Machine learning reveals chronic graft-versus-host disease phenotypes and stratifies survival after stem cell transplant for hematologic malignancies

Programmable biomaterials for bone regeneration

Programmable biomaterials are distinguished by their ability to adjust properties and functions on demand, in a periodic, reversible, or sequential manner. This contrasts with traditional biomaterials, which undergo irreversible, uncontrolled changes. This review synthesizes key advances in programmable biomaterials, examining their design principles, functionalities and applications in bone regeneration. It charts the transition from traditional to programmable biomaterials, emphasizing their enhanced precision, safety and control, which are critical from clinical and biosafety standpoints. We then classify programmable biomaterials into six types: dynamic nucleic acid-based biomaterials, electrically responsive biomaterials, bioactive scaffolds with programmable properties, nanomaterials for targeted bone regeneration, surface-engineered implants for sequential regeneration and stimuli-responsive release materials. Each category is analyzed for its structural properties and its impact on bone tissue engineering. Finally, the review further concludes by highlighting the challenges faced by programmable biomaterials and suggests integrating artificial intelligence and precision medicine to enhance their application in bone regeneration and other biomedical fields.

Oral Chronic Graft-Versus-Host Disease: Pathogenesis, Diagnosis, Current Treatment, and Emerging Therapies

Chronic graft-versus-host disease (cGvHD) is a multisystem disorder that occurs in recipients of allogeneic hematopoietic (alloHCT) stem cell transplants and is characterized by both inflammatory and fibrotic manifestations. It begins with the recognition of host tissues by the non-self (allogeneic) graft and progresses to tissue inflammation, organ dysfunction and fibrosis throughout the body. Oral cavity manifestations of cGVHD include mucosal features, salivary gland dysfunction and fibrosis. This review synthesizes current knowledge on the pathogenesis, diagnosis and management of oral cGVHD, with a focus on emerging trends and novel therapeutics. Data from various clinical studies and expert consensus are integrated to provide a comprehensive overview.

Distinct Type 1 Immune Networks Underlie the Severity of Restrictive Lung Disease after COVID-19

The variable etiology of persistent breathlessness after COVID-19 have confounded efforts to decipher the immunopathology of lung sequelae. Here, we analyzed hundreds of cellular and molecular features in the context of discrete pulmonary phenotypes to define the systemic immune landscape of post-COVID lung disease. Cluster analysis of lung physiology measures highlighted two phenotypes of restrictive lung disease that differed by their impaired diffusion and severity of fibrosis. Machine learning revealed marked CCR5+CD95+ CD8+ T-cell perturbations in mild-to-moderate lung disease, but attenuated T-cell responses hallmarked by elevated CXCL13 in more severe disease. Distinct sets of cells, mediators, and autoantibodies distinguished each restrictive phenotype, and differed from those of patients without significant lung involvement. These differences were reflected in divergent T-cell-based type 1 networks according to severity of lung disease. Our findings, which provide an immunological basis for active lung injury versus advanced disease after COVID-19, might offer new targets for treatment.

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.

Chronic graft-versus-host disease. Part I: Epidemiology, pathogenesis, and clinical manifestations

Chronic graft-versus-host disease is a major complication of allogeneic hematopoietic cell transplantation and a leading cause of long-term morbidity, nonrelapse mortality, and impaired health-related quality of life. The skin is commonly affected and presents heterogeneously, making the role of dermatologists critical in both diagnosis and treatment. In addition, new clinical classification and grading schemes inform treatment algorithms, which now include 3 U.S. Food and Drug Administration-approved therapies, and evolving transplant techniques are changing disease epidemiology. Part I reviews the epidemiology, pathogenesis, clinical manifestations, and diagnosis of chronic graft-versus-host disease. Part II discusses disease grading and therapeutic management.

Anti-Idiotypic mRNA Vaccine to Treat Autoimmune Disorders

The 80+ existing autoimmune disorders (ADs) affect billions with little prevention or treatment options, except for temporary symptomatic management, leading to enormous human suffering and a monumental financial burden. The autoantibodies formed in most ADs have been identified, allowing the development of novel anti-idiotypic antibodies to mute the autoantibodies using vaccines. Nucleoside vaccines have been successfully tested as antigen-specific immunotherapies (ASI), with mRNA technology offering multi-epitope targeting to mute multiple autoantibodies. This paper proposes using mRNA technology to produce anti-idiotypic antibodies with broad effectiveness in preventing and treating them. This paper delves into the state-of-the-art mRNA design strategies used to develop novel ASIs by selecting appropriate T cell and B cell epitopes to generate anti-idiotypic antibodies. The low cost and fast development of mRNA vaccines make this technology the most affordable for the global control of ADs.

