12-month uptake of PD-L1 testing and atezolizumab (atezo) + nab-paclitaxel (nab-pac) treatment in metastatic triple-negative breast cancer (mTNBC) following accelerated FDA-approval in the United States.
e13103 Background: In March 2019, atezo + nab-pac received accelerated FDA-approval for the treatment (tx) of patients (pts) with unresectable locally advanced or metastatic TNBC who express PD-L1 in immune cells (IC), based on the results from the randomized IMpassion130 clinical trial. A companion diagnostic, the Ventana SP142 PD-L1 assay, was concurrently approved. This real-world data analysis describes the uptake of both PD-L1 testing and therapy with atezo + nab-pac from the time of approval. Methods: This is a retrospective study using the Flatiron Health electronic health record (EHR) derived de-identified database, representing > 2.4 million US cancer pts from > 280 cancer clinics. Pts diagnosed with mTNBC starting any line of systemic therapy between March 9, 2019 and December 31, 2019 (to be updated in April 2020 for 12-months follow-up) were included. Two cohorts were examined: pts receiving either atezo-based tx, or tx without atezo. Demographics were described in the pt group overall and by line of therapy (LOT). Steroid tx at baseline was classified as any order/administration of a steroid within the first three administrations of a systemic therapy in a given LOT. Any record of a PD-L1 test was reported. Results: 228 pts diagnosed with mTNBC were included; 65 pts (28%) received any atezo combination, the majority in combination with nab-pac: 57% (n = 37) in first-line (1L), 25% (n = 16) in second-line (2L), and 18% (n = 12) in third-line or later (3L+). Median age at metastatic diagnosis (mDx) for atezo-treated pts was: 1L, 65 y (IQR: 49, 72); 2L, 56 y (IQR: 48, 64); 3L+, 52 y (IQR: 42, 60). Median age at mDx for non-atezo-treated patients was: 1L, 63 y (IQR: 53, 74); 2L, 59 y (IQR: 51, 65); 3L+, 61 y (IQR: 53, 71). Of 228 pts, 158 (69%) had any PD-L1 test recorded. Among atezo-treated pts, 62 (95%) had a documented PD-L1 test; of those 40 (62%) were tested using the SP142 assay. Overall, 205 tests were recorded: SP142, n = 96 (46.8%); unknown/not documented, n = 44 (21.5%); 22C3, n = 36 (17.6%); lab-developed test, n = 26 (12.7%); SP263, n = 2 (1.0%); and 28-8, n = 1, (0.5%). 37 (57%) atezo-treated pts received a steroid at baseline. Conclusions: Atezo tx and testing with SP142 is being implemented following accelerated FDA-approval. A considerable proportion of pts received steroids at baseline and had nab-pac as chemotherapy backbone in clinical practice. Clinical outcomes of atezo-treated pts in the real-world setting remain to be further explored pending longer pt follow-up times.