Search

Introduction: Prediction of outcomes following allogeneic hematopoietic cell transplantation (HCT) remains a major challenge. Machine learning (ML) is a computational procedure that may facilitate the generation of HCT prediction models. We sought to investigate the prognostic potential of multiple ML algorithms when applied to a large single-center allogeneic HCT database. Methods: Our registry included 2,697 patients that underwent allogeneic HCT from January 1976 to December 2017. 45 pretransplant baseline variables were included in the predictive assessment of each ML algorithm on overall survival (OS) as determined by area under the curve (AUC). Pretransplant variables used in the EBMT ML study (Shouval et al., 2015) were used as a benchmark for comparison. Results: On the entire dataset, the random forest (RF) algorithm performed best (AUC 0.71 ± 0.04) compared to the second-best model, logistic regression (LR) (AUC = 0.69 ± 0.04) (p < 0.001). Both algorithms demonstrated improved AUC scores using all 45 variables compared to the limited variables examined by the EBMT study. Survival at 100 days post-HCT using RF on the full dataset discriminated patients into different prognostic groups with different 2-year OS (p < 0.0001). We then examined the ML methods that allow for significant individual variable identification, including LR and RF, and identified matched related donors (HR = 0.49, p < 0.0001), increasing TBI dose (HR = 1.60, p = 0.006), increasing recipient age (HR = 1.92, p < 0.0001), higher baseline Hb (HR = 0.59, p = 0.0002), and increased baseline FEV1 (HR = 0.73, p = 0.02), among others. Conclusion: The application of multiple ML techniques on single-center allogeneic HCT databases warrants further investigation and may provide a useful tool to identify variables with prognostic potential. [ABSTRACT FROM AUTHOR]

Background: Allogeneic hematopoietic cell transplantation (HCT) is curative for high-risk acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) but with significant non-relapse mortality (NRM) and relapse. We compared the combination of anti-thymocyte globulin (ATG; 4.5 mg/kg) and post-transplant cyclophosphamide (PTCy; 50 mg/kg on day +3 and +4) with other graft-versus-host disease (GvHD) prophylaxis regimens used for these patients. Methods: We retrospectively analyzed 159 patients, aged 22–73 (median 56) years, having undergone transplantation for high-risk AML (n = 120) or MDS (n = 39). The donors were matched related (33%), unrelated (55%) and haploidentical (12%). Almost all patients used peripheral blood stem cells. Conditioning was myeloablative (34%) or reduced intensity (66%). ATG + PTCy was used in 69 patients (43%), and other GvHD prophylaxis regimens in 90 patients (57%). Results: Grade III–IV acute GvHD occurred in 4% of the ATG + PTCy patients versus 20% of those using other regimens (p = 0.004), and chronic GvHD in 19% of the ATG + PTCy patients versus 41% of those using other regimens (p = 0.003). Two-year GvHD-free relapse-free survival (GRFS) was 30% with ATG + PTCy versus 18% with other regimens (p = 0.04). Multivariable analysis demonstrated that while ATG + PTCy had no significant influence on overall survival, cumulative incidence of relapse or NRM, there was a significant influence on GRFS in favor of ATG + PTCy (HR = 0.69, 95% CI 0.45–0.99, p = 0.04). Conclusions: We conclude that the ATG + PTCy combination significantly improved GRFS in allogeneic HCT for high-risk AML and MDS without influencing other outcomes. [ABSTRACT FROM AUTHOR]