Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is an effective treatment method to reconstruct the normal hematopoiesis and immune function of patients. However, posttransplantation complications remain an obstacle for recipients to achieve long-term survival. Dyslipidemia after allo-HSCT is a common event, and it is defined as disorders of lipoprotein metabolism, including high total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), triglycerides (TGs), and low high-density lipoprotein cholesterol (HDL-C) [1, 2]. In the general population, the incidence of dyslipidemia increases with age, and it can lead to a rising incidence of abdominal obesity, cardiovascular disease (CVD), and atherosclerotic cardiovascular disease (ASCVD) [3].

Previous studies have indicated that dyslipidemia after allo-HSCT shares some similar prevalent characteristics within the general population. A nonignorable question in clinical practice is that dyslipidemia is often underestimated and undertreated for transplantation recipients. It has been reported that the prevalence of dyslipidemia in transplant recipients ranges from 40 to 80% [4, 5], especially after kidney transplantation, which can be as high as 80% [5]. Oudin et al. analyzed 170 patients who received allo-HSCT for childhood leukemia. They found that the cumulative incidence of metabolic syndrome (MS) was 13.4% at 25 years and 35.5% at 35 years, which was significantly higher than that among the French population (4% and 5.6%, respectively) [6]. Another study confirmed that patients who underwent allo-HSCT were at higher risk of developing CVD more often and earlier than the age and sex matched healthy population [2]. Therefore, a better understanding of the prevalence and pathogenesis of dyslipidemia and the provision of effective prophylactic and therapeutic strategies are essential and urgent to improve the longevity and outcome of allo-HSCT recipients.

In recent decades, the most widely accepted etiologies of dyslipidemia after allo-HSCT have included age, sex, immunosuppressive agent application, and endocrine dysfunction. Recently, some studies have proposed that there might be an interaction between dyslipidemia and acute and/or chronic graft-versus-host disease (aGVHD and/or cGVHD) [2]. However, literature that interprets the exact molecular mechanism and signaling pathways involved in dyslipidemia during the development of GVHD is sparse. DeFilipp et al. published a comprehensive and practical guideline on screening and preventing MS and CVD in recipients following HSCT on bone marrow transplantation in 2017 [7]. This recommendation included multiple aspects, including lifestyle improvement, risk factor assessment, and the monitoring frequencies of the associated hematological parameters. However, it does not address many aspects of lipid-lowering medications. In this review, we summarized the pathogenesis of dyslipidemia and focused on the association with GVHD in patients who underwent allo-HSCT. The existing lipid-lowering treatments and promising target agents will also be discussed in this review.

Prevalence of dyslipidemia after HSCT

Compared with the healthy population, dyslipidemia is more common in allo-HSCT recipients. It has a younger age of onset in patients who have undergone HSCT, especially in allo-HSCT recipients [8]. Premstaller et al. reported that the prevalence of dyslipidemia in patients with hematopoietic diseases before transplantation was 36% and 28% in the autologous and allogeneic groups, respectively. This might be explained by the fact that patients who underwent auto-HSCT were older than those in the allo-HSCT group [8]. Dyslipidemia is a frequent long-term complication with underestimated secondary consequences, that could be very severe and potentially life-threatening, such as thromboembolic events and arteriovascular disease. Compared to other complications, such as infection and GVHD, dyslipidemia has often been ignored by many clinicians because it usually develops slowly and insidiously.

The next question is which period is the most high-risk onset for dyslipidemia posttransplantation. Premstaller et al. concluded from a retrospective study that the prevalence rose to 62% and 74% 3 months after HSCT. At 25 years, it was 67% and 89% [8]. Bis et al. reported that the most common disturbances in the first half of the year after HSCT included low HDL levels (41.36%), increased levels of TGs (68.35% of the tested population) and TC (38.46%) [9]. HDL is often recognized as a protective factor against lipid metabolism. Consistent with the general population, female patients exhibited relatively higher HDL levels in the first 6 months than male patients after HSCT [9]. This difference might be explained by the fact that estrogens are considered a protective factor to retain the activity of the hydrolytic enzymes of lipids.

Overall, dyslipidemia is a more common complication post-transplantation than we realized, especially in allo-HSCT recipients. In this review, we will narrow our review scope to allo-HSCT recipients.

