The personal impact of living with a myeloproliferative neoplasm

The aim of this study is to gain insight into the physical, psychological and social impact of having a myeloproliferative neoplasm (MPN), a rare type of cancer with an often chronic course.


| INTRODUCTION
Philadelphia-negative myeloproliferative neoplasms (MPNs) are rare haematopoietic diseases characterised by abnormal bone marrow proliferation.[3] The mutated malignant stem cells induce chronic inflammation, leading to a high symptom burden in a significant proportion of patients. 4MPNs are classified as cancers 5 with three main types: essential thrombocytosis (ET), polycythaemia vera (PV), and myelofibrosis (MF).
The disease is often chronic but can progress to acute myeloid leukaemia or myelodysplasia. 6,7MPNs are not curable, except by stem cell transplantation.Current medical treatment focuses primarily on preventing thromboembolic complications and reducing the risk of the disease progression. 1,80][11][12] The crude annual incidence rate of MPN ranges from 1.15 to 4.99 per 100,000. 13It is slightly more common in men than in women. 14e average age at diagnosis is between 65 and 75 years, 8 although MPNs are increasingly being diagnosed in younger people. 15o landmark studies on MPN-related symptoms, also known as MPN symptom burden, 1,9 have shown that patients with a high disease burden, often have a reduced quality of life (QoL).Fatigue is the most common and severe symptom. 1,9Pruritus and night sweats are also commonly reported.In addition, MPNs can affect the emotional and social well-being, such as anxiety, depression 16 or impairment at work. 1,9e impact of rare and chronic diseases such as MPNs, can make life complicated.An international literature study 17 shows that the rarity of a disease presents specific challenges, not only physically, but also psychologically and socially.Research by NIVEL 18 has shown that patients with a rare chronic disease have a poorer QoL than patients with a more common chronic disease.Half of the patients seemed to suffer from anxiety, depressive feelings, tension and social problems.It is not known how the suffering of MPN patients relates to more common cancers.
Research in chronic diseases has shown that QoL may be related to sociodemographic and disease-related characteristics.Being female, unmarried, less educated, older, receiving disability benefits and having a physical disability were associated with a poorer QoL. 19,20Limited research in MPNs has also shown that females perceive higher disease burden, 9,21,22 but results for QoL are conflicting. 21,22Conflicting results have also been found between the MPN subtypes when measuring the disease burden using the MPN Symptom Assessment Form Total Symptom Score (MPN-SAF TSS). 23,24e extent to which this disease impact occurs in MPN patients has not been investigated comprehensively across different domains of life.The aim of this study is to gain insight into the most important physical, psychological and social problems faced by people with MPN, in order to provide tools for targeted (psychosocial) interventions leading to improved QoL.
To better understand the impact and meaning of having an MPN, the Hierarchical QoL Model 25 provided the theoretical framework (Figure 1).This model distinguishes three domains of QoL, physical, psychological and social.In addition, a hierarchical distinction is made from general QoL to specific aspects of QoL, such as pain and social functioning.Previous research suggests that different domains of QoL may be affected, depending on the kind of disease. 26,27We slightly adapted this framework by using the validated EORTC QLQ-C30 28 to operationalise QoL.This questionnaire included specific sub-dimensions of QoL, denoted by 'A' in the figure.The most relevant and representative MPN symptoms (MPN-SAF TSS) 23 are also included in this model ('B').
This study focuses on the personal impact of having an MPN on QoL in general and in the different domains as presented in the model.How do these results relate to the impact of other cancer patients?Are there differences in the disease impact related to sociodemographic and disease-related characteristics of the MPN group?And, are there associations between symptoms, both within and between the domains?

| Design, participants and procedure
We conducted a cross-sectional survey among 455 Dutch MPN patients.We recruited them via the Dutch patient organisation and two Dutch private online platforms for MPN patients.After giving informed consent, participants were invited to complete an online questionnaire anonymously.Exclusion criteria were <18 years of age, not living in the Netherlands, having undergone a stem cell transplantation, and never been diagnosed with an MPN.Data were collected from 30 November 2022 to 7 February 2023.

