Adenoid Cystic Carcinoma (ACC): Comprehensive Research Report

1. Disease Information

Overview. Adenoid cystic carcinoma (ACC) is a rare, histologically distinctive malignant epithelial neoplasm that most commonly arises in the major and minor salivary glands of the head and neck but also occurs in the breast, lacrimal gland, trachea/bronchus, skin, and uterine cervix. It is a biphasic tumor composed of malignant epithelial (ductal/luminal) and myoepithelial (abluminal/basaloid) cells, and it is genomically defined by recurrent rearrangements activating the MYB transcription-factor family. Clinically, ACC is characterized by an unusual combination of indolent growth with relentless local infiltration, a striking propensity for perineural invasion, frequent late local recurrence, and delayed distant metastasis (commonly to lung, bone, and liver) that can occur many years to decades after initial diagnosis (PMC3597152, PMC11387731).

Key identifiers: - MONDO: MONDO:0004971 (adenoid cystic carcinoma, general); site-specific children include MONDO:0003175 (salivary gland ACC) and MONDO:0003181 (lung ACC) (monarchinitiative.org) - ICD-O-3 morphology: 8200/3 (malignant) — combined with topography codes for site (e.g., C07 parotid gland, C08.0 submandibular gland, C05.0 hard palate, C50.9 breast) - ICD-10: varies by site (e.g., C07, C08.0, C50.91) - MeSH: D003528 (Carcinoma, Adenoid Cystic) - OMIM: No dedicated OMIM phenotype entry exists — ACC is predominantly a sporadic, somatically driven malignancy rather than a classic monogenic Mendelian disorder, so it is not catalogued in OMIM the way inherited syndromes are. - Orphanet: site-specific Orphanet entries exist (e.g., ORPHA:213823 for cervical ACC); a general salivary-gland ACC entry is also indexed under Orphanet's rare tumor nomenclature.

Synonyms: cylindroma (historical term, now largely reserved for the benign cutaneous adnexal tumor to avoid confusion), adenocystic carcinoma, cribriform carcinoma (older term reflecting histology), "adenoid cystic basal cell carcinoma" (obsolete).

Evidence provenance: Most quantitative data on ACC (incidence, survival, treatment response) derive from aggregated disease-level resources — national cancer registries (SEER, National Cancer Database), multi-institutional retrospective cohorts, and pooled genomic sequencing cohorts — rather than individual electronic health records, reflecting the tumor's rarity and the resulting reliance on multi-center consortia (e.g., the Adenoid Cystic Carcinoma Research Foundation's genomic sequencing program) (PMC11476411, JCI:128227).

2. Etiology

Disease causal factors. ACC is a genetically/mechanistically driven malignancy rather than one with an established infectious or classical toxic etiology. Its defining molecular event is activation of the MYB transcription-factor family — predominantly via a t(6;9)(q22-23;p23-24) chromosomal translocation producing an MYB-NFIB gene fusion, first reported by Persson et al. (Proc Natl Acad Sci USA, 2009;106:18740–18744; PMCID PMC2773970). This translocation truncates MYB, removing its 3′ untranslated region and thereby escaping microRNA-mediated (miR-15a/16, miR-150) negative feedback regulation, leading to MYB overexpression (PMC11387731).

Genetic risk/causal factors: - MYB-NFIB fusion — the genomic hallmark, detected in ~28–86% of primary ACCs and up to 35% of metastatic ACCs depending on cohort and assay (meta-analysis: Mitani et al., systematic review PMID:30269389). Serves as "a specific and sensitive marker to distinguish ACC from other salivary gland tumors." - MYBL1 (A-MYB) rearrangements (MYBL1-NFIB, MYBL1-YTHDF3) — found in MYB-fusion-negative tumors (~2.4–8% of cases), mutually exclusive with MYB alterations, and reported to skew toward submandibular gland primaries (PMID:29149504, "MYBL1 rearrangements and MYB amplification in breast adenoid cystic carcinomas lacking the MYB-NFIB fusion gene"). - High-level MYB amplification (copy-number gain) as an alternative, fusion-independent route to MYB overexpression. - Nonclassical fusions (MYB-TGFBR3, MYB-RAD51B*) in a small subset (~2.2%) (PMC11387731). - NOTCH1 activating mutations in the negative regulatory region (NRR) and PEST domain hotspots shared with T-cell acute lymphoblastic leukemia — found in ~13.7–20% of primary tumors and markedly enriched (26.3% vs. 8.5%) in recurrent/metastatic disease (Ferrarotto et al., J Clin Oncol 2017;35(3):352-360, DOI:10.1200/JCO.2016.67.5264; Ho et al., genetic hallmarks study, J Clin Invest 2019, DOI:10.1172/JCI128227). - TERT* promoter mutations — present in ~13.1% of recurrent/metastatic ACC and mutually exclusive with both NOTCH1 mutations and MYB/MYBL1 fusions, suggesting a distinct alternative oncogenic route (JCI:128227). - Chromatin-remodeling gene alterations enriched in recurrent/metastatic disease relative to primary tumors: KDM6A (15.2% vs 3.4%), KMT2C/MLL3 (14.3% vs 4.0%), ARID1B (14.1% vs 4.0%), ARID1A (13.7% vs 2.3%) (JCI:128227). - DNA-damage-repair gene alterations (ATM, enriched 6.8% vs 1.7% in recurrent/metastatic disease) (JCI:128227). - Rare germline predisposition: isolated familial reports associated with germline BRCA2 mutations, but no established hereditary cancer syndrome accounts for a meaningful fraction of cases; genetic predisposition otherwise plays a limited, largely unproven role.