Etiology of dyslipidemia in patients after allo-HSCT

During chemotherapy or conditioning, cytotoxic drugs, such as alkylating agents, anthracyclines, antimetabolites, and vinca alkaloids, can induce mitochondrial dysfunction, and endothelial cytotoxicity leads to insulin resistance, steatosis, and hypertension [10]. Furthermore, noncytotoxic drugs, such as steroids, can cause hyperglycemia and dyslipidemia [3]. Their meaningful but incomplete investigation helped us have a better understanding of dyslipidemia. In addition to the common reasons within the general population, such as age, obesity, diet, less physical activity, and menopause, some specific factors can also lead to or aggravate dyslipidemia in allo-HSCT recipients (Table 1).

Table 1 Etiology of dyslipidemia in patient after allo-HSCTInsulin resistance (IR) and diabetes

Previous studies have reported a prevalence of type 2 diabetes mellitus (T2DM), impaired glucose tolerance, and IR in allo-HSCT recipients of 17%, 26%, and 52%, respectively [11]. An Israeli study reported that 3.3% of survivors, at a mean of 6.2 years after receiving a bone marrow transplant (BMT) in childhood or adolescence, developed T2DM [12]. However, the prevalence of T2DM in general adolescence is only 0.0029% in Europe. A multiple risk factor analysis indicated that total body irradiation (TBI), immunosuppressants (such as glucocorticoids), and calcineurin inhibitors (CNIs) were closely associated with T2DM or IR [13]. TBI is popularly used in either myeloablative or nonmyeloablative conditioning. Traggiai et al. reported posttransplant diabetes (PTDM) in 6 of 74 BMT patients (8%) who received TBI and 0 of 11 patients who did not receive TBI [14]. In a study of 34 survivors of auto- or allo-HSCT procedures in childhood, IR was significantly more prevalent in patients who received TBI than in patients who received thoracoabdominal irradiation, patients not receiving irradiation, and healthy control subjects [15]. The association between IR and immunosuppressants will be discussed in the following part of this review.

IR can lead to impaired de novo fat synthesis and dyslipidemia. IR in the liver leads to the export of free fatty acids (FFAs) to the muscles and, in turn, is aggravated by IR. In the case of IR, muscle tissue obstructs sugar utilization. Enhanced activity of hormone-sensitive lipase results in many fatty acids being released from the fat tissue. For patients with T2DM, such free nonesterified fatty acids cannot be converted into ketone bodies but act as raw materials to increase the synthesis of very low-density lipoprotein(VLDL), TGs, and TC in the liver. Therefore, patients with abnormal glucose tolerance or DM before or after transplantation are at a higher risk of dyslipidemia [10, 11, 14, 23].

Immunosuppressants

Many immunosuppressants, such as glucocorticoids, CNIs, and mammalian target of rapamycin (mTOR) inhibitors, affect both TC and TGs metabolism in a dose-dependent manner [3]. CNIs (such as cyclosporine and tacrolimus) have been popularly used in immunosuppressive therapy (IST) since they were approved by the Food and Drug Administration (FDA) in 1983. In a trial, 36 transplant patients treated with cyclosporine for just 2 months resulted in increases of average total cholesterol and LDL of 21% and 31%, respectively [16]. Cyclosporine (CSA) is associated with impaired glucose tolerance in approximately 35% of kidney transplant patients. Sirolimus (SRL), a classic mTOR inhibitor, also exhibits a significant adverse effect on lipid metabolism [17]. Morrisett et al. found that treatment with SRL resulted in an expansion in hepatic VLDL synthesis. Fuhrmann et al. observed the effects of treatment with CSA or SRL on glucolipid metabolism in Wistar rats [18]. Their study showed significantly increased TGs levels in the serum, liver, and skeletal muscle resulting from tissue steatosis caused by the drug.

Despite the potential immunosuppressive effect, these immunosuppressants are a "double sword" when treating GVHD. Furthermore, most of them share a similar metabolic pathway through the cytochrome P450 3A4 (CYP3A4) system with many lipid-lowering agents, including statins, by decreasing the clearance of statins. Therefore, we should pay more attention to the cross-linking effect when we simultaneously use immunosuppressants and statins.

