6121 Background: Although ACC commonly arises in the salivary glands, nearly 40% arise at extra-salivary sites and are often managed as site-specific malignancies rather than as a unified histologic entity. Population-based data suggest that thoracic and head and neck ACC are associated with poorer survival compared with breast ACC; however, the factors underlying these differences remain poorly understood. Methods: We retrospectively identified all patients (pts) with ACC, irrespective of primary site, who were followed at a large cancer center between 2011 and 2024. Collected data included staging at diagnosis, primary site, local treatment (i.e. surgery, radiation), histologic subtype (solid, cribriform or tubular), lymphovascular invasion (LVI), perineural invasion (PNI), margin status (R0, R1), and clinical outcomes including local recurrence-free survival (LRFS), distant metastasis free-survival (DMFS), and overall survival (OS). Primary sites were grouped as major salivary glands (MASG), oral cavity or oropharynx (OCOP), sinonasal region (SINO), breast (BRST), thorax-lung (THLU), lacrimal gland (LCGL), and others (i.e., skin OTHR). Clinicopathologic characteristics were compared using Fisher tests, and associations between primary site and clinical outcomes were evaluated with multivariable Cox models comparing each site versus all others combined adjusted for local treatment type, margins, and solid histology, defined a priori following a causal inference framework. Results: Of 220 pts, 151 (69%) were female, and 199 (90.5%) had localized or locally advanced disease at diagnosis. The most common primary site was OCOP (n=67, 30%), followed by MASG (n=60, 27%), SINO (n=25, 11%), THLU (n=22, 10%), BRST (n=20, 9.1%), LCGL (n=6, 2.7%), and OTHR (n=20, 9.1%). Among 210 pts that received local treatment, 41 (19%) had surgery only, 134 (61%) surgery plus radiation, and 35 (16%) radiation alone. Patients with THLU primary had less up-front surgery (50% vs 76-100%, p<0.001). Among surgical cases, 58% had R1 resections; PNI was present in 48%. BRST pts had lower rates of positive margins (R1 7.7%, p=0.04) and PNI (28% vs 35-83%, p=0.019). Histologic subtype was reported in 108 (49%) pts with 72 (78%) cribriform, 58 (54%) solid, and 49 (45%) tubular. Median LRFS, DMFS, and OS were 103 months mo (95%CI 77-141), 87 mo (95%CI 70-131), and 153 mo (95%CI 124-222) respectively. Pts with SINO experience worse LRFS (HR 2.01, p=0.03), DMFS (HR 2.07, p=0.02), and OS (HR 3.51, p<0.001) while MASG had potentially improved DMFS (HR 0.63, p=0.07) and THLU OS (HR 0.31, p=0.07) in adjusted models. Conclusions: Primary site-specific clinical outcomes in ACC are not fully explained by treatment approach or histologic features. These findings suggest biological heterogeneity across ACC by primary site and underscore the need for molecular characterization of site-specific disease drivers.
Passarini et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: