The use of machine learning (ML) and deep learning (DL) methods in hematology includes diagnostic, prognostic, and therapeutic applications. This increase is due to the improved access to ML and DL tools and the expansion of medical data. The utilization of ML remains limited in clinical practice, with some disciplines further along in their adoption, such as radiology and histopathology. In this review, we discuss the current uses of ML in diagnosis in the field of hematology, including image-recognition, laboratory, and genomics-based diagnosis. Additionally, we provide an introduction to the fields of ML and DL, highlighting current trends, limitations, and possible areas of improvement.

In αβ T-cell/CD19 B-cell depleted hematopoietic stem cell transplantation (αβhaplo-HSCT) recipients, anti--thymocyte globulin (ATG, Thymoglobulin®) is used for preventing graft rejection and graft-versus-host disease (GvHD). However, optimal dosing has yet to be established. Here, we present the first comparative analysis of three different ATG dosing strategies and their impact on immune reconstitution and GvHD. Our study aims to evaluate the effects of three distinct dosing strategies of ATG on engraftment success, αβ+ and γδ+ T-cells immune reconstitution, and the incidence and severity of acute GvHD in recipients of αβhaplo-HSCT. This comparative analysis includes three cohorts of pediatric patients with malignant (36) or non-malignant diseases (8). Cohorts 1 and 2 were given fixed ATG doses, while cohort 3 received doses via a new nomogram, based on absolute lymphocyte count (ALC) and body weight (BW). Cohort 3 showed 0% incidence of Day-100 grade II-IV acute GvHD, as opposed to 48% and 27% in cohort 1 and 2, respectively. Further, cohort 3 (the ALC/BW-based cohort) had a significant increase in CD4+ and CD8+ naïve T-cells by Day 90 (P=0.04; P=0.03). Additionally, we found that the reconstitution and maturation of γδ+ T-cells post-HSCT was not impacted across all three cohorts. Cumulative ATG exposure in all cohorts was lower than previously reported in T-cell replete settings, with a lower pre-HSCT exposure (<40 AU*day/mL) correlating with engraftment failure (p=0.007). Conversely, a post-HSCT ATG exposure between 10-15 AU*day/mL was optimal for improving Day-100 CD4+ (p= 0.058) and CD8+ (p=0.03) immune reconstitution, without increasing relapse or non-relapse mortality risks. This study represents the first comparative analysis of ATG Thymoglobulin® exposure in αβhaplo-HSCT recipients. Our findings indicate that i) a 1-2-fold ATG to ATLG bioequivalence is more effective than previously established standards, and ii) ATG exposure post-HSCT does not adversely affect γδ+ T-cell immune reconstitution. Furthermore, a model-based ATG dosing strategy effectively reduces graft rejection and Day-100 acute GvHD while also promoting early CD4+/CD8+ immune reconstitution. These insights suggest that further optimization, including more distal administration of higher ATG doses within an ALC/BW-based strategy, will yield even greater improvements in outcomes.

Background. Leukemic relapse remains the primary cause of treatment failure and death after allogeneic hematopoietic stem cell transplant. Changes in post-transplant donor chimerism have been identified as a predictor of relapse. A better predictive model of relapse incorporating donor chimerism has the potential to improve leukemia-free survival by allowing earlier initiation of post-transplant treatment on individual patients. We explored the use of machine learning, a suite of analytical methods focusing on pattern recognition, to improve post-transplant relapse prediction. Methods. Using a cohort of 63 pediatric patients with acute lymphocytic leukemia (ALL) and 46 patients with acute myeloid leukemia (AML) who underwent stem cell transplant at a single institution, we built predictive models of leukemic relapse with both pre-transplant and post-transplant patient variables (specifically lineage-specific chimerism) using the random forest classifier. Local Interpretable Model-Agnostic Explanations, an interpretable machine learning tool was used to confirm our random forest classification result. Results. Our analysis showed that a random forest model using these hyperparameter values achieved 85% accuracy, 85% sensitivity, 89% specificity for ALL, while for AML 81% accuracy, 75% sensitivity, and 100% specificity at predicting relapses within 24 months post-HSCT in cross validation. The Local Interpretable Model-Agnostic Explanations tool was able to confirm many variables that the random forest classifier identified as important for the relapse prediction. Conclusions. Machine learning methods can reveal the interaction of different risk factors of post-transplant leukemic relapse and robust predictions can be obtained even with a modest clinical dataset. The random forest classifier distinguished different important predictive factors between ALL and AML in our relapse models, consistent with previous knowledge, lending increased confidence to adopting machine learning prediction to clinical management.