GVHD

In addition to the most concerning target organs/systems of aGVHD or cGVHD, such as the skin, liver, intestinal tract, lungs, and bone marrow, lipid metabolism is also a critical target system of GVHD. However, few studies have reported the detailed mechanism of dyslipidemia during the process of GVHD. The relationship between dyslipidemia and GVHD is often underestimated until some researchers found a decreased frequency of aGVHD if statins were given to patients simultaneously [24]. In 2012, Kaguya et al. characterized the prevalence and risk factors for dyslipidemia in allo-HSCT recipients and found that almost half of them had hypercholesterolemia or hypertriglyceridemia [2]. Monica et al. conducted a retrospective analysis of 121 allo-HSCT recipients. Among them, 25 (22%) patients with aGVHD presented high hypercholesterolemia [19]. Previous researchers have proposed that there might be a mutually causal role between dyslipidemia and GVHD. However, the accurate interactive mechanism of these two abnormal biological processes is uncertain.

To better explore the association between the abnormal immune response and GVHD, T-cell activation and metabolic reprogramming have been the “hot points” in recent years [20]. It is well known that T cells play an essential role and are the primary effector cells during the whole process of aGVHD or cGVHD. T-cell activation is the first and most important step to drive an immune response, attacking the host immune system by mounting potent cytokines. The metabolic changes of T cells are shown in a dynamic model. Initial T cells only rely on oxidative phosphorylation (OXPHOS) to meet their energy requirements [25]. After antigen recognition, initial T cells differentiate into effector T cells (Teff), and the primary energy source of Teff metabolism is converted to anaerobic glycolysis. Anaerobic glycolysis produces only 2 mol of adenylate triphosphate (ATP) per gram of glucose through independent mitochondrial metabolism, oxidizing to pyruvate to produce lactic acid. However, OXPHOS may produce up to 30 ATPs per glucose molecule [20]. T cells from patients with aGVHD polarize toward proinflammatory T cells and have higher glycolytic activity than those from patients who do not have aGVHD. Compared with resting T cells, T cells involved in the process of aGVHD have to increase their oxygen consumption to meet the energy requirements, exhibit hyperpolarized mitochondrial membrane potential, and increase reactive oxygen species (ROS) production. In turn, the release of ROS enhances T-cell activation and expansion, resulting in the development of aGVHD [20]. Inhibition of mTOR with rapamycin reduces glycolysis and enhances fatty acid oxidation (FAO) in donor T cells, reducing alloreactive T cells and enhancing regulatory T-cell (Treg) function; the latter has lower glycolytic activity [26]. In brief, glycolysis is a major metabolic pathway after T-cell activation, and Tregs primarily use the fatty acid and pyruvate oxidation (mitochondrial oxidation) pathways. Therefore, reducing glycolysis might be feasible to interrupt the process of T-cell activation and improve aGVDH.

Similar to aGVHD, dyslipidemia is also common during cGVHD. Previous studies have confirmed that cGVHD in the liver is usually accompanied by significantly increased levels of TC and TGs in the serum [2]. Yoshihiro et al. reported an increase in lipoprotein X (LP-X) in recipients [1]. LP-X is separated from LDL and is mainly composed of free cholesterol and phospholipids. The presence of LP-X in liver disease is regarded as the most sensitive and specific biochemical marker of cholestasis. Furthermore, LP-X may play a crucial role in the development of cholestatic hypercholesterolemia because it cannot inhibit the synthesis of new cholesterol in the liver [27]. LP-X increased in patients with cholestasis and was closely related to biliary obstruction. Thereby, bile salts and cholesterol cannot be removed through the bile duct, leading to gallbladder disease and decreased hepatiltriglyceridase (HTGL) activity [1]. HTGL decompose the TGs and cholesterol esters in cells into FFAs. A low level of HTGL is closely associated with hypertriglyceridemia and hypercholesterolemia. Impaired liver function and destruction of fat cells via autoimmunity after the onset of aGVHD are also causes of hyperlipidemia [2, 28]. In some rare situations, the manifestation of cGVHD involving the kidney is nephrotic syndrome, thus resulting in dyslipidemia.

