Longitudinal Patterns of Fatigue in Long-Term Survivors of Childhood and Adolescent Cancers: A Report From the Swiss Childhood Cancer Survivor Study

Pediatric Blood & Cancer RESEARCH ARTICLE Longitudinal Patterns of Fatigue in Long-Term Survivors of Childhood and Adolescent Cancers: A Report From the Swiss Childhood Cancer Survivor Study Salome Christen1Luzius Mader2André O. von Bueren3,4Eva Maria Tinner5,6Grit Sommer7 Christina Schindera7,8Claudia E. Kuehni5,7Katharina Roser1Gisela Michel1 1Faculty of Health Sciences and Medicine, University of Lucerne, Lucerne, Switzerland 2Cancer Registry Bern Solothurn, University of Bern, Bern, Switzerland 3Division of General Pediatrics, Pediatric Hematology and Oncology Unit, Department of Pediatrics, Gynecology and Obstetrics, University Hospitals of Geneva, Geneva, Switzerland 4Department of Pediatrics, Gynecology and Obstetrics, CANSEARCH Research Laboratory, Faculty of Medicine, University of Geneva, Geneva, Switzerland 5Division of Pediatric Hematology/Oncology, Department of Pediatrics, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland 6Cantonal Hospital Baselland, University Institute of Internal Medicine, Liestal, Switzerland 7Childhood Cancer Research Group, Institute of Social and Preventive Medicine, University of Bern, Bern, Switzerland 8Division of Paediatric Oncology/Haematology, University Children’s Hospital Basel, University of Basel, Basel, Switzerland Correspondence: Gisela Michel ([email protected]) Received: 26 July 2025 Revised: 24 September 2025 Accepted: 24 September 2025 Funding: The SCCSS was supported by the Swiss Cancer League and Swiss Cancer Research (KLS-3644-02-2015, KFS-4722-02-2019, KLS/KFS-482501-2019, KLS/KFS-5711-01-2022, KFS-6105-02-2024), Kinderkrebshilfe Schweiz (www.kinderkrebshilfe.ch), Kinderkrebs Schweiz (www.kinderkrebs-schweiz.ch)and Stiftung für krebskranke Kinder - Regio Basiliensis (https://www.stiftung-kinderkrebs.ch). The data collection of the general population was funded by the Swiss National Science Foundation (100019_153268 / 1). S.C. has received funding from Vontobel-Stiftung, Krebsliga Zentralschweiz, and Avenira Stiftung. Keywords: childhood cancer | fatigue | longitudinal | registry | survivorship | Switzerland ABSTRACT Background: Fatigue negatively affects quality of life. We aimed to compare the prevalence of fatigue in survivors of childhood cancer with the Swiss general population, describe longitudinal patterns of fatigue, and identify characteristics associated with persistent fatigue in survivors. Procedure: In this cohort study, we used data from the Swiss Childhood Cancer Registry and the Swiss Childhood Cancer Survivor Study, including survivors (≥5 years since diagnosis; diagnosed between 1976 and 2015 at <20 years of age) aged ≥20 years at study entry, using data from the baseline and follow-up survey. A representative sample of the general population was used as a comparison group. Fatigue prevalence and fatigue severity were measured using the SF-36 vitality scale, from which we derived longitudinal patterns (no/low fatigue, late onset,improving,persistent). We used multivariable logistic regression to identify clinical, psychosocial and demographic characteristics associated with persistent fatigue. Results: Overall, 1846 survivors participated at baseline (52% male), and 684 survivors also participated at follow-up (median 9 years from baseline; 52% male). From the general population, 863 persons participated (42% male). Survivors had similar fatigue prevalence at baseline/follow-up (26%/29%) as the general population (26%). No/Low fatigue was experienced by 64%, late onset by Abbreviations: BMI, body mass index; BSI-18, Brief Symptom Inventory-18; CNS, central nervous system; IQR, interquartile range; OR, odds ratio; SCCSS, Swiss Childhood Cancer Survivor Study; SF-36, Short Form-36; SMN, second malignant neoplasm. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2025 The Author(s). Pediatric Blood & Cancer published by Wiley Periodicals LLC. This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA. Pediatric Blood & Cancer,2025;0:e32110 https://doi.org/10.1002/pbc.32110 1of12 14%, improving by 7%, and persistent fatigue by 15% of survivors. More late effects, psychological distress, pain, and less time spent on moderate-intensity physical activity were associated with persistent fatigue. Conclusions: This study provides data on longitudinal patterns of fatigue in survivors and identifies factors associated with persistent fatigue that can be used to identify survivors at risk and as a target for interventions aimed at improving fatigue. 1 Introduction Fatigue is prevalent among survivors of childhood and adolescent cancers, and significantly impacts survivors’ functioning and quality of life [1–3]. Cancer-Related fatigue is defined by the National Comprehensive Cancer Network as a ‘distressing, persistent, subjective sense of physical, emotional and/or cognitive tiredness or exhaustion related to cancer or cancer treatment that is not proportional to recent activity and interferes with usual functioning’ [4]. Fatigue negatively affects participation in activities of daily life [2, 5], working ability [6], and quality of life [2] in survivors. At the same time, it is often difficult for survivors with persisting fatigue to get their symptoms and condition recognised [7] or to obtain financial support from health or disability insurances [4]. More than 50% of patients experience fatigue during treatment for childhood and adolescent cancers [8, 9], and during treatment, fatigue is perceived as the most severe and interfering symptomatic adverse effect by children and adolescents affected by cancer [10]. Fatigue generally improves after treatment [11], but many survivors experience fatigue in the long-term [1, 3, 12]. In a previous systematic review, we found that the prevalence of fatigue in survivors ranges from 10% to 85% [1], but limited data are available from Switzerland [3]. The longitudinal patterns of fatigue in long-term survivors (≥5 years) of childhood and adolescent cancer are unclear, and several studies have called for research that clarifies the course of fatigue [1, 4, 11]. Four studies have investigated longitudinal patterns of fatigue in survivors of childhood cancer around treatment end [11, 13–15]. Fewer studies have investigated this in long-term survivors: A Norwegian study found that, in 5-year survivors of lymphomas and leukaemia, 31% of survivors were persistently fatigued [16]. A Dutch study in brain tumour survivors found that cognitive fatigue worsened over time, whereas general fatigue first improved, then worsened [17]. However, both studies focused on selected diagnostic groups. As fatigue can occur after any type of cancer [1], it is relevant to assess longitudinal patterns for all diagnoses. We used data from a nationwide, population-based cohort study, the Swiss Childhood Cancer Survivor Study (SCCSS) [18], to (a) compare prevalence and severity of fatigue in survivors to the general population, (b) describe longitudinal patterns of fatigue in survivors, and (c) identify characteristics associated with persistent fatigue in survivors. 