CK is widely distributed in tissues and is generally associated with intracellular energy metabolism, muscle contraction, and ATP regeneration [21]. In the clinical setting, serum CK is a biomarker mainly for identifying myocardial infarction and diseases related to muscle injury. The prognostic value of CK is still under investigation. Several studies concerning the prognostic role of serum CK in other cancers, including hepatocellular cancer [15], esophageal cancer [12], and gastric cancer [13], had indicated low serum CK concentration was a predictor for poor OS and recurrence‑free survival (RFS) in male patients rather than female patients. However, discussion and investigation for mechanisms behind these interesting findings remain insufficient.

Our previous study found that PC could lead to a decrease in serum CK [14]. Considering CK skeletal muscle isoenzyme (CK-MM) is the CK isozyme constituting over 90% of the total serum CK activity [22], which mainly comes from skeletal muscle [23], we believe decreased CK could be the result of muscle wasting due to cancer-related cachexia, as reported by Weber et al. [24] and Takamori et al. [10]. Therefore, serum CK may have a prognostic significance for patients with pancreatic cancer, which is worthy of further investigation. The results of the present study showed a significantly higher incidence of sarcopenia in the low CK group compared to the high CK group in male patients. Additionally, patients in the low CK group had poorer OS and DFS than those in the high CK group. We further investigated the relationship between CK and body composition. We found a significant correlation between CK levels and SMA in male patients, whereas there was no significant relationship in female patients, which explained the absence of association between CK and sarcopenia in female patients.

Two core questions should be addressed before elucidating the mechanisms underlying the prognostic value of CK: (1) Why does CK decreases in cancer patients? (2) why are there sex differences in the prognostic significance of CK? We tried to answer these two questions based on existing evidence, as shown in Fig. 4. Serum CK is predominantly derived from the efflux of CK-MM in muscle fibers [22], and the reason for the decrease in CK should be traced back to the physiopathological changes in skeletal muscle. In cancer patients, both the quality and quantity of skeletal muscle fibers decline, and the decrease of serum CK could possibly be the result of both aspects. Our results show a positive correlation between serum CK and abdominal skeletal muscle area (quantity), although the correlation coefficient was moderate. Considering CK plays a critical role in energy metabolism of muscle fiber, and the association between muscle quality and serum CK levels requires further investigation. In the study by Yen et al. [11], sarcopenia was diagnosed according to a combination of appendicular SMI, skeletal muscle strength, and endurance in subjects without cancer. Similarly, their study found that low serum CK was significantly associated with the occurrence of sarcopenia. Therefore, the decrease in serum CK may be a common phenomenon associated with muscle wasting, and not unique to cancer patients. The explanation that low serum CK related to poor prognosis possibly due to tumor cells consuming serum CK leading to decreased CK, as stated in previous study [15], is doubtful. Sarcopenia in cancer patients does not simply indicate the presence of skeletal muscle wasting but reflects a disturbance of immunonutritional status. In the systemic inflammatory environment triggered by tumors, TNF-α can induce the suppression of myogenic differentiation factor D (MyoD), which is a vital regulator of the synthesis of CK-MM [24,25,26]. Consequently, the efflux of CK-MM may decrease, leading to reduced serum CK activity. Therefore, a decrease in CK levels could be the result of muscle atrophy and the repression of CK-MM expression.

Muscle wasting, serum CK decrease, and sex. In cancer patients, skeletal muscle wasting leads to a decrease in SMA, with a more pronounced reduction in type II muscle fibers compared to type I muscle fibers. Additionally, the synthesis of CK-MM in muscle fibers is inhibited in the inflammatory environment caused by cancer, which further contributes to the reduction of CK-MM efflux from muscle fibers into the bloodstream. Estrogen may be a factor contributing to the lack of prognostic significance of CK in females. CK, creatine kinase; SMA, skeletal muscle area

The limited prognostic effects of CK in females of our study were in accordance with former studies. Serum CK concentration is influenced by race, sex, and age [27], which is similar to the characteristics of skeletal muscle mass [28]. The pathophysiological differences in muscle growth [29], muscle phenotype [30], muscle regeneration [31], and muscle atrophy [32] exist between sexes, and muscle wasting mode also exhibits distinct characteristics in male and female patients [33]. These phenomena suggest that research concerning the prognostic value of skeletal muscle should perform sex-stratified analyses rather than studying males and females as a single group. The study by Yamazaki et al. [13] found the prognostic role of CK in gastric cancer patients; however, further subgroup analysis based on sex did not identify significant prognostic effects in female patients, indicating that the insignificant prognostic effect in females was overshadowed by that of males when evaluating males and females as a whole.

