Osteopenia and osteoporosis (OP) refer to conditions characterized by decreased bone mineral density (BMD); however, they differ in the severity of bone loss. Osteopenia represents a milder form of bone loss than OP. The global prevalence of osteopenia and OP is estimated to reach 40.4% and 19.7%, respectively (1), resulting in a substantial health and financial burden. OP can lead to fragile fractures, which can lead to disability and even death in older adults (2, 3).

Lactoferrin (Lf) is primarily found in milk. It is a natural iron-binding glycoprotein with a molecular weight of approximately 78 kDa and consists of >690 amino acids. Lf supplementation has been reported to be beneficial against various diseases, including obesity, type 2 diabetes, atherosclerosis, non-alcoholic liver disease, alcoholic liver disease, and some cancers (4, 5). Studies have reported the positive effect of Lf on OP. To obtain a comprehensive understanding of the association between Lf and OP, this review summarizes the protective effects and underlying mechanisms of Lf treatment in OP.

Endogenous Lf is present in serum, neutrophils, and saliva. Although reference values for serum Lf levels have not been established in the general population, most studies have indicated circulating Lf concentrations of ~500 ng/ml (6–10); however, some studies have reported that Lf circulates at concentrations as low as 500 pg/ml or as high as 3,500 ng/ml (11, 12). Lf levels in biological samples are associated with several diseases, including inflammatory bowel disease (13), Alzheimer's disease, allergic rhinitis (9), and rheumatoid arthritis (11). Lf is a potential biomarker for these diseases; however, studies on the association between serum Lf and bone health are limited, and their interactions remain unclear. In a cross-sectional study, although there was no association between serum Lf concentration and BMD or N-terminal propeptide of type 1 precollagen (P1NP), a positive correlation of Lf with parathyroid hormone and β-crosslaps (β-CTx) was observed in older women (14). Specifically, circulating Lf was associated with bone resorption markers (14). Hanna et al. evaluated the saliva Lf levels in patients with OP and in healthy controls without OP (15). The results indicated that, although not statistically significant, Lf levels decreased in both unstimulated and stimulated saliva from OP patients compared with the control group (15); however, this was a preliminary analysis with no adjustments performed. The predictive value of Lf in OP still requires investigation through large-scale studies.

Studies on the effects of exogenous Lf supplementation for OP have been conducted primarily in cells or animals. Although Bharadwaj et al. found that a milk ribonuclease-enriched Lf supplement could restore the balance of bone turnover within a short period in postmenopausal women (16), the study failed to isolate Lf and report BMD. Therefore, direct evidence from clinical trials remains lacking. Low research priority and limited market attention might be two important reasons. On the one hand, more studies on OP mostly focused on bisphosphonates, denosumab, and hormonal therapy (17), and Lf was regarded as a relatively low priority in scientific resource allocation. On the other hand, Lf had a smaller market scale than medicines or other classical nutrients (e.g., 247 million for Lf in 2025 vs. 1.3 billion for vitamin D in 2022), resulting in insufficient support for clinical trials (18, 19). Moreover, cohort studies on the associations between Lf intake and BMD were also hard to conduct. Since almost all Lf intake was from milk, Lf intake inevitably coincided with increased calcium intake. Consequently, even though calcium intake could be adjusted to some extent by statistical methods, the confounding factor cannot be entirely eliminated.

Endogenous Lf is also present in breast milk, saliva, and neutrophils (4, 20). Immunohistochemical analyses have indicated that fetal osteoblasts (OBs) exhibit Lf immunoreactivity, whereas adult OBs do not (21, 22). In the fetus, Lf was detected up to the 18th week of gestation and disappeared after the 30th week (22, 23). Thus, Lf may be involved in bone growth regulation during the early phases (23) but not as an optional biomarker for OP in adults. In addition, Lf may be expressed in osteocartilagineous tumors, chondroblastomas, chondromyxoid fibromas, and osteoid osteomas but not in osteosarcomas, chondrosarcomas, ossifying fibromas, osteochondroma, and enchondromas, which may reflect a mature phenotype of these tumors (21, 23, 24).