2 Methods 2.1 Study Population and Procedure 2.1.1 Survivor Population The Childhood Cancer Registry (https://www. childhoodcancerregistry.ch/) is a nationwide, populationbased cancer registry for all Swiss residents diagnosed younger than 20 years of age with leukaemia, lymphoma, central nervous system (CNS) tumours, malignant solid tumours or Langerhans cell histiocytosis since 1976 [19]. The SCCSS is a nationwide, population-based, prospective cohort study including all children and adolescents registered in the Childhood Cancer Registry who survived at least 5 years since diagnosis (https://www.swiss-ccss.ch/en/)[18]. For the current analysis, we included survivors aged ≥20 years at study entry because of a different survey version used for younger survivors. A detailed description of the SCCSS is provided elsewhere [18]. In short, survivors first received a letter from their former treatment centre with study information and the option to refuse participation. Two weeks later, the SCCSS research team at the University of Bern sent survivors a paper-based questionnaire with a pre-paid return envelope. If they did not respond, they received a reminder letter with another questionnaire 4–6 weeks later. Non-responding survivors were contacted by phone for a second reminder. Baseline data were collected from 2007 to 2022, with an overall response rate of 56%. Follow-up data were collected from 2016 to 2022 from participants who had completed the baseline survey (response rate of the follow-up questionnaire 53%). 2.1.2 General Population Previously, our group collected health-related data from a random and representative sample of the Swiss general population in the context of another study [20, 21]. The sample was obtained from the Swiss Federal Statistical Office. It was drawn according to the representative distribution of age, sex and language region in Switzerland and included 3000 households (n=7052 persons). Household members were eligible if they were aged 18–75 years in 2015 (n=5644 eligible persons). They were contacted individually by postal mail from May 2015 to June 2016. Overall, 1255 persons participated in the survey (response rate 24%). For the current 2of12 Pediatric Blood & Cancer,2025 analysis, we restricted the sample to participants within the same age range as the survivor population (aged ≥20 and <60 years at study). 2.2 Measurements 2.2.1 Fatigue Fatigue was measured with the vitality scale of the Short Form36 (SF-36) Version 2 [22]. The SF-36 vitality scale is a reliable and valid measure of fatigue [23], is sensitive to change [23], and has been validated and successfully used in survivors of childhood cancer [24]. Participants were asked, ‘How much of the time during the past 4 weeks (a) ... have you felt full of life?’; ‘(b) ... did you have a lot of energy?’; ‘(c) ...did you feel worn out?’; and ‘(d) ...did you feel tired?’. Participants could choose from a five-point Likert scale ranging from ‘never’ to ‘all of the time’. We scored the SF-36 and subscales according to the scoring manual [25], and generated standardised scores ranging from 0 to 100. Lower scores on the SF-36 vitality scale indicate greater fatigue (fatigue severity). Scores ≤45 have been established as representing clinically significant fatigue [26]. We generated a caseness variable for fatigue prevalence, with participants scoring ≤45 as having fatigue. We generated longitudinal fatigue patterns based on fatigue prevalence at baseline and follow-up: no/low fatigue (no fatigue at baseline and follow-up), late-onset fatigue (no fatigue at baseline, fatigue at follow-up), improving fatigue (fatigue at baseline, no fatigue at follow-up), and persistent fatigue (fatigue at baseline and follow-up). The SF-36 was administered twice to the SCCSS participants (at baseline and follow-up), and once to the general population. 2.2.2 Explanatory Variables We obtained prospectively collected information from the Childhood Cancer Registry: sex (male, female), cancer diagnosis (according to the International Classification of Childhood Cancer - third edition) [27], and recoded into the major tumour groups (leukaemia, lymphoma, CNS tumours, other tumours), surgery (yes/no), chemotherapy (yes/no), radiotherapy (yes/no), stem cell transplantation (yes/no), relapse or second malignant neoplasm (SMN; yes/no), age at diagnosis (years), and year of diagnosis (1976–2015, 10-year increments). The following data were assessed at baseline in the SCCSS: health status (good–excellent, fair–poor), any late effects (yes, no), organ-specific late effects (yes, no; assessed for neurological, musculoskeletal, cardiovascular, vision, hearing, endocrine, pulmonary, digestive or urinary system), number of organ systems affected by late effects (range =0–9), psychological distress (Tscore ≥63 on at least two dimensions or the Global Severity Index of the Brief Symptom Inventory [BSI]-18) [20, 28, 29] pain (none– mild, moderate–very severe), body mass index (BMI), time spent on vigorous-intensity physical activity (in half hours per week), time spent on moderate-intensity physical activity (in half hours per week), year of study participation (year), age at study (years), language region (German, French, Italian), migration background (yes, no), civil status (single, married, widowed/divorced), highest educational achievement (compulsory school, vocational training, upper secondary education, university education), and employment situation (employed, unemployed, in education). We generated time since diagnosis (years) by subtracting age at diagnosis (years) from age at baseline (years), and time to followup (years) by subtracting age at baseline (years) from age at follow-up (years). A more detailed description of the coding of the explanatory variables can be found in Appendix 1. For the general population, the following data were assessed through self-report: sex, age at study, language region, migration background, civil status, educational achievement and employment situation. 2.3 Data Analysis For all analyses, we used Stata Statistical Software: Release 18. We excluded participants with incomplete information in the SF-36 vitality scale (survivors: n=117 [6%] at baseline, n=7[1%]at follow-up; general population: n=7 [1%]) and analysed available cases (pairwise deletion). Aim A: We used descriptive statistics to describe the fatigue prevalence and fatigue severity in survivors and the general population. We used logistic regression to test for differences in fatigue prevalence between survivors and the general population, and linear regression to test for differences in fatigue severity between survivors and the general population, adjusting for sex, age and socio-demographic characteristics (language region, migration background, civil status, educational achievement and employment situation). Aim B: We only included survivors who participated in both the baseline and follow-up questionnaires. We calculated the proportion of survivors with specific longitudinal fatigue patterns: no/low fatigue (no fatigue at baseline and follow-up), late onset fatigue (no fatigue at baseline, fatigue at follow-up), improving fatigue (fatigue at baseline, no fatigue at follow-up), and persistent fatigue (fatigue at baseline and follow-up). We used https:// sankeymatic.com/ to build Figure 1. Aim C: In survivors participating in both the baseline and followup questionnaire, we used univariable and multivariable logistic regression to identify characteristics associated with persistent fatigue (reference group: no/low fatigue). We included independent variables (measured at baseline) that were associated with fatigue in previous studies [1, 3, 15, 30–33]: cancer diagnosis, surgery, chemotherapy, radiotherapy, stem cell therapy, relapse/SMN, time since diagnosis, health status, organ-specific late effects, number of late effects, psychological distress, pain, BMI, vigorous-intensity physical activity, moderate-intensity physical activity, age at study, sex, civil status and employment situation. We a priori included age at diagnosis, year of diagnosis, educational achievement, and year of study participation as independent variables. In the multivariable model, we ran a full model, only excluding covariables that were correlated with others, or categorical covariables that had a cell size of n<25. 