1
|
Yu A, Lee L, Yi T, Fice M, Achar RK, Tepper S, Jones C, Klein E, Buac N, Lopez-Hisijos N, Colman MW, Gitelis S, Blank AT. Development and external validation of a machine learning model for prediction of survival in extremity leiomyosarcoma. Surg Oncol 2024:102057. [PMID: 38462387 DOI: 10.1016/j.suronc.2024.102057] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/14/2023] [Revised: 01/25/2024] [Accepted: 02/28/2024] [Indexed: 03/12/2024]
Abstract
PURPOSE Machine learning (ML) models have been used to predict cancer survival in several sarcoma subtypes. However, none have investigated extremity leiomyosarcoma (LMS). ML is a powerful tool that has the potential to better prognosticate extremity LMS. METHODS The Surveillance, Epidemiology, and End Results (SEER) database was queried for cases of histologic extremity LMS (n = 634). 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 extremity LMS patients (n = 46). 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 0.75-0.76 at the 5-year time point. The Random Forest (RF) model was the best performing model and used for external validation. This model also performed best at 1-year and worst at 5-year on external validation with c-statistics of 0.90 and 0.87, respectively. The RF model was well calibrated on external validation. This model has been made publicly available at https://rachar.shinyapps.io/lms_app/ CONCLUSIONS: ML models had excellent performance for survival prediction of extremity LMS. Future studies incorporating a larger institutional cohort may be needed to further validate the ML model for LMS prognostication.
Collapse
Affiliation(s)
- Austin Yu
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| | - Linus Lee
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| | - Thomas Yi
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| | - Michael Fice
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| | - Rohan K Achar
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| | - Sarah Tepper
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| | - Conor Jones
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| | - Evan Klein
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| | - Neil Buac
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| | | | - Matthew W Colman
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| | - Steven Gitelis
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| | - Alan T Blank
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL, USA.
| |
Collapse
|
2
|
Lee L, Yi T, Fice M, Achar RK, Jones C, Klein E, Buac N, Lopez-Hisijos N, Colman MW, Gitelis S, Blank AT. Development and external validation of a machine learning model for prediction of survival in undifferentiated pleomorphic sarcoma. Musculoskelet Surg 2024; 108:77-86. [PMID: 37658174 DOI: 10.1007/s12306-023-00795-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/01/2022] [Accepted: 08/20/2023] [Indexed: 09/03/2023]
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.
Collapse
Affiliation(s)
- L Lee
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, 1611 W. Harrison St., Suite 300, Chicago, IL, USA.
| | - T Yi
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, 1611 W. Harrison St., Suite 300, Chicago, IL, USA
| | - M Fice
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, 1611 W. Harrison St., Suite 300, Chicago, IL, USA
| | - R K Achar
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, 1611 W. Harrison St., Suite 300, Chicago, IL, USA
| | - C Jones
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, 1611 W. Harrison St., Suite 300, Chicago, IL, USA
| | - E Klein
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, 1611 W. Harrison St., Suite 300, Chicago, IL, USA
| | - N Buac
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, 1611 W. Harrison St., Suite 300, Chicago, IL, USA
| | - N Lopez-Hisijos
- Department of Pathology, Rush University Medical Center, Chicago, IL, USA
| | - M W Colman
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, 1611 W. Harrison St., Suite 300, Chicago, IL, USA
| | - S Gitelis
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, 1611 W. Harrison St., Suite 300, Chicago, IL, USA
| | - A T Blank
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, 1611 W. Harrison St., Suite 300, Chicago, IL, USA
| |
Collapse
|
3
|
