The phase II PHERFLOT trial has demonstrated remarkable efficacy in treating HER2-positive localized esophagogastric adenocarcinoma, achieving a pathological complete response (pCR) rate of 48.4% with the combination of pembrolizumab, trastuzumab, and FLOT chemotherapy. The results, presented at ESMO Congress 2025 and published in Nature Medicine, represent a significant advancement over current standard-of-care treatments.
Trial Design and Patient Population
The open-label, multicenter study enrolled 31 patients across 11 German centers between March 2023 and May 2024. Patients received pembrolizumab (200 mg) and trastuzumab (initial loading dose of 8 mg/kg, followed by 6 mg/kg) every 3 weeks for 3 cycles before surgery, combined with standard FLOT chemotherapy (docetaxel, oxaliplatin, leucovorin, and 5-FU) administered every 2 weeks for 4 cycles.
The median patient age was 65 years, with most tumors located at the gastroesophageal junction (77%). HER2 overexpression was confirmed by immunohistochemistry, with 80.6% scoring 3+. PD-L1 combined positive score (CPS) was ≥1 in 85% of evaluated cases, reflecting a generally inflamed tumor phenotype.
Pathological Response Outcomes
Among the 30 patients who underwent surgery, all achieved R0 resection. The pCR rate reached 50.0% (95% CI 31.3-68.7) in resected patients and 48.4% (95% CI 30.2-66.9) in the intention-to-treat population. An additional 19.4% achieved subtotal regression, resulting in a major pathological response rate of 67.7%.
These outcomes substantially exceed historical benchmarks: 15% pCR in FLOT4, 19% in MATTERHORN, and up to 35% in HER2-targeted regimens such as PETRARCA. Dr. Joseph Tintelnot, who presented the interim analysis, noted that the regimen shows "clear biological synergy and potential superiority over current standards."
Safety Profile and Surgical Outcomes
The safety profile was consistent with expectations, though certain toxicities were elevated. Grade 3 diarrhea occurred in 38.7% of patients, compared to approximately 6% in trials of FLOT alone. Most cases of diarrhea were attributed to FLOT (83%), while 50% were considered related to pembrolizumab or trastuzumab.
Surgery was performed in 30 patients with a mean time to surgery of 88.9 days after enrollment. While 70% of patients experienced no postoperative complications, the reoperation rate was 26.7%, higher than the 10% reported in FLOT4. However, the median hospitalization time was 14 days, comparable to FLOT4, and no 30-day postoperative mortality occurred.
Two patients experienced fatal serious adverse events, both unrelated to therapy: one died from hyperglycemia and sepsis within 60 days postoperatively, and another from acute respiratory distress syndrome approximately 3.5 months after surgery.
Biomarker-Driven Responses
Exploratory analyses revealed differential responses across molecular subgroups. Patients with PD-L1 CPS ≥10 demonstrated higher pCR rates (63.6%) compared to those with CPS 1-9 (40.0%). Notably, even PD-L1-negative patients showed activity, with 33.3% achieving pCR and 50% achieving subtotal regression.
HER2 IHC 3+ tumors had higher pCR rates (52.0%) than IHC 2+/ISH+ tumors (33.3%). Among early-stage tumors (T1 and T2), the pCR rate reached 77.8%, while T3 and T4 tumors achieved a combined pCR and subtotal regression rate of 66.7%.
All three patients with deficient mismatch repair (dMMR) achieved complete pathological responses, suggesting enhanced treatment sensitivity in this biomarker-enriched subgroup.
Clinical Implications and Future Directions
The PHERFLOT trial's co-primary endpoints include both pCR rate and 2-year disease-free survival. While the pCR endpoint has been met, longer follow-up is needed to assess survival outcomes. The median follow-up time of 14.8 months is insufficient to evaluate the 2-year disease-free survival rate.
The high response rates observed raise important questions about treatment de-escalation and organ-preserving approaches. With nearly 50% of patients achieving complete pathological response and an additional 20% showing subtotal regression, the findings may prompt discussions about nonoperative management strategies, similar to those established for dMMR/MSI-high rectal cancer.
Study Limitations and Context
The trial's single-arm design and conduct in a single country may limit generalizability. Additionally, the absence of mature disease-free survival data represents a key limitation of the current analysis. The increased perioperative morbidity, particularly the higher reoperation rate, raises questions about whether the improved pathological responses will translate into meaningful survival benefits.
Despite these limitations, the PHERFLOT trial provides compelling evidence for the efficacy of combining immunotherapy and HER2-targeted therapy with intensive chemotherapy in localized HER2-positive esophagogastric adenocarcinoma. The results support further investigation in larger randomized trials and may inform future treatment paradigms for this molecularly defined patient population.
