Emiltatug ledadotin

Emiltatug ledadotin (Emi-Le, XMT-1660) is investigational and has no marketing approval in any jurisdiction. It is being studied in NCT05377996, a phase 1/2 first-in-human trial with dose escalation, dose expansion, and a phase 2 part (EMBLEM-1) in aggressive adenoid cystic carcinoma (ACC); the registry record was last updated in June 2026 with status recruiting, an estimated enrollment of 360 participants, and an estimated primary completion date of February 2027. The FDA granted Breakthrough Therapy designation on or before 12 May 2026 for locally advanced, recurrent or metastatic ACC with solid histology or high-grade transformation. Earlier FDA Fast Track designations cover advanced or metastatic triple-negative breast cancer and HER2-low or HER2-negative breast cancer after a topoisomerase-1 inhibitor ADC. EMBLEM-1 is testing 80 mg/m2 every 4 weeks (capped at 160 mg) with objective response rate as the primary endpoint, and the sponsor described it as a “pivotal Phase 2 study”. In its February 2026 annual report, Day One stated that it “may seek” orphan drug designation for Emi-Le; no orphan drug designation, biologics license application, or regulatory submission outside the United States was identified. Under the acquisition agreement, contingent payments are tied to first dosing in a registrational ACC-1 trial by 31 December 2027 and FDA approval in ACC-1 by 31 December 2030.

Emiltatug ledadotin has no approved indication. The FDA Breakthrough Therapy designation covers locally advanced, recurrent or metastatic ACC with solid histology or high-grade transformation. The phase 1 ACC cohort enrolled adults whose disease met a clinical definition of aggressive ACC and/or a molecular definition (activating NOTCH1 to NOTCH4 mutations, c-Myc positivity, or p63-negative or low tumors). EMBLEM-1 requires both a histopathologic criterion (solid or basaloid histology, or high-grade transformation) and a clinical criterion (progression within 3 years or extrapulmonary metastasis), which matches the stricter definition used in the post hoc analysis of 32 evaluable participants presented at ASCO 2026. Positive B7-H4 expression is not listed as an EMBLEM-1 enrollment criterion in the sources identified; tissue is submitted for B7-H4 testing. The same trial also enrolled participants with breast, endometrial, and ovarian cancers, and two dose expansion cohorts enrolled participants with triple-negative breast cancer previously treated with a topoisomerase-1 inhibitor ADC.

No prescribing information, boxed warning, contraindication, or risk evaluation and mitigation strategy exists because emiltatug ledadotin is investigational. In the ASCO 2026 abstract safety set (221 participants, data cutoff 1 October 2025), the most common treatment-related adverse events were proteinuria (49.8%), transient AST increase (49.3%), and fatigue (41.2%); grade 3 AST increase and grade 3 proteinuria each occurred in 17.6%, and 3.6% discontinued treatment because of treatment-related adverse events. The oral presentation reported 180 participants with transient AST increase (56%), proteinuria (54%), fatigue (42%), and nausea (33%), with no grade 4 or 5 events. Among 48 participants with ACC, treatment-related adverse events led to dose reduction in 45.8% and dose delay in 56.3%, and proteinuria (characterized on biopsy as glomerular podocytopathy) was the leading cause of dose modification. After a protocol amendment that added monitoring and renal-protective medications, proteinuria-related dose reductions decreased from 38.9% to 20% and delays from 50.0% to 6.7% among participants with ACC. Neutropenia (2.2%), keratitis (3.9%), and peripheral neuropathy (2.8%) were reported at low rates. Severe adverse events observed with UpRi, an earlier Mersana ADC that used the same auristatin payload, included hemorrhage, proteinuria, pneumonitis, and death.

Dose escalation evaluated 7.2 to 115 mg/m2 per cycle on three schedules (every 4 weeks, every 3 weeks, and days 1 and 8 of a 28-day cycle). Participants with ACC received 57.4 to 89 mg/m2 per cycle every 3 or 4 weeks. The first expansion dose (Dose A) was 67.4 mg/m2 every 4 weeks, selected from the intermediate dose range (38.1 to 67.4 mg/m2); at doses at or above 76.2 mg/m2, proteinuria-related dose delays prevented confirmation of some responses. A second expansion dose (Dose B) used 44.5 mg/m2 on days 1 and 8 of the first 4-week cycle followed by 80 mg/m2 every 4 weeks, and an 80 mg/m2 every-4-week cohort was also evaluated. The dose selected for the EMBLEM-1 phase 2 part is 80 mg/m2 every 4 weeks, capped at 160 mg. No labeled dose, dose-adjustment table, or recommended dose for any approved use exists.

