Table 2.
Nomograms for Patients With RPS.
| Study | Development Series Characteristics | Nomogram Details | Internal Validation | External Validation | |||||
|---|---|---|---|---|---|---|---|---|---|
| Selection Criteria | Timeframe | No. of Centers | Predicted Outcomes | No. of Patients | Nomogram’s Covariates (a) | Concordance Index | Yes/No | Concordance Index | |
| Anaya [48] 2010 |
Primary or recurrent, nonmetastatic, resected | 1996–2006 | 1 | Median OS, 3-y OS, and 5-y OS | 343 | Histology (3 categories), completeness of surgical resection, age (dichotomic; cutoff at 65 y), multifocality, tumor size (dichotomic; cutoff at 15 cm), presentation (primary vs. recurrent) | 0.73 (95% CI, 0.71–0.75) | No | - |
| Ardoino [10] 2010 |
Primary, localized, resected | 1985–2007 | 1 | 5-y OS and 10-y OS | 192 | Histology (5 categories), FNCLCC grade, size (continuous), surgical resection margins (complete vs. incomplete), age (continuous) | 0.73 | No | - |
| Gronchi [2] 2013 |
Primary, localized, resected | 1999–2009 | 3 | 7-y OS | 523 | FNCLCC grade, tumor size (continuous), histology (7 categories), patient age (continuous), multifocality (yes vs. no), extent of surgical resection (complete vs. incomplete) | 0.74 | Yes | 0.67–0.73 (b) |
| 7-y DFS | 475 | FNCLCC grade, tumor size (continuous), histology (7 categories), multifocality (yes vs. no) | 0.71 | Yes | 0.63–0.69 (b) | ||||
| Tan [24] 2016 |
Primary, localized, resected | 1982–2010 | 1 | 3-y, 5-y, and 10-y DSD | 632 | Histology (7 categories), extent of surgical resection (R0/R1 vs. R2), no. of organs resected (dichotomic, cutoff at 3 organs), size (3 categories), RT (yes vs. no) | 0.71 (95% CI, 0.66–0.74) | No | - |
| 3-y, 5-y, and 10-y LR rate | 574 | Histology (7 categories), size (3 categories), age (dichotomic; cutoff at 65 y), surgical resection (R0 vs. R1), location (pelvis vs. other), vascular resection (yes vs. no), no. of resected organs (dichotomic; cutoff at 3 organs) | 0.71 (95% CI, 0.67–0.75) | No | - | ||||
| 3-y, 5-y, and 10-y DR rate | 632 | Histology (7 categories), no. of resected organs (0 vs. 1–2 vs. 3 organs), size (3 categories), RT (yes vs. no), vascular resection (yes vs. no) | 0.74 (95% CI, 0.69–0.77) | No | - | ||||
| Raut [49] 2019 |
First local relapse, nonmetastatic, resected with curative intent | 2002–2011 | 22 (TARPSWG centers) | 6-y DFS | 602 | Histology (6 categoies), number of resected organs at first surgery (0–7), multifocality at second surgery (yes vs. no), quality of surgery at second surgery (complete vs. incomplete), grade (1/2 vs. 3), ChT after first surgery (yes vs. no), RT after first surgery (yes vs. no) | 0.67 | No | - |
| 22 (TARPSWG centers) | 6-y OS | 602 | Histology (6 categories), age at second surgery (continuous), quality of surgery at second surgery (complete vs. incomplete), number of resected organs at first surgery (0–7), grade (1/2 vs. 3), multifocality at second surgery (yes vs. no), RT after first surgery (yes vs. no) | 0.70 | No | - | |||
| Callegaro [13] 2021 |
Primary (non-recurrent), non-metastatic, resected with curative intent | 2002–2017 | 4 | 5-y OS at different time points (0–60 months) throughout the first 5 years of follow-up | 1357 | Landmark time, FNCLCC grade (1/2/3), occurrence of LR (yes vs. no), occurrence of DM (yes vs. no), age at first surgery (continuous), and completeness of resection (complete vs. incomplete) | 0.75–0.85 | Yes | 0.72–0.79 (c) |
| 4 | 5-y DFS at different time points (0–60 months) throughout the first 5 years of follow-up | 1357 | Tumour size (continuous), FNCLCCgrade (1/2/3), multifocality (yes vs. no), landmark time, histology and the interaction terms between landmark time and tumour size, tumour grade, and multifocality. | 0.64–0.72 | Yes | 0.62–0.68 (c) | |||
Abbreviations: 95% CI, 95% confidence interval; ChT, chemotherapy; DFS, disease-free survival; DR, distant recurrence; DSD, disease-specific death; FNCLCC, Federation Francaise des Centres de Lutte Contre le Cancer; LR, local; recurrence; OS, overall survival; RPS, retroperitoneal sarcoma; RT, radiotherapy; TARPSWG, TransAtlantic Retroperitoneal Sarcoma Working Group. (a) Variables are listed according to the size of their score range, which reflects their relative effect on the predicted outcome, on a decreasing basis (the first variable exerts the strongest influence on the predicted outcome). (b) External validations were performed on 135 patients from the Institut Gustave Roussy in Villejuif, France (Harrell C-index of 0.67 for the OS nomogram and 0.68 for the DFS nomogram) [2]; 631 patients from a multicentric cohort of the Trans-Atlantic Retroperitoneal Sarcoma Working Group (Harrell C-index of 0.73 for the OS nomogram and 0.69 for the DFS nomogram) [50]; 144 patients from Taipei Veterans General Hospital in Taipei, Taiwan (Harrell C-index, 0.72 for the OS nomogram) [51]; 502 patients who underwent resection of primary RPS at nine high-volume academic US institutions (data from the US Sarcoma Collaborative database) (Harrell C-index of 0.69 for the OS nomogram and 0.65 for the DFS nomogram) [25]; and 109 patients who underwent complete resection for primary retroperitoneal sarcoma at the National Cancer Centre Singapore (Harrell C-index of 0.73 fot the OS nomogram and 0.63 for the DFS nomogram.) [26]. (c) External validation was performed on 487 patients for OS and 452 patients for DFS aged ≥16 years affected by primary (non-recurrent), non-metastatic, RPS who underwent surgery with curative intent at Brigham and Women’s Hospital, Dana-Farber Cancer Institute, Harvard Medical School (Boston, MA, USA), Institut Bergonie (Bordeaux, France), Institut Curie (Paris, France), Leiden University Medical Center (Leiden, The Netherlands), Maria Sklodowska-Curie Institute-Oncology Center (Warsaw, Poland), and The Netherlands Cancer Institute (Amsterdam, The Netherlands).