Outcomes from a Range of Neoadjuvant Radiotherapy Doses in Myxoid Liposarcoma

Summary: The DOREMY trial showed that neoadjuvant radiotherapy (NA-RT) dose deescalation to 36Gy resulted in acceptable rates of treatment effect in resected tumors. However, as a phase 2 non-randomized study, this regimen has not been accepted as standard of care. We describe clinical outcomes of mild dose deescalation for myxoid liposarcoma (ML), comparing patients who received NA-RT equivalent dose in 2Gy fractions (EQD2) < 50Gy versus EQD2 = 50Gy. Medical records were queried for patients at our institution treated with NA-RT between 2011 and 2024 with biopsy confirmation of ML and FISH showing DDIT3 mutation. Records were kept of tumor location (extremity versus other), size, total RT dose, fraction size, % viable tumor post-RT, date of surgery, date of last follow-up (FU) and date of recurrence. Total dose was converted to EQD2 using an alpha/beta for ML of 5. Disease free survival (DFS) intervals in months were calculated from date of surgery to date of recurrence or date of last FU. Wilcoxon rank sum and Fisher exact tests evaluated differences between EQD2 groups by tumor location, tumor size and treatment effect >50%. Log-rank and Cox proportional hazard tests assessed correlations between EQD2, DFS and other variables. 22 patients were included in the study. 17 received EQD2 = 50Gy and 5 received EQD2 50% or tumor location. There was a difference in size, with mean maximal dimension 18.7cm in EQD2 < 50Gy (SD 4.5) versus 10cm in EQD2 = 50Gy (SD 1.4) (p= 0.040). There was no difference in median DFS between the two groups. In the EQD2 < 50Gy group, DFS was 30 months and was not reached in the EQD2 = 50Gy group (p=0.184). DFS was inversely associated with tumor size (hazard ratio 0.89, p = 0.027). Between the two EQD2 groups, there was no difference in recurrences (6 total, 3 per group). All recurrences were distant, with one patient having a local component. Larger tumors were more likely to be treated with EQD2 < 50Gy, seemingly reflecting an effort to avoid excessive dose to adjacent organs at risk in large targets. Despite statistically significant differences in size between the two EQD2 groups, there was no difference in DFS. For the entire ML cohort, there was an inverse association between size and DFS, confirming the importance of size in stage and outcome. In our data set, all recurrences were distant, reflecting the metastatic potential of ML and suggesting that systemic therapy may play a primary role. In terms of NA-RT, while we await randomized studies, we conclude that it is reasonable to prescribe lower dose NA-RT, particularly in larger tumors to reduce toxicities. Erin McCammack (she/her/hers), MD (Presenting Author) - University of Minnesota; Sara Alcorn (she/her/hers), MD, MPH, PhD (Co-Author) - University of Minnesota; Stephanie Terezakis, MD (Co-Author) - University of Minnesota