Emiltatug ledadotin is available only within the NCT05377996 phase 1/2 trial, which the registry lists as recruiting, with an estimated enrollment of 360 participants aged 18 years or older. The registry record indicates that expanded access is not available. No commercial distribution model, specialty pharmacy arrangement, or site-of-care restriction has been described.

No systemic therapy is approved for recurrent or metastatic ACC, so no approved comparator exists. Emiltatug ledadotin is the only B7-H4-directed ADC identified with an FDA Breakthrough Therapy designation in ACC and a sponsor-led phase 2 trial in ACC (EMBLEM-1). In the ASCO 2026 abstract, the objective response rate was 40% (10 of 25 evaluable participants with ACC); the oral presentation reported 35.6% in 45 evaluable participants with ACC and 46.9% in 32 evaluable participants with aggressive ACC. A patient foundation described historical response rates in ACC trials as rarely above 15%. Other B7-H4-directed ADCs use topoisomerase I inhibitor payloads and have been developed mainly in gynecologic and breast cancers. Puxitatug samrotecan (AZD8205) produced 90% complete responses in two ACC-I patient-derived xenograft models and is in an investigator-sponsored phase 2 trial in ACC; in its first-in-human study (46 participants, no ACC cohort), grade 3 or higher neutropenia occurred in 37.0%. Mocertatug rezetecan (GSK5733584) is entering phase 3 trials in ovarian and endometrial cancers, and its grade 3 or higher adverse events were described as predominantly hematologic. Felmetatug vedotin (SGN-B7H4V), whose phase 1 trial listed ACC among its conditions, was terminated. In ACC, the reported safety profile of emiltatug ledadotin is dominated by proteinuria and transient AST increase, with neutropenia in 2.2% of all participants. Non-ADC investigational agents in ACC include the oral MYB mRNA degrader REM-422 (3 partial responses in 7 biomarker-positive participants at the recommended phase 2 dose), the MDM2 inhibitor alrizomadlin, and the gamma-secretase inhibitor AL101. No head-to-head or indirect comparisons were identified.

Adenoid cystic carcinoma (ACC) is a malignancy of secretory glands that arises most often in the major and minor salivary glands of the head and neck, and less often in the tracheobronchial tree, breast, lacrimal glands, and other sites. It accounts for approximately 1% of head and neck malignancies. Tumors are biphasic (ductal and myoepithelial cells) and grow in cribriform, tubular, and solid patterns, often mixed within a single tumor. No lifestyle risk factors such as smoking or alcohol use have been linked to ACC.

The t(6;9) translocation producing the MYB-NFIB fusion is the hallmark genetic event. The reported frequency varies by detection method (28% of primary tumors by RT-PCR in one series of 123 salivary cancers; 40% to 70% by FISH), while MYB RNA expression is detected in 80% to 90% of tumors regardless of fusion status. Activating NOTCH pathway alterations, most often in NOTCH1, occur in a subset of tumors; in a genomic analysis of 1,045 ACCs, NOTCH1 alterations were present in 26.3% of recurrent or metastatic tumors and 8.5% of primary tumors.

Proteogenomic profiling of 54 tumors at MD Anderson Cancer Center defined two molecular subtypes: ACC-I (20 of 54, 37%), with MYC upregulation, enrichment of NOTCH-activating mutations, enrichment for solid histology, and median overall survival of 3.44 years; and ACC-II (34 of 54, 63%), with TP63 upregulation, predominantly cribriform and tubular histology, and median overall survival of 23.2 years. MYC and p63 immunohistochemistry reproduced this classification. Solid histology and high-grade transformation (a rare progression with nuclear enlargement and irregularity, higher mitotic counts, and loss of the biphasic ductal-myoepithelial differentiation) define the aggressive ACC population addressed in this report. Day One Biopharmaceuticals reports that up to 40% of patients have this aggressive form.

