This retrospective single-institution cohort study evaluated the role of conversion surgery after first-line immunotherapy combined with chemotherapy and/or trastuzumab in patients with gastric cancer who had peritoneal metastasis (P1) and/or positive peritoneal lavage cytology (CY1) at initial diagnosis.
The cohort included 144 patients confirmed by staging laparoscopy and treated between November 2019 and March 2024. The study compared outcomes between patients who underwent conversion surgery after an observed response to systemic therapy and those who continued with palliative therapy.
The analysis was retrospective and conducted at a single institution. Inclusion required confirmation of P1 and/or CY1 by staging laparoscopy at initial diagnosis. All patients received first-line immunotherapy with chemotherapy and/or trastuzumab. Patients were categorized into a conversion therapy group or a palliative therapy group based on response assessment and subsequent management.
Patients received combined systemic therapy consisting of immunotherapy plus chemotherapy, with trastuzumab added when indicated. Initial staging laparoscopy established the presence of peritoneal metastasis and/or positive cytology. Those who demonstrated a clinical response to systemic treatment underwent a second-look laparoscopy to reassess peritoneal disease status prior to considering surgical conversion.
Out of 144 patients treated, 75 showed a clinical response and proceeded to second-look laparoscopy. Negative peritoneal metastasis and cytology were confirmed in 36 patients; all 36 subsequently underwent conversion surgery. The decision to proceed to surgery was based on the negative findings at reassessment laparoscopy, indicating clearance of detectable peritoneal disease and cytology.
Among the 36 patients who underwent conversion surgery, 27 (75.0%) achieved R0 resection—defined as complete resection with no residual tumor. The abstract reports these operative outcomes without additional detail on surgical technique, perioperative morbidity, or length of stay; such details were not provided in the source summary.
The study reported substantial survival differences favoring the conversion surgery group. Median progression-free survival (PFS) was 22.9 months in the conversion group versus 5.8 months in the palliative group (P < 0.001). Median overall survival (OS) was 33.4 months for the conversion group compared with 12.3 months for the palliative group (P < 0.001). These comparisons indicate a statistically significant survival benefit associated with conversion surgery after effective systemic therapy in this cohort.
Peritoneal tumor burden, measured by the peritoneal cancer index (PCI), influenced outcomes among operated patients. Those with PCI ≤2 had significantly longer median OS than patients with PCI >2. For the PCI ≤2 subgroup, median OS was not reached during follow-up, whereas median OS for PCI >2 was 28.5 months (P < 0.006). The abstract does not provide additional subgroup sizes or time-to-event curves.
The authors report that no serious treatment-related adverse events led to treatment discontinuation or death in the study population. The abstract does not enumerate specific toxicities, grades, or rates of non-fatal adverse events, so detailed safety profiling is not available from the source text.
The investigators conclude that conversion surgery following immunochemotherapy provided a significant survival benefit for gastric cancer patients with P1 and/or CY1 in their cohort. They recommend active reassessment with staging laparoscopy when immunochemotherapy is effective to determine surgical feasibility. The abstract emphasizes that conversion surgery can be considered when peritoneal disease and cytology become negative after systemic therapy, and that lower PCI (≤2) identifies patients with particularly favorable outcomes.
Notes on limitations reported in the source summary: the abstract reflects a retrospective, single-institution design and provides limited details on treatment regimens, surgical procedures, morbidity, and subgroup sizes. Those specifics were not reported in the provided source abstract.