Adenoid cystic carcinoma (ACC) is the second most common histologic subtype of sinonasal malignancies after squamous cell carcinoma (SCC), accounting for 5%–15% of malignant paranasal sinus tumors (Husain et al., 2013; Dulguerov et al., 2001). It is characterized by slow growth, a tendency for perineural invasion, late recurrence, and distant metastasis (Bradley, 2004; Chummun et al., 2001). More than 50% of cases show notable invasiveness with a propensity for distant metastasis, commonly to the lungs, bones, and livers, ultimately resulting in low survival rates (Seok et al., 2019; Cavalieri et al., 2020). Given its slow-growing biological nature and non-specific symptoms in the early stage, the diagnosis of sinonasal adenoid cystic carcinoma (SNACC) is often delayed and usually found in advanced stages (65%–76% T3-T4 tumors). A meta-analysis involving 17 studies with 2,300 patients showed that high rates of locally advanced tumor stages at diagnosis: 23% cT3, 53% cT4 (Mauthe et al., 2023). This leads to a poorer prognosis compared with other head and neck ACCs, such as salivary gland counterparts (Michel et al., 2013).

The optimal treatment consists of surgical resection and postoperative adjuvant radiotherapy for SNACC patients (Husain et al., 2013; Rhee et al., 2006; Wiseman et al., 2002). Platinum-based single-agents and combination regimens are usually administered to advanced diseases, including unresectable, recurrent or metastatic tumors, while limited data on the effectiveness of these treatments are available, mainly due to the indolent course of the disease and the lack of large-sized studies or prospective trials to evaluate the effect of chemotherapy (Papaspyrou et al., 2011; Laurie et al., 2011) (Rentschler et al., 1977; Dodd and Slevin, 2006; Vermorken et al., 1993; Ross et al., 2009). At present, cisplatin, doxorubicin, cyclophosphamide (CAP) was the most studied regimen in ACC, achieving an objective response rate of 25% (Licitra et al., 1996). Therefore, the value of chemotherapy in reducing distant metastasis and improving long-term survival remain controversial for ACC patients (Coca-Pelaz et al., 2015).

Due to the rarity of the disease, very few clinical reports have investigated the effect of chemotherapy in managing SNACC, and its impact on distant metastasis and long-term survival remains unclear. Moreover, the clinical usefulness of chemotherapy in patients with SNACC has been a topic of debate because of the absence of high-level evidence. Additionally, patients with stage T4 lesions exhibit significantly worse overall survival (OS) compared with those with T1-T3 disease in SNACC (Lupinetti et al., 2007). Given the urgent need to improve the OS for stage T4 SNACC, it is essential to explore treatment strategies that include chemotherapy within a multidisciplinary framework. The primary goal of this study was to determine whether chemotherapy could reduce the risk of distant metastasis and improve the long-term survival outcomes for patients with T4-stage SNACC.

Survival data were shown in Supplementary Figure S1. The median follow-up duration was 71.2 months, with a range of 3.7–353.4 months. Among the 87 patients, 45 (51.7%) experienced treatment failures, which included local recurrence in 22 cases (25.3%), distant metastasis in 13 cases (14.9%) or both failures in 10 cases (11.5%). Additionally, 33 (37.9%, 33/87) patients ultimately died of the disease. The OS rates for all the patients’ 5-, 10-, and 15-year were 74.3%, 47.5%, and 33.0%, respectively. Similarly, the PFS rates at the same time points were 60.6%, 31.9%, and 22.8%, respectively.

Multivariate Cox regression analysis revealed that the lymph node status was an independent prognostic factor for worse PFS (HR = 3.844, 95% CI: 1.252–11.801, P = 0.019) in T4 stage SNACC (Table 2). Furthermore, chemotherapy failed to significantly affect either OS (HR = 0.902, 95% CI: 0.425–1.917, P = 0.789) or PFS (HR = 1.212, 95% CI: 0.638–2.302, P = 0.556).

To examine whether chemotherapy can reduce the risk of distant metastasis or not in SNACC patients, we performed univariate and multvariate logistic regression analyses. We enrolled clinical factors that were potentially related to metastasis, such as T stage, N stage, and anatomical sites. The results of the multivariate analysis revealed that patients with nasopharyngeal invasion were more likely to develop distant metastasis (P = 0.028), while chemotherapy did not reduce the risk of distant metastasis in both univariate (P = 0.383) and multivariate analyses (P = 0.213) (Figure 1).

Adjuvant chemotherapy in head and neck adenoid cystic carcinoma (ACC) is typically reserved for advanced-stage disease. However, there is a paucity of studies exclusively focusing on ACC originating from the paranasal sinuses. Due to the aggregation of data across various anatomical sites in most head and neck ACC studies, it is difficult to draw specific conclusions about SNACC. Although the chemotherapy might have a palliative benefit (Laurie et al., 2011), its long-term survival effects remain unclear, particularly in T4-stage SNACC. To address this critical clinical question, we conducted a retrospective study evaluating the role of platinum-based chemotherapy in T4-stage sinonasal adenoid cystic carcinoma.

