CONTRAST ENHANCED HARMONIC EUS-GUIDED VERSUS CONVENTIONAL EUS-GUIDED FINE NEEDLE BIOPSY WITH MACROSCOPIC ON-SITE EVALUATION FOR SOLID PANCREATIC LESIONS: A MULTICENTER RANDOMIZED TRIAL

Background:
While endoscopic ultrasound (EUS) guided fine needle biopsy (FNB) is the preferred method of tissue acquisition for pancreatic solid lesions (PSL), false negative result may occur if FNB is performed in the avascular (necrotic) area. Contrast enhanced harmonic EUS (CEH EUS) allows better real-time delineation of lesion vascularity and may help to avoid avascular (necrotic) area during EUS-FNB. This multicenter randomized controlled trial aims to compare the diagnostic performance of CEH EUS-guided vs conventional EUS-guided FNB with macroscopic on-site evaluation (MOSE) for PSL.

Methods:
From 6/2021 to 8/2023, patients aged 18 to 80 years with PSL ≥1cm undergoing EUS-guided tissue acquisition in 3 referral centers in Hong Kong, Italy and Korea were randomized to undergo CEH EUS-guided FNB or conventional EUS-guided FNB. In the CEH EUS group, CEH EUS with SonoVue was used to identify non-enhancing, avascular (necrotic) area in target lesion and guide FNB to the viable area of the lesion. MOSE was used in both study arms to assess specimen adequacy. EUS-FNB was deemed complete if the obtained macroscopic visible core (MVC) was ≥4mm. If MVC was <4mm, FNB was repeated until a total MVC length ≥4mm was obtained. Primary outcome is the false negative rate of the assigned EUS-FNB technique. Secondary outcomes are sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), procedural time and procedure-related complications of the assigned technique.

Results:
128 patients were randomized (64 in CEH EUS group, 64 in conventional EUS group). The baseline patient and PSL characteristics were similar in both groups, except the rate of pancreatic adenocarcinoma (59.5% vs 79.7%, p=0.013) (Table 1). Avascular areas were identified on B mode EUS in 25.0% of PSL in CEH EUS group and 20.3% in conventional EUS group (p=0.526). There was a non-statistically significant increase in avascular area detection to 31.3% in CEH EUS group. The number of passes needed to achieve a MCV length ≥4mm (1 vs 1.5, p=0.480) and the mean MCV length (16.8mm vs 20.6mm, p=0.259) were similar in both groups. The false negative rates of CEH EUS-guided and conventional EUS-guided FNB were low and similar (6.0% vs 7.9%, p>0.999) (Table 2). The sensitivity (94.0% vs 92.1%, p>0.999), specificity (100% vs 100%, p>0.999) and diagnostic accuracy (95.3% vs 92.2%, p=0.718) were high and comparable in both groups. The procedure time was slightly longer in CEH EUS group (28.9min vs 24.4min, p=0.039). The procedure related adverse event rate was low in both groups (1.6% vs 1.6%, p>0.999).

Conclusion:
In this study, while CEH EUS could increase the detection of avascular (necrotic) area, the diagnostic performance of FNB with MOSE guided by CEH EUS or conventional EUS for PSL were comparable with low false negative rate and high diagnostic accuracy.