xRead Articles - November 2022

128

Otolaryngology–Head and Neck Surgery 154(1)

Table 7. Recommendations for Elective Treatment of the Neck (cN0).

Elective Treatment Recommended

Elective Treatment Not Recommended

High-grade carcinoma ex pleomorphic adenoma

Low-grade carcinoma ex pleomorphic adenoma

High-grade mucoepidermoid carcinoma

Low-grade mucoepidermoid carcinoma

High-grade acinar cell carcinoma

Low-grade acinar cell carcinoma

High-grade adenocarcinoma not otherwise specified

Low-grade adenocarcinoma not otherwise specified

High-grade adenoid cystic carcinoma High-grade salivary ductal carcinoma

Low-grade adenoid cystic carcinoma Low-grade salivary ductal carcinoma All epithelial myoepithelial carcinomas All basal cell adenocarcinomas

regional metastasis in multivariate analysis for acinar cell car cinoma and carcinoma ex pleomorphic adenoma but not for adenoid cystic carcinoma and salivary ductal carcinomas. Grade alone was significant in multivariate analysis for ductal carcinoma NOS. In summary, it is clear that T stage and grade are significant and important predictors of nodal disease. The prevalence of occult neck disease is a key compo nent to planning therapy of the cN0 neck, but large or pro spective studies are not available to provide conclusive evidence that could guide therapy. Treating the cN0 neck is still debated, with some authors advocating routine elective neck dissection 19 versus others with a more conservative approach, reserving lymphadenectomy for high-grade tumors, high-risk histologies, and T3/T4 disease. 22 Radiotherapy is another alternative that has been proposed for treatment of the cN0 neck and/or in an adjuvant setting for high risk or N 1 neck. 21,30 Frankenthaler et al recommend elective radia tion for cN0 high-risk tumors (since pre- and perioperative typing is unreliable), thus controlling for microscopic disease if nodal disease is later found on pathologic analysis. 20 Herman et al recently found that for high-grade salivary malignancies, patients undergoing surgery and postoperative radiation likely do not benefit from planned neck dissec tion. 31 The incidence of true occult disease is difficult to ascertain and has a wide range of reported values, all based on small studies. Moreover, the degree of scrutiny in lympha denectomy specimens can differ significantly among institu tions. 32 We found an overall rate of occult disease of 10.2%, lower than literature estimates of 14% to 33%. 32 Our study’s biggest strength is that it provides a significantly larger sample set for all histopathologies and a separate analysis of each histopathology and excludes minor and submandibular gland sites. Using a threshold of . 10% occult nodal disease, we found that ductal carcinoma NOS (23.6%), adenocarcinoma NOS (19.9%), and carcinoma ex pleomorphic adenoma (11.8%) have high rates of overall occult disease overall. Out of this high-risk group, all low-grade variants had occult disease incidences \ 10%. Despite its traditionally ‘‘high risk’’ classification, carcinoma ex pleomorphic ade nomas behave less aggressively at lower grades, 22 suggested by the low incidence of nodal metastasis (7%) and occult

disease (7.5%); in stark contrast, high-grade disease had N 1 and occult rates of 42.2% and 19.2%, respectively. As such, previous authors have described ‘‘intracapsular’’ carci noma ex pleomorphic adenoma and minimally invasive car cinoma ( \ 1.5-mm invasion) ex pleomorphic adenoma as 2 indolent variants. 22 This is important to consider since elec tive neck dissection may be performed for all ‘‘high risk’’ histologies, despite their low-grade variants having low inci dences of occult disease. On the basis of these data, we would not recommend routine elective neck dissection in low-grade malignancy without evidence of regional metasta sis. Last, increasing T stage is significantly associated with occult nodal incidence except for epithelial-myoepithelial carcinoma and salivary ductal carcinoma. This is another variable that the clinician should factor in when considering neck dissection. Table 7 summarizes our recommendations for elective treatment of the clinically N0 neck. This study has a few limitations, some inherent to large national database studies. As discussed previously, we included only patients with some form of lymph node sampling—the extent of which we cannot verify. Patients may have received only a lymph node biopsy or have under gone extensive neck dissection. This is a selection bias that may select for more aggressive cancers while increasing occult node detection frequency. In this sense, our data may represent a ‘‘worst case’’ scenario for nodal disease. However, we feel that without pathologic confirmation of neck metastasis, the study would be weaker and with less accurate reporting of the N0 neck. The NCDB is a heterogeneous database and draws infor mation from a wide array of hospitals. Pathologic diagnosis of salivary cancers is inherently challenging, better left to experienced surgeons and pathologists due to controversy surrounding the diagnosis via fine-needle aspiration biopsy and pathologic subtyping. Without the ability to verify every stage, grade, and histopathologic diagnosis, this is a potential weakness inherent in large database studies, and it represents an unknown degree of error in histologic misclas sification; we rely solely on accurate coding practices. However, the NCDB has a high degree of quality control and takes great measures to ensure high-quality and consis tent reporting. 33