Clinical and Genetic Risk Prediction of Cognitive Impairment After Blood or Marrow Transplantation for Hematologic Malignancy

Factors associated with perceived cognitive function in breast cancer patients treated with chemotherapy: A multicenter cross-sectional study

PURPOSE

This study aimed to investigate the factors associated with perceived cognitive function among breast cancer patients treated with chemotherapy in China.

METHODS

The study was a multicenter cross-sectional design. Data were collected from 10 public hospitals in China between April 2022 and February 2023. A total of 741 participants completed questionnaires assessing sociodemographic and medical characteristics, perceived cognitive function, sleep quality, fatigue, anxiety, and depression. Hierarchical multiple regression analysis was used to assess the determinants of cognitive function.

RESULTS

The hierarchical multiple regression model accounted for 31.5% of variation in perceived cognitive function (sociodemographic 4.5%; medical 6.6%; exercise frequency 6.6%; sleep quality 2.1%; fatigue 2.8%; anxiety combined with depression 9.0%). Education level, chemotherapy type, number of chemotherapy cycles, and cyclophosphamide drug use were significant predisposing factors of perceived cognitive function (p < 0.001). Exercising ≥3 times/week (p < 0.001) was a significant factor positively influencing perceived cognitive function, meanwhile, anxiety (p < 0.001) and depression (p < 0 0.001) were negative factors.

CONCLUSION

Our findings suggest that patients with low education levels, postoperative chemotherapy, cyclophosphamide treatment, and a greater number of chemotherapy cycles need more assessment. Sedentary patients, those who have never exercised, and those with anxiety or depression all showed greater cognitive decline. By identifying susceptible populations, encouraging regular exercise, and addressing anxiety and depression, healthcare professionals can contribute significantly to prevent patients' cognitive decline throughout chemotherapy.

Association of geriatric measures and global frailty with cognitive decline after allogeneic hematopoietic cell transplantation in older adults

INTRODUCTION

Allogeneic hematopoietic cell transplantation (alloHCT) is increasingly offered to older adults, and its potential impact on cognition in this population is understudied. This work aims to evaluate the ability of cancer-specific geriatric assessments (cGA) and a global frailty index based on accumulation of deficits identified in the cGA to predict the risk of cognitive decline after alloHCT in older adults.

MATERIALS AND METHODS

AlloHCT recipients aged 50 years or older completed a cGA, including a cognitive evaluation by the Blessed Orientation Memory Concentration (BOMC) test, at baseline prior to alloHCT and then at 3, 6, and 12 months after transplant. Baseline frailty was assessed using a deficit accumulation frailty index (DAFI) calculated from the cGA. A multinomial logit model was used to examine the association between predictors (individual cGA measures, DAFI) and the following three outcomes: alive with stable or improved cognition, alive with cognitive decline, and deceased. In post-hoc analyses, analysis of variance was used to compare BOMC scores at baseline, 3, 6, and 12 months across frailty categories.

RESULTS

In total, 148 participants were included, with a median age of 62 (range 50-76). At baseline, 12% had cognitive impairment; at one year, 29% of survivors had improved BOMC scores, 33% had stable BOMC, and 37% had worse BOMC. Prior to transplant, 25% were pre-frail and 11% were frail. Individual baseline cGA measures were not associated with cognitive change at one year as assessed by BOMC. Adjusting for age, sex, and education, those who were frail at baseline were 7.4 times as likely to develop cognitive decline at one year than those who were non-frail, although this finding did not reach statistical significance (95% confidence interval [CI] 0.74-73.8, p = 0.09). The probability of being alive with stable/improved cognition at 12 months for the non-frail, pre-frail, and frail groups was 43%, 34%, and 8%, respectively.

DISCUSSION

Baseline geriatric measures and frailty were not significantly associated with cognitive change as assessed by BOMC in adults aged 50 or older after alloHCT. However, the study was underpowered to detect clinically meaningful differences, and future work to elucidate potential associations between frailty and cognitive outcomes is warranted.

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.

