High-risk mucosal human papillomavirus-associated squamous cell carcinoma of the fingers presents distinctive clinicopathological characteristics and outcomes: a bicentric retrospective study.

BACKGROUND: High-risk mucosal human papillomavirus (HR-HPV) seems to play a role in cutaneous squamous cell carcinoma (cSCC), particularly in nail tumours, where genitodigital transmission has been suggested. The role of HR-HPV in nonungual cSCC of the finger needs to be clarified. AIM: To evaluate the prevalence, clinicopathological characteristics, surrogates and outcomes of HR-HPV in cSCC of the finger. METHODS: This was an observational bicentric study including patients with an excised in situ or invasive cSCC located on the finger. Differences in HR-HPV and non-HR-HPV tumours were evaluated. RESULTS: Forty-five patients (45 tumours) were included. HR-HPV was detected in 33% of cases (22% HPV type 16). The mean age was lower in patients with HR-HPV than in those with non-HR-HPV (62·4 vs. 81·1 years, P = 0·001). HR-HPV tumours were smaller (10 mm vs. 15 mm, P = 0·07) and more frequently intraepidermal (60% vs. 20%, P = 0·004). The absence of elastosis (P = 0·030) and inflammation (P = 0·026) and the presence of basaloid morphology (P = 0·003) were surrogates of HR-HPV detection. Mean p16 positivity was 61% in HR-HPV and 36% in non-HR-HPV tumours (P = 0·061). Recurrence after surgery was more common in HR-HPV tumours (58% vs. 34%), although this was not statistically significant. HR-HPV was detected in 27% of the nonungual tumours. CONCLUSION: HR-HPV-associated cSCC of the finger appears in younger patients, is smaller and is less infiltrative than non-HR-HPV tumours. The presence of a basaloid morphology and the absence of elastosis and inflammation could be used as markers for HR-HPV detection. The high prevalence of HR-HPV in nonungual cSCC suggests its aetiopathogenic role in these tumours.

Performance of Salamanca refinement of the T3-AJCC8 versus the Brigham and Women's Hospital and Tübingen alternative staging systems for high-risk cutaneous squamous cell carcinoma.

INTRODUCTION: The Brigham and Women's Hospital and the Tübingen cutaneous squamous cell carcinoma (SCC) stratification systems propose different criteria from the American Joint Committee on Cancer, eighth edition. Our group identified prognostic subgroups within T3 stage according to the American Joint Committee on Cancer eighth edition's classification, the most common classification for high-risk cutaneous SCCs. OBJECTIVE: To compare the performance and prognostic accuracy of these staging systems in a subset of high-risk cutaneous SCCs. METHODS: Homogeneity, monotonicity, and McNemar tests for pairwise comparisons were assessed. Distinctiveness and relative risk of poor outcome were calculated by stage. Prognostic accuracy was compared with respect to quality (Akaike and Bayesian information criteria), concordance (Harrell C-index and Gönen and Heller concordance probability estimate), and predictive accuracy (sensitivity, specificity, negative predictive value, positive predictive value, and global accuracy). RESULTS: The Brigham and Women's Hospital and Salamanca systems were more distinctive, homogeneous, and monotonic than the Tübingen system. The Tübingen system was the most specific, whereas the Salamanca and Brigham and Women's Hospital systems were more sensitive. Negative predictive value was high in all 3 systems, but positive predictive value and accuracy were low overall. CONCLUSIONS: Alternative staging systems may partially overcome the heterogeneity and low prognostic accuracy of the American Joint Committee on Cancer, eighth edition and enable high-risk cutaneous SCCs to be stratified more reliably, but their prognostic accuracy is still low. Considering the accumulation of risk factors may improve high-risk cutaneous SCC risk stratification.

Definition of prognostic subgroups in the T3 stage of the eighth edition of the American Joint Committee on Cancer staging system for cutaneous squamous cell carcinoma: Tentative T3 stage subclassification.

BACKGROUND: Although the eighth edition of the American Joint Committee on Cancer staging system (AJCC8) provides improved prognosis stratification of cutaneous squamous cell carcinoma (CSCC) over AJCC7, T3 has a variable prognosis. OBJECTIVE: To define prognostic subgroups in T3-AJCC8 CSCC. METHODS: Retrospective cohort study of 196 primary T3-AJCC8 CSCCs. We conducted multidimensional scaling analysis using the 6 risk factors that define T3 CSCCs. The prognoses of the groups obtained were analyzed by means of competing risk analysis. RESULTS: Group 1 was characterized by a tumor thickness greater than 6 mm (without invasion beyond the subcutaneous fat), alone or in combination with a tumor width of at least 4 cm. Group 2 was characterized by the presence of either invasion beyond the subcutaneous fat or by the involvement of nerves (≥0.1 mm, or deeper than the dermis). Group 3 was characterized by the combination of both T3b risk factors, or of 3 or more risk factors. Group 3 (tentatively named T3c) patients had the worst prognosis for disease-specific poor outcome events and major events, Group 2 (T3b) had intermediate risk, and Group 1 (T3a) had the best prognosis (disease-specific poor outcome events: hazard ratio [HR], 1.94; P = .00009; major events: HR, 2.55; P = .00001; disease-specific death: HR, 10.25; P = .0009). LIMITATIONS: Retrospective study. CONCLUSIONS: There is statistically significant evidence that T3-AJCC8 may be classified into distinct prognostic subgroups.