| Measures
To gain insight into the impact of an MPN on daily life, we measured MPN symptoms, such as fatigue and pruritus, using the MPN-SAF TSS 23 (Figure 1).Each item has a score ranging from 0 (absent/as good as it gets) to 10 (worst imaginable/as bad as it gets).The TSS is calculated by summing the 10 MPN symptom scores, 23 hereafter called the MPN symptom burden (Cronbach's alpha = 0.84).QoL, and its subdomains, were assessed using the EORTC QLQ-C30, 26 a validated questionnaire for cancer patients.The 30 items were divided into five functional scales, three symptom scales, the overall QoL/global health status (renamed as QoL) and six single items (see also Figure 1).The first 28 items are rated on a 4-point Likert scale ranging from 1 (not at all) to 4 (very much); the last two questions are rated on a 7-point Likert scale ranging from 1 (not at all) to 7 (very much).The scores were converted to scales ranging from 0 to 100.A higher score on the QoL and functional scales reflects better health/well-being; a higher score on the symptom scales or single items represents a more severe condition. 29The reliability of QoL was (r = 0.83, p < 0.001), other subscales ranged from r = 0.54 to α = 0.88 (p < 0.001).Due to low reliability (r = 0.41, p < 0.001), 'nausea and vomiting' was excluded from further analysis.An additional question was added to measure the impact of MPN on ability to work, such as reduced working hours.
To gain insight into how the disease burden of MPN patients relates to that of other cancer patients, we compared our EORTC QLQ-C30 scores with data from the reference group of all cancer patients at all stages. 30 measure sociodemographic variables, questions were asked about age, sex, level of education (low, middle or high), and civil state (having a partner and living together or separated, being single).
To measure disease-related factors questions were asked about the type of MPN, time since diagnosis (in years), treatment (wait-and-see, phlebotomy, phlebotomy combined with cytoreduction or cytoreduction only), possible side effects (none, a little, a lot), complications (no or yes), and comorbidities (no or yes).

| Statistical analyses
Descriptive statistics were used to explore the MPN population, sociodemographic and disease-related characteristics, MPN symptom burden (individual items and total score), and (different aspects of) QoL in the three domains.T-tests were used to examine differences in disease impact between groups that differed in sociodemographic and disease-related variables or in the impact of MPN on working life.ANOVA analyses were performed to explore possible differences in disease impact between patients who differed in MPN type, treatment and side effects.Pearson's correlation tests were used to examine the associations between the disease impact within and across the domains of QoL, looking at the MPN symptom burden and (aspects of) QoL.Finally, hierarchical regression analysis was performed in three blocks using the 'enter' method to examine whether sociodemographic and disease-related characteristics (block 1), the MPN symptom burden (block 2), and subscales and single items of QoL (block 3) explained perceived QoL.
We used IBM SPSS Statistics (version 29) to conduct the analyses.
F I G U R E 1 Hierarchical QoL Model, 25 adapted to the EORTC QLQ-C30 26 and MPN-SAF-TSS, 23 with an additional question on impact on working life ('C').† Functional scales, ‡ Symptom scales.
The majority of patients reported suffering from side effects: 46.6% a little, 19.6% a lot.Complications occurred in 23.3% of the participants.Finally, 52.5% reported comorbidities (see Supporting Information S1 for additional patient characteristics).