Bone remodeling is tightly regulated through crosstalk between bone-forming OBs and bone-resorbing osteoclasts (OCs) (25). Compared with OCs, more in vitro studies have focused on the effects of Lf on OBs. Nagashima et al. found that human recombinant Lf promotes MC3T3-E1 cell differentiation and calcification (26). Another study indicated that Lf mediates the enhanced osteogenesis of adipose-derived stem cells (27). Mechanistically, the mitogen-activated protein kinase (MAPK) signaling pathway (28) and BCL2-Beclin1 signaling-mediated autophagy (29) participate in OB formation. In contrast, Owen et al. compared the anabolic effects of five compounds on OBs. No effect on osteogenic differentiation was observed, and even a high dose of Lf (1 mg/ml) produced an adverse effect (30). Furthermore, certain Lf-derived peptide fractions (fragment residues 624 to 632, also called LPF-C, and amino acids 97–122 from the N-terminus) also induce OB proliferation (31, 32), which warrants further investigation.

Some studies have indicated that Lf not only promotes OBs but also inhibits OC development (33–35). However, Lf does not alter bone resorption in calvarial organ culture, which suggests that Lf does not affect mature OC function (34). In other words, Lf is able to affect immature OCs but not mature ones. Lorget et al. found that Lf inhibited osteoclastogenesis and bone resorption through a mechanism independent of the osteoprotegerin/receptor activator of nuclear factor κB ligand/receptor activator of nuclear factor κB (OPG/RANKL/RANK) (36).

Many factors might lead to discrepancies in the effects of Lf on OBs or OCs, including dose, source, and intervention time. Additionally, cell type might also be an important reason. For example, Lf at the same dose could promote differentiation and calcification in MC3T3-E1 cells (26) but not in human mesenchymal progenitor-derived OBs (30). Iron saturation might further play a role, since some researchers observed that the biological effects of Lf varied with iron saturation levels in other diseases (37). Meanwhile, we should hold a cautious attitude toward the results from the cell-based studies due to the inherent limitations in their evidence hierarchy.

Most studies on Lf have primarily focused on the bovine or human form. Generally, they appear to exhibit comparable activity (38); however, it should be emphasized that the activities are not always interchangeable, because their modes of intestinal receptor recognition is inconsistent (39).

Structure-function relationship studies suggest that the differences are minimal for the effects of the various Lf forms on osteogenic activity (40). The iron saturation level of Lf is not a key factor affecting OB function or mitogenic activity in MC3T3-E1 cells (41). In addition, the glycosylated forms and source of Lf do not alter its mitogenic activity (42). Although Wang et al. found that the osteogenic activity of Lf decreased with increasing iron saturation (43), and Zhang et al. found that bovine Lf appears to have more proliferative capacity compared with human Lf (44), the differences may be minor. Studies on Lf in OCs are relatively insufficient, whereas its osteoclastogenic activity appears to be located in the N-lobe of recombinant Lf (38).

Numerous studies have examined the underlying mechanisms of Lf in OP, as summarized in Table 1. However, many conclusions remain speculative, and the exact mechanisms are poorly understood. A lack of high-quality studies also is an issue. A comprehensive exploration on the mechanisms is necessary for the applications of Lf. Emerging technologies such as spatial metabolomics and single-cell sequencing offer new opportunities for mechanism exploration. Meanwhile, although many pathways are involved in the protective effects of Lf in OP, regulation of the balance between osteoblasts and osteoclasts remains a fundamental mechanism.

Lf is nearly completely degraded in the stomach; however, some fragments resist further digestion (104, 105). Therefore, the biological activity of Lf may depend on its peptide fragments. Lactoferricin (Lfcin) and lactoferrampin (Lfampin) are two Lf fragments of interest that exhibit antimicrobial effects (106). Moreover, the anticancer and immunomodulatory effects of Lfcin have been reported (107, 108). Whether Lfcin or Lfampin influences bone metabolism has not yet been established.

LFP-C (FKSETKNLL) is a peptide from bovine Lf hydrolysates generated through pepsin digestion. Its osteogenesis activity was demonstrated in vitro (31). Molecular docking suggested that the osteogenesis of LFP-C may result from its binding to the key domain (Lys13-Thr15-Gln16-Leu17-Gly18-Asp22) of the epidermal growth factor receptor (EGFR), which activates the MAPK pathway; however, biological validation has not been performed (31).

