Predictive risk model of mild cognitive impairment in patients with malignant haematological diseases after haematopoietic stem cell transplantation.

Introduction

Hematopoietic stem cell transplantation (HSCT) is an effective treatment for numerous malignant and non-malignant hemopathies [ 1 ]. Since its inception as a treatment modality, more than one million HSCTs have been performed worldwide, with a substantial increase in the number of HSCT over the last three decades [ 2 ]. It remains the only curative therapy for many aggressive diseases, with high response rates and prolonged progression-free survival in relapsed chemosensitive patients [ 3 ]. However, the benefits of HSCT can be compromised by graft-versus-host disease (GVHD), regimen-related endothelial toxicity, and severe infection [ 4 , 5 ]. In addition, physical late effects including altered growth stature, chronic pain, and psychological late effects have all been reported [ 6 , 7 ]. Patients undergoing HSCT have been found to experience psychological distress, including depression, anxiety, distress, post-traumatic stress disorder, delirium, and cognitive impairment (CI), which can negatively affect recovery, function, and health outcomes, including mortality and increased risk of GVHD [ 8 , 9 ].

CI is characterised by deficits in executive function, memory, and attention and ranges in severity from mild cognitive impairment (MCI) to severe dementia [ 10 ]. MCI generally refers to cognitive impairment beyond that seen with normal age-related cognitive decline, but not severe enough to cause significant impairment of daily function. Previous studies have shown that HSCT has a lasting and widespread effect on brain health, causing cognitive dysfunction, fatigue, mood, and sleep disturbances. Patients’ autonomy, return to work, relationships, and quality of life may be affected [ 11 ]. Data have shown that changes in cognitive function have been identified as a significant problem among patients and caregivers following HSCT and as an established sequela of HSCT [ 12 ]. MCI may also affect survivors’ independence and sense of personal agency. Prediction and early diagnosis of MCI is important to improve patient survival outcomes.

Previous studies have investigated the relationship between MCI and HSCT. In a study of 69 HSCT survivors’ perceptions of CI and its impact on daily living function, Lacy et al. found that 49 (71%) reported cognitive impairment after transplantation, including 38 with memory impairment (55%), 29 with attention and concentration problems (42%), and smaller numbers in other aspects [ 13 ]. Studies on the time dimension found that deficits in learning, memory and motor functions were found one month after HSCT, and the above functions decreased almost half of HSCT recipients compared with those before HSCT [ 14 ]. These acute deficits often develop into chronic neurocognitive changes. When analysing populations of allogeneic and autogenous HSCT recipients 2 to 6 months after HSCT, declines were found in verbal memory, complex attention, writing motor processing speed, verbal fluency, and motor function [ 15 , 16 ]. At 3 months after HSCT, Jones and colleagues found that almost half of their 53 HSCT patients showed significant declines in learning, memory, and/or executive function compared with performance at 1 month [ 14 ]. In another group of HSCT recipients assessed 6 months after transplantation, half of the recipients had deficits in at least one cognitive domain, with psychomotor speed being most impaired, followed by complex executive function [ 17 ]. At 36 months after HSCT, patients who received full-intensity allogenic HSCT showed poorer performance in executive function, verbal speed, processing speed, auditory memory, and fine motor flexibility compared with healthy controls [ 18 ]. Studies have shown that isolation and prolonged hospitalization after HSCT affect patients’ cognitive function and be associated with adverse clinical outcomes [ 19 ]. Prediction model may be an effective way to identify individuals at high risk of MCI. There are many predictive models of risk factors associated with CI. However, these models have been limited by moderate accuracy and have mostly focused on mental function [ 20 ].

The aim of this study was to develop and validate a robust model for predicting the risk of MCI in patients with malignant haematological diseases after allogeneic HSCT and to provide clinicians with an evidence-based framework for the care of patients with MCI and their caregivers. This study is based on the Health Ecology Model (HEM) [ 21 ], which consists of five dimensions: individual characteristics, behaviours and lifestyles, interpersonal networks, work-life conditions, and policy environment. Factors regarding individual characteristics (gender, age, education, height, weight, BMI), behaviours and lifestyles (sleep disorders, anxiety, depression), interpersonal networks (marital status), work-life conditions, and policy environment (medical insurance) were selected as criteria in the present study to estimate cognitive impairment in patients with HSCT. Some drugs used in pre-treatment regimens before HSCT, including methotrexate, cytarabine [ 22 ] may upregulate the expression of specific inflammatory markers (tumour necrosis factor [TNF]-α, interleukin[IL]−1) [ 12 , 23 ], leading to neuroinflammation and cognitive dysfunction. Factors related to HSCT treatment may also influence cognitive outcomes. Factors such as prior chemotherapy regimens, donor relationship, length of hospital stay, and days before transplantation are associated with poorer cognitive outcomes [ 14 , 18 , 24 ].

