Machine learning model to identify prognostic factors in glioblastoma: a SEER-based analysis

Abstract

Purpose: Analyzing and interpreting large amounts of complex health care data are becoming more insufficient by traditional statistical approaches. However, analyzing Big Data (BD) by machine learning (ML) supports the storage, classification of patient information. Therefore, improves disease identification, treatment evaluation, surgical planning, and outcome prediction. The current study aims to create a competing risk model to identify prognostic factors in glioblastoma (GB). Materials and methods: The study included 31663 patients diagnosed with GB between 2007 and 2018. The data in the study were taken from the Surveillance, Epidemiology, and End Results (SEER) database. Overall survivals (OS), age, race, gender, primary site, laterality, surgery and tumor size at the time of diagnosis, vital status, and follow-up time (months) were selected for the analyzes. Results: The median OS of the patients was found to be 9.00±0.09 months. In addition, all variables in the table were statistically significant risk factors for survival except gender. Therefore, surgery, age, laterality, primary site, tumor size, race, gender variables were used as independent risk factors, and vital status was used as a dependent variable for ML analysis. Looking at the ML results, hybrid model gave the best results according to Accuracy, F-measure, and MCC performance criteria. According to hybrid model, which has the best performance, the diagnosis of alive/dead in 84 and 74 out of 100 patients can be interpreted as correct for 1- and 2-year, respectively. Conclusions: The model created by ML was 84.9% and 74.1% successful in predicting 1- and 2-year survival in GB patients, respectively. Recognition of the fundamental ideas will allow neurosurgeons to understand BD and help evaluate the extraordinary amount of data within the associated healthcare field.

Keywords

• Machine learning
• big data
• glioblastoma
• SEER

Glioblastomda prognostik faktörleri tanımlamak için makine öğrenmesi modeli: SEER tabanlı analiz

Abstract

Amaç: Büyük miktarlardaki karmaşık sağlık hizmeti verilerinin analiz edilmesi ve yorumlanmasında geleneksel istatistiksel yaklaşımlar giderek yetersiz kalmaktadır. Bununla birlikte, Büyük Verinin makine öğrenmesi ile analiz edilmesi, hasta bilgilerinin depolanmasını, sınıflandırılmasını destekler. Bu nedenle hastalık tanımlamasını, tedavi değerlendirmesini, cerrahi planlamayı ve sonuç tahminini geliştirir. Mevcut çalışma, glioblastomda (GB) prognostik faktörleri tanımlamak için bir risk modeli oluşturmayı amaçlamaktadır. Gereç ve yöntem: Çalışmaya 2007-2018 yılları arasında GB tanısı konan 31663 hasta dahil edilmiştir. Çalışmadaki veriler Surveillance, Epidemiology, and End Results (SEER) veri tabanından alınmıştır. Analizler için genel sağ kalımlar, yaş, ırk, cinsiyet, primer bölge, lateralite, cerrahi ve tanı anındaki tümör boyutu, vital durum ve takip süresi (ay) seçildi. Bulgular: Hastaların ortanca sağ kalımı 9,00±0,09 ay olarak bulundu. Ayrıca tablodaki tüm değişkenler cinsiyet dışında sağ kalım için istatistiksel olarak anlamlı risk faktörleriydi. Bu nedenle, makine öğrenmesi analizi için bağımsız risk faktörleri olarak cerrahi, yaş, lateralite, primer bölge, tümör boyutu, ırk, cinsiyet değişkenleri ve vital durum bağımlı değişken olarak kullanıldı. Makine öğrenmesi sonuçlarına bakıldığında, doğruluk, F-ölçümü ve MCC performans kriterlerine göre Hibrit Model en iyi sonuçları vermiştir. En iyi performansa sahip olan hibrit modele göre 100 hastanın 84'ünde canlı/ölü tanısı sırasıyla 1 ve 2 yıl için doğru olarak yorumlanabilmektedir. Sonuç: Makine öğrenmesi ile oluşturulan model GB hastalarında 1 ve 2 yıllık sağ kalımı öngörmede sırasıyla %84,9 ve %74,1 başarılıydı. Temel fikirlerin tanınması, beyin cerrahlarının Büyük Veriyi anlamalarına ve ilgili sağlık hizmetleri alanındaki olağanüstü miktarda veriyi değerlendirmelerine yardımcı olacaktır.