Despite the frequent occurrence of late recurrence and distant metastasis, SNACC patients may still exhibit prolonged survival, with a 5-year OS rate of 55%–70% (details seen in Supplementary Table S2) (Lupinetti et al., 2007; Unsal et al., 2017; Akbaba et al., 2019; Trope et al., 2019; Jang et al., 2025; Hu et al., 2020). However, this rate drops significantly to 40% at 10 years and further decreases to 15% by 20 years (Husain et al., 2013; Mays et al., 2018). Previous research has identified the T4 stage as an independent prognostic factor associated with decreased local control in SNACC patients (Michel et al., 2013; Akbaba et al., 2019). T4 stage tumors frequently involve surrounding organs at risk, such as orbit, cranial nerves, dura, cavernous sinus, carotid arteries, and brain, limiting the possibility of radical resection (Unsal et al., 2017). The involvement of the organ at risk makes treatment challenging. Additionally, pathological subtypes are also closely associated with patients’ prognosis. Classic ACC often shows a combination of cribriform and tubular patterns. In most studies, a solid growth pattern is a predictive factor for poor prognosis, advanced stage and development of distant metastases (Ferrarotto et al., 2021). Clinically, managing T4 lesions represents a significant therapeutic dilemma. In our study, despite receiving multidisciplinary treatment, up to 25.3% of patients still experienced local recurrence, and 14.9% faced distant metastasis. Consistent with the literature (Husain et al., 2013; Mays et al., 2018), the 5-, 10-, and 15-year OS rates for T4 stage SNACC were 74.3%, 47.6%, and 33.0%, respectively. During the first 5 years of follow-up, the results suggested a high survival rate. However, the OS and PFS rates were shown to decrease drastically during the subsequent follow-up periods, which may be attributed to the high rate of later recurrence and distant metastasis.

Chemotherapy plays a limited role in ACC, primarily used for patients with unresectable, recurrent or metastatic disease (Dillon et al., 2016). Chemotherapeutic agents such as Cisplatin, Adriamycin, and Vinorelbine (CAP) can be used as monotherapy or in combination chemotherapy schedules (Rentschler et al., 1977; Dodd and Slevin, 2006; Vermorken et al., 1993; Ross et al., 2009). CAP is the most commonly used combination chemotherapy regimen; however, it is not regarded as a standard treatment due to the balance between its high toxicity profile and treatment efficacy (Papaspyrou et al., 2011; Ha et al., 2020). Hyerim et al. (Ha et al., 2021) observed that concurrent chemoradiation with cisplatin was effective for local control and had tolerable toxicity in ten cases of locally advanced unresectable ACC. In our study, patient with orbital/nasopharyngeal/cavernous sinus involvement were prone to receive platinum-based chemotherapy as a component of their treatment (Table 1). A combination of cisplatin, doxorubicin, and cyclophosphamide was the most common chemotherapy regimen (46%, 23/50). Unfortunately, logistic regression analysis indicated that chemotherapy could not reduce the risk of distant metastasis in both univariate (P = 0.383) and multivariate analyses (P = 0.213). Pooled data from studies on ACC indicate that chemotherapy demonstrates a low objective response rate (0%–30%), and stable disease (SD) and symptomatic relief are more commonly achieved outcomes (Papaspyrou et al., 2011; Dodd and Slevin, 2006). In the Surveillance, Epidemiology, and End Results (SEER) database, Micha et al. (Trope et al., 2019) found that surgery, but not radiation therapy or chemotherapy, predicted better OS for SNACC patients. In Figure 3, during the univariate Kaplan-Meier curve analysis, no statistical difference was observed between the chemotherapy group and the non-chemotherapy group, and the curves showed a crossing phenomenon. Thus, we adopted landmark analysis and found that there was no statistical difference in 5-year OS between the two groups (P = 0.080). However, as the follow-up time extended, patients in the chemotherapy group exhibited worse long-term survival. The lower 10-year OS (P = 0.034) in the chemotherapy group may be due to the toxic effects of chemotherapy and its tendency to favor patients with less favorable prognostic indicators (orbital/nasopharyngeal/cavernous sinus involvement) who were difficult to achieve negative surgical margins. Risk factors such as positive lymph nodes (N⁺), orbital involvement, and nasopharyngeal invasion often influence treatment decisions, including the use of chemotherapy (Megwalu and Sirjani, 2017; International and Neck Scientific, 2016; Qazi et al., 2016). Our subgroup analyses indicate that chemotherapy does not provide a survival benefit among specific populations, such as those with N⁺, orbital invasion or nasopharyngeal invasion.

This study has several limitations that require consideration. The primary weakness of our research was that it was a single-center retrospective study, which introduced the potential for selection and information bias. For instance, the choice of whether to administer chemotherapy may have been influenced by prognostic factors that were not included in the multivariate analysis, highlighting the need for prospective studies, that are currently lacking. Additionally, prognosis is associated with histologic subtypes (i.e., tubular subtype, cribriform or solid), but the absence of data regarding histological classification made it impossible to perform a subgroup analysis. Despite these limitations, to our knowledge, this is the only reported series of patients with T4 stage SNACC treated with chemotherapy.

In this retrospective analysis, platinum-based chemotherapy did not seem to provide a long-term survival advantage for patients with T4 SNACC and might be associated with worse outcomes after 5 years, although these findings need confirmation through prospective studies.