Objectives The four-tied peritoneal regression grading score (PRGS) is increasingly used to evaluate the response of peritoneal metastases (PM) to chemotherapy. The minimal number of peritoneal biopsies needed for PRGS determination remains unclear. Methods A prospective cohort of 89 PM patients treated with 210 pressurized intraperitoneal aerosol chemotherapy (PIPAC) cycles was investigated. Four biopsies from every abdominal quadrant were recommended. Histological tumor response was defined as a stable or decreasing mean PRGS between therapy cycles, progression increasing. We compared the diagnostic uncertainty induced by missing biopsies to the histological response. Results A total of 49 patients had at least two PIPAC and were eligible for therapy response assessment. Mean PRGS decreased from 2.04 (CI 5–95% 1.85–2.27) to 1.79 (CI 5–95% 1.59–2.01), p=0.14, as a proof of therapy effectiveness. 35 (71.4%) patients had a stable or decreasing PRGS (therapy response), 14 (28.6%) a PRGS increase (disease progression). Histology showed agreement between four biopsies in 42/210 laparoscopies (20%), between ≥3 biopsies in 103 (49%), and between ≥2 biopsies in 169 laparoscopies (81%). Mean loss of information with one missing biopsy was 0.11 (95% CI=0.13) PRGS points, with two missing biopsies 0.18 (95% CI 0.21). In 9/49 patients (18.3%), the loss of information with one less biopsy exceeded the change in PRGS under therapy. Conclusions A minimum of three biopsies is needed to diagnose PM progression with an accuracy superior to 80%. Missing biopsies often result in a false diagnosis of tumor progression.
