Impact of lower extremity muscle strength on the clinical outcome of patients with allogeneic hematopoietic stem cell transplantation

Purpose Muscle strength decline is reported to predict mortality in many cancers. However, there is little knowledge of the relation between muscle strength decline and clinical outcome of allogeneic hematopoietic stem cell transplantation (allo-HSCT). This study aimed to determine the impact of pretransplant lower extremity muscle strength (LEMS) on post-transplant overall survival (OS) and nonrelapse mortality (NRM). Methods In this retrospective cohort study, 86 adult patients underwent allo-HSCT during 2012-2020. LEMS was dened as knee extension force divided by patient’s body weight. The patients were divided into low, middle, and high LEMS groups based on pre-transplant LEMS. OS was measured using the Kaplan–Meier method and the Cox proportional hazards model. Cumulative incidence of NRM was evaluated using the Fine and Gray method, with relapse considered as a competing risk event. Results Probability of OS was signicantly lower in the low and middle LEMS groups (hazard ratio [HR] 2.98, p=0.020 and HR 2.68, p=0.034, respectively) than in the high LEMS group on multivariate analysis. OS at 5 years was 26.0%, 36.5%, and 72.5% in the low, middle, and high LEMS group, respectively. Risk of NRM was signicantly lower in the low LEMS group (HR 5.62, p=0.038) than in the high LEMS group. Conclusions Pre-transplant LEMS was a signicant factor in predicting OS and NRM.

Introduction Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a potential curative treatment for patients with high-risk hematological malignancies. However, patients who have undergone allo-HSCT may have severe treatment-related side effects and complications including infections and graft versus host disease (GVHD). Transplant-related mortality rate is an additional major problem. In Japan, an "Annual Report of Nationwide Survey 2019" [1] reports a 1-year survival rate of 64.2% in patients with any disease who received allogeneic transplants, with a premature mortality rate of one in three.
Pre-transplant physical function has been reported to be related to post-transplant survival [2,3], and muscle strength is a factor that directly affects these physical functions. Although muscle strength has been reported to be associated with survival in healthy adults [4] and in patients with cancer [5], there are no reports on its association with long-term prognosis in allo-HSCT. As a factor related to muscle strength, skeletal muscle mass measured by computed tomography (CT) imaging has been reported to be associated with long-term prognosis after transplantation [6][7][8]. However, problems with this measurement method are that there is risk of radiation exposure and only a few facilities can perform the measurement because only they have the required software for measurement analysis. In contrast, muscle strength measurements are low risk, can be performed at any time, and are useful for screening physical function before transplantation.
In recent years, maintenance or improvement in lower extremity muscle strength (LEMS) has been reported to be possible by exercise intervention during induction therapy before transplantation [9]. If the relationship between pre-transplant LEMS and long-term prognosis can be clari ed, then it can be a target of pre-transplant exercise intervention. Therefore, in this study, we investigated the impact of pretransplant LEMS on OS, non-relapse mortality (NRM), and early adverse events after allo-HSCT.

Patients
This was a retrospective cohort study involving 96 consecutive patients (aged ≥ 18 years) who had undergone allo-HSCT between January 1, 2012 and November 30, 2020. LEMS data were missing in 10 patients, so 86 patients with complete LEMS data were included in the analysis. After discharge, these patients were followed up periodically at our hospital's outpatient clinic. The study was approved by the Institutional Review Board of Tokushima University Hospital and performed in accordance with the ethical standards established in the 1964 Declaration of Helsinki and later amendments. Informed consent was obtained from all patients prior to the study.

LEMS measurement
We evaluated LEMS by measuring the muscle power of the quadriceps femoris (QF) using a hand-held dynamometer (μ-TAS F-1; ANIMA, Tokyo, Japan) [10,11]. The patient sat on a training bench with their feet hanging slightly above the oor with their knee joint at 90°. The trunk was perpendicular to the bench and both hands touched the bench surface on either side of the trunk. The dynamometer was attached to the distal surface of the tibia and a belt secured the dynamometer to one of the bench legs.
The patient performed knee extension with maximum effort for 5 s. Isometric knee extension force (IKEF) was measured twice bilaterally, and the average of the highest values for the right and left IKEFs were recorded. Intraclass coe cients for this method were 0.94 (95%CI: 0.89-0.97) [12]. The average values of the IKEF (kg) were divided by the patient's own body weight (kg) to account for differences in the patients' body sizes. IKEF per body weight (%) was selected for analysis as LEMS. In all patients, LEMS was measured within 2 weeks before allo-HSCT. Measurements were performed by 4 trained physiotherapists.

Study endpoints and variables
The patients' clinical and follow-up data were obtained from medical records. OS and NRM were set as the endpoint at 5 years follow-up. OS was de ned as time from the date of transplantation to the date of death. NRM was de ned as death from any cause in continuous complete remission or no progression. Acute GVHD was diagnosed and graded according to the criteria described previously [13]. Respiratory failure was de ned as requiring mechanical ventilation, and renal failure as requiring renal replacement therapy. We collected the following variables: age, sex, body mass index (BMI), diagnosis, disease risk, time from diagnosis to transplant, comorbidity (hematopoietic cell transplantation comorbidity index; HCT-CI), stem cell source, graft types, conditioning regimen, GVHD prophylaxis, nutritional status (Mini Nutritional Assessment -Short Form), and general condition (Eastern Cooperative Oncology Group Performance Status [PS]).

