The potential application of the random forest method to explore the association between microRNAs and tissue and molecular genetic biomarkers of breast cancer

Aim. Exploring the potential of using a random forest method (RFM) to identify associations between serum levels of microRNAs miR-181a and miR-25 and the expression of the following markers: E-cadherin (CDH1), type II collagen (CII), integrin beta-1 (ITGB1), cyclin D1 (CCND1), and T-lymphocyte costimulator (CD86), as well as their relationship with clinically relevant molecular genetic markers (ER, PR, HER2, Ki-67) in patients with breast cancer (BC).
Materials and methods. We recruited 44 BC patients with invasive ductal carcinoma who had not received neoadjuvant treatment. Serum levels of miR-181a and miR-25 were measured using digital droplet PCR, and the expression of CDH1, CII, ITGB1, CCND1, CD86, ER, PR, HER2, and Ki-67 in tumor tissue samples was assessed via immunohistochemistry. Spearman’s correlation analysis and the RFM were applied to assess associations between variables.
Results. Spearman’s correlation analysis did not reveal any significant linear relationships between the studied parameters (p > 0.05). However, the RFM identified nonlinear associations: for miR-181a, the optimal predictive model included CDH1 and CCND1 (MAE = 4.267); for miR-25, ITGB1 and CCND1 (MAE = 16.255). Among clinical and pathological characteristics, statistically significant positive correlations were found between CII and ER (r = 0.498; p = 0.001) and PR (r = 0.354; p = 0.018). The RFM models demonstrated the highest accuracy in predicting the HER2 status (82.6%) when using a combination of ITGB1, miR-181a, and CDH1.
Conclusion. The RFM revealed complex nonlinear associations between microRNAs and tissue markers of tumor progression that were not detected by conventional statistical methods. These findings are exploratory in nature and provide a foundation for further research aimed at validating these biomarkers and elucidating their role in the pathogenesis of various breast cancer subtypes.

1. Rani V., Sengar R.S. Biogenesis and mechanisms of micro RNA-mediated gene regulation. Biotechnol. Bioeng. 2022;119 (3):685–692. DOI: 10.1002/bit.28029.

2. Kumar S., Ranga A. Role of miRNAs in breast cancer development and progression: Current research. BioFactors. 2025;51(1):e2146. DOI: 10.1002/biof.2146.

3. Zhai Z., Mu T., Zhao L., Li Y., Zhu D., Pan Y. MiR-181a-5p facilitates proliferation, invasion, and glycolysis of breast cancer through NDRG2-mediated activation of PTEN/AKT pathway. Bioengineered. 2022;13(1):83–95. DOI: 10.1080/21655979.2021.2006974.

4. Zhao T., Meng W., Chin Y., Gao L., Yang X., Sun S. et al. Identification of miR-25-3p as a tumor biomarker: Regulation of cellular functions via TOB1 in breast cancer. Mol. Med. Rep. 2021; 23(6):406. DOI: 10.3892/mmr.2021.12045.

5. Kalinina T.S., Kononchuk V.V., Alekseenok E.Y., Obukhova D., Sidorov S.V., Strunkin D. et al. Expression of estrogen receptor- and progesterone receptor-regulating micro RNAs in breast cancer. Genes. 2021;12(4):582. DOI: 10.3390/genes12040582.

6. Andreeva O.E., Sorokin D.V., Mikhaevich E.I., Bure I.V., Shchegolev Y.Y., Nemtsova M.V. et al. Towards unravelling the role of ERα-targeting miRNAs in the exosome-mediated transferring of the hormone resistance. Molecules. 2021;26(21):6661. DOI: 10.3390/molecules26216661.

7. Gilam A., Shai A., Ashkenazi I., Sarid L.A., Drobot A., Bickel A. et al. MicroRNA regulation of progesterone receptor in breast cancer. Oncotarget. 2017;8(16):25963–25976. DOI: 10.18632/oncotarget.15657.