Environmental risk factors: - Prior therapeutic ionizing radiation to the head/neck (e.g., childhood radiotherapy for benign conditions or other malignancies) is the most established environmental/iatrogenic risk factor, with elevated risk manifesting 10–20 years after exposure, attributed to radiosensitivity of salivary gland tissue. - No consistent association has been demonstrated with tobacco smoking or alcohol consumption — in contrast to most other head and neck carcinomas. - No established viral or infectious etiology (unlike, e.g., HPV-driven oropharyngeal SCC or EBV-driven nasopharyngeal carcinoma). - Age and sex: peak incidence in the 4th–6th decades (median age ~58 years per SEER; PMID:39410002), with a female predominance (female:male ratio approximately 3:2, and 63.3% female in the SEER cohort).

Protective factors: No specific genetic or environmental protective factors have been established in the literature; this is an area of relative research gap for ACC (in contrast to more common cancers where lifestyle/dietary protective factors are better characterized).

Gene-environment interactions: No well-documented gene-environment interaction has been established for ACC; the disease is best modeled as a somatic-driver-defined malignancy with radiation as the principal known extrinsic contributor, acting independently of germline genotype in essentially all reported cases.

3. Phenotypes

ACC's phenotype spectrum is dominated by mass-effect and neurotropic (perineural) manifestations rather than systemic/laboratory abnormalities, consistent with a locally aggressive solid tumor.

Age of onset: ACC spans the first to ninth decades but is most frequent in middle-aged and older adults (peak 45–60 years); rare pediatric cases occur (PMC11387731).

Severity/progression: Best characterized as "indolent but aggressive" — slow radiographic growth juxtaposed with a high propensity for perineural spread, positive margins, and eventual distant relapse. Course is typically chronic and progressive with a prolonged natural history (median overall survival reported around 16 years in some cohorts) but a persistent risk of recurrence that does not plateau even after 10–15 years, mandating indefinite surveillance (PMC10163974, SEER-based studies).

Quality of life impact: Cranial nerve deficits (facial paralysis, numbness, masticatory dysfunction), dysphagia, and disfigurement from radical surgery substantially affect quality of life; long-term surveillance imaging and the psychological burden of a disease with a very long "tail" of recurrence risk are notable but under-quantified in standardized instruments (EQ-5D/SF-36 data specific to ACC are sparse in the literature reviewed).

4. Genetic/Molecular Information

Causal/driver genes: - MYB (HGNC:7545; 6q23.3) — via t(6;9)(q22-23;p23-24) fusion to NFIB (HGNC:7784; 9p23-24), or via 3′UTR truncation/amplification. This is the dominant genomic hallmark (Persson et al. 2009, PMCID PMC2773970). - MYBL1 (HGNC:7548; 8q13.1) — alternative driver in MYB-fusion-negative ACC, fusing to NFIB or YTHDF3; mutually exclusive with MYB alterations (PMID:29149504). - NOTCH1 (HGNC:7881; 9q34.3) — recurrent activating mutations clustering in the negative regulatory region (heterodimerization domain) and PEST domain, analogous to T-ALL hotspots (Ferrarotto et al., J Clin Oncol 2017;35:352-360). - TERT promoter — recurrent hotspot promoter mutations in a mutually exclusive subset (JCI:128227).

Variant classification/type: MYB-NFIB and MYBL1 alterations are chromosomal rearrangements/gene fusions (structural variants), essentially always somatic. NOTCH1 and TERT promoter alterations are somatic point mutations/small indels; NOTCH1 mutations are gain-of-function (activating), analogous to leukemia-associated NOTCH1 mutations. There is no established ClinVar/ACMG germline pathogenic-variant framework for ACC, since virtually all reported drivers are somatic tumor events rather than germline predisposition alleles.

Somatic vs. germline: ACC driver alterations are overwhelmingly somatic. COSMIC and TCGA-style sequencing (whole-exome sequencing of lacrimal gland ACC, PMC5562266; genomic landscape studies) confirm this. Rare germline BRCA2 variants have been reported anecdotally in familial clusters but are not an established recurrent germline predisposition mechanism.

Additional genomic alterations (enriched particularly in recurrent/metastatic disease per Ho et al., J Clin Invest 2019, DOI:10.1172/JCI128227): - Chromatin remodeling: KDM6A, KMT2C/MLL3, ARID1A, ARID1B, SMARCA4, SMARCB1, PBRM1 - DNA damage/checkpoint: ATM - FGF/IGF/PI3K signaling pathway alterations - TP53 mutations (associated with recurrent/metastatic disease and solid histologic subtype) - TP63 activation (paradoxically associated with a better prognosis subgroup; induces AXL, EGFR, MET expression) - RAS pathway mutations (associated with worse disease-free and overall survival) - FAT1/FAT3 (Hippo-YAP pathway) mutations; YAP1 alterations common in lung-primary ACC - EGFR mutations (~1–2%), with EGFR overexpression linked to poor prognosis and PD-L1 induction via c-Myc - 1p36 locus deletion — reported as an independent adverse prognostic marker (Virchows Arch, 2018)

Molecular subtyping: Ferrarotto and colleagues proposed an "ACC-I" molecular class characterized by NOTCH-MYC pathway activation, solid histology, minor salivary gland origin, and co-mutation of SPEN, CREBBP, EP300 — associated with significantly worse prognosis (PMC11387731).

Epigenetic information: MYB overexpression is reinforced by a positive-feedback super-enhancer loop in which MYB binds its own enhancer elements; disruption of the normal miRNA-mediated (miR-15a/16, miR-150) post-transcriptional brake via 3′UTR loss is a key epigenetic-adjacent mechanism of MYB dysregulation. Downregulation of chromatin remodeling complex components (SWI/SNF family: SMARCA4, SMARCB1, ARID1A/B, PBRM1) has been reported, implicating global chromatin dysregulation as ACC progresses.

Chromosomal abnormalities: The signature t(6;9)(q22-23;p23-24) translocation generating MYB-NFIB; 1p36 deletions as a secondary recurrent copy-number event associated with prognosis.