| MPN symptom burden and QoL
Looking at the MPN-SAF TSS, a high number of patients (N = 443) reported having MPN symptoms, although there was variability in the extent of symptoms.Only 12 patients (2.6%) reported no symptoms.
Table 1 illustrates the variety of health problems experienced by MPN patients.
Considering the different domains of QoL (Figure 1), the most severe MPN symptom (MPN-SAF TSS) in the physical domain was fatigue, followed by inactivity.A high score for fatigue was also found in the EORTC QLQ-C30 measurement, followed by insomnia.In the psychological domain, patients reported problems with cognitive and emotional functioning (EORTC QLQ-C30) and concentration (MPN-SAF TSS).In the social domain, problems with social and role functioning (EORTC QLQ-C30) were reported.Almost 50% also reported that having an MPN affected their work life.
When we compared the EORTC QLQ-C30 scores of the MPN patients with the reference group of cancer patients (Table 1), it appeared that the MPN patients scored similarly to other cancer patients on many items.However, we also observed some differences: MPN patients seemed to suffer much more from fatigue, cognitive functioning and insomnia, but less from appetite loss and financial difficulties.

| Sociodemographic and disease-related differences in MPN symptom burden and QoL
Table 2 shows the sociodemographic and disease-related differences in the perceived MPN symptom burden and QoL.
The MPN symptom burden was significantly higher in women than in men.There were no significant differences in MPN symptom burden/QoL between the different MPN subtypes.However, the presence of comorbidities, complications or disease worsening was associated with a significantly higher MPN symptom burden and lower QoL than the absence of these conditions.
T A B L E 1 Scores on MPN-SAF-TSS, EORTC QLQ-C30 and impact on working life (N = 455).Been voluntarily terminated from job 36 7.9

Variable
Been involuntarily terminated from job 27 5.9 Gone on partial disability living allowance 35 7.7 The type of MPN treatment did not result in significant differences in the MPN symptom burden or QoL, but the extent to which patients experienced side effects did: MPN patients who reported no side effects scored significantly lower on MPN symptom burden and higher on QoL compared to patients who reported some or many side effects.No significant differences in MPN symptom burden/QoL were found in relation to education level.Finally, when an MPN interfered with patients' working life, MPN symptom burden was significantly higher and QoL was significantly lower compared with no interference with working life.

| Correlations and explained variance in QoL
Pearsons correlation test showed that the MPN symptom burden and subscales and single items of the EORTC were correlated with each other and with QoL (see Supporting Information S2).Some correlations were strong, such as between fatigue and QoL (−0.653, p < 0.001), MPN symptom burden (0.746, p < 0.001), and role functioning (−0.775, p < 0.001).Role and social functioning were also strongly correlated (0.755, p < 0.001).
In hierarchical regression analysis the final model explained almost 60% of the variance in QoL (R 2 = 0.58, F(33,389) = 16.06,p < 0.001).All three blocks were significant (Table 3).The sociodemographic and disease-related variables in block 1 explained the most variance (29%), the added variance of MPN symptom burden was 14% (block 2), and finally the EORTC subscales and single items added 15% of the explained variance in QoL (block 3).The variance of the following variables was significant in model 3: type of disease progression (worse vs. stable (−)), presence of comorbidities (none vs. yes (þ)), MPN symptom burden (−), social functioning (þ), emotional functioning (þ), role functioning (þ), dyspnoea (−), and financial difficulties (þ).A closer look at the subsequent models showed that ET (vs.PV(þ)), many side effects (vs.few (−)), and impact on working life (vs.none (−)) explained variance in QoL in both models 1 and 2, but that these effects disappeared when symptoms and subscales were added in model 3.The explained variance of MPN symptom burden (−) was large in model 2, but became significantly smaller in model 3.