Chronic GvHD NIH Consensus Project Biology Task Force: evolving path to personalized treatment of chronic GvHD

Chronic graft-versus-host disease (cGvHD) remains a prominent barrier to allogeneic hematopoietic stem cell transplantion as the leading cause of nonrelapse mortality and significant morbidity. Tremendous progress has been achieved in both the understanding of pathophysiology and the development of new therapies for cGvHD. Although our field has historically approached treatment from an empiric position, research performed at the bedside and bench has elucidated some of the complex pathophysiology of cGvHD. From the clinical perspective, there is significant variability of disease manifestations between individual patients, pointing to diverse biological underpinnings. Capitalizing on progress made to date, the field is now focused on establishing personalized approaches to treatment. The intent of this article is to concisely review recent knowledge gained and formulate a path toward patient-specific cGvHD therapy.

Prognostic Value of Cutaneous Disease Severity Estimates on Survival Outcomes in Patients With Chronic Graft-vs-Host Disease

Importance

Prior studies have demonstrated an association between cutaneous chronic graft-vs-host disease (cGVHD) and mortality. Assessment of the prognostic value of different measures of disease severity would assist in risk stratification.

Objective

To compare the prognostic value of body surface area (BSA) and National Institutes of Health (NIH) Skin Score on survival outcomes stratified by erythema and sclerosis subtypes of cGVHD.

Design, Setting, and Participants

Multicenter prospective cohort study from the Chronic Graft-vs-Host Disease Consortium including 9 medical centers in the US, enrolled from 2007 through 2012 and followed until 2018. Participants were adults and children with a diagnosis of cGVHD requiring systemic immunosuppression and with skin involvement during the study period, who had longitudinal follow-up. Data analysis was performed from April 2019 to April 2022.

Exposures

Patients underwent continuous BSA estimation and categorical NIH Skin Score grading of cutaneous cGVHD at enrollment and every 3 to 6 months thereafter.

Main Outcomes and Measures

Nonrelapse mortality (NRM) and overall survival (OS), compared between BSA and NIH Skin Score longitudinal prognostic models, adjusted for age, race, conditioning intensity, patient sex, and donor sex.

Results

Of 469 patients with cGVHD, 267 (57%) (105 female [39%]; mean [SD] age, 51 [12] years) had cutaneous cGVHD at enrollment, and 89 (19%) developed skin involvement subsequently. Erythema-type disease had earlier onset and was more responsive to treatment compared with sclerosis-type disease. Most cases (77 of 112 [69%]) of sclerotic disease occurred without prior erythema. Erythema-type cGVHD at first follow-up visit was associated with NRM (hazard ratio, 1.33 per 10% BSA increase; 95% CI, 1.19-1.48; P < .001) and OS (hazard ratio, 1.28 per 10% BSA increase; 95% CI, 1.14-1.44; P < .001), while sclerosis-type cGVHD had no significant association with mortality. The model with erythema BSA collected at baseline and first follow-up visits retained 75% of the total prognostic information (from all covariates including BSA and NIH Skin Score) for NRM and 73% for OS, with no statistical difference between prognostic models (likelihood ratio test χ2, 5.9; P = .05). Conversely, NIH Skin Score collected at the same intervals lost significant prognostic information (likelihood ratio test χ2, 14.7; P < .001). The model incorporating NIH Skin Score instead of erythema BSA accounted for only 38% of the total information for NRM and 58% for OS.

Conclusions and Relevance

In this prospective cohort study, erythema-type cutaneous cGVHD was associated with increased risk of mortality. Erythema BSA collected at baseline and follow-up predicted survival more accurately than the NIH Skin Score in patients requiring immunosuppression. Accurate assessment of erythema BSA may assist in identifying patients with cutaneous cGVHD at high risk for mortality.

Day 4 collection of granulocyte colony-stimulating factor-mobilized HLA-matched sibling donor peripheral blood allografts demonstrates no long-term increase in chronic graft-versus-host disease or relapse rates

BACKGROUND AIMS

In a previous pilot study of HLA-matched sibling donor hematopoietic cell transplantation (HCT), the authors determined the feasibility of day 4 versus day 5 granulocyte colony-stimulating factor (G-CSF)-mobilized peripheral blood stem cell (PBSC) collection compared with a historical cohort. Given identified differences in the PBSC product (day 4 cohort with significantly lower infused total nucleated, mononuclear and CD3 cells compared with other collection cohorts), the authors performed a follow-up study to determine long-term post-HCT outcomes, including detailed characterization of chronic graft-versus-host disease (GVHD).

METHODS

This was a prospective observational study, and the authors collected data on chronic GVHD, staging, sites of involvement and treatments. Performance status, incidence of relapse, overall survival and duration of immunosuppressive therapy (IST) were also evaluated. Data were examined retrospectively. To account for differences in length of follow-up among cohorts, the authors also determined performance status and chronic GVHD staging, sites and treatment at 2 years post-HCT.