Improving efficiency of fitting Cox proportional hazards models for time-to-event outcomes in genome-wide association studies (GWAS)

Summary

Technologies identifying single nucleotide polymorphisms (SNPs) in DNA sequencing yield an avalanche of data requiring analysis and interpretation. Standard methods may require many weeks of processing time. The use of statistical methods requiring data sorting, matrix inversions of a high-dimension and replication in subsets of the data on multiple outcomes exacerbate these times.A method which reduces the computational time in problems with time-to-event outcomes and hundreds of thousands/millions of SNPs using Cox-Snell residuals after fitting the Cox proportional hazards model (PH) to a fixed set of concomitant variables is proposed. This yields coefficients for SNP effect from a Cox-Snell adjusted Poisson model and shows a high concordance to the adjusted PH model.The method is illustrated with a sample of 10 000 SNPs from a genome-wide association study in a diabetic population. The gain in processing efficiency using the proposed method based on Poisson modelling can be as high as 62%. This could result in saving of over three weeks processing time if 5 million SNPs require analysis. The method involves only a single predictor variable (SNP), offering a simpler, computationally more stable approach to examining and identifying SNP patterns associated with the outcome(s) allowing for a faster development of genetic signatures. Use of deviance residuals from the PH model to screen SNPs demonstrates a large discordance rate at a 0.2% threshold of concordance. This rate is 15 times larger than that based on the Cox-Snell residuals from the Cox-Snell adjusted Poisson model.

Availability and implementation

The method is simple to implement as the procedures are available in most statistical packges. The approach involves obtaining Cox-Snell residuals from a PH model, to a binary time-to-event outcome, for factors which need to be common when assessing each SNP. Each SNP is then fitted as a predictor to the outcome of interest using a Poisson model with the Cox-Snell as the exposure variable.

Cancer-related cognitive impairment in racial and ethnic minority groups: a scoping review

PURPOSE

Disparities in cognitive function among racial and ethnic groups have been reported in non-cancer conditions, but cancer-related cognitive impairment (CRCI) in racial and ethnic minority groups is poorly understood. We aimed to synthesize and characterize the available literature about CRCI in racial and ethnic minority populations.

METHODS

We conducted a scoping review in the PubMed, PsycInfo, and Cumulative Index to Nursing and Allied Health Literature databases. Articles were included if they were published in English or Spanish, reported cognitive functioning in adults diagnosed with cancer, and characterized the race or ethnicity of the participants. Literature reviews, commentaries, letters to the editor, and gray literature were excluded.

RESULTS

Seventy-four articles met the inclusion criteria, but only 33.8% differentiated the CRCI findings by racial or ethnic subgroups. There were associations between cognitive outcomes and the participants' race or ethnicity. Additionally, some studies found that Black and non-white individuals with cancer were more likely to experience CRCI than their white counterparts. Biological, sociocultural, and instrumentation factors were associated with CRCI differences between racial and ethnic groups.

CONCLUSIONS

Our findings indicate that racial and ethnic minoritized individuals  may be disparately affected by CRCI. Future research should use standardized guidelines for measuring and reporting the self-identified racial and ethnic composition of the sample; differentiate CRCI findings by racial and ethnic subgroups; consider the influence of structural racism in health outcomes; and develop strategies to promote the participation of members of racial and ethnic minority groups.

Genome-wide variants and polygenic risk scores for cognitive impairment following blood or marrow transplantation

Cognitive impairment is prevalent in blood or marrow transplantation (BMT) recipients, albeit with inter-individual variability. We conducted a genome-wide association study of objective cognitive function assessed longitudinally in 239 adult BMT recipients for discovery and replicated in an independent cohort of 540 BMT survivors. Weighted genome-wide polygenic risk scores (PRS) were constructed using linkage disequilibrium pruned significant SNPs. Forty-four genome-wide significant SNPs were identified using additive (n = 3); codominant (n = 20) and genotype models (n = 21). Each additional copy of a risk allele was associated with a 0.28-point (p = 1.07 × 10-8) to a 1.82-point (p = 6.7 × 10-12) increase in a global deficit score. We replicated two SNPs (rs11634183 and rs12486041) with links to neural integrity. Patients in the top PRS quintile were at increased risk of cognitive impairment in discovery (RR = 1.95, 95%CI: 1.28-2.96, p = 0.002) and replication cohorts (OR = 1.84, 95%CI, 1.02-3.32, p = 0.043). Associations were stronger among individuals with lowest clinical risk for cognitive impairment. These findings support potential utility of PRS-based risk classification in the development of targeted interventions aimed at improving cognitive outcomes in BMT survivors.