GVHD can cause severe immune dysregulation, multi-organ dysfunction, and reduced quality of life. Research has shown [ 25 ] that chronic GVHD may affect psychological function after transplantation. Cancer patients and survivors often experience cognitive impairment alongside other symptoms such as sleep disturbance, fatigue, anxiety, and depression [ 26 – 28 ]. The occurrence of two or more of these symptoms is known as a psychoneurological symptom cluster [ 26 , 29 – 31 ]. Symptom clusters have a greater impact on quality of life and functional status than individual symptoms [ 32 – 34 ], and increased symptom burden is associated with decreased survival rates [ 35 , 36 ].

Materials and methods

Study design

This is a clinical follow-up study in which questionnaires were issued and collected in the first month of patients receiving HSCT. The questionnaires were completed independently by the patients. The pre-transplant data of patients with MCI confirmed by the questionnaire were collected, except for GVHD and HSCT comorbidities. The relevant clinical data were analysed to identify the risk factors and to build a prediction model (Fig. 1 ).

Patient selection

In this study, 282 patients with malignant haematological diseases from the affiliated Huaian No.1 People’s Hospital of Nanjing Medical University, the affiliated hospital of Xuzhou Medical University, and the First People’s Hospital of Yancheng City were enrolled between April 2019 and February 2022. Signed informed consent was obtained before the study procedures. This research was approved by the Ethics Committees of the affiliated Huaian No.1 People’s Hospital of Nanjing Medical University (2023-K-115) and conducted in accordance with the Declaration of Helsinki.

Eligible patients were as follows: (1) patients older than 18 years; (2) haematopoietic stem cell transplantation was completed; and (3) could complete the questionnaire survey. Patients with the following conditions were excluded: (1) mental impairment that could not cooperate with completion; (2) complicated with heart, brain, kidney, and other organ failure and other tumours; (3) confidential treatment; and (4) treatment discontinuation.

Sample size estimation

Logistic Events Per Variable (EPV) was used to calculate the sample size. When the Wald method was applied, only an EPV greater than 5 could ensure the stability of the results [ 37 ]. According to the previous investigation results, if the number of independent variables to be included in the binary regression model is 20, and the EPV is 6, then the number of cases in the positive group is at least (20 × 6) = 120. According to the data analysis results, the incidence of MCI after HSCT was 40%, and the sample size required for modelling the data was at least (20 × 5) ÷ 40% = 250 cases. In this study, the sample size of the final model was 282 cases.

Assessment of cognitive function

Cognitive function after HSCT was assessed with the Functional Assessment after Cancer Cognition (FACT-Cog) [ 38 ]. FACT-Cog is a questionnaire designed to assess perceived cognitive function and its impact on quality of life in cancer patients. It contains 33 items and ranges from 0 to 4, with higher scores indicating better function. Patients with scores between 33 and 66% of total sores were defined as having MCI. The Cronbach’s α of the FACT-Cog is 0.868, and the cumulative contribution rate is 64.416%, indicating good reliability and validity.

The Generalized Anxiety Disorder-7 has been studied to detect anxiety disorders in patients (GAD-7, scores range from 0 to 21 with higher scores indicating more severe illness) [ 39 ]. The GAD-7 was originally developed in English in 2006 as a brief screening tool to detect anxiety. Using the optimal cut-off point of 10, sensitivity reached 89.0%, and specificity was as high as 82.0%. To date, the GAD-7 is recommended as a screening tool for anxiety. In this study, the Cronbach’s α of GAD-7 is 0.799, and the cumulative contribution rate is 62.636%, indicating good reliability and validity.

Depression was assessed using the Patient Health Questionnaire-9 (PHQ-9) [ 40 ]. PHQ-9 is a nine-item questionnaire designed to screen for depression in primary care and other medical settings [ 40 ]. The standard cut-off score for screening to identify possible major depression is 10 or above. A previous study has found that PHQ-9 to be the most reliable screening for tool depression [ 41 ]. A study of the PHQ-9 in patients with malignant tumours showed good reliability and validity, with a cut-off value of 7, sensitivity and specificity of 83% and 61%, and positive and negative predictive values of 10% and 99%, respectively [ 42 ]. In this study, the Cronbach's α of PHQ-9 is 0.871, and the cumulative contribution rate is 49.529% indicating good reliability and validity.