Studies indicating a correlation between CK and SMA did not perform sub-group analyses based on sex [10, 34]. Our sex-stratified analyses found a significant correlation between CK and SMA in male patients and an insignificant correlation in female patients, although the result was significant in the entire cohort (P < 0.001, data not shown in the results). The different primary sex hormones in males and females mediate the sex-specific myogenic processes and responses: testosterone has an anabolic effect that promotes muscle growth, while estrogen may exert its muscle-protective effects through various pathways [35]. There are several explanations for the protective effects of estrogen on muscle [36, 37]: (1) Its structural similarity to potent antioxidants like vitamin E likely confers high antioxidant capacity, limiting oxidative damage; (2) Similar to cholesterol, estrogen can intercalate into membrane phospholipids, providing a stabilizing effect on cell membranes; (3) Estrogen reduces the inflammatory response following muscle damage. Thus, in female patients, the efflux of CK in muscle fibers may decline due to the effect of estrogen, and the correlation between serum CK and skeletal muscle mass is subsequently weakened, which explains the insignificant correlation between the serum CK concentration and SMA in our results. Since most of the female patients in our study were postmenopausal, the effect of menopause on serum CK concentration should be considered. Studies have reported that postmenopausal females had higher baseline CK concentrations [38, 39]. However, a large-scale study showed that CK levels in females were stable with increasing age and were unaffected by menopause [27]. In contrast, males experienced a decline in serum CK concentration with aging [27]. The inconsistent findings, possibly due to small sample sizes and underlying confounding factors, highlight the need for higher-quality studies to fully elucidate how menopause impacts female CK levels.

Moreover, females appear to be inherently less prone to muscle wasting. It was found that estradiol prevented cancer-induced skeletal muscle wasting by reducing the actions of related cytokines such as myostatin, TGF-β, and TNF-α [40]. In pancreatic ductal adenocarcinoma, the abundance of endogenous inhibitors of the activin-family ligands (such as follistatin related protein-1, follistatin like-3, and GASP2/WFIKKN2) protects females from the drivers of muscle wasting, and as a result, males showed earlier and more severe sarcopenia than females [41]. In our cohort, the incidence of sarcopenia in male patients was significantly higher than that in female patients (48.5% vs. 32.8%, P = 0.045). These differences and their connection to the sexual disparity in the prognostic value of CK warrant further investigation.

Body composition is increasingly being valued by oncologists due to the prognostic information it conveys. More attention is now given to individual patient characteristics than ever before, in addition to tumor-specific factors [42]. For patients with resectable PC, the skeletal muscle index assessed by CT images could stratify patients into different prognostic categories [6, 43]. Compounds released from muscles may provide insight into the existence of sarcopenia, serving as a convenient alternative to traditional methods of skeletal muscle mass measurement. For male patients with a low CK level, intensive postoperative follow-up or postoperative adjuvant therapy may be needed. It has been reported that PC patients with decreased skeletal muscle mass have a lower rate of chemotherapy completion compared to those with normal skeletal muscle mass [8], indicating that patients with sarcopenia are more susceptible to the detrimental effects of chemotherapy. The serum CK may provide insight into selecting a chemotherapy regimen and dose intensity as a simple and convenient marker, which requires further investigation to confirm.

In summary, the results of the present study demonstrate the prognostic value of preoperative serum CK in male patients with resectable pancreatic cancer, providing new evidence for the sex dimorphism in CK’s prognostic significance. Additionally, we conducted an in-depth investigation into the association between CK and body composition in cancer patients, offering a possible explanation for the mechanisms underlying the sex dimorphism in CK. There were some limitations to our study. First, due to the retrospective nature of the study, detailed data about CK isozyme levels were not available, which makes it impossible to determine the variation in the CK isozymes. Second, the cutoff values for serum CK were calculated based on subjects from a single institution, the effectiveness of these values requires validation in an external cohort before application in clinic. Third, since CK levels are influenced by race, assessing the prognostic value of CK in different racial groups would be necessary. Additionally, dynamic screening of the CK variation in cancer patients may help to better understand the role CK plays. Hence, further prospective studies are required in the future.