Nuno Rui Paulino Pereira1, Stein J Janssen2, Eva van Dijk3, Mitchel B Harris4, Francis J Hornicek5, Marco L Ferrone6, Joseph H Schwab7. 1. Department of Orthopaedic Surgery, Orthopaedic Oncology Service, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts nunorui.pp@gmail.com. 2. Department of Orthopaedic Surgery, Orthopaedic Oncology Service, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts steinjanssen@gmail.com. 3. Department of Orthopaedic Surgery, Orthopaedic Oncology Service, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts e.c.vandijk2@students.uu.nl. 4. Department of Orthopaedic Surgery, Orthopaedic Oncology Service, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts mbharris@partners.org. 5. Department of Orthopaedic Surgery, Orthopaedic Oncology Service, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts fhornicek@mgh.harvard.edu. 6. Department of Orthopaedic Surgery, Orthopaedic Oncology Service, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts mferrone@bwh.harvard.edu. 7. Department of Orthopaedic Surgery, Orthopaedic Oncology Service, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts jhschwab@partners.org.
Abstract
BACKGROUND: Current prognostication models for survival estimation in patients with metastatic spine disease lack accuracy. Identifying new risk factors could improve existing models. We assessed factors associated with survival in patients surgically treated for spine metastases, created a classic scoring algorithm, nomogram, and boosting algorithm, and tested the predictive accuracy of the three created algorithms at estimating survival. METHODS: We included 649 patients from two tertiary care referral centers in this retrospective study (2002 to 2014). A multivariate Cox model was used to identify factors independently associated with survival. We created a classic scoring system, a nomogram, and a boosting (i.e., machine learning) algorithm and calculated their accuracy by receiver operating characteristic analysis. RESULTS: Older age (hazard ratio [HR], 1.01; p = 0.009), poor performance status (HR, 1.54; p = 0.001), primary cancer type (HR, 1.68; p < 0.001), >1 spine metastasis (HR, 1.32; p = 0.009), lung and/or liver metastasis (HR, 1.35; p = 0.005), brain metastasis (HR, 1.90; p < 0.001), any systemic therapy for cancer prior to a surgical procedure (e.g., chemotherapy, immunotherapy, hormone therapy) (HR, 1.65; p < 0.001), higher white blood-cell count (HR, 1.03; p = 0.002), and lower hemoglobin levels (HR, 0.92; p = 0.009) were independently associated with decreased survival. The boosting algorithm was best at predicting survival on the training data sets (p < 0.001); the nomogram was more reliable at estimating survival on the test data sets, with an accuracy of 0.75 (30 days), 0.73 (90 days), and 0.75 (365 days). CONCLUSIONS: We identified risk factors associated with survival that should be considered in prognostication. Performance of the boosting algorithm and nomogram were comparable on the testing data sets. However, the nomogram is easier to apply and therefore more useful to aid surgical decision-making. LEVEL OF EVIDENCE: Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.
BACKGROUND: Current prognostication models for survival estimation in patients with metastatic spine disease lack accuracy. Identifying new risk factors could improve existing models. We assessed factors associated with survival in patients surgically treated for spine metastases, created a classic scoring algorithm, nomogram, and boosting algorithm, and tested the predictive accuracy of the three created algorithms at estimating survival. METHODS: We included 649 patients from two tertiary care referral centers in this retrospective study (2002 to 2014). A multivariate Cox model was used to identify factors independently associated with survival. We created a classic scoring system, a nomogram, and a boosting (i.e., machine learning) algorithm and calculated their accuracy by receiver operating characteristic analysis. RESULTS: Older age (hazard ratio [HR], 1.01; p = 0.009), poor performance status (HR, 1.54; p = 0.001), primary cancer type (HR, 1.68; p < 0.001), >1 spine metastasis (HR, 1.32; p = 0.009), lung and/or liver metastasis (HR, 1.35; p = 0.005), brain metastasis (HR, 1.90; p < 0.001), any systemic therapy for cancer prior to a surgical procedure (e.g., chemotherapy, immunotherapy, hormone therapy) (HR, 1.65; p < 0.001), higher white blood-cell count (HR, 1.03; p = 0.002), and lower hemoglobin levels (HR, 0.92; p = 0.009) were independently associated with decreased survival. The boosting algorithm was best at predicting survival on the training data sets (p < 0.001); the nomogram was more reliable at estimating survival on the test data sets, with an accuracy of 0.75 (30 days), 0.73 (90 days), and 0.75 (365 days). CONCLUSIONS: We identified risk factors associated with survival that should be considered in prognostication. Performance of the boosting algorithm and nomogram were comparable on the testing data sets. However, the nomogram is easier to apply and therefore more useful to aid surgical decision-making. LEVEL OF EVIDENCE: Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.
Authors: Andrew J Schoenfeld; Joseph H Schwab; Marco L Ferrone; Justin A Blucher; Tracy A Balboni; Lauren B Barton; John H Chi; James D Kang; Elena Losina; Jeffrey N Katz Journal: Clin Neurol Neurosurg Date: 2019-11-04 Impact factor: 1.876
Authors: Andrew J Schoenfeld; Marco L Ferrone; Peter G Passias; Justin A Blucher; Lauren B Barton; John H Shin; Mitchel B Harris; Joseph H Schwab Journal: Spine J Date: 2019-05-22 Impact factor: 4.166
Authors: Andrew J Schoenfeld; Marco L Ferrone; Joseph H Schwab; Justin A Blucher; Lauren B Barton; Mitchel B Harris; James D Kang Journal: Clin Neurol Neurosurg Date: 2019-04-22 Impact factor: 1.876
Authors: Andrew J Schoenfeld; Elena Losina; Marco L Ferrone; Joseph H Schwab; John H Chi; Justin A Blucher; Genevieve S Silva; Angela T Chen; Mitchel B Harris; James D Kang; Jeffrey N Katz Journal: Cancer Date: 2019-04-15 Impact factor: 6.860
Authors: Ori Barzilai; Anne L Versteeg; Arjun Sahgal; Laurence D Rhines; Mark H Bilsky; Daniel M Sciubba; James M Schuster; Michael H Weber; Peter Pal Varga; Stefano Boriani; Chetan Bettegowda; Michael G Fehlings; Yoshiya Yamada; Michelle J Clarke; Paul M Arnold; Ziya L Gokaslan; Charles G Fisher; Ilya Laufer Journal: Cancer Date: 2018-11-29 Impact factor: 6.860
Authors: Robert J Rothrock; Ori Barzilai; Anne S Reiner; Eric Lis; Adam M Schmitt; Daniel S Higginson; Yoshiya Yamada; Mark H Bilsky; Ilya Laufer Journal: Neurosurgery Date: 2021-01-13 Impact factor: 4.654
Authors: Grace X Xiong; Miles W A Fisher; Joseph H Schwab; Andrew K Simpson; Lananh Nguyen; Daniel G Tobert; Tracy A Balboni; John H Shin; Marco L Ferrone; Andrew J Schoenfeld Journal: Spine (Phila Pa 1976) Date: 2022-04-01 Impact factor: 3.468
Authors: Andrew J Schoenfeld; Gordon P Bensen; Justin A Blucher; Marco L Ferrone; Tracy A Balboni; Joseph H Schwab; Mitchel B Harris; Jeffrey N Katz; Elena Losina Journal: J Bone Joint Surg Am Date: 2021-07-21 Impact factor: 5.284