across CLL and other B-cell malignancies has a target enrollment of approximately 108 across cohorts; the proof-of-concept zanubrutinib cohort is roughly 20-30 patients [1][12]. The B-cell program is one of several active emavusertib indications. Relapsed/refractory primary CNS lymphoma is in Phase 1/2 (NCT03328078, TakeAim Lymphoma, with monotherapy and ibrutinib combination arms), and the same protocol includes Waldenstrom's macroglobulinemia cohorts, the highest-MYD88-L265P-prevalence indication (>90%) [2][13]. A pembrolizumab combination in checkpoint-refractory urothelial cancer is in NCI-sponsored Phase 1 (NCT06439836) [3]. A FOLFOX plus bevacizumab combination in metastatic colorectal is in NCI-sponsored Phase 1 (NCT06696768) [4]. Emavusertib is also being slotted into the NCI MyeloMATCH umbrella for myeloid disease (NCT05564390) [5]. The asset carries earlier fast track and orphan designations from the AML/MDS programs, which ran into a partial clinical hold in 2022 over rhabdomyolysis cases [9]. No breakthrough therapy designation. Curis has not publicly guided to a specific BLA timeline; the Phase 2 zanubrutinib combination would need a key expansion or a separate registrational trial. Recent 8-K activity through 2025 and into 2026 has centered on financings, a reverse stock split, and corporate updates rather than registrational data drops [6][7].

IRAK4 (interleukin-1 receptor-associated kinase 4) is a kinase that sits inside immune cells like a relay switch. When a toll-like receptor or the IL-1 receptor on the cell surface gets pulled, IRAK4 fires next, triggering the NF-kB transcription factor to flip on inflammatory and survival genes. The cancer angle: in a sizable chunk of B-cell lymphomas, a mutation in MYD88 (the adapter protein that recruits IRAK4) jams that switch into the 'on' position, so the malignant B cell gets a constant survival signal with no pathogen around. Blocking IRAK4 cuts that wire. MYD88 L265P prevalence varies sharply by tumor type. Waldenstrom's macroglobulinemia carries the mutation in over 90% of cases, primary CNS lymphoma in roughly three-quarters, and CLL in only a small minority [13]. In CLL, BTK is the dominant addiction. BTK inhibitors (ibrutinib, acalabrutinib, zanubrutinib) have rewritten the CLL field. Resistance shows up via BTK C481S point mutations or by tumors rerouting survival signals through PLCG2 or NF-kB. IRAK4 is one of those rerouting nodes. The combination thesis is that IRAK4 inhibition pinches off the parallel survival signal that BTK inhibitors leave standing, deepening responses to the point of measurable residual disease negativity. Genetic validation for IRAK4 in B-cell malignancies is real but uneven. The MYD88 L265P story is well-anchored in Waldenstrom's and PCNSL biology. Whether that translates into deeper CLL responses with a BTKi combination, in a tumor where MYD88 mutation is uncommon, is the experiment.

NCT07271667 is a Phase 2, multi-cohort, open-label trial with a target enrollment of approximately 108 across CLL and other B-cell malignancies. The proof-of-concept zanubrutinib cohort is roughly 20-30 patients [1][12]. Cohort 1 enrolls CLL patients in partial response or partial response with lymphocytosis who are MRD-positive by ClonoSEQ NGS assay after at least 12 months on zanubrutinib; emavusertib is then added on. Primary endpoint in Cohort 1 is undetectable measurable residual disease rate, a stringent bar [1]. The trial's uMRD endpoint uses the ClonoSEQ next-generation sequencing assay, which detects one tumor cell in 10^6 cells - more sensitive than the flow-cytometry-at-10^-4 standard used in most historical ibrutinib monotherapy benchmarks. That assay mismatch matters for cross-trial comparison: a 5-10% ibrutinib monotherapy uMRD rate at flow-cytometry 10^-4 in peripheral blood is not directly comparable to a uMRD readout at NGS 10^-6, where the bar is mechanically harder to clear. The endpoint choice is a tell. Curis is not chasing overall response rate, which BTK inhibitors already produce as monotherapy. They are going after depth of response, which is where BTKi monotherapy plateaus and where combinations with venetoclax (a BCL2 inhibitor that flips B cells back into apoptosis) have reset the bar [10]. The enrollment design is also a tell: the BTKi-pretreated, MRD-positive, partial-responder population is exactly the subgroup where BTKi monotherapy has hit a ceiling, which gives the add-on a fair shot but does not test contribution-of-effect against frontline BTKi-venetoclax combinations [10]. Concerns: no comparator arm, so any uMRD rate gets benchmarked against historical BTKi monotherapy data, which is heterogeneous in assay and tissue source. Curis has disclosed that any approved BTK inhibitor can be used in other B-cell malignancy cohorts, which adds heterogeneity across cohorts; the lead CLL cohort is anchored to zanubrutinib specifically [1][12].

