(P 223) ASSOCIATION OF POSTOPERATIVE INFECTION (POI) WITH SOFT TISSUE SARCOMA (STS) RESECTION OUTCOMES

Objective: Soft tissue sarcomas (STS) are a heterogenous group of rare malignancies of mesenchymal origin with more than 100 histologic subtypes. Advances in imaging, radiation, and surgical technique have resulted in major changes in the management of STS in recent years. Approximately 90% of previously unresectable extremity STS (ESTS) cases have been converted to limb-preservation surgeries. However, STS resections can be complicated by major wound complications (MWC). Risk factors associated with MWC include patient characteristics (age, BMI, smoking history, diabetes), tumor characteristics (size, location, skin proximity), treatment variables (radiation, chemotherapy), and surgical closure types (primary, staged, skin grafts, free vs rotational flap). Postoperative infection (POI) is a subset of MWC that happens in 5-12% of primary STS patients. POI following osteosarcoma resection has been shown to be associated with potential survival benefits, with infections possibly serving as immune checkpoint inhibitors. The association of POI with STS resection outcomes has not been well studied. A single 2014 study by Behnke et al evaluated STS patients with and without POI and found no difference in survival, local recurrence, or metastasis.

Our goal was to expand this line of inquiry with a larger cohort and a different patient population to determine whether POIs were correlated with distant metastasis (DM), 5-year overall survival (OS), or 5-year disease-specific survival (DSS) in patients with ESTS.

Methods: A retrospective case review was conducted on a cohort of patients with ESTS that underwent surgical resection at a single tertiary referral and sarcoma center from 2008-2023. Patients were divided into two groups: those who developed POI following STS resection and those who did not. Patient characteristics, tumor characteristics, radiation therapy, chemotherapy, and presence of metastatic disease at presentation were compared between the two groups using Pearson’s Chi-squared test and Welch Two Sample t-tests.

A total of 691 participants were identified. To determine DSS and OS, we utilized Surveillance, Epidemiology, and End Results (SEER) Cause-Specific Death Classification. Seven participants were excluded due to missing or unknown cause of death for further analysis. Primary outcomes metastasis-free survival, OS, and DSS were compared using Pearson’s Chi-squared test. Survival curves for metastasis-free survival, OS, and DSS were analyzed using Cox Proportional Hazard (CPH) models and presented using Kaplan-Meier methods (n= 684 participants; n=172 with POI) Multivariable adjusted CPH models were used to calculate hazard ratios (HR). Statistical software R was used for analyses. P-values of less than 0.05 were considered to indicate statistical significance.

Results: Out of 691 participants, 175 with POI and 516 without. Patient groups were equally matched, in terms of sex (p=0.11) and age at the time of surgery (p=0.75). Median BMI was significantly larger in the POI group compared to without (p< 0.001). Smoking status (p=0.077) and diagnosis of diabetes (p=0.056) were not statistically significant between the groups. The presence of STS in the lower extremity (p< 0.001) and tumor grade (p=0.015) were significantly higher and tumor size was significantly larger (p< 0.001) in the POI group. Microscopic margins were not statistically different between the two (p=0.26). The number of patients who received radiation therapy (p< 0.001), regardless of timing (p< 0.001), received chemotherapy (p=0.001), or who presented with a metastatic disease (p=0.008) was significantly higher in the POI group.

OS was significantly lower in the POI group (p=0.019), however presence of POI did not have a statistically significant effect on the DM (p=0.37) or DSS (p=0.082) (Table 2). After multivariable adjustment of CPH models, the risk of DM, OS, and DSS among those who developed POI was 0.97 (95% CI: 0.45, 2.12), 1.60 (95%CI: 0.73, 3.53), and 1.01 (95% CI: 0.73, 1.41) respectively. There was no difference in DM (p=0.95), OS (p=0.25), and DSS (p=0.95) between patients with or without a POI (Figure 1).

A competing risk model demonstrated positive margins, increased tumor size, higher tumor grade, and metastasis at presentation increased the risk of all oncologic outcomes. In contrast, receiving radiation therapy was associated with improved DM, OS, and DSS (Table 3).

Conclusion: In our study, tumor grade was the greatest predictor of all oncologic outcomes followed by metastasis at presentation. Previous data showed getting radiation therapy is a risk factor for MWC including POI. However, our study showed protective effect of the XRT in completing risk analysis. This apparent protective effect is likely not causative but rather due to lack of patients who qualified for XRT due to advanced disease course.

In conclusion, there are no significant differences in long term oncologic outcomes between the ESTS resection patients who developed POI and those who did not. Nonetheless, developing POI delays adjuvant care, increases comorbidities, impacts quality of life, and causes burden to health care system.