The role of macroscopic on-site evaluation in the era of biopsy needles: is it still useful?

Macroscopic on-site evaluation (MOSE) involves visual assessment of samples obtained during endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA). In brief, the presence or length of the macroscopically visible core is investigated to decide whether an appropriate sample for pathological evaluation has been obtained [1] [2]. The approach is similar to that of rapid on-site evaluation (ROSE), which differs from MOSE in terms of cytological evaluation. ROSE overcomes the disadvantage of EUS-FNA, in which more than three or four needle passes are required for sufficient diagnostic accuracy, by reducing the number of needle passes [3]. However, because of limited pathology staffing, ROSE has not been implemented in many hospitals, especially in Europe and Asia. MOSE may therefore be an alternative in such situations [1] [2].

A prospective study of 19G standard needles revealed that a macroscopically visible core length of ≥ 4 mm can be an indicator of specimen adequacy [2]. Furthermore, a randomized controlled trial (RCT) using this criterion showed that EUS-FNA with MOSE provided a similar diagnostic yield to EUS-FNA without MOSE with fewer needle passes (2 vs. 3; P < 0.001) [4]. This criterion is however not applicable to different types and sizes of needles. In recent years, biopsy needles such as Franseen-tipped and fork-tipped needles have become commercially available, innovations that have offered a greater quantity of obtained sample. The procedure using these needles is called EUS-guided fine-needle biopsy (EUS-FNB). A retrospective study using 22G Franseen needles showed that a macroscopically visible core length of ≥ 10 mm can be an indicator of diagnostic accuracy [5]. However, the clinical utility of MOSE, such as reduction in the number of needle passes, procedure time, and enhancement of safety, when using these biopsy needles is unknown.

“This is the first RCT to confirm the utility of MOSE when using a Franseen needle to reduce the number of needle passes while maintaining diagnostic accuracy.”

Mangiavillano et al. conducted a noninferiority multicenter RCT comparing EUS-FNB with and without MOSE, with the results being presented in this issue of Endoscopy [6]. In this well-designed study, 370 EUS-FNB procedures using 22G Franseen needles for pancreatic or nonpancreatic masses were evaluated, and diagnostic accuracy was compared as a primary end point. This study aimed to demonstrate the noninferiority of EUS-FNB with MOSE, based on a noninferiority limit of 6 %. The number of needle passes in the MOSE group was determined by the achievement of a total macroscopically visible core length ≥ 10 mm, whereas three needle passes were performed in the control group. No significant difference was found between the two groups for diagnostic accuracy (MOSE group 90.0 % vs. control group 87.8 %; P = 0.49), and the difference in accuracy between the two groups was 2.2 % (−2.4 % to 3.8 %), showing noninferiority for the primary outcome. In addition, the median number of needle passes was significantly lower in the MOSE group (1 vs. 3; P < 0.001). This is the first RCT to confirm the utility of MOSE when using a Franseen needle to reduce the number of needle passes while maintaining diagnostic accuracy.

When planning an RCT evaluating MOSE, setting the number of needle passes in the control group is a critical issue; however, this is controversial as few studies have evaluated the number of needle passes required for sufficient diagnostic accuracy when using such biopsy needles. The authors referred to the European Society of Gastrointestinal Endoscopy (ESGE) guideline, which suggests two to three needle passes with the FNB needle when ROSE and MOSE are unavailable [3]. However, these guidelines are based on studies that used reverse-bevel needles. Recently, an RCT showed that the use of a 22G Franseen needle is superior to that of a 20G bevel needle in both the quantity of tissue obtained and diagnostic accuracy [7]. Furthermore, a prospective study evaluating a 22G Franseen needle showed that the sensitivity reached 93.3 % with one needle pass and exceeded 95 % with two needle passes; other prospective studies have also revealed that an accuracy rate > 95 % was achieved with two needle passes, demonstrating that two passes are sufficient when using a 22G Franseen needle in clinical practice [8] [9].

The Achilles’ heel of MOSE is false positives as the macroscopically visible core can contain normal tissue, such as fibrous tissue and pancreatic parenchyma, and discrimination between this and tumor tissue is impossible. The aim of measuring the macroscopically visible core length is to reduce the possibility of false positives and, although it is useful, it does not guarantee an accuracy of 100 %. In fact, false positives occurred in 6.9 % (13/190) of the MOSE group in this study. In nine of these 13 patients, a correct diagnosis was obtained by additional passes after the macroscopically visible core length had reached 10 mm, and the accuracy rate increased from 90.0 % to 94.7 %. In other words, these nine patients would have been disadvantaged without additional passes. Considering this, it may not be realistic for the procedure to be terminated after one needle pass based on the MOSE result. When performing EUS-FNB with 22G Franseen needles, rather than using MOSE to reduce the number of needle passes, it may be better to always perform two needle passes without MOSE to save on the labor required to perform MOSE. Therefore, an RCT with two needle passes in the control group would be more meaningful for clinical practice.

Comprehensive genome profiling (CGN) using next-generation sequencing has been used clinically for solid tumors, and genotype-matched therapies provide improved prognoses. A larger sample size is required for this test compared with that required for pathological diagnosis. Therefore, more needle passes are probably needed in EUS-FNB, which might lead to an increased rate of adverse events or become a burden on endoscopists. A prospective study showed that measurement of the macroscopically visible core can predict the quantity of sample obtained when using a 22G Franseen needle, indicating the possibility of reducing the number of needle passes by using MOSE when samples are to be submitted to CGN [9]. The sample quantity cannot be evaluated by ROSE, and the prediction of tissue quantity plays a role only in MOSE. As a result, the role of MOSE may have changed in the era of new biopsy needles.

Publication History

Article published online:
10 October 2022

• References

• 1 Ishiwatari H, Sato J, Fujie S. et al. Gross visual inspection by endosonographers during endoscopic ultrasound-guided fine needle aspiration. Pancreatology 2019; 19: 191-195
• 2 Iwashita T, Yasuda I, Mukai T. et al. Macroscopic on-site quality evaluation of biopsy specimens to improve the diagnostic accuracy during EUS-guided FNA using a 19-gauge needle for solid lesions: a single-center prospective pilot study (MOSE study). Gastrointest Endosc 2015; 81: 177-185
• 3 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 – March 2017. Endoscopy 2017; 49: 989-1006
• 4 Chong CCN, Lakhtakia S, Nguyen N. et al. Endoscopic ultrasound-guided tissue acquisition with or without macroscopic on-site evaluation: randomized controlled trial. Endoscopy 2020; 52: 856-863
• 5 Kaneko J, Ishiwatari H, Sasaki K. et al. Macroscopic on-site evaluation of biopsy specimens for accurate pathological diagnosis during EUS-guided fine needle biopsy using 22-G Franseen needle. Endosc Ultrasound 2020; 9: 385-391
• 6 Mangiavillano B, Crino SF, Facciorusso A. et al. Endoscopic ultrasound-guided fine-needle biopsy with or without macroscopic on-site evaluation: a randomized controlled noninferiority trial. Endoscopy 2023; 55: 129-137
• 7 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
• 8 Mita N, Iwashita T, Uemura S. et al. Endoscopic ultrasound-guided fine needle biopsy using 22-gauge Franseen needle for the histological diagnosis of solid lesions: a multicenter prospective pilot study. Dig Dis Sci 2020; 65: 1155-1163
• 9 Kaneko J, Ishiwatari H, Sasaki K. et al. Macroscopic visible core length can predict the histological sample quantity in endoscopic ultrasound-guided tissue acquisition: multicenter prospective study. Dig Endosc 2022; 34: 622-631