Real-world (RW) systemic treatment patterns and clinical outcomes in locally advanced (LA) and recurrent or metastatic (R/M) cutaneous squamous cell carcinoma (cSCC) patients (pts).

Abstract

e19284 Background: Data is limited on RW systemic treatments, objective response rate (ORR), response duration (DOR), overall survival (OS), and time-to-treatment discontinuation (TTD) for cSCC. Clinical trials (CTs) use RECIST criteria for response assessment; however, frequency of follow-up imaging and availability/precision of tumor lesion measures outside of trials can vary across pts. Understanding the types of image-informed responses and outcomes available in clinical practice may inform RW study design and interpretation. Methods: Adults with LA and R/M cSCC receiving first-line (1L) systemic treatment from 09/04/2014 to 06/30/2017 in US Oncology community practices, with follow-up to 06/30/2018, were included. Baseline radiologic scan was required; pts in CTs or who had concurrent cancers or immunosuppressive disorders/treatments were excluded. The primary endpoint was RW ORR using RECIST 1.1, based on tumor lesion sizes (if available) in imaging reports from the iKnowMed electronic health record. If RECIST could not be applied, physician-assessed responses were used. Pts not evaluable for response were classified as non-responders. DOR, OS, and TTD were assessed using Kaplan-Meier methods. Results: 51 (11 LA + 40 R/M) cSCC pts were eligible. Median age was 76 yrs, 80% male, 86% Caucasian, 65% ECOG PS of 0/1. The primary location was head/neck in 67%. The most common 1L regimens were cetuximab (51%), carboplatin + paclitaxel (22%), and cisplatin (6%). Conclusions: Cetuximab was the most common treatment. OS and TTD in advanced cSCC were short. RW ORR was lower using RECIST than with physician assessment; however, few pts were evaluable by RECIST due to differences in available data in prospective vs. retrospective settings. [Table: see text]