Suggested ontology terms: - Genes (HGNC): hgnc:7545 MYB, hgnc:7784 NFIB, hgnc:7548 MYBL1, hgnc:7881 NOTCH1, hgnc:11730 TERT, hgnc:11998 TP53, hgnc:12558 TP63, hgnc:3236 EGFR

5. Environmental Information

• Environmental/occupational factors: No consistently reproduced association with specific toxins, industrial exposures, or air pollutants has been established, unlike some other salivary or sinonasal malignancies (e.g., wood-dust exposure and sinonasal adenocarcinoma). The strongest documented environmental contributor remains prior ionizing radiation exposure to the head and neck.
• Lifestyle factors: Tobacco and alcohol use — unlike most head and neck squamous cell carcinomas — have not been shown to increase ACC risk in the literature reviewed.
• Infectious agents: No infectious/microbial trigger (viral, bacterial, fungal, parasitic) has been implicated in ACC pathogenesis; this distinguishes it from HPV-associated oropharyngeal cancers and EBV-associated nasopharyngeal carcinoma, both of which can present in overlapping anatomic sites.

6. Mechanism / Pathophysiology

Causal chain overview: Trigger (MYB-family transcription-factor dysregulation, principally via MYB-NFIB/MYBL1 fusion or amplification) → constitutive transcriptional activation of proliferative and anti-apoptotic MYB target genes → cooperating/parallel NOTCH1 pathway hyperactivation (particularly in aggressive/recurrent disease) → downstream MYC upregulation, EMT induction, angiogenesis, and immune evasion → biphasic tumor growth with intrinsic myoepithelial/luminal cellular heterogeneity → perineural invasion and local infiltration → (in a subset) late hematogenous metastasis, chiefly to lung.

Molecular pathways: - MYB transcriptional program: MYB (with cooperating partner NFIB) drives a transcriptional program promoting cell cycle progression and inhibiting apoptosis partly through MYC; also upregulates VEGFA and vimentin, contributing to angiogenesis, EMT, and metastatic competence (PMC11387731). - NOTCH-MYC/HES-HEY axis: Activating NOTCH1 mutations (or wild-type pathway hyperactivation via paracrine ligand signaling) drive NOTCH intracellular domain (NICD)-mediated transcription of HES1, HEY1, REST; the NOTCH1-HEY1 axis is implicated in proliferation, invasion, metastasis, and apoptosis resistance. Notably, MYB and NOTCH pathways appear to have opposing effects on cellular differentiation, and single-cell data show that metastatic lesions have elevated MYB/lower NOTCH1 expression while recurrent tumors show the inverse (increased NOTCH signaling, reduced MYB), suggesting dynamic pathway switching across the disease course (Cell Reports single-cell study, PMC9714264/PMID:36465348). - EGFR-PI3K-AKT / MEK-ERK signaling: EGFR overexpression (mutation rate only ~1–2%) is linked to poor prognosis; EGFR activation induces PD-L1 expression via c-Myc, offering one mechanistic link between growth-factor signaling and immune evasion. - Hippo-YAP pathway: FAT1/FAT3 mutations and YAP1 alterations (particularly common in lung-primary ACC). - TGF-β signaling: implicated in the biphasic epithelial/myoepithelial differentiation program. - RAS/MAPK: RAS pathway mutations associated with poorer disease-free and overall survival.

Cellular processes and cell-of-origin. Single-cell RNA-sequencing studies (Lin et al., PMID:36465348; Cell Reports, PMID/PMC9714264) profiled ~49,948 cells from paracarcinoma/carcinoma tissue and identified: - Myoepithelial-like cells (CD49f+) — higher tumorigenic potential than ductal cells, serving as progenitors for ductal differentiation, and expressing NOTCH ligands (DLL1, JAG1, JAG2) that paracrine-signal to adjacent luminal cells. - Intercalated duct-like cells (KRT19+/AQP5+/KIT+) and duct-like cells (KRT19+/AQP5−/KIT−), expressing high levels of NOTCH receptors and NOTCH target genes. - Pre-malignant cells (~47% of paracarcinoma epithelial cells) with copy-number alterations affecting the MYB family and EN1; pseudotime trajectory analysis suggests these differentiate into malignant cells over time, implying an intercalated-duct cell of origin. - Tumor microenvironment composition: epithelial cells ~40.1%, fibroblasts ~27.9% (subdivided into proliferative, myogenic, inflammatory, and fibrotic cancer-associated fibroblast [CAF] subtypes), T/NK cells ~11.3%, endothelial cells ~10.8%.

Immune system involvement / "cold" tumor phenotype: ACC exhibits significantly reduced tumor-infiltrating lymphocytes compared with normal tissue, downregulated CD8+ and CD4+ T-cell populations, absent IgA plasma cells, inhibited canonical PD-1/PD-L1 signaling but activated immunosuppressive PD-L2 and HLA-G, and elevated M2 macrophages/myeloid-derived suppressor cells (with macrophage communication mediated via the MIF-CD74 axis rather than classical HLA-MHC signaling). This immunologically "cold," low-tumor-mutational-burden phenotype underlies the generally poor response of ACC to immune checkpoint inhibition (PMC11387731).

Tissue damage / perineural invasion mechanism: Perineural invasion — one of ACC's most defining pathophysiological features — has been mechanistically linked to brain-derived neurotrophic factor (BDNF) signaling promoting neurotropism (Kulasegaran/Vered et al., PMID:12378520, "Perineural invasion in adenoid cystic carcinoma: Its causation/promotion by brain-derived neurotrophic factor"). PNI classically shows a "target-like arrangement" of tumor cells around nerve fibers and independently predicts worse disease-free and overall survival.

