Trends in allogeneic transplantation for favorable risk acute myeloid leukemia in first remission: a longitudinal study of >15 years from the ALWP of the EBMT

We assessed outcomes of allogeneic transplantation (HSCT) in favorable risk AML in CR1 over 3 time periods. 1850 patients were included, 2005 to 2009- 222, 2010 to 2014 -392, and 2015 to 2021-1236; 526 with t (8:21), 625 with inv (16), and 699 with NPM1mutFLT3WT. Patients transplanted in 2015-2021 were older (p < 0.0001) with more patients ≥60 years of age (p < 0.0001). The most frequent diagnosis in 2015-2021 was NPM1mutFLT3WT vs. t (8:21) in the 2 earlier periods, (p < 0001). Haploidentical transplants (Haplo) increased from 5.9% to 14.5% (p < 0.0001). Graft-versus-host disease (GVHD) prophylaxis with post-transplant cyclophosphamide (PTCy) was more frequent in 2015-2021 vs. the other 2 periods (p < 0.0001). On multivariate analysis, incidence of total chronic GVHD was reduced in HSCTs performed ≥2015 vs. those performed in 2005-2009, hazard ratio (HR) = 0.74 (95% CI 0.56-0.99, p = 0.046) and GVHD-free, relapse-free survival (GRFS) improved for patients transplanted from 2010-2014 vs. those transplanted in 2005-2009, HR = 0.74 (95% CI 0.56-0.98, p = 0.037). Other HSCT outcomes did not differ with no improvement ≥2015. LFS, OS, and GRFS were inferior in patients with t (8:21) with HR = 1.32 (95% CI 1.03-1.68, p = 0.026), HR = 1.38 (95% CI 1.04-1.83, p = 0.027) and HR = 01.25 (95% CI 1.02-1.53, p = 0.035), respectively. In conclusion, this retrospective analysis of HSCT in patients with favorable risk AML, transplanted over 16 years showed an increased number of transplants in patients ≥60 years, from Haplo donors with PTCy. Most importantly, 3-year GRFS improved ≥2010 and total chronic GVHD reduced ≥2015, with no significant change in other HSCT outcomes.

Dynamic Relapse Prediction by Peripheral Blood WT1mRNA after Allogeneic Hematopoietic Cell Transplantation for Myeloid Neoplasms

Although various relapse prediction models based on pretransplant information have been reported, they cannot update the predictive probability considering post-transplant patient status. Therefore, these models are not appropriate for deciding on treatment adjustment and preemptive intervention during post-transplant follow-up. A dynamic prediction model can update the predictive probability by considering the information obtained during follow-up. This study aimed to develop and assess a dynamic relapse prediction model after allogeneic hematopoietic cell transplantation (allo-HCT) for acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) using peripheral blood Wilms' tumor 1 messenger RNA (WT1mRNA). We retrospectively analyzed patients with AML or MDS who underwent allo-HCT at our institution. To develop dynamic models, we employed the landmarking supermodel approach, using age, refined disease risk index, conditioning intensity, and number of transplantations as pretransplant covariates and both pre- and post-transplant peripheral blood WT1mRNA levels as time-dependent covariates. Finally, we compared the predictive performances of the conventional and dynamic models by area under the time-dependent receiver operating characteristic curves. A total of 238 allo-HCT cases were included in this study. The dynamic model that considered all pretransplant WT1mRNA levels and their kinetics showed superior predictive performance compared to models that considered only pretransplant covariates or factored in both pretransplant covariates and post-transplant WT1mRNA levels without their kinetics; their time-dependent areas under the curve were 0.89, 0.73, and 0.87, respectively. The predictive probability of relapse increased gradually from approximately 90 days before relapse. Furthermore, we developed a web application to make our model user-friendly. This model facilitates real-time, highly accurate, and personalized relapse prediction at any time point after allo-HCT. This will aid decision-making during post-transplant follow-up by offering objective relapse forecasts for physicians.

Impact of waitlist time on post-HSCT survival: a cohort study at a hospital in southern Brazil

INTRODUCTION

The time elapsed from diagnosis to hematopoietic stem cell transplantation (HSCT) is influenced by numerous factors. In Brazil, patients using the public health system are also dependent on the availability of HSCT-specific beds in the hematology ward.

OBJECTIVE AND METHODS

We conducted a cohort study of listed patients who underwent allogeneic HSCT at a Brazilian public hospital to investigate the impact of the waitlist time on post-HSCT survival.

RESULTS

The median time from diagnosis to HSCT was 19 months (IQR, 10 - 43), of which 6 months (IQR, 3 - 9) were spent on the waitlist. The time on the waitlist for HSCT appeared to influence mainly the survival of adult patients (≥ 18 years), with an increasing risk according to this time (RR, 3.53 and 95%CI, 1.81 - 6.88 for > 3 and ≤ 6 months; RR 5.86 and 95%CI, 3.26 - 10.53 for > 6 and ≤ 12 months, and; RR 4.24 and 95%CI, 2.32 - 7.75 for > 12 months).

CONCLUSION

Patients who remained on the waitlist for less than 3 months had the highest survival (median survival, 856 days; IQR, 131 - 1607). The risk of reduced survival was about 6-fold higher (95%CI, 2.8 - 11.5) in patients with malignancies.

Management of ALL in adults: 2024 ELN recommendations from a European expert panel

ABSTRACT

Experts from the European Leukemia Net (ELN) working group for adult acute lymphoblastic leukemia have identified an unmet need for guidance regarding management of adult acute lymphoblastic leukemia (ALL) from diagnosis to aftercare. The group has previously summarized their recommendations regarding diagnostic approaches, prognostic factors, and assessment of ALL. The current recommendation summarizes clinical management. It covers treatment approaches, including the use of new immunotherapies, application of minimal residual disease for treatment decisions, management of specific subgroups, and challenging treatment situations as well as late effects and supportive care. The recommendation provides guidance for physicians caring for adult patients with ALL which has to be complemented by regional expertise preferably provided by national academic study groups.

Development and validation of prognostic machine learning models for short- and long-term mortality among acutely admitted patients based on blood tests

Several scores predicting mortality at the emergency department have been developed. However, all with shortcomings either simple and applicable in a clinical setting, with poor performance, or advanced, with high performance, but clinically difficult to implement. This study aimed to explore if machine learning algorithms could predict all-cause short- and long-term mortality based on the routine blood test collected at admission.

METHODS

We analyzed data from a retrospective cohort study, including patients > 18 years admitted to the Emergency Department (ED) of Copenhagen University Hospital Hvidovre, Denmark between November 2013 and March 2017. The primary outcomes were 3-, 10-, 30-, and 365-day mortality after admission. PyCaret, an automated machine learning library, was used to evaluate the predictive performance of fifteen machine learning algorithms using the area under the receiver operating characteristic curve (AUC).

RESULTS

Data from 48,841 admissions were analyzed, of these 34,190 (70%) were randomly divided into training data, and 14,651 (30%) were in test data. Eight machine learning algorithms achieved very good to excellent results of AUC on test data in a of range 0.85-0.93. In prediction of short-term mortality, lactate dehydrogenase (LDH), leukocyte counts and differentials, Blood urea nitrogen (BUN) and mean corpuscular hemoglobin concentration (MCHC) were the best predictors, whereas prediction of long-term mortality was favored by age, LDH, soluble urokinase plasminogen activator receptor (suPAR), albumin, and blood urea nitrogen (BUN).

CONCLUSION

The findings suggest that measures of biomarkers taken from one blood sample during admission to the ED can identify patients at high risk of short-and long-term mortality following emergency admissions.

Longitudinal clinical data improve survival prediction after hematopoietic cell transplantation using machine learning

ABSTRACT

Serial prognostic evaluation after allogeneic hematopoietic cell transplantation (allo-HCT) might help identify patients at high risk of lethal organ dysfunction. Current prediction algorithms based on models that do not incorporate changes to patients' clinical condition after allo-HCT have limited predictive ability. We developed and validated a robust risk-prediction algorithm to predict short- and long-term survival after allo-HCT in pediatric patients that includes baseline biological variables and changes in the patients' clinical status after allo-HCT. The model was developed using clinical data from children and young adults treated at a single academic quaternary-care referral center. The model was created using a randomly split training data set (70% of the cohort), internally validated (remaining 30% of the cohort) and then externally validated on patient data from another tertiary-care referral center. Repeated clinical measurements performed from 30 days before allo-HCT to 30 days afterwards were extracted from the electronic medical record and incorporated into the model to predict survival at 100 days, 1 year, and 2 years after allo-HCT. Naïve-Bayes machine learning models incorporating longitudinal data were significantly better than models constructed from baseline variables alone at predicting whether patients would be alive or deceased at the given time points. This proof-of-concept study demonstrates that unlike traditional prognostic tools that use fixed variables for risk assessment, incorporating dynamic variability using clinical and laboratory data improves the prediction of mortality in patients undergoing allo-HCT.

