Pediatrik malign non-seminamatöz germ hücreli testis tümörlerinde sağkalım tahmini için makine öğrenme tekniklerinin uygulanması: bir SEER veritabanı çalışması

Abstract

Amaç: Çocukluk çağı testis kanserleri, tüm çocukluk çağı tümörlerinin %1-2'sini oluşturur. 2013-2019 yılları arasındaki verilere dayanan Surveillance, Epidemiology, and End Results (SEER) veritabanına göre, 5 yıllık sağkalım oranı %95,2 olarak görülmüştür. İkinci en yaygın testis tümörü türü ise malign non-seminomatöz germ hücreli tümördür. Son yıllarda, hastalığın prognozu ve sağkalımı hakkında bilgi elde etmek için çeşitli istatistiksel teknikler ve geniş veritabanları kullanılmaktadır. Bu çalışmada, malign non-seminomatous germ hücreli testis tümörü olan hastaların genel sağkalımını doğru bir şekilde tahmin etmek için yapay zeka ve makine öğrenimi tekniklerini kullanarak yazılım geliştirmeyi amaçladık. Gereç ve yöntem: Çalışmamız, Ocak 1975 ile Aralık 2019 tarihleri arasında malign non-seminomatous germ hücreli testis kanseri tanısı konmuş 0-18 yaş arası 788 hastanın verilerini içermektedir. Çalışmanın temel hipotezi, tanı tarihinden ölüm tarihi veya sağ kalan hastalar için son takip tarihi itibarıyla genel sağkalımı yıllar olarak gösterebilmekti. Sağkalım yanı sıra, tanı anındaki hasta yaşı, ırk, tümörün yerleşim yeri, tanı yılı, tümör histolojik tipi, T evresi, N evresi, M evresi, tümör büyüklüğü, mortalite ve takip süresi gibi faktörleri de analiz ettik. Bulgular: Genel sağkalım 41,29±0,43 yıl olarak bulundu. Medyan sağkalım süresi, 15 yaşından küçük hastalar için 43,21±0,62 yıl ve 15 yaş ve üzeri hastalar için 40,34±0,52 yıl olarak bulundu. Tüm hastalar için genel sağkalıma ek olarak, hastaya özgü sağkalım sağlayan bir yazılım geliştirdik. Sonuç: Son zamanlarda, makine öğrenimi gibi yapay zeka teknikleri, diğer istatistiksel yöntemlere kıyasla dikkate değer ilerlemeler göstermiştir. Sonuç olarak bu çalışmamızda pediatrik NSCGT’de tümörün 2000 yılından sonra teşhis edilmesi, boyutunun 2 cm. den küçük olması, T1M0N0 evre yolk sak tümör olmasının hayatta kalma oranının daha yüksek olduğunu saptadık. Sonuçlarla 10 yıllık sağ kalma tahmin modeli oluşturduk ve bu modelin prognoz, nüks ve hayatta kalma analizinde yapay zeka çalışmalarının ilerlemesine katkı sağlayacagını düşündük.

Keywords

• Non-seminomatöz germ hücreli tümör
• testis tümörü
• SEER
• machine learning
• sağkalım

Application of machine learning techniques for survival prediction in pediatric malignant non-seminomatous germ cell testicular tumors: a SEER database study

Abstract

Purpose: Childhood testicular cancers constitute 1-2% of all childhood tumors. According to the Surveillance, Epidemiology, and End Results (SEER) database, based on data from 2013 to 2019, the 5-year survival rate is 95.2%. The second most common type of testicular tumor is malignant non-seminomatous germ cell tumor. In recent years, various statistical techniques and extensive databases have been used to obtain information on disease prognosis and survival. In this study, we aimed to develop software using artificial intelligence and machine learning techniques to accurately predict the overall survival of patients with malignant nonseminomatous germ cell testicular tumors. Materials and methods: Our study included data from 788 patients aged 0-18 diagnosed with malignant nonseminomatous germ cell testicular cancer between January 1975 and December 2019. The main hypothesis of the study was to provide overall survival (OS) in years from the date of diagnosis to the date of death or the last follow-up date for surviving patients. In addition to survival analysis, we also analyzed patient age at diagnosis, race, laterality, year of diagnosis, tumor histological type, T stage, N stage, M stage, tumor size, mortality, and follow-up duration. Results: The OS was found to be 41.29±0.43 years. The median survival time was 43.21±0.62 years for patients <15 and 40.34±0.52 years for patients aged ≥15. We developed software that enabled the provision of patient-specific survival in addition to OS for all patients. Conclusion: Recently, artificial intelligence techniques such as machine learning, have shown remarkable advancements compared to other statistical methods. As a result, in this study, we found that the survival rate in pediatric NSCGT was higher if the tumor was diagnosed after 2000, was less than 2 cm in size, and had a T1M0N0 stage yolk sac tumor. We created a 10-year survival prediction model with the results and thought that this model would contribute to the advancement of artificial intelligence studies in prognosis, recurrence and survival analysis.

