With or without ROSE: does fine-needle biopsy settle the debate on rapid on-site evaluation?

Rapid onsite evaluation (ROSE) involves an assessment of samples obtained by fine-needle aspiration (FNA) for adequacy and preliminary diagnosis that is performed by a cytopathologist or cytotechnician attending the procedure. Originally used to optimize the yield of percutaneous FNA, ROSE was quickly adopted for endoscopic ultrasound (EUS)-guided FNA soon after the technique became available. A debate has ensued on whether ROSE improves the results of EUS-FNA and, if so, whether its advantages are worth the effort and cost of organizing the service. More than two decades later the debate continues.

“Altogether, the study provides a solid evidence base that FNB is no less accurate than FNA with ROSE, a technique considered by many to be the long-standing gold standard for diagnostic accuracy in EUS-guided sampling of pancreatic masses.”

There is no doubt that ROSE allows a reduction in the number of needle passes required to obtain a diagnostic sample. This has been consistently shown in observational studies, randomized controlled trials (RCTs), and meta-analyses [1] [2] [3] [4]; however, the clinical significance of this finding is uncertain. Theoretically, fewer needle passes mean less trauma to the tissue and better safety. In fact, an RCT found that patients with lymphadenopathy who had five passes performed without ROSE more often reported self-limited post-procedural pain compared with those who had a lower number of passes as a result of ROSE [1]. This observation aside, however, there is no proof that ROSE influences the safety of EUS-FNA [2] [3]. There is also no proof that it saves time or money. RCTs comparing FNA with ROSE to FNA with a fixed number of five or seven passes have failed to demonstrate significant differences in procedural time and cost [1] [2] [3]. RCTs have also not confirmed the other advantages of ROSE that were demonstrated in observational studies and some (but not all) meta-analyses, such as superior diagnostic yield, sensitivity, and accuracy for malignancy [1] [2] [3] [4].

While the evidence in favor of ROSE, especially from RCTs, is not conclusive, studies have never suggested that the results achieved with ROSE are any worse than without it. And – a tribute to its value – ROSE is advocated by experts and used in many centers worldwide. In a 2016 international survey, 65 % of 186 endosonographers, working mostly in an academic setting, declared that they had access to ROSE [5]. In addition, 53 % of expert endosonographers surveyed in 2020 believed ROSE increases EUS-FNA sensitivity of cancer detection and should be used; 38 % held the opposite view [6].

The landscape of EUS-guided sampling has changed considerably with the advent a few years ago of fine-needle biopsy (FNB). When compared with FNA, FNB provides samples that are larger, of higher histological quality, more often suitable for immunostaining studies, and more often diagnostic [7]. There is a body of evidence that suggests fewer needle passes are required to obtain a diagnostic sample with FNB needles. FNB samples can be assessed for adequacy by macroscopic evaluation and ROSE does not improve the results [8]. Last, but not least, FNB offers better diagnostic sensitivity and accuracy for malignancy, when compared with FNA without ROSE [9] [10] [11].

On this background, in this issue of Endoscopy, Chen et al. present the results of a non-inferiority RCT comparing FNB to FNA with ROSE [12]. In this well designed and well conducted study, a total of 235 individuals with solid pancreatic masses, recruited at seven centers in Canada, were assigned to FNA with the number of needle passes guided by ROSE or to FNB with two passes of a fork-tip needle (a third pass was allowed when the endosonographer felt, based on macroscopic evaluation, that the specimen from the two standard passes was insufficient). ROSE was performed by a cytotechnician.

The primary analysis concerned the accuracy for diagnosis of malignancy, which is considered the most important outcome in studies evaluating diagnostic tests. With an accuracy of 92 % (95 % confidence interval [CI] 87 %–97 %), FNB was found to be non-inferior to FNA with ROSE, which had an accuracy of 93 % (95 %CI 89 %–98 %). The sensitivity of both techniques exceeded 92 % and specificity was 100 %.

While both techniques had excellent diagnostic performance, FNB offered significant advantages over FNA with ROSE in respect to the number of needle passes (mean 2.3 vs. 3.0), procedural time (mean 19.3 vs. 22.7 minutes), and histological quality of the specimen (histologically interpretable specimen in 88 % vs. 39 % of cases). On the other hand, a cost-minimization analysis showed that FNB was more expensive, in both the US and Canadian settings, by a margin of 6 % and 19 %, respectively (45 US dollars or 102 Canadian dollars). Two adverse events were observed (both in the FNA with ROSE group), but obviously, as with other RCTs in the field, the study was not powered to evaluate safety.

The study allowed the use of needles other than 22-gauge. As a result, thinner, 25-gauge needles were used in the majority of FNA procedures (56 %), and significantly more often than in FNB procedures (4 %). This imbalance might have biased the assessment of the histological quality of the specimen in favor of FNB (thinner needles are less likely to provide samples of high histological quality); however, an impact on diagnostic accuracy, the primary outcome of the study, was unlikely [4]. In fact, the diagnostic performance of FNA with ROSE in this study was excellent, therefore setting the comparison bar for FNB very high.