The Pittsburgh Sleep Quality Index (PSQI) was used to reflect sleep quality (scores range from 0 to 21, with higher scores indicating worse sleep) [ 43 ]. The PSQI is the most widely used sleep quality questionnaire in research and clinical practice. This self-reported 18-item measure assesses sleep quality within the past month and includes a global score consisting of seven sleep components [ 44 ]. Psychometric studies have demonstrated that the PSQI is a reliable and valid standardized measure of sleep quality in clinical and nonclinical populations [ 45 ]. The Cronbach’s α of PSQI is 0.836, and cumulative contribution rate is 50.583% indicating good reliability and validity.

Data collection

Data included patient characteristics (age, medical insurance type, gender, marital status, education level, hospital stays, medications, and body mass index), haematological diagnosis, and transplant complications. In the first month after discharge, relevant questionnaires were administered and collected by full-time nurses and completed independently by the patients. Records shall be made after interrogating items of the scale one by one. After the investigation is completed, the questionnaire results shall be checked on the spot, and any questionnaire omitted or incorrectly filled out shall be found, and the patients shall be instructed to improve the questionnaire after obtaining their consent. At the same time, to ensure the accuracy, integrity, and authenticity of the data, the study implemented two-person input and verification; if errors or discrepancies are found, the researchers should be notified for timely correction and explanation. The primary endpoint was the incidence of MCI 1 month after HSCT. Cognitive function was assessed 1 month after hospital discharge with a battery of psychological tests. Four subscales were administered.

Statistical analysis

All analyses were performed with the Statistical Package (SPSS 26.0 Inc., Chicago, Illinois) and R software (version 3.5.3). Age, length, weight, body mass index (BMI), type of transplant, and hospital stays were compared using unpaired t test or Mann–Whitney U -test. Differences of categorical variables (medical insurance type, gender, marital status, education level, haematological diagnosis, medication, anxiety, depression, GVHD) were analysed using the Fisher exact test. Odds ratios (ORs) were calculated using logistic regression analysis. Univariate and multivariate logistic regression proportional hazards models were performed to identify significant prognostic predictors. The hazard ratio (HR) and 95% confidence interval (CI) were calculated. The significant variables with p < 0.05 defined in the univariate analyses (by log rank test) were included in the multivariate analyses to validate the prognostic value. The p value < 0.05 (two-tailed) indicated statistical significance.

Results

Characteristics of the participants

A total of 282 patients were included in the analyses. All sociodemographic and clinical characteristics of the sample are summarised in Table 1 . The median age was 48 (18–76) years and 150 (53.19%) were male. Subjects were divided into MCI ( N = 135) and non-MCI ( N = 147) cohorts. The distribution of characteristics such as age, medical insurance type, gender, marital status, education level, haematological diagnosis, medication, and body related indexes, such as height, weight, and body mass index (BMI), were not significantly different between the two groups. As cognitive functional characteristics parameters, the occurrence of anxiety and depression were different between two groups ( p < 0.001, p < 0.001, respectively). Moreover, transplant type and transplant complication, such as GVHD and hospital stays, were also different between the two groups ( p < 0.001, p < 0.001, p < 0.001, respectively).

Construction of predictive model

Univariate logistic regression analyses

A total of 282 patients in the training set were divided into cognitive impairment group and non-cognitive impairment group. There were no statistically significant differences in cognitive impairment among different types of medical insurance, gender, marriage, education, disease diagnosis, cyclophosphamide, cytarabine, methotrexate, cyclosporine, sleep disorders, weight, height, BMI, implantation days, etc. There were statistically significant differences in the occurrence of MCI between the two groups in terms of age, type of transplant, anxiety, depression, GVHD, length of hospital stays (time in transplant bin), etc., as shown in Table 1 .

