Abstract
Purpose
We used five machine-learning algorithms to predict cancer-specific mortality after surgical resection of primary non-metastatic invasive breast cancer.
Methods
This study was a secondary analysis of data for 1661 women with primary non-metastatic invasive breast cancer. The overall patient population was divided into a training group and a test group at a ratio of 8:2 and python was used for machine learning to establish the prognosis model.
Results
The machine-learning Gbdt algorithm for cancer-specific death caused by various factors showed the five most important factors, ranked from high to low as follows: the number of regional lymph node metastases, LDH, triglyceride, plasma fibrinogen, and cholesterol. Among the five algorithm models in the test group, the highest accuracy rate was by DecisionTree (0.841), followed by the gbm algorithm (0.838). Among the five algorithms, the AUC values from high to low were GradientBoosting (0.755), gbm (0.755), Logistic (0.733), Forest (0.715), and DecisionTree (0.677).
Conclusion
Machine learning can predict cancer-specific mortality after surgery for patients with primary non-metastatic invasive breast.
Similar content being viewed by others
References
Peto R, Davies C, Godwin J, Gray R, Pan HC, Clarke M, et al. Comparisons between different polychemotherapy regimens for early breast cancer: meta-analyses of long-term outcome among 100,000 women in 123 randomised trials. Lancet (London, England). 2012;379:432–44.
Di LA, Jerusalem G, Petruzelka L, Torres R, Bondarenko IN, Khasanov R, et al. Final overall survival: fulvestrant 500 mg vs 250 mg in the randomized CONFIRM trial. J Natl Cancer Inst. 2014;106:djt337.
Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA. 2011;61:69–90.
Acs G, Paragh G, Chuang S-T, Laronga C, Zhang PJ. The presence of micropapillary features and retraction artifact in core needle biopsy material predicts lymph node metastasis in breast carcinoma. Am J Surg Pathol. 2009;33:202–10.
Shah TS, Kaag M, Raman JD, Chan W, Tran T, Kunchala S, et al. Clinical significance of prominent retraction clefts in invasive urothelial carcinoma. Hum Pathol. 2017;61:90–6.
Beam AL, Kohane IS. Big data and machine learning in health care. JAMA. 2018;319:1317–8.
Delahanty RJ, Kaufman D, Jones SS. Development and evaluation of an automated machine learning algorithm for in-hospital mortality risk adjustment among critical care patients. Crit Care Med. 2018;46:e481–8.
Obermeyer Z, Emanuel EJ. Predicting the future—big data, machine learning, and clinical medicine. N Engl J Med. 2016;375:1216–9.
Ravi D, Wong C, Deligianni F, Berthelot M, Andreu-Perez J, Lo B, et al. Deep learning for health informatics. IEEE J Biomed health Inf. 2017;21:4–21.
Sarkans U, Gostev M, Athar A, Behrangi E, Melnichuk O, Ali A, et al. The BioStudies database-one stop shop for all data supporting a life sciences study. Nucleic Acids Res. 2018;46:D1266–70.
Viale G. Breast cancer. Lancet. 2005;365:1727–41.
Li L, Chen L-Z. Factors influencing axillary lymph node metastasis in invasive breast cancer. Asian Pac J Cancer Prev. 2012;13:251–4.
Fisher B, Bauer M, Wickerham DL, Redmond CK, Fisher ER, Cruz AB, et al. Relation of number of positive axillary nodes to the prognosis of patients with primary breast cancer. NSABP Update Cancer. 1983;52:1551–7.
Fisher B, Jeong J-H, Anderson S, Bryant J, Fisher ER, Wolmark N. Twenty-five-year follow-up of a randomized trial comparing radical mastectomy, total mastectomy, and total mastectomy followed by irradiation. N Engl J Med. 2002;347:567–75.
Voordeckers M, Vinh-Hung V, Van DSJ, Lamote J, Storme G. The lymph node ratio as prognostic factor in node-positive breast cancer. Radiother Oncol J Eur Soc Ther Radiol Oncol. 2004;70:225–30.
Vinnicombe S, Pinto PSM, McCormack VA, Shiel S, Perry N, Dos SSIM. Full-field digital versus screen-film mammography: comparison within the UK breast screening program and systematic review of published data. Radiology. 2009;251:347–58.
Hong YC. Serum high-density lipoprotein cholesterol and breast cancer risk by menopausa status, body mass index, and hormonal receptor in Korea. Cancer Epidemiol Biomarkers Prev. 2009;18:508–15.
Lin X, Hong S, Huang J, Chen Y, Chen Y, Wu Z. Plasma apolipoprotein A1 levels at diagnosis are independent prognostic factors in invasive ductal breast cancer. Discov Med. 2017;23:247–58.
Carlson SE. An empirically derived dietary pattern associated with breast cancer risk is validated in a nested case-control cohort from a randomized primary prevention trial. Clin Nutr Espen. 2017;17:8–17.
Hux JE. Diabetes mellitus and breast cancer: a retrospective population-based cohort study. Breast Cancer Res Treat. 2006;98:349–56.
Cook R. Serum lactate dehydrogenase (LDH) is a significant prognostic variable for survival in patients with metastatic breast cancer-a multivariate analysis. Eur J Cancer Suppl. 2008;6:189.
Kamby C, Bruun Rasmussen B, Kristensen B. Prognostic indicators of metastatic bone disease in human breast cancer. Cancer. 1991;68:2045–50.