Population-level incidence data are reported for ACC overall, not for the solid or high-grade transformation subgroup. Reported incidence of ACC is 3 to 4.5 per million, and approximately 1,200 to 1,300 new cases are diagnosed annually in the United States. In the Surveillance, Epidemiology, and End Results (SEER) program, 3,026 head and neck ACC cases were recorded between 1973 and 2007 (mean age at diagnosis 57.4 years), with a decline in incidence over that period. ACC represents 10% to 25% of malignant salivary gland tumors.

No registry reports incidence of aggressive ACC directly. Estimates of the aggressive share come from institutional series: ACC-I accounted for 20 of 54 tumors (37%) in a proteogenomic cohort; high-grade transformation was identified in 23 of 412 ACC cases (5.6%) at a single center in China; and solid-type histology was present in 27.5% of 58 cases at a Japanese institution. Day One Biopharmaceuticals reports that up to 40% of patients have aggressive ACC.

No published source reports the prevalence of ACC with solid histology or high-grade transformation. For ACC overall, Servier reports that more than 200,000 people have ACC worldwide. For all salivary gland cancers, the estimated global 1-year prevalence is about 42,000 cases (GLOBOCAN 2022). A 2020 ASCO abstract on the epidemiology of ACC in the United States could not be retrieved for quotation. Because survival in aggressive ACC is shorter than in ACC-II (median overall survival 3.44 versus 23.2 years in a proteogenomic cohort of 54 participants), the aggressive subgroup is expected to represent a smaller share of prevalent cases than of incident cases.

ACC overall has an indolent early course with continued late recurrence and mortality. In SEER (3,026 head and neck cases), 5-, 10-, and 15-year survival was 90.3%, 79.9%, and 69.2%. In institutional series, 5-year survival of 74% to 85% declines to 50% to 64% at 10 years and 20% at 25 years, and more than 50% of tumors recur or metastasize. Mortality continues beyond 20 years: a review of long-followed series reports 5-, 10-, and 20-year survival of 68%, 52%, and 28% in one cohort of 105 participants, 40% of participants alive at 20 years in a United Kingdom series in which the actuarial primary site recurrence rate at 30 years was 100%, and attributes the low long-term survival to failure to control distant disease. In a contemporary cohort of 161 participants at one center, 86% of whom had recurrent disease, median overall survival from diagnosis was 13.9 years (95% CI 9.39 to 17.8), and NOTCH1 status was associated with survival from diagnosis (P = 0.040) and from recurrence (P = 0.007).

Solid histology and ACC-I status are associated with shorter survival. In the original Szanto grading series (79 participants), 15-year survival was 39%, 26%, and 5% for grades I, II, and III (predominantly solid), and grade III tumors caused death within 4 years. In a proteogenomic cohort of 54 participants, median overall survival was 3.44 years for ACC-I and 23.2 years for ACC-II; a 2026 meta-analysis of 247 participants from 5 cohorts found a pooled hazard ratio for death of 3.88 (95% CI 2.55 to 5.90) for ACC-I versus ACC-II. NOTCH1-mutant tumors had median overall survival of 30 months versus 122 months for wild-type tumors. High-grade transformation carries the poorest reported outcomes. In a 9-participant series, 5 died within 5 years and median overall survival was 12 months. In a 1:1 matched series of 23 transformed and 23 non-transformed cases drawn from 412 ACCs, mortality was 47.8% (11 of 23) versus 8.7% (2 of 23), 3-year mortality was 34.8% (8 of 23) versus 4.3% (1 of 23), average survival was 35.9 versus 55.9 months, and median progression-free survival was 18.0 versus 37.0 months, although median follow-up was shorter in the transformed group (29.0 versus 39.0 months); distant metastasis occurred in 78.3% (18 of 23), the lung was the first site in 83.3% (15 of 18), and median survival after metastasis was 8.0 versus 16.5 months. A pathology review describes salivary gland carcinomas with high-grade transformation as more aggressive than conventional carcinomas, with higher local recurrence, a high propensity for cervical lymph node metastasis, and poorer prognosis. After distant metastasis, solid-type ACC had shorter survival and more frequent multiorgan metastases (17 of 28, 61%) than non-solid ACC, and metastases outside the lung were associated with shorter survival (median 11.04 versus 44.52 months in one series of 51 participants).

