Abstract
Purpose
Machine learning (ML) algorithms to predict cancer survival have recently been reported for a number of sarcoma subtypes, but none have investigated undifferentiated pleomorphic sarcoma (UPS). ML is a powerful tool that has the potential to better prognosticate UPS.
Methods
The Surveillance, Epidemiology, and End Results (SEER) database was queried for cases of histologically confirmed undifferentiated pleomorphic sarcoma (UPS) (n = 665). Patient, tumor, and treatment characteristics were recorded, and ML models were developed to predict 1-, 3-, and 5-year survival. The best performing ML model was externally validated using an institutional cohort of UPS patients (n = 151).
Results
All ML models performed best at the 1-year time point and worst at the 5-year time point. On internal validation within the SEER cohort, the best models had c-statistics of 0.67–0.69 at the 5-year time point. The Multi-Layer Perceptron Neural Network (MLP) model was the best performing model and used for external validation. Similarly, the MLP model performed best at 1-year and worst at 5-year on external validation with c-statistics of 0.85 and 0.81, respectively. The MLP model was well calibrated on external validation. The MLP model has been made publicly available at https://rachar.shinyapps.io/ups_app/.
Conclusion
Machine learning models perform well for survival prediction in UPS, though this sarcoma subtype may be more difficult to prognosticate than other subtypes. Future studies are needed to further validate the machine learning approach for UPS prognostication.
Similar content being viewed by others
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
Weiss SW, Enzinger FM (1978) Malignant fibrous histiocytoma: an analysis of 200 cases. Cancer 41(6):2250–2266
Widemann BC, Italiano A (2018) Biology and management of undifferentiated pleomorphic sarcoma, myxofibrosarcoma, and malignant peripheral nerve sheath tumors: state of the art and perspectives. J Clin Oncol 36(2):160–167
Vodanovich DA, Spelman T, May D, Slavin J, Choong PFM (2019) Predicting the prognosis of undifferentiated pleomorphic soft tissue sarcoma: a 20-year experience of 266 cases. ANZ J Surg 89(9):1045–1050
Le Doussal V, Coindre JM, Leroux A, Hacene K, Terrier P, Bui NB et al (1996) Prognostic factors for patients with localized primary malignant fibrous histiocytoma: a multicenter study of 216 patients with multivariate analysis. Cancer 77(9):1823–1830
Kearney MM, Soule EH, Ivins JC (1980) Malignant fibrous histiocytoma a retrospective study of 167 cases. Cancer 45(1):167–178
Pezzi CM, Rawlings MS Jr, Esgro JJ, Pollock RE, Romsdahl MM (1992) Prognostic factors in 227 patients with malignant fibrous histiocytoma. Cancer 69(8):2098–2103
Bertoni F, Capanna R, Biagini R, Bacchini P, Guerra A, Ruggieri P et al (1985) Malignant fibrous histiocytoma of soft tissue. An analysis of 78 cases located and deeply seated in the extremities. Cancer 56(2):356–367
O’Sullivan B, Davis AM, Turcotte R, Bell R, Catton C, Chabot P et al (2002) Preoperative versus postoperative radiotherapy in soft-tissue sarcoma of the limbs: a randomised trial. The Lancet 359(9325):2235–2241
Davis AM, O’Sullivan B, Turcotte R, Bell R, Catton C, Chabot P et al (2005) Late radiation morbidity following randomization to preoperative versus postoperative radiotherapy in extremity soft tissue sarcoma. Radiother Oncol J Eur Soc Ther Radiol Oncol 75(1):48–53
Winchester D, Lehman J, Tello T, Chimato N, Hocker T, Kim S et al (2018) Undifferentiated pleomorphic sarcoma: factors predictive of adverse outcomes. J Am Acad Dermatol 79(5):853–859
Lehnhardt M, Daigeler A, Homann HH, Schwaiberger V, Goertz O, Kuhnen C et al (2009) MFH revisited: outcome after surgical treatment of undifferentiated pleomorphic or not otherwise specified (NOS) sarcomas of the extremities – an analysis of 140 patients. Langenbecks Arch Surg 394(2):313–320
Xu F, Zhao F, Feng X, Li C, Han D, Zheng S et al (2021) Nomogram for predicting cancer-specific survival in undifferentiated pleomorphic sarcoma: a surveillance, epidemiology, and end results -based study. Cancer Control J Moffitt Cancer Cent 28:10732748211036776