Ryberg M, Nielsen D, Osterlind K, Andersen PK, Skovsgaard T, Dombernowsky P. Predictors of central nervous system metastasis in patients with metastatic breast cancer. A competing risk analysis of 579 patients treated with epirubicin-based chemotherapy. Breast Cancer Res Treat. 2005;91:217–25.
Kołodziejczyk J, Ponczek MB. The role of fibrinogen, fibrin and fibrin(ogen) degradation products (FDPs) in tumor progression. Contemp Oncol. 2013;17:113–9.
Altiay G, Ciftci A, Demir M, Kocak Z, Sut N, Tabakoglu E, et al. High plasma D-dimer level is associated with decreased survival in patients with lung cancer. Clin Oncol. 2007;19:494–8.
Takeuchi H, Ikeuchi S, Kitagawa Y, Shimada A, Oishi T, Isobe Y, et al. Pretreatment plasma fibrinogen level correlates with tumor progression and metastasis in patients with squamous cell carcinoma of the esophagus. J Gastroenterol Hepatol. 2007;22:2222–7.
Matsuda S, Takeuchi H, Fukuda K, Nakamura R, Takahashi T, Wada N. Clinical significance of plasma fibrinogen level as a predictive marker for postoperative recurrence of esophageal squamous cell carcinoma in patients receiving neoadjuvant treatment. Dis Esophagus. 2014;27:654–61.
Fujii T, Tokuda S, Nakazawa Y, Kurozumi S, Obayashi S, Yajima R, et al. Implications of low serum albumin as a prognostic factor of long-term outcomes in patients with breast cancer. In Vivo. 2020;34:2033–6.
Takaaki F, Reina Y, Takahiro T, Toshinaga S, Hiroki M, Satoru Y, et al. Serum albumin and prealbumin do not predict recurrence in patients with breast cancer. Anticancer Res. 2014;34:3775–9.
Chung L, Moore K, Phillips L, Boyle FM, Marsh DJ, Baxter RC, et al. Novel serum protein biomarker panel revealed by mass spectrometry and its prognostic value in breast cancer. Breast Cancer Res. 2014;16:R63.
Zhou T, He X, Fang W, Zhan J, Hong S, Qin T, et al. Pretreatment albumin/globulin ratio predicts the prognosis for small-cell lung cancer. Medicine (Baltimore). 2016;95(12):e3097.
Liu X, Meng QH, Ye Y, Hildebrandt MA, Gu J, Wu X. Prognostic significance of pretreatment serum levels of albumin, LDH and total bilirubin in patients with non-metastatic breast cancer. Carcinogenesis. 2015;36(2):243–8.
Xue F, Lin F, Yin M, Feng N, Zhang X, Cui YG, et al. Preoperative albumin/globulin ratio is a potential prognosis predicting biomarker in patients with resectable gastric cancer. Turk J Gastroenterol. 2017;28(6):439–45.
Yang S, He X, Liu Y, Ding X, Jiang H, Tan Y, et al. Prognostic Significance of serum uric acid and gamma-glutamyltransferase in patients with advanced gastric cancer. Dis Markers. 2019;2019:1415421.
Yue C-F, Feng P-N, Yao Z-R, Yu X-G, Lin W-B, Qian Y-M, et al. High serum uric acid concentration predicts poor survival in patients with breast cancer. Clin Chim Acta. 2017;473:160–5.
Fisher ER, Wang J, Bryant J, Fisher B, Mamounas E, Wolmark N. Pathobiology of preoperative chemotherapy: findings from the National Surgical Adjuvant Breast and Bowel (NSABP) protocol B-18. Cancer. 2002;95(4):681.
Hortobagyi GN. Comprehensive management of locally advanced breast cancer. Cancer. 1990;66(Supplement S14):1387.
von MG, Huang C-S, Mano MS, Loibl S, Mamounas EP, Untch M, et al. Trastuzumab emtansine for residual invasive HER2-positive breast cancer. N Engl J Med. 2019;380:617–28.
Caponio R, Ciliberti MP, Graziano G, Necchia R, Scognamillo G, Pascali A, et al. Waiting time for radiation therapy after breast-conserving surgery in early breast cancer: a retrospective analysis of local relapse and distant metastases in 615 patients. Eur J Med Res. 2016;21:32.
Zhang W-W, Wu S-G, Sun J-Y, Li F-Y, He Z-Y. Long-term survival effect of the interval between mastectomy and radiotherapy in locally advanced breast cancer]. Cancer Manag Res. 2018;10:2047–54.
Wen J, Ye F, Li S, Huang X, Yang L, Xiao X, et al. The Practicability of a Novel Prognostic Index (PI) Model and Comparison with Nottingham Prognostic Index (NPI) in Stages I–III breast cancer patients undergoing surgical treatment. PLoS ONE. 2015;10:e0143537.
Acknowledgements
We thank the BioStudies database (public database) for including and providing Professor Xie's original data [41].
Funding
None.
Author information
Authors and Affiliations
Contributions
All authors provided critical feedback and helped shape the research, analysis, and manuscript.
Corresponding authors
Ethics declarations
Conflict of interest
All authors declare that they have no conflict of interest.
Availability of data and material
Data are available at the BioStudies database: https://www.ebi.ac.uk/biostudies/studies?query=S-EPMC4658156, accession number: S-EPMC4658156.
Ethics approval and consent to participate
This was a secondary analysis using data from the BioStudies database, which is a public database.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Zhou, CM., Xue, Q., Wang, Y. et al. Machine learning to predict the cancer-specific mortality of patients with primary non-metastatic invasive breast cancer. Surg Today 51, 756–763 (2021). https://doi.org/10.1007/s00595-020-02170-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00595-020-02170-9