Conventional ACC is composed of luminal ductal cells and abluminal myoepithelial cells. MYB activation, through MYB-NFIB or MYBL1-NFIB fusion or other mechanisms, is the common oncogenic event, and MYB/MYBL1 fusion rates do not differ between ACC-I (65%) and ACC-II (64%). The aggressive phenotype is associated with loss of myoepithelial differentiation, solid growth, MYC upregulation (through NOTCH activation or MYC amplification), and suppression of p63 signaling. NOTCH1 alterations are enriched in recurrent or metastatic tumors (26.3% versus 8.5% in primary tumors). High-grade transformation shows chromosomal gains confined to the transformed areas, including C-MYC gains. ACC-I tumors are immune-excluded and overexpress the immune checkpoint B7-H4, whose expression level was an independent predictor of poor survival.

ACC is diagnosed by biopsy with histology and immunohistochemistry, after CT or MRI to define local extent and stage. Grading is based on the solid component: the Perzin/Szanto system uses a >30% solid cutoff, the Spiro system >50%, the van Weert system the presence of any solid component, and the WHO notes a worse course when the solid component exceeds one third of the tumor. Interobserver agreement was moderate for the Perzin/Szanto (kappa 0.393) and Spiro (kappa 0.433) systems and 0.990 for the van Weert system. High-grade transformation is distinguished from solid ACC by nuclear enlargement and irregularity, higher mitotic counts, and loss of biphasic differentiation.

Molecular testing supports diagnosis and subtyping. MYB::NFIB FISH had 42.9% sensitivity and 100% specificity, and a MYB RNA in situ hybridization model had 93.2% sensitivity and 97.1% specificity in 79 cases. ACC-I and ACC-II can be assigned with MYC and p63 immunohistochemistry (AUROC 0.99). NOTCH pathway activation can be identified by genomic sequencing or NICD1 immunohistochemistry.

Long-term morbidity in ACC results from the tumor (perineural spread with facial pain, paralysis, and cranial neuropathies), from repeated local therapy for recurrences, and from the need for lifelong follow-up because distant metastases can occur more than 20 years after diagnosis. In a multinational study of 60 survivors of salivary gland cancer (all histologies) more than 5 years after diagnosis, the most frequently reported long-term problem was dry mouth (21 participants, about one third), and the most frequent grade 2 to 3 clinician-assessed toxicities were hearing impairment (13), soft tissue fibrosis (9), dry mouth (8), and cranial neuropathies (8). No study reporting long-term morbidity specifically for solid or high-grade ACC was identified.

The only ACC-specific HRQoL study identified used the EQ-5D-5L in 161 participants (563 questionnaires, 2019 to 2023) at an experimental medicine center: median EQ-5D value was 0.81 (mean 0.79) and median visual analogue scale score was 70, and a decrease in value from 1 to 0 was associated with an eightfold increase in the risk of death. The mean value of 0.79 is below the age-matched general population mean of 0.86. Utility declined significantly over time, and adding NOTCH1 status, primary site, sex, age, recurrence status, or type of recurrence to the time model did not improve its fit, so no clinical predictor of decline was identified. In a 5-year prospective cohort of 252 participants with salivary gland cancer (all histologies), 41 (16.2%) had a declining quality-of-life trajectory and 79 (31.3%) had delayed recovery; recurrence, worse baseline pain, female sex, and older age were associated with unstable trajectories. The only interventional HRQoL data in recurrent or metastatic ACC come from the lenvatinib phase II trial, in which general quality-of-life scores fell by more than 10 points from baseline to 6 months for role functioning, global health, fatigue, pain, and diarrhea and by more than 10 points to 12 months for those domains plus physical functioning, social functioning, and appetite loss; head and neck scores fell by more than 10 points for sexuality, dry mouth, sticky saliva, coughing, and weight gain; EQ-5D-derived utilities ranged from 0.7 to 0.9 across the 12 months; and health status was below that of the general Italian population at every time point. No HRQoL data were identified separately for solid or high-grade transformation ACC.