Construction of risk prediction model for CI after HSCT in patients

The progressive forward (conditional) screening method was adopted, the inclusion criterion was 0.05, and the exclusion criterion was 0.10. Variables with P < 0.1 in the results of univariate analysis and those with potential effects shown in previous studies were taken as independent variables, including 11 variables such as gender, age, transplant type (dummy variable), anxiety, depression, disease diagnosis, educational background, GVHD, length of hospital stays, type of medical insurance, and BMI. The results of the regression are shown in Table 2 . Finally, 6 variables were entered into the model, and the results are shown in Table 3 . MCI risk scores were calculated based on normalised levels for each predictor. Finally, the predictive formula for calculating the risk score was as follows: MCI risk = −7.488 + (1.133* hospital stays) + (0.951* GAD-7 scores) + (1.582*PHQ-9 scores) + (1.165* GVHD) + (1.90* allogeneic HSCT) + (−0.212* microtransplantation) + (0.083* age), and is plotted as a nomogram (see Fig. 2 ) . For each selected variable, the corresponding points (0 ~ 100 points by default) are obtained on the scoring scale at the top of the column diagram through the vertical line, and then the total points of all variables are added to obtain the total score, and then the corresponding predicted risk value is obtained through the total score line on the prediction line at the bottom of the column diagram. The nomogram showed that age contributed the most to the prediction of MCI. The nomogram was evaluated using receiver operating characteristic (ROC) curves, and concordance index (C-index) and area under the curve (AUC) values were calculated to determine the predictive potential of the nomogram. Calibration was used to determine the consistency of the model, and decision curve analysis (DCA) was used to assess the clinical utility of the model. An area under the ROC curve of 0.5 to 0.7 indicates low accuracy of the model, > 0.7 to 0.9 indicates moderate accuracy, and ≥ 0.9 indicates high accuracy.

Internal evaluation of the model

The AUC for development data was 0.897 as shown in Fig. 3 . The results show that the models had excellent consistent performance in predicting the risk of MCI. The calibration curves are shown in Fig. 4 . The plots show high agreement between the predicted and observed probability of MCI. The DCA curve also shows that the model has good predictive power and stability (Fig. 5 ).

External verification of the prediction model

Data of patients with malignant haematological diseases between March 2022 and July 2023 from the affiliated Huaian No.1 people’s hospital of Nanjing Medical University were collected as an external cohort to validate the prognostic value of the model. MCI scores were calculated using the same formula for patients in the dataset and ROC analysis showed that the AUC was 0.789 (Fig. 6 ). The calibration curves are shown in Fig. 7 . The DCA curve also shows that the model has good predictive ability and stability (Fig. 8 ).

Discussion

As HSCT technology continues to mature, an increasing number of people are undergoing this procedure, leading to a gradual increase in side effects. Post-transplant cognitive impairment is a particularly concerning area as it has a significantly impact on patients’ quality of life in aspects such as social reintegration, and the ability to achieve occupational or educational goals [ 46 – 48 ]. Early identification and intervention for cognitive impairment are crucial for this population. This study aims to develop and validate a simple and reliable predictive model for the risk of cognitive impairment in HSCT recipients, aiding clinical professionals in promptly and accurately identifying high-risk individuals for early intervention.

The research indicates that older age correlates with a higher likelihood of cognitive impairment after transplantation. Consistent with similar studies [ 24 ], the prevalence of subjective cognitive impairment increases with age, possibly due to brain atrophy, cortical thinning, particularly in regions closely related to cognition such as the temporoparietal and hippocampal areas [ 49 ]. Additionally, age-related reduction in neurotransmitters in the nervous system contribute to decreased cognitive reserve.

This study found that cognitive impairments was more likely to occur after allogeneic transplantation compared with autologous transplantation, a finding that is broadly consistent with previous research [ 18 , 24 ]. The reason for this may be related to the increased likelihood of developing GVHD after allogeneic transplantation. GVHD is a common complication of allogeneic HSCT, with an incidence rate of 50–70%, in which donor T lymphocytes recognise tissue compatibility antigens on host cells, leading to secondary inflammatory reactions. GVHD can lead to severe immune dysregulation, multiple organ failure, and reduced quality of life. Studies have shown that patients with severe GVHD are more susceptible to cognitive impairments [ 50 ]; chronic GVHD may also affect psychological function after transplantation [ 51 ]. Additionally, research suggests [ 52 ] that after allogeneic HSCT, activated T lymphocytes from the donor source can cross the blood–brain barrier and enter the central nervous system. T-cells activation is the effector output unit of GVHD and a primary factor triggering multisystem dysfunction, mainly by interacting with small glial cells in the central nervous system, particularly in the hippocampal area long-term potentiation (LTP).