3Results We included 1846 survivors with baseline information (52% male, 48% female). Of the survivors invited to participate in Pediatric Blood & Cancer,2025 3of12 FIGURE 1 Longitudinal patterns of fatigue in survivors of childhood and adolescent cancer participating in both the baseline and follow-up survey (N=684). Note: The numbers in the boxes indicate the number and proportion of survivors with fatigue at baseline and follow-up, and the number and proportion of survivors with the distinct longitudinal fatigue patterns: no/low fatigue (no fatigue at baseline and follow-up), late-onset fatigue(nofatigue at baseline, fatigue at follow-up), improving fatigue (fatigue at baseline, no fatigue at follow-up), and persistent fatigue (fatigue at baseline and follow-up). the follow-up survey, 684 survivors (52% male, 48% female) responded to the follow-up questionnaire. From the general population, we included 863 participants (42% male, 58% female). At baseline, survivors were on average 26.8 years old (interquartile range [IQR] 9.9 years) and 18.0 years from diagnosis (IQR 12.5 years). At follow-up, survivors were on average 36.7 years old (IQR 9.5), 8.9 years from the baseline survey (IQR 4.1), and 28.8 years from diagnosis (IQR 10.4). Participants of the general population were on average 43.7 years old (IQR 16.9) (Table 1). Fatigue prevalence was similar in survivors at baseline (26%) and follow-up (29%), and the general population (26%; Table 1). After adjusting for sex, age and socio-demographic characteristics, we did not find any differences in fatigue prevalence between the general population and survivors at baseline (odds ratio [OR] = 0.9, 95% CI: 0.7–1.2), or follow-up (OR =1.0, 95% CI: 0.7–1.3; Tables S1 and S2). Fatigue severity was similar in survivors at baseline (mean 50.2, SD 10.8) and follow-up (mean 49.0, SD 10.9), and the general population (mean 49.3, SD 10.2; Table 1). After adjusting for sex, age and socio-demographic characteristics, we found that the score for fatigue severity was slightly higher in survivors at baseline than in the general population (b=1.2, 95% CI: 0.1– 2.2; indicating less fatigue symptoms in survivors), and we found no differences between survivors at follow-up and the general population (b=0.2, 95% CI: −1.1to1.5;TablesS3 and S4). Fatigue prevalence and fatigue severity stratified by participants’ characteristics are presented in Tables S5 and S6. On an individual level, 64% (n=437) of survivors had consistently low or no fatigue,14%(n=93) had late-onset fatigue,7%(n=49) had improving fatigue,and15%(n=105) experienced persistent fatigue (Figure 1,TableS7). Experiencing late effects in a greater number of organ systems (OR =1.5, 95% CI: 1.2–1.9), psychological distress (OR =14.5, 95% CI: 5.9–35.6), and pain (OR =6.1, 95% CI: 2.1–17.9) was associated with a higher likelihood of experiencing persistent fatigue.In contrast, more time spent on moderate-intensity physical activity (half hours per week; OR =0.95, 95% CI: 0.9–1.0) was associated with a lower likelihood of experiencing persistent fatigue (Table 2). The results from the univariable analysis are presented in Table S8. 4 Discussion Around one in four survivors (26%) experienced fatigue, which was comparable to the general population. Most survivors (64%) experienced low or no fatigue,14%hadlate-onset fatigue,7% had improving fatigue, and 15% experienced persistent fatigue. More late effects, psychological distress, pain and spending less time on moderate-intensity physical activity were associated with persistent fatigue. Fatigue is a significant problem in long-term survivorship, with 26% of survivors experiencing clinically significant fatigue symptoms at baseline (on average 18 years from diagnosis) and 29% at follow-up (on average 29 years after diagnosis). Fatigue in this study is similarly prevalent as reported by systematic reviews in survivors of childhood and adolescent cancers [1, 12], and slightly more prevalent than what a previous study in a smaller cohort of survivors in Switzerland found [3]. Of note, being fatigued may be a barrier to survey participation [33], which may result in an underestimation of fatigue prevalence both in survivors and the general population. Our results show that in a population-based cohort of survivors, fatigue in long-term survivorship (median 18 years after diagnosis) is comparable to fatigue in the general population. This is in contrast to our previous systematic review, where we found that survivors had higher fatigue prevalence and fatigue levels than comparisons [1]. One possible explanation is that we compared survivors to a representative group of the general population rather than to healthy controls. Fatigue is common in the general population [34], where some individuals may have chronic health conditions [35], and previous studies have shown 4of12 Pediatric Blood & Cancer,2025 TABLE 1 Characteristics of the study population. Survivors General population Baseline Follow-Upa Total N(%) Total N(%) Total N(%) 1846 (100.0) 684 (100.0) 863 (100.0) Sex Male 961 (52.1) 358 (52.3) 360 (41.7) Female 885 (47.9) 326 (47.7) 503 (58.3) Age at study (years) Median (IQR; range) 26.8 (9.90; 20–59) 36.7 (9.54; 25–58) 43.7 (16.88; 20–59) 20–29 years 1204 (65.2) 91 (13.3) 139 (16.1) 30–39 years 472 (25.6) 367 (53.6) 212 (24.6) 40–49 years 148 (8.0) 198 (29.0) 256 (29.7) 50–59 years 22 (1.2) 28 (4.1) 256 (29.7) Time from baseline (years) Median (IQR; range) n.a. 8.9 (4.12; 3–10) n.a. Time since diagnosis (years) Median (IQR; range) 18.0 (12.48; 5–42) 28.8 (10.44; 9–42) n.a. Language region German 1279 (69.3) 479 (70.0) 633 (73.3) French 497 (26.9) 188 (27.5) 181 (21.0) Italian 56 (3.0) 17 (2.5) 49 (5.7) Missing 14 (0.8) 0 (0.0) 0 (0.0) Migration background No 1434 (77.7) 568 (83.1) 657 (76.1) Yes 319 (17.3) 87 (12.7) 206 (23.9) Missing 93 (5.0) 29 (4.2) 0 (0.0) Civil status Single 1393 (75.5) 504 (73.7) 309 (35.8) Married 381 (20.6) 152 (22.2) 406 (47.0) Divorced/Widowed 49 (2.7) 22 (3.2) 108 (12.5) Missing 23 (1.2) 6 (0.9) 40 (4.6) Educational achievement Compulsory school 217 (11.8) 67 (9.8) 49 (5.7) Vocational training 1051 (56.9) 414 (60.5) 376 (43.6) Upper secondary education 353 (19.1) 134 (19.6) 227 (26.3) University education 189 (10.2) 61 (8.9) 167 (19.4) Missing 36 (2.0) 8 (1.2) 44 (5.1) Employment status Employed 1411 (76.4) 538 (78.7) 732 (84.8) Unemployed 142 (7.7) 49 (7.2) 82 (9.5) In education 233 (12.6) 77 (11.3) 28 (3.2) Missing 60 (3.3) 20 (2.9) 21 (2.4) Fatigue prevalence No 1365 (73.9) 486 (71.1) 638 (73.9) Yes 481 (26.1) 198 (28.9) 225 (26.1) (Continues) Pediatric Blood & Cancer,2025 5of12 TABLE 1 (Continued) Survivors General population Baseline Follow-Upa Total N(%) Total N(%) Total N(%) 1846 (100.0) 684 (100.0) 863 (100.0) Fatigue severity Mean (SD; range) 50.2 (10.81; 13–71) 49.0 (10.9; 13–71) 49.3 (10.17; 13–71) BMI Mean (SD; range) 23.7 (4.7; 13–68) n.a. n.a. Missing 36 (2.0) n.a. n.a. Vigorous-intensity physical activity (30 min/week) Median (IQR; range) 4c(7; 0–210) n.a. n.a. Missing 328 (17.8) n.a. n.a. Moderate-intensity physical activity (30 min/week) Median (IQR; range) 4c(6; 0–224) n.a. n.a. Missing 315 (17.1) n.a. n.a. Health status Good–Excellent 1754 (95.0) n.a. n.a. Fair–Poor 92 (5.0) n.a. n.a. Pain None–Mild 1669 (90.4) n.a. n.a. Moderate–Very severe 175 (9.5) n.a. n.a. Missing 2 (0.1) n.a. n.a. Late effects Number of organ systems affected:bMedian (IQR) 1 (2; 0–9) n.a. n.a. No 546 (29.6) n.a. n.a. Any 1295 (70.2) n.a. n.a. Neurological 617 (33.4) n.a. n.a. Musculoskeletal 407 (22.0) n.a. n.a. Cardiovascular 176 (9.5) n.a. n.a. Vision 246 (13.3) n.a. n.a. Hearing 207 (11.2) n.a. n.a. Endocrine 248 (13.4) n.a. n.a. Pulmonary 519 (28.1) n.a. n.a. Digestive 266 (14.4) n.a. n.a. Urinary 64 (3.5) n.a. n.a. Missing 5 (0.3) n.a. n.a. Psychological distressd No 1566 (84.8) n.a. n.a. Yes 248 (13.4) n.a. n.a. Missing 32 (1.7) n.a. n.a. Diagnosis Leukaemia 493 (26.7) 189 (27.6) n.a. Lymphoma 459 (24.9) 167 (24.4) n.a. CNS tumour 276 (14.9) 73 (10.7) n.a. Other tumour 618 (33.5) 255 (27.3) n.a. (Continues) 6of12 Pediatric Blood & Cancer,2025 TABLE 1 (Continued) Survivors General population Baseline Follow-Upa Total N(%) Total N(%) Total N(%) 1846 (100.0) 684 (100.0) 863 (100.0) Surgery No 529 (28.7) 200 (29.2) n.a. Yes 1229 (66.6) 448 (65.5) n.a. Missing 88 (4.8) 36 (5.3) n.a. Chemotherapy No 376 (20.4) 117 (17.1) n.a. Yes 1372 (74.3) 526 (76.9) n.a. Missing 98 (5.3) 41 (6.0) n.a. Radiotherapy No 1092 (59.2) 386 (56.4) n.a. Yes 641 (34.7) 254 (37.1) n.a. Missing 113 (6.1) 44 (6.4) n.a. Stem cell transplantation No 1652 (89.5) 613 (89.6) n.a. Yes 59 (3.2) 15 (2.2) n.a. Missing 135 (7.3) 56 (8.2) n.a. Relapse/SMN No 1492 (80.8) 557 (81.4) n.a. Yes 354 (19.2) 127 (18.6) n.a. Age at diagnosis (years) Median (IQR; range) 12 (9; 0–20) 10 (9; 0–20) n.a. Year of diagnosis 1976-1985 453 (24.5) 228 (33.3) n.a. 1986–1995 687 (37.2) 299 (43.7) n.a. 1996–2005 413 (22.4) 149 (21.8) n.a. 2006–2015 293 (15.9) 8 (1.2) n.a. Note: ‘Missing’ is only displayed if there are missing values for the respective variable. Standardised fatigue severity scores range from 0 to 100, with lower scores indicating greater fatigue. Fatigue prevalence and fatigue severity stratified by participants’ characteristics are presented in Tables S5 and S6. Abbreviations: BSI-18, Brief Symptom Inventory-18; CNS, central nervous system; n.a., not available; SD, standard deviation; SMN, second malignant neoplasm. aAll variables (except age at study, time from baseline, time since diagnosis and fatigue) were measured at baseline. bNumber of organ systems affected by late effects (in neurological, musculoskeletal, cardiovascular, vision, hearing, endocrine, pulmonary, digestive or urinary). cThe unit is ’30 min/week’. A score of 4 means: 4 ×30 min of physical activity per week =2 h of physical activity per week dT-score ≥63 on at least two dimensions or the Global Severity Index of the Brief Symptom Inventory (BSI)-18. that fatigue is prevalent not only among cancer survivors, but also in persons with other chronic diseases [36]. Another reason may be a response shift, which has been shown to occur in cancer patients [37]. It is possible that survivors subconsciously compare their current fatigue to previously experienced more severe fatigue, for example, during treatment, and thus underreport symptoms of fatigue. Regarding longitudinal patterns of fatigue, our results are in line with studies in patients closer to diagnosis and treatment, showing that fatigue at baseline predicts fatigue at follow-up [11, 16, 33]. In our study, the majority of survivors either experienced no/low fatigue or persistent fatigue, and only a minority changed their fatigue status. Similarly, a study from the United States found that fatigue in Hodgkin lymphoma survivors did not change significantly from the end of therapy to 36 months posttherapy [15]. Another study found a larger proportion of survivors with persistent fatigue (31%) or improving fatigue (21%), and a smaller proportion of survivors with no/low fatigue (39%) [16]. These differences may be due to the assessment used (Fatigue Questionnaire vs. SF-36 vitality scale in our study) and different lengths of follow-up periods (median 2.7 vs. 8.9 years in our Pediatric Blood & Cancer,2025 7of12 TABLE 2 Characteristics associated with persistent fatigue in survivors of childhood and adolescent cancer (from multivariable logistic regression). Persistent fatigue (Ref. no/low fatigue) nORa95% CI p-value Total 433 Diagnosis Leukaemia Ref. Lymphoma 0.11 0.01 1.18 0.068 CNS tumour 0.51 0.04 6.25 0.602 Other tumour 0.12 0.01 1.27 0.078 Treatment Surgery (Ref. No) 4.44 0.44 44.61 0.205 Chemotherapy (Ref. No) 1.36 0.47 3.95 0.577 Radiotherapy (Ref. No) 1.28 0.63 2.60 0.487 SCT (Ref. No) n.a.b Relapse/SMN: Yes (Ref. No) 0.42 0.17 1.04 0.060 Age at diagnosis [continuous] 1.01 0.88 1.15 0.933 Year of diagnosis 1976–1985 Ref. 1986–1995 1.02 0.30 3.44 0.976 1996–2015 1.58 0.17 14.55 0.687 Health status: Fair–Poor (Ref. Good–Excellent) n.a.d Late effects: Number of organ systems affectedc1.49 1.19 1.87 0.001 Psychological distress:dYes (Ref. No) 14.53 5.94 35.57 <0.001 Pain: Moderate–Very severe (Ref. None–Mild) 6.05 2.05 17.87 0.001 BMI [continuous] 1.04 0.96 1.11 0.349 Moderate-intensity physical activity (30 min/week) 0.95 0.90 1.00 0.038 Year of study participation (year) 1.08 0.90 1.31 0.416 Age at study: (years) 1.06 0.94 1.19 0.365 Sex: Female (Ref. Male) 1.06 0.52 2.13 0.876 Civil status Single/Divorced/Widowed Ref. Married 0.59 0.25 1.41 0.237 Educational achievement Compulsory school 2.04 0.64 6.53 0.231 Vocational training Ref. Upper secondary education 1.39 0.61 3.14 0.432 University education 1.17 0.38 3.60 0.786 Employment situation Employed Ref. Unemployed 0.62 0.15 2.59 0.509 In education 1.83 0.68 4.93 0.229 Note: Models adjusted for all variables displayed in the table. Explanatory variables were measured at baseline. Bold font indicates results with p-values less than 0.05. Time since diagnosis, vigorous-intensity physical activity, and the organ-specific late effects variables were excluded from this analysis due to collinearity with other variables. Abbreviations: BMI, body mass index; CI, confidence interval, CNS, central nervous system; OR, odds ratio; SMN, second malignant neoplasm. aHigher odds ratio indicates a higher likelihood to be in the ‘persistent fatigue’ group as compared to the ‘no/low fatigue’ group. bToo few observations (n<25 in one category), variable excluded from model. cNumber of organ systems affected by late effects (range: 0–9; neurological, musculoskeletal, cardiovascular, vision, hearing, endocrine, pulmonary, digestive or urinary). dT-score ≥63 on at least two dimensions or the Global Severity Index of the Brief Symptom Inventory (BSI)-18. 8of12 Pediatric Blood & Cancer,2025 study). LeBlanc and colleagues in the United States also used the SF-36 vitality scale to measure fatigue in adult survivors of non-Hodgkin lymphoma and found very similar results to ours, although survivors were significantly older (mean age 62 years) [33]. It is unclear why fatigue persists in survivors of childhood cancers. First, it may be difficult to treat fatigue symptoms. However, there is enough evidence of interventions that improve fatigue symptoms [38]. Thus, treatment resistance should not be a major concern. Other theories point to methodological problems, as the SF-36 is assessed retrospectively for 4 weeks, and the follow-up time in our study was, on average, 9 years. Specifically, fatigue status may have changed several times over the unobserved period. Finally, it is possible that survivors did not receive adequate treatment, as evidence suggests that there are several communication barriers between healthcare practitioners and survivors for discussing fatigue in appointments [4, 39]. With current practice, fatigue symptoms improve only for a small proportion of fatigued survivors. It is therefore recommended to strengthen current practice by implementing a more consistent screening and treatment approach: healthcare professionals responsible for the follow-up care of survivors of childhood cancer are encouraged to utilise published clinical practice guidelines for the surveillance and treatment of fatigue [1, 4, 38]. Simple, low-threshold interventions such as providing survivors with an information booklet can help to raise awareness and encourage survivors to take action or seek help [40]. In Switzerland, such information booklets are provided by the Swiss Cancer League and are available for free online (www.krebsliga.ch). In addition, capacity building in paediatric oncology rehabilitation with healthcare professionals who can provide effective interventions for fatigue (occupational therapists, physiotherapists, exercise therapists and