Statistical analyses
The patients were divided into 3 groups according to tertiles of their baseline LEMS. The IKEF per body weight in the low tertile (low LEMS) group was 15.1%-34.1% in females and 25.6%-46.6% in males, that in the middle tertile (middle LEMS) group was 34.8%-50.1% in females and 46.7%-59.6% in males, and that in the high tertile (high LEMS) group was 51.3%-66.7% in females, 62.8%-92.7% in males. Patient characteristics were compared between the LEMS groups using Fisher's exact test for categorical variables and ANOVA for continuous variables. The probability of OS was estimated using the Kaplan-Meier method, and the log-rank test was used to evaluate differences between the groups. Cox proportional-hazards regression model was used to analyze OS. The probability of NRM was estimated using cumulative incidence curves, with relapse de ned as a competing risk. The Fine and Gray method was used to evaluate the differences between the groups.  Table 2). The cumulative incidence of NRM at 5 years was 40.9% in the low LEMS group, 32.0% in the middle LEMS group, and 6.9% in the high LEMS group (p = 0.0189, Figure 1b). Multivariate analysis for NRM showed that the low LEMS group (HR = 5.62, 95% CI = 1.10-28.76, p = 0.038) was independently associated with risk of NRM (Table 3).
In acute adverse events after allo-HSCT, the incidences of acute GVHD (≥ Grade 2) and sepsis were signi cantly higher in the low LEMS group than in the high LEMS group (Table 4). In the other items, the incidences were highest in the low LEMS group and lowest in the high LEMS group.

Discussion
This study is the rst to investigate the impact of muscle strength on OS and NRM in patients who have undergone allo-HSCT. We found that pre-transplant LEMS signi cantly in uenced post-transplant clinical outcome. The middle and low LEMS groups showed a signi cantly lower rate of OS than the high LEMS group, and the 5-year survival rate in the high LEMS group was 2.0-fold that of the middle LEMS group and 2.8-fold that of the low LEMS group. The incidences of post-transplant acute GVHD (≥ Grade 2) and sepsis were signi cantly different between the groups. As a result, the incidence of NRM was signi cantly higher in the low LEMS group than in the high LEMS group.
Measuring LEMS using the hand-held dynamometer evaluates pre-transplant physical function easily, objectively, and quantitatively. This measurement may help predict the mortality risks of transplantation that could not be detected by previously reported methods. Previous studies have assessed physical function by patient self-assessment [3] or as observed by a medical doctor [2,14]. Wood [14] reported that PS ≥ 2 was a signi cant risk factor for NRM (HR = 1.44, 95% CI = 1.26-1.63, p < 0.001) and OS (HR = 1.77, 95% CI = 1.60-1.96, p < 0.001). In the present study, although all patients in the middle LEMS group were PS 0 or 1, they were found to be at risk of OS and NRM. Therefore, objective assessment of LEMS could provide useful information for decision making when evaluating not only for patients with poor PS, but also for patients with relatively good PS. Furthermore, increasing the physical status of such patients should be one of the aims of pre-transplant rehabilitation.
Proin ammatory cytokines and immune function may be related to LEMS and the development of posttransplant adverse events. A signi cant association between decreased LEMS and increased TNF-α and IL-6 has been reported in older Caucasians [19]. At 1 week after all-HSCT, increased levels of TNF-α receptor I, a surrogate marker for TNF-α, is strongly correlated with the later development of GVHD [20]. IL-6 is also associated with the development of GVHD [21]. Patients in the low or middle LEMS groups may have increased TNF-α and IL-6 levels due to decreased muscle strength, which may lead to the development of GVHD. The GVHD is itself a risk factor for mortality [22], and systemic steroid therapy for GVHD treatment can cause infection [23]. Muscle strength itself is a risk factor for mortality in various clinical populations [24][25][26][27]. In the present study, the incidence of sepsis, thrombotic microangiopathy and/or veno-occulusive disease, and respiratory failure increased as muscle strength decreased, and the incidence of NRM within 1 year was signi cantly higher in the low LEMS group than in the middle and high LEMS groups.
Patients undergoing allo-HSCT have many factors that reduce muscle strength before transplantation, including induction and consolidation therapy. During that period, patients experience disuse syndrome due to fatigue, which causes nutritional disturbance due to nausea. LEMS in patients undergoing HSCT has already declined pre-transplantation [10,11]. The median period between diagnosis and transplantation in this study was 7 months in all LEMS groups, and adequate physical exercise and good nutritional control during that period could improve the outcome of allo-HSCT.
This study has several limitations. First, this was a single institutional and small cohort study, so future multi-center studies with larger cohorts are necessary. Second, we measured and analyzed the muscle strength of only the knee extensors, but comprehensive assessment and analysis of the skeletal muscles (including muscle mass) are required to clarify the impact of sarcopenia on post-transplant outcomes. Third, patients whose muscle strength was not measured pre-transplantation and were excluded from the study may have had a poor medical condition and prognosis, which may also have in uenced outcomes.
In conclusion, LEMS before allo-HSCT was associated with post-transplant acute adverse events and could be a predictor of OS and NRM. Encouraging patient physical activity and maintaining muscle strength pre-transplantation could lead to improved post-transplant outcomes.     Figure 1 (a) Probability of overall survival and (b) cumulative incidence of non-relapse mortality in patients with low, middle, and high pre-transplant lower extremity muscle strength