RESULTS

At 2 years post-HCT, the overall survival rate was 71.7% in the day 4 cohort compared with 61.5%, 52% and 56% in the day 5, 2-day and historical cohorts, respectively (P = 0.283). The cumulative incidence of chronic GVHD was 65.2% in the day 4 cohort versus 46.4% in the day 5 cohort, 51.1% in the 2-day cohort and 65% in the historical cohort (P = 0.26). There was no significant difference in the maximum overall stage of chronic GVHD (P = 0.513), median number of sites involved (P = 0.401) or cumulative incidence of discontinuation of IST (P = 0.32). Death from chronic GVHD was less common in the day 4 and day 5 cohorts compared with the 2-day and historical cohorts, though this did not reach statistical significance.

CONCLUSIONS

The authors' preliminary results demonstrated that collection of allogeneic matched sibling donor PBSCs on day 4 of G-CSF was feasible, reduced donor exposure to growth factor and was associated with an initial cost savings. Importantly, the authors now demonstrate that transplantation of day 4 mobilized PBSCs is not associated with any adverse outcomes post-HCT, including late effects such as chronic GVHD. Further investigation of donor G-CSF collection algorithms is merited in other HCT settings, including unrelated and mismatched related donors.

Chronic gvhd dictionary-eurograft cost action initiative consensus report

Chronic graft versus host disease (cGVHD) affects patients after allogeneic hematopoietic stem cell transplantation (alloHSCT). This orphan disease poses a challenge for clinicians and researchers. The purpose of the cGVHD Dictionary is to provide a standardized structure for cGVHD databases on an international level, reconciling differences in data retrieval and facilitate database merging. It is derived from several consensus meetings of the EUROGRAFT consortium (European Cooperation in Science and Technology-COST Action CA17138) followed by a consensus process involving European Society for Blood and Marrow Transplantation (EBMT), US GvHD consortium and Center for International Bone Marrow Transplant Registry (CIBMTR). Databases used for the dictionary were: the National Institutes of Health (NIH) database, the Center for International Blood and Marrow Transplant Research, Applying Biomarkers to Minimize Long Term Effects of Childhood/Adolescent Cancer Treatment - Pediatric Blood and Marrow Transplant Consortium database, EBMT registry, the German-Austrian-Swiss GvHD registry, Italian Blood and Marrow Transplantation Society registry and Regensburg-Göttingen-Newcastle HSCT dataset. A four-part cGVHD Dictionary was formed based on the databases, consensus, and evidence in the literature. The Dictionary is divided into: (1) Patient characteristics, (2) Transplant characteristics, (3) cGVHD characteristics and (4) patient-reported quality of life, symptom burden and functional indicators.

Expanding Personalized, Data-Driven Dermatology: Leveraging Digital Health Technology and Machine Learning to Improve Patient Outcomes

The current revolution of digital health technology and machine learning offers enormous potential to improve patient care. Nevertheless, it is essential to recognize that dermatology requires an approach different from those of other specialties. For many dermatological conditions, there is a lack of standardized methodology for quantitatively tracking disease progression and treatment response (clinimetrics). Furthermore, dermatological diseases impact patients in complex ways, some of which can be measured only through patient reports (psychometrics). New tools using digital health technology (e.g., smartphone applications, wearable devices) can aid in capturing both clinimetric and psychometric variables over time. With these data, machine learning can inform efforts to improve health care by, for example, the identification of high-risk patient groups, optimization of treatment strategies, and prediction of disease outcomes. We use the term personalized, data-driven dermatology to refer to the use of comprehensive data to inform individual patient care and improve patient outcomes. In this paper, we provide a framework that includes data from multiple sources, leverages digital health technology, and uses machine learning. Although this framework is applicable broadly to dermatological conditions, we use the example of a serious inflammatory skin condition, chronic cutaneous graft-versus-host disease, to illustrate personalized, data-driven dermatology.

Prediction and recommendation by machine learning through repetitive internal validation for hepatic veno-occlusive disease/sinusoidal obstruction syndrome and early death after allogeneic hematopoietic cell transplantation

Using traditional statistical methods, we previously analyzed the risk factors and treatment outcomes of veno-occlusive disease/sinusoidal obstruction syndrome (VOD/SOS) after allogeneic hematopoietic cell transplantation. Within the same cohort, we applied machine learning to create prediction and recommendation models. We analyzed 2572 transplants using eXtreme Gradient Boosting (XGBoost) to predict post-transplant VOD/SOS and early death. Using the XGBoost and SHapley Additive exPlanations (SHAP), we found influential factors and devised recommendation models, which were internally verified by repetitive ten-fold cross-validation. SHAP values suggested that gender, busulfan dosage, age, forced expiratory volume, and Disease Risk Index were significant factors for VOD/SOS. The areas under the receiver operating characteristic curves and the areas under the precision-recall curve of the models were 0.740, 0.144 for all VOD/SOS, 0.793, 0.793 for severe to very severe VOD/SOS, and 0.746, 0.304 for early death. According to our single feature recommendation, following the busulfan dosage was the most effective for preventing VOD/SOS. The recommendation method for six adjustable feature sets was also validated, and a subgroup corresponding to five to six features showed significant preventive power for VOD/SOS and early death. Our personalized treatment set recommendation showed reproducibility in repetitive internal validation, but large external cohorts should prospectively validate our model.