Keywords

• Makine öğrenmesi
• büyük veri
• glioblastoma
• SEER

References

1. 1. Yakar F, Egemen E, Çeltikçi E, et al. The big data awareness of Turkish neurosurgeons: a national survey. J Nervous Sys Surgery 2022;8:9-16. https://doi.org/10.54306/SSCD.2022.200
2. 2. Hinton GE, Osindero S, Teh YW. A fast-learning algorithm for deep belief nets. Neural Comput 2006;1:1527-1554. https://doi.org/10.1162/neco.2006.18.7.1527
3. 3. White SE. A review of big data in healthcare: challenges and opportunities. Open Access Bioinf 2014;6:13-18. https://doi.org/10.2147/OAB.S50519
4. 4. Hashem IAT, Yaqoob I, Anuar NB, Mokhtar S, Gani A, Ullah Khan S. The rise of 'Big Data' on cloud computing: review and open research issues. Inf Syst 2015;47:98-115. https://doi.org/10.1016/j.is.2014.07.006
5. 5. Ostrom QT, Cioffi G, Gittleman H, et al. CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2012-2016. Neuro-Oncology 2019;21:1-100. https://doi.org/10.1093/neuonc/noz150
6. 6. Stupp R, Hegi ME, Mason WP, et al. Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomized phase III study: 5-year analysis of the EORTC-NCIC trial. Lancet Oncol 2009;10:459-466. https://doi.org/10.1016/S1470-2045(09)70025-7
7. 7. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987-996. https://doi.org/10.1056/NEJMoa043330
8. 8. Johnson DR, O'Neill BP, Glioblastoma survival in the United States before and during the temozolomide era. J Neuro-Oncol 2012;107:359-364. https://doi.org/10.1007/s11060-011-0749-4

