Perineural invasion (PNI) occurring in non-melanoma skin cancers (NMSC) is associated with an increased risk of locoregional recurrence and reduced disease-free survival. This necessitates early and accurate diagnosis, appropriate risk-stratification and a clear management strategy. The diagnosis of PNI is based on careful clinical assessment, imaging and histopathology. Surgery, preferably with margin control, and definitive or adjuvant radiotherapy (ART) are established treatment strategies for PNI. Clinical uncertainty remains over the role of ART in incidental PNI. This review synthesises current literature to ascertain which clinicopathological features impart a higher risk to individuals with PNI in NMSC, in order to provide treatment algorithms, including the identification of patient subsets that are most likely to benefit from ART. This includes those with extratumoural PNI, involvement of larger-calibre nerves, tumour invasion beyond dermis, recurrent tumour or diffuse intratumoural spread. Patients with clinical PNI may be optimally managed by a multidisciplinary head and neck cancer service that is best placed to offer skull base surgery and intensity-modulated radiation therapy (IMRT). The management options presented are stratified by histological subtype and a new classification of PNI into low-risk, medium-risk and high-risk groups.

This article by the Perineural Invasion (PNI) Registry Group aims to clarify clinical and histopathological ambiguities surrounding PNI in non-melanoma skin cancer (NMSC). PNI is reportedly present in approximately 2-6% of cases of NMSC and is associated with greater rates of morbidity and mortality. The distinction between clinical PNI and incidental PNI is somewhat unclear, especially in regard to management and prognosis. One important objective of the PNI Registry is to develop a standardised method of classifying perineural invasion. Hence, in this article we propose a definition for PNI and for its sub-classification. This article also provides a critical analysis of the current literature on the treatment of incidental PNI by evaluating the key cohort studies that have investigated the use of surgery or radiotherapy in the management of incidental PNI. At present, there are no universal clinical guidelines that specify the acceptable treatment of NMSC exhibiting incidental PNI. Consequently, patients often receive surgery with varying wider margins, or radiotherapy despite the limited evidence substantiating such management options. It is evident from the existing literature that current opinion is divided over the benefit of adjuvant radiotherapy. Certain prognostic factors have been proposed, such as the size and depth of tumour invasion, nerve diameter, the presence of multifocal PNI and the type of tumour. The PNI Registry is a web-based registry that has been developed to assist in attaining further data pertaining to incidental PNI in NMSC. It is envisaged that this information will provide the foundation for identifying and defining best practice in managing incidental PNI.

A retrospective study of 813 histological specimens reported as seborrhoeic keratoses included 43 (5.3%) associated with non-melanoma skin cancer. Intraepidermal carcinoma (squamous cell carcinoma in situ) was the most common of these (36). There were five basal cell carcinomas (one with intraepidermal carcinoma also) and two invasive squamous cell carcinomas. No melanomas were reported. Twenty-seven of the intraepidermal carcinomas appeared to arise within the seborrhoeic keratosis as did one of the invasive squamous cell carcinomas. Of these 28 lesions, the head was the most common site. Fourteen were clinically diagnosed as a non-melanoma skin cancer with only nine clinically felt to be a seborrhoeic keratosis. These lesions may represent malignant transformation within the seborrhoeic keratosis. Twelve specimens reported adjacent dual pathologies, with the trunk and limbs the most common sites. Seven were diagnosed clinically as a skin malignancy, whereas three were thought to be solar keratoses. Clinically, the remaining two were seborrhoeic keratoses. The origin of the malignancy in these cases is less obvious and may represent collision tumours. Three curette specimens could not be assessed for architecture.

No abstract

The authors have indicated no significant interest with commercial supporters