2023 / 06

Next Paper
DOI: 10.24075/brsmu.2023.050

ORIGINAL RESEARCH

The possibility of evaluation mRNA expression profiling to predict progression of local stage colorectal cancer

Goncharov SV, Bozhenko VK, Zakharenko MV, Chaptykov AA, Kulinich TM, Solodkiy VA
About authors

Russian Scientific Center for Roentgenoradiology, Moscow, Russia

Correspondence should be addressed: Sergey V. Goncharov
Profsoyuznaya, 86, 117997, Moscow, Russia; ur.liam@5109

About paper

Funding: the study was supported through the grant by RSF 22-15-00448.

Author contribution: Goncharov SV, Bozhenko VK, Solodkiy VA — study concept and design; Bozhenko VK, Goncharov SV, Kulinich TM, Zakharenko MV, Solodkiy VA — data acquisition and processing; Goncharov SV, Chaptykov AA, Bozhenko VK, Kulinich TM — manuscript writing; Goncharov SV, Chaptykov AA, Bozhenko VK, Kulinich TM, Zakharenko MV, Solodkiy VA — editing.

Received: 2023-09-13 Accepted: 2023-12-01 Published online: 2023-12-23
|
Fig. 1. Prediction accuracy when using conventional clinical and morphological risk factors of CRC progression
Fig. 2. Prediction accuracy of the logit model using mRNA expression profiles of eight genes
Fig. 3. Individual prognosis distribution in the model using clinical and morphological risk factors of progression
Table 1. Genes included in the analysis
Table 2. Logit model using conventional criteria to predict the risk of CRC progression
Note: Grade (G) is “0” when grade is G1 or G2 and “1” when grade is G3; the variable is “0” when histological assessment reveals no signs of angiolymphatic and/or lymphovascular invasion and “1” when there is at least one such sign; the ratio of the number of metastatic lymph nodes to the total number of resected lymph nodes in the surgical specimen is 0–1; Age is the patient’s age in whole years; Localization is tumor localization, which takes the value of “0” in cases of right-sided tumors (cecum, ascending colon, transverse colon) and “1” in cases of left-sided tumors (descending colon, sigmoid, rectum); 0.332 is a constant coefficient (constant) used in the model.
Table 3. Classification matrix of the logit model based on conventional clinical and morphological risk factors of CRC progression
Table 4. Logit model using mRNA expression profiles
Note: Grade (G) is “0” when grade is G1 or G2 and “1” when grade is G3; the variable is “0” when histological assessment reveals no signs of angiolymphatic and/or lymphovascular invasion and “1” when there is at least one such sign; the ratio of the number of metastatic lymph nodes to the total number of resected lymph nodes in the surgical specimen is 0–1; Age is the patient’s age in whole years; CCNB1, Ki67, GRB7, IGF1, Il2, Il6, Il8, GATA3 are expression levels of appropriate genes in the tumor; 11.05 is a constant coefficient used in the model.
Table 5. Classification matrix of the logit model using mRNA expression profiles of eight genes