Unlocking the Potential of Artificial Intelligence in Acute Myeloid Leukemia and Myelodysplastic Syndromes

PURPOSE OF THE REVIEW

This review aims to elucidate the transformative impact and potential of machine learning (ML) in the diagnosis, prognosis, and clinical management of myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML). It further aims to bridge the gap between current advances of ML and their practical application in these diseases.

RECENT FINDINGS

Recent advances in ML have revolutionized prognostication, diagnosis, and treatment of MDS and AML. ML algorithms have proven effective in predicting disease progression, optimizing treatment responses, and in the stratification of patient groups. Particularly, the use of ML in genomic and epigenomic data analysis has unveiled novel insights into the molecular heterogeneity of MDS and AML, leading to better-informed therapeutic strategies. Furthermore, deep learning techniques have shown promise in analyzing complex patterns in bone marrow biopsy images, providing a potential pathway towards early and accurate diagnosis. While still in the nascent stages, ML applications in MDS and AML signify a paradigm shift towards precision medicine. The integration of ML with traditional clinical practices could potentially enhance diagnostic accuracy, refine risk stratification, and improve therapeutic approaches. However, challenges related to data privacy, standardization, and algorithm interpretability must be addressed to realize the full potential of ML in this field. Future research should focus on the development of robust, transparent ML models and their ethical implementation in clinical settings.

Artificial intelligence methods to estimate overall mortality and non-relapse mortality following allogeneic HCT in the modern era: an EBMT-TCWP study

Allogeneic haematopoietic cell transplantation (alloHCT) has curative potential counterbalanced by its toxicity. Prognostic scores fail to include current era patients and alternative donors. We examined adult patients from the EBMT registry who underwent alloHCT between 2010 and 2019 for oncohaematological disease. Our primary objective was to develop a new prognostic score for overall mortality (OM), with a secondary objective of predicting non-relapse mortality (NRM) using the OM score. AI techniques were employed. The model for OM was trained, optimized, and validated using 70%, 15%, and 15% of the data set, respectively. The top models, "gradient boosting" for OM (AUC = 0.64) and "elasticnet" for NRM (AUC = 0.62), were selected. The analysis included 33,927 patients. In the final prognostic model, patients with the lowest score had a 2-year OM and NRM of 18 and 13%, respectively, while those with the highest score had a 2-year OM and NRM of 82 and 93%, respectively. The results were consistent in the subset of the haploidentical cohort (n = 4386). Our score effectively stratifies the risk of OM and NRM in the current era but do not significantly improve mortality prediction. Future prognostic scores can benefit from identifying biological or dynamic markers post alloHCT.

Predicting Patient Death after Allogeneic Stem Cell Transplantation for Inborn Errors Using Machine Learning (PREPAD): A European Society for Blood and Marrow Transplantation Inborn Errors Working Party Study

Allogeneic hematopoietic stem cell transplantation (HSCT) is a curative treatment for many inborn errors of immunity, metabolism, and hematopoiesis. No predictive models are available for these disorders. We created a machine learning model using XGBoost to predict survival after HSCT using European Society for Blood and Marrow Transplant registry data of 10,888 patients who underwent HSCT for inborn errors between 2006 and 2018, and compared it to a simple linear Cox model, an elastic net Cox model, and a random forest model. The XGBoost model had a cross-validated area under the curve value of .73 at 1 year, which was significantly superior to the other models, and it accurately predicted for countries excluded while training. It predicted close to 0% and >30% mortality more often than other models at 1 year, while maintaining good calibration. The 5-year survival was 94.7% in the 25% of patients at lowest risk and 62.3% in the 25% at highest risk. Within disease and donor subgroups, XGBoost outperformed the best univariate predictor. We visualized the effect of the main predictors-diagnosis, performance score, patient age and donor type-using the SHAP ML explainer and developed a stand-alone application, which can predict using the model and visualize predictions. The risk of mortality after HSCT for inborn errors can be accurately predicted using an explainable machine learning model. This exceeds the performance of models described in the literature. Doing so can help detect deviations from expected survival and improve risk stratification in trials.

Data-driven grading of acute graft-versus-host disease

Despite advances in allogeneic hematopoietic cell transplantation, acute graft-versus-host disease (aGVHD) remains its leading complication, yet with heterogeneous outcomes. Here, we analyzed aGVHD phenotypes and clinical classifications in depth in large, multicenter cohorts involving 3019 patients and addressed prevailing gaps by developing data-driven models. We compared, tested and verified these along with all conventional classifications in independent cohorts and found that data-driven grading outperformed conventional grading in Akaike information criterion and concordance index metrics. Data-driven classifications refined aGVHD assessment with up to 12 severity grades, which were associated with distinct nonrelapse mortality (NRM) and confirmed the key role of intestinal aGVHD. We developed an online calculator for physicians to implement principal component-derived grading (PC1). These results provide substantial insight into the evaluation of aGVHD phenotypes and multiorgan involvement, which relegates the exclusive reporting of overall aGVHD severity grades in transplant registries and clinical trials. Data-driven aGVHD grading provides an expandable platform to refine classification and transplant risk assessment.

A machine learning tool for the diagnosis of SARS-CoV-2 infection from hemogram parameters

Monocytes and neutrophils play key roles in the cytokine storm triggered by SARS-CoV-2 infection, which changes their conformation and function. These changes are detectable at the cellular and molecular level and may be different to what is observed in other respiratory infections. Here, we applied machine learning (ML) to develop and validate an algorithm to diagnose COVID-19 using blood parameters. In this retrospective single-center study, 49 hemogram parameters from 12,321 patients with clinical suspicion of COVID-19 and tested by RT-PCR (4239 positive and 8082 negative) were analysed. The dataset was randomly divided into training and validation sets. Blood cell parameters and patient age were used to construct the predictive model with the support vector machine (SVM) tool. The model constructed from the training set (5936 patients) achieved an accuracy for diagnosis of SARS-CoV-2 infection of 0.952 (95% CI: 0.875-0.892). Test sensitivity and specificity was 0.868 and 0.899, respectively, with a positive (PPV) and negative (NPV) predictive value of 0.896 and 0.872, respectively (prevalence 0.50). The validation set model (4964 patients) achieved an accuracy of 0.894 (95% CI: 0.883-0.903). Test sensitivity and specificity was 0.8922 and 0.8951, respectively, with a positive (PPV) and negative (NPV) predictive value of 0.817 and 0.94, respectively (prevalence 0.34). The area under the receiver operating characteristic curve was 0.952 for the algorithm performance. This algorithm may allow to rule out COVID-19 diagnosis with 94% of probability. This represents a great advance for early diagnostic orientation and guiding clinical decisions.

Machine learning applications and challenges in graft-versus-host disease: a scoping review

PURPOSE OF REVIEW

This review delves into the potential of artificial intelligence (AI), particularly machine learning (ML), in enhancing graft-versus-host disease (GVHD) risk assessment, diagnosis, and personalized treatment.

RECENT FINDINGS

Recent studies have demonstrated the superiority of ML algorithms over traditional multivariate statistical models in donor selection for allogeneic hematopoietic stem cell transplantation. ML has recently enabled dynamic risk assessment by modeling time-series data, an upgrade from the static, "snapshot" assessment of patients that conventional statistical models and older ML algorithms offer. Regarding diagnosis, a deep learning model, a subset of ML, can accurately identify skin segments affected with chronic GVHD with satisfactory results. ML methods such as Q-learning and deep reinforcement learning have been utilized to develop adaptive treatment strategies (ATS) for the personalized prevention and treatment of acute and chronic GVHD.

SUMMARY

To capitalize on these promising advancements, there is a need for large-scale, multicenter collaborations to develop generalizable ML models. Furthermore, addressing pertinent issues such as the implementation of stringent ethical guidelines is crucial before the widespread introduction of AI into GVHD care.

What Influences the Decision to Proceed to Transplant for Patients With AML in First Remission?

