臨床研究等提出・公開システム

Feasibility for consensus molecular subtype classification of colorectal cancer using preoperative biopsy tissues; A pilot study

COLLAB study

8

A total of eight patients were enrolled in the study. During preoperative colonoscopy, two biopsy specimens were obtained from different tumor regions using two types of biopsy forceps (standard and jumbo forceps), resulting in a total of 16 specimens for analysis. The median age of the patients was 59 years, and five of the eight patients were male. Tumor locations included two right-sided colon cancers, two left-sided colon cancers, and four rectal cancers. All tumors were histologically diagnosed as tubular adenocarcinoma. Colonoscopic procedures were completed safely, and no procedure-related adverse events were observed.

The planned enrollment period was three months from the start of the study. However, following the first patient enrollment on June 2, 2025, a total of eight patients were enrolled by July 2, 2025, and the protocol-specified examinations were completed in all cases.

No adverse events or study-related medical conditions occurred during the study period.

[Primary endpoint] The primary endpoint of this study was the proportion of biopsy specimens successfully classified into CMS based on gene expression analysis using preoperative biopsy samples. CMS classification was performed using the CMScaller algorithm. The success rate of CMS classification was calculated separately for each biopsy forceps (standard and jumbo), and the corresponding 95% confidence intervals were estimated using the Clopper-Pearson method. CMS classification was successfully achieved in all biopsy specimens obtained with both forceps types. Therefore, the success rate was 100% (95% confidence interval: 63-100%) for both the standard and jumbo forceps. [Secondary endpoints] Secondary endpoints included: 1. concordance of CMS classification between biopsy specimens obtained from different regions of the same tumor, and 2. the occurrence of colonoscopy-related adverse events. Discordance in CMS classification between biopsy specimens obtained from different regions of the same tumor was observed in three of the eight patients, suggesting the presence of intratumoral heterogeneity. No colonoscopy-related adverse events occurred during the study.

This prospective pilot study evaluated the feasibility of CMS classification using preoperative biopsy specimens in colorectal cancer. Eight patients with primary CRC underwent colonoscopic biopsies using two forceps. CMS classification using CMScaller was successful in all samples. However, discordant CMS results were observed in three patients, suggesting intratumoral heterogeneity.

Aug. 04, 2026

July. 25, 2026

https://www.jstage.jst.go.jp/article/jarc/10/3/10_2026-004/_article/-char/en

No

No plan to share IPD

https://jrct.mhlw.go.jp/latest-detail/jRCT1042250012

Shiomi Akio

Shizuoka Cancer Center

1007 Shimonagakubo, Nagaizumi-cho, Sunto-gun, Shizuoka

a.shiomi@scchr.jp

Kasai Shunsuke

Shizuoka Cancer Center

1007 Shimonagakubo, Nagaizumi-cho, Sunto-gun, Shizuoka

+81-55-989-5222

s.kasai@scchr.jp

Complete

April. 23, 2025

Interventional

single arm study

open(masking not used)

uncontrolled control

single assignment

treatment purpose

(1) Patients 18 years and older with PS (ECOG) of 0 or1.
(2) Patients with primary colorectal cancer deeper than cT2.

(1) Patients with a history of chemotherapy or radiotherapy, including treatment for other cancers.
(2) Patients taking antiplatelet or anticoagulant medications.

Both

Primary colorectal cancer

Two types of biopsy forceps of different sizes (EndoJaw (FB-220U, OLYMPUS) / EndoJaw Jumbo (FB-214U, OLYMPUS)) are used to collect one additional tissue at each location during preoperative colonoscopy.

Percentage of tissue collected using two types of biopsy forceps during preoperative colonoscopy that can be classified as CMS

(1) Comparison of CMS classification by tissue collection site using two types of biopsy forceps
(2) Adverse events

+81-55-989-5222

Mar. 31, 2025

none