Overall, the relationship of GVHD and dyslipidemia has attracted great attention to better understand their interaction mechanism, occurrence sequence, and therapeutic target for GVHD. A schematic diagram of dyslipidemia and GVHD is shown in Fig. 1.

Fig. 1

Relationship between dyslipidemia and GVHD (designed with BioRender). A: Preconditioning results in tissue damage, releasing of various inflammatory factors, expansion, and differentiation of T cells. B: GVHD leads to a decrease in serum HTGL content, makes triglyceride fail to break down, and ultimately develops hypertriglyceridemia. Increased LP-X content leads to cholestasis, blocks cholesterol excretion, and eventually leads to hyperlipidemia. C: Extensive immune expansion can result in hypoproteinemia and hyperlipidemia. D: GVHD alters the intestinal microenvironment and affects the synthesis of enzymes that regulate substances

Intestinal microflora

The intestinal flora plays a vital role in maintaining immune homeostasis, regulating intestinal function, and regulating metabolism. The intestinal flora is a complex and colossal ecosystem called the "second gene pool" of human beings—approximately 1014 bacteria in the normal human intestinal tract [29], with more than 800 types [30]. Intestinal inflammation can alter the diversity and composition of intestinal microorganisms, creating a state of dysbiosis [31]. A previous study confirmed that the gut microbiome is related to overweight, dyslipidemia, and IR [32]. Margaret et al. summarized microbial metabolites associated with hyperlipidemia, such as short-chain fatty acids (SCFAs), bile acids, and trimethylamine-N-oxide (TMO) [31]. SCFAs can activate the cyclic adenosine phosphate (AMP)/protein kinase A (PKA)/cyclic adenosine phosphate (cAMP) reaction element-binding protein pathway in the liver, enhancing oxidative metabolism, inhibiting liver fat production, and improving lipid levels. SCFAs can also stimulate the secretion of glucagon-like-1 (GLP-1) and gastroenteric peptide YY (PYY) and reduce the occurrence of hyperlipidemia [32]. TMO downregulates the enzymes CYP7A1 and CYP27A1 that are involved in bile acid synthesis, which might be mediated by the activation of the farnesoid X receptor and small heterodimer partner [31, 33]. Intestinal flora can upregulate the element-binding protein-3 gene, promote cholesterol 7α-hydroxylase (CYP7A1) expression, stimulate bile acid synthesis, and reduce de novo fat synthesis in the liver. The combined action of choleidochrome acid and TGR5 can increase the level of cAMP, stimulate the secretion of type II deiodinase, increase thyroid hormone, promote the consumption of brown adipose tissue and increase heat production, thus improving lipid metabolism and preventing the occurrence of hyperlipidemia and other diseases [34].

In allo-HSCT recipients, the gastrointestinal tract can be damaged, resulting in intestinal microbiota dysbiosis due to the conditioning process (chemotherapy or radiotherapy) or intestinal infection post allo-HSCT [35]. Disturbances of intestinal microbiota may be associated with the development and progression of infectious or noninfectious inflammation, including aGVHD. As a common target organ in aGVHD, intestinal tissue damage can, in turn, aggravate dysbiosis, resulting in an alteration of the enzyme profile in the metabolism of sugar, fat, and proteins.

Posttransplant nutrition and living habits

During conditioning and before hematopoietic reconstitution, the diet of the patients is restricted to maximally conserve digestive and absorptive functions and avoid potential intestinal infection. Even after hematopoietic reconstitution, recipients are also at a high risk of various pathogen infections due to the applications of immunosuppressants and immunodeficiency [36]. Severe mucositis or gastrointestinal complications of GVHD allow many patients to only receive partial or complete parenteral nutrition [37]. Gabriela et al. followed up 198 pediatric patients from 1995 to 2016, with a median follow-up of 3.8 years after HSCT. They concluded that many children required total parenteral nutrition due to severe mucosal toxicities or gastrointestinal complications of aGVHD. The return to the regular oral diet is prolonged and difficult [9]. The input of a large amount of high-sugar and high-fat nutrient solutions dramatically increases the TC and TGs levels in the blood of patients.

