Three main and one additional data collection time points were determined: T0 was the date of enrollment into the study and part of the pre-transplantation workup. These screening procedures included measurements of cardiopulmonary performance (Cycle ergometry), body composition, muscle strength and a motor skill test. Blood samples were taken and examined. In addition, participants were asked to fill out questionnaires regarding quality of life, state of health and - beginning with at hospital admission T_ha- fatigue, muscle weakness and physical activity were documented.

The period around engraftment was defined as T1. Engraftment was defined as a leukocyte concentration >0.5 x 10⁹/L on at least three consecutive days after the transplantation. Accordingly, T1 in this study was the day on which the leukocyte concentration was above the threshold value for the first time (± 2 days). T2 was the last measurement on the day of patients discharge or the day before. At T1 and T2, blood samples were analyzed, grip strength was measured and a motor skill test was performed. During the inpatient stay, the total weekly glucocorticoid dose and daily physical activity were documented.

Statistical analysis was performed using the SPSS software package (SPSS Statistics 21.0, IBM, Ehningen, Germany) and GraphPad PRISM (Version 9.4, La Jolla, USA). Data were tested for Gaussian distribution with Kolmogorov-Smirnov test and time point differences were tested using by One-way ANOVA with repeated measures in the case normality could be assumed. Otherwise, a Friedman’s test was used. Where data of endpoints variables were missing for at least one time point, a mixed-model approach with Tukey’s multiple comparisons test was used. Data of inpatient time in hospital were analyzed using the t-test. To test a possible interdependence of inflammation/cardiovascular function and strength, correlation analysis between Grip strength and inflammatory markers hsCRP, Kynurenine and TNF-alpha as well as the myokine Musclin and Galectin-3 were analysed with Spearman’s rank correlation coefficient. Descriptive statistics of the data are presented as mean ± standard deviation if not otherwise stated. Statistical differences were considered to be significant for values of *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

Tinetti test decreased at all-time points (n=26): T0 (27.6 ± 0.8), T1 (25.8 ± 2.7; p=0.002) and T2 (25.9 ± 3.7; p=0.009). While in T0 no patient had a low or moderate risk to fall, in T1 and T2 two patients developed an increased risk of falling. Grip strength (n=25) decreased significantly at T2 (p=0.002), grip strength percentiles decreased continuously from T0 to T2 (T0 (50.8 ± 30.4), T1 (26.8 ± 20.9; p=0.003) and T2 (22.8 ± 15.8; p<0.0001)). Chair-Rising, one-leg-stand and tandem-test did not differ significantly at any time-point ( Figure 2 ).

The patients reported in BSA-questionnaire that they walked within four weeks prior to T0 about 42.96 ± 40.1 floors per day (n=25). In this time the duration of physical activity in daily and leisure time life was 7789.3 ± 5899.5 Metabolic Equivalent of Task (MET) min/week (n=25). Thus, 65.4% had a high activity level, 34.6% had a moderate activity level and nobody had a low activity level.

In IPAQ the participants stated with an average of physical activity of 235.9 ± 210.1 min/week (n=25) and the additional sports activity was 60.35 ± 233.5 min/week (n=25).

As already mentioned, steroid myopathy caused by the direct catabolic effect of glucocorticoids is a common side effect during alloSCT. Besides the general wide range and individual risk factors (12), myopathy can contribute to the development of GVHD (70). We examined our cohort only during the inpatient stay, so we cannot draw any conclusion about the connection between GVHD and physical activity, as chronic GVHD is normally diagnosed up to day 100 (13) and longer assessment would be needed in our subjects.

Our patients received an average glucocorticoid dose of 4.5 ± 3.9 mg/kg body weight per week and a mean of 27.9 ± 27.7 mg/kg body weight during the entire inpatient stay. Therapy of choice for acute GVHD includes high-dose glucocorticoids (71, 72) and the German Working Group for Bone Marrow and Blood Stem-cell Transplantation recommends (in addition to other immunosuppressive medications) 1-2 mg of prednisolone/kg/body weight/day for up to seven days, then a dose reduction of 20% every 5 days (73). Regarding our mean inpatient stay of around 77 days, the administered dose thus fell well into the guideline recommendations.