A Clinical Prognostic Model Based on Machine Learning from the Fondazione Italiana Linfomi (FIL) MCL0208 Phase III Trial

BACKGROUND

Multicenter clinical trials are producing growing amounts of clinical data. Machine Learning (ML) might facilitate the discovery of novel tools for prognostication and disease-stratification. Taking advantage of a systematic collection of multiple variables, we developed a model derived from data collected on 300 patients with mantle cell lymphoma (MCL) from the Fondazione Italiana Linfomi-MCL0208 phase III trial (NCT02354313).

METHODS

We developed a score with a clustering algorithm applied to clinical variables. The candidate score was correlated to overall survival (OS) and validated in two independent data series from the European MCL Network (NCT00209222, NCT00209209); Results: Three groups of patients were significantly discriminated: Low, Intermediate (Int), and High risk (High). Seven discriminants were identified by a feature reduction approach: albumin, Ki-67, lactate dehydrogenase, lymphocytes, platelets, bone marrow infiltration, and B-symptoms. Accordingly, patients in the Int and High groups had shorter OS rates than those in the Low and Int groups, respectively (Int→Low, HR: 3.1, 95% CI: 1.0-9.6; High→Int, HR: 2.3, 95% CI: 1.5-4.7). Based on the 7 markers, we defined the engineered MCL international prognostic index (eMIPI), which was validated and confirmed in two independent cohorts; Conclusions: We developed and validated a ML-based prognostic model for MCL. Even when currently limited to baseline predictors, our approach has high scalability potential.

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.

Machine Learning in Oncology: What Should Clinicians Know?

The volume and complexity of scientific and clinical data in oncology have grown markedly over recent years, including but not limited to the realms of electronic health data, radiographic and histologic data, and genomics. This growth holds promise for a deeper understanding of malignancy and, accordingly, more personalized and effective oncologic care. Such goals require, however, the development of new methods to fully make use of the wealth of available data. Improvements in computer processing power and algorithm development have positioned machine learning, a branch of artificial intelligence, to play a prominent role in oncology research and practice. This review provides an overview of the basics of machine learning and highlights current progress and challenges in applying this technology to cancer diagnosis, prognosis, and treatment recommendations, including a discussion of current takeaways for clinicians.

Deep phenotyping of synovial molecular signatures by integrative systems analysis in rheumatoid arthritis

OBJECTIVE

RA encompasses a complex, heterogeneous and dynamic group of diseases arising from molecular and cellular perturbations of synovial tissues. The aim of this study was to decipher this complexity using an integrative systems approach and provide novel insights for designing stratified treatments.

METHODS

An RNA sequencing dataset of synovial tissues from 152 RA patients and 28 normal controls was imported and subjected to filtration of differentially expressed genes, functional enrichment and network analysis, non-negative matrix factorization, and key driver analysis. A naïve Bayes classifier was applied to the independent datasets to investigate the factors associated with treatment outcome.

RESULTS

A matrix of 1241 upregulated differentially expressed genes from RA samples was classified into three subtypes (C1-C3) with distinct molecular and cellular signatures. C3 with prominent immune cells and proinflammatory signatures had a stronger association with the presence of ACPA and showed a better therapeutic response than C1 and C2, which were enriched with neutrophil and fibroblast signatures, respectively. C2 was more occupied by synovial fibroblasts of destructive phenotype and carried highly expressed key effector molecules of invasion and osteoclastogenesis. CXCR2, JAK3, FYN and LYN were identified as key driver genes in C1 and C3. HDAC, JUN, NFKB1, TNF and TP53 were key regulators modulating fibroblast aggressiveness in C2.

CONCLUSIONS

Deep phenotyping of synovial heterogeneity captured comprehensive and discrete pathophysiological attributes of RA regarding clinical features and treatment response. This result could serve as a template for future studies to design stratified approaches for RA patients.