9. 9. Ostrom QT, Gittleman H, Farah P, et al. CBTRUS statistical report: primary brain and central nervous system tumors diagnosed in the United States in 2006-2010. Neuro Oncol 2013;15:1-56. https://doi.org/10.1093/neuonc/not151
10. 10. Thakkar JP, Dolecek TA, Horbinski C, et al. Epidemiologic and molecular prognostic review of glioblastoma. Cancer Epidemiol Biomarkers Prev 2014;23:1985-1996. https://doi.org/10.1158/1055-9965.EPI-14-0275
11. 11. Filippini G, Falcone C, Boiardi A, et al. Prognostic factors for survival in 676 consecutive patients with newly diagnosed primary glioblastoma. Neuro-Oncol 2008;10:79-87. https://doi.org/10.1215/15228517-2007-038
12. 12. González Bonet LG, Piqueras Sánchez C, Roselló Sastre E, Broseta Torres R, de Las Peñas R. Long-term survival of glioblastoma: a systematic analysis of literature about a case. Neurocirugia 2021;33:227-236. https://doi.org/10.1016/j.neucie.2021.11.001
13. 13. Surveillance Research Program, National Cancer Institute SEER*Stat software. version 8.3.6.1. Available at: https://seer.cancer.gov/seerstat. Accessed April 29, 2020
14. 14. Dangeti P. Statistics for machine learning. Packt Publishing Ltd 2017.
15. 15. Albon C. Python Machine Learning Cookbook. 1st Ed. O’Reilly Media 2018.
16. 16. Kantardzic M. Data Mining: Concepts, Models, Methods, and Algorithms. John Wiley & Sons 2011.
17. 17. Breiman L. Bagging predictors. Machine Learning 1996;24:123-140.
18. 18. Quinlan JR. Induction of decision trees. Machine Learning 1986;1:81-106.
19. 19. Tian M, Ma W, Chen Y, et al. Impact of gender on the survival of patients with glioblastoma. Biosci Rep 2018;38:BSR20180752.e1-9. https://doi.org/10.1042/BSR20180752
20. 20. Goldman DA, Reiner AS, Diamond EL, DeAngelis LM, Tabar V, Panageas KS. Lack of survival advantage among re-resected elderly glioblastoma patients: a SEER-Medicare study. Neuro-Oncol Adv 2020;3:vdaa159.e1-10. https://doi.org/10.1093/noajnl/vdaa159
21. 21. Soon WC, Goacher E, Solanki S, et al. The role of sex genotype in paediatric CNS tumour incidence and survival. Childs Nerv Syst 2021;37:2177-2186. https://doi.org/10.1007/s00381-021-05165-0
22. 22. Liu ZY, Feng SS, Zhang YH, et al. Competing risk model to determine the prognostic factors and treatment strategies for elderly patients with glioblastoma. Sci Rep 2021;11:9321.e1-10. https://doi.org/10.1038/s41598-021-88820-5
23. 23. Lin J, Bytnar JA, Theeler BJ, McGlynn KA, Shriver CD, Zhu K. Survival among patients with glioma in the US Military Health System: A comparison with patients in the Surveillance, Epidemiology, and End Results program. Cancer 2020;126:3053-3060. https://doi.org/10.1002/cncr.32884
24. 24. Bohn A, Braley A, Rodriguez de la Vega P, Zevallos JC, Barengo NC. The association between race and survival in glioblastoma patients in the US: a retrospective cohort study. PLoS One 2018;13:0198581.e1-10. https://doi.org/10.1371/journal.pone.0198581
25. 25. Patel NP, Lyon KA, Huang JH. The effect of race on the prognosis of the glioblastoma patient: a brief review. Neurol Res 2019;41:967-971. https://doi.org/10.1080/01616412.2019.1638018
26. 26. Li H, He Y, Huang L, Luo H, Zhu X. The Nomogram model predicting overall survival and guiding clinical decision in patients with glioblastoma based on the SEER database. Front Oncol 2020;10:1051.e1-14. https://doi.org/10.3389/fonc.2020.01051
27. 27. Shu C, Yan X, Zhang X, Wang Q, Cao S, Wang J. Tumor-induced mortality in adult primary supratentorial glioblastoma multiforme with different age subgroups. Future Oncol 2019;15:1105-1114. https://doi.org/10.2217/fon-2018-0719
28. 28. Forjaz G, Barnholtz Sloan JS, Kruchko C, et al. An updated histology recode for the analysis of primary malignant and nonmalignant brain and other central nervous system tumors in the surveillance, epidemiology, and end results program. Neuro-Oncol Adv 2020;3:vdaa175. https://doi.org/10.1093/noajnl/vdaa175
29. 29. Roa W, Kepka L, Kumar N, et al. International atomic energy agency randomized phase iii study of radiation therapy in elderly and frail patients with newly diagnosed glioblastoma multiforme. J Clin Oncol 2015;33:4145-4150. https://doi.org/10.1200/JCO.2015.62.6606
30. 30. Laperriere N, Weller M, Stupp R, et al. Optimal management of elderly patients with glioblastoma. Cancer Treat Rev 2013;39:350-357. https://doi.org/10.1016/j.ctrv.2012.05.008
31. 31. Barnholtz Sloan JS, Maldonado JL, Williams VL, et al. Racial/ethnic differences in survival among elderly patients with a primary glioblastoma. J Neuro-Oncol 2007;85:171-180. https://doi.org/10.1007/s11060-007-9405-4
32. 32. Ostrom QT, Rubin JB, Lathia JD, et al. Females have the survival advantage in glioblastoma. Neuro Oncol 2018;20:576-577. https://doi.org/10.1093/neuonc/noy002
33. 33. Ostrom QT, Gittleman H, Liao P, et al. CBTRUS Statistical Report: primary brain and other central nervous system tumors diagnosed in the United States in 2010-2014. Neuro Oncol 2017;19:1-88. https://doi.org/10.1093/neuonc/nox158
