Is complete really complete? Prognostic divergence of pathological complete response after FLOT versus CROSS in esophageal adenocarcinoma.

OBJECTIVE: To review whether pathological complete response (pCR) in esophageal adenocarcinoma (AC) may be regime-dependent and to outline its implications for surveillance, adjuvant therapy and trial design. BACKGROUND: ESOPEC established perioperative FLOT as superior to preoperative CROSS for survival and distant disease control in resectable esophageal AC, but neither analysis stratified outcomes by pCR status. To our knowledge, this is the first synthesis focused on the prognostic meaning of pCR by treatment modality. METHODS: Semi-structured search of MEDLINE, Embase, Cochrane Library, and Scopus (2010 to June 2026) for randomized trials, propensity-matched cohorts, registry analyses, individual patient data (IPD) meta-analyses, and circulating tumor DNA (ctDNA) studies. Squamous-cell carcinoma (SCC) was supportive context. RESULTS: Four patient-level cohorts (Cools-Lartigue, n = 465 pCR; Capovilla, n = 102 pCR of 563 total; Shridhar, n = 328 pCR of 1007 matched pairs; Alhayo, n = 297 pCR [150 CROSS, 147 FLOT]) and one SCC IPD meta-analysis (Okui, n = 1044), included as supportive cross-histology context only (Methods), suggest superior outcomes within the pCR subgroup after chemotherapy (5-y RFS 87.1 % vs 75.3 %, HR 1.70; 5-y OS 97.5 % vs 70.4 % in SCC). In the 14-centre European Alhayo cohort, adjusted overall survival was significantly better after FLOT (HR 0.34, 95 % CI 0.16-0.69) with markedly lower recurrence (8.8 % vs 24.7 %). In ESOPEC, FLOT reduced 3-year distant recurrence to 31.5 % versus 47.2 % (HR 0.59). Pathological complete response remains a favorable prognostic indicator relative to residual disease after neoadjuvant therapy, but postoperative ctDNA may further stratify recurrence risk within the pCR population. Patients with pCR who remain ctDNA-positive carry a 40 % 1-year recurrence risk in a small squamous-cell-carcinoma post-hoc cohort (not yet confirmed in adenocarcinoma), consistent with occult molecular residual disease undetectable by pathology alone. All current evidence is retrospective. CONCLUSIONS: The available data are consistent with, but do not prove, regime-dependent prognostic meaning of pCR. Whether recurrence patterns after pCR differ at the randomized-trial level remains unanswered. A multicenter IPD pooled analysis of approximately 1630 pCR patients with ctDNA assessment is the next research step.