Unsupervised machine learning reveals key immune cell subsets in COVID-19, rhinovirus infection, and cancer therapy

For an emerging disease like COVID-19, systems immunology tools may quickly identify and quantitatively characterize cells associated with disease progression or clinical response. With repeated sampling, immune monitoring creates a real-time portrait of the cells reacting to a novel virus before disease-specific knowledge and tools are established. However, single cell analysis tools can struggle to reveal rare cells that are under 0.1% of the population. Here, the machine learning workflow Tracking Responders EXpanding (T-REX) was created to identify changes in both rare and common cells across human immune monitoring settings. T-REX identified cells with highly similar phenotypes that localized to hotspots of significant change during rhinovirus and SARS-CoV-2 infections. Specialized MHCII tetramer reagents that mark rhinovirus-specific CD4+ cells were left out during analysis and then used to test whether T-REX identified biologically significant cells. T-REX identified rhinovirus-specific CD4+ T cells based on phenotypically homogeneous cells expanding by ≥95% following infection. T-REX successfully identified hotspots of virus-specific T cells by comparing infection (day 7) to either pre-infection (day 0) or post-infection (day 28) samples. Plotting the direction and degree of change for each individual donor provided a useful summary view and revealed patterns of immune system behavior across immune monitoring settings. For example, the magnitude and direction of change in some COVID-19 patients was comparable to blast crisis acute myeloid leukemia patients undergoing a complete response to chemotherapy. Other COVID-19 patients instead displayed an immune trajectory like that seen in rhinovirus infection or checkpoint inhibitor therapy for melanoma. The T-REX algorithm thus rapidly identifies and characterizes mechanistically significant cells and places emerging diseases into a systems immunology context for comparison to well-studied immune changes.

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.

Ruxolitinib for chronic steroid-refractory graft versus host disease: a single center experience

BACKGROUND

Chronic Graft versus Host Disease (GvHD) is a serious complication of allogeneic hematopoietic stem cell transplant that severely impacts quality of life and long-term survival. About 50-to-60 % of patients treated with steroids require a further line of therapy due to lack of sustained response. Ruxolitinib, a JAK1/2 inhibitor, has recently been approved for the treatment of acute GvHD.

METHODS

We aimed to retrospectively evaluate ruxolitinib efficacy and safety in a cohort of patients diagnosed with moderate (25 %) or severe (75 %) steroid-refractory or steroid-dependent chronic GvHD. Response evaluation was performed at three and six months.

RESULTS

Thirty-six patients received ruxolitinib after a median of three previous lines (range, r 1-11) for a median of 8.6 months (r 1-51.6). Cutaneous GvHD was the most frequent presentation. We observed an overall response of 59 % (CR 9%, PR 50 %) at three months and 62 % (CR 15 %, PR 46 %) at six months. Two patients had hematologic disease recurrence and were censored at relapse; no other permanent discontinuation due to adverse events were documented. Cutaneous, oral, genital and ocular GvHD significantly improved after treatment. 2-year overall survival and 2-year transplant related mortality were 74 % and 19 % respectively. Ruxolitinib was associated with a significant reduction of steroid dose.

CONCLUSION

Ruxolitinib was confirmed to be a safe and effective option as salvage treatment also for advanced stages of chronic GvHD. Longer follow up is needed to evaluate durability of response. Prospective analyses on larger cohorts are ongoing.

Methods to Assess Disease Activity and Severity in Cutaneous Chronic Graft-versus-Host Disease: A Critical Literature Review

Chronic graft-versus-host disease (cGVHD), a potentially debilitating complication of hematopoietic cell transplantation, confers increased risk for mortality. Whereas treatment decisions rely on an accurate assessment of disease activity/severity, validated methods of assessing cutaneous cGVHD activity/severity appear to be limited. In this study, we aimed to identify and evaluate current data on the assessment of disease activity/severity in cutaneous cGVHD. Using modified PRISMA methods, we performed a critical literature review for relevant articles. Our literature search identified 1741 articles, of which 1635 were excluded as duplicates or failure to meet inclusion criteria. Of the included studies (n = 106), 39 (37%) addressed clinical and/or histopathologic parameters, 53 (50%) addressed serologic parameters, 8 (7.5%) addressed imaging parameters, and 6 (5.5%) addressed computer-based technologies. The only formally validated metric of disease activity/severity assessment in cutaneous cGVHD is the National Institutes of Health consensus scoring system, which is founded on clinical assessment alone. The lack of an objective marker for cGVHD necessitates further studies. An evaluation of the potential contributions of serologic, imaging, and/or computer-based technologies is warranted.

