This novel class includes two drugs, luspatercept and sotatercept. They are recombinant fusion proteins that act by inhibiting negative regulators of late-stage erythropoiesis, like activin B and other transforming growth factor beta (TGF β) superfamily ligands (65–70). Thus, they can be viewed as molecular traps that facilitate erythropoiesis by removing biological brakes to the process (24, 26, 71). At variance with EPO, which stimulates the proliferation of RBC progenitors during the early stages of erythropoiesis, ACVR ligand traps distinctly favor late-stage erythropoiesis. Their effect, also through modulation of marrow stromal cells (66), is prevalent in differentiation rather than proliferation of RBC progenitors. This explains why these compounds work particularly well in conditions of pathologically expanded but ineffective erythropoiesis, like thalassemic syndromes and low-risk MDS (see below).

Of note, sotatercept was initially developed to increase bone mineral density in malignant bone disease and in osteoporosis (72, 73), with the unexpected observation of an erythroid response (73, 74). Later, the same effect was observed with the analog luspatercept (75). Improvement of ineffective erythropoiesis was subsequently confirmed in murine models of b-thalassemia (69). This recently translated into one of the major achievements in the treatment of patients with beta-thalassemic syndromes, where luspatercept has been proven to increase Hb and substantially decrease transfusion needs (76). Indeed, luspatercept has been preferentially developed in clinical studies because of a higher specificity activin B (rather than for activin A) as compared to sotatercept, with a lower probability of determining off-target effects (77). The sustained effect on erythropoiesis has also prompted the use of luspatercept for anemia associated with hematologic malignancies (77, 78). Low-risk MDS have been selected because of the need for alternative strategies to frequent RBC transfusions in anemic patients with prolonged anticipated survival. The MEDALIST trial enrolled 229 transfusion-dependent patients affected by MDS with ringed sideroblasts, who were either refractory or not candidates for ESAs because of high baseline EPO level (79). Transfusion independence for ≥8 weeks was observed in 38 percent of the patients in the luspatercept group, as compared with 13 percent in the placebo group (P<0.001). Such positive results have been recently confirmed by the COMMANDS trial, which enrolled 356 patients randomized to receive either luspatercept or epoetin alfa and showed a higher rate of transfusion independence in the luspatercept group (80). Luspatercept shows a favorable safety profile (79), and is now approved for low-risk MDS by both FDA and EMA. A post-hoc analysis of the MEDALIST study provided support for the efficacy of Luspatercept also in a subgroup of patients affected by myelodysplastic syndromes/myeloproliferative neoplasm with ring sideroblasts and thrombocytosis (MDS/MPN-RT-T) (81); Luspatercept is currently under active investigation for other types of anemia in hematological malignancies, like myeloproliferative neoplasm-associated myelofibrosis (NCT03194542). Preliminary results of the latter study have been recently presented showing that transfusion-independency was reached in nearly one-third of patients on ruxolitinib that were initially transfusion dependent (82). Such promises and successes in anemia associated with hematological malignancies make ACVR ligand traps an attractive option also for anemia in other malignancies. To date, only limited/incomplete data are available for sotatercept in CCRA. Two Phase II placebo-controlled studies evaluated this anti-anemic drug in patients with metastatic breast cancer receiving myelosuppressive chemotherapy or with advanced/metastatic solid tumors receiving platinum-based chemotherapy (83). Unfortunately, both studies were terminated early due to slow patient accrual. Nevertheless, preliminary results indicated that mean hemoglobin levels increased ≥10 g/L in up to 66.7 percent of patients receiving sotatercept, as compared to 0 percent in placebo groups (83). In both studies, the safety profile was comparable to that of placebo groups. While caution is needed in interpretation due to the small populations studied, both studies point out the potential benefit of ACVR ligand traps for CCRA.

As mentioned above, hepcidin overexpression plays a major role in the anemia of inflammation (39, 84) and often contributes to anemia in cancer patients by reducing iron availability for erythropoiesis because of iron sequestration in cells, mainly in macrophages (39). This condition of FID can be, in principle, reversed by counteracting hepcidin activity (85). However, inhibiting hepcidin is not as simple as it could be theoretically anticipated. This hormone, which represents the master regulator of iron homeostasis (53), is continuously produced by the liver in discrete amounts (nM concentration). Direct hepcidin binders have been developed (85, 86), including a fully humanized monoclonal antibody (LY2787106), and a structured mirror-image RNA oligonucleotide (NOX-H94) (87), both with high affinity for human hepcidin. A phase 1 multicenter trial evaluated LY2787106 in 33 patients with different malignancies and anemia, 19 of whom received chemotherapy (88). LY2787106 was well-tolerated and induced initially dose-dependent increases in serum iron and transferrin saturation in patients with high hepcidin concentration. However, iron parameters quickly returned to baseline (within 1 week), and the primary efficacy endpoint (increased Hb level) was not reached (88). No further development of the drug has appeared in the literature.