Altogether, the study provides a solid evidence base, which has been missing until now, that FNB is no less accurate than FNA with ROSE, a technique considered by many to be the long-standing gold standard for diagnostic accuracy in EUS-guided sampling. Fewer needle passes, shorter procedural time, and a higher quantity and quality of the specimen are additional benefits of FNB. And because FNB is much easier to implement in practice than FNA with ROSE – and simple wins over complicated – FNB has the full potential to be the method of choice, not only in centers that have no established ROSE service, but also in those that have been successfully using ROSE until now. There is one caveat, however: evidence is accumulating that not all FNB needles are equal. The 22-gauge fork-tip needle, evaluated in this study, and 22-gauge Franseen-tip needle have been shown in RCTs to offer similar performance and to outperform the reverse bevel and forward-facing bevel needles [11] [13] [14] [15]. This should be considered when choosing the FNB needle to use.

Publication History

Article published online:
10 August 2021

• References

• 1 Kappelle WFW, van Leerdam ME, Schwartz MP. et al. Rapid on-site evaluation during endoscopic ultrasound-guided fine-needle aspiration of lymph nodes does not increase diagnostic yield: A randomized, multicenter trial. Am J Gastroenterol 2018; 113: 677-685
• 2 Wani S, Mullady D, Early DS. et al. The clinical impact of immediate on-site cytopathology evaluation during endoscopic ultrasound-guided fine needle aspiration of pancreatic masses: a prospective multicenter randomized controlled trial. Am J Gastroenterol 2015; 110: 1429-1439
• 3 Lee LS, Nieto J, Watson RR. et al. Randomized noninferiority trial comparing diagnostic yield of cytopathologist-guided versus 7 passes for EUS-FNA of pancreatic masses. Dig Endosc 2016; 28: 469-475
• 4 Polkowski M, Jenssen C, Kaye P. et al. Technical aspects of endoscopic ultrasound (EUS)-guided sampling in gastroenterology: European Society of Gastrointestinal Endoscopy (ESGE) Technical Guideline. Endoscopy 2017; 49: 989-1006
• 5 van Riet PA, Cahen DL, Poley JW. et al. Mapping international practice patterns in EUS-guided tissue sampling: outcome of a global survey. Endosc Int Open 2016; 4: E360-E370
• 6 Guo J, Sahai AV, Teoh A. et al. An international, multi-institution survey on performing EUS-FNA and fine needle biopsy. Endosc Ultrasound 2020; 9: 319-328
• 7 Bang JY, Hebert-Magee S, Navaneethan U. et al. EUS-guided fine needle biopsy of pancreatic masses can yield true histology. Gut 2018; 67: 2081-2084
• 8 Fabbri C, Fuccio L, Fornelli A. et al. The presence of rapid on-site evaluation did not increase the adequacy and diagnostic accuracy of endoscopic ultrasound-guided tissue acquisition of solid pancreatic lesions with core needle. Surg Endosc 2017; 31: 225-230
• 9 Oppong KW, Bekkali NLH, Leeds JS. et al. Fork-tip needle biopsy versus fine-needle aspiration in endoscopic ultrasound-guided sampling of solid pancreatic masses: a randomized crossover study. Endoscopy 2020; 52: 454-461
• 10 van Riet PA, Larghi A, Attili F. et al. A multicenter randomized trial comparing a 25-gauge EUS fine-needle aspiration device with a 20-gauge EUS fine-needle biopsy device. Gastrointest Endosc 2019; 89: 329-339
• 11 Young Bang J, Krall K, Jhala N. et al. Comparing needles and methods of endoscopic ultrasound-guided fine-needle biopsy to optimize specimen quality and diagnostic accuracy for patients with pancreatic masses in a randomized trial. Clin Gastroenterol Hepatol 2021; 19: 825-835
• 12 Chen YI, Chatterjee A, Berger R. et al. Endoscopic ultrasound (EUS)-guided fine needle biopsy alone vs. EUS-guided fine needle aspiration with rapid onsite evaluation of cytopathology in pancreatic lesions: a multicenter randomized trial. Endoscopy 2022; 54: 4-12
• 13 Ashat M, Klair JS, Rooney SL. et al. Randomized controlled trial comparing the Franseen needle with the Fork-tip needle for EUS-guided fine-needle biopsy. Gastrointest Endosc 2021; 93: 140-150
• 14 Karsenti D, Palazzo L, Perrot B. et al. 22G Acquire vs. 20G Procore needle for endoscopic ultrasound-guided biopsy of pancreatic masses: a randomized study comparing histologic sample quantity and diagnostic accuracy. Endoscopy 2020; 52: 747-753
• 15 Kurita A, Yasukawa S, Zen Y. et al. Comparison of a 22-gauge Franseen-tip needle with a 20-gauge forward-bevel needle for the diagnosis of type 1 autoimmune pancreatitis: a prospective, randomized, controlled, multicenter study (COMPAS study). Gastrointest Endosc 2020; 91: 373-381