This study was the first to look at the relationship between microtransplantation and cognitive impairment. It found that compared with autologous transplantation, cognitive impairment was more likely to occur after autologous transplantation. Micro transplantation (MST) was first proposed internationally in 2011 by China’s Ai Huisheng and Guo Mei team as a new cellular transplantation and immunotherapy for leukaemia [ 53 – 55 ]. It differs from conventional allogeneic transplantation in that MST emphasises the establishment of donor microchimerism in the patient’s body (donor ratio < 1%), without the establishment of complete donor engraftment. MST involves the infusion of granulocyte colony-stimulating factor (G-CSF)-mobilized human leukocyte antigen (HLA) disparate donor hematopoietic stem cells without GVHD prevention, with the aim of inducing transplant anti-tumour (leukaemia) and/or recipient anti-tumour (leukaemia) effects while avoiding GVHD. MST pretreatment focuses on preserving the recipient's immune function while clearing tumours and leukaemia cells as much as possible. MST is simple, safe, cost-effective, and has shown good efficacy, with successful application in numerous haematological treatment centres worldwide [ 56 – 59 ]. With an increasing number of MST cases in recent years, this study has added research on the impact of MST on cognitive impairment. The reasons for this may be related to the minimal drug toxicity of MST, the extremely low incidence of GVHD, and the rapid hematopoietic recovery.

In this study, we found that anxiety and depression increased the risk of cognitive impairment after HSCT. It has been reported that depression, anxiety, and sleep disturbance worsen cognitive function in cancer patients [ 60 – 64 ]. Conversely, cognitive decline exacerbates depression and anxiety, with implications for the treatment of these symptoms [ 65 – 67 ]. Our study also investigated the impact of sleep disturbance on cognitive impairment after transplantation, but it did not show statistical significance in univariate analysis, possibly due to differences in the assessment scales used in similar studies. It was also observed that the risk of cognitive impairment after HSCT increased with longer hospital stays. This finding is consistent with similar research results [ 24 ]. The reason for this could be that patients with longer hospital stays are mostly associated with slower recovery after allogeneic transplantation and are patients with GVHD.

The ROC curve area under the curve of the model is 0.897 in the modelling group and 0.789 in the validation group, indicating the high accuracy of the model. It can effectively differentiate between post-transplant patients with and without cognitive impairment in clinical practice. The calibration curve closely aligns with the diagonal line (ideal curve), indicating good consistency of the model. The DCA curve also shows that the model has good predictive ability and stability. Overall, considering various analyses of predictive performance, the model demonstrates good risk prediction value with good stability, and it holds certain clinical utility.

This study has some limitations. Patient cognition, anxiety, and depression are not routinely measured, so it is not known whether these factors affect cognitive impairment before HSCT, and the follow-up time cannot be relatively fixed, which is limited by the outpatient follow-up time, therefore measuring patient cognition, anxiety, depression, sleep, and fixed follow-up time to provide more reliable data for further optimisation of this prediction model.

Conclusion

The risk prediction model for MCI after HSCT in patients with malignant haematological diseases developed in this study is easy to use and has good predictive power. Staff can use the model to see patients on the first day of HSCT, during and after HSCT, identify individuals at high risk of post-transplant MCI in a timely manner and intervene early, and focus limited care resources on high-risk groups. Early risk prediction, assessment, and management of MCI after HSCT can be initially achieved with this model.

Author contribution

AUTHOR CONTRIBUTIONS Study design: S.C and XW.L Data collection: Y.Z, YY. F, LL.S, XQ. Q, TT. C, Y. L. Data analysis: S. C Manuscript writing: S.C. Participation in revising section, language editing:Y.J. Writing—review and editing: X. L

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interest

The authors declare no competing interests.

Footnotes

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Saha A, Blazar BR (2022) Antibody based conditioning for allogeneic hematopoietic stem cell transplantation. Front Immunol 13

2. Gratwohl A, Pasquini MC, Aljurf M et al (2015) One million haemopoietic stem-cell transplants: a retrospective observational study. Lancet Haematol 2(3):e91-100

3. Yang G, Wang X, Huang S et al (2022) Generalist in allogeneic hematopoietic stem cell transplantation for MDS or AML: epigenetic therapy. Front Immunol 13

4. van Lier YF, Vos J, Blom B, Hazenberg MD (2023) Allogeneic hematopoietic cell transplantation, the microbiome, and graft-versus-host disease. Gut Microbes 15(1)

5. Jerry Teng CL, Wang PN, Chen YC, Ko BS (2021) Cytomegalovirus management after allogeneic hematopoietic stem cell transplantation: a mini-review. J Microbiol Immunol Infect 54(3):341–348

6. Sinatora F, Traverso A, Zanato S et al (2017) Quality of life and psychopathology in adults who underwent hematopoietic stem cell transplantation (HSCT) in childhood: a qualitative and quantitative analysis. Front Psychol 8