psychotherapists), as well as strengthening the collaboration between paediatric oncologists and paediatric oncology rehabilitation professionals, could help to provide adequate care for survivors. Persistent fatigue was associated with clinical characteristics: more organ systems affected by late effects, psychological distress, pain, and less time spent on moderate-intensity physical activity. In line with this, a study from Belgium found that psychological symptoms predicted fatigue 1 year later [41], and a study in adult non-Hodgkin lymphoma survivors found post-traumatic stress symptoms and a higher number of comorbidities to be associated with persistent fatigue [33]. We found that more time spent in moderate-intensity physical activity was associated with a lower likelihood of persistent fatigue, consistent with the Physical Activity in Childhood Cancer Survivors (PACCS) study, which reported that light physical activity was associated with improvement in fatigue in survivors [42] Examples of moderate-intensity or light physical activity are walking briskly, cycling with low effort, or water aerobics. However, the relationship between fatigue and physical activity is likely complex: A study from the Netherlands showed that fatigue is a causal factor for decreased physical activity [43]. On the other hand, evidence from intervention studies shows that physical activity interventions are effective in reducing fatigue symptoms [38]. Thus, the relationship is likely bidirectional. Overall, our study confirms that it is primarily factors that are at least partly modifiable, such as health problems, psychological distress, pain and physical inactivity, that contribute to fatigue in long-term survivorship. The few studies available on childhood cancer survivors suggest that targeting modifiable factors, for example, physical activity [44] or psychological distress [45], may reduce fatigue symptoms. These factors, and additional contributors like poor sleep [3], can be targeted with interventions [1, 4, 38]. Our study adds evidence that these factors are associated with longitudinal fatigue patterns in survivorship. The major strength of our study is its nationwide and populationbased design. Other strengths are the good response rate of the SCCSS, which is comparable to other survivor cohorts [46–48], and the availability of general population data. A limitation of our study is the use of a generic fatigue measure (SF-36) instead of a cancer-specific and/or multidimensional measure. However, generic measures may be used to allow for comparison with other groups [49]. Also, the SF-36 assesses symptoms over a 4week period. It is possible that the fatigue status of survivors changed over the unobserved period. It is not predictable in which direction this may bias our results. In addition, the variability in the timing of follow-up assessments may have influenced placement of participants into fatigue patterns, and future studies with uniform assessment intervals may clarify these patterns. Attrition bias may be an issue, as not all baseline participants participated in follow-up. Another limitation is that survivors are prone to response bias when completing quality-of-life surveys [50]. This may cause an underestimation of the prevalence and severity of fatigue. Finally, the response rate in the general population was relatively low, which may have further biased our results. In conclusion, our study provides data on longitudinal patterns of fatigue in survivors and identifies factors associated with the different patterns. Although fatigue prevalence and severity of survivors were comparable to the general population, fatigue is a considerable problem for long-term survivors, with a quarter of survivors experiencing fatigue symptoms. Almost two-thirds of survivors (64%) had consistently low or no fatigue at both timepoints, but 15% of survivors were persistently fatigued. It is recommended to strengthen current clinical practice by implementing surveillance and treatment guidelines for fatigue. Managing late effects, pain and psychological distress, and promoting physical activity can help to improve fatigue symptoms and quality of life of survivors. Author Contributions Salome Christen: conceptualisation (equal), formal analysis (equal), funding acquisition (equal), methodology (equal), project administration (equal), software (equal), validation (equal), visualisation (lead), writing – original draft (lead), writing – review and editing (lead). Luzius Mader: conceptualisation (equal), methodology (equal), supervision (equal), data curation (equal), investigation (equal), supervision (equal), validation (equal), writing – review and editing (equal). André O. von Bueren: resources (equal), validation (equal), writing – review and editing (equal). Eva Maria Tinner: resources (equal), validation (equal), writing – review and editing (equal). Grit Sommer: data curation (equal), investigation (equal), resources (equal), validation (equal), writing – review and editing (equal). Christina Schindera: resources (equal), validation (equal), writing – review and editing (equal). Claudia Kuehni: conceptualisation Pediatric Blood & Cancer,2025 9of12 Fatigue in survivors of childhood cancer. Supplemental Material. Christen et al. 2025 4 Supplemental Table S3. Fatigue severity (SF-36 vitality scale score) in survivors of childhood and adolescent cancer at baseline as compared to the Swiss general population (N=2490; from multivariable linear regression). Fatigue severity b 95%CI p - value Population General population Ref. Survivors at baseline 1.15 0.06 2.24 0.038 Sex Male Ref. Female - 2.84 -3.66 - 2.02 <0.001 Age 20-29 years Ref. 30-39 years -0.80 -1.95 0.35 0.171 40-49 years -0.35 -1.90 1.19 0.654 50-59 years -1.11 -2.97 0.75 0.242 Language region German Ref. French - 3.15 -4.10 - 2.19 <0.001 Italian -0.66 -2.76 1.44 0.539 Migration background No migration background Ref. Migration background - 1.97 -3.01 - 0.94 <0.001 Civil status Single Ref. Married 1.46 0.33 2.58 0.011 Divorced/Widowed 1.19 -0.77 3.16 0.233 Educational achievement Compulsory school - 2.32 -3.76 - 0.88 0.002 Vocational training Ref. Upper secondary education -0.22 -1.26 0.82 0.681 University education -0.30 -1.56 0.97 0.646 Employment situation Employed Ref. Unemployed - 3.45 -4.96 - 1.94 <0.001 In education -0.86 -2.29 0.57 0.238 Note: Results from multivariable linear regression with fatigue severity (SF-36 vitality scale score) as the dependent variable, adjusted for all variables displayed in the table (population, sex, age, language region, migration background, civil status, educational achievement, and employment situation). Standardized fatigue severity scores range from 0 to 100, with lower scores indicating greater fatigue. Bold font indicates results with p-values <0.05. Abbreviations: b=unstandardized regression coefficient, CI=confidence interval. Fatigue in survivors of childhood cancer. Supplemental Material. Christen et al. 2025 5 Supplemental Table S4. Fatigue severity (SF-36 vitality scale score) in survivors of childhood and adolescent cancer at follow-up as compared to the Swiss general population (N=1447; from multivariable linear regression). Fatigue severity b 95%CI p - value Population General population Ref. Survivors at follow-up 0.23 -1.05 1.50 0.728 Sex Male Ref. Female - 2.31 - 3.39 - 1.23 <0.001 Age 20-29 years Ref. 30-39 years 0.10 -1.64 1.84 0.910 40-49 years 0.46 -1.48 2.41 0.640 50-59 years -0.19 -2.34 1.96 0.861 Language region German Ref. French - 3.52 - 4.80 - 2.24 <0.001 Italian 0.53 -2.15 3.20 0.699 Migration background No migration background Ref. Migration background -0.65 -2.06 0.75 0.364 Civil status Single Ref. Married 1.38 0.01 2.74 0.049 Divorced/Widowed 1.02 -1.17 3.21 0.362 Educational achievement Compulsory school - 2.51 - 4.60 - 0.43 0.018 Vocational training Ref. Upper secondary education -0.46 -1.78 0.87 0.499 University education -0.19 -1.79 1.40 0.814 Employment situation Employed Ref. Unemployed - 3.87 - 5.82 - 1.92 <0.001 In education 0.41 -1.79 2.61 0.713 Note: Results from multivariable linear regression with fatigue severity (SF-36 vitality scale score) as the dependent variable, adjusted for all variables displayed in the table (population, sex, age, language region, migration background, civil status, educational achievement, and employment situation). Standardized fatigue severity scores range from 0 to 100, with lower scores indicating greater fatigue (fatigue severity). Bold font indicates results with p-values <0.05. Abbreviations: b=unstandardized regression coefficient, CI=confidence interval. Fatigue in survivors of childhood cancer. Supplemental Material. Christen et al. 2025 6 Supplemental Table S5. Fatigue prevalence (SF-36 vitality scale score≤45) stratified by the characteristics of the study population. Survivors General population Baseline Follow - up* Total With fatigue Total With fatigue Total With fatigue N (% a ) N (% b ) N (% a ) N (% b ) N (% a ) N (% b ) 1846 (100.0%) 684 (100.0%) 863 (100.0%) Sex Male 961 (52.1) 211 (22.0) 358 (52.3) 85 (23.7) 360 (41.7) 64 (17.8) Female 885 (47.9) 270 (30.5) 326 (47.7) 113 (34.7) 503 (58.3) 161 (32.0) Age at study [years] Median (IQR; range) 26.8 (9.9; 20-59) 26.7 (9.6; 20-55) 36.7 (9.5; 25-58) 37.7 (9.8; 26-58) 43.7 (16.8;20-60) 41.5 (17.2;20-60) 20 - 29 years 1204 (65.2) 315 (26.2) 91 (13.3) 26 (28.6) 139 (16.1) 41 (29.5) 30 - 39 years 472 (25.6) 113 (23.9) 367 (53.7) 104 (28.3) 212 (24.6) 60 (28.3) 40 - 49 years 148 (8.0) 45 (30.4) 198 (28.9) 57 (28.8) 256 (29.7) 67 (26.2) 50 - 59 years 22 (1.2) 8 (36.4) 28 (4.1) 11 (39.3) 256 (29.7) 57 (22.3) Time from baseline [years] Median (IQR; range) n.a. n.a. 8.9 (4.1; 3-10) 8.8 (4.8; 4-10) n.a. n.a. Time since diagnosis [years] Median (IQR; range) 18.0 (12.5; 5-42) 17.1 (12.6; 6-42) 28.8 (10.5; 9-42) 28.9 (10.1; 10-42) n.a. n.a. Language region German 1279 (69.3) 281 (22.0) 479 (70.0) 116 (24.2) 633 (73.3) 145 (22.9) French 497 (26.9) 172 (34.6) 188 (27.5) 78 (41.5) 181 (21.0) 74 (40.9) Italian 56 (3.0) 23 (41.1) 17 (2.5) 4 (23.5) 49 (5.7) 6 (12.2) Missing 14 (0.8) . (.) . (.) Migration background No 1434 (77.7) 344 (24.0) 568 (83.0) 151 (26.6) 657 (76.1) 164 (25.0) Yes 319 (17.3) 113 (35.4) 87 (12.7) 35 (40.2) 206 (23.9) 61 (29.6) Missing 93 (5.0) 29 (4.2) . (.) Civil status Single 1393 (75.5) 367 (26.3) 504 (73.7) 145 (28.8) 309 (35.8) 85 (27.5) Married 381 (20.6) 91 (23.9) 152 (22.2) 46 (30.3) 406 (47.0) 101 (24.9) Divorced/Widowed 49 (2.7) 13 (26.5) 22 (3.2) 5 (22.7) 108 (12.5) 30 (27.8) Missing 23 (1.2) 6 (0.9) 40 (4.6) Educational achievement Compulsory school 217 (11.8) 86 (39.6) 67 (9.8) 28 (41.8) 49 (5.7) 21 (42.9) Vocational training 1051 (56.9) 248 (23.6) 414 (60.5) 113 (27.3) 376 (43.6) 94 (25.0) Upper secondary education 353 (19.1) 84 (23.8) 134 (19.6) 37 (27.6) 227 (26.3) 59 (26.0) University education 189 (10.2) 56 (29.6) 61 (8.9) 17 (27.9) 167 (19.4) 41 (24.6) Missing 36 (2.0) 8 (1.2) 44 (5.1) Employment status Employed 1411 (76.4) 333 (23.6) 538 (78.7) 139 (25.8) 732 (84.8) 181 (24.7) Unemployed 142 (7.7) 58 (40.8) 49 (7.2) 28 (57.1) 82 (9.5) 33 (40.2) In education 233 (12.6) 70 (30.0) 77 (11.3) 24 (31.2) 28 (3.2) 6 (21.4) Missing 60 (3.3) 20 (2.9) 21 (2.4) Fatigue prevalence Yes 1365 (73.9) n.a. 486 (71.1) n.a. 638 (73.9) n.a. No 481 (26.1) n.a. 198 (28.9) n.a. 225 (26.1) n.a. Fatigue severity Mean (SD; range) 50.2 (10.80) 36.0 (6.60) 49.0 (10.90) 35.3 (6.60) 49.3 (10.20) 35.8 (6.50) Body Mass Index Mean (SD; range) 23.7 (4.70) 23.6 (4.3) n.a. n.a. n.a. n.a. Missing 36 (2) n.a. n.a. Vigorous - intensity physical activity [30 minutes/week] Median (IQR; range) 4.0 (7.0; 0-210) 3.0 (6.5; 0-155) n.a. n.a. n.a. n.a. Missing 328 (17.8) n.a. n.a. Moderate - intensity physical activity [30 minutes/week] Median (IQR; range) 4.0 (6.0; 0-224) 3.0 (6.0; 0-112) n.a. n.a. n.a. n.a. Missing 315 (17.1) n.a. n.a. Current health status Good - Excellent 1754 (95.0) 402 (22.9) n.a. n.a. n.a. n.a. Fair - Poor 92 (5.0) 79 (85.9) n.a. n.a. n.a. n.a. Fatigue in survivors of childhood cancer. Supplemental Material. Christen et al. 2025 7 Supplemental Table S5. (continued) Pain None - Mild 1669 (90.4) 361 (21.6) n.a. n.a. n.a. n.a. Moderate - Very severe 175 (9.5) 120 (68.6) n.a. n.a. n.a. n.a. Missing 2 (0.1) n.a. n.a. Late effects Nr. of organ systems affected:a Median (IQR; range) 1 (2; 0-9) 2 (2; 0-9) n.a. n.a. n.a. n.a. No 546 (29.6) 85 (15.6) n.a. n.a. n.a. n.a. Any 1295 (70.2) 394 (30.4) n.a. n.a. n.a. n.a. Neurological 617 (33.4) 240 (38.9) n.a. n.a. n.a. n.a. Musculoskeletal 407 (22.0) 147 (36.1) n.a. n.a. n.a. n.a. Cardiovascular 176 (9.5) 62 (35.2) n.a. n.a. n.a. n.a. Vision 246 (13.3) 84 (34.1) n.a. n.a. n.a. n.a. Hearing 207 (11.2) 85 (41.1) n.a. n.a. n.a. n.a. Endocrine 248 (13.4) 95 (38.3) n.a. n.a. n.a. n.a. Pulmonary 519 (28.1) 170 (32.8) n.a. n.a. n.a. n.a. Digestive 266 (14.4) 117 (44.0) n.a. n.a. n.a. n.a. Urinary 64 (3.5) 25 (39.1) n.a. n.a. n.a. n.a. Missing 5 (0.3) n.a. n.a. Psychological distress c No 1566 (84.8) 286 (18.3) n.a. n.a. n.a. n.a. Yes 248 (13.4) 184 (74.2) n.a. n.a. n.a. n.a. Missing 32 (1.7) n.a. n.a. Diagnosis Leukaemia 493 (26.7) 123 (24.9) 189 (27.6) 54 (28.6) n.a. n.a. Lymphoma 459 (24.9) 107 (23.3) 167 (24.4) 40 (24.0) n.a. n.a. CNS tumour 276 (15.0) 96 (34.8) 73 (10.7) 36 (49.3) n.a. n.a. Neuroblastoma 46 (2.5) 15 (32.6) 19 (2.8) 3 (15.8) n.a. n.a. Retinoblastoma 25 (1.4) 5 (20.0) 10 (1.5) 4 (40.0) n.a. n.a. Renal tumour 66 (3.6) 8 (12.1) 30 (4.4) 4 (13.3) n.a. n.a. Hepatic tumour 11 (0.6) 1 (9.1) 3 (0.4) . (.) n.a. n.a. Malignant bone tumour 112 (6.1) 30 (26.8) 46 (6.7) 12 (26.1) n.a. n.a. Soft tissue sarcoma 120 (6.5) 33 (27.5) 61 (8.9) 18 (29.5) n.a. n.a. Germ cell tumour 124 (6.7) 29 (23.4) 40 (5.8) 12 (30.0) n.a. n.a. Other tumour 66 (3.6) 22 (33.3) 29 (4.2) 10 (34.5) n.a. n.a. Langerhans cell histiocytosis 48 (2.6) 12 (25.0) 17 (2.5) 5 (29.4) n.a. n.a. Surgery No 529 (28.7) 137 (25.9) 200 (29.2) 54 (27.0) n.a. n.a. Yes 1229 (66.6) 311 (25.3) 448 (65.5) 129 (28.8) n.a. n.a. Missing 88 (4.8) 36 (5.3) n.a. Chemotherapy No 376 (20.4) 116 (30.9) 117 (17.1) 45 (38.5) n.a. n.a. Yes 1372 (74.3) 336 (24.5) 526 (76.9) 136 (25.9) n.a. n.a. Missing 98 (5.3) 41 (6.0) n.a. Radiotherapy No 1092 (59.2) 265 (24.3) 386 (56.4) 98 (25.4) n.a. n.a. Yes 641 (34.7) 180 (28.1) 254 (37.1) 83 (32.7) n.a. n.a. Missing 113 (6.1) 44 (6.4) n.a. Stem cell transplantation No 1652 (89.5) 421 (25.5) 613 (89.6) 171 (27.9) n.a. n.a. Yes 59 (3.2) 16 (27.1) 15 (2.2) 6 (40.0) n.a. n.a. Missing 135 (7.3) 56 (8.2) n.a. Relapse / SMN No 1492 (80.8) 390 (26.1) 557 (81.4) 161 (28.9) n.a. n.a. Yes 354 (19.2) 91 (25.7) 127 (18.6) 37 (29.1) n.a. n.a. Age at diagnosis [years] Median (IQR; range) 12.0 (9.0; 0-20) 13.0 (9.0; 0-20) 10.0 (9.0; 0-20) 11.5 (9.0; 0-20) n.a. n.a. Year of diagnosis 1976 - 1985 453 (24.5) 102 (22.5) 228 (33.3) 61 (26.8) n.a. n.a. 1986 - 1995 687 (37.2) 172 (25.0) 299 (43.7) 90 (30.1) n.a. n.a. 1996 - 2005 413 (22.4) 114 (27.6) 149 (21.8) 45 (30.2) n.a. n.a. 2006 - 2015 293 (15.9) 93 (31.7) 8 (1.2) 2 (25.0) n.a. n.a. Note: “Missing” is only displayed if there are missing values for the respective variable. Abbreviations: CNS=central nervous system, n.a.