Biochemical/protein-level abnormalities: - c-KIT (CD117) overexpression in ~90% of ACC — associated with enhanced invasiveness, though targeting KIT with imatinib/dasatinib has shown minimal clinical efficacy, implying KIT expression is a bystander/marker rather than a driver. - p53 overexpression in ~50% of cases, associated with solid histologic subtype, higher metastatic potential, and reduced 5-year overall survival. - Bcl-2 overexpression associated with perineural invasion and worse prognosis. - Beclin-1 (autophagy regulator) — lower expression correlates with higher histologic grade. - SOX2/SOX-10 — broadly expressed; SOX2 correlates with clinical progression.

Molecular/omics profiling: - Transcriptomics: Single-cell RNA-seq (above) plus bulk expression studies show MYB target gene programs and NOTCH pathway signatures distinguishing primary from recurrent/metastatic disease. - Genomics: Whole-exome sequencing of lacrimal gland ACC (PMC5562266) and personalized oncogenomic whole-genome sequencing of metastatic ACC (Molecular Case Studies, molecularcasestudies.cshlp.org/content/4/2/a002626) have informed individualized treatment decisions. - Proteomics: Comparative proteomic analysis distinguishes breast ACC from basal-like triple-negative breast cancer despite morphologic and IHC overlap (PMC9366086). - Spatial transcriptomics: Recent work has mapped molecular heterogeneity by pathological grade using combined whole-exome sequencing and spatial transcriptomics (ScienceDirect, S0046817725000450).

Suggested ontology terms: - GO Biological Process: GO:0007219 (Notch signaling pathway), GO:0038095 (Fc-epsilon receptor... not relevant); more precisely GO:0045747 (positive regulation of Notch signaling), GO:0001525 (angiogenesis), GO:0001837 (epithelial to mesenchymal transition), GO:0006915 (apoptotic process), GO:0007050 (cell cycle arrest — inverse), GO:0035633 (maintenance of blood-brain barrier — not relevant), GO:0021675 (nerve development, for perineural invasion biology) - GO Molecular Function: GO:0003700 (DNA-binding transcription factor activity) for MYB/MYBL1 - Cell Ontology (CL): CL:0002326 (luminal epithelial cell of mammary gland — analog for ductal/luminal ACC cells), CL:0000185 (myoepithelial cell), CL:0000066 (epithelial cell) - UBERON: UBERON:0001830 (major salivary gland), UBERON:0001831 (minor salivary gland), UBERON:0001911 (mammary gland)

7. Anatomical Structures Affected

Organ level: - Primary sites (per SEER 2000–2019 cohort of 5,150 cases, PMID:39410002): head and neck 70.1% (parotid and submandibular glands together ~37% of all ACC, oral cavity/minor salivary glands ~22%), breast 14.1%, thoracic (trachea, bronchus, lung) 6.8%, genitourinary (cervix uteri, Bartholin gland, prostate) 2.2%, miscellaneous sites (skin, lacrimal gland, esophagus) 6.8%. - Secondary/metastatic involvement: lung (most common distant metastatic site), bone, liver; regional lymph node metastasis is comparatively uncommon relative to squamous cell carcinoma of the head and neck, reflecting ACC's preference for hematogenous over lymphatic spread. - Body systems involved: exocrine gland tissue (salivary, lacrimal, mammary, tracheobronchial submucosal glands), integumentary system (rare cutaneous ACC), and — via perineural spread — the peripheral and cranial nervous system (facial [VII], trigeminal [V], and other cranial nerves).

Tissue and cell level: - Epithelial (ductal/luminal) and myoepithelial (basaloid/abluminal) cell populations, arranged in cribriform, tubular, or solid architectural patterns. - Perineural tissue — Schwann cells and nerve fibers infiltrated by tumor cells in a "target-like" pattern.

Suggested Cell Ontology terms: CL:0000185 (myoepithelial cell), CL:0000068 (duct epithelial cell), CL:0002327 (mammary luminal progenitor cell — analog).

Subcellular level: Nuclear localization of MYB/MYBL1 and NICD (Notch intracellular domain) as transcription factors (GO Cellular Component: GO:0005634 nucleus); mitochondrial/ER stress pathways implicated in ferroptosis-based therapeutic vulnerability research (GPX4 inhibitor studies).

Localization/lateralization: Typically unilateral at presentation (e.g., unilateral parotid or submandibular mass); bilateral or midline presentations occur with minor salivary gland (palate) or tracheal primaries. Perineural spread can cross the skull base bilaterally in advanced disease via named neural foramina.

Suggested UBERON terms: UBERON:0001831 (minor salivary gland), UBERON:0001830 (major salivary gland; parotid UBERON:0006330, submandibular UBERON:0006331), UBERON:0001911 (mammary gland), UBERON:0003126 (trachea), UBERON:0002048 (lung), UBERON:0000948 (heart — not typically involved), UBERON:0000948... (for cervix: UBERON:0000002 uterine cervix).

8. Temporal Development

Onset: Adult-onset predominant, with a broad age range (first through ninth decades) and median age at diagnosis of ~58 years (IQR 47–70, SEER cohort). Onset is typically insidious — patients often present with a slow-growing, painless mass for months to years before diagnosis, though pain and neurologic symptoms may prompt earlier presentation when perineural invasion is present.

Progression: - Histologic grading correlates with growth pattern and prognosis: Grade I (predominantly tubular) — most favorable; Grade II (predominantly cribriform, <30% solid) — intermediate; Grade III (>30% solid component) — least favorable, associated with higher proliferation, greater local invasiveness, and reduced survival. Historical cumulative 15-year survival by grade: 39% (Grade I), 26% (Grade II), 5% (Grade III), with Grade III tumors often proving fatal within 4 years in early cohorts. More recent analyses suggest tumor stage may be a more robust prognostic indicator than histologic grade alone. - High-grade transformation (HGT): a described phenomenon in which a conventional low/intermediate-grade ACC acquires markedly increased mitotic activity, Ki-67 index, and p53 positivity, associated with a more aggressive clinical course. - Progression rate: locally, growth is slow, but the disease is notable for continuing to progress/recur over a very long time horizon; distant metastasis, when it occurs, has a median time-to-recurrence of approximately 50–51 months post-primary treatment, and events can occur even after 10–15+ years of apparent disease-free survival. - Disease course pattern: best described as chronic-progressive with a long "tail" — unlike many carcinomas that either recur within 2–5 years or are cured, ACC carries persistent recurrence risk indefinitely, necessitating lifelong surveillance.

