Eva Versteijne1, Mustafa Suker2, Jesse V Groen2, Marc G Besselink3, Bert A Bonsing4, Koop Bosscha5, Olivier R Busch3, Ignace H J T de Hingh6, Koert P de Jong7, I Quintus Molenaar8, Hjalmar C van Santvoort9, Helena M Verkooijen10, Casper H Van Eijck2, Geertjan van Tienhoven1. 1. Department of Radiation Oncology, Cancer Center Amsterdam, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands. 2. Department of Surgery, Erasmus MC Cancer Institute, Rotterdam, the Netherlands. 3. Department of Surgery, Cancer Center Amsterdam, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands. 4. Department of Surgery, Leiden University Medical Center, Leiden, the Netherlands. 5. Department of Surgery, Jeroen Bosch Hospital, Den Bosch, the Netherlands. 6. Department of Surgery, Catharina Hospital, Eindhoven, the Netherlands. 7. Department of Surgery, University Medical Center Groningen, University of Groningen, Groningen, the Netherlands. 8. Department of Surgery, RAKU, University Medical Center Utrecht, Utrecht, the Netherlands. 9. Department of Surgery, RAKU, St Antonius Hospital, Nieuwegein, the Netherlands. 10. Imaging Division, University Medical Center Utrecht, Utrecht, the Netherlands.
Abstract
OBJECTIVES: To investigate the accrual proportion and patients' reasons for not participating in the PREOPANC trial on neoadjuvant chemoradiotherapy versus immediate surgery in resectable and borderline resectable pancreatic cancer, and to compare these patients' outcomes with those of patients who had been randomized in the trial. SUMMARY OF BACKGROUND DATA: The external validity of multicenter randomized trials in cancer treatment has been criticized for suboptimal non-representative inclusion. In trials, it is unclear how outcomes compare between randomized and nonrandomized patients. METHODS: At 8 of 16 participant centers, this multicenter observational study identified validation patients, who had been eligible but not randomized during recruitment for the PREOPANC trial. We assessed the accrual proportion, investigated their most common reasons for not participating in the trial, and compared resection rates, radical (R0) resection rates, and overall survival between the validation patients and PREOPANC patients, who had been randomized in the trial to immediate surgery. RESULTS: In total, 455 patients had been eligible during the recruitment period, 151 of whom (33%) had been randomized. Fifty-five percent of the 304 validation patients had refused to participate. Median overall survival in the validation group was 15.2 months, against 15.5 months in the PREOPANC group (P = 1.00). The respective resection rates (76% vs 73%) and R0 resection rates (51% vs 46%) did not differ between the groups. CONCLUSIONS: The PREOPANC trial included a reasonable percentage of 33% of eligible patients. In terms of the outcomes survival, resection rate, and R0 resection rate, this appeared to be a representative group.
OBJECTIVES: To investigate the accrual proportion and patients' reasons for not participating in the PREOPANC trial on neoadjuvant chemoradiotherapy versus immediate surgery in resectable and borderline resectable pancreatic cancer, and to compare these patients' outcomes with those of patients who had been randomized in the trial. SUMMARY OF BACKGROUND DATA: The external validity of multicenter randomized trials in cancer treatment has been criticized for suboptimal non-representative inclusion. In trials, it is unclear how outcomes compare between randomized and nonrandomized patients. METHODS: At 8 of 16 participant centers, this multicenter observational study identified validation patients, who had been eligible but not randomized during recruitment for the PREOPANC trial. We assessed the accrual proportion, investigated their most common reasons for not participating in the trial, and compared resection rates, radical (R0) resection rates, and overall survival between the validation patients and PREOPANC patients, who had been randomized in the trial to immediate surgery. RESULTS: In total, 455 patients had been eligible during the recruitment period, 151 of whom (33%) had been randomized. Fifty-five percent of the 304 validation patients had refused to participate. Median overall survival in the validation group was 15.2 months, against 15.5 months in the PREOPANC group (P = 1.00). The respective resection rates (76% vs 73%) and R0 resection rates (51% vs 46%) did not differ between the groups. CONCLUSIONS: The PREOPANC trial included a reasonable percentage of 33% of eligible patients. In terms of the outcomes survival, resection rate, and R0 resection rate, this appeared to be a representative group.