The Primary Result of Prospective Randomized Multicenter Trial of New Spray-Type Bio-absorbable Adhesion Barrier System (TCD-11091) Against Postoperative Adhesion Formation

This study was designed as a prospective, multicenter, randomized (in a 1:1 ratio) single-blind study using a non-treatment group as the control, and enrolled 124 patients.

TCD-11091 consists of a polymer kit and sprayer kit. To apply TCD-11091, powder A solution (NHS-modified carboxymethyl dextrin) and B solution (sodium carbonate/sodium hydrogen carbonate) are prepared with the polymer kit and loaded into a double syringe, which is attached to the nozzle. The two solutions are mixed uniformly with the assistance of compressed medical air at the nozzle tip and sprayed in the form of mist. The two sprayed solutions gelate rapidly over the living tissues. The gel adheres to the injury site and forms a physical barrier to provide the adhesion prevention effect. After serving its purpose, the gel gradually degrades, is absorbed, and is then excreted, mainly into urine. The sprayer of this system was designed to enable the abdominal cavity to be accessed through a 5-mm trocar in line with its use during laparoscopy. Consequently, the formed gel contains microbubbles and appears white. Therefore, the surgeon can visually confirm the extent of gel coverage through the laparoscope.

This study was approved by the institutional review board (IRB) at each participated institutes. All patients signed the informed consent form. Subjects were patients with primary rectal cancer who were scheduled to undergo temporary ileostomy. Rectal resection and ileostomy (the initial surgery) through midline laparotomy and ileostomy closure (the second-look surgery) were performed. Patients were allocated at the time of enrollment in the study after informed consent was obtained. Randomized allocation was performed according to the minimization method by the dynamic allocation and using clinical site as allocation factor. Then, the initial surgery was performed according to the allocation. At the second-look surgery, laparoscopic video recordings of the area under the midline incision were made in both groups. For the efficacy evaluation, based on the laparoscopic images of the abdominal cavity at the time of the second-look surgery, the presence, extent, and severity of adhesions were assessed by the efficacy assessment committee which was independent of the study site and the sponsor. For the initial surgery, the operation was performed under general anesthesia according to the standard practice of each site. At the end of the surgery, TCD-11091 was sprayed to fully cover the organs under the midline incision, and pictures were taken from directly above this incision. After pictures were taken, the peritoneum was sutured with an absorbable synthetic monofilament suture to close the abdomen. Taking pictures of adhesions through the laparoscope at the second-look surgery was an essential part of the efficacy evaluation in this study, and variations in resolution were minimized by employing a photographic method that was standardized across all sites. Movies were produced laparoscopically so that the upper edge (cranial end) and the lower edge (caudal end) of the midline incision were included in the images and the location of the midline incision could be confirmed. The independent efficacy assessment committee was established for determination of adhesions and assessed the presence, extent, and severity of adhesions. The extent and severity of adhesions under the midline incision were assessed on a 4-grade scale (Tables 1 and 2).

Major inclusion criteria were the following: 20 years and older, scheduled to undergo the initial surgery and to undergo the second-look surgery 3 to 6 months after the initial surgery, able to undergo the first- and second-look surgeries under general anesthesia, and indicated for laparotomy. Major exclusion criteria were active cancer other than rectal cancer, history of radiotherapy in the abdominal cavity or pelvic cavity, peritonitis, severe diabetes or abnormal glucose metabolism, glycogenosis, serious liver disorder, and serious renal disorder. All inclusion and exclusion criteria are summarized in Table 3.

The primary endpoint of this study was the incidence of adhesions under the midline incision and was defined as “the number of patients who are determined to have adhesions/the number of patients who underwent the second-look surgery.” The secondary endpoints included the following: (1) the extent of adhesions under the midline incision, (2) the severity of adhesions under the midline incision, (3) major adverse events (MAEs: intestinal obstruction, abdominal abscess, peritonitis, postprocedural site wound infection), (4) serious adverse events, (5) adverse events, (6) adverse events which occur from the initiation of the initial surgery to the day of discharge after the second-look surgery and for which a causal relationship to TCD-11091 cannot be ruled out, and (7) malfunction of the study device.

