OBJECTIVE(S):
We sought to determine the prognostic significance of histologic subtype in a large series of patients with primary liposarcoma and to use this information to construct a liposarcoma-specific postoperative nomogram (LPN) for disease-specific survival (DSS).
METHODS:
We identified 852 patients with primary extremity, truncal, or retroperitoneal liposarcoma from a prospective database (7/1982-6/2005). Histology was subtyped by a sarcoma pathologist into five groups according to strict criteria. The influence of clinicopathological factors on DSS was analyzed. A LPN predictive of 5- and 12-year DSS was developed based on a Cox regression model.
RESULTS:
585 patients were alive with a median follow-up time of 41 months (1-259 months) and 50 months for surviving patients. The median tumor burden (sum of maximum tumor diameters) was 14.5 cm (1-139 cm). Table 1 shows 5- and 12-year DSS by histologic subtype. The significant predictors of DSS were age (p=0.008), histologic subtype (p<0.0001), tumor burden (p=0.0002), location (p=0.002), margin status (p<0.0001), and presentation status (p=0.02). The nomogram, internally validated, had excellent calibration (Fig. 1). The concordance index for the LPN and original sarcoma nomogram for 12-year DSS was 0.791 and 0.777, respectively.
CONCLUSIONS:
For primary liposarcoma, the LPN based on histologic subtype discriminates DSS better than the original sarcoma nomogram. This will aid in more accurate counseling of patients and in stratification of patients for clinical trials and molecular analysis.

Disease-Specific Survival by Histologic Subtype

Histologic Subtype	n (%)		5-yr DSS (95% CI)	12-yr DSS (95% CI)
Well-differentiated	367 (43.6)		0.92 (0.87-0.94)	0.74 (0.64-0.81)
De-differentiated	133 (15.8)		0.40 (0.28-0.51)	0.30 (0.16-0.46)
Myxoid	154 (18.3)		0.90 (0.83-0.95)	0.82 (0.72-0.89)
Round cell	80 (9.5)		0.73 (0.60-0.82)	0.57 (0.39-0.71)
Pleomorphic	107 (12.7)		0.64 (0.53-0.73)	0.58 (0.46-0.67)
Overall	852 (100)		0.87 (0.83- 0.90)	0.76 (0.70-0.80)