No cost-of-illness or healthcare resource utilization study specific to ACC or to aggressive ACC in the United States was identified. In United States administrative claims data (2005 to 2007 index dates) for 6,812 participants with oral cavity, oropharyngeal, or salivary gland cancer, total annual healthcare spending in the year after diagnosis was $79,151 in the commercially insured population, and spending exceeded that of matched comparators by payer type ($71,732 commercial, $35,890 Medicare, $44,541 Medicaid). Multimodality treatment cost twice as much as single-modality treatment, and employed participants had 44.9 more short-term disability days than comparators. In a German patient-level cost-effectiveness analysis in head and neck ACC, adding a carbon ion boost to IMRT increased overall costs by €18,076 with an ICER of €26,863 per life-year gained.

ACC is diagnosed mainly in adults (mean age at diagnosis 57.4 years in SEER; range 11 to 99 years), so family economic impact arises chiefly through caregiving and lost income rather than pediatric care. No study of caregiver or family costs specific to ACC was identified. In a pilot survey of participants with head and neck cancer at a United States comprehensive cancer center, high financial toxicity (COST score below 17.5) was reported by 25.9% of 27 patients and 44.4% of 9 caregivers. Among commercially insured employees with oral cavity, oropharyngeal, or salivary gland cancer, 281 participants had 44.9 more short-term disability days than matched comparators.

Identification of aggressive ACC relies on histologic assessment of the solid component and of high-grade transformation on routine histologic sections, supported by immunohistochemistry. Molecular subtyping by MYC and p63 immunohistochemistry uses stains available in standard clinical pathology laboratories. Ancillary tests used in ACC include MYB::NFIB FISH, MYB RNA in situ hybridization, and next-generation sequencing for NOTCH1 and other alterations; MYB immunohistochemistry has been described as an inexpensive alternative to FISH. No ACC-specific study of diagnostic testing costs was identified. As a reference price for comprehensive genomic profiling, the Medicare payment amount for FoundationOne CDx (0037U) during its new advanced diagnostic laboratory test initial period was $3,500.00.

No systemic therapy is approved by the FDA for recurrent or metastatic ACC, and the ESMO-EURACAN guideline states that no systemic treatment has been shown to improve overall survival. Guideline-endorsed systemic options (multitargeted VEGFR tyrosine kinase inhibitors and platinum-based chemotherapy) are supported by single-arm phase II trials and one randomized phase II trial, and NCCN lists lenvatinib and axitinib plus avelumab as category 2B options.

Objective responses are infrequent. A meta-analysis of 17 VEGFR inhibitor studies (560 participants) reported a pooled objective response rate of 6%, with stable disease as the most frequent best response (82%). Response rates were 15.6% with lenvatinib (32 evaluable participants), 11.5% in a second lenvatinib phase II trial (26 evaluable participants), 21.7% in a real-world lenvatinib cohort (46 evaluable participants), 9.1% with axitinib in ACC (33 participants), 8% with axitinib in a single-center trial in salivary gland carcinomas (26 participants), 0.0% with axitinib in the randomized trial versus observation, 11% with sorafenib (19 evaluable participants), 18% with axitinib plus avelumab (28 evaluable participants), and 9.7% by blinded independent review with rivoceranib (72 evaluable participants). Progression-free survival captures activity that the response rate does not: in the randomized axitinib trial the 6-month PFS rate was 73.0% with axitinib and 23.0% with observation and the disease control rate was 100.0% versus 51.9%, and median PFS was 10.8 versus 2.8 months (hazard ratio 0.25), yet neither arm produced a response and there was no statistically significant overall survival difference (P = 0.226), which the investigators attribute in part to crossover. Cytotoxic chemotherapy produced objective responses in 22 of 163 participants (16%) across single-agent studies and 25% with CAP, with greater toxicity for combinations.