Callegaro D, Miceli R, Bonvalot S, Ferguson P, Strauss DC, Levy A et al (2016) Development and external validation of two nomograms to predict overall survival and occurrence of distant metastases in adults after surgical resection of localised soft-tissue sarcomas of the extremities: a retrospective analysis. Lancet Oncol 17(5):671–680
Balachandran VP, Gonen M, Smith JJ, DeMatteo RP (2015) Nomograms in oncology: more than meets the eye. Lancet Oncol 16(4):e173-180
Kattan MW, Leung DHY, Brennan MF (2002) Postoperative nomogram for 12-year sarcoma-specific death. J Clin Oncol Off J Am Soc Clin Oncol 20(3):791–796
Alabi RO, Mäkitie AA, Pirinen M, Elmusrati M, Leivo I, Almangush A (2021) Comparison of nomogram with machine learning techniques for prediction of overall survival in patients with tongue cancer. Int J Med Inf 145:104313
Connor CW (2019) Artificial intelligence and machine learning in anesthesiology. Anesthesiology 131(6):1346–1359
Abbasi B, Goldenholz DM (2019) Machine learning applications in epilepsy. Epilepsia 60(10):2037–2047
Erickson BJ, Korfiatis P, Akkus Z, Kline TL (2017) Machine learning for medical imaging. Radiogr Rev Publ Radiol Soc N Am Inc 37(2):505–515
Do BH, Langlotz C, Beaulieu CF (2017) Bone tumor diagnosis using a Naïve Bayesian model of demographic and radiographic features. J Digit Imaging 30(5):640–647
Forsberg JA, Sjoberg D, Chen QR, Vickers A, Healey JH (2013) Treating metastatic disease: Which survival model is best suited for the clinic? Clin Orthop 471(3):843–850
Forsberg JA, Wedin R, Boland PJ, Healey JH (2017) Can we estimate short- and intermediate-term survival in patients undergoing surgery for metastatic bone disease? Clin Orthop 475(4):1252–1261
Thio QCBS, Karhade AV, Ogink PT, Raskin KA, De Amorim BK, Lozano Calderon SA et al (2018) Can machine-learning techniques be used for 5-year survival prediction of patients with chondrosarcoma? Clin Orthop 476(10):2040–2048
Bongers MER, Thio QCBS, Karhade AV, Stor ML, Raskin KA, Lozano Calderon SA et al (2019) Does the SORG algorithm predict 5-year survival in patients with chondrosarcoma? An external validation. Clin Orthop 477(10):2296–2303
Bongers MER, Karhade AV, Setola E, Gambarotti M, Groot OQ, Erdoğan KE et al (2020) How does the skeletal oncology research group algorithm’s prediction of 5-year survival in patients with chondrosarcoma perform on international validation? Clin Orthop 478(10):2300–2308
Ogink PT, Groot OQ, Karhade AV, Bongers MER, Oner FC, Verlaan JJ et al (2021) Wide range of applications for machine-learning prediction models in orthopedic surgical outcome: a systematic review. Acta Orthop 92(5):526–531
Stekhoven DJ, Bühlmann P (2012) MissForest–non-parametric missing value imputation for mixed-type data. Bioinforma Oxf Engl 28(1):112–118
Nascimento AF, Raut CP (2008) Diagnosis and management of pleomorphic sarcomas (so-called “MFH”) in adults. J Surg Oncol 97(4):330–339
Jiang J, Pan H, Li M, Qian B, Lin X, Fan S (2021) Predictive model for the 5-year survival status of osteosarcoma patients based on the SEER database and XGBoost algorithm. Sci Rep 11(1):5542
Chen W, Zhou C, Yan Z, Chen H, Lin K, Zheng Z et al (2021) Using machine learning techniques predicts prognosis of patients with Ewing sarcoma. J Orthop Res Off Publ Orthop Res Soc 39(11):2519–2527
Kamalapathy PN, Ramkumar DB, Karhade AV, Kelly S, Raskin K, Schwab J et al (2021) Development of machine learning model algorithm for prediction of 5-year soft tissue myxoid liposarcoma survival. J Surg Oncol 123(7):1610–1617
Bhambhvani HP, Zamora A, Velaer K, Greenberg DR, Sheth KR (2021) Deep learning enabled prediction of 5-year survival in pediatric genitourinary rhabdomyosarcoma. Surg Oncol 36:23–27
Soule EH, Enriquez P (1972) Atypical fibrous histiocytoma, malignant fibrous histiocytoma, malignant histiocytoma, and epithelioid sarcoma. A comparative study of 65 tumors. Cancer 30(1):128–143
Soleymani T, Aasi SZ, Novoa R, Hollmig ST (2019) Atypical fibroxanthoma and pleomorphic dermal sarcoma: updates on classification and management. Dermatol Clin 37(3):253–259