National Institutes of Health Consensus Development Project on Criteria for Clinical Trials in Chronic Graft-versus-Host Disease: IIa. The 2020 Clinical Implementation and Early Diagnosis Working Group Report

Recognition of the earliest signs and symptoms of chronic graft-versus-host disease (GVHD) that lead to severe manifestations remains a challenge. The standardization provided by the National Institutes of Health (NIH) 2005 and 2014 consensus projects has helped improve diagnostic accuracy and severity scoring for clinical trials, but utilization of these tools in routine clinical practice is variable. Additionally, when patients meet the NIH diagnostic criteria, many already have significant morbidity and possibly irreversible organ damage. The goals of this early diagnosis project are 2-fold. First, we provide consensus recommendations regarding implementation of the current NIH diagnostic guidelines into routine transplant care, outside of clinical trials, aiming to enhance early clinical recognition of chronic GVHD. Second, we propose directions for future research efforts to enable discovery of new, early laboratory as well as clinical indicators of chronic GVHD, both globally and for highly morbid organ-specific manifestations. Identification of early features of chronic GVHD that have high positive predictive value for progression to more severe manifestations of the disease could potentially allow for future pre-emptive clinical trials.

Application of machine learning in the management of acute myeloid leukemia: current practice and future prospects

Machine learning (ML) is rapidly emerging in several fields of cancer research. ML algorithms can deal with vast amounts of medical data and provide a better understanding of malignant disease. Its ability to process information from different diagnostic modalities and functions to predict prognosis and suggest therapeutic strategies indicates that ML is a promising tool for the future management of hematologic malignancies; acute myeloid leukemia (AML) is a model disease of various recent studies. An integration of these ML techniques into various applications in AML management can assure fast and accurate diagnosis as well as precise risk stratification and optimal therapy. Nevertheless, these techniques come with various pitfalls and need a strict regulatory framework to ensure safe use of ML. This comprehensive review highlights and discusses recent advances in ML techniques in the management of AML as a model disease of hematologic neoplasms, enabling researchers and clinicians alike to critically evaluate this upcoming, potentially practice-changing technology.

Unsupervised machine learning reveals key immune cell subsets in COVID-19, rhinovirus infection, and cancer therapy

For an emerging disease like COVID-19, systems immunology tools may quickly identify and quantitatively characterize cells associated with disease progression or clinical response. With repeated sampling, immune monitoring creates a real-time portrait of the cells reacting to a novel virus before disease specific knowledge and tools are established. However, single cell analysis tools can struggle to reveal rare cells that are under 0.1% of the population. Here, the machine learning workflow Tracking Responders Expanding (T-REX) was created to identify changes in both very rare and common cells in diverse human immune monitoring settings. T-REX identified cells that were highly similar in phenotype and localized to hotspots of significant change during rhinovirus and SARS-CoV-2 infections. Specialized reagents used to detect the rhinovirus-specific CD4+ cells, MHCII tetramers, were not used during unsupervised analysis and instead 'left out' to serve as a test of whether T-REX identified biologically significant cells. In the rhinovirus challenge study, T-REX identified virus-specific CD4+ T cells based on these cells being a distinct phenotype that expanded by ≥95% following infection. T-REX successfully identified hotspots containing virus-specific T cells using pairs of samples comparing Day 7 of infection to samples taken either prior to infection (Day 0) or after clearing the infection (Day 28). Mapping pairwise comparisons in samples according to both the direction and degree of change provided a framework to compare systems level immune changes during infectious disease or therapy response. This revealed that the magnitude and direction of systemic immune change in some COVID-19 patients was comparable to that of blast crisis acute myeloid leukemia patients undergoing induction chemotherapy and characterized the identity of the immune cells that changed the most. Other COVID-19 patients instead matched an immune trajectory like that of individuals with rhinovirus infection or melanoma patients receiving checkpoint inhibitor therapy. T-REX analysis of paired blood samples provides an approach to rapidly identify and characterize mechanistically significant cells and to place emerging diseases into a systems immunology context.

A Systematic Review of Machine Learning Techniques in Hematopoietic Stem Cell Transplantation (HSCT)

Machine learning techniques are widely used nowadays in the healthcare domain for the diagnosis, prognosis, and treatment of diseases. These techniques have applications in the field of hematopoietic cell transplantation (HCT), which is a potentially curative therapy for hematological malignancies. Herein, a systematic review of the application of machine learning (ML) techniques in the HCT setting was conducted. We examined the type of data streams included, specific ML techniques used, and type of clinical outcomes measured. A systematic review of English articles using PubMed, Scopus, Web of Science, and IEEE Xplore databases was performed. Search terms included "hematopoietic cell transplantation (HCT)," "autologous HCT," "allogeneic HCT," "machine learning," and "artificial intelligence." Only full-text studies reported between January 2015 and July 2020 were included. Data were extracted by two authors using predefined data fields. Following PRISMA guidelines, a total of 242 studies were identified, of which 27 studies met the inclusion criteria. These studies were sub-categorized into three broad topics and the type of ML techniques used included ensemble learning (63%), regression (44%), Bayesian learning (30%), and support vector machine (30%). The majority of studies examined models to predict HCT outcomes (e.g., survival, relapse, graft-versus-host disease). Clinical and genetic data were the most commonly used predictors in the modeling process. Overall, this review provided a systematic review of ML techniques applied in the context of HCT. The evidence is not sufficiently robust to determine the optimal ML technique to use in the HCT setting and/or what minimal data variables are required.