Morishita et al. examined 113 patients who received alloSCT and observed correlations between the dose of steroids administered with a decrease in strength of the hand and knee extensors (74). This result could not be replicated in our study where the grip strength significantly decreased compared to baseline measurements, regardless of steroid dosage. One explanation for this could be that the patients in the study by Morishita et al. received an average of 57.6 mg/kg glucocorticoids, which is almost twice the dose received by patients in our study. Thus, the total dose per kilogram of body weight during the observation period is larger in Morishita et al. with presumably higher impact for developing myopathic symptoms

Unfortunately, there was a reduction in the physical activity level in our alloSCT patients after engraftment, nearly after two weeks of inpatient stay. The daily time of endurance training decreased from 14 min/day before engraftment to 7 min/day after engraftment. Strength training also decreased steadily, but from 12 to 10 min/day, thus only the total time of strength training was significantly different. The cause of this development can be explained by the weakened immune system. A possible reason that our patients did not concomitantly report an increased load intensity during training with regard to the Borg scale could be the low daily resistances or intensities when they are active or exercise. It could be speculated that the activity time would drop maybe even lower if there would not be supervised physiotherapy nearly every three days during inpatient stay.

One explanation for the decreased activity time would normally have been increased fatigue, but this was not reported in this case. According to the literature, one would have expected a higher fatigue towards the end of the inpatient stay due to the reduced physical activity (75, 76), but this was not supported by our questionnaire evaluation with similar fatigue during the whole stay.

The evidence of beneficial effects of training – especially for alloSCT patients – is growing. Baumann et al. showed that patients receiving alloSCT can reduce strength losses of the knee extensors by increased training (23) with probably positive outcomes for gait speed and falling risk reduction. Furthermore, a review by Wiskemann et al. reported significant improvement by exercise interventions for physical performance, quality of life and fatigue as well as faster recurrence of immune cells or reduced severity of therapy-related side effects can be estimated. (77). Hence, it seems advisable to lay an additional focus of treatment before, during and after the treatment to reduce alloSCT side effects and reduce hospitalization time.

Unfortunately, the independently performed training documentation by the patients in our study was inadequately accomplished in most cases. Accordingly, the results of the sports documentation can only be assessed to a limited extent. In order to achieve more objectivity, the number of days on which the patients received physiotherapy was also recorded. As a result, this showed that a higher number and more working hours for physiotherapists are necessary to avoid the negative consequences of a lack of supervised exercise training for patients.

The immunosuppressive effect of alloSCT treatment (including a conditioning phase with aplasia) and glucocorticoids can have an immense impact on physical abilities and therapy outcome. Patients receiving alloSCT are in a special situation: their hematopoietic system has to be specifically eliminated and replaced by donor stem-cells due to the underlying hematological disease. In this case, the immune system must re-establish itself. Immune cells comprise approximately 2-6% of the cells in healthy skeletal muscle. Evidence exists that (chronically) elevated inflammatory markers are associated with the development of sarcopenia (35, 78–80) and that immune cells are important for the maintenance of skeletal muscle function and muscle plasticity (81). A randomized control trial with allogeneic bone marrow transplant patients was able to increase the total lymphocyte count by 40.9 cells/μl through a series of bed exercises (30 minutes daily over a period of 6 weeks). The non-exercising control group of the same study showed a decrease of 640.7 cells/μl) without affecting the composition of CD4+ and CD8+ T-cell subsets (28). Thus, fluorescence activated cell sorting (FACS) analysis to assess immune cell subgroups during alloSCT treatment and monitoring might be added to the clinical routine. Unfortunately, as was also the case in the presented study, the required blood volume, time and expertise for clinical FACS analysis is seldom available and leads to potential important missing information.

The association between inflammatory status on the decrease in muscular function in alloSCT patients, especially grip strength and sport motor skills, have already been mentioned. To assess the inflammatory status, the changes of the highly sensitive C-reactive protein (hsCRP), Kynurenine and TNF-alpha were examined in this study. CRP is an acute phase protein produced in the liver in response to inflammation or tumors, and elevated CRP levels before stem-cell transplantation correlate with a reduced probability of survival (82).

Beside the known correlation of elevated CRP levels before stem-cell transplantation with a reduced probability of survival (82), hsCRP can be used for the diagnosis of acute infections and partly for chronic inflammation - such as vasculitis - (83). We observed both an increase in hsCRP and Kynurenine at engraftment compared to baseline measurements, indicating the still high susceptibility of patients after alloSCT. This may be aggravated by the interdependence between severe treatment, increased pro-inflammation and declining physical performance capacity during inpatient stay. In addition to the well-known and established CRP, growing evidence indicate Kynurenine as a reliable biomarker to monitor disease status. Kynurenine is a metabolite during tryptophan metabolism synthesized mainly by the Indoleamine 2,3-Dioxygenase (IDO). Increased IDO activity, on the other hand, is associated with increased mortality (84), while an increase in Kynurenine levels is seen in older adults with reduced physical activity (85) as well as in the context of GVHD-related inflammation (86). Several studies report an association between elevated inflammatory markers such as CRP or Kynurenine and increased fatigue (87–89). The concentration of these markers can be reduced with training (therapy) (90–93) and it was shown that strength and endurance training can reduce fatigue during a hospitalization for alloSCT (75, 76).