Most recipients cannot return to daily exercise in time due to poor performance status and multiple transplant-related complications. Psychological or social factors constrict them from initiating activity, even though their physical condition has recovered well. Yu, J et al. analyzed 686 consecutive patients with acute leukemia who received allo-HSCT between January 2008 and December 2017. Their results showed that 56.4% of patients had standard body mass indices, 17.3% were underweight, 20.4% were overweight, and 5.8% were obese [38]. A previous study reported that one-quarter of relatively younger survivors smoke after HSCT, further deteriorating lipid metabolism. Overall, increased comorbidity was associated with a healthier lifestyle, resulting in the development of dyslipidemia [39].

Others

Genetic factors are also involved in dyslipidemia. Primary hypothyroidism has been reported in 10% to 50% of allo-HSCT recipients. Secondary autoimmunity from graft donors may also cause autoimmune thyroid disease in allo-HSCT recipients, leading to hypothyroidism or hyperthyroidism, especially in HLA-related donors. Hypothyroidism, elevated blood pressure, and hyperlipidemia occur in 45% of patients after liver transplantation [21, 22]. Furthermore, hypogonadism and growth hormone deficiency may predispose to IR, MS, and dyslipidemia. This conclusion is consistent with the fact that sex dysfunction (SD) and hormone deficiency increase with age in the general population. Previous studies have confirmed that the age of SD and dyslipidemia is far younger in sex-matched allo-HSCT recipients [22].

Management of hyperlipidemia after allo-HSCT

To date, there are no widely recognized guidelines for dyslipidemia management post-HSCT. Clinicians usually rely on procedures for managing dyslipidemia after solid organ transplantation. The most commonly used measures to prevent and intervene in dyslipidemia include dietary control, lifestyle improvement, and drug intervention. The guidelines recommend the monitoring content, testing frequencies, and suggested agents for the treatment of hyperlipemia. Recently, new lipid-lowering drugs, such as proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors, have been confirmed to be safe, effective, and tolerant in these populations. Here, we summarize the current therapeutic options for the prophylaxis or treatment of dyslipidemia in HSCT recipients.

General treatment

Like the general population, HSCT recipients should abandon bad habits such as alcoholism and smoking and taste preferences for salt, fat, and sugar. With guidance from the nutrition specialist, recipients should control total energy intake to meet essential daily nutrient requirements, allocate the proportion of nutrient elements, control weight, and perform regular and moderate metabolic exercise [38].

Changing the immunosuppressant regimen

Withdrawal of the immunosuppressants cannot be conducted in most situations. However, it might be feasible to adjust the dose or type of immunosuppressant [40]. Basiliximab and ruxolitinib have been popularly accepted by clinicians to improve steroid-refractory or steroid-resistant aGVHD. Recently, Janus kinase inhibitors, such as ruxolitinib, have been approved by the FDA for application in the treatment of SR-aGVHD. When used as a single agent or combined with calcineurin inhibitor (CNI) reduction, these immunosuppressive agents rarely cross-react with statins. Many new anti-GVHD choices, such as mesenchymal stem cells (MSCs), integrin inhibitors, cytokine modulators, and brentuximab vedotin, have been investigated in multiple clinical trials. Although CNIs cannot be replaced, we might explore more effective and safe combination strategies to avoid adverse effects on lipid metabolism.

Lipid-lowering drugs

At present, the most commonly used lipid-lowering agents are statins, ezetimibe, fibrates, and niacin. In recent years, new lipid-lowering drugs, such as PCSK9 inhibitors, have been gradually applied to the clinic. The lipid-lowering mechanisms, main side effects, and drug interactions of these agents are listed in Table 2. Lipid-lowering drugs and their binding targets are shown in Fig. 2.

Table 2 Lipid-lowering drugs after allo-HSCTFig. 2

Targets of lipid-lowering drugs (designed with BioRender)

Statins

Statins are currently the most popularly used lipid-lowering drugs. Statins inhibit the conversion of 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA), transforming HMG-CoA into mevalonic acid. Stains reduce the production of cholesterol in the liver and lead to upregulation of LDL receptors, thereby increasing LDL clearance. Statins can lower LDL cholesterol in a range of 25% to 60% and reduce lower TGs in a range of 10% to 37% in the general population [44]. Therefore, statins might be beneficial for patients with hyperlipidemia posttransplant, especia