Acute myeloid leukemia and artificial intelligence, algorithms and new scores

Artificial intelligence, and more narrowly machine-learning, is beginning to expand humanity's capacity to analyze increasingly large and complex datasets. Advances in computer hardware and software have led to breakthroughs in multiple sectors of our society, including a burgeoning role in medical research and clinical practice. As the volume of medical data grows at an apparently exponential rate, particularly since the human genome project laid the foundation for modern genetic inquiry, informatics tools like machine learning are becoming crucial in analyzing these data to provide meaningful tools for diagnostic, prognostic, and therapeutic purposes. Within medicine, hematologic diseases can be particularly challenging to understand and treat given the increasingly complex and intercalated genetic, epigenetic, immunologic, and regulatory pathways that must be understood to optimize patient outcomes. In acute myeloid leukemia (AML), new developments in machine learning algorithms have enabled a deeper understanding of disease biology and the development of better prognostic and predictive tools. Ongoing work in the field brings these developments incrementally closer to clinical implementation.

Machine learning for predicting long-term kidney allograft survival: a scoping review

Supervised machine learning (ML) is a class of algorithms that "learn" from existing input-output pairs, which is gaining popularity in pattern recognition for classification and prediction problems. In this scoping review, we examined the use of supervised ML algorithms for the prediction of long-term allograft survival in kidney transplant recipients. Data sources included PubMed, the Cumulative Index to Nursing and Allied Health Literature, and the Institute for Electrical and Electronics Engineers (IEEE) Xplore libraries from inception to November 2019. We screened titles and abstracts and potentially eligible full-text reports to select studies and subsequently abstracted the data. Eleven studies were identified. Decision trees were the most commonly used method (n = 8), followed by artificial neural networks (ANN) (n = 4) and Bayesian belief networks (n = 2). The area under receiver operating curve (AUC) was the most common measure of discrimination (n = 7), followed by sensitivity (n = 5) and specificity (n = 4). Model calibration examining the reliability in risk prediction was performed using either the Pearson r or the Hosmer-Lemeshow test in four studies. One study showed that logistic regression had comparable performance to ANN, while another study demonstrated that ANN performed better in terms of sensitivity, specificity, and accuracy, as compared with a Cox proportional hazards model. We synthesized the evidence related to the comparison of ML techniques with traditional statistical approaches for prediction of long-term allograft survival in patients with a kidney transplant. The methodological and reporting quality of included studies was poor. Our study also demonstrated mixed results in terms of the predictive potential of the models.

Machine learning in haematological malignancies

Machine learning is a branch of computer science and statistics that generates predictive or descriptive models by learning from training data rather than by being rigidly programmed. It has attracted substantial attention for its many applications in medicine, both as a catalyst for research and as a means of improving clinical care across the cycle of diagnosis, prognosis, and treatment of disease. These applications include the management of haematological malignancy, in which machine learning has created inroads in pathology, radiology, genomics, and the analysis of electronic health record data. As computational power becomes cheaper and the tools for implementing machine learning become increasingly democratised, it is likely to become increasingly integrated into the research and practice landscape of haematology. As such, machine learning merits understanding and attention from researchers and clinicians alike. This narrative Review describes important concepts in machine learning for unfamiliar readers, details machine learning's current applications in haematological malignancy, and summarises important concepts for clinicians to be aware of when appraising research that uses machine learning.

Machine learning and artificial intelligence in haematology

Digitalization of the medical record and integration of genomic methods into clinical practice have resulted in an unprecedented wealth of data. Machine learning is a subdomain of artificial intelligence that attempts to computationally extract meaningful insights from complex data structures. Applications of machine learning in haematological scenarios are steadily increasing. However, basic concepts are often unfamiliar to clinicians and investigators. The purpose of this review is to provide readers with tools to interpret and critically appraise machine learning literature. We begin with the elucidation of standard terminology and then review examples in haematology. Guidelines for designing and evaluating machine-learning studies are provided. Finally, we discuss limitations of the machine-learning approach.

