235P - Evaluation of a prognostic model for head and neck cutaneous squamous cell carcinoma using a cumulative number of risk factors

Background

Cutaneous squamous cell carcinoma (cSCC) is the second most common cutaneous malignant tumor, occurring frequently in the head and neck. The latest National Comprehensive Cancer Network (NCCN) Guidelines classify cSCC into low-, high-, and very-high-risk groups using multiple risk factors. There were no detailed and accurate prognostic models for the newly established “very-high-risk” group. We identified the following four important risk factors for local relapse, nodal relapse, and SCC-related death in a previous study: recurrent tumor, high-risk histologic features, >6-mm tumor thickness or invasion beyond the subcutaneous fat layer, and lymphatic or vascular involvement. We evaluated whether the number (no.) of NCCN very-high-risk factors or our important risk factors could predict disease prognosis.

Methods

We retrospectively examined very-high-risk head and neck (H&N) cSCC patients (pts) without clinical nodal involvement. Pts were categorized into 3 groups based on the no. of very-high-risk factors (1, 2, or ≥3) or no. of our important risk factors (0, 1, or ≥2). Primary outcomes were local relapse, regional relapse, and SCC-related death. Cumulative incidence function (CIF) was used to describe SCC-related death probability and competing risk mortality. Gray’s test was used to compare differences in CIF between the groups.

Results

In total, 249 very-high-risk H&N cSCC pts were included. The no. of pts with 1, 2, or ≥3 very-high-risk factors was 176, 62, and 11, respectively. The no. of pts with 0, 1, or ≥2 our important risk factors was 27, 177, and 45, respectively. Neither the no. of very-high-risk factors nor the no. of our important risk factors was significantly associated with local relapse (P=0.60 and 0.14) and SCC-related death (P=0.10 and 0.12). Likewise, the no. of very-high-risk factors was not significantly associated with regional relapse (P=0.63). However, pts with a higher no. of our important risk factors were significantly more likely to develop regional relapse (P=0.03).

Conclusions

Although the no. of very-high-risk factors did not predict the risk of relapse and death, the no. of our important risk factors may predict the higher risk of regional relapse in pts with H&N cSCC.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

S. Matsushita: Financial Interests, Personal, Invited Speaker: Bristol Myers-Squibb, Merck Sharp & Dohme, Novartis Pharma, Ono Pharmaceutical. H. Kato: Financial Interests, Personal, Invited Speaker: Novartis Pharma, Ono Pharmaceutical. S. Saito: Financial Interests, Personal, Invited Speaker: Ono Pharma. M. Yasuda: Financial Interests, Personal, Invited Speaker: Bristol Myers-Squibb, Ono Pharmaceutical. N. Fujimoto: Financial Interests, Personal, Invited Speaker: Novartis Pharma, Ono Pharmaceutical. Y. Kawahara: Financial Interests, Personal, Invited Speaker: Maruho. Y. Nakamura: Financial Interests, Personal, Invited Speaker: Bristol Myers-Squibb, Ono Pharmaceutical, Novartis Pharma, Merck Sharp & Dohme, Maruho, Kyowa Kirin, Tanabe Mitsubishi Pharma, Novartis Pharma; Financial Interests, Personal, Advisory Board: Novartis Pharma; Financial Interests, Institutional, Funding: Ono Pharmaceutical, Kaken Pharma, Torii, POLA Pharma. All other authors have declared no conflicts of interest.