Patterns: - Remission: Achievable with complete surgical resection ± adjuvant radiotherapy in localized disease, but "cure" is difficult to declare definitively given the long natural history; spontaneous remission is not described. - Critical periods: The window for achieving durable local control is at initial surgical resection (negative margins); once perineural spread has occurred along a named nerve to the skull base, the opportunity for complete resection narrows considerably, making early, meticulous margin-negative surgery a key point of intervention leverage.

9. Inheritance and Population

Epidemiology: - Incidence: approximately 3–4.5 cases per million people per year; ACC represents ~1% of all head and neck malignancies and ~10% of all salivary gland neoplasms (PMC11387731, PMC11476411). - SEER-based cohort (2000–2019, N=5,150; PMID:39410002) provides the most detailed contemporary US population data: - Median age 58 years (IQR 47–70) - Sex: 63.3% female, 36.7% male (female:male ≈ 1.7:1, broadly consistent with the commonly cited ~3:2 ratio) - Race/ethnicity: White 75.9%, Asian 11.2%, Black 10.8%, other/unknown 2.1% - Stage at diagnosis: localized 77.5%, regional 14.3%, distant 8.2% - Site distribution: head/neck 70.1%, breast 14.1%, thoracic 6.8%, genitourinary 2.2%, miscellaneous 6.8%

Inheritance pattern: ACC is essentially always a sporadic, somatically driven malignancy; there is no established Mendelian inheritance pattern (autosomal dominant/recessive, X-linked, mitochondrial) for the disease as a whole. Rare familial clustering has been anecdotally linked to germline BRCA2 mutations, but this does not constitute an established hereditary cancer syndrome specific to ACC, and penetrance/expressivity data for such associations are not robust in the literature. Genetic anticipation, germline mosaicism, founder effects, and consanguinity are not applicable/not established for this predominantly somatic-driver disease.

Population demographics: - Geographic distribution: No strong endemic geographic clustering is reported; SEER regional analysis found some survival variation by US region (Northeast longest median OS at 171 months vs. Midwest 132 months), likely reflecting access-to-care and treatment-pattern differences rather than incidence differences. - Sex ratio: Female predominance (~3:2 to ~1.7:1 depending on cohort/site — notably even more female-predominant in breast-primary ACC, which is inherently sex-skewed given anatomic site). - Age distribution: Broad (1st–9th decades), peak in middle-to-older adulthood (45–70 years). - Ethnicity/hospital-based series suggest some site-distribution variation (e.g., a higher proportion of Hispanic patients showing modestly better 6-year OS: 73.9% vs. 70.4% non-Hispanic in the SEER cohort), though causal interpretation is limited by retrospective registry design.

10. Diagnostics

Clinical/laboratory tests: No specific serum biomarker or routine laboratory test is diagnostic for ACC; diagnosis rests on tissue biopsy with histopathology and immunohistochemistry, supported by imaging for local/perineural extent and staging.

Imaging studies: - MRI is the preferred modality for staging, treatment planning, and surveillance because of superior soft-tissue contrast and multiplanar capability, and is particularly valuable for detecting perineural spread — appearing as replacement of normal fat within neural foramina and pathologic nerve enhancement/thickening. MRI sensitivity/specificity for perineural spread has been reported at ~100%/85%, compared with ~88%/89% for CT (JAMA Otolaryngol Head Neck Surg, PMID:17576903). - CT remains useful for assessing bony invasion/erosion (e.g., skull base, mandible, maxillary sinus) and for chest staging (screening for pulmonary metastases, the most common distant metastatic site). - PET/CT may be used for staging and surveillance, though ACC's typically low proliferative index in low-grade tumors can limit FDG-avidity sensitivity.

Biopsy/pathology findings: Core needle or incisional biopsy demonstrating the characteristic biphasic population of ductal/luminal and myoepithelial/basaloid cells arranged in cribriform (most common; "Swiss cheese" pattern with basement-membrane-like/glycosaminoglycan-filled pseudocysts), tubular, or solid patterns. Perineural invasion is frequently identified histologically as a target-like tumor cuff around nerve fascicles.

Immunohistochemistry (diagnostic panel): - MYB nuclear expression (by IHC) — a highly sensitive and specific surrogate for MYB-NFIB fusion status, useful for distinguishing ACC from morphologic mimics (e.g., basal cell adenoma/adenocarcinoma, polymorphous adenocarcinoma) and for detecting the solid variant of breast ACC among triple-negative breast cancers. - c-KIT (CD117) — positive in ~90% of ACC. - p63/p40 and calponin/SMA/S100 — myoepithelial cell markers, highlighting the biphasic architecture. - Ki-67 proliferation index — higher indices correlate with solid pattern and worse prognosis. - p53 — overexpressed in ~50% of cases, associated with solid subtype and worse survival.

Molecular/genetic testing: - FISH or NanoString-based detection of MYB-NFIB / MYBL1 rearrangements — used diagnostically in ambiguous cases and increasingly to guide clinical trial eligibility (e.g., MYB-inhibitor trials). - Targeted NGS panels covering NOTCH1, NOTCH2, NOTCH3 hotspots — used to identify NOTCH-activating mutations that may render patients eligible for gamma-secretase inhibitor (GSI) trials (e.g., AL101, CB103). - Whole-exome/whole-genome sequencing has been used investigationally (including for individualized treatment selection in metastatic cases, e.g., the personalized oncogenomic case study at molecularcasestudies.cshlp.org).