Unsupervised machine learning reveals risk stratifying glioblastoma tumor cells

A goal of cancer research is to reveal cell subsets linked to continuous clinical outcomes to generate new therapeutic and biomarker hypotheses. We introduce a machine learning algorithm, Risk Assessment Population IDentification (RAPID), that is unsupervised and automated, identifies phenotypically distinct cell populations, and determines whether these populations stratify patient survival. With a pilot mass cytometry dataset of 2 million cells from 28 glioblastomas, RAPID identified tumor cells whose abundance independently and continuously stratified patient survival. Statistical validation within the workflow included repeated runs of stochastic steps and cell subsampling. Biological validation used an orthogonal platform, immunohistochemistry, and a larger cohort of 73 glioblastoma patients to confirm the findings from the pilot cohort. RAPID was also validated to find known risk stratifying cells and features using published data from blood cancer. Thus, RAPID provides an automated, unsupervised approach for finding statistically and biologically significant cells using cytometry data from patient samples.

Biomarkers for Allogeneic HCT Outcomes

Allogeneic hematopoietic cell transplantation (HCT) remains the only curative therapy for many hematological malignant and non-malignant disorders. However, key obstacles to the success of HCT include graft-versus-host disease (GVHD) and disease relapse due to absence of graft-versus-tumor (GVT) effect. Over the last decade, advances in "omics" technologies and systems biology analysis, have allowed for the discovery and validation of blood biomarkers that can be used as diagnostic test and prognostic test (that risk-stratify patients before disease occurrence) for acute and chronic GVHD and recently GVT. There are also predictive biomarkers that categorize patients based on their likely to respond to therapy. Newer mathematical analysis such as machine learning is able to identify different predictors of GVHD using clinical characteristics pre-transplant and possibly in the future combined with other biomarkers. Biomarkers are not only useful to identify patients with higher risk of disease progression, but also help guide treatment decisions and/or provide a basis for specific therapeutic interventions. This review summarizes biomarkers definition, omics technologies, acute, chronic GVHD and GVT biomarkers currently used in clinic or with potential as targets for existing or new drugs focusing on novel published work.

Organ Changes Associated with Provider-Assessed Responses in Patients with Chronic Graft-versus-Host Disease

Assessments of overall improvement and worsening of chronic graft-versus-host disease (GVHD) manifestations by the algorithm recommended by National Institutes of Health (NIH) response criteria do not align closely with those reported by providers, particularly when patients have mixed responses with improvement in some manifestations but worsening in others. To elucidate the changes that influence provider assessment of response, we used logistic regression to generate an overall change index based on specific manifestations of chronic GVHD measured at baseline and 6 months later. We hypothesized that this overall change index would correlate strongly with overall improvement as determined by providers. The analysis included 488 patients from 2 prospective observational studies who were randomly assigned in a 3:2 ratio to discovery and replication cohorts. Changes in bilirubin and scores of the lower gastrointestinal tract, mouth, joint/fascia, lung, and skin were correlated with provider-assessed improvement, suggesting that the main NIH response measures capture relevant information. Conversely, changes in the eye, esophagus, and upper gastrointestinal tract did not correlate with provider-assessed response, suggesting that these scales could be modified or dropped from the NIH response assessment. The area under the receiver operator characteristic curve in the replication cohort was 0.72, indicating that the scoring algorithm for overall change based on NIH response measures is not well calibrated with provider-assessed response.

Changes in Immunosuppressive Treatment of Chronic Graft-versus-Host Disease: Comparison of 2 Surveys within Allogeneic Hematopoietic Stem Cell Transplant Centers in Germany, Austria, and Switzerland

Chronic graft-versus-host disease (cGVHD) remains the leading cause of late morbidity and mortality. Despite the growing number of treatment options in cGVHD, evidence remains sparse. The German-Austrian-Swiss GVHD Consortium performed a survey on clinical practice in treatment of cGVHD among transplant centers in Germany, Austria, and Switzerland in 2009 and 2018 and compared the results. The survey performed in 2009 contained 20 questions on first-line treatment and related issues and 4 questions on second-line scenarios followed by a survey on all systemic and topic treatment options known and applied, with 31 of 36 transplant centers (86%) responding. The survey in 2018 repeated 7 questions on first-line treatment and 3 questions on second-line scenarios followed by an updated survey on all current systemic treatment options known and applied, with 29 of 66 centers (43%) responding. In summary, the results show a large overlap of first-line treatment practice between centers and the 2 surveys because of a lack of new data that changes practice, except significant heterogeneity of treatment of cGVHD progressive onset type, which can be explained by the lack of trials focusing on this high-risk entity. In contrast, treatment options applied to second-line therapy vary considerably, with new agents like ibrutinib and ruxolitinib entering clinical practice. Moreover, treatment of bronchiolitis obliterans syndrome demonstrates heterogeneity in applied therapeutic options and sequence because of a lack of controlled data and different conclusions from already existing evidence. In summary, the survey results demonstrate an increasing number of treatment options applied to cGVHD accompanied by a significant heterogeneity in second-line treatment and underline the urgent need for clinical trials and registry analyses on rare entities with high mortality like progressive onset type and lung involvement of cGVHD.