Our observed increased Kynurenine levels at engraftment with subsequent baseline values at inpatient release show that Kynurenine may be a valuable addition to hsCRP to examine the acute inflammatory status of alloSCT. Furthermore, it was shown by Westbrook et al. that elevated levels of Kynurenine were associated with increased inflammatory cytokines, decreased grip strength, and decreased gait speed in humans (94). However, it remains problematic in this context that there exists no standardized measurement procedure for Kynurenine in clinical routine, which complicates the comparison of study results. In addition, we could not show an interdependence between hsCRP or Kynurenine with grip strength, so a direct link between these inflammatory markers and strength decline appears unlikely.

The biomarker panel of CRP and Kynurenine could be complemented by TNF-alpha measurements during alloSCT. TNF-α, formerly only seen as pro-inflammatory cytokine, is now regarded as a pleiotropic molecule with anti-inflammatory and immunomodulatory effects (95). Increasing TNF-alpha plays a key role in aGVHD (96) by preceding the onset of aGVHD, before peaking at the time of its development (97–99). It can also be used for the discrimination of patients with and without GVHD, the prediction of survival as well as the estimation of therapy efficacy are reliable application fields of TNF-alpha measurements (99–101). We observed decreased TNF-alpha already at T1, which further decreased at inpatient release T2. As mentioned by the colleagues Kitko et al, TNF-alpha cannot sufficiently be used as only biomarker for the prediction of the development of GVHD (96), but the combination with hsCRP and Kynurenine seems promising.

For a more sensitive view of muscle wasting during alloSCT, we examined the development of the novel biomarkers Musclin and Galectin-3 during the inpatient stay. The synthesis of the myokine Musclin is up-regulated in skeletal muscle in response to physical activity and subsequently secreted into the systemic circulation promoting skeletal muscle mitochondrial biogenesis and exercise endurance. As it was observed that Musclin may be able to alleviate cancer-induced muscle wasting during cachexia, which is characterized by systemic inflammation through elevated levels of pro-inflammatory cytokines such as TNF-alpha, it might serve as a novel marker for cachexia development and potential therapeutic target. We observed decreased levels of Musclin at engraftment, which indicates the decline in muscle synthesis during the inpatient stay and is in line with our measured declining grip strength, although no direct correlation could be observed. Most studies examining the mechanistic effect of Musclin have been determined in rodents, so further data are needed to finally elucidate the monitoring value in humans (36).

In contrast, Galectin-3 as marker for cardiovascular disease, showed highest values at inpatient release. In previous studies, this molecule was upregulated in human cancer development and progression (102). Thus, a recent study by Caputo et al. suggested Galectin-3 as a potential therapeutic target especially in the early phases of prostate cancer progression and metastasis (41). In addition, our observed values may indicate Galectin-3 as potential long-term biomarker of apoptosis/neogenesis during alloSCT therapy and cardiovascular health status due to its additional involvement in exercise and stress induced damage (103). But in how far these observations can be transferred to alloSCT and subsequently transplantation risk and health monitoring, has still to be evaluated in further controlled clinical studies.

In this study, motor skill skills were monitored using grip strength measurements as well as the Tinetti test, the chair-rising test, and various balance exercises during the course.

At T0, physical status was also assessed using a bioimpedance analysis, muscle strength measurements of the knee extensors, and other sport motor tests such as the SPPB test. For a more precise assessment of the progression, a renewed measurement of the body composition and muscle strength measurement of the knee extensors would have offered more information but was not possible due to the lack of equipment on the ward and for possible safety reasons (possible fall with dislocation of the catheter) as well as due to the expected false high results (the patients would have to additionally move an infusion stand in contrast to T0). Finally, in this study, all patients reported being very active before their diagnosis with no diagnoses low activity level before study start.

The above pre-transplant screening with assessment of sarcopenia was performed in about 2 hours and assessed by sports physicians. Additional resources are needed to introduce such procedures in transplant centers, but the benefit seems given the expenses of peri-transplant complications very valuable.