The ability to accurately predict leukemic relapse post-HSCT would improve outcomes by allowing pre-emptive therapeutic strategies. Recent studies have identified post-transplant T- and CD34 cell chimerism as predictors of relapse in patients, who had undergone HSCT for hematologic malignancies (Preuner et al, 2016; Lee et al, 2015). However, these studies assess relapse risk looking at only a single threshold of chimerism using standard regression analysis, which permits only limited consideration of other patient variables. As the result, the findings of these analysis are frequently not applicable to patients generally. Machine learning methods offer the possibility to capture nonlinear relationships and simultaneous interactions between multiple variables, thus better recapitulate the dynamics and nuances of the relapse process in different patients. We use machine learning methods, specifically random forest classification (RF), to build a predictive model of post-transplant relapse and to analyze the data from a cohort of 46 pediatric patients, who received HSCT for acute lymphoblastic leukemia (ALL) and had serial lineage-specific chimerism testing post-transplant. Our model achieved 58 % sensitivity and 98% specificity at predicting relapses in cross validation compared to a baseline model (24% sensitivity, 76% specificity). Consistent with previous reports, our model implicates both peripheral blood (PB) donor CD34 and CD3 chimerism as important variables for relapse. More importantly, the RF showed how different variables interacted with each other, providing additional insights into how to best interpret post-transplant chimerism results. To our knowledge, this is the first study featuring RF machine learning methods in the clinical setting of relapse after HSCT. We use a dataset of patients with ALL undergoing HSCT at Lucile Packard Children's Hospital from 2012 to 2018. Variables collected are summarized in Table 1. The analytical sensitivity of STR-based chimerism testing is 1%. Chimerism results on the same day of relapse were excluded from the analysis. The RF model is based on a set of 500 individual decision trees, each based on a bootstrapped sample of the patient data. A 5-fold cross-validation was used to test predictive skill, with 20% of patients excluded from each fold. We compared results with a Monte Carlo baseline model in which relapse status was repeatedly assigned randomly to each patient with a probability based on the prevalence of relapse in our cohort. Patients, transplantation, and relapse characteristics are summarized in Table 2. Chimerism data are summarized in Table 3. The cross-validation results show a robust predictive skill of relapse within 2 years post-transplant. Our RF achieved 58% sensitivity and 98% specificity, greatly improving the predictive values from the base model (Table 4). Variable importance, the ability of a variable to decrease the error of the prediction model, was calculated for all variables used in our RF (Figure 1). Our analysis shows that the age at the time of transplant has the highest importance, followed by PB donor CD34 chimerism. Bone marrow chimerism generally has lower importance suggesting PB monitoring only is adequate in the clinical setting. We showcase the relationships of 1) age at transplant, 2) donor PB CD34, and 3) donor PB CD3 chimerism to the odds of relapse using a partial dependence plot. Younger patients relapse less often. Donor PB CD34 chimerism exhibits a threshold effect, in which the odds of relapse dramatically decreases when it is above 95% while donor PB CD3 chimerism has a more gradual linear profile (Figure 2). 2D dependence plot of donor PB CD34 and PB CD3 chimerism shows the interaction of the two variables (Figure 3) as continuous variables; relapse risk remaining low with even if donor PB CD3 chimerism is as low as 50% as long as donor PB CD34 chimerism is > 95%. Our study shows that machine learning methods such as RF can be very useful at making accurate predictive model of post-HSCT complications that incorporates multiple variables, allowing for more granular differentiation between different patients. Such analyses can enable more effective deployment of risk-adapted, personalized treatment. By building hundreds of independent decision trees, the RF is also able provide useful insights to the interaction between different variables in a clinically relevant manner. Disclosures: No relevant conflicts of interest to declare.

Graft versus host disease (GVHD) is one of the most serious complications following stem cell transplant in children and is a major cause of morbidity and mortality. Corticosteroids remain the mainstay of treatment, and although a majority of children respond to systemic steroids, those refractory to or dependent upon corticosteroids suffer from complications secondary to long-term steroid administration. This problem has prompted consideration of steroid-sparing treatment strategies, although the time to clinical remission can be variable. Intraarterial corticosteroid delivery has been used in adults as a rescue therapy in steroid-resistant patients, but its use in children has been limited. We investigated the feasibility of intraarterial steroid administration into the bowel and/or liver in a cohort of six pediatric patients with acute GVHD. All patients successfully underwent treatment with no serious adverse effects. Five of five (100%) patients with gastrointestinal bleeding due to GVHD had rapid symptom improvement by 48 h, which was durable up to three weeks. Three of four (75%) patients with hepatic GVHD had improved cholestasis following intraarterial steroid administration. Our experience with this small cohort preliminarily demonstrated the feasibility and safety of intraarterial steroid administration in children with acute GVHD. This approach warrants consideration as a rescue therapy in steroid-refractory cases and as a “bridge” therapy for children with severe acute GVHD who are transitioning to steroid-sparing regimens.