Feasibility of Endoscopic Ultrasound-Guided Tissue Acquisition for Next-Generation Sequencing in Pancreatic Cancer Patients

Aims Next-generation sequencing (NGS) of tumors can identify prognostic and therapeutic molecular targets. Endoscopic ultrasound-guided fine needle aspiration/biopsy (EUS-FNA/B) is the established modality of choice to diagnose pancreatic cancer. However, reported feasibility outcomes of EUS-FNA/B for NGS have been variable. The aims of this study were 1) to determine whether EUS-FNA/B provides adequate genomic material for NGS and 2) to assess whether tumor clinical phenotype and procedural factors of tissue acquisition affect NGS success rates.

Methods We describe a retrospective study of all EUS-FNA/B specimens with pancreatic cancer at Memorial Sloan Kettering Cancer Center in New York City between 2014-2023. NGS was performed with a comprehensive paired tumor-normal, large panel, hybridization-capture based, NGS known as Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT). Relevant data were obtained from the electronic medical record, and MSK-IMPACT results were analyzed.

Results MSK-IMPACT testing was requested on 464 EUS-FNA/B specimens with pancreatic adenocarcinoma and was successful on 356 (77%) of them. Success rates of the assay significantly improved over time with changes in DNA extraction and sequencing optimization (88% in 2021-2023 vs 62% in 2015-2017, p<0.01). Most failed specimens were due to a low DNA yield, indicating low tissue sampling. Although 21% had failed MSK-IMPACT testing, all specimens were adequate for routine diagnostic cytology. MSK-IMPACT success did not significantly differ based on tumor size, location within the pancreas, disease stage, and exposure to prior therapy (i.e. radiation, chemotherapy, or both). With regards to procedural parameters, the number of passes did not significantly affect MSK-IMPACT success. However, the use of a 22G needle to acquire tissue resulted in significantly improved MSK-IMPACT performance compared to the use of a 25G needle. Of the successful cases, 96% identified a significant hotspot mutation, of which 93% identified a KRAS driver mutation.

Conclusions This is the largest retrospective review to date investigating the feasibility of EUS-FNA/B for NGS. EUS-FNA/B reliably provided adequate tissue for NGS with improved success rates over time. This is also the first study to determine that disease stage and exposure to prior therapy did not significantly affect NGS performance. The usage of a 22G needle to acquire tissue yielded better NGS success compared to a 25G needle. Most hotspot mutations identified were KRAS driver mutations, which has significant implications for further development and implementation of KRAS directed therapy to treat pancreatic cancer.

Conflicts of Interest

Authors do not have any conflict of interest to disclose.

Publication History

Article published online:
27 March 2025