7. Wilhelmsson M, Vatanen A, Borgström B et al (2015) Adverse health events and late mortality after pediatric allogeneic hematopoietic SCT-two decades of longitudinal follow-up. Bone Marrow Transplant 50(6):850–857

8. Nakamura ZM, Nash RP, Quillen LJ, Richardson DR, McCall RC, Park EM (2019) Psychiatric care in hematopoietic stem cell transplantation. Psychosomatics 60(3):227–237

9. Amonoo HL, Massey CN, Freedman ME et al (2019) Psychological considerations in hematopoietic stem cell transplantation. Psychosomatics 60(4):331–342

10. Viggiano D, Wagner CA, Martino G et al (2020) Mechanisms of cognitive dysfunction in CKD. Nat Rev Nephrol 16(8):452–469

11. Harrison RA, Sharafeldin N, Rexer JL et al (2021) Neurocognitive impairment after hematopoietic stem cell transplant for hematologic malignancies: phenotype and mechanisms. Oncologist 26(11):e2021–e2033

12. Buchbinder D, Kelly DL, Duarte RF et al (2018) Neurocognitive dysfunction in hematopoietic cell transplant recipients: expert review from the late effects and Quality of Life Working Committee of the CIBMTR and complications and Quality of Life Working Party of the EBMT. Bone Marrow Transplant 53(5):535–555

13. Lacy M, Fong M, Bolton C, Maranzano M, Bishop M, Artz A (2021) Cognitive functioning of older adults prior to hematopoietic stem cell transplantation. Bone Marrow Transplant 56(10):2575–2581

14. Jones D, Vichaya EG, Wang XS, Sailors MH, Cleeland CS, Wefel JS (2013) Acute cognitive impairment in patients with multiple myeloma undergoing autologous hematopoietic stem cell transplant. Cancer 119(23):4188–4195

15. Friedman MA, Fernandez M, Wefel JS, Myszka KA, Champlin RE, Meyers CA (2009) Course of cognitive decline in hematopoietic stem cell transplantation: a within-subjects design. Arch Clin Neuropsychol 24(7):689–698

16. Syrjala KL, Langer SL, Abrams JR et al (2004) Recovery and long-term function after hematopoietic cell transplantation for leukemia or lymphoma. JAMA 291(19):2335–2343

17. Booth-Jones M, Jacobsen PB, Ransom S, Soety E (2005) Characteristics and correlates of cognitive functioning following bone marrow transplantation. Bone Marrow Transplant 36(8):695–702

18. Sharafeldin N, Bosworth A, Patel SK et al (2018) Cognitive functioning after hematopoietic cell transplantation for hematologic malignancy: results from a prospective longitudinal study. J Clin Oncol 36(5):463–475

19. Smith PJ, Lew M, Lowder Y et al (2022) Cognitive impairment in candidates for allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplant 57(1):89–94

20. Mayo SJ, Wozniczka I, Brennenstuhl S et al (2021) Late cognitive outcomes among allogeneic stem cell transplant survivors: follow-up data from a 6-year longitudinal study. Support Care Cancer 29(5):2621–2630

21. Guo L, Zhang M, Namassevayam G et al (2023) Effectiveness of health management among individuals at high risk of stroke: an intervention study based on the health ecology model and self-determination theory (HEM-SDT). Heliyon 9(11)

22. Correa DD, Wang Y, West JD et al (2016) Prospective assessment of white matter integrity in adult stem cell transplant recipients. Brain Imaging Behav 10(2):486–496

23. Kelly DL, Buchbinder D, Duarte RF et al (2018) Neurocognitive dysfunction in hematopoietic cell transplant recipients: expert review from the late effects and quality of life working committee of the center for international blood and marrow transplant research and complications and quality of life working party of the European Society for Blood and Marrow Transplantation. Biol Blood Marrow Transplant 24(2):228–241

24. Jim HS, Small B, Hartman S et al (2012) Clinical predictors of cognitive function in adults treated with hematopoietic cell transplantation. Cancer 118(13):3407–3416

25. Kurosawa S, Oshima K, Yamaguchi T et al (2017) Quality of life after allogeneic hematopoietic cell transplantation according to affected organ and severity of chronic graft-versus-host disease. Biol Blood Marrow Transplant 23(10):1749–1758

26. Kim HJ, Barsevick AM, Fang CY, Miaskowski C (2012) Common biological pathways underlying the psychoneurological symptom cluster in cancer patients. Cancer Nurs 35(6):E1–E20