Although allogeneic hematopoietic cell transplantation (allo-HCT) remains the backbone of curative treatment for the majority of fit adults diagnosed with AML, there is indeed a subset of patients for whom long-term remission may be achieved without transplantation. Remarkable changes in our knowledge of AML biology in recent years has transformed the landscape of diagnosis, management, and treatment of AML. Specifically, markedly increased understanding of molecular characteristics of AML, the expanded application of minimal/measurable residual diseases testing, and an increased armamentarium of leukemia-directed therapeutic agents have created a new paradigm for the medical care of patients with AML. An attempt is herein made to decipher the decision to proceed to transplant for patients with AML in first complete remission on the basis of the current best available evidence. The focus is on factors affecting the biology and treatment of AML itself, rather than on variables related to allo-HCT, an area characterized by significant advancements that have reduced overall therapy-related complications. This review seeks to focus on areas of particular complexity, while simultaneously providing clarity on how our current knowledge and treatment strategies may, or may not, influence the decision to pursue allo-HCT in patients with AML.

Application of the CIBMTR One Year Survival Outcomes Calculator as a tool for retrospective analysis

The Web-based One Year Survival Outcomes Calculator developed by the Center for International Blood and Marrow Transplant Research (CIBMTR) applies large-scale registry data to generate individualized estimates of overall survival (OS) probability 1 year after first allogeneic hematopoietic cell transplant (HCT) and can therefore provide a data-driven foundation for personalized patient counseling. We assessed the calibration of the CIBMTR One Year Survival Outcomes Calculator when applied to retrospective data among adult recipients of first allogeneic HCT for acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), or myelodysplastic syndrome (MDS) with peripheral blood stem cell transplant (PBSCT) from a 7/8- or 8/8-matched donor from 2000 through 2015 at a single center. Predicted 1 year OS was estimated for each patient using the CIBMTR Calculator. Corresponding observed 1 year OS was estimated for each group by the Kaplan-Meier method. A weighted Kaplan-Meier estimator was used to visually display the average of observed 1 year survival estimates over the continuous range of predicted OS. In the first analysis of its kind, we demonstrated that the CIBMTR One Year Survival Outcomes Calculator could be applied to larger patient cohorts and predicted 1 year prognosis with general agreement between predicted and observed survival.

Machine Learning for the Prediction of Survival Post-Allogeneic Hematopoietic Cell Transplantation: A Single-Center Experience

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.

Deep learning-enabled quantification of simultaneous PET/MRI for cell transplantation monitoring

Current methods of in vivo imaging islet cell transplants for diabetes using magnetic resonance imaging (MRI) are limited by their low sensitivity. Simultaneous positron emission tomography (PET)/MRI has greater sensitivity and ability to visualize cell metabolism. However, this dual-modality tool currently faces two major challenges for monitoring cells. Primarily, the dynamic conditions of PET such as signal decay and spatiotemporal change in radioactivity prevent accurate quantification of the transplanted cell number. In addition, selection bias from different radiologists renders human error in segmentation. This calls for the development of artificial intelligence algorithms for the automated analysis of PET/MRI of cell transplantations. Here, we combined K-means++ for segmentation with a convolutional neural network to predict radioactivity in cell-transplanted mouse models. This study provides a tool combining machine learning with a deep learning algorithm for monitoring islet cell transplantation through PET/MRI. It also unlocks a dynamic approach to automated segmentation and quantification of radioactivity in PET/MRI.

Matched related versus unrelated versus haploidentical donors for allogeneic transplantation in AML patients achieving first complete remission after two induction courses: a study from the ALWP/EBMT

We compared transplants (HSCT) from matched related siblings (MSD) with those from matched 10/10 and mismatched 9/10 unrelated (UD) and T-replete haploidentical (Haplo) donors in acute myeloid leukemia (AML) in first complete remission (CR1) achieved after two inductions, a known poor prognostic factor. One thousand two hundred and ninety-five patients were included: MSD (n = 428), UD 10/10 (n = 554), UD 9/10 (n = 135), and Haplo (n = 178). Acute GVHD II-IV was higher in all groups compared to MSD. Extensive chronic (c) GVHD was significantly higher in UD 9/10 (HR = 2.52; 95% CI 1.55-4.11, p = 0.0002) and UD 10/10 (HR = 1.48; 95% CI 1.03-2.13, p = 0.036) and cGVHD all grades were higher in UD 9/10 vs MSD (HR = 1.77; 95% CI 1.26-2.49, p = 0.0009). Non-relapse mortality was higher in all groups compared to MSD. Relapse incidence, leukemia-free, and overall survival did not differ significantly between donor types. Finally, GVHD-free relapse-free survival was lower in HSCT from UD 9/10 (HR = 1.56, 95% CI 1.20-2.03, p = 0.0009) but not in those from UD 10/10 (HR = 1.13, p = 0.22) and Haplo donors (HR = 1.12, p = 0.43) compared to MSD. In conclusion, in AML patients undergoing HSCT in CR1 achieved after two induction courses 10/10 UD and Haplo but not 9/10 UD donors are comparable alternatives to MSD.

A convolutional neural network-based model that predicts acute graft-versus-host disease after allogeneic hematopoietic stem cell transplantation

BACKGROUND

Forecasting acute graft-versus-host disease (aGVHD) after allogeneic hematopoietic stem cell transplantation (HSCT) is highly challenging with conventional statistical techniques due to complex parameters and their interactions. The primary object of this study was to establish a convolutional neural network (CNN)-based prediction model for aGVHD.

METHOD

We analyzed adult patients who underwent allogeneic HSCT between 2008 and 2018, using the Japanese nationwide registry database. The CNN algorithm, equipped with a natural language processing technique and an interpretable explanation algorithm, was applied to develop and validate prediction models.

RESULTS

Here, we evaluate 18,763 patients between 16 and 80 years of age (median, 50 years). In total, grade II-IV and grade III-IV aGVHD is observed among 42.0% and 15.6%. The CNN-based model eventually allows us to calculate a prediction score of aGVHD for an individual case, which is validated to distinguish the high-risk group of aGVHD in the test cohort: cumulative incidence of grade III-IV aGVHD at Day 100 after HSCT is 28.8% for patients assigned to a high-risk group by the CNN model, compared to 8.4% among low-risk patients (hazard ratio, 4.02; 95% confidence interval, 2.70-5.97; p < 0.01), suggesting high generalizability. Furthermore, our CNN-based model succeeds in visualizing the learning process. Moreover, contributions of pre-transplant parameters other than HLA information to the risk of aGVHD are determined.

CONCLUSIONS

Our results suggest that CNN-based prediction provides a faithful prediction model for aGVHD, and can serve as a valuable tool for decision-making in clinical practice.

Nomogram for Predicting Early Mortality after Umbilical Cord Blood Transplantation in Children with Inborn Errors of Immunity

PURPOSE

Pediatric patients with inborn errors of immunity (IEI) undergoing umbilical cord blood transplantation (UCBT) are at risk of early mortality. Our aim was to develop and validate a prediction model for early mortality after UCBT in pediatric IEI patients based on pretransplant factors.

METHODS

Data from 230 pediatric IEI patients who received their first UCBT between 2014 and 2021 at a single center were analyzed retrospectively. Data from 2014-2019 and 2020-2021 were used as training and validation sets, respectively. The primary outcome of interest was early mortality. Machine learning algorithms were used to identify risk factors associated with early mortality and to build predictive models. The model with the best performance was visualized using a nomogram. Discriminative ability was measured using the area under the curve (AUC) and decision curve analysis.

RESULTS

Fifty days was determined as the cutoff for distinguishing early mortality in pediatric IEI patients undergoing UCBT. Of the 230 patients, 43 (18.7%) suffered early mortality. Multivariate logistic regression with pretransplant albumin, CD4 (absolute count), elevated C-reactive protein, and medical history of sepsis showed good discriminant AUC values of 0.7385 (95% CI, 0.5824-0.8945) and 0.827 (95% CI, 0.7409-0.9132) in predicting early mortality in the validation and training sets, respectively. The sensitivity and specificity were 0.5385 and 0.8154 for validation and 0.7667 and 0.7705 for training, respectively. The final model yielded net benefits across a reasonable range of risk thresholds.

CONCLUSION

The developed nomogram can predict early mortality in pediatric IEI patients undergoing UCBT.