Tolerability limits treatment duration and dose intensity. In the lenvatinib phase II trial reported by Tchekmedyian, 23 of 32 participants required dose modification and 18 of 32 discontinued for drug-related issues; in the Italian lenvatinib phase II trial, 96% had a treatment-related adverse event, grade 3 events occurred in approximately half of participants, dose reductions were required in 24 of 28, and 88% of first reductions occurred within 12 weeks. In the real-world lenvatinib cohort, every participant had a treatment-related adverse event, the most frequent grade 3 or higher events were hypertension (25.49%), stomatitis (17.65%), and fatigue (15.69%), 32 of 51 participants (63.6%) required a dose reduction, one fatal drug-related bleeding occurred, and only 16 of 51 (31.37%) started at the full 24 mg dose. With sorafenib, 13 of 23 participants (57%) had grade 3 toxicity and 74% required a dose reduction, and the investigators concluded that single-agent sorafenib could not be recommended for further evaluation. With axitinib in salivary gland carcinomas, all participants had a drug-related adverse event, grade 3 or higher events occurred in 12 of 26 (46%), median dose intensity was 59%, and the trial did not meet its primary endpoint. With rivoceranib, 70.0% had grade 3 or higher treatment-related adverse events and 85.0% required dose modification. The pooled dose reduction rate for VEGFR inhibitors was 59%. Because participants with recurrent or metastatic ACC often have few symptoms and prolonged life expectancy, even grade 1 to 2 adverse events can reduce quality of life, which deteriorated for fatigue and dry mouth by 6 months on lenvatinib.

Outcomes are worse in the aggressive subgroup targeted by emiltatug ledadotin. With axitinib plus avelumab, median PFS was 1.8 months in ACC-I versus 11.4 months in ACC-II, and in a real-world cohort ACC-I was associated with worse outcomes across pooled systemic treatment categories (HR 3.06). NOTCH inhibitors in NOTCH-activated disease (86% solid histology) produced a median PFS of 4.2 months. Pulmonary metastasis-directed local therapy did not increase overall survival in unselected patients, and active surveillance is recommended mainly for asymptomatic, low-burden, lung-only disease, which is less typical of solid or high-grade transformed ACC. Subtype-specific comparator evidence is otherwise scarce: the randomized axitinib trial did not capture detailed information on pathologic subtype, and the claims-based comparison of ACC-I and ACC-II by treatment type had limited sample sizes per treatment cohort and no information on clinical response.

Emiltatug ledadotin is investigational. Its anticipated population is locally advanced, recurrent or metastatic ACC with solid histology or high-grade transformation, the population named in the FDA Breakthrough Therapy designation. In current guidelines, systemic therapy for R/M ACC is started when metastases are symptomatic and not amenable to local therapy, threaten organ function, or have grown more than 20% in 6 months; clinical trial participation is the preferred option, followed by multitargeted VEGFR tyrosine kinase inhibitors or platinum-based chemotherapy. No agent is approved or designated as preferred, so emiltatug ledadotin, if approved, would enter a setting without an established first-line systemic standard. Based on the existing sequencing framework, it would most likely be used in progressive or symptomatic aggressive ACC, either before or after a VEGFR tyrosine kinase inhibitor; the ASCO 2026 interim analysis did not define a line of therapy, and the sponsor has stated that a pivotal phase 2 study is planned.

Emiltatug ledadotin is administered intravenously, which places it in a hospital-based or clinic-based infusion setting, in contrast with the oral VEGFR tyrosine kinase inhibitors. Given the rarity of ACC, guideline recommendations for management by multidisciplinary teams in specialized head and neck units apply.

Molecular subtype and histology are the principal modifiers described. ACC-I, enriched for NOTCH1-activating mutations and MYC, has a median overall survival of approximately 3 years versus 23 years for ACC-II. NOTCH1-mutant tumors are associated with solid histology, liver and bone metastases, and shorter overall survival (median 30 versus 122 months). ACC-I tumors overexpress B7-H4, and B7-H4 expression is an independent predictor of poor survival; in patient-derived xenografts, the B7-H4-directed ADC AZD8205 produced complete responses only in ACC-I/B7-H4-high models.

With axitinib plus avelumab, objective response rates were similar in ACC-I and ACC-II (15% and 20%), but median PFS was 1.8 versus 11.4 months and progression as best response occurred in 62% versus 0%. In real-world data, ACC-I was associated with worse outcomes across systemic treatments (HR 3.06). High-grade transformation (solid pattern in 69.6%) was associated with distant metastasis in 78.3%. Solid histology was the strongest predictor of shorter survival after pulmonary metastasis. Among emiltatug ledadotin-treated participants, the objective response rate was 35.6% in all evaluable ACC (45 participants) and 46.9% in the post hoc aggressive ACC subgroup (32 participants), a comparison that is not randomized. With rivoceranib, investigator-assessed response rates differed by prior VEGFR inhibitor exposure (18.6% versus 0%), and bone metastasis was an independent poor prognostic factor with VEGFR-TKIs.

