Ultrasonography may improve lymph node metastasis detection among patients with high-risk cutaneous squamous cell carcinoma of the head and neck.
Clinical Pearls
This was a retrospective cohort study of 223 patients with 246 high-risk cutaneous squamous cell carcinomas (cSCC) at a single tertiary care center in the Netherlands between 2015 and 2017.
There were 20 cytologically-confirmed metastases at baseline and 2 found within the 6-months of follow-up. Clinical exam had a sensitivity of 50% (95% confidence interval 28-72%), specificity of 96% (95%CI 93-98%), positive predictive value (PPV) 55% (95%CI 36-72%) and negative predictive value (NPV) of 95% (95%CI 93-97%). Ultrasonography (US) had a sensitivity of 91% (95%CI 71-99), specificity of 78% (95%CI 72-83%), PPV 29% (95%CI 23-35%) and NPV of 99% (96-100%).
In patients with a negative baseline clinical assessment, US detected 9/11 metastases.
Discussion:
Cutaneous squamous cell carcinoma (cSCC) is the second most common non-melanoma skin cancer (NMSC) among non-Hispanic white patients. While the prognosis of cSCC is good for many patients, up to 4-5% may have metastasis of the primary cSCC. This may be especially true of higher risk tumors of the head and neck.
This retrospective diagnostic study investigated the use of baseline ultrasonography and clinical exam among a cohort with 233 patients with 246 high-risk cSCC of the head and neck that were referred to a tertiary care center in the Netherlands between January 1, 2015-December 31, 2017.
The authors found that, while staging did not differ significantly between cSCCs with and without metastases, cases with metastases did have greater depth of invasion (DOI) (p<.001) with metastases primarily located in or around the parotid gland (54.5%), cervical glands (22.7%), both parotid and cervical glands (18.2%), or retro-aurically (4.5%).
20/246 tumors were observed to have suspicious lymphadenopathy with 11 confirmed via fine needle aspiration cytology (FNAC) leading to 9 false positive results. Clinical exam therefore had a sensitivity of 50% (95% confidence interval 28-72%), specificity of 96% (95%CI 93-98%), PPV of 55% (95%CI 36-72%) and NPV of 95% (95%CI 93-97%).
US detected suspicious lymphadenopathy in 70/246 tumors, 21 of which were confirmed after FNAC (1/21 required repeat FNAC due to insufficient tissue). Ultrasonography (US) had a sensitivity of 91% (95%CI 71-99), specificity of 78% (95%CI 72-83%), PPV 29% (95%CI 23-35%) and NPV of 99% (96-100%). Furthermore, among the 54/226 tumors with negative clinical exam but suspicious US exam, 9/54 had FNAC-positive metastasis. In this sub-cohort, US had a sensitivity of 82% (95%CI 48-98%), specificity 79% (95%CI 73-84%), and NPV 99% (95%CI 96-100%) but a low PPV of 17% (95%CI 12-23%).
Limitations of this study included its retrospective design and confinement to a single study location. The authors also note that disease may have been missed by both clinical and US exam and therefore overestimate sensitivity analysis and comparative gold-standard assessment could have been augmented with CT or MRI imaging. Authors also imputed data regarding perineural invasion if it was not available.
Overall, the major limitation is lack of true gold-standard assessment and minimal sample size. A larger study that also used more comprehensive imaging or included patients who underwent sentinel lymph node biopsy may provide a more accurate assessment of US application for lymph node metastasis detection. It is also important to keep in mind that US is also highly user dependent and, given the high false positive rate the authors noted, may be exacerbated depending on user capabilities. Overall, US may provide an additional step when used in series with clinical assessment to provide real-time risk-stratification. At this time, however, further studies are needed to identify risk factors and patients who would most benefit from this technology.
Comments