Fit for Duty: Lessons Learned from Outpatient and Homebound Hematopoietic Cell Transplantation to Prepare Family Caregivers for Home-Based Care

In the past decade, the demand for home-based care has been amplified by the Coronavirus disease 2019 pandemic. Home-based care has significant benefits for patients, their families, and healthcare systems, but it relies on the often-invisible workforce of family and friend caregivers who shoulder essential health care responsibilities, frequently with inadequate training and support. Hematopoietic cell transplantation (HCT), a potentially curative but intensive treatment for many patients with blood disorders, is being increasingly offered in home-based care settings and necessitates the involvement of family caregivers for significant patient care responsibilities. However, guidelines for supporting and preparing HCT caregivers to effectively care for their loved ones at home have not yet been established. Here, informed by the literature and our collective experience as clinicians and researchers who care for diverse patients with hematologic malignancies undergoing HCT, we provide considerations and recommendations to better support and prepare family caregivers in home-based HCT and, by extension, family caregivers supporting patients with other serious illnesses at home. We suggest tangible ways to screen family caregivers for distress and care delivery challenges, educate and train them to prepare for their caregiving role, and create an infrastructure of support for family caregivers within this emerging care delivery model.

Applicability of the acute leukemia (AL) - EBMT score as a prognostic model for allogeneic hematopoietic stem cell transplantation: a single-center, prospective, cohort study at a reference transplant center in Brazil

INTRODUCTION

The Acute Leukemia-European Society for Blood and Marrow Transplantation (AL-EBMT) risk score was recently developed and validated by Shouval et al. OBJECTIVE: To assess the ability of this score in predicting the 2-year overall survival (OS-2), leukemia-free survival (LFS-2) and transplant-related mortality (TRM) in acute leukemia (AL) adult patients undergoing a first allogeneic hematopoietic stem cell transplant (HSCT) at a transplant center in Brazil.

METHODS

In this prospective, cohort study, we used the formula published by Shouval et al. to calculate the AL-EBMT score and stratify patients into three risk categories.

RESULTS

A total of 79 patients transplanted between 2008 and 2018 were analyzed. The median age was 38 years. Acute myeloid leukemia was the most common diagnosis (68%). Almost a quarter of the cases were at an advanced stage. All hematopoietic stem cell transplantations (HSCTs) were human leukocyte antigen-matched (HLA-matched) and the majority used familial donors (77%). Myeloablative conditioning was used in 92% of the cases. Stratification according to the AL-EBMT score into low-, intermediate- and high-risk groups yielded the following results: 40%, 12% and 47% of the cases, respectively. The high scoring group was associated with a hazard ratio of 2.1 (p = 0.007), 2.1 (p = 0.009) and 2.47 (p = 0.01) for the 2-year OS, LFS and TRM, respectively.

CONCLUSION

This study supports the ability of the AL-EBMT score to reasonably predict the 2-year post-transplant OS, LFS and TRM and to discriminate between risk categories in adult patients with AL, thus confirming its usefulness in clinical decision-making in this setting. Larger, multicenter studies may further help confirm these findings.

Time-dependent prediction of mortality and cytomegalovirus reactivation after allogeneic hematopoietic cell transplantation using machine learning

Allogeneic hematopoietic cell transplantation (HCT) effectively treats high-risk hematologic diseases but can entail HCT-specific complications, which may be minimized by appropriate patient management, supported by accurate, individual risk estimation. However, almost all HCT risk scores are limited to a single risk assessment before HCT without incorporation of additional data. We developed machine learning models that integrate both baseline patient data and time-dependent laboratory measurements to individually predict mortality and cytomegalovirus (CMV) reactivation after HCT at multiple time points per patient. These gradient boosting machine models provide well-calibrated, time-dependent risk predictions and achieved areas under the receiver-operating characteristic of 0.92 and 0.83 and areas under the precision-recall curve of 0.58 and 0.62 for prediction of mortality and CMV reactivation, respectively, in a 21-day time window. Both models were successfully validated in a prospective, non-interventional study and performed on par with expert hematologists in a pilot comparison.

Survival Prediction of Children Undergoing Hematopoietic Stem Cell Transplantation Using Different Machine Learning Classifiers by Performing Chi-Square Test and Hyperparameter Optimization: A Retrospective Analysis

Bone marrow transplant (BMT) is an effective surgical treatment for bone marrow-related disorders. However, several associated risk factors can impair long-term survival after BMT. Machine learning (ML) technologies have been proven useful in survival prediction of BMT receivers along with the influences that limit their resilience. In this study, an efficient classification model predicting the survival of children undergoing BMT is presented using a public dataset. Several supervised ML methods were investigated in this regard with an 80-20 train-test split ratio. To ensure prediction with minimal time and resources, only the top 11 out of the 59 dataset features were considered using Chi-square feature selection method. Furthermore, hyperparameter optimization (HPO) using the grid search cross-validation (GSCV) technique was adopted to increase the accuracy of prediction. Four experiments were conducted utilizing a combination of default and optimized hyperparameters on the original and reduced datasets. Our investigation revealed that the top 11 features of HPO had the same prediction accuracy (94.73%) as the entire dataset with default parameters, however, requiring minimal time and resources. Hence, the proposed approach may aid in the development of a computer-aided diagnostic system with satisfactory accuracy and minimal computation time by utilizing medical data records.

The Clinical Value of Procalcitonin in the Neutropenic Period After Allogeneic Hematopoietic Stem Cell Transplantation

The diagnostic value of procalcitonin and the prognostic role of PCT clearance remain unclear in neutropenic period after allogeneic hematopoietic stem cell transplantation introduction. This study evaluated 219 febrile neutropenic patients (116, retrospectively; 103, prospectively) who underwent allo-HSCT from April 2014 to March 2016. The area under the receiver operator characteristic curve (AUC) of PCT for detecting documented infection (DI) was 0.637, and that of bloodstream infection (BSI) was 0.811. In multivariate analysis, the inability to decrease PCT by more than 80% within 5–7 days after the onset of fever independently predicted poor 100-day survival following allo-HSCT (P = 0.036). Furthermore, the prognostic nomogram combining PCTc and clinical parameters showed a stable predictive performance, supported by the C-index of 0.808 and AUC of 0.813 in the primary cohort, and C-index of 0.691 and AUC of 0.697 in the validation cohort. This study demonstrated the diagnostic role of PCT in documented and bloodstream infection during the neutropenic period after allo-HSCT. PCTc might serve as a predictive indicator of post-HSCT 100-day mortality. A nomogram based on PCTc and several clinical factors effectively predicted the 100-day survival of febrile patients and may help physicians identify high-risk patients in the post-HSCT neutropenic period.

Warfarin sensitivity is associated with increased hospital mortality in critically Ill patients

Background

Warfarin is a widely used anticoagulant with a narrow therapeutic index and large interpatient variability in the therapeutic dose. Warfarin sensitivity has been reported to be associated with increased incidence of international normalized ratio (INR) > 5. However, whether warfarin sensitivity is a risk factor for adverse outcomes in critically ill patients remains unknown. In the present study, we aimed to evaluate the utility of different machine learning algorithms for the prediction of warfarin sensitivity and to determine the impact of warfarin sensitivity on outcomes in critically ill patients.

Methods

Nine different machine learning algorithms for the prediction of warfarin sensitivity were tested in the International Warfarin Pharmacogenetic Consortium cohort and Easton cohort. Furthermore, a total of 7,647 critically ill patients was analyzed for warfarin sensitivity on in-hospital mortality by multivariable regression. Covariates that potentially confound the association were further adjusted using propensity score matching or inverse probability of treatment weighting.

Results

We found that logistic regression (AUC = 0.879, 95% CI: 0.834–0.924) was indistinguishable from support vector machine with a linear kernel, neural network, AdaBoost and light gradient boosting trees, and significantly outperformed all the other machine learning algorithms. Furthermore, we found that warfarin sensitivity predicted by the logistic regression model was significantly associated with worse in-hospital mortality in critically ill patients with an odds ratio (OR) of 1.33 (95% CI, 1.01–1.77).

Conclusions

Our data suggest that the logistic regression model is the best model for the prediction of warfarin sensitivity clinically and that warfarin sensitivity is likely to be a risk factor for adverse outcomes in critically ill patients.

Predicting Long-term Survival After Allogeneic Hematopoietic Cell Transplantation in Patients With Hematologic Malignancies: Machine Learning-Based Model Development and Validation

Background

Scoring systems developed for predicting survival after allogeneic hematopoietic cell transplantation (HCT) show suboptimal prediction power, and various factors affect posttransplantation outcomes.

Objective

A prediction model using a machine learning–based algorithm can be an alternative for concurrently applying multiple variables and can reduce potential biases. In this regard, the aim of this study is to establish and validate a machine learning–based predictive model for survival after allogeneic HCT in patients with hematologic malignancies.