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.

Cognitive adverse effects of chemotherapy and immunotherapy: are interventions within reach?

One in three people will be diagnosed with cancer during their lifetime. The community of cancer patients is growing, and several common cancers are becoming increasingly chronic; thus, cancer survivorship is an important part of health care. A large body of research indicates that cancer and cancer therapies are associated with cognitive impairment. This research has mainly concentrated on chemotherapy-associated cognitive impairment but, with the arrival of immunotherapies, the focus is expected to widen and the number of studies investigating the potential cognitive effects of these new therapies is rising. Meanwhile, patients with cognitive impairment and their healthcare providers are eagerly awaiting effective approaches to intervene against the cognitive effects of cancer treatment. In this Review, we take stock of the progress that has been made and discuss the steps that need to be taken to accelerate research into the biology underlying cognitive decline following chemotherapy and immunotherapy and to develop restorative and preventive interventions. We also provide recommendations to clinicians on how to best help patients who are currently experiencing cognitive impairment.

Elevated Oxidative Stress and DNA Damage in Cortical Neurons of Chemotherapy Patients

The unintended neurologic sequelae of chemotherapy contribute to significant patient morbidity. Chemotherapy-related cognitive impairment (CRCI) is observed in up to 80% of cancer patients treated with chemotherapy and involves multiple cognitive domains including executive functioning. The pathophysiology underlying CRCI and the neurotoxicity of chemotherapy is incompletely understood, but oxidative stress and DNA damage are highly plausible mechanisms based on preclinical data. Unfortunately, validating pathways relevant to CRCI in humans is limited by an absence of relevant neuropathologic studies of patient brain tissue. In the present study, we stained sections of frontal lobe autopsy tissue from cancer patients treated with chemotherapy (n = 15), cancer patients not treated with chemotherapy (n = 10), and patients without history of cancer (n = 10) for markers of oxidative stress (nitrotyrosine, 4-hydroxynonenal) and DNA damage (pH2AX, pATM). Cancer patients treated with chemotherapy had increased staining for markers of oxidative stress and DNA damage in frontal lobe cortical neurons compared to controls. We detected no statistically significant difference in oxidative stress and DNA damage by the duration between last administration of chemotherapy and death. The study highlights the potential relevance of oxidative stress and DNA damage in the pathophysiology of CRCI and the neurotoxicity of chemotherapy.

Neurocognitive Impairment After Hematopoietic Stem Cell Transplant for Hematologic Malignancies: Phenotype and Mechanisms

Hematopoietic stem cell transplant (HSCT) plays a central role in the treatment of hematologic cancers. With the increasing survival of patients after HSCT, survivorship issues experienced by this population have become an important outcome. Cognitive impairment is an established sequela of HSCT, with studies to date establishing its presence, associated risk factors, and clinical phenotype. There are multiple potential contributors to cognitive impairment after HSCT. Efforts are ongoing to further characterize its clinical phenotype, associated biomarkers, and biologic underpinnings. A fundamental knowledge of post-HSCT cognitive impairment is of value for all clinicians who interface with this population, and further academic efforts are needed to more fully understand the impact of this cancer treatment on brain health. IMPLICATIONS FOR PRACTICE: As survival outcomes after hematopoietic stem cell transplant (HSCT) improve, an awareness of the post-treatment challenges faced by this population has become central to its care. HSCT can have a sustained and broad impact on brain health, causing cognitive dysfunction, fatigue, disturbed mood, and sleep. In affected patients, autonomy, return to work, relationships, and quality of life may all be affected. A fundamental fluency in this area is important for clinicians interfacing with HSCT survivors, facilitating the identification and management of cognitive dysfunction and concurrent symptom clusters, and stimulating interest in these sequelae as areas for future clinical research.

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.