National Trends in Treatment for Retroperitoneal Soft Tissue Sarcoma: A Modern Appraisal of Variability in Therapeutic Strategies.

2021 
BACKGROUND Little level 1 evidence exists to guide multimodality treatment in retroperitoneal soft tissue sarcoma (RPS), which may lead to significant variation in therapeutic approaches. This analysis aimed to describe national RPS treatment trends and explore potential variability among low-/high-volume hospitals (LVH/HVHs). PATIENTS AND METHODS In total, 5992 patients who underwent resection for primary RPS were retrospectively identified in the National Cancer Database (2004-2017). Time trend analyses examined rates of multivisceral resection (MVR), radiation, and chemotherapy use. LVHs were defined as those carrying out fewer than ten resections per year (N = 5433), whereas HVHs were defined as those carrying out ten or more (N = 559). Descriptive statistics and logistic regression models compared trends between groups. RESULTS MVR was more frequent at HVHs (63.7% versus 43.5%, p < 0.001). Use of radiation varied widely by hospital volume. HVHs more frequently employed preoperative radiation as compared with LVHs (14.7% versus 8.1%, p < 0.001). Throughout the study period, LVHs increased utilization of preoperative radiation (2.6% to 12.0%, p < 0.001) whereas rates at HVHs remained stable. Overall, LVHs utilized postoperative radiation significantly more frequently as compared with HVHs (14.7% versus 2.7%, respectively, p < 0.001). Postoperative radiation at LVHs remained stable until 2013 and sharply declined thereafter (16.7% to 6.9%, p < 0.001). Rates of postoperative radiation use at HVHs remained lower than those at LVHs at all time points. CONCLUSIONS Strategies for resection and radiation use at LVHs have trended towards those of HVHs. Current national implementation of preoperative radiation, MVR, and chemotherapy remains heterogeneous. These findings inform future trial design and support standardization of care.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    40
    References
    1
    Citations
    NaN
    KQI
    []