Methods

Data from 1470 patients with hematologic malignancies who underwent allogeneic HCT between December 1993 and June 2020 at Asan Medical Center, Seoul, South Korea, were retrospectively analyzed. Using the gradient boosting machine algorithm, we evaluated a model predicting the 5-year posttransplantation survival through 10-fold cross-validation.

Results

The prediction model showed good performance with a mean area under the receiver operating characteristic curve of 0.788 (SD 0.03). Furthermore, we developed a risk score predicting probabilities of posttransplantation survival in 294 randomly selected patients, and an agreement between the estimated predicted and observed risks of overall death, nonrelapse mortality, and relapse incidence was observed according to the risk score. Additionally, the calculated score demonstrated the possibility of predicting survival according to the different transplantation-related factors, with the visualization of the importance of each variable.

Conclusions

We developed a machine learning–based model for predicting long-term survival after allogeneic HCT in patients with hematologic malignancies. Our model provides a method for making decisions regarding patient and donor candidates or selecting transplantation-related resources, such as conditioning regimens.

A systematic review of the predicted outcomes related to hematopoietic stem cell transplantation: focus on applied machine learning methods' performance

INTRODUCTION

: Hematopoietic stem cell transplantation (HSCT) is a critical therapeutic procedure in blood diseases, and the investigation of HSCT data can provide valuable information. Machine learning (ML) techniques are novel and useful data analysis tools that have been applied in many studies to predict HSCT survival and estimate the risk of transplantation.

AREAS COVERED

: A systematic review was performed with a search of PubMed, Science Direct, Embase, Scopus, and the European Society for Blood and Marrow Transplantation, the Center for International Blood and Marrow Transplant Research, and the American Society for Transplantation and Cellular Therapy publications for articles published by September 2020.

EXPERT OPINION

: After investigating the results, 24 papers that met eligibility criteria were included in this study. The applied ML algorithms with the highest performance were Random Survival Forests (AUC=0.72) for survival-related, Random Survival Forests and Logistic Regression (AUC=0.77) for mortality-related, Deep Learning (AUC=0.8) for relapse, L2-Regularized Logistic Regression (AUC=0.66) for Acute-Graft Versus Host Disease, Random Survival Forests (AUC=0.88) for sepsis, Elastic-Net Regression (AUC=0.89) for cognitive impairment, and Bayesian Network (AUC=0.997) for oral mucositis outcome. This review reveals the potential of ML techniques to predict HSCT outcomes and apply them to developing clinical decision support systems.

Prediction of Tacrolimus Dose/Weight-Adjusted Trough Concentration in Pediatric Refractory Nephrotic Syndrome: A Machine Learning Approach

Purpose

Tacrolimus (TAC) is a first-line immunosuppressant for patients with refractory nephrotic syndrome (NS). However, there is a high inter-patient variability of TAC pharmacokinetics, thus therapeutic drug monitoring (TDM) is required. In this study, we aimed to employ machine learning algorithms to investigate the impact of clinical and genetic variables on the TAC dose/weight-adjusted trough concentration (C₀/D) in Chinese children with refractory NS, and then develop and validate the TAC C₀/D prediction models.

Patients and Methods

The association of 82 clinical variables and 244 single nucleotide polymorphisms (SNPs) with TAC C₀/D in the third month since TAC treatment was examined in 171 children with refractory NS. Extremely randomized trees (ET), gradient boosting decision tree (GBDT), random forest (RF), extreme gradient boosting (XGBoost), and Lasso regression were carried out to establish and validate prediction models, respectively. The best prediction models were validated on a cohort of 30 refractory NS patients.

Results

GBDT algorithm performed best in the whole group (R²=0.444, MSE=591.032, MAE=20.782, MedAE=18.980) and CYP3A5 nonexpresser group (R²=0.264, MSE=477.948, MAE=18.119, MedAE=18.771), while ET algorithm performed best in the CYP3A5 expresser group (R²=0.380, MSE=1839.459, MAE=31.257, MedAE=19.399). These prediction models included 3 clinical variables (ALB0, AGE0, and gender) and 10 SNPs (ACTN4 rs3745859, ACTN4 rs56113315, ACTN4 rs62121818, CTLA4 rs4553808, CYP3A5 rs776746, IL2RA rs12722489, INF2 rs1128880, MAP3K11 rs7946115, MYH9 rs2239781, and MYH9 rs4821478).

Conclusion

The association between the clinical and genetic variables and TAC C₀/D was described, and three TAC C₀/D prediction models integrating clinical and genetic variables were developed and validated using machine learning, which may support individualized TAC dosing.

Potential applications and performance of machine learning techniques and algorithms in clinical practice: A systematic review

PURPOSE

The advent of clinically adapted machine learning algorithms can solve numerous problems ranging from disease diagnosis and prognosis to therapy recommendations. This systematic review examines the performance of machine learning (ML) algorithms and evaluates the progress made to date towards their implementation in clinical practice.

METHODS

Systematic searching of databases (PubMed, MEDLINE, Scopus, Google Scholar, Cochrane Library and WHO Covid-19 database) to identify original articles published between January 2011 and October 2021. Studies reporting ML techniques in clinical practice involving humans and ML algorithms with a performance metric were considered.

RESULTS

Of 873 unique articles identified, 36 studies were eligible for inclusion. The XGBoost (extreme gradient boosting) algorithm showed the highest potential for clinical applications (n = 7 studies); this was followed jointly by random forest algorithm, logistic regression, and the support vector machine, respectively (n = 5 studies). Prediction of outcomes (n = 33), in particular Inflammatory diseases (n = 7) received the most attention followed by cancer and neuropsychiatric disorders (n = 5 for each) and Covid-19 (n = 4). Thirty-three out of the thirty-six included studies passed more than 50% of the selected quality assessment criteria in the TRIPOD checklist. In contrast, none of the studies could achieve an ideal overall bias rating of 'low' based on the PROBAST checklist. In contrast, only three studies showed evidence of the deployment of ML algorithm(s) in clinical practice.

CONCLUSIONS

ML is potentially a reliable tool for clinical decision support. Although advocated widely in clinical practice, work is still in progress to validate clinically adapted ML algorithms. Improving quality standards, transparency, and interpretability of ML models will further lower the barriers to acceptability.

Conditioning Regimens for Frail Patients with Acute Leukemia Undergoing Allogeneic Stem Cell Transplant: How to Strike Gently

Despite the recent dramatic progress in acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL) therapy, allogeneic transplant remains a mainstay of treatment for patients with acute leukemia. The availability of novel compounds and low intensity chemotherapy regimens made it possible for a significant proportion of elderly and comorbid patients with AML or ALL to undergo curative treatment protocols. In addition, the expansion of donor availability and the recent dramatic progress in haploidentical stem cell transplant, allow the identification of an available donor for nearly every patient. Therefore, an increasing number of transplants are currently performed in elderly and frail patients with AML or ALL. However, allo-Hematopoietic stem cell transplant (HSCT) in this delicate setting represents an important challenge, especially regarding the selection of the conditioning protocol. Ideally, conditioning intensity should be reduced as much as possible; however, in patients with acute leukemia relapse remains the major cause of transplant failure. In this article we present modern tools to assess the patient health status before transplant, review the available data on the outcome of frail AML an ALL patients undergoing allo-HSCT, and discuss how preparatory regimens can be optimized in this setting.

Establishment of a Predictive Model for GvHD-free, Relapse-free Survival after Allogeneic HSCT using Ensemble Learning

Stacked ensemble of machine-learning algorithms could establish more accurate prediction model for survival analysis than existing methods.

Stacked ensemble model can be applied to personalized prediction of HSCT outcomes from pretransplant characteristics.

Graft-versus-host disease-free, relapse-free survival (GRFS) is a useful composite end point that measures survival without relapse or significant morbidity after allogeneic hematopoietic stem cell transplantation (allo-HSCT). We aimed to develop a novel analytical method that appropriately handles right-censored data and competing risks to understand the risk for GRFS and each component of GRFS. This study was a retrospective data-mining study on a cohort of 2207 adult patients who underwent their first allo-HSCT within the Kyoto Stem Cell Transplantation Group, a multi-institutional joint research group of 17 transplantation centers in Japan. The primary end point was GRFS. A stacked ensemble of Cox Proportional Hazard (Cox-PH) regression and 7 machine-learning algorithms was applied to develop a prediction model. The median age for the patients was 48 years. For GRFS, the stacked ensemble model achieved better predictive accuracy evaluated by C-index than other state-of-the-art competing risk models (ensemble model: 0.670; Cox-PH: 0.668; Random Survival Forest: 0.660; Dynamic DeepHit: 0.646). The probability of GRFS after 2 years was 30.54% for the high-risk group and 40.69% for the low-risk group (hazard ratio compared with the low-risk group: 2.127; 95% CI, 1.19-3.80). We developed a novel predictive model for survival analysis that showed superior risk stratification to existing methods using a stacked ensemble of multiple machine-learning algorithms.