Our model estimates a 15% 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 a non-randomized design 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.

Capital risk is the dominant near-term concern. Curis ended Q1 2026 with $15.0M cash, used $9.0M in operations that quarter, and disclosed substantial doubt about going concern within twelve months absent additional capital [12]. Management has stated that current and anticipated proceeds, including up to $20.2M from PIPE Series B warrant exercise, should fund operations into the second half of 2027, but that runway depends on warrant exercise that is not guaranteed. A reverse stock split was approved by stockholders in June 2026, signaling listing-compliance pressure [7]. Efficacy risk is the next concern. CLL is not the indication with the cleanest IRAK4 biology. MYD88 L265P is more of a Waldenstrom's and PCNSL story than a CLL story [13]. The TakeAim CLL design enriches for MRD-positive partial responders on zanubrutinib, which is a reasonable population for an add-on, but it does not pre-select by MYD88 mutation status. If the uMRD readout in Q4 2026 comes back flat or modest, the signal does not differentiate from BTKi alone. Safety risk is concrete and historical. The earlier emavusertib AML/FLT3 program hit an FDA partial clinical hold in April 2022 after cases of rhabdomyolysis, a skeletal-muscle breakdown syndrome that releases myoglobin into the blood and damages kidneys [9]. The hold was eventually resolved and dosing strategies were adjusted, but the signal exists and any recurrence in the B-cell program would be disqualifying. CLL patients are often older and on multiple medications, which raises the bar for tolerability of any combination. Competitive class risk has narrowed: Kymera's KT-413, the IRAK4-IMiD heterobifunctional degrader that was the most direct clinical competitor in MYD88-mutant B-cell lymphoma, was discontinued by Kymera after Phase 1 despite hitting expected pharmacology, leaving emavusertib as the lead IRAK4 oncology clinical asset [11]. KT-474 (Sanofi-licensed) remains in autoimmune development. Commercial risk: BTKi plus venetoclax, with or without obinutuzumab, is the new CLL standard for frontline patients seeking time-limited therapy [10]. Pirtobrutinib owns the BTKi-resistant space. The window for an IRAK4 add-on is narrow and patient-specific.

Watch, with skepticism. The signal that would move emavusertib from interesting to credible is the Q4 2026 interim readout showing uMRD rates clearly above the BTKi monotherapy historical benchmark in the BTKi-pretreated, MRD-positive population, paired with a clean safety profile and no rhabdomyolysis recurrence [10]. Note the assay caveat: TakeAim CLL uses ClonoSEQ NGS at 10^-6 sensitivity, which is more stringent than the typical flow-cytometry-at-10^-4 ibrutinib benchmark, so any cross-trial comparison needs to adjust for assay floor. A pre-specified biomarker hypothesis (MYD88 mutation status or post-BTKi resistance genotype) would sharpen the read substantially. The PCNSL and Waldenstrom's arms of the TakeAim Lymphoma trial (NCT03328078) remain the more biologically anchored shots on goal, given MYD88 L265P prevalence of roughly three-quarters in PCNSL and over 90% in WM [2][13]. The Kymera KT-413 discontinuation leaves emavusertib with less direct class competition in oncology, but it also removes external validation of the IRAK4-degrader thesis in MYD88-mutant lymphoma [11]. Check back at the Q4 2026 data drop and at each quarterly earnings report through then, with attention to two things: cash runway and PIPE warrant exercise status against announced trial expansions, and any disclosed interim data from the zanubrutinib combination or the PCNSL and WM cohorts [6][7][12]. For Curis as an equity, emavusertib is functionally the whole story, and the going-concern overhang means any signal moves the stock both ways. For the IRAK4 thesis more broadly, the field needs a clean monotherapy or combination win somewhere in oncology before this becomes a target other large companies seriously revisit.