Abstract 1093P
Background
In recent years, tumor microenvironment (TMe) and tumor budding (TB) have gained greater attention in the prognostication of tumors, including cutaneous squamous cell carcinoma (CSCC). A better characterization of TMe and TB might increase our knowledge on high-risk CSCC and improve future staging systems. Objective: To define prognostic subgroups according to the features of the TMe and the TB.
Methods
The features of TMe (peritumoral inflammatory infiltrate, tumoral stroma and desmoplasia) and TB were evaluated in a retrospective cohort study of 190 primary HR-CSCC. Unsupervised multidimensional scaling (MDS) analysis was conducted to identify subgroups based on the combination of these risk factors. The time-dependent outcomes of the groups were analyzed considering competing risks. Multivariate analysis was conducted to check for the independence of TMe and TB in prognosis together with the AJCC8 risk factors.
Results
The features that define TMe and TB were associated among them and thus we aimed to search for clusters based on the aforementioned variables. MDS analysis identified three subsets of CSCC according to TMe and TB. Group-3 (very high risk) showed absent peritumoral inflammation, aberrant stroma, desmoplasia and tumor budding at the same time; Group-2 (high risk) displayed any but not all of these risk factors combined; and Group-1 (low risk) was characterized by the absence of all these risk factors. The combination of the TMe+TB subgroups with the AJCC8 high-risk features in multivariate Cox regression models, demonstrated the independence of TMe and TB in the prognosis of CSCC. Indeed, the very-high risk group showed 2.5 times greater risk for local recurrence and 6.5 times greater risk for metastasis and disease-specific death compared with the low risk group in the multivariate analysis.
Conclusions
There is statistically significant evidence to conclude that the features of TMe and the TB may identify prognostic subgroups in HR-CSCC, and they may be considered for upcoming staging systems.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
PI18/00587 (Instituto de Salud Carlos III, Cofinanciado con Fondos FEDER).
Disclosure
All authors have declared no conflicts of interest.