Comparison of Machine Learning Techniques for Mortality Prediction in a Prospective Cohort of Older Adults

As global demographics change, ageing is a global phenomenon which is increasingly of interest in our modern and rapidly changing society. Thus, the application of proper prognostic indices in clinical decisions regarding mortality prediction has assumed a significant importance for personalized risk management (i.e., identifying patients who are at high or low risk of death) and to help ensure effective healthcare services to patients. Consequently, prognostic modelling expressed as all-cause mortality prediction is an important step for effective patient management. Machine learning has the potential to transform prognostic modelling. In this paper, results on the development of machine learning models for all-cause mortality prediction in a cohort of healthy older adults are reported. The models are based on features covering anthropometric variables, physical and lab examinations, questionnaires, and lifestyles, as well as wearable data collected in free-living settings, obtained for the "Healthy Ageing Initiative" study conducted on 2291 recruited participants. Several machine learning techniques including feature engineering, feature selection, data augmentation and resampling were investigated for this purpose. A detailed empirical comparison of the impact of the different techniques is presented and discussed. The achieved performances were also compared with a standard epidemiological model. This investigation showed that, for the dataset under consideration, the best results were achieved with Random UnderSampling in conjunction with Random Forest (either with or without probability calibration). However, while including probability calibration slightly reduced the average performance, it increased the model robustness, as indicated by the lower 95% confidence intervals. The analysis showed that machine learning models could provide comparable results to standard epidemiological models while being completely data-driven and disease-agnostic, thus demonstrating the opportunity for building machine learning models on health records data for research and clinical practice. However, further testing is required to significantly improve the model performance and its robustness.

Exploring the Current Trends of Artificial Intelligence in Stem Cell Therapy: A Systematic Review

The concept of healing in medicine has been taking a new form where scientists and researchers are in pursuance of regenerative medicine. Until now, doctors have "reacted" to disease by treating the symptoms; however, modern medicine is transforming toward regeneration rather than reactive treatment, which is where stem cell therapy comes into the play-the concept of replacing damaged cells with brand new cells that perform the same function better. Stem cell treatment is currently being used to treat autoimmune, inflammatory, neurological, orthopedic, and traumatic disorders, with various research being undertaken for a wide range of diseases. It could also be the answer to anti-aging and a disease-free state. Despite the benefits, numerous errors could prevail in treating patients with stem cells. With the advancement of technology and research in the modern period, medicine is beginning to turn to artificial intelligence (AI) to address the complicated errors that could occur in regenerative medicine. For successful treatment, one must achieve precision and accuracy when analyzing healthy and productive stem cells that possess all the properties of a native cell. This review intends to discuss and study the application of AI in stem cell therapy and how it influences how medicine is practiced, thus creating a path to a regenerative future with negligible adverse effects. The following databases were used for a literature search: PubMed, Google Scholar, PubMed Central, and Institute of Electrical and Electronics Engineers (IEEE) Xplore. After a thorough analysis, studies were chosen, keeping in mind the inclusion and exclusion criteria set by the authors of this review, which comprised reports published within the last six years in the English language. The authors also made sure to include studies that sufficed the quality of each report assessed using appropriate quality appraisal tools, after which eight reports were found to be eligible and were included in this review. This research mainly revolves around machine learning, deep neural networks (DNN), and other subclasses of AI encompassed in these categories. While there are concerns and limitations in implementing various mediums of AI in stem cell therapy, the analysis of the eligible studies concluded that artificial intelligence provides significant benefits to the global healthcare ecosystem in numerous ways, such as determining the viability, functionality, biosafety, and bioefficacy of stem cells, as well as appropriate patient selection. Applying AI to this novelty brings out the precision, accuracy, and a revolution in regenerative medicine. In addition, stem cell therapy is not currently FDA approved (except for the blood-forming stem cells) and, to date, is considered experimental with no clear outline of risks and benefits. Given this limitation, studies are conducted regularly around the world in hopes for a concrete conclusion where technological advances such as AI could help in shaping the future of regenerative medicine.

GITMO Registry Study on Allogeneic Transplantation in Patients Aged ≥60 Years from 2000 to 2017: Improvements and Criticisms

Today, allogeneic stem cell transplantation (allo-SCT) can be offered to patients up to age 70 to 72 years and represents one of the most effective curative treatments for many hematologic malignancies. The primary objective of the study was to collect data from the allo-SCTs performed in Italy between 2000 and 2017 in patients aged ≥60 years to evaluate the changes in safety and efficacy outcomes, as well as their distribution and characteristics over time. The Italian Group for Bone Marrow Transplantation, Hematopoietic Stem Cells and Cell Therapy (GITMO) AlloEld study (ClinicalTrials.gov identifier NCT04469985) is a retrospective analysis of allo-SCTs performed at 30 Italian transplantation centers in older patients (age ≥60 years) between 2000 and 2017 (n = 1996). For the purpose of this analysis, patients were grouped into 3 time periods: time A, 2000 to 2005 (n = 256; 12%); time B, 2006 to 2011 (n = 584; 29%); and time C, 2012 to 2017 (n = 1156; 59%). After a median follow-up of 5.6 years, the 5-year nonrelapse mortality (NRM) remained stable (time A, 32.8%; time B, 36.2%; and time C, 35.0%; P = .5), overall survival improved (time A, 28.4%; time B, 31.8%; and time C, 37.3%; P = .012), and the cumulative incidence of relapse was reduced (time A, 45.3%; time B, 38.2%; time C, 30.0%; P < .0001). The 2-year incidence of extensive chronic graft-versus-host disease was reduced significantly (time A, 17.2%; time B, 15.8%; time C, 12.2%; P = .004). Considering times A and B together (2000 to 2011), the 2-year NRM was positively correlated with the Hematopoietic Cell Transplantation Comorbidity Index (HCT-CI) score; NRM was 25.2% in patients with an HCT-CI score of 0, 33.9% in those with a score of 1 or 2, and 36.1% in those with a score of 3 (P < .001). However, after 2012, the HCT-CI score was not significantly predictive of NRM. This study shows that the transplantation procedure in elderly patients became more effective over time. Relapse incidence remains the major problem, and strategies to prevent it are currently under investigation (eg, post-transplantation maintenance). The selection of patients aged ≥60 could be improved by combining HCT-CI and frailty assessment to better predict NRM.

A novel Iowa-Mayo validated composite risk assessment tool for allogeneic stem cell transplantation survival outcome prediction

Allogeneic hematopoietic stem cell transplantation (HSCT) is a curative option for many hematologic conditions and is associated with considerable morbidity and mortality. Therefore, prognostic tools are essential to navigate the complex patient, disease, donor, and transplant characteristics that differentially influence outcomes. We developed a novel, comprehensive composite prognostic tool. Using a lasso-penalized Cox regression model (n = 273), performance status, HCT-CI, refined disease-risk index (rDRI), donor and recipient CMV status, and donor age were identified as predictors of disease-free survival (DFS). The results for overall survival (OS) were similar except for recipient CMV status not being included in the model. Models were validated in an external dataset (n = 378) and resulted in a c-statistic of 0.61 and 0.62 for DFS and OS, respectively. Importantly, this tool incorporates donor age as a variable, which has an important role in HSCT outcomes. This needs to be further studied in prospective models. An easy-to-use and a web-based nomogram can be accessed here: https://allohsctsurvivalcalc.iowa.uiowa.edu/ .

Machine Learning Applications in the Diagnosis of Benign and Malignant Hematological Diseases

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.

Revisiting performance metrics for prediction with rare outcomes

Machine learning algorithms are increasingly used in the clinical literature, claiming advantages over logistic regression. However, they are generally designed to maximize the area under the receiver operating characteristic curve. While area under the receiver operating characteristic curve and other measures of accuracy are commonly reported for evaluating binary prediction problems, these metrics can be misleading. We aim to give clinical and machine learning researchers a realistic medical example of the dangers of relying on a single measure of discriminatory performance to evaluate binary prediction questions. Prediction of medical complications after surgery is a frequent but challenging task because many post-surgery outcomes are rare. We predicted post-surgery mortality among patients in a clinical registry who received at least one aortic valve replacement. Estimation incorporated multiple evaluation metrics and algorithms typically regarded as performing well with rare outcomes, as well as an ensemble and a new extension of the lasso for multiple unordered treatments. Results demonstrated high accuracy for all algorithms with moderate measures of cross-validated area under the receiver operating characteristic curve. False positive rates were <1%, however, true positive rates were <7%, even when paired with a 100% positive predictive value, and graphical representations of calibration were poor. Similar results were seen in simulations, with the addition of high area under the receiver operating characteristic curve (>90%) accompanying low true positive rates. Clinical studies should not primarily report only area under the receiver operating characteristic curve or accuracy.