Lin Q, Huang Q, Wang Q, Yan W, Sun Y (2021) Novel nomograms-based prediction models for patients with primary undifferentiated pleomorphic sarcomas resections. Cancers 13(8):1917
Rööser B, Willén H, Gustafson P, Alvegård TA, Rydholm A (1991) Malignant fibrous histiocytoma of soft tissue. A population-based epidemiologic and prognostic study of 137 patients. Cancer 67(2):499–505
Fromm J, Klein A, Baur-Melnyk A, Knösel T, Lindner L, Birkenmaier C et al (2018) Survival and prognostic factors in conventional central chondrosarcoma. BMC Cancer 18(1):849
Giuffrida AY, Burgueno JE, Koniaris LG, Gutierrez JC, Duncan R, Scully SP (2009) Chondrosarcoma in the United States (1973 to 2003): an analysis of 2890 cases from the SEER database. J Bone Joint Surg Am 91(5):1063–1072
Björnsson J, McLeod RA, Unni KK, Ilstrup DM, Pritchard DJ (1998) Primary chondrosarcoma of long bones and limb girdles. Cancer 83(10):2105–2119
Staals EL, Bacchini P, Bertoni F (2006) Dedifferentiated central chondrosarcoma. Cancer 106(12):2682–2691
Grimer RJ, Gosheger G, Taminiau A (2007) Dedifferentiated chondrosarcoma: prognostic factors and outcome from a European group. Eur J Cancer 43(14):2060–2065
Lazarides AL, Visgauss JD, Nussbaum DP, Green CL, Blazer DG, Brigman BE et al (2018) Race is an independent predictor of survival in patients with soft tissue sarcoma of the extremities. BMC Cancer 18(1):488
Alamanda VK, Song Y, Schwartz HS, Holt GE (2015) Racial disparities in extremity soft-tissue sarcoma outcomes: a nationwide analysis. Am J Clin Oncol 38(6):595–599
Funding
None.
Author information
Authors and Affiliations
Contributions
LL (data collection, analysis, interpretation, conception, drafting, editing, revisions); TY (analysis, interpretation, drafting, editing, revisions); MF (data collection, interpretation, drafting, editing, revisions); CJ (data collection, interpretation, drafting, editing, revisions); EK (data collection, interpretation, drafting, editing, revisions); NB (data collection, interpretation, drafting, editing, revisions); NLH (data collection, oversight, editing, revisions); MWC (oversight, editing, revisions); SG (oversight, editing, revisions); ATB (conception, oversight, editing, revisions).
Corresponding author
Ethics declarations
Conflict of interest
ATB: (BMJ Case Reports: Editorial or governing board; Clinical Orthopaedics and Related Research: Editorial or governing board; exparel/pacira: Stock or stock Options; Journal of Oncology Practice: Editorial or governing board; Journal of Surgical Oncology: ad hoc reviewer; Lancet—Oncology: Editorial or governing board; Musculoskeletal Tumor Society: Board or committee member; Onkos Surgical: Paid consultant; Pediatric Blood and Cancer: Editorial or governing board; Rare Tumors: Editorial or governing board; Rush Orthopedic Journal: Editorial or governing board; Swim Across America Cancer Research Grant: Research support); SG: (Onkos Surgical: Paid consultant; Stock or stock Options; USMI: Stock or stock Options); MWC: (Alphatec Spine: IP royalties; Paid consultant; AO Spine North America: Board or committee member; Research support; Cervical Spine Research Society: Board or committee member; CSRS: Research support; DePuy, A Johnson & Johnson Company: Paid presenter or speaker; K2M: Paid presenter or speaker; Musculoskeletal Tumor Society: Board or committee member; North American Spine Society: Board or committee member; Orthofix, Inc.: Paid presenter or speaker; Spinal Elements: Paid consultant. All other authors have no pertinent financial disclosures or pertinent conflicts of interest.
Ethics approval
Rush University Medical Center obtained individual Institutional Review Board approval prior to beginning any research efforts.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Lee, L., Yi, T., Fice, M. et al. Development and external validation of a machine learning model for prediction of survival in undifferentiated pleomorphic sarcoma. Musculoskelet Surg 108, 77–86 (2024). https://doi.org/10.1007/s12306-023-00795-w
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12306-023-00795-w