Lee L, Buac N, Colman MW, Gitelis S, Blank AT. Total Knee Arthroplasty for Osteoarthritis Is Uncommon after Intralesional Curettage in Giant Cell Tumor of Bone. J Knee Surg 2023; 36:1218-1223. [PMID: 35901797 DOI: 10.1055/s-0042-1750749] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 02/07/2023]
Abstract
Giant cell tumor of bone (GCTB) is most often treated with intralesional curettage; however, periarticular lesions have been shown to increase risk for osteoarthritis. Additionally, the location of these lesions may occasionally preclude a joint-sparing procedure in recurrent tumors. This study sought to investigate rates of secondary arthroplasty in long-term follow-up of knee GCTB. Cases of knee GCTB treated at our institution were reviewed. Rates of recurrence and secondary arthroplasty were recorded, and Kaplan-Meier survival analyses were performed. The records of 40 patients were reviewed. Local recurrence occurred in 25% of patients. The 1-, 5-, and 10-year recurrence-free survival (RFS) probability was 87.4% (95% CI, 77.0-97.7), 72.4% (95% CI, 57.6-87.2), and 72.4% (95% CI, 57.6-87.2), respectively. Function improved after surgery with a mean preoperative MSTS score of 14.9 (standard deviation [SD] 8.4) and mean postoperative MSTS score of 25.1 (SD 5.6) (p <0.001). Three patients had evidence of radiographic osteoarthritis at the last follow-up though they did not require arthroplasty. Arthroplasty was performed as a secondary procedure in six patients. Five patients underwent arthroplasty for recurrent tumors after initial treatment with curettage and one patient underwent patellar arthroplasty for osteoarthritis after initial treatment with an allograft composite arthroplasty. Arthroplasty is performed as a secondary procedure in patients with GCTB at a relatively infrequent rate and more often for cases of recurrent disease than for osteoarthritis. Overall, patients treated for GCTB have improved functional outcomes after surgery than before. Large, multi-institutional studies may be required to assess the incidence of secondary osteoarthritis requiring arthroplasty as this was an infrequent finding in our cohort.
Collapse
Affiliation(s)
- Linus Lee
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, Illinois
| | - Neil Buac
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, Illinois
| | - Matthew W Colman
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, Illinois
| | - Steven Gitelis
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, Illinois
| | - Alan T Blank
- Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, Illinois
| |
Collapse
|
4
|
Lee L, Yi T, Fice M, Jones C, Klein E, Buac N, Lopez-Hisijos N, Colman MW, Gitelis S, Blank AT. External validation of machine learning models for prediction of survival in undifferentiated pleomorphic sarcoma. J Clin Oncol 2022. [DOI: 10.1200/jco.2022.40.16_suppl.e13551] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022] Open
Abstract
e13551 Background: 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 from 2004 to 2015 for cases of histologically confirmed undifferentiated pleomorphic sarcoma (UPS) and malignant fibrous histiocytoma (MFH). Patient, tumor, and treatment characteristics were recorded, and various machine learning (ML) models were built to predict 1-, 3-, and 5-year survival. The best performing ML models were externally validated using an institutional cohort of UPS patients. 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 0.71 to 0.73 at the 5-year time point. Similarly, all ML models performed best at 1-year and worst at 5-year on external validation. The best performing models had c-statistics of 0.81 at the 5-year time point on external validation, demonstrating good performance in survival prediction. Conclusions: 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. [Table: see text]
Collapse
Affiliation(s)
- Linus Lee
- Rush University Medical Center, Chicago, IL
| | - Thomas Yi
- Rush University Medical Center, Chicago, IL
| | | | | | - Evan Klein
- Rush University Medical Center, Chicago, IL
| | - Neil Buac
- Rush University Medical Center, Chicago, IL
| | | | | | | | | |
Collapse
|
5
|
Becerra A, Buac N, Greydanus M, Sturgis M, Cao D, Roadman D, Coogan C, Cherullo E, Vourganti S, Stephenson A, Chow A. MP61-19 FIVE YEAR SURVIVAL OUTCOME COMPARISON AMONGST PATIENTS WITH UNFAVORABLE VS. FAVORABLE RENAL CELL CARCINOMA SUBTYPES. J Urol 2021. [DOI: 10.1097/ju.0000000000002101.19] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/25/2022]
|