A Review on the Role of Artificial Intelligence in Stem Cell Therapy: An Initiative for Modern Medicines

Stem Cells (SCs) show a wide range of applications in the treatment of numerous diseases, including neurodegenerative diseases, diabetes, cardiovascular diseases, cancer, etc. SC related research has gained popularity owing to the unique characteristics of self-renewal and differentiation. Artificial Intelligence (AI), an emerging field of computer science and engineering, has shown potential applications in different fields like robotics, agriculture, home automation, healthcare, banking, and transportation since its invention. This review aims to describe the various applications of AI in SC biology, including understanding the behavior of SCs, recognizing individual cell type before undergoing differentiation, characterization of SCs using mathematical models and prediction of mortality risk associated with SC transplantation. This review emphasizes the role of neural networks in SC biology and further elucidates the concepts of machine learning and deep learning and their applications in SC research.

The future of bone regeneration: integrating AI into tissue engineering

Tissue engineering is a branch of regenerative medicine that harnesses biomaterial and stem cell research to utilise the body's natural healing responses to regenerate tissue and organs. There remain many unanswered questions in tissue engineering, with optimal biomaterial designs still to be developed and a lack of adequate stem cell knowledge limiting successful application. Advances in artificial intelligence (AI), and deep learning specifically, offer the potential to improve both scientific understanding and clinical outcomes in regenerative medicine. With enhanced perception of how to integrate artificial intelligence into current research and clinical practice, AI offers an invaluable tool to improve patient outcome.

Accuracy and Efficiency of Deep-Learning-Based Automation of Dual Stain Cytology in Cervical Cancer Screening

Background

With the advent of primary human papillomavirus testing followed by cytology for cervical cancer screening, visual interpretation of cytology slides remains the last subjective analysis step and suffers from low sensitivity and reproducibility.

Methods

We developed a cloud-based whole-slide imaging platform with a deep-learning classifier for p16/Ki-67 dual-stained (DS) slides trained on biopsy-based gold standards. We compared it with conventional Pap and manual DS in 3 epidemiological studies of cervical and anal precancers from Kaiser Permanente Northern California and the University of Oklahoma comprising 4253 patients. All statistical tests were 2-sided.

Results

In independent validation at Kaiser Permanente Northern California, artificial intelligence (AI)-based DS had lower positivity than cytology (P < .001) and manual DS (P < .001) with equal sensitivity and substantially higher specificity compared with both Pap (P < .001) and manual DS (P < .001), respectively. Compared with Pap, AI-based DS reduced referral to colposcopy by one-third (41.9% vs 60.1%, P < .001). At a higher cutoff, AI-based DS had similar performance to high-grade squamous intraepithelial lesions cytology, indicating a risk high enough to allow for immediate treatment. The classifier was robust, showing comparable performance in 2 cytology systems and in anal cytology.

Conclusions

Automated DS evaluation removes the remaining subjective component from cervical cancer screening and delivers consistent quality for providers and patients. Moving from Pap to automated DS substantially reduces the number of colposcopies and also achieves excellent performance in a simulated fully vaccinated population. Through cloud-based implementation, this approach is globally accessible. Our results demonstrate that AI not only provides automation and objectivity but also delivers a substantial benefit for women by reduction of unnecessary colposcopies.

Allogeneic Stem Cell Transplantation for Acute Myeloid Leukemia: Who, When, and How?

Although the majority of patients with acute myeloid leukemia (AML) treated with intensive chemotherapy achieve a complete remission (CR), many are destined to relapse if treated with intensive chemotherapy alone. Allogeneic stem cell transplant (allo-SCT) represents a pivotally important treatment strategy in fit adults with AML because of its augmented anti-leukemic activity consequent upon dose intensification and the genesis of a potent graft-versus-leukemia effect. Increased donor availability coupled with the advent of reduced intensity conditioning (RIC) regimens has dramatically increased transplant access and consequently allo-SCT is now a key component of the treatment algorithm in both patients with AML in first CR (CR1) and advanced disease. Although transplant related mortality has fallen steadily over recent decades there has been no real progress in reducing the risk of disease relapse which remains the major cause of transplant failure and represents a major area of unmet need. A number of therapeutic approaches with the potential to reduce disease relapse, including advances in induction chemotherapy, the development of novel conditioning regimens and the emergence of the concept of post-transplant maintenance, are currently under development. Furthermore, the use of genetics and measurable residual disease technology in disease assessment has improved the identification of patients who are likely to benefit from an allo-SCT which now represents an increasingly personalized therapy. Future progress in optimizing transplant outcome will be dependent on the successful delivery by the international transplant community of randomized prospective clinical trials which permit examination of current and future transplant therapies with the same degree of rigor as is routinely adopted for non-transplant therapies.

Interactive Web Application for Plotting Personalized Prognosis Prediction Curves in Allogeneic Hematopoietic Cell Transplantation Using Machine Learning

BACKGROUND

Allogeneic hematopoietic cell transplantation (allo-HCT) is a curative treatment option for malignant hematological disorders. Transplant clinicians estimate patient-specific prognosis empirically in clinical practice based on previous studies on similar patients. However, this approach does not provide objective data. The present study primarily aimed to develop a tool capable of providing accurate personalized prognosis prediction after allo-HCT in an objective manner.

METHODS

We developed an interactive web application tool with a graphical user interface capable of plotting the personalized survival and cumulative incidence prediction curves after allo-HCT adjusted by 8 patient-specific factors, which are known as prognostic predictors, and assessed their predictive performances. A random survival forest model using the data of patients who underwent allo-HCT at our institution was applied to develop this application.

RESULTS

We succeeded in showing the personalized prognosis prediction curves of 1-year overall survival, progression-free survival, relapse/progression, and nonrelapse mortality (NRM) interactively using our web application (https://predicted-os-after-transplantation.shinyapps.io/RSF_model/). To assess its predictive performance, the entire cohort (363 cases) was split into a training cohort (70%) and a test cohort (30%) time-sequentially based on the patients' transplant dates. The areas under the receiver-operating characteristic curves for 1-year overall survival, progression-free survival, relapse/progression, and nonrelapse mortality in test cohort were 0.70, 0.72, 0.73, and 0.77, respectively.

CONCLUSIONS

The new web application could allow transplant clinicians to inform a new allo-HCT candidate of the objective personalized prognosis prediction and facilitate decision-making.

Machine Learning to Assess the Risk of Multidrug-Resistant Gram-Negative Bacilli Infections in Febrile Neutropenic Hematological Patients

INTRODUCTION

We aimed to assess risk factors for multidrug-resistant Gram-negative bacilli (MDR-GNB) from a large amount of data retrieved from electronic health records (EHRs) and determine whether machine learning (ML) may be useful in assessing the risk of MDR-GNB infection at febrile neutropenia (FN) onset.

METHODS

Retrospective study of almost 7 million pieces of structured data from all consecutive episodes of FN in hematological patients in a tertiary hospital in Barcelona (January 2008-December 2017). Conventional multivariate analysis and ML algorithms (random forest, gradient boosting machine, XGBoost, and GLM) were done.

RESULTS

A total of 3235 episodes of FN in 349 patients were documented; MDR-GNB caused 180 (5.6%) infections in 132 patients. The most frequent MDR-GNBs were MDR-Pseudomonas aeruginosa (53%) and extended-spectrum beta-lactamase-producing Enterobacterales (46%). According to conventional logistic regression analysis, independent factors associated with MDR-GNB infection were age older than 45 years (OR 2.07; 95% CI 1.31-3.24), prior antibiotics (2.62; 1.39-4.92), first-ever FN in this hospitalization (2.94; 1.33-6.52), prior hospitalizations for FN (1.72; 1.02-2.89); at least 15 prior hospital visits (2.65; 1.31-5.33), high-risk hematological diseases (3.62; 1.12-11.67), and hospitalization in a room formerly occupied by patients with MDR-GNB isolation (1.69; 1.20-2.38). ML algorithms achieved the following AUC and F1 score for MDR-GNB prediction: random forest, 0.79-0.9711; GMB, 0.79-0.9705; XGBoost, 0.79-0.9670; and GLM, 0.78-0.9716.