34. 34. Noone AM, Lund JL, Mariotto A, et al. Comparison of SEER treatment data with Medicare claims. Med Care 2016;54:55-64. https://doi.org/10.1097/MLR.0000000000000073
35. 35. Fyllingen EH, Bø LE, Reinertsen I, et al. Survival of glioblastoma in relation to tumor location: a statistical tumor atlas of a population-based cohort. Acta Neurochir (Wien) 2021;163:1895-1905. https://doi.org/10.1007/s00701-021-04802-6
36. 36. Liu S, Wang Y, Fan X, Ma J, Qiu X, Jiang T. Association of MRI-classified subventricular regions with survival outcomes in patients with anaplastic glioma. Clin Radiol 2017;72:426.e1-6. https://doi.org/10.1016/j.crad.2016.11.013 37. Ben Nsir A, Gdoura Y, Thai QA, Zhani Kassar A, Hattab N, Jemel H. Intraventricular Glioblastomas. World Neurosurg 2016;88:126-131. https://doi.org/10.1016/j.wneu.2015.12.079
37. 38. Yang W, Xu T, Garzon Muvdi T, Jiang C, Huang J, Chaichana KL. Survival of ventricular and periventricular high-grade gliomas: a surveillance, epidemiology, and end results program-based study. World Neurosurg 2018;111:323-334. https://doi.org/10.1016/j.wneu.2017.12.052
38. 39. Liu H, Qin X, Zhao L, Zhao G, Wang Y. Epidemiology and survival of patients with brainstem gliomas: a population-based study using the seer database. Front Oncol 2021;11:692097. https://doi.org/10.3389/fonc.2021.692097
39. 40. Dayani F, Young JS, Bonte A, et al. Safety and outcomes of resection of butterfly glioblastoma. Neurosurg Focus 2018;44:4.e1-8. https://doi.org/10.3171/2018.3.FOCUS1857
40. 41. Babu R, Sharma R, Karikari IO, Owens TR, Friedman AH, Adamson C. Outcome and prognostic factors in adult cerebellar glioblastoma. J Clin Neurosci 2013;20:1117-1121. https://doi.org/10.1016/j.jocn.2012.12.006
41. 42. Jeswani S, Nuno M, Folkerts V, Mukherjee D, Black KL, Patil CG. Comparison of survival between cerebellar and supratentorial glioblastoma patients: surveillance, epidemiology, and end results (SEER) analysis. Neurosurgery 2013;73:240-246. https://doi.org/10.1227/01.neu.0000430288.85680.37
42. 43. Chandra A, Lopez Rivera V, Dono A, et al. Comparative analysis of survival outcomes and prognostic factors of supratentorial versus cerebellar glioblastoma in the elderly: does location really matter? World Neurosurg 2021;146:755-767. https://doi.org/10.1016/j.wneu.2020.11.003
43. 44. Adams H, Chaichana KL, Avendano J, Liu B, Raza SM, Quinones Hinojosa A. Adult cerebellar glioblastoma: understanding survival and prognostic factors using a population-based database from 1973 to 2009. World Neurosurg 2013;80:237-243. https://doi.org/10.1016/j.wneu.2013.02.010
44. 45. Levine SA, McKeever PE, Greenberg HS. Primary cerebellar glioblastoma multiforme. J Neuro-Oncol 1987;5:231-236. https://doi.org/10.1007/BF00151226
45. 46. Palpan Flores A, Vivancos Sanchez C, Roda JM, et al. Assessment of pre-operative measurements of tumor size by mri methods as survival predictors in wild type idh glioblastoma. Front Oncol 2020;10:1662.e1-12. https://doi.org/10.3389/fonc.2020.01662
46. 47. Lacroix M, Abi-Said D, Fourney DR, et al. A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg 2001;95:190-198. https://doi.org/10.3171/jns.2001.95.2.0190
47. 48. Hess KR. Extent of resection as a prognostic variable in the treatment of gliomas. J Neuro-Oncol 1999;42:227-231. https://doi.org/10.1023/a:1006118018770
48. 49. Youngblood MW, Stupp R, Sonabend AM. Role of Resection in glioblastoma management. Neurosurg Clin N Am 2021;32:9-22. https://doi.org/10.1016/j.nec.2020.08.002
49. 50. Peeken JC, Goldberg T, Pyka T, et al. Combining multimodal imaging and treatment features improves machine learning-based prognostic assessment in patients with glioblastoma multiforme. Cancer Med 2019;8:128-136. https://doi.org/10.1002/cam4.1908
50. 51. Upadhaya T, Morvan Y, Stindel E, Le Reste PJ, Hatt M. A framework for multimodal imaging-based prognostic model building: preliminary study on multimodal MRI in glioblastoma multiforme. IRBM 2015;36:345-350. https://doi.org/10.1016/j.irbm.2015.08.001
51. 52. Chang K, Zhang B, Guo X, et al. Multimodal imaging patterns predict survival in recurrent glioblastoma patients treated with bevacizumab. Neuro-Oncology 2016;18:1680-1687. https://doi.org/10.1093/neuonc/now086
52. 53. Sanghani P, Ang BT, King NKK, Ren H. Overall survival prediction in glioblastoma multiforme patients from volumetric, shape and texture features using machine learning. Surg Oncol 2018;27:709-714. https://doi.org/10.1016/j.suronc.2018.09.002
53. 54. Zacharaki EI, Morita N, Bhatt P, O'Rourke DM, Melhem ER, Davatzikos C. Survival analysis of patients with high-grade gliomas based on data mining of imaging variables. AJNR Am J Neuroradiol 2012;33:1065-1071. https://doi.org/10.3174/ajnr.A2939
54. 55. Tewarie IA, Senders JT, Kremer S, et al. Survival prediction of glioblastoma patients-are we there yet? A systematic review of prognostic modeling for glioblastoma and its clinical potential. Neurosurg Rev 2021;44:2047-2057. https://doi.org/10.1007/s10143-020-01430-z