(P 91) BLOOD TRANSFUSION AND SURVIVAL AMONG CHILD, ADOLESCENT, AND YOUNG ADULT PATIENTS WITH OSTEOSARCOMA: A RETROSPECTIVE COHORT STUDY

Objective: Prior data suggest that pretransplant blood transfusion markedly increases survival of kidney allograft in a dose-dependent fashion, presumably by creating immune tolerance in the recipients. Such data has raised concerns that blood transfusion among oncologic patients may subdue anti-cancer immunity and thereby worsen survival outcomes. Many observational studies and meta-analyses reported that transfusion was associated with decreased survival and increased recurrence of various tumors including colorectal cancer and lung cancer. However, scarce data are available for pediatric malignancies. The primary objective of this study, therefore, is to investigate the association between post-diagnosis blood transfusion and survival of child, adolescent, and young adult (CAYA, defined as 39 years of age or younger) patients with osteosarcoma.

Methods: This retrospective cohort study included patients with osteosarcoma at our institution from 2007 to 2022. Search queries for osteosarcoma were built using eligibility criteria consisting of diagnostic codes for bone cancer (C40 or C41) and a medication code for cisplatin (2555) as per the International Classification of Diseases 10th Revision. Cisplatin is a standard-of-care first-line chemotherapy for osteosarcoma but rarely given for other bone sarcomas. Identified patients were manually reviewed to confirm the diagnosis of osteosarcoma and classified according to whether transfusion was required during the initial 10 weeks of chemotherapy. Kaplan-Meier estimates of 3- and 5-year overall survival (OS) and event-free survival (EFS) were calculated. Inter-group survival functions were compared with the log-rank test while point-wise comparisons were calculated using Z-tests.

Results: Among 79 CAYA patients identified with osteosarcoma, six were excluded due to unclear diagnosis date, comorbidity of Diamond-Blackfan anemia, and/or considerable treatment occurring at the outside facility. Of the remaining 73 patients, 34 (46.6%) required blood transfusion during the first 10 weeks of chemotherapy prior to surgery. Among transfused patients, 30 (88.2%) received red blood cells (RBCs) and 14 (41.2%) received platelets, while 10 (29.4%) received both platelets and RBCs. Table 1 describes demographic and clinical characteristics by transfusion status. Median (range) age at diagnosis for transfused and non-transfused patients was 15 (9 to 28) and 14 (4 to 29), respectively. Sex at birth was evenly distributed among the non-transfused group while the majority of the transfused group was male (74.4%). There were no significant inter-group differences in race and ethnicity, primary tumor location, stage, or exposure to ifosfamide. Median (range) survival time for the entire cohort was 2.9 (0.5 to 11.4) years. Among the transfused group, median (range) survival was 3.8 (0.8 to 11.4) years and among the non-transfused group, median (range) survival was 2.8 (0.5 to 6.4) years. Table 2 & 3 and Figure 1 summarize OS and EFS of patients by transfusion type and status. As seen in Table 2, there were no observed inter-group differences in OS and EFS at 3- (OS: 81.2% vs. 81.0%, p>.9; EFS: 74.9% vs. 73.6%, p=.9) or 5-years (OS: 76.7% vs. 70.8%, p=.6; EFS: 58.2% vs. 41.7%, p=.3): Three-year EFS of patients not transfused with platelets differed significantly from patients transfused with platelets (70% vs. 91.7%, p=.04). There were no other significant associations observed between 3- and 5-year EFS or OS and RBC or platelet transfusion.

Conclusion: These data did not demonstrate a significant association between blood transfusion and OS or EFS of CAYA patients with osteosarcoma. Given the small sample size and retrospective nature of the study, further investigation with a larger, prospective cohort should be considered.