HSCs, also known as CD34⁺ cells, are the simplest stem cell source for generating platelets. At present, HSCs are mainly obtained from the BM, umbilical cord blood (UCB) and peripheral blood (27,28).

UCB-derived HSCs typically have a higher ability to proliferate compared with the other two types of cells (29). However, a previous study demonstrated that UCB-derived CD34⁺ cells were difficult to mature fully, with <10% MKs induced from UCB identified as polyploid after one week of thrombopoietin (TPO) induction (30). Current UCB availability is also limited.

Compared with BM and UCB, peripheral blood is easier to obtain and the procedure is less invasive. An increasing number of studies preferentially use peripheral blood as a source of HSCs (31,32). Following procurement, HSCs may be isolated and further cultured into MKs. However, the main problem with peripheral blood is that HSCs are rare (<0.1% of all nucleated cells in the circulation), resulting in an insufficient yield of MKs (33–35). In addition, cell culture is very slow and labor-intensive for large-scale production (36).

iPSCs are created by artificially inducing non-pluripotent cells to express specific genes (37). Due to the limited availability of HSCs, further studies on iPSCs have been undertaken. Over the past years, iPSCs have shown great potential in biomedical research (38,39).

Feng et al (40) created a three-step protocol to generate MKs and functional platelets from iPSCs in a scalable manner within 20 days. In addition, the MK progenitors produced with their method may be stored at low temperatures and proliferate rapidly within a short time. Further analyses demonstrated that iPSC-derived platelets exhibited no major differences with platelets in the circulation. Subsequently, iPSC-derived platelets without major histocompatibility antigen (HLA) were successfully generated by knocking out the β2-microglobulin gene (40). Another study generated a stable HLA-universal iPSC line by silencing the expression of HLA class I up to 82% successfully (41). This HLA-universal iPSC line was able to renew MKs and functional platelets with low immunogenicity. More importantly, iPSC-derived HLA-universal MKs had the ability to escape antibody-dependent cell-mediated cytotoxicity and produce platelets for transfusion. Once the HLA-universal MKs and platelets are efficiently generated, the problem of PTR and limited platelet supply may be resolved.

hESCs are primitive pluripotent stem cells derived from a human blastocyst inner cell mass. They can be propagated indefinitely in vitro, providing an ideal unlimited source for large-scale production of platelets (2). Several methods for differentiating hESCs into MKs have been developed over time (42). Gaur et al (42) demonstrated that hESC-derived CD41⁺/CD42⁺ MKs expressed von Willebrand factor and released functional platelets, but the final number of platelets was small. Subsequently, different culture approaches were described to improve the platelet yield. Lu et al (2) achieved 60 MKs per starting hESC with a feeder-free hESC culture approach. These hESC-derived platelets were demonstrated to have the same characteristics of platelets in the blood. Recently, the TPO gene was inserted into the adeno-associated virus integration site 1 locus of the hESC genome (43), creating cell lines stably expressing and secreting TPO. As a result, the production of hESC-derived platelets was increased, and the function of these platelets was comparable with that in the peripheral blood.

With further studies, more strategies for producing platelets have been designed for clinical use (13). ASCs represent an attractive choice for platelet production in vitro. As ASCs contain certain essential genes indispensable for MK differentiation and platelet production, they can differentiate without gene transfer (44). Furthermore, ASCs may secrete endogenous TPO that promotes platelet production (45). Tozawa et al (46) reported a manufacturing system for platelets from the ASC line (ASCL). In the aforementioned study, ASCs were cultured in MK lineage induction media. At day 8 of culture, the maximum number of ASCL-derived MKs was achieved, and ASCL-derived platelets were obtained, with a peak at day 12 of culture. The inspection results of platelet-related functions are satisfactory.

The dimensions, hardness, matrix components and other conditions of the BM microenvironment accurately mediate the effects of environmental factors on platelet production. A variety of associated studies have improved our understanding of this process (65–68).

A three-dimensional environment expands the contact area of MKs with the surrounding environment. Platelet production may be promoted through the interaction of proplatelets with the microenvironment (65,66).

Transient receptor potential cation channel subfamily V member 4, sensitive to ion channels, can trigger calcium influx, β1 integrin activation and internalization, and human Akt phosphorylation in order to promote platelet production (67); this process only occurs when MKs adhere to a softer instead of a harder matrix. Experiments demonstrated that lysyl oxidase (LOX) may modulate the stiffness of the BM matrix via collagen crosslinking (68). Therefore, appropriate conditions, including increased LOX levels and a softer matrix, favor platelet production. Prior to blood cell release in the circulation, the interaction between progenitor cells and the vasculature is crucial (68). In order to simulate the vascular network, a custom perfusion chamber containing a multi-channel lyophilized silk sponge was constructed (68), which increased platelet production efficiently.

Previous studies observed in vivo flow dynamics and have conducted a series of experiments (24,69). It was reported that turbulent flow was a crucial physical factor for platelet release (24). Based on this result, a novel bioreactor with a flow chamber and multiple pillars was developed, and further experiments produced a higher number of platelets (69).

The factors most widely used to promote platelet production include interleukin-3 (IL-3), IL-6, IL-9, IL-11 and TPO (70–72).

IL-3, IL-6, IL-9 and IL-11 affect TPO-induced MK production indirectly (70–72). Experiments in vitro indicated that the addition of mixed cytokines mentioned above could stimulate platelet production (73–75). IL-3 and TPO act synergistically in promoting MK differentiation (73). In the inflammatory state, IL-6 promotes proplatelet formation by increasing the level of TPO (74). Stem cell factor also plays an important role in promoting cell proliferation in the early stage of MK differentiation (73).

TPO is a major regulator of platelet production and is mainly produced by liver cells in serum (75). TPO combines with Mpl to regulate the differentiation, development, maintenance and proliferation of HSCs and MKs (75). The level of free TPO in the plasma increases when the platelet count decreases, stimulating hematopoietic progenitor cells in the BM to differentiate into MK lines to produce more platelets (76).

Eltrombopag (EP) is a second-generation TPO receptor (TPO-R) agonist that promotes the differentiation and proliferation of MKs and platelet generation. It is combined with the c-Mpl transmembrane area of the TPO-R of MKs, leading to the activation of Janus kinase 2 and tyrosine kinase 2, and phosphorylation of related signaling pathways. All these processes induce proliferation and differentiation of MKs and platelet generation (77,78). All these processes induce proliferation and differentiation of MKs, as well as platelet generation (77). EP was first approved by the Food and Drug Administration for clinical use in 2008. Reported results to date have indicated that thrombocytopenia caused by chronic immune thrombocytopenic purpura (78), severe aplastic anemia (79) and chronic infection with the hepatitis C virus (80) notably improved following administration of EP.

Romiplostim is a synthetic polypeptide that activates downstream signals and stimulates platelet production by binding to the TPO-R on MKs (81). A multicenter study of romiplostim for chemotherapy-induced thrombocytopenia was conducted in solid tumors and hematological malignancies; the results demonstrated that 71% of patients responded to romiplostim and weekly dosing was found to be superior to intracycle dosing (82). An analysis of five clinical trials proved that romiplostim self-administration could achieve 95% response without adverse effects (83). Hosokawa et al (84) found that high-dose romiplostim was highly effective in patients with AA who were refractory to EP. Furthermore, sequential therapy with EP followed by romiplostim may further improve the prognosis of patients with AA who are refractory to conventional therapy (84).