8. Feng J., Hu S., Liu K., Sun G., Zhang Y. The role of microRNA in the regulation of tumor epithelial-mesenchymal transition. Cells. 2022;11(13):1981. DOI: 10.3390/cells11131981.

9. Mendez M.J., Hoffman M.J., Cherry E.M., Lemmon C.A., Weinberg S.H. Cell fate forecasting: a data-assimilation approach to predict epithelial-mesenchymal transition. Biophys. J. 2020;118(7):1749–1768. DOI: 10.1016/j.bpj.2020.02.011.

10. Manfioletti G., Fedele M. Epithelial-mesenchymal transition (EMT). Int. J. Mol. Sci. 2023;24(14):11386. DOI: 10.3390/ijms241411386.

11. Ali A.H., Haider G., Ahmed K., Hashmi M., Kerio P., Irshad R. Correlation between e-cadherin and hormone receptor status among breast cancer patients. Journal of the College of Physicians and Surgeons Pakistan. 2020;30(10):1030–1034. DOI: 10.29271/jcpsp.2020.10.1030.

12. Autenshlyus A.I., Perepechaeva M.L., Studenikina A.A., Grishanova A.Y., Lyakhovich V.V. Serum miR-181а and miR-25 levels in patients with breast cancer or benign breast disease. Dokl. Biochem. Biophys. 2023;512(1):279–283. DOI: 10.1134/S1607672923700436.

13. Студеникина А.А., Архипов С.А., Вараксин Н.А., Проскура А.В., Аутеншлюс А.И. Биомаркеры молочной железы в выявлении малигнизации при доброкачественных заболеваниях у женщин. Профилактическая медицина. 2024;27(1):63–66. DOI: 10.17116/profmed20242701163.

14. Студеникина А.А., Архипов С.А., Жураковский И.П., Проскура А.В., Аутеншлюс А.И. Экспрессия маркеров эпителиально-мезенхимального перехода при раке молочной железы. Архив патологии. 2023;85(4):12–17. DOI: 10.17116/patol20238504112.

15. Jeffreys S.A., Becker T.M., Khan S., Soon P., Neubauer H., de Souza P. et al. Prognostic and predictive value of CCND1/Cyclin D1 Amplification in Breast Cancer with a Focus on Postmenopausal Patients: A Systematic Review and Meta-Analysis. Frontiers in Endocrinology. 2022;13:895729. DOI: 10.3389/fendo.2022.895729.

16. Wennhold K., Thelen M., Lehmann J., Schran S., Preugszat E., Garcia-Marquez M. et al. CD86+ antigen-presenting B cells are increased in cancer, localize in tertiary lymphoid structures, and induce specific T-cell responses. Cancer Immunol. Res. 2021;9(9):1098–1108. DOI: 10.1158/2326-6066.CIR-20-0949.

17. Yang H.L., Tsai Y.C., Korivi M., Chang C.T., Hseu Y.C. Lucidone Promotes the cutaneous wound healing process via activation of the PI3K/AKT, Wnt/β-catenin and NF-κB signaling pathways. Biochimica et Biophysica Acta – Molecular Cell Research. 2017;1864(1):151–168. DOI: 10.1016/j.bbamcr.2016.10.021.

18. Li X., Jin Y., Xue J. Unveiling collagen’s role in breast cancer: insights into expression patterns, functions and clinical implications. International Journal of General Medicine. 2024;17:1773–1787. DOI: 10.2147/IJGM.S463649.

19. Khan M.J., Singh A.K., Sultana R., Singh P.P., Khan A., Saxena S. Breast cancer: A comparative review for breast cancer detection using machine learning techniques. Cell Biochem. Funct. 2023;41(8):996–1007. DOI: 10.1002/cbf.3868.

20. Breiman L. Random forests. Mach. Learn. 2001;45:5–32. DOI: 10.1023/A:1010933404324.

21. Zuo D., Yang L., Jin Y., Qi H., Liu Y., Ren L. Machine learning-based models for the prediction of breast cancer recurrence risk. BMC Med. Inform. Decis. Mak. 2023;23(1):276. DOI: 10.1186/s12911-023-02377-z.