27. Tometich DB, Small BJ, Carroll JE et al (2019) Pretreatment psychoneurological symptoms and their association with longitudinal cognitive function and quality of life in older breast cancer survivors. J Pain Symptom Manage 57(3):596–606

28. Mandelblatt JS, Zhai W, Ahn J et al (2020) Symptom burden among older breast cancer survivors: The Thinking and Living With Cancer (TLC) study. Cancer 126(6):1183–1192

29. Kim HJ, McGuire DB, Tulman L, Barsevick AM (2005) Symptom clusters: concept analysis and clinical implications for cancer nursing. Cancer Nurs 28(4):270–282 ( quiz 283–4 )

30. Dodd MJ, Miaskowski C, Paul SM (2001) Symptom clusters and their effect on the functional status of patients with cancer. Oncol Nurs Forum 28(3):465–470

31. Starkweather AR, Lyon DE, Elswick RK Jr, Montpetit AJ, Conley Y, McCain NL (2013) A Conceptual model of psychoneurological symptom cluster variation in women with breast cancer: bringing nursing research to personalized medicine. Curr Pharmacogenomics Person Med 11(3):224–230

32. Dodd MJ, Cho MH, Cooper BA, Miaskowski C (2010) The effect of symptom clusters on functional status and quality of life in women with breast cancer. Eur J Oncol Nurs 14(2):101–110

33. Cleeland CS, Mayer M, Dreyer NA et al (2014) Impact of symptom burden on work-related abilities in patients with locally recurrent or metastatic breast cancer: Results from a substudy of the VIRGO observational cohort study. Breast 23(6):763–769

34. Kim HJ, Barsevick AM, Beck SL, Dudley W (2012) Clinical subgroups of a psychoneurologic symptom cluster in women receiving treatment for breast cancer: a secondary analysis. Oncol Nurs Forum 39(1):E20-30

35. Jiménez A, Madero R, Alonso A et al (2011) Symptom clusters in advanced cancer. J Pain Symptom Manage 42(1):24–31

36. Aktas A, Walsh D, Rybicki L (2012) Symptom clusters and prognosis in advanced cancer. Support Care Cancer 20(11):2837–2843

37. Peduzzi P, Concato J, Feinstein AR, Holford TR (1995) Importance of events per independent variable in proportional hazards regression analysis. II. Accuracy and precision of regression estimates. J Clin Epidemiol. 48(12):1503–1510

38. Costa D, Loh V, Birney DP et al (2018) The structure of the FACT-Cog v3 in cancer patients, students, and older adults. J Pain Symptom Manage 55(4):1173–1178

39. Byrd-Bredbenner C, Eck K, Quick V (2021) GAD-7, GAD-2, and GAD-mini: psychometric properties and norms of university students in the United States. Gen Hosp Psychiatry 69:61–66

40. Levis B, Benedetti A, Thombs BD (2019) Accuracy of Patient Health Questionnaire-9 (PHQ-9) for screening to detect major depression: individual participant data meta-analysis. BMJ 365

41. Costantini L, Pasquarella C, Odone A et al (2021) Screening for depression in primary care with Patient Health Questionnaire-9 (PHQ-9): A systematic review. J Affect Disord 279:473–483

42. Hartung TJ, Friedrich M, Johansen C et al (2017) The Hospital Anxiety and Depression Scale (HADS) and the 9-item Patient Health Questionnaire (PHQ-9) as screening instruments for depression in patients with cancer. Cancer 123(21):4236–4243

43. Pilz LK, Keller LK, Lenssen D, Roenneberg T (2018) Time to rethink sleep quality: PSQI scores reflect sleep quality on workdays. Sleep 41(5):zsy029

44. Sancho-Domingo C, Carballo JL, Coloma-Carmona A, Buysse DJ (2021) Brief version of the Pittsburgh Sleep Quality Index (B-PSQI) and measurement invariance across gender and age in a population-based sample. Psychol Assess 33(2):111–121

45. Manzar MD, BaHammam AS, Hameed UA et al (2018) Dimensionality of the Pittsburgh Sleep Quality Index: a systematic review. Health Qual Life Outcomes 16(1):89

46. Ahles TA, Saykin A (2001) Cognitive effects of standard-dose chemotherapy in patients with cancer. Cancer Invest 19(8):812–820

47. Ahles TA, Saykin AJ, Noll WW et al (2003) The relationship of APOE genotype to neuropsychological performance in long-term cancer survivors treated with standard dose chemotherapy. Psychooncology 12(6):612–619