=not available, SD=standard deviation, SMN=second malignant neoplasm * All variables (except age at study, time from baseline, time since diagnosis, and fatigue) were measured at baseline. a total percentage b row percentage c T-score≥63 on at least two dimensions or the Global Severity Index of the Brief Symptom Inventory BSI-18 Fatigue in survivors of childhood cancer. Supplemental Material. Christen et al. 2025 8 Supplemental Table S6. Fatigue severity (SF-36 vitality scale score) stratified by the characteristics of the study population. Survivors General population Baseline: N=1846 Follow - up: N=684* N=863 Total Without fatigue With fatigue Total Without fatigue With fatigue Total Without fatigue With fatigue Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Sex Male 51.7 (10.4) 55.8 (6.9) 37.0 (6.3) 50.4 (10.5) 54.9 (6.7) 36.0 (6.7) 50.9 (9.5) 54.2 (6.2) 35.8 (7.5) Female 48.6 (11.1) 54.4 (6.8) 35.3 (6.7) 47.5 (11.2) 54.1 (6.3) 34.9 (6.6) 48.2 (10.5) 54.0 (6.2) 35.8 (6.1) Age at study [years] 20 - 29 years 50.3 (10.6) 55.2 (6.8) 36.5 (6.4) 49.5 (11.6) 55.1 (7.5) 35.5 (7.3) 48.1 (10.2) 53.2 (6.5) 36.0 (6.3) 30 - 39 years 50.3 (10.8) 54.9 (6.9) 35.5 (6.7) 49.4 (10.5) 54.7 (6.2) 36.0 (6.4) 49.0 (9.7) 53.8 (6.0) 36.8 (5.6) 40 - 49 years 49.7 (12.0) 56.0 (7.6) 35.2 (6.5) 48.6 (11.2) 54.4 (6.6) 34.4 (6.9) 50.1 (9.1) 54.4 (5.9) 37.9 (4.5) 50 - 59 years 46.2 (14.0) 55.2 (6.0) 30.4 (8.7) 44.7 (10.7) 52.1 (5.6) 33.4 (5.0) 49.5 (11.5) 54.5 (6.4) 32.2 (8.0) Time since diagnosis [years] 5 - 14 years 49.5 (10.8) 54.8 (6.6) 36.0 (6.8) 47.3 (12.7) 55.3 (8.0) 34.7 (7.5) n.a. n.a. n.a. 15 - 24 years 50.8 (10.5) 55.3 (7.0) 36.6 (5.9) 49.4 (10.3) 54.3 (6.4) 36.2 (6.4) n.a. n.a. n.a. 25 - 34 years 50.4 (11.5) 55.6 (7.4) 35.4 (7.1) 48.7 (10.6) 54.3 (6.1) 35.4 (6.5) n.a. n.a. n.a. 35 - 44 years 47.9 (12.3) 54.6 (5.8) 31.3 (6.7) 49.9 (11.9) 55.5 (7.2) 34.0 (7.1) n.a. n.a. n.a. Time from baseline [years] 0 - 4 years n.a. n.a. n.a. 47.3 (11.5) 54.0 (6.7) 34.3 (6.9) n.a. n.a. n.a. 5 - 9 years n.a. n.a. n.a. 49.5 (10.8) 54.9 (6.5) 35.5 (6.6) n.a. n.a. n.a. 10+ years n.a. n.a. n.a. 49.4 (7.9) 52.7 (6.0) 39.4 (3.0) n.a. n.a. n.a. Language region German 51.3 (10.3) 55.4 (6.9) 36.7 (6.5) 50.1 (10.3) 54.7 (6.3) 35.9 (6.5) 50.0 (9.8) 54.2 (6.0) 35.9 (6.3) French 47.8 (11.4) 54.5 (6.9) 35.1 (6.5) 46.2 (11.9) 54.3 (7.1) 34.8 (6.7) 46.0 (10.7) 52.9 (6.8) 36.0 (6.6) Italian 47.0 (12.3) 55.1 (7.7) 35.4 (7.1) 48.0 (12.2) 53.4 (6.9) 30.4 (8.1) 52.6 (9.8) 55.2 (6.0) 33.7 (11.7) Missing n=14 (0.8%) n=0 (0.0%) n=0 (0.0%) Migration background No 50.9 (10.4) 55.4 (6.9) 36.8 (6.1) 49.5 (10.6) 54.5 (6.5) 35.6 (6.8) 49.5 (10.1) 54.1 (6.1) 35.7 (6.8) Yes 47.2 (12.1) 54.5 (6.7) 33.8 (7.1) 46.8 (11.6) 54.9 (6.6) 34.9 (5.6) 48.6 (10.3) 53.9 (6.5) 36.1 (5.6) Missing n=93 (5.0%) n=29 (4.2%) n=0 (0.0%) Civil status Single 50.1 (10.6) 55.0 (6.8) 36.3 (6.5) 49.0 (10.8) 54.4 (6.4) 35.5 (6.7) 48.9 (10.2) 54.0 (6.0) 35.6 (6.5) Married 50.9 (11.1) 55.6 (7.3) 35.7 (6.5) 49.3 (10.9) 55.0 (6.8) 36.2 (5.7) 49.6 (9.8) 54.0 (6.2) 36.5 (6.1) Widowed/Divorced 50.5 (12.8) 56.8 (7.0) 33.0 (8.2) 48.2 (14.5) 55.1 (6.5) 24.7 (6.5) 48.9 (11.6) 54.8 (6.5) 33.8 (7.2) Missing n=23 (1.2%) n=6 (0.9%) n=40 (4.6%) Educational achievement Compulsory school 47.2 (12.6) 55.6 (7.1) 34.3 (7.2) 45.8 (12.9) 54.6 (7.6) 33.7 (7.7) 46.1 (13.9) 56.3 (7.7) 32.7 (7.1) Vocational training 50.7 (10.5) 55.1 (6.9) 36.2 (6.5) 49.7 (10.7) 54.9 (6.6) 35.7 (6.2) 49.2 (10.2) 53.9 (6.0) 35.2 (6.7) Upper secondary education 51.1 (10.4) 55.6 (6.9) 37.0 (6.2) 49.2 (10.0) 54.3 (5.3) 35.8 (6.2) 49.2 (9.5) 53.6 (5.7) 36.6 (6.4) University education 49.2 (10.4) 54.6 (6.6) 36.5 (6.0) 47.3 (11.4) 52.5 (7.3) 33.6 (8.3) 50.2 (9.6) 54.4 (6.4) 37.2 (5.1) Missing n=36 (2.0%) n=8 (1.2%) n=44 (5.1%) Employment status Employed 50.8 (10.4) 55.2 (6.9) 36.6 (6.3) 49.5 (10.6) 54.5 (6.3) 35.1 (6.7) 49.7 (9.7) 54.1 (6.2) 36.5 (5.6) Unemployed 46.1 (13.1) 55.0 (7.9) 33.2 (6.8) 42.9 (11.5) 53.9 (5.8) 34.7 (6.9) 45.8 (13.5) 54.8 (6.8) 32.4 (9.1) In education 49.3 (10.6) 54.8 (6.6) 36.6 (6.3) 49.5 (10.7) 55.0 (7.4) 37.3 (5.0) 49.2 (9.9) 53.1 (5.8) 34.7 (8.0) Missing n=60 (3.3%) n=20 (2.9%) n=21 (2.4%) Fatigue prevalence No 55.2 (6.9) n.a. n.a. 54.5 (6.5) n.a. n.a. 54.1 (6.2) n.a. n.a. Yes 36.0 (6.6) n.a. n.a. 35.3 (6.6) n.a. n.a. 35.8 (6.5) n.a. n.a. BMI <18 46.3 (10.5) 52.1 (5.4) 34.6 (8.0) n.a. n.a. n.a. n.a. n.a. n.a. 18 - 24 50.8 (10.5) 55.3 (7.0) 36.6 (6.1) n.a. n.a. n.a. n.a. n.a. n.a. 25 - 29 50.2 (11.0) 55.4 (6.7) 35.7 (6.9) n.a. n.a. n.a. n.a. n.a. n.a. >30 47.7 (11.8) 54.3 (7.2) 34.6 (7.2) n.a. n.a. n.a. n.a. n.a. n.a. Missing n=36 (2.0%) n.a. n.a. Time spent on vigorous - intensity physical activity [minutes/week] Below average (<120) 48.4 (11.1) 54.4 (6.7) 35.3 (6.7) n.a. n.a. n.a. n.a. n.a. n.a. Above average (120+) 51.9 (10.5) 55.8 (7.0) 36.3 (6.5) n.a. n.a. n.a. n.a. n.a. n.a. Missing n=328 (17.8%) n.a. n.a. Fatigue in survivors of childhood cancer. Supplemental Material. Christen et al. 2025 9 Supplemental Table S6. (continued) Time spent on moderate - intensity physical activity [minutes/week] Below average (<120) 49.3 (11.0) 54.9 (6.8) 35.9 (6.8) n.a. n.a. n.a. n.a. n.a. n.a. Above average (120+) 51.2 (10.5) 55.4 (7.0) 36.0 (6.2) n.a. n.a. n.a. n.a. n.a. n.a. Missing n=315 (17.1%) n.a. n.a. Current health status Good - Excellent 51.1 (10.1) 55.2 (6.9) 37.2 (5.6) n.a. n.a. n.a. n.a. n.a. n.a. Fair - Poor 32.9 (10.1) 49.5 (4.9) 30.1 (7.8) n.a. n.a. n.a. n.a. n.a. n.a. Pain None - Mild 51.5 (9.9) 55.4 (6.9) 37.4 (5.5) n.a. n.a. n.a. n.a. n.a. n.a. Moderate - Very severe 37.9 (11.2) 50.9 (5.2) 32.0 (7.7) n.a. n.a. n.a. n.a. n.a. n.a. Missing n=2 (0.1%) n.a. n.a. Late effects None 53.5 (10.1) 56.6 (7.2) 36.6 (6.3) n.a. n.a. n.a. n.a. n.a. n.a. At least one 48.8 (10.8) 54.4 (6.6) 35.9 (6.6) n.a. n.a. n.a. n.a. n.a. n.a. Neurological 46.6 (11.5) 54.0 (6.7) 35.0 (7.0) n.a. n.a. n.a. n.a. n.a. n.a. Musculoskeletal 47.1 (11.2) 54.0 (6.3) 35.0 (6.7) n.a. n.a. n.a. n.a. n.a. n.a. Cardiovascular 47.2 (12.1) 54.5 (6.6) 33.9 (7.7) n.a. n.a. n.a. n.a. n.a. n.a. Vision 47.5 (11.4) 54.1 (6.8) 34.8 (6.9) n.a. n.a. n.a. n.a. n.a. n.a. Hearing 45.7 (11.5) 53.5 (6.5) 34.4 (6.5) n.a. n.a. n.a. n.a. n.a. n.a. Endocrine 46.2 (11.2) 53.2 (6.6) 34.8 (6.6) n.a. n.a. n.a. n.a. n.a. n.a. Pulmonary 48.2 (10.7) 54.1 (6.6) 36.2 (6.4) n.a. n.a. n.a. n.a. n.a. n.a. Digestive 45.4 (11.8) 54.0 (6.2) 34.4 (7.2) n.a. n.a. n.a. n.a. n.a. n.a. Urinary 45.4 (10.3) 52.5 (4.4) 34.5 (6.7) n.a. n.a. n.a. n.a. n.a. n.a. Missing n=5 (0.3%) n.a. n.a. Psychological distress a No 52.2 (9.6) 55.4 (6.9) 37.7 (5.7) n.a. n.a. n.a. n.a. n.a. n.a. Yes 37.8 (9.6) 50.2 (4.9) 33.5 (6.8) n.a. n.a. n.a. n.a. n.a. n.a. Missing n=32 (1.7%) n.a. n.a. Diagnosis Leukaemia 50.9 (10.5) 55.6 (6.7) 36.7 (6.1) 49.7 (10.0) 54.7 (6.0) 37.0 (5.9) n.a. n.a. n.a. Lymphoma 51.0 (10.6) 55.5 (7.1) 36.4 (6.1) 50.3 (10.4) 54.9 (6.8) 36.0 (5.9) n.a. n.a. n.a. CNS tumour 48.2 (10.9) 54.5 (6.9) 36.4 (6.2) 43.9 (12.6) 54.3 (5.9) 33.1 (7.7) n.a. n.a. n.a. Neuroblastoma 48.6 (13.6) 55.8 (8.3) 33.6 (9.7) 52.9 (10.8) 56.3 (7.9) 34.9 (3.6) n.a. n.a. n.a. Retinoblastoma 52.2 (10.1) 56.0 (6.7) 36.6 (3.8) 45.5 (9.3) 51.3 (5.0) 36.7 (7.0) n.a. n.a. n.a. Renal tumour 53.9 (9.6) 56.3 (7.2) 36.3 (5.8) 51.1 (8.1) 53.4 (5.1) 35.8 (7.5) n.a. n.a. n.a. Hepatic tumour 46.8 (10.7) 49.9 (4.0) 16.7 b 56.8 (3.6) 56.8 (3.6) 0.0 (0.0) n.a. n.a. n.a. Malignant bone tumour 49.3 (11.0) 54.6 (6.3) 34.8 (7.3) 48.5 (11.6) 53.7 (7.3) 33.7 (8.0) n.a. n.a. n.a. Soft tissue sarcoma 49.8 (10.3) 54.6 (6.6) 37.3 (7.1) 48.7 (11.2) 54.2 (7.0) 35.5 (7.8) n.a. n.a. n.a. Germ cell tumour 49.6 (12.2) 54.9 (7.2) 32.3 (8.2) 48.4 (11.6) 54.2 (7.9) 34.9 (6.2) n.a. n.a. n.a. Other tumour 47.6 (9.8) 53.3 (5.9) 36.1 (4.7) 47.5 (12.8) 55.6 (5.8) 32.0 (5.9) n.a. n.a. n.a. Langerhans cell histiocytosis 50.1 (10.7) 54.8 (7.0) 35.8 (6.2) 49.1 (9.3) 53.7 (6.4) 37.8 (3.0) n.a. n.a. n.a. Surgery No 50.7 (10.6) 55.6 (6.9) 36.7 (6.2) 49.9 (10.2) 54.9 (6.1) 36.6 (6.1) n.a. n.a. n.a. Yes 50.2 (10.8) 55.1 (7.0) 35.9 (6.6) 48.7 (11.0) 54.3 (6.6) 35.0 (6.7) n.a. n.a. n.a. Missing n=88 (4.8%) n=36 (5.3%) n.a. Chemotherapy No 48.8 (11.0) 54.5 (7.1) 36.0 (6.7) 46.2 (11.8) 53.9 (6.2) 33.9 (7.2) n.a. n.a. n.a. Yes 50.7 (10.7) 55.4 (6.9) 36.2 (6.4) 49.8 (10.4) 54.6 (6.5) 36.0 (6.3) n.a. n.a. n.a. Missing n=98 (5.3%) n=41 (6.0%) n.a. Radiotherapy No 50.7 (10.7) 55.3 (6.9) 