NOX-H94 was initially evaluated in human volunteers subjected to experimental endotoxemia by lipopolysaccharide administration, with an observed increase of serum iron consistent with hepcidin inhibition (89). The drug was further shown to effectively inhibit hepcidin in a dose-dependent manner and was apparently well-tolerated (90). In 2014, a phase-II pilot study on 12 patients with anemia and hematological malignancies reported increased hemoglobin levels in 5 (91). However, again, no further developments have been reported. Another crucial point that makes hepcidin antagonization far from simple is represented by the substantial complexity of molecular regulation of hepcidin production (92, 93). Many strategies alternative to direct hepcidin neutralization have been reported, including antagonists of the bone morphogenetic protein 6 (BMP6) pathway with heparin derivatives (94–97), anti-BMP6 antibodies (98), dorsomorphin (99), and ferroportin stabilizers (100, 101). Further drugs that modulate the hepcidin-ferroportin axis include anti-hemojuvelin (HJV) antibodies (currently under evaluation in two clinical trials NCT05745883 and NCT05320198), and momelotinib. This latter drug is a promising and potent inhibitor of the ACVR1/ALK2 receptor, which has been developed for the treatment of anemia in patients with myelofibrosis (102–104). In a broad sense, numerous other drug classes indirectly regulate hepcidin (for a recent review, see (105)). Among them are interleukin 6 (IL-6) and interleukin-1 beta (IL-1b) inhibitors, signal transducer and activator of transcription 3 (STAT3) inhibitors, and Hypoxia-inducible factor–prolyl hydroxylase inhibitors (HIF-PHIs) (see below). While several clinical trials are underway (90), no data are available for patients with CCRA.

Hypoxia-inducible factors (HIFs) are transcription factors that mediate the cellular response to hypoxic conditions via the upregulation of several genes that are key to erythropoiesis, including those for EPO, the EPO receptor, and even for some critical proteins involved in iron metabolism (25, 106, 107). HIFs are heterodimers consisting of a constitutively expressed HIF-1β subunit and an O2-regulated HIF-α subunit. Under normoxic conditions, HIF is degraded by the hydroxylation of prolyl residues of the HIF-α subunit by HIF prolyl hydroxylases (PH). By preventing HIF degradation, HIF-PHIs stimulate endogenous EPO and promote erythropoiesis (25). Several HIF prolyl hydroxylase inhibitors have been developed and successfully used to treat anemia in chronic kidney disease (CKD) patients (108).

HIF-PHIs induce endogenous EPO at a concentration near the physiological range, representing a substantial difference with peaks determined by rhEPO administration and may explain the lower rate of cardiovascular adverse events. Other advantages over traditional ESAs include oral administration, lower costs, and lower immunogenicity. As mentioned above, it is worth noting that HIF-PHIs have pleiotropic favorable action on erythropoiesis beyond EPO stimulation (25). Indeed, critical players in iron metabolism like the Transferrin Receptor, Divalent Metal Transporter 1, and Ferroportin (the cell membrane receptor of hepcidin) are controlled by HIF through binding to their hypoxia-responsive elements, resulting in potent transcriptional activation (109–111). In this way, HIF-PHIs improve iron absorption from the gut and iron uptake and utilization by erythroid precursors through upregulation of transferrin receptor, ultimately leading to decreased levels of hepcidin, ferritin, and transferrin saturation (112). Roxadustat, one of the main HIF-PHIs already approved for anemia in CKD (113, 114), is currently being investigated in oncologic patients with anemia. MATTERHORN (NCT03263091) is a phase 3, randomized, double-blind, placebo-controlled study to assess the efficacy and safety of roxadustat in anemia of lower risk-MDS. In the dose-selection stage, Roxadustat was well-tolerated, and 37.5% of pts (9/24) with LR-MDS and low RBC transfusion burden achieved transfusion independence (TI) (115). In the MATTERHORN double-blind stage, a more significant percentage of pts in the Roxadustat arm compared with the placebo arm were TI responders (47.5% vs. 33.3%). However, this difference did not reach statistical significance (p=0.22) (116). A recent phase 2 open-label trial has investigated the efficacy and safety of Roxadustat for CCRA in various non-myeloid malignancies, including pancreas, breast, lung, and ovarian cancer, who have planned concurrent myelosuppressive therapies (117). Subgroup analyses demonstrated the efficacy of Roxadustat in correcting Hb regardless of tumor type and chemotherapy received. Adverse events were consistent with observations in patients with advanced-stage malignancies.

Despite their undoubtful benefit in the short-term correction of anemia, concerns have been raised about the possible off-target effects of HIF-PHIs during long-term use. Indeed, HIFs are ubiquitously expressed and regulate a broad spectrum of cell functions beyond erythropoiesis. Theoretically, HIF-PHIs may foster latent cancers since HIF activation in a hypoxic environment could promote cell proliferation and angiogenesis. This calls for caution, especially in older people. Nevertheless, no data thus far have revealed any effect of HIF-PHIs on tumor progression. On the other hand, chronic hypoxia caused by prolyl hydroxylase inhibition may inactivate cancer-associated fibroblasts, leading to decreased tumor metastatic spread. Ideally, future compounds able to selectively stimulate HIF2 could reduce undesirable off-target effects.