(P 373) EVALUATION OF HISTOLOGIC ORGAN INFILTRATION (HOI) AND VISCERAL RESECTION MARGINS (VRM) IN PRIMARY RETROPERITONEAL SARCOMAS: PROGNOSTIC IMPLICATIONS FOR LOCAL RECURRENCE

Objective: Retroperitoneal sarcomas (RPS) are rare and heterogeneous tumors, often exhibiting local recurrence (LR) despite extensive surgical procedures. Key features in prognostic evaluation include tumor histology, tumor grade, and completeness of surgical resection. This study aimed to investigate the prognostic implications of histologic organ infiltration (HOI) and visceral resection margin (VRM) status for LR in RPS and to develop a predictive model.

Methods: We retrospectively analyzed all consecutive patients with primary retroperitoneal liposarcoma (LPS) and leiomyosarcoma (LMS) operated at IRCCS Humanitas Research Hospital (Rozzano, Milan) from 1998 to 2023. Patients with multifocality and metastasis at diagnosis were excluded. Data encompassing clinical, surgical, and neoadjuvant therapy details were collected. Attention was given to histological parameters such as the degree of HOI and assessment of VRM; HOI was classified into 4 degrees (absent/HOI 0, perivisceral/HOI 1, initial/HOI 2, advanced/HOI 3). A predictive tool for LR was developed based on the findings.

Results: We included 182 patients. Demographic and tumor characteristics are delineated in Table 1. HOI 0 comprised 13.2% of the total, with LMS representing 19.6% of these cases. HOI 1 and HOI 2 accounted for 17.6% and 7.1% of cases, respectively, with a prevalence of well-differentiated LPS (WDLPS) at 21.0% and 15.8%. HOI 3 was the most common, observed in 62.1% of cases, with a median tumor size of 20 cm and dedifferentiated LPS (DDLPS) as the most prevalent histotype.
Figures 1A and 1B depict the median overall survival (OS) and disease-free survival (DFS) for HOI 0/1 and HOI 3. While results were not statistically significant, the hazard ratios (HR) showed consistently worse prognosis for patients with higher HOI compared to those with lower HOI (DFS HR: 1 for HOI 0/1, 0.5 for HOI 2, and 1.5 for HOI 3).
For HOI 0/1, LR-free survival at 1-, 3-, and 4-years was 91.0%, 70.0%, and 62.0%, respectively. For HOI 3, the rates were 84.0%, 70.0%, and 67.0%. HR was 1 for HOI 0/1, 0.6 for HOI 2, and 1.5 for HOI 3.
Of the 158 patients considered for VRM, 28 (17.7%) exhibited infiltrated margins: 15 (53.6%) DDLPS, 6 (21.4%) WDLPS, and 7 (25.0%) LMS. Renal vein margins demonstrated a higher frequency of infiltration across all histologies compared to other organs (see Figure 2).
Figure 3 shows that the group with negative VRM had better OS and DFS compared to the positive VRM group, although these differences were not statistically significant (OS p=0.55; DFS p=0.16). In LPS, among the 21 cases with positive VRM, 10 (47.6%) experienced LR, while 11 (52.4%) did not. Among the 96 cases with negative VRM, 26 (27.1%) experienced LR, while 70 (72.9%) did not (P = 0.07). In LMS cases, of the 7 patients with positive VRM, 2 (28.6%) experienced LR while 5 (71.4%) did not. Among the 34 patients with negative VRM, 12 (35.3%) experienced LR while 22 (64.7%) did not (P = 0.73).
We have created a predictive tool where users can input the number of cRO, HOI level, VRM status, and histology. The output will provide the rate of LR.

Conclusion: Our study highlights the significance of HOI and VRM status in predicting LR in RPS patients. While the prognostic differences were not statistically significant, higher HOI and positive VRM were associated with worse outcomes. The predictive tool we developed can assist clinicians in estimating LR risk based on individual patient profiles. Further research with larger sample sizes is necessary to validate these findings.