36.1 (6.8) 49.7 (10.5) 54.6 (6.4) 35.5 (6.4) n.a. n.a. n.a. Yes 49.7 (10.8) 55.0 (7.1) 36.2 (6.0) 48.2 (11.1) 54.3 (6.6) 35.5 (6.9) n.a. n.a. n.a. Missing n=113 (6.1%) n=44 (6.4%) n.a. Stem cell transplantation No 50.3 (10.8) 55.2 (6.9) 36.0 (6.6) 49.2 (10.8) 54.5 (6.4) 35.4 (6.7) n.a. n.a. n.a. Yes 50.4 (10.6) 55.3 (7.5) 37.2 (4.8) 45.8 (9.9) 52.3 (6.3) 36.1 (5.0) n.a. n.a. n.a. Missing n=135 (7.3%) n=56 (8.2%) n.a. Relapse/SMN No 50.3 (10.8) 55.3 (6.8) 36.1 (6.7) 49.1 (10.9) 54.6 (6.5) 35.4 (6.5) n.a. n.a. n.a. Yes 49.8 (10.8) 54.7 (7.3) 35.8 (6.2) 48.5 (11.0) 54.1 (6.6) 34.9 (7.2) n.a. n.a. n.a. Age at diagnosis [years] 0 - 5 years 51.2 (10.3) 55.3 (6.9) 36.7 (6.4) 50.5 (9.8) 54.3 (6.2) 36.2 (6.9) n.a. n.a. n.a. 6 - 10 years 50.3 (11.2) 55.5 (7.4) 35.8 (6.1) 48.9 (11.2) 55.1 (6.4) 35.4 (6.8) n.a. n.a. n.a. 11 - 15 years 50.1 (11.0) 55.2 (7.0) 36.0 (6.8) 48.7 (11.3) 55.0 (6.9) 35.6 (6.2) n.a. n.a. n.a. 16 - 20 years 49.2 (10.7) 54.6 (6.4) 35.7 (6.8) 47.1 (11.3) 53.4 (6.4) 33.8 (7.0) n.a. n.a. n.a. Fatigue in survivors of childhood cancer. Supplemental Material. Christen et al. 2025 10 Supplemental Table S6. (continued) Year of diagnosis 1976 - 1985 51.4 (11.3) 56.0 (7.5) 35.5 (6.6) 49.5 (11.5) 55.1 (6.7) 34.1 (7.1) n.a. n.a. n.a. 1986 - 1995 50.3 (10.4) 55.0 (6.8) 36.4 (6.1) 48.4 (9.9) 53.6 (5.8) 36.3 (6.1) n.a. n.a. n.a. 1996 - 2005 49.6 (10.9) 54.9 (6.7) 35.7 (7.0) 49.6 (11.7) 55.7 (7.2) 35.5 (6.8) n.a. n.a. n.a. 2006 - 2015 48.8 (10.7) 54.6 (6.3) 36.4 (6.8) 45.8 (13.3) 51.9 (7.2) 27.6 (10.3) n.a. n.a. n.a. Note: Standardized scores range from 0 to 100, with lower scores indicating greater fatigue (fatigue severity). Abbreviations: CNS=central nervous system, BSI-18= Brief Symptom Inventory-18, n.a.=not available, SMN=second malignant neoplasm * All variables (except age at study, time from baseline, time since diagnosis, and fatigue) were measured at baseline. a T-score≥63 on at least two dimensions or the Global Severity Index of the Brief Symptom Inventory BSI-18 b Only one observation in this category, calculation of SD therefore omitted Fatigue in survivors of childhood cancer. Supplemental Material. Christen et al. 2025 11 Supplemental Table S7. Frequency of specific longitudinal patterns of fatigue in survivors of childhood and adolescent cancer, and corresponding fatigue severity (SF-36 vitality scale score), and fatigue prevalence and severity in the general population. Survivors General population Baseline Follow - up n (%) Fatigue severity Fatigue severity Fatigue prevalence Fatigue severity Mean (SD) Mean (SD) n (%) a Mean (SD) a No/low fatigue 437 (63.9) 56.0 (6.8) 55.0 (6.5) 638 (73.9) 54.1 (6.2) Late onset fatigue 93 (13.6) 52.4 (6.3) 37.0 (6.1) n.a. n.a. Improving fatigue 49 (7.2) 38.6 (4.6) 50.8 (5.0) n.a. n.a. Persistent fatigue 105 (15.3) 35.6 (6.3) 33.8 (6.8) 225 (26.1) 35.8 (6.5) a Fatigue was only assessed at one timepoint for participants from the general population. Those who experienced fatigue are displayed in the persistent fatigue category, and those without in the no/low fatigue category. Fatigue in survivors of childhood cancer. Supplemental Material. Christen et al. 2025 12 Supplemental Table S8. Baseline characteristics associated with persistent fatigue (from univariable logistic regression). Persistent fatigue (vs. no/low fatigue) n= OR 95% CI p-value Diagnosis 539 Leukaemia Ref. Lymphoma 0.87 0.47 1.64 0.670 CNS tumour 3.19 1.59 6.39 0.001 Other tumour 0.97 0.55 1.70 0.919 Treatment Surgery (Ref. No) 511 1.19 0.72 1.96 0.504 Chemotherapy (Ref. No) 506 0.48 0.29 0.81 0.006 Radiotherapy (Ref. No) 502 1.41 0.89 2.23 0.141 Stem cell transplantation (Ref. No) 492 2.70 0.77 9.43 0.120 Relapse/SMN: Yes (Ref. No) 542 1.05 0.62 1.79 0.856 Age at diagnosis [continous] 542 1.06 1.02 1.10 0.003 Year of diagnosis 537 1976-1985 Ref. 1986-1995 1.11 0.68 1.84 0.672 1996-2005 1.37 0.78 2.42 0.275 2006-2015 n.a. c Time since diagnosis: [years] 542 1.01 0.98 1.05 0.363 Health status: Fair-Poor (Ref. Good-Excellent) 542 54.38 12.51 236.27 <0.001 Late effects: Number a 541 1.73 1.49 2.02 <0.001 Late effects, any (Ref. No) 541 3.26 1.85 5.74 <0.001 Neurological (Ref. No) 538 3.91 2.51 6.11 <0.001 Musculoskeletal (Ref. No) 540 2.00 1.25 3.21 0.004 Cardiovascular (Ref. No) 535 1.97 1.01 3.84 0.048 Vision (Ref. No) 537 3.09 1.74 5.48 <0.001 Hearing (Ref. No) 533 1.93 1.03 3.62 0.042 Endocrine (Ref. No) 537 3.02 1.77 5.17 <0.001 Pulmonary (Ref. No) 536 1.31 0.82 2.09 0.253 Digestive (Ref. No) 535 4.42 2.60 7.50 <0.001 Urinary (Ref. No) 536 3.32 1.36 8.09 0.008 Psychological distress: b Yes (Ref. No) 538 22.88 11.84 44.22 <0.001 Pain: Moderate-Very severe (Ref. None-Mild) 542 15.53 7.65 31.53 <0.001 BMI [continuous] 537 1.06 1.02 1.11 0.009 Vigorous - intensity physical activity [30 minutes/week] 491 0.98 0.96 1.00 0.047 Moderate - intensity physical activity [30 minutes/week] 486 0.97 0.94 1.00 0.051 Year of study participation: [year] 542 1.18 1.08 1.28 <0.001 A ge at study: [years] 542 1.05 1.02 1.08 0.001 Sex: Female (Ref. Male) 542 1.56 1.02 2.39 0.042 Civil status 537 Single Ref. Married 1.11 0.67 1.84 0.695 Divorced/Widowed 1.16 0.37 3.59 0.801 Educational achievement 536 Compulsory school 0.95 0.31 1.59 0.003 Vocational training Ref. Upper secondary education 0.21 -0.35 0.76 0.467 University education 0.39 -0.37 1.16 0.312 Employment situation 528 Employed Ref. Unemployed 4.21 2.07 8.58 <0.001 In education 1.35 0.70 2.63 0.370 Note: Bold font indicates results with p-values <0.05. Abbreviations: BMI=body mass index, CI=confidence interval, CNS=central nervous system, OR=odds ratio, SMN=second malignant neoplasm a number of organ systems affected by late effects (neurological, musculoskeletal, cardiovascular, vision, hearing, endocrine, pulmonary, digestive, or urinary) b T-score≥63 on at least two dimensions or the Global Severity Index of the Brief Symptom Inventory BSI-18 c Predicts fatigue pattern perfectly, observations not used. Fatigue in survivors of childhood cancer. Supplemental Material. Christen et al. 2025 13 Appendix 1. Detailed information on coding and scoring of explanatory variables from the SCCSS and the general population  Year of diagnosis: We recoded year of diagnosis into 10-year categories (19761985, 1986-1995, 1996-2005, 2006-2015).  Health status: Current health status was assessed with one item of the SF-36 (“In general, would you say your health is: 1 – Excellent; 2 - Very good; 3 – Good; 4 – Fair; 5 – Poor”) and recoded into “good-excellent” for scores 1-3 and “fair-poor” for scores 4-5. We scored the SF-36 with subscales according to the scoring manual.1  Organ-specific late effects: Prevalence of specific late effects was assessed by asking survivors whether they currently experienced specific health conditions of different organ systems (yes, no): neurological, musculoskeletal, cardiovascular, vision, hearing, endocrine, pulmonary, digestive, or urinary. For each organ system, we coded prevalence “yes” if participants endorsed at least one of the specific health conditions assessed. Participants who answered “no” or who left a question blank were included in the “no” categories.2 If participants did not complete any question on health conditions, they were coded to have missing information.  Any late effects: We generated an “any prevalence of late effects” variable coding “yes” if a late effect of any organ system (neurological, musculoskeletal, cardiovascular, vision, hearing, endocrine, pulmonary, digestive, or urinary) was prevalent.  Number of organ systems affected by late effects: We generated a variable “number of organ systems affected by late effects” by counting how many organ systems were affected by late effects for each participant (possible range: 0-9 organ systems (neurological, musculoskeletal, cardiovascular, vision, hearing, endocrine, pulmonary, digestive, or urinary)).