Keywords

• Non-seminomatous germ cell tumors
• tescicular tumors
• SEER
• machine learning
• survival

References

1. 1. SEER*Explorer Application [Internet]. Available from: https://seer.cancer.gov/statistics-network/explorer/application.html?site=67&data_type=1&graph_type=3&compareBy=race&chk_race_1=1&rate_type=2&hdn_sex=2&advopt_precision=1&advopt_show_ci=on&hdn_view=1#resultsRegion1. Accessed Apr 2, 2023
2. 2. Ahmed HU, Arya M, Muneer A, Mushtag I, Sebire NJ, Testicular and paratesticular tumors in the prepubertal population. Lancet Oncol 2010;11:476-483. https://doi.org/10.1016/S1470-2045(10)70012-7
3. 3. Facts About Testicular Cancer | Testicular Cancer Statistics [Internet]. Available at: https://www.cancer.org/cancer/testicular-cancer/about/key-statistics.html. Accessed Mar 26, 2023
4. 4. Cancer today [Internet]. Available from: http://gco.iarc.fr/today/home. Accessed September 2, 2023 5. Turabian J. Prognosis-based medicine-the importance of psychosocial factors: conceptualization from a case of acute pericarditis. Trends Gen Pract 2018;1. https://doi.org/10.15761/TGP.1000101
5. 6. SEER Incidence Database - SEER Data & Software [Internet]. SEER. Available from: https://seer.cancer.gov/data/index.html. Accessed April 7, 2023
6. 7. McGlynn KA, Trabert B. Adolescent and adult risk factors for testicular cancer. Nat Rev Urol 2012;9:339-349. https://doi.org/10.1038%2Fnrurol.2012.61
7. 8. Cancer statistics for adolescents and young adults, 2020 - Miller - 2020 - CA: A Cancer Journal for Clinicians - Wiley Online Library [Internet]. Available from: https://acsjournals.onlinelibrary.wiley.com/doi/10.3322/caac.21637 Accessed April 20, 2023
8. 9. Key Statistics for Childhood Cancers [Internet]. Available from: https://www.cancer.org/cancer/cancer-in-children/key-statistics.html. Accesed April 20, 2023

9. 10. Statistical Methods in the Cancer Surveillance Research Program (SRP).
10. 11. Srivastava A, Patel HV, Koehne E, et al. Survival of nonseminomatous germ cell tumors in pediatric patients and young adults - a stage group stratified analysis. Urol Oncol 2022;40:169.e1-169.e12. https://doi.org/10.1016/j.urolonc.2021.12.012
11. 12. Mao W, Wu J, Kong Q, Li J, Xu B, Chen M. Development and validation of prognostic nomogram for germ cell testicular cancer patients. Aging 2020;12:22095-22111. https://doi.org/10.18632%2Faging.104063
12. 13. Li M, Wang J, Li J, et al. Develop and validate nomogram to predict cancer-specific survival for patients with testicular yolk sac tumors. Front Public Health 2022;10:1038502. https://doi.org/10.3389/fpubh.2022.1038502
13. 14. Wu P, Yang Y, Yu Z, Zhao L, Feng S. Clinical features and survival outcomes in children and adolescents with malignant mediastinal germ cell tumors based on surveillance, epidemiology, and end results database analysis. J Surg Res 2023;288:362-371. https://doi.org/10.1016/j.jss.2023.03.018
14. 15. Berney DM, Warren AY, Verma M, et al. Malignant germ cell tumours in the elderly: a histopathological review of 50 cases in men aged 60 years or over. Mod Pathol 2008;21:54-59. https://doi.org/10.1038/modpathol.3800978
15. 16. Li Y, Lu Q, Wang Y, Ma S. Racial differences in testicular cancer in the United States: descriptive epidemiology. BMC Cancer 2020;20:284. https://doi.org/10.1186/s12885-020-06789-2
16. 17. Caballero Mora FJ, Muñoz Calvo MT, García Ros M, et al. Testicular and paratesticular tumors during childhood and adolescence. An Pediatr (Barc) 2013;78:6-13. https://doi.org/10.1016/j.anpedi.2012.05.018
17. 18. Geng R, Zheng Z, Lin Y, et al. Clinical characteristics and prognostic factors of male yolk sac tumor: a Surveillance, Epidemiology, and End Results program study. World J Urol 2021;39:1211-1217. https://doi.org/10.1007/s00345-020-03311-y
18. 19. Curto ML, Lumia F, Alaggio R, et al. Malignant germ cell tumors in childhood: results of the first Italian cooperative study “TCG 91”. Med Pediatr Oncol 2003;41:417-425. https://doi.org/10.1002/mpo.10324
19. 20. Song G, Xiong GY, Fan Y, et al. The role of tumor size, ultrasonographic findings, and serum tumor markers in predicting the likelihood of malignant testicular histology. Asian J Androl 2019;21:196-200. https://doi.org/10.4103%2Faja.aja_119_18