CONCLUSION

Data generated in EHRs proved useful in assessing risk factors for MDR-GNB infections in patients with FN. The great number of analyzed variables allowed us to identify new factors related to MDR infection, as well as to train ML algorithms for infection predictions. This information may be used by clinicians to make better clinical decisions.

Optimizing Transplant Approaches and Post-Transplant Strategies for Patients With Acute Myeloid Leukemia

Acute Myeloid Leukemia (AML) is the commonest indication for allogeneic stem cell transplantation (allo-SCT) worldwide. The increasingly important role of allo-SCT in the management of AML has been underpinned by two important advances. Firstly, improvements in disease risk stratification utilizing genetic and Measurable Residual Disease (MRD) technologies permit ever more accurate identification of allo-mandatory patients who are at high risk of relapse if treated by chemotherapy alone. Secondly, increased donor availability coupled with the advent of reduced intensity conditioning (RIC) regimens has substantially expanded transplant access for patients with high risk AML In patients allografted for AML disease relapse continues to represent the commonest cause of transplant failure and the development of novel strategies with the potential to reduce disease recurrence represents a major unmet need.

Comparison of the Prognostic Ability of the HCT-CI, the Modified EBMT, and the EBMT-ADT Pre-transplant Risk Scores for Acute Leukemia

BACKGROUND

Allogeneic hematopoietic cell transplantation (HCT) outcomes may be predicted by published risk scores; however, the ideal system has not been identified for acute leukemias.

PATIENTS AND METHODS

We retrospectively examined the Hematopoietic Cell Transplantation-Comorbidity Index (HCT-CI), modified European Group for Blood and Marrow Transplantation (mEBMT), EBMT-Alternating Decision Tree (ADT), and others on 231 patients with acute leukemia.

RESULTS

Acute myeloid leukemia was diagnosed in 200 patients, and acute lymphocytic leukemia was diagnosed in 31 patients. For HCT-CI, patients were grouped as 0 to 1, 2 to 3, and > 3. For mEBMT, patients were grouped as 0 to 2, 3, and > 3. For EBMT-ADT, the 100-day mortality was calculated and grouped as ≤ 4.1%, 4.1% to 11.5%, and > 11.5%. Higher HCI-CI demonstrated inferior overall survival (P = .04; c-statistic, 0.57), whereas mEBMT and EBMT-ADT did not stratify well. A new weighted score was developed that assigned 1 point for age ≥ 60 years, acute lymphocytic leukemia diagnosis, mismatch unrelated or haploidentical donor, cardiovascular comorbidity, and pre-transplant diabetes, whereas arrhythmia received 2 points. The new weighted score assigned 0 points to 88 (38%), 1 to 2 points to 121 (52%) and ≥ 3 points to 22 (10%) patients, and demonstrated improved prognostic capability compared with the other scores (P = .0001; c-statistic, 0.61).

CONCLUSIONS

The HCT-CI stratifies patients with leukemia for overall survival but is inferior to our single-center score, which is influenced by cardiac comorbidity and arrhythmia. Differences in pre-transplant risk scores may be related to different transplant practices.

Development and validation of a disease risk stratification system for patients with haematological malignancies: a retrospective cohort study of the European Society for Blood and Marrow Transplantation registry

BACKGROUND

Diagnosis and remission status at the time of allogeneic haematopoietic stem-cell transplantation (HSCT) are the principal determinants of overall survival following transplantation. We sought to develop a contemporary disease-risk stratification system (DRSS) that accounts for heterogeneous transplantation indications.

METHODS

In this retrospective cohort study we included 55 histology and remission status combinations across haematological malignancies, including acute leukaemia, lymphoma, multiple myeloma, and myeloproliferative and myelodysplastic disorders. A total of 47 265 adult patients (aged ≥18 years) who received an allogeneic HSCT between Jan 1, 2012, and Dec 31, 2016, and were reported to the European Society for Blood and Marrow Transplantation registry were included. We divided EBMT patients into derivation (n=25 534), tuning (n=18 365), and geographical validation (n=3366) cohorts. Disease combinations were ranked in a multivariable Cox regression for overall survival in the derivation cohort, cutoff for risk groups were evaluated for the tuning cohort, and the selected system was tested on the geographical validation cohort. An independent single-centre US cohort of 660 patients transplanted between Jan 1, 2010, and Dec 31, 2015 was used to externally validate the results.

FINDINGS

The DRSS model stratified patients in the derivation cohort (median follow-up was 2·1 years [IQR 1·0-3·2]) into five risk groups with increasing mortality risk: low risk (reference group), intermediate-1 (hazard ratio for overall survival 1·26 [95% CI 1·17-1·36], p<0·0001), intermediate-2 (1·53 [1·42-1·66], p<0·0001), high (2·03 [1·86-2·22], p<0·0001), and very high (2·87 [2·63-3·13], p<0·0001). DRSS levels were also associated with a stepwise increase in risk across the tuning and geographical validation cohort. In the external validation cohort (median follow-up was 5·7 years [IQR 4·5-7·1]), the DRSS scheme separated patients into 4 risk groups associated with increasing risk of mortality: intermediate-2 risk (hazard ratio [HR] 1·34 [95% CI 1·04-1·74], p=0·025), high risk (HR 2·03 [95% CI 1·39-2·95], p=0·00023) and very-high risk (HR 2·26 [95% CI 1·62-3·15], p<0·0001) patients compared with the low risk and intermediate-1 risk group (reference group). Across all cohorts, between 64% and 65% of patients were categorised as having intermediate-risk disease by a previous prognostic system (ie, the disease-risk index [DRI]). The DRSS reclassified these intermediate-risk DRI patients, with 855 (6%) low risk, 7111 (51%) intermediate-1 risk, 5700 (41%) intermediate-2 risk, and 375 (3%) high risk or very high risk of 14 041 patients in a subanalysis combining the tuning and internal geographic validation cohorts. The DRI projected 2-year overall survival was 62·1% (95% CI 61·2-62·9) for these 14 041 patients, while the DRSS reclassified them into finer prognostic groups with overall survival ranging from 45·7% (37·4-54·0; very high risk patients) to 73·1% (70·1-76·2; low risk patients).

INTERPRETATION

The DRSS is a novel risk stratification tool including disease features related to histology, genetic profile, and treatment response. The model should serve as a benchmark for future studies. This system facilitates the interpretation and analysis of studies with heterogeneous cohorts, promoting trial-design with more inclusive populations.

FUNDING

The Varda and Boaz Dotan Research Center for Hemato-Oncology Research, Tel Aviv University.

Long-Term Outcomes After Autologous Versus Allogeneic Stem Cell Transplantation in Molecularly-Stratified Patients With Intermediate Cytogenetic Risk Acute Myeloid Leukemia: A PETHEMA Study

Acute myeloid leukemia (AML) with intermediate risk cytogenetics (IRcyto) comprises a variety of biological entities with distinct mutational landscapes that translate into differential risks of relapse and prognosis. Optimal postremission therapy choice in this heterogeneous patient population is currently unsettled. In the current study, we compared outcomes in IRcyto AML recipients of autologous (autoSCT) (n = 312) or allogeneic stem cell transplantation (alloSCT) (n = 279) in first complete remission (CR1). Molecular risk was defined based on CEBPA, NPM1, and FLT3-ITD mutational status, per European LeukemiaNet 2017 criteria. Five-year overall survival (OS) in patients with favorable molecular risk (FRmol) was 62% (95% confidence interval [CI], 50-72) after autoSCT and 66% (95% CI, 41-83) after matched sibling donor (MSD) alloSCT (P = .68). For patients of intermediate molecular risk (IRmol), MSD alloSCT was associated with lower cumulative incidence of relapse (P < .001), as well as with increased nonrelapse mortality (P = .01), as compared to autoSCT. The 5-year OS was 47% (95% CI, 34-58) after autoSCT and 70% (95% CI, 59-79) after MSD alloSCT (P = .02) in this patient subgroup. In a propensity-score matched IRmol subcohort (n = 106), MSD alloSCT was associated with superior leukemia-free survival (hazard ratio [HR] 0.33, P = .004) and increased OS in patients alive 1 year after transplantation (HR 0.20, P = .004). These results indicate that, within IRcyto AML in CR1, autoSCT may be a valid option for FRmol patients, whereas MSD alloSCT should be the preferred postremission strategy in IRmol patients.