This study was conducted by Terumo Corporation (Tokyo, Japan), and the study management and monitoring and source document verification of all data, statistical analyses, and data management were conducted by A2 Healthcare Corporation (Tokyo, Japan), the CRO. An independent efficacy assessment committee was established for determination of adhesion status and assessed the presence, extent, and severity of adhesions. Safety data and important efficacy endpoints were evaluated at appropriate intervals. The data safety monitoring board (DSMB) was established for the purpose of recommending continuation or discontinuation of the study or changes in the study to the sponsor.

To calculate the sample size to be enrolled in this study, the incidence of adhesions was estimated to be 80% or higher in the control group based on the results of the preceding five studies (86.2 to 100%).3‐8 The incidence of adhesions was estimated to be reduced to 50% when TCD-11091 was used based on the report of the study conducted by Becker et al., which had the largest sample size among the clinical studies of Seprafilm.8 Therefore, the incidence of adhesions with the study device was set at 50%. Based on the above, the incidences of adhesions with and without the study device were assumed to be 50 and 80%, respectively. The number of patients required for a group was calculated to be 52 patients with a power of 90%. Consequently, the planned sample size per group was set at 62, and the total sample size required for this study was set at 124 based on the assumption that 15% of the patients might drop out. In the statistical analyses, summary statistics including mean, standard deviation, minimum, median, and maximum were calculated. Regarding patient background, comparisons were conducted between groups: for continuous values, the t test was used to compare differences in the mean values; for ordered categorical variables, the Wilcoxon rank sum test was used to compare differences in the median values; and for nominal variables, the chi-squared test was used to assess differences in the frequencies. The incidences of adhesions, adverse events, and MAEs were tested using Fisher’s exact test, and the extent and severity of adhesions were tested by between-group comparisons using the Wilcoxon rank sum test. The two-sided significance level was set at 5%. Study data were collected as EDC using DATATRAK ONE (version 13.2.5.; DATATRAK International, OH, USA) and analyzed using SAS (version 9.2; SAS Institute, Cary, NC, USA). The incidence of adhesions, the primary endpoint, was compared using Fisher’s exact test, and the extent and severity of adhesions were compared using the Wilcoxon rank sum test.

The incidences of adhesions under the midline incision, the primary endpoint, were 52.7% in the TCD-11091 group (29/55) and 90.7% in the control group (39/43); the incidence of adhesions in the TCD-11091 group was significantly reduced compared with that in the control group (p < 0.001) (Fig. 2). For the secondary endpoints, adhesions of wide extent (grade 2 or higher) occurred 38.2% (21/55) and 79.1% (34/43); the incidence of wide extent adhesions was significantly reduced (p < 0.001) (Fig. 3). In terms of adhesion severity, grade 2 or higher adhesions were noted 47.3% (26/55) and 88.4% (38/43); the severity of adhesions was significantly reduced by TCD-11091 (p < 0.001) (Fig. 4).

The incidences of adverse events were 85.2% in the TCD-11091 group (52/61) and 75.4% in the control group (46/61); no significant difference was noted (p = 0.225). Adverse events that occurred at a relatively high frequency were “pyrexia,” “peripheral neuropathy,” “intestinal obstruction,” “nausea,” and “palmar-plantar erythrodysaesthesia syndrome.” The incidences of MAEs were 21.3% (13/61) and 14.8% (9/61); no significant difference was noted (p = 0.360). The frequently reported MAE was functional intestinal obstruction: 11.5% (7/61) and 9.8% (6/61); no significant difference was noted (p = 0.779). Serious adverse events were reported in 26.2% of the patients (16/61) and 23.0% (14/61); no significant difference was noted. There were no adverse events that resulted in death or were life-threatening. No cases of carcinogenesis or infections related to TCD-11091 were reported. The success rate for the use of the TCD-11091 was 93.0% (53/57). A total of four cases (five kits) were considered unsuccessful. However, these cases were not considered to meet the criteria for malfunction because all of them were caused by misuse of the device (Table 5).