48. Fliessbach K, Urbach H, Helmstaedter C et al (2003) Cognitive performance and magnetic resonance imaging findings after high-dose systemic and intraventricular chemotherapy for primary central nervous system lymphoma. Arch Neurol 60(4):563–568

49. Fleischman DA, Leurgans S, Arfanakis K et al (2014) Gray-matter macrostructure in cognitively healthy older persons: associations with age and cognition. Brain Struct Funct 219(6):2029–2049

50. Monk JP, Phillips G, Waite R et al (2006) Assessment of tumor necrosis factor alpha blockade as an intervention to improve tolerability of dose-intensive chemotherapy in cancer patients. J Clin Oncol 24(12):1852–1859

51. Braamse AM, Yi JC, Visser OJ et al (2016) Developing a risk prediction model for long-term physical and psychological functioning after hematopoietic cell transplantation. Biol Blood Marrow Transplant 22(3):549–556

52. Rizzo JD, Wingard JR, Tichelli A et al (2006) Recommended screening and preventive practices for long-term survivors after hematopoietic cell transplantation: joint recommendations of the European Group for Blood and Marrow Transplantation, the Center for International Blood and Marrow Transplant Research, and the American Society of Blood and Marrow Transplantation. Biol Blood Marrow Transplant 12(2):138–151

53. Guo M, Hu KX, Yu CL et al (2011) Infusion of HLA-mismatched peripheral blood stem cells improves the outcome of chemotherapy for acute myeloid leukemia in elderly patients. Blood 117(3):936–941

54. (2012) HLA-mismatched stem-cell microtransplantation as postremission therapy for acute myeloid leukemia: long-term follow-up. J Clin Oncol Off J Am Soc Clin Oncol 30(33): 4084–90

55. Guo M, Chao NJ, Li JY et al (2018) HLA-mismatched microtransplant in older patients newly diagnosed with acute myeloid leukemia: results from the microtransplantation interest group. JAMA Oncol 4(1):54

56. Lindbergh CA, Dishman RK, Miller LS (2016) Functional disability in mild cognitive impairment: a systematic review and meta-analysis. Neuropsychol Rev 26(2):129–159

57. Hu KX, Du X, Guo M et al (2020) Comparative study of micro-transplantation from HLA fully mismatched unrelated and partly matched related donors in acute myeloid leukemia. Am J Hematol 95(6):630–636

58. Cai B, Sun Q, Qiao J et al (2020) Comparison of microtransplantation, chemotherapy and allogeneic transplantation in post-remission therapy for Philadelphia chromosome-positive acute lymphoblastic leukemia. Am J Cancer Res 10(11):3852–3866

59. Pan B, Lazarus HM, Gale RP (2020) Microtransplantation for acute myeloid leukemia: a systematic review. JAMA Oncol 6(10):1614–1620

60. Yang Y, Hendrix CC (2018) Cancer-related cognitive impairment in breast cancer patients: influences of psychological variables. Asia Pac J Oncol Nurs 5(3):296–306

61. Bedillion MF, Ansell EB, Thomas GA (2019) Cancer treatment effects on cognition and depression: the moderating role of physical activity. Breast 44:73–80

62. Henneghan AM, Carter P, Stuifbergan A, Parmelee B, Kesler S (2018) Relationships between self-reported sleep quality components and cognitive functioning in breast cancer survivors up to 10 years following chemotherapy. Psychooncology 27(8):1937–1943

63. Henneghan A, Stuifbergen A, Becker H, Kesler S, King E (2018) Modifiable correlates of perceived cognitive function in breast cancer survivors up to 10 years after chemotherapy completion. J Cancer Surviv 12(2):224–233

64. Tager FA, McKinley PS, Schnabel FR et al (2010) The cognitive effects of chemotherapy in post-menopausal breast cancer patients: a controlled longitudinal study. Breast Cancer Res Treat 123(1):25–34

65. Caudle DD, Senior AC, Wetherell JL et al (2007) Cognitive errors, symptom severity, and response to cognitive behavior therapy in older adults with generalized anxiety disorder. Am J Geriatr Psychiatry 15(8):680–689

66. Thompson DG, Kesler SR, Sudheimer K et al (2015) FMRI activation during executive function predicts response to cognitive behavioral therapy in older, depressed adults. Am J Geriatr Psychiatry 23(1):13–22

67. Vinkers DJ, Gussekloo J, Stek ML, Westendorp RG, van der Mast RC (2004) Temporal relation between depression and cognitive impairment in old age: prospective population based study. BMJ 329(7471):881

Associated Data

Data Availability Statement

No datasets were generated or analysed during the current study.