3474 - High Rates of Local Failure in the Triangle Volume for Non-Irradiated Borderline Resectable and Locally Advanced Pancreatic Cancer
Presenter(s)
I. C. Liu1, S. Sehgal1, S. Mao1, T. A. Lin2, A. V. Reddy3, C. Hill4, M. B. Roumeliotis1, J. He5, D. Laheru6, J. J. Meyer1, C. Hu7, and A. Narang1; 1Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, 2Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, 3Northside Radiation Oncology Consultants, Canton, GA, 4Department of Radiation Oncology, New York University Grossman School of Medicine, New York, NY, 5Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, MD, 6Sidney Kimmel Comprehensive Cancer Center, Department of Oncology, Johns Hopkins University School of Medicine, Baltimore, MD, 7Division of Biostatistics and Bioinformatics, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD
Purpose/Objective(s): For patients with borderline resectable and locally advanced pancreatic cancer (BRPC/LAPC) who are candidates for surgical resection, the benefit of preoperative radiotherapy (RT) beyond neoadjuvant chemotherapy alone remains controversial. While randomized studies have yielded mixed results, recent institutional data suggest that expanding the RT target volume to electively cover the “Triangle Volume” (TV) - defined by the extra-pancreatic perineural tract anatomy at risk for microscopic involvement – may significantly improve locoregional control. Given the controversy of pre-operative RT for pancreatic cancer, we herein analyze locoregional control and patterns of failure in a cohort of BRPC/LAPC patients who underwent resection following neoadjuvant chemotherapy alone, as these data may help demonstrate to the multi-disciplinary community the impact of omission of pre-operative RT in this setting.
Materials/Methods: We retrospectively reviewed BRPC/LAPC patients who underwent gross total resection without preoperative RT at our institution from 2015 through 2022. Baseline patient and disease characteristics, treatment details, surgical outcomes, and survival data were collected. Endpoints included freedom from local failure (FFLF) and local progression free survival (LPFS), all defined from date of surgery. A FFLF event was defined as radiographic evidence of locoregional recurrence, while LPFS events also included death. Censoring was done at the last imaging date for both endpoints. Association between clinical characteristics and LPFS were analyzed using univariate and multivariable Cox proportional hazards models.
Results: A total of 43 patients met inclusion criteria, including 34 with BRPC (79%) and 9 with LAPC (21%). All patients received neoadjuvant chemotherapy, with 53% receiving =4 months. FOLFIRINOX was the most common regimen (60%). On surgical pathology, 91% achieved an R0 resection. Despite this high rate of margin-negative resection, 28 patients (65.1%) experienced local failure, with 2-year FFLF and LPFS rates of 33% (95% CI: 18 – 48%) and 25% (95% CI: 12 – 38%), respectively. Notably, 25 (89%) of the failures occurred in the TV. On univariate and multivariable Cox analysis, risk factors for LPFS included positive margin (HR 9.3, 95% CI: 2.5 – 34), perineural invasion (HR 2.9, 95% CI 1.1 – 7.7), and lymphovascular invasion (HR 2.5, 95% CI: 1.2 – 5.4).
Conclusion: Despite a high R0 resection rate, locoregional control remained poor in BRPC/LAPC patients who were managed with neoadjuvant chemotherapy alone prior to resection. Importantly, nearly all locoregional failures occurred within the TV, reinforcing the potential benefit of including this region in the RT clinical target volume. These findings support further investigation into the optimal integration and delivery of preoperative RT in the multi-disciplinary treatment of pancreatic cancer.