Endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) has high diagnostic accuracy for evaluating solid pancreatic lesions (SPLs), independent of rapid on-site evaluation (ROSE), investigators reported in Gastroenterology.
The randomized, noninferiority study was conducted at 14 centers in 8 countries and included patients aged ≥18 years with SPLs who were referred for EUS evaluation. Participants were randomly assigned (1:1) to undergo EUS-FNB or EUS-FNB+ROSE, with use of new-generation FNB needles. ROSE was performed using touch imprint cytology.
The study cohort included 771 patients in total (mean age, 67.5±11.5 years; 56.4% men), with 385 in the EUS-FNB+ROSE arm and 386 in the EUS-FNB alone arm.
In the EUS-FNB+ROSE arm, the diagnostic accuracy was 96.4% (95% CI, 94.0%-98.0%), with a sensitivity of 96.0% (95% CI, 93.5%-97.7%), and specificity of 100% (95% CI, 73.5%-100.0%). In the EUS-FNB alone arm, the diagnostic accuracy was 97.4% (95% CI, 95.3%-98.8%), with a sensitivity of 97.3% (95% CI, 95.1%-98.7%), and specificity of 100% (95% CI, 78.2%-100.0%). The difference in accuracy between study arms was 1.0% (2-sided 90% CI, –1.1%-3.1%). Noninferiority was shown for diagnostic accuracy (P <.001).
In addition, a significantly higher rate of tissue core was observed in the EUS-FNB alone arm compared with the EUS-FNB+ROSE arm (78.0% vs 70.7%, respectively; P =.021), with a significantly shortened procedural time (mean time, 11.7±6.0 vs 17.9±8.8 minutes; P <.0001).
A nonrandomized comparison found lower accuracy and core procurement rates, along with a higher blood contamination of samples taken with the 20G side-fenestrated needles vs the 22G end-cutting needles. The safety profile was similar for all needles.
Researchers noted some study limitations. Some pathologists had limited experience with touch imprint cytology, which could result in bias in favor of EUS-FNB alone. Additionally, all centers were highly experienced, potentially reducing reproducibility of results. Finally, the study was performed with the use of 3 different histologic needles without randomization.
“[I]n patients with SPLs our study demonstrated that diagnostic accuracy of EUS-FNB alone with three needle passes reached 97% and was non-inferior compared to EUS-FNB+ROSE, rendering ROSE not routinely recommended,” the researchers advised. “Based on these results, we strongly believe that EUS-FNB should become standard of care for SPLs evaluation.”
Disclosure: Some of the study authors declared affiliations with device companies. Please see the original reference for a full list of authors’ disclosures.
Crinò SF, Di Mitri R, Nguyen NQ, et al. EUS-guided fine-needle biopsy with or without rapid on-site evaluation for diagnosis of solid pancreatic lesions: a randomized controlled non-inferiority trial. Gastroenterol. Published online June 8, 2021. doi:10.1053/j.gastro.2021.06.005