The MIN, newly designed for this study, is a 25-gauge inner needle embedded in a 19-gauge outer needle (Fig. 1A). The inner needle is advanced from a hole located 7.5 mm from the outer needle tip. MIN is capable of injecting the drug into the tumor in multiple directions through combined manipulation of the inner needle and rotation of the outer needle at a puncture site (Fig. 1B).

Animal experiments in the present study were approved by the Animal Research Committee of the Graduate School of Medicine, Kyoto University (MedKyo 12074). Male athymic nude rats (n=32; 4 weeks old) purchased from CLEA Japan, Inc. (Tokyo, Japan) were used in this study.

Human BxPC-3 pancreatic cancer cells (1×10⁷ cells) (DS Pharma Biomedical Co., Ltd., Osaka, Japan) resuspended in 150 μl of PBS were transplanted subcutaneously into the flank region of nude rats. After 6 weeks, tumors were measured with a caliper and the tumor volume (mm³) was calculated using the formula: length × width² × 0.5. Animals were randomized into 3 groups (n=10).

After piercing the tumor with the MIN, the inner needle was advanced and ethanol (99.5%) (Nacalai Tesque Inc., Kyoto, Japan) was slowly infused over a period of 30 sec with a KDS-100 syringe pump (KD Scientific Inc., New Hope, PA, USA). After retracting the inner needle, the needle shaft was rotated 90°, the inner needle was advanced again and an additional 50 μl of ethanol was injected. Injection was performed 4 times with a total of 200 μl of ethanol being injected into the tumor, as shown in Fig. 1B. The same volume (200 μl) of ethanol was injected using a single-end-hole 19-gauge needle (conventional needle; THERUMO Corp., Tokyo, Japan) and a 4-side-hole needle (4SHN; EUSN-20-CPN; Cook Medical Inc., Bloomington, IN, USA), under the same conditions.

The rats were sacrificed 7 days after ethanol injection. Tumors were then excised and fixed in formalin, sectioned perpendicular to the needle axis at 0.5-mm intervals and stained with hematoxylin and eosin (H&E). Images of the sections were captured using a VS120 scanning system (Olympus Corp., Tokyo, Japan) and boundaries of the injury area were determined within the image. The injury area was defined as the presence of necrosis, fibrosis and regeneration. Tumors and the injury areas were quantified with the ImageJ software.

The injury volume (V, mm³) was calculated using the sum of the injury area in each section [S(x)], the number of sections (N) and the distance between each section (0.5 mm): V = 1 2 ∑ i = 1 N S ( i 2 )

Two orthotopic model of human pancreatic tumor xenografts in rats were prepared using BxPC-3 cells, as described previously (8). Ten weeks after transplantation, ultrasound observations of ethanol injection into pancreatic tumors using the MIN were performed under anesthesia with an open surgical approach, in which the needle was inserted through the gastric wall under ultrasound guidance (GF-UCT2000-OL5, EU-C2000; Olympus Medical Systems Corp., Tokyo, Japan).

To evaluate ethanol distribution following injection with three types of needle, sequential sections of subcutaneous tumors were prepared with H&E staining. As shown in Fig. 2A, MIN resulted in a cross-shaped injury area. However, 4SHN produced an ellipsoidal distribution (Fig. 2B) and the conventional needle produced a circular distribution (Fig. 2C). The range of injury volume was wide in each group, although the mean injury volume with MIN was greater compared to 4SHN (Fig. 3A). Tumor volume analysis was therefore performed on the day of ethanol injection using the Ward method (JMP8 software; SAS Institute Inc., Cary, NC, USA), a mathematical classification based on the distance among the values (Fig. 3B and C). In the cluster group for large tumors (n=4) (Fig. 3C), the difference among the three needles was small. However, in the cluster group for small tumors (n=6) (Fig. 3B), the difference between the MIN and straight-type needles increased.

EUS was used to observe ethanol injection into pancreatic orthotopic tumors using the MIN. Insertion of the outer needle into the tumor, advancement of the inner needle from the outer needle and ethanol infusion were observed on ultrasound images and were visualized as a highly echoic image (Fig. 4).