Differential diagnosis: basal cell adenoma/adenocarcinoma, polymorphous adenocarcinoma (PAC), epithelial-myoepithelial carcinoma, dermal cylindroma (shares the MYB-NFIB fusion but is histologically and clinically distinct — an important molecular-vs-morphologic diagnostic nuance), and — for breast primaries — basal-like triple-negative breast cancer (distinguished by MYB IHC/FISH and generally much better prognosis for ACC-of-breast despite triple-negative receptor status).

Screening: No population-based or high-risk-group screening program exists for ACC, reflecting its rarity and lack of a clearly definable high-risk population (in contrast to, e.g., BRCA1/2-driven breast/ovarian cancer screening).

11. Outcome/Prognosis

Survival: - Long-term reported survival is notably better than raw 5-year figures suggest, owing to the tumor's protracted natural history: one classic cohort reported mean survival of 16 years, with actuarial survival of 77% at 5 years, 66% at 10 years, and 56% at 15 years. - Another cohort reported overall survival of 80.4% at 5 years, 61.3% at 10 years, and 29.4% at 15 years — illustrating cohort-to-cohort variability tied to stage/site mix. - SEER (2000–2019) 6-year overall survival by primary site (PMID:39410002): breast 86.5% ± 1.4% (best prognosis site), genitourinary 75.3% ± 4.5%, miscellaneous 73.8% ± 2.7%, head/neck 68.2% ± 0.9%, thoracic 60.5% ± 3.0% (worst). - Median overall survival varies substantially by US region in the SEER cohort: Northeast 171 months, West 144 months, South 136 months, Midwest 132 months — likely reflecting access/treatment-pattern effects. - Independent adverse prognostic factors (SEER multivariable analysis): thoracic primary site, regional disease (HR 1.63), distant metastasis at diagnosis (HR 2.55), chemotherapy use (HR 1.76, likely reflecting more advanced/refractory disease selection), Western US region of treatment (HR 2.93), and unmarried status (HR 1.27, a social-support proxy). - Favorable factors: surgical resection (HR 0.43) and breast primary site.

Distant metastasis and lung-specific outcomes: - Mean survival after the appearance of distant metastases is reported at approximately 11 years in one cohort. - For patients undergoing pulmonary metastasectomy, mean survival from the time of distant metastasis appearance was 72 months, versus 62 months for those who did not undergo resection — suggesting a potential (though selection-biased) survival benefit from surgical management of oligometastatic pulmonary disease (Eur J Cardiothorac Surg, PMID:18343149).

Recurrence patterns: Approximately one-third of patients experience recurrence, predominantly at distant sites rather than locoregionally; median time to locoregional recurrence and to distant metastasis are both approximately 50–51 months, but late events (>10 years) are well documented, reinforcing the need for indefinite long-term surveillance rather than the typical 5-year "cure" window used for many other carcinomas.

Prognostic factors/biomarkers: - Histologic grade/pattern (solid > cribriform > tubular for aggressiveness), though tumor stage may be a more robust predictor in contemporary series. - Perineural invasion — independently associated with worse disease-free and overall survival. - NOTCH1-activating mutations — independently associated with markedly worse survival (median OS 55.1 vs. 204.5 months in NOTCH1-mutant vs. wild-type recurrent/metastatic ACC per Ho et al., J Clin Invest 2019). - KDM6A mutation — associated with markedly worse overall survival (HR = 3.428, p = 0.0012) in one cohort. - 1p36 deletion — independent adverse prognostic marker. - p53 overexpression, RAS mutation, TP53 mutation — each associated with worse outcomes. - TP63 activation — paradoxically associated with a more favorable prognostic subgroup. - Margin status and surgical resection — consistently favorable.

Morbidity: Cranial neuropathies (facial, trigeminal), disfigurement from radical resection, and functional deficits (dysphagia, speech impairment, airway compromise for tracheal disease) constitute the principal non-fatal morbidity burden; standardized quality-of-life instrument data specific to ACC are limited in the literature surveyed.

12. Treatment

Surgery — the primary and preferred treatment modality regardless of anatomic site. Complete surgical resection with tumor-free margins is considered the "gold standard," though achieving negative margins is frequently challenged by the tumor's propensity for microscopic perineural extension well beyond the grossly visible tumor margin, often necessitating sacrifice of, or careful dissection around, critical nerves (e.g., facial nerve in parotid ACC). MAXO term: MAXO:0000004 (surgical procedure).

Radiation therapy: - Postoperative radiotherapy (PORT) is widely used for advanced, high-grade, or margin-positive/close-margin ACC, typically to a conventional photon dose around 60 Gy; recommended for all cases except T1N0 disease with clear margins. Evidence suggests PORT improves local control and may improve quality of life, though its impact on overall survival is debated in some series. MAXO:0000014 (radiation therapy). - Particle/advanced radiotherapy modalities show particular promise for ACC given its relatively radioresistant, low-proliferative biology: - Carbon-ion radiotherapy (CIRT) — an oxygen enhancement ratio near 1.0 (vs. 2.5–3.0 for photons) and reported 1.5–3× higher biological efficacy; evaluated in the phase II ACCO trial (adenoid cystic Carcinoma and Carbon ion Only irradiation), a prospective randomized two-arm study (PMC8281682). - Proton therapy (IMPT) — superior depth-dose distribution permitting dose escalation (70.0–79.1 CGE) with reported enhanced local control and reduced neurotoxicity. - Fast neutron therapy — high linear energy transfer, effective against the low-metabolic-rate biology typical of ACC. - Boron neutron capture therapy (BNCT) — investigational for skull-base ACC.