| DISCUSSION
This cross-sectional study revealed that patients with MPN experience a high symptom burden.From the perspective of the hierarchical QoL model 25 negative personal consequences of the disease were found in all three domains of QoL.Consistent with previous research 1, 9 MPN patients suffered the most from fatigue, but also from problems related to concentration and cognitive functioning.
This suffering appears to be much higher than in the reference group of cancer patients in general.MPN patients also reported suffering more from insomnia, though less from some other symptoms such as loss of appetite than the reference group.Other outcomes, such as QoL, were comparable.The rare and chronic nature of MPN may have contributed to the burden of disease, as previous studies revealed. 17,18nsistent with previous findings, 1,9 our study showed that patients with MPN were limited in their ability to work.These patients reported significantly higher symptom burden and lower QoL compared to patients whose MPN did not affect their work life.The high percentage (48.8%) of patients who were restricted in their ability to work raises the question of what is behind this.Based on what has been found in the literature, 30 it is plausible that fatigue, which is a common complaint of patients in our study, plays an important role.It has been noted that cancer-related fatigue can have a profound impact on QoL, not only in terms of occupational functioning, but also in terms of personal and social roles.This may also be the case for MPN patients, as our study showed that fatigue in the physical domain was associated with worser role and social functioning in the social domain.
Another notable finding is the large explanation of variance in QoL (58%).Disease progression, comorbidities, role and social functioning and the MPN symptom burden significantly contributed to the explanation of QoL.Considering that higher MPN symptom burden was associated with lower QoL also in previous studies, 1,9 symptoms should be taken more seriously when treating patients, as well as the other explanatory factors.
Although sociodemographic variables did not contribute significantly to the explained variance in QoL, our study showed that female gender was associated with higher symptom burden, as found in previous studies. 9,22Notably, this did not result in a lower QoL compared to the male gender.Previous findings on this topic has shown conflicting results.One study 21 found poorer QoL in women, whereas another study 22 found no sex differences.Our results showed no significant differences in symptom burden or QoL by civil state or educational level, in contrast to a previous study in chronic diseases. 20oking at disease-related variables we found no significant differences in MPN symptom burden regarding MPN subtype.Previous research on this topic showed conflicting results. 23,24Also, perceived suffering was not related to the treatment patients received, but to other aspects of the disease, such as side effects, complications and comorbidities.

| Study limitations
This study is self-reported, which may bias the results, and the crosssectional design does not allow conclusions to be drawn about causality.
T The majority of participants were female, whereas males are more often diagnosed. 14Therefore, the results (i.e.women reported a significantly higher MPN symptom burden) may not be representative of the overall MPN population.This study did not identify the type of comorbidities that patients had.As it is plausible that certain comorbidities promote an inflammatory environment and thus increase the specific disease burden in MPN, 31,32  should be interpreted with caution and no conclusions or generalisations can be made.A final limitation concerns the way in which patients were recruited.Our sample may over-represent patients with a high symptom burden, as one study 9 showed that patients recruited through patient organisations had a higher symptom burden than those recruited by physicians.This, and the impossibility to obtain insight into the response rate due to our recruitment strategy, may have implications for generalisability to the general MPN population.

| Clinical implications
The variety of health complaints in the different domains and their impact on overall QoL reflects the suffering of MPN patients in their daily lives and their needs in these domains, for example, in terms of interventions and support.Although MPNs are a unique type of cancer with often a long life expectancy, the suffering of MPN patients seem to be similar to other cancer patient groups, and even worse in terms of fatigue and cognitive functioning.
Support is needed for all subtypes of MPN, not just the most severe, as the negative impact of the disease is similar across subtypes.Healthcare providers should add symptom assessment to regular blood testing to monitor disease burden.Gender differences, the impact of comorbidities and the rare and chronic nature of the disease should be considered.There may be a role for a specialist nurse with MPN expertise to guide the patient in this area, in addition to consultation with the specialist.The use of mobile applications may be considered to reduce the burden of MPN symptoms.

| CONCLUSIONS
The personal impact of an MPN on daily life can be profound, although there are sociodemographic and disease-related differences in the impact of the disease.
This study was approved by the Ethics Committee of the Faculty of Psychology of the Open Universiteit in the Netherlands (11-10-2022) and pre-registered in Open Science Framework (OSF, 29-02-2023).
Fatigue and cognitive problems were the most frequent and disabling symptoms.MPNs have a negative impact on QoL, regardless of MPN subtype or treatment option, with a large explained variance in QoL.Having comorbidities, MPN symptom burden and role, emotional and social functioning significantly contributed to this.Assessment of symptoms, social support and further research are needed to explore ways to improve QoL in MPN patients.
Regression analysis a showing predictors and explained variance in QoL (N = 423).