22. Jin Y., Lan A., Dai Y., Jiang L., Liu S. Development and testing of a random forest-based machine learning model for predicting events among breast cancer patients with a poor response to neoadjuvant chemotherapy. Eur. J. Med. Res. 2023;28(1):394. DOI: 10.1186/s40001-023-01361-7.

23. Arika R.N., Mindila A., Cheruiyo W. Machine learning algorithms for breast cancer diagnosis: challenges, prospects and future research directions. J. Oncol. 2023;5(1):1–31. DOI: 10.30564/jor.v5i1.4977.

24. Kondov B., Milenkovikj Z., Kondov G., Petrushevska G., Basheska N., Bogdanovska-Todorovska M. et al. Presentation of the molecular subtypes of breast cancer detected by immunohistochemistry in surgically treated patients. J. Med. Sci. 2018;6(6): 961–967. DOI: 10.3889/oamjms.2018.231.

25. Yumpu [Internet]. ER/PR PharmDx Interpretation Manual. – Dako. URL: https://www.yumpu.com/en/document/read/18500132/er-pr-pharmdxtm-interpretation-manual-dako. Available at: 09.20.2013.

26. Willmott C.J., Matsuura K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Clim. Res. 2005;30(1):79–82. DOI: 10.3354/cr030079.

27. Li L., Xu Q.H., Dong Y.H., Li G.X., Yang L., Wang L.W. et al. MiR-181a upregulation is associated with epithelial-to-mesenchymal transition (EMT) and multidrug resistance (MDR) of ovarian cancer cells. Eur. Rev. Med. Pharmacol. Sci. 2016;20(10):2004–2010.

28. Strotbek M., Schmid S., Sánchez-González I., Boerries M., Busch H., Olayioye M.A. miR-181 elevates Akt signaling by co-targeting PHLPP2 and INPP4B phosphatases in luminal breast cancer. Int. J. Cancer. 2017;140(10):2310–2320. DOI: 10.1002/ijc.30661.

29. Govatati S., Singamsetty G.K., Nallabelli N., Malempati S., Rao P.S., Madamchetty V.K.K. et al. Contribution of cyclin D1 (CCND1) and E-cadherin (CDH1) alterations to colorectal cancer susceptibility: a case-control study. Tumor Biol. 2014;35(12):12059–12067. DOI: 10.1007/S13277-014-2505-9.

30. Yan X., Gao M., Zhang P., Ouyang G., Mu Q., Xu K. MiR-181a functions as an oncogene by regulating CCND1 in multiple myeloma. Oncol. Lett. 2020;20(1):758–764. DOI: 10.3892/OL.2020.11579.

31. Yao J., Li G., Liu M., Yang S., Su H., Ye C. lnc-MICAL2 1 sponges miR 25 to regulate DKK3 expression and inhibits activation of the Wnt/β catenin signaling pathway in breast cancer. Int. J. Mol. Med. 2022;49(2):23. DOI: 10.3892/ijmm.2021.5078.

32. Jiang X., Wang J., Zhang K., Tang S., Ren C., Chen Y. The role of ITGB1-ILK-Akt signaling-mediated epithelial-mesenchymal transition of liver epithelial cells and chemoresistance and radioresistance in hepatocellular carcinoma cells. Med. Oncol. 2015;32(5):141. DOI: 10.1007/S12032-015-0595-X.

33. Vora H., Gavali M., Patel N. A Study of Col2A1 expression in triple-negative brest cancer patients. Glob. J. Clinic. Medici. Medica. Res. 2023;1:1–5.

34. Li X., Jin Y., Xue J. Unveiling collagen’s role in breast cancer: insights into expression patterns, functions and clinical implications. Int. J. Gen. Med. 2024;17:1773–1787. DOI: 10.2147/IJGM.S463649.