Systemic therapy: - Cytotoxic chemotherapy has limited efficacy in ACC; platinum-based regimens (e.g., cisplatin + paclitaxel) achieve response rates of only ~15–25%, and chemotherapy is generally reserved for symptomatic, rapidly progressive, or otherwise unresectable/unirradiable metastatic disease. - Tyrosine kinase inhibitors (targeting angiogenesis/VEGFR predominantly, given limited actionable driver mutations): - Lenvatinib (VEGFR/FGFR/PDGFR) — disease control and stability in most recurrent/metastatic cases, with 40.6% (13/32) achieving ≥6-month clinical benefit in one series. - Axitinib — 18% objective response rate (5/28), median PFS 7.3 months, median OS 16.6 months. - Apatinib (selective VEGFR2 inhibitor) — one series reported a 92.3% response rate; an international phase II trial (NCT02775370) reported a more modest but still notable 15.3% ORR with 14.9-month median response duration, reported as superior to other TKIs and chemotherapy in that comparison. - Sorafenib — ~10% ORR, with frequent adverse reactions limiting tolerability. - c-KIT-targeted agents (imatinib, dasatinib) and EGFR-targeted agents (gefitinib, cetuximab, lapatinib) have shown minimal-to-no objective response despite target expression, underscoring that IHC positivity (e.g., CD117) does not equate to oncogenic dependence in ACC. - NOTCH pathway inhibitors (for the NOTCH1-mutant subgroup, ~13–20% of patients): - AL101 (osugacestat), a pan-NOTCH gamma-secretase inhibitor — the ACCURACY phase II trial reported clinical activity at 4 mg once weekly with a disease control rate of 68% (15% partial response); AL101 received FDA Orphan Drug (2019) and Fast Track designations. Preclinical/mechanistic rationale published in Cell Death & Disease (PMID from PMC9355983). - CB103 (pan-NOTCH inhibitor blocking the CSL-NICD interaction) — phase I reported 58% disease stabilization, median PFS 2.5 months, OS 18.4 months (NCT03422679). - Brontictuzumab (anti-NOTCH1 monoclonal antibody) — phase I: 2 partial responses and 3 stable disease among 12 ACC patients. - Emerging MYB-directed strategies: - All-trans retinoic acid (ATRA) — reduces MYB enhancer binding, attenuating the MYB overexpression feedback loop (trials NCT03999684, NCT04433169). - RGT-61159 — an oral RNA-splicing modulator selectively inhibiting c-MYB, effective in ACC patient-derived xenograft (PDX) models, now in phase I (NCT06462183). - Preclinical MYB-targeting approaches: monensin A (ionophore screen hit), Bcr-TMP/oprozomib (proteasome inhibition, p300-dependent), and ferroptosis-inducing GPX4 inhibitors (ML162, ML210, RSL3) correlated with MYBL1 dependency; HDAC inhibitors (vorinostat) implicated in ferroptosis sensitivity regulation. - Immunotherapy: Given ACC's immunologically "cold" phenotype (low mutational burden, minimal lymphocyte infiltration, rare PD-L1 expression, active PD-L2/HLA-G immunosuppression), checkpoint inhibitor monotherapy has shown limited efficacy — pembrolizumab achieved only a 12% partial response rate among 26 advanced salivary gland cancer patients (including 2 ACC) in KEYNOTE-028, and pembrolizumab + radiotherapy (NCT03087019) achieved 0% objective response in 20 metastatic ACC patients. Combination strategies (e.g., axitinib + avelumab, a checkpoint inhibitor, in an ongoing phase II trial) and a MYB peptide vaccine + anti-PD-1 trial (NCT03287427) are being explored to convert the cold tumor microenvironment. - Other investigational targeted agents: cabozantinib (multikinase AXL/MET/VEGFR inhibitor, NCT03729297, rationale tied to TP63-activated AXL/EGFR/MET expression conferring EGFR-inhibitor resistance), everolimus (mTOR inhibitor; 0% ORR but 65.5% disease stabilization), bortezomib (proteasome inhibitor; 0% ORR, 68% stable disease), PARP inhibitors (proposed for the ACC-I molecular subtype with high PARP/CHK1/CHK2 expression), CDK9 inhibitor KB-0742 (53.8% disease control in 18 patients), and the STING agonist TAK-676 (NCT04879849). - Combination approaches: triple combination of linsitinib (IGF1R) + gefitinib (EGFR) + crizotinib (ALK/MET) significantly reduced MYB expression preclinically; combination strategies pairing NOTCH inhibitors with MYC-targeting agents (e.g., PRMT5 inhibitor GSK3326595, which produced partial responses in 3/14 patients) and BCL2 inhibitors with gamma-secretase inhibitors are being explored to address tumor heterogeneity.

Supportive/rehabilitative care: Facial nerve rehabilitation (physical therapy, nerve grafting/reanimation surgery when the facial nerve is sacrificed), speech and swallowing therapy for oropharyngeal/laryngotracheal disease, and pain management for perineural neuropathic symptoms.

Suggested MAXO terms: MAXO:0000004 (surgical procedure), MAXO:0000014 (radiation therapy), MAXO:0000647 (chemotherapy), MAXO:0000011 (physical therapy), MAXO:0000950 (supportive care); with the pharmacotherapy modality (NCIT:C15986) plus therapeutic_agent bindings to CHEBI/NCIT terms for lenvatinib, axitinib, apatinib, sorafenib, imatinib, cetuximab, pembrolizumab, and the investigational AL101/CB103 gamma-secretase inhibitors.

13. Prevention

Primary prevention: No established primary prevention strategy exists, since ACC lacks a clearly modifiable lifestyle risk factor (unlike smoking-associated head and neck cancers). The one identified modifiable/avoidable risk factor is minimizing unnecessary ionizing radiation exposure to the head and neck, particularly in children, given the well-documented latency-associated risk of radiation-induced salivary gland malignancy.

Secondary prevention (screening/early detection): No population-based screening program exists for ACC due to its rarity and lack of a definable high-risk population; early detection instead relies on prompt clinical evaluation (and imaging/biopsy) of persistent head/neck masses, unexplained cranial neuropathy, or slow-growing breast/skin lesions.