The LP2 peptide (RKVRGPPVSCIKRDSPIQ) from human Lf has self-assembly properties and skeletal bioavailability. LP2 stimulates OB differentiation through a BMP-dependent mechanism and osteoblastic production of OPG. Moreover, the subcutaneous administration of LP2 accelerates bone healing and bone formation in vivo (73); however, the underlying molecular mechanisms were not examined. Because of the low resistance of Lf to digestion, Lf peptides may exhibit a higher simulation effect than Lf, particularly in vitro. Additionally, the active form of Lf may be its digestive hydrolysate rather than the intact molecule. Therefore, using intact Lf for in vitro experiments may not fully simulate the real in vivo effects, and the peptides may have higher research value for exploring the underlying mechanisms of Lf.

The main source of Lf is milk in the daily diet. In milk, the Lf contents mainly concentrate on 0.1~0.2 mg/ml (109, 110). Thus, an adult can ingest 50-100 mg per day through a regular diet. In a rat study, oral administration of Lf at 2 g/kgBW/d (equivalent to 20 g/d for an adult) for 13 weeks did not produce adverse effects (111). Another study also found that a daily intake of up to 9 g Lf is safe for humans (112). Due to its proven safety, Lf has been approved to be added to infant formula in many countries (113), and FAO and WHO recommend the level of Lf supplementation is 500 mg/kg in infant formula (114). Moreover, for adults, an expert consensus indicated that the recommended daily supplementation of Lf is 200–600 mg (115). It should be noted that these doses are intended for the general population. To date, there is still no evidence-based recommended Lf dose for OP prevention and treatment. Although some studies indicated that excessive doses of Lf may have potential negative effects (78, 116), this concern is likely of limited practical relevance. Due to the high cost of Lf supplements, excessive intake is virtually impossible in the real world.

So far, most evidence for the beneficial effects of Lf on bone metabolism derives from preclinical studies, and the clinical trials are markedly lacking. Although no clinical trial has directly focused on Lf and OP, one randomized controlled trial investigated milk ribonuclease-enriched Lf on bone turnover markers. In this trial, milk ribonuclease-enriched Lf supplementation displayed positive effects on serum bone turnover biomarkers in postmenopausal women aged 40–60 years (16). However, it cannot be confirmed that the changes were attributed to Lf; meanwhile, the study was performed in a special population, and BMD was not determined, which might limit its generalizability and reliability. Despite these limitations, Lf is still remains a promising agent for OP protection due to its broad biological activities and high safety (5, 117). In the future, large-scale randomized controlled trials are required to validate the efficacy of Lf in OP. Of course, Lf, as a natural food component, is not as therapeutically effective as conventional pharmaceuticals. Lf may serve as a preventive agent or adjunctive therapy rather than a primary therapeutic agent in clinical practices.

Due to the poor oral bioavailability of Lf, some groups have developed different formulations for bone health. Lf-embedded type 1 collagen membranes retain their pro-calcification effects during osteogenic differentiation in vitro (118). Injectable scaffolds are also considered efficient Lf delivery systems. Kim et al. reported that Lf-loaded porous polymicrospheres promote osteogenic differentiation by controlling Lf release (119). Liposomes are another important delivery system (120, 121). Recently, these topics have been thoroughly reviewed by other scholars (122, 123) and are beyond the scope of this review. Therefore, we have not discussed them in detail in this review.

Several mechanisms have been reported to explain the effects of Lf on OP (Figure 1); however, these mechanisms are primarily derived from cell or animal experiments. The lack of clinical evidence remains the largest pain point for the applications of Lf. It is valuable to conduct a randomized, placebo-controlled, double-blinded trial to assess the efficacy of Lf. In this trial, BMD is a more valuable outcome besides serum biomarkers such as ALP, β-CTx, and TP1NP. Considering that the immediate effects of Lf (as a natural food component) may be weak, a long-term intervention is recommended. And other proteins without medical effects can be selected as a placebo. It should be noted that we need to maintain a “cautiously” optimistic attitude until the effectiveness of Lf on OP is confirmed. Moreover, the mechanisms through which lactoferrin regulates these pathways remain a “black box,” and deconstructing this box is a research topic that warrants further exploration.