Perioperative chemotherapy for operable gastro-esophageal or gastric cancer: Anthracycline triplets versus FLOT

The MAGIC trial (2006) showed improved survival of patients with gastric or gastro-esophageal junction (GEJ) carcinoma treated with perioperative chemotherapy (epirubicin, cisplatin and fluorouracil) compared to surgery alone. This resulted in anthracycline triplets becoming standard of care in the Netherlands. In 2019, the FLOT4-AIO study showed further improved survival in patients treated with perioperative fluorouracil, leucovorin, oxaliplatin and docetaxel (FLOT). The aim of this study was to compare the outcomes of anthracycline triplets to FLOT chemotherapy in the Netherlands, using real-world population level data.

Patients diagnosed with resectable (cT2-4a/cTx N0-3/Nx M0) gastric/GEJ carcinoma between 2015 and 2020 who were treated with at least neoadjuvant anthracycline triplets or FLOT were selected from the Netherlands Cancer Registry. The primary outcome was overall survival (OS) in months from start of neoadjuvant therapy. Secondary outcomes included the pathological complete response (pCR), proportion of patients that fully completed neoadjuvant chemotherapy (100% of scheduled cycles permitting dose reductions), proportion of patients that underwent surgical resection and with radical resection (R0), and proportion of patients receiving adjuvant therapy. Multivariable cox regression analysis was performed to assess OS adjusted for sex, age, comorbidities, performance status, clinical T-stage/N-stage and tumor grade.

In total, 1691 patients were included. 913 (54%) patients were treated with anthracycline triplets and 778 (46%) with FLOT. Median age was respectively 66 years (interquartile range [IQR] 59-71) and 67 years (IQR 59-73) (p=0.032). The majority of patients was male (67%) and classified as WHO 0-1 at baseline (anthracycline triplets 76.2% vs. FLOT 80.2%, p < 0.001). Patients treated with anthracycline triplets underwent fewer diagnostic laparoscopies (DLS) than patients treated with FLOT (44.14% vs. 73.52%, p < 0.0001). Multivariable adjusted OS was better in patients treated with FLOT compared to anthracycline triplets (HR = 0.84, 95% CI 0.72-0.98, p=0.03). Median OS was 39.6 months (95% CI 35.2-47.2) for anthracycline triplets and 49.7 months (95% CI 41.3-NA) for FLOT, with 3-year OS rates of 52.7% and 56.4% respectively (p=0.06). The proportion of patients that completed the full neoadjuvant chemotherapy regimen was higher with FLOT (78.5% vs. 73.1%, p=0.009). More patients underwent surgical resection after FLOT (90.4% vs. 87.3%, p=0.06). Of the 1500 patients that underwent resection, patients treated with FLOT showed a similar proportion of pCR (10.4% vs. 7.8%, p=0.09), had a higher percentage of R0 resections (86.2% vs. 85.2%, p=0.007), and showed no difference in adjuvant therapy rate (63.6% vs. 58.9%, p=0.06).

In real-world population level data showed better overall survival of patients treated with FLOT chemotherapy compared to anthracycline triplets. No statistically significant difference was seen in pCR rates. Patients treated with FLOT did complete the full neoadjuvant chemotherapy regimen significantly more often and had more R0 resections. Not every outcome as described in the FLOT4-AIO trial could be reproduced in a real-world population, despite the large patient cohort and improved diagnostics (DLS) in the FLOT group, perhaps due to the poorer performance status and/or less intensive neoadjuvant treatment that is seen in the real world.

The authors.

Has not received any funding.

All authors have declared no conflicts of interest.