Tertiary prevention: Given the long natural history and persistent late-recurrence risk, indefinite long-term clinical and radiographic surveillance (including periodic chest imaging for pulmonary metastasis surveillance) after primary treatment functions as the principal tertiary-prevention strategy to enable early detection and potential resection of oligometastatic (particularly pulmonary) recurrence.

Genetic counseling: Not routinely indicated given the absence of an established hereditary ACC syndrome; counseling would only be considered in the rare context of a family history suggestive of a BRCA2-associated cancer predisposition syndrome, and even then would be directed at the broader hereditary breast/ovarian cancer risk rather than ACC specifically.

Immunization/prophylaxis: Not applicable — no infectious trigger has been identified.

14. Other Species / Natural Disease

Taxonomy and natural disease: Adenoid cystic-type carcinomas arising from salivary and other exocrine glandular tissue are recognized in veterinary comparative oncology, most notably in dogs (NCBITaxon:9615) and cats (NCBITaxon:9685), where salivary gland adenocarcinomas — including adenoid cystic-type histology — occur as naturally occurring tumors, and mammary gland carcinomas (the most common tumor type in intact female dogs) can show adenoid cystic-like ("mucinous") morphologic patterns. The literature search did not surface a dedicated OMIA (Online Mendelian Inheritance in Animals) entry specifically for canine ACC as a distinct catalogued entity, though comparative-oncology reviews of canine mammary carcinoma emphasize strong molecular parallels with human breast malignancies in cell-cycle regulatory and oncogene/tumor-suppressor gene expression patterns (PMC5644615), supporting the general utility of the dog as a comparative model for glandular carcinomas, even though ACC-specific MYB-NFIB comparative genomic data in veterinary species were not identified in this search.

Comparative biology: No strong evidence was found in this search for a dedicated invertebrate, zebrafish, or other non-mammalian natural-disease correlate of ACC, consistent with its origin from exocrine/glandular epithelial-myoepithelial architecture that is a mammalian anatomic feature.

Zoonotic potential/transmission: Not applicable — ACC is a non-infectious, non-transmissible somatic malignancy.

15. Model Organisms

Genetically engineered mouse models (GEMMs): Because the MYB-NFIB fusion is the genomic hallmark of human ACC, at least three GEMMs have been engineered to test its in vivo oncogenic role (as reviewed in PMC11387731 and by the Adenoid Cystic Carcinoma Research Foundation, accrf.org/tools-for-researchers/gemms/): - Mikse et al. (2016) — crossed bi-allelic MYB-NFIB/MMTV-Cre mice (driving fusion expression in salivary and mammary tissue) with p53^fl/fl^ mice. Notably, no mice developed salivary gland cancer, but poorly differentiated mammary tumors developed specifically in MYB-NFIB/MMTV-Cre/p53^+/fl^ mice — indicating that MYB-NFIB expression alone is insufficient for tumorigenesis and requires cooperating p53 pathway loss, at least in this model, and highlighting a human-model mismatch: the model did not recapitulate the salivary-gland tropism of human ACC despite transgene expression there. - Jiang et al. (2019) — crossed a MYB-NFIB GEMM with Ink4a^+/−^/Arf^+/−^ mice; one resulting mouse developed a mammary tumor with cribriform ACC-like adenocarcinoma histology, overexpressing Myb protein and positive for the human MYB-NFIB transgene, providing partial phenotypic recapitulation of human cribriform-pattern ACC. - A third model exists per the ACCRF GEMM resource page, underscoring active community effort to build faithful in vivo models (accrf.org/tools-for-researchers/gemms/).

In vitro / cell line models: Established ACC cell lines (e.g., ACC2, SACC-83) are used widely but have documented significant genetic drift with passage, and some historically used lines have been found to be misidentified/contaminated, a recognized limitation flagged in recent reviews (PMC11387731) — an important caveat for interpreting older cell-line-based mechanistic and drug-screening literature.

Patient-derived organoids (PDOs): Report a relatively low establishment success rate (~19%), but successfully derived organoids maintain MYB(L1)/NFIB rearrangement status, making them a molecularly faithful (if logistically challenging) platform.

Patient-derived xenografts (PDX): Higher establishment success rate (~60%) and maintain genomic alterations, though production remains resource- and time-intensive; PDX models have been used to validate emerging MYB-directed agents such as RGT-61159.

Model limitations: No single current model (GEMM, cell line, organoid, or PDX) fully recapitulates the combination of (a) faithful anatomic tropism (salivary gland vs. mammary gland), (b) the full spectrum of cribriform/tubular/solid histologic heterogeneity, (c) the biphasic epithelial-myoepithelial cellular architecture, and (d) the immunologically "cold" tumor microenvironment seen in human disease — motivating ongoing development of orthotopic transplantation models (implantation directly into mouse salivary gland) and next-generation genetically diverse cell lines and 3D organoid systems (PMC11387731).

Applications: Current models are primarily used to (1) validate MYB-directed and NOTCH-directed small-molecule/biologic therapeutics preclinically, (2) study the biology of the MYB-NFIB/MYBL1 fusion's transforming potential and its dependency on cooperating lesions (e.g., p53, Ink4a/Arf), and (3) explore mechanisms of perineural invasion and the cold immune microenvironment for future immunotherapy-sensitization strategies.

Summary of Key Evidence Sources

Data gaps flagged for curation: standardized QoL instrument data specific to ACC; robust germline predisposition data beyond anecdotal BRCA2 reports; a dedicated OMIA veterinary entry; comprehensive epigenomic (ENCODE/Roadmap-style) profiling specific to ACC; and confirmed PMID numbers for the Persson 2009 and select other papers cited here by PMCID/DOI only — verify final PMIDs directly on PubMed before entering evidence.reference fields.