Most comparator studies did not analyze effect by subtype, and the modifiers they did examine were mainly clinical. The randomized axitinib trial found that mutation burden and MYB fusion did not correlate with response, progression-free survival, or overall survival, that FANCM and PCTH2 mutations were respectively positive and negative predictors of progression-free survival, and that it had not captured detailed information on pathologic subtype. In the real-world lenvatinib cohort of 51 participants, site of metastatic disease, including bone and liver, was not associated with activity, younger age was associated with a higher probability of disease control, and the higher response rate seen with a 24 mg starting dose was not statistically significant. In the sorafenib trial, 19 of 23 participants had documented progression at entry, and of the 4 with stable disease at entry none responded, which indicates that apparent benefit depends on whether progression is required for enrollment. In the axitinib trial in salivary gland carcinomas, the only responder with four mutated genes was a non-ACC tumor and best response was stable or progressive disease in all 6 tumors without recurrent mutations. The claims-based ACC-I and ACC-II comparison found similar survival with chemotherapy and with VEGFR tyrosine kinase inhibitors across subtypes, but had limited sample sizes per treatment cohort and no clinical response data. In high-grade transformation, therapeutic targets such as HER2 overexpression may be confined to the high-grade component, so biopsy of metastatic disease may be needed to confirm which clone the metastasis derives from.

Guidelines recommend multidisciplinary management in specialized head and neck units, tumor sequencing at the time of recurrent or metastatic disease, and consideration of clinical trial participation, including NOTCH inhibitor trials for tumors with activating NOTCH mutations. Patients with residual, recurrent, or metastatic disease are imaged 2 to 4 times per year, with lower frequency when growth is slow, and systemic therapy is initiated on defined symptomatic or growth criteria. During VEGFR tyrosine kinase inhibitor therapy, dose modification is frequent (85.0% with rivoceranib; 23 of 32 participants with lenvatinib; 32 of 51 participants, 63.6%, in a real-world lenvatinib cohort), and hypertension, stomatitis, fatigue, and bleeding events require monitoring, including one fatal drug-related bleeding in the real-world cohort. For emiltatug ledadotin, the most common grade 3 treatment-related adverse events in the phase 1 analysis were AST increase (20%) and proteinuria (23%), which indicates a need for liver enzyme and urine protein monitoring; treatment-related adverse events led to discontinuation in 3.9% of participants.

The goals of systemic therapy in R/M ACC are tumor control, symptom palliation, and prolongation of progression-free survival, because no systemic treatment has been shown to improve overall survival. Survival benchmarks differ by subgroup: after pulmonary metastasis, median overall survival was 7.2 years in a cohort of 219 participants, whereas ACC-I has a median overall survival of approximately 3 years, and high-grade transformation was associated with an average survival of 35.9 months. After surgery and radiotherapy for high-grade transformation, median progression-free survival was 18.0 months compared with 37.0 months in matched non-transformed ACC. Treatment benchmarks include a pooled VEGFR inhibitor objective response rate of 6% and 6-month disease control rate of 54%; a 6-month PFS rate of 73.0% with axitinib versus 23.0% with observation and a median PFS of 10.8 versus 2.8 months in the randomized trial, with a median PFS of 14.5 months in participants who crossed over to axitinib after progression; a 6-month PFS rate of 69.3% and a 12-month rate of 46.2% with sorafenib, with median PFS of 11.3 months and median overall survival of 19.6 months; a median PFS of 14.8 months, a 6-month PFS rate of 75.7%, and a median overall survival of 16.1 months in a real-world lenvatinib cohort of 51 participants; and a median PFS of 1.8 months and median overall survival of 11.3 months in ACC-I treated with axitinib plus avelumab.

Against these benchmarks, the interim phase 1 analysis of emiltatug ledadotin reported, in 32 evaluable participants with aggressive ACC (solid histology or high-grade transformation), an objective response rate of 46.9%, a disease control rate of 81.3%, and a median PFS of 7.8 months. In the same post hoc subset, median duration of response was 6.4 months. At the earlier October 1, 2025 data cut-off, median overall survival had not been reached in participants with ACC. These are single